University of Southern California Dissertations and Theses

Improving antitumor efficacy of chimeric antigen receptor-engineered immune cell therapy with synthetic biology and combination therapy approaches

(USC Thesis Other)

Improving antitumor efficacy of chimeric antigen receptor-engineered immune cell therapy with synthetic biology and combination therapy approaches

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

IMPROVING ANTITUMOR EFFICACY OF CHIMERIC ANTIGEN
RECEPTOR-ENGINEERED IMMUNE CELL THERAPY WITH
SYNTHETIC BIOLOGY AND COMBINATION THERAPY
APPROACHES

by

Günce Ezgi Cinay

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BIOMEDICAL ENGINEERING)

December 2021

Copyright 2021 Günce Ezgi Cinay

ii
Epigraph
“Yaşamak şakaya gelmez,
büyük bir ciddiyetle yaşayacaksın
bir sincap gibi meselâ,
yani, yaşamanın dışında ve ötesinde hiçbir şey beklemeden,
yani, bütün işin gücün yaşamak olacak.
Yaşamayı ciddiye alacaksın,
yani, o derecede, öylesine ki,
meselâ, kolların bağlı arkadan, sırtın duvarda,
yahut kocaman gözlüklerin,
beyaz gömleğinle bir laboratuvarda
insanlar için ölebileceksin,
hem de yüzünü bile görmediğin insanlar için,
hem de hiç kimse seni buna zorlamamışken,
hem de en güzel, en gerçek şeyin
yaşamak olduğunu bildiğin halde.
Yani, öylesine ciddiye alacaksın ki yaşamayı,
yetmişinde bile, meselâ, zeytin dikeceksin,
hem de öyle çocuklara falan kalır diye değil,
ölmekten korktuğun halde ölüme inanmadığın için,
yaşamak, yani ağır bastığından…”

iii
(“Living is no laughing matter:
you must live with great seriousness
like a squirrel, for example—
I mean without looking for something beyond and above living,
I mean living must be your whole occupation.
Living is no laughing matter:
you must take it seriously,
so much so and to such a degree
that, for example, your hands tied behind your back,
your back to the wall,
or else in a laboratory
in your white coat and safety glasses,
you can die for people—
even for people whose faces you've never seen,
even though you know living
is the most real, the most beautiful thing.
I mean, you must take living so seriously
that even at seventy, for example, you'll plant olive trees—
and not for your children, either,
but because although you fear death you don't believe it,
because living, I mean, weighs heavier…”)

- Nazim Hikmet, 1947, from the poem “Yaşamaya Dair/On Living” (From Poems of Nazim
Hikmet, translated by Randy Blasing and Mutlu Konuk)

iv
Dedication
This dissertation is dedicated to my mother Necla Cinay and my father Sami Cinay. Witnessing
your life taught me to choose service to humanity as the purpose of my life. Seeing you as life-long learners
led me to choose education and science as my way of achieving this purpose. Watching your attitude
towards life’s challenges taught me to never give up and to stay focused on my path. I cannot thank you
enough for paving the way with your enlightened view of life and raising me as a world citizen.
Bu doktora tezi annem Necla Cinay ve babam Sami Cinay’a adanmıştır. Insanlık için yararlı işler
yapmayı bir hayat amacı olarak almayı, eğitim ve bilimi bunun için yol olarak seçmeyi, bu yolda pes
etmeden ve umutla devam etmeyi ben sizin hayatınıza şahit olarak öğrendim. Aydın hayat görüşünüzle
önümü açtığınız ve beni bir dünya insanı olarak yetiştirdiğiniz için ne kadar teşekkür etsem az.

v
Acknowledgements
The doctor of philosophy is the highest academic level awarded in a field of study. The seeds of
this success had been planted very early on by my grandparents, parents, and all my prior teachers. This
work could not have been possible without your efforts and support.
First and foremost, I would like to express my deepest gratitude to my PhD advisor, Professor Pin
Wang, for giving me the opportunity to do research in his laboratory in the field and topics I am excited
about. Throughout my PhD he has given me guidance to explore the core research questions at hand, while
providing the much needed freedom and space to grow as an independent researcher. He put in place the
structure in which my only concern has been my research and teaching duties and not other formalities of
the institution. Because of him I was able to work closely and collaborate with peers within our research
group and other research labs at USC. The lab culture he formed in our research group has been conducive
to collaboration and team work, and those who contributed to a project were rewarded fairly. With these
supportive components in place, I have been able to grow professionally in my own way and pace. I have
also had the opportunity to explore my other passions with the resources that USC offers. Dr. Wang’s
multifaceted education background, research record and more currently his entrepreneurship efforts,
inspired me to hold onto and cherish my versatile side. I respect him a lot for consistently setting an
incredible example for me as a curious and open-minded scientist who continuously keeps himself up-to-
date with the latest developments in the field.
Next, I would like to thank my committee members, Professor Megan McCain, Professor Keyue
Shen, and Professor Nicholas Graham for serving in my committee and offering me valuable advice during
my qualifying exam. Their guidance helped me plan my in vivo studies more realistically and graduate
timely. Professor Megan McCain also set an outstanding example as an academician throughout the times
I served as her teaching assistant. I am grateful for Professor Keyue Shen for his supervision in our
collaboration project. I appreciate Professor Nicholas Graham for his insightful questions and comments

vi
during my qualifying exam, which helped me practice critical thinking and dive deeper into my research
projects.
I would also like to express my gratitude for all my teachers who had educated me and added value
to my life throughout my education. I am grateful for all my teachers who inspired me to continue my career
in science and engineering, starting from my primary school teacher, Yeter Aktas, to my undergraduate
research advisors Dr. Olcay Anac and Dr. Fatma Nese Kök, to my masters research advisor Dr. Seda Kizilel.
Very importantly, my heartfelt thanks go to all my lab mates. It has been an honor and privilege to
work and study with them. Dr. Xiaolu Han and Dr. Elizabeth Siegler trained me on the essential skills to
carry out my research. My collaborator, Dr. Xiaolu Han and Dr. Yun Qu, who I am deeply grateful for,
helped make our projects successful, and the process of researching meaningful. I would like to thank to
Dr. Yuta Ando, my first research mate at USC and my collaborator from Shen Lab, who made the time we
spent in lab as pleasant as the time we spent outside school. I also would like to thank to Dr. Xianhui Chen,
Zachary Dunn, Dr. Jennifer A. Rohrs, Dr. Natnaree Siriwon, Dr. Paul D. Bryson, Dr. John Mac, Dr.
Xiaoyang Zhang, Dr. Si Li, Dr. Yu Jeong Kim, Jiangyue Liu, Chumeng Cheng and Melanie MacMullan
for our reciprocal attempts to make each other’s PhD journeys easier and for all the moments we smiled
together. This work would not have been possible without their genuine support. My PhD years would have
been dull without their companionship. I would also like to thank my undergraduate student mentees
Neelesh Bagrodia and Nicolas Ritcheson for their assistance in my research.
I would also like to express my gratitude to USC Graduate School for funding me with the Provost’s
Fellowship in the first two years of my program. I also thank the staff of USC’s Department of Biomedical
Engineering and the Mork Family Department of Chemical Engineering and Materials Science, specifically
Annie Lee Houang, Mischalgrace Diasanta and William Yang, for their help facilitating the behind the
scenes work regarding my course registration and teaching/research assistantship duties.
Also, I want to express my thanks for my fellow BME PhD students. It was a great pleasure for me
to be a part of the BME family. I wholeheartedly thank Dr. Elizabeth Siegler, Dr. Jennifer A. Rohrs, Dr.
Yuta Ando, Dr. John F. Sunwoo, Dr. Qianhui (Jessica) Wu, Dr. Davi Lyra Leite, Dr. Marilena Dimotsantou,

vii
Dr. Toey Thuptimdang, Dr. Julio E. Villalon-Reina, Dr. Andrew Petersen, Dr. Joycelyn Yip, Dr. Nathan
Cho, Dr. Nethika Ariyasinghe, Dr. Jeffrey Santoso, Dr. Gene Yu, Virat Agrawal, Adam Mergenthal, Hydari
Masuma Begum, and Alireza Marjaninezhad, for their friendship and support throughout my PhD years.
Your unceasing encouragement has kept me strong.
I wholeheartedly thank all my friends who have both grieved and celebrated with me from miles
away over the years through this PhD journey. My warmest thanks go to Nisa Nildan Dilaver and Bahar
Temizer. It has been a joy growing up with you since the age of seven. Without your genuine support, I
would not have adjusted to life in the US and finished this degree. Also, I would like to thank my beloved
friends Zeynep Esencan, Ece Hasret Sönmez, Irina Crivet, Dr. Pelin Erkoc, Dr. Canan Ipek, Dr. Gozde
Barim, Dr. Alperen Ozdemir, Dr. Beril Kiragasi, Dr. Betul Mutlugün, Dr. Sözen Ozkan Grigoras, Dr Mert
Besken, Dr. Andres Stucky, Agnieszka Satola, , Dr. Majid Monji, Tarun Mundluru, Michael Hopek, Melih
Iseri, Irmak Balcik, Ulubilge Ulusoy, Xiaobei Wu, Dr. Jeremy Liu, Dr. Zümra Peksaglam, Dr. Alireza
Delfarah, Dr. Amirhosein Mousavi, Dr. Bilgenur Baloglu, Rene Prubes, Aarju Goyal, Nripsuta Saxena, and
many others, for their unwavering support and close friendship.
Canim kardesim Guvenc Cinay, aramizdaki bagin yillar gectikce nasil evrilip yeni anlamlar
kazandigini heyecanla izliyorum. Beraberken yakaladigimiz cocuksu heyecan ve gulusler icin ve bana on
yil onceki beni hatirlatip hayatima tazelik kattigin icin kocaman sicacik bir tesekkur.
My dear brother, Guvenc Cinay, I am excited to watch how our relationship has evolved over the
years. I appreciate how you remind me of me ten years ago. It brings me a fresh perspective. Big and warm
thanks to you for the childlike enthusiasm we create when we are together.
Son olarak, en derin tesekkurlerimi aileme sunmak istiyorum. Annem Necla Cinay ve babam Sami
Cinay’a, bana kendim olabilmem icin alan sagladiklari ve kendimi var edebilecegim araclari verdikleri icin,
bana hep destek olup bana inanmayi hic birakmadiklari icin, ve bugun, bundan sonra da hep yanimda
olacaklarini bilmenin ic huzurunu tasiyabildigim icin cok tesekkur ederim. Anne, bana yillardir “Turkiye
bir gun seninle gurur duyacak!” diyerek bir Kutup Yildizi verdin, yolumu sasirdigimda oraya bakiyorum.
Kucuklugumden beri elimi tutup beni goturdugun tiyatrolar, sinemalar, muzeler, beni bugunku cok yonlu

viii
ve duyarli insan yapti, bunlar beni ben yapan ozelliklerden ve senin eserinler. Baba, hayatimdaki duzenin,
dengenin, umudun ve sertlik yerine tatli sozu sectigim anlarin mimari sensin. Simdi geriye bakinca
anliyorum ki senin adimlarini takip ederek, kabugumu kirip dogdugum yerden uzakta kök salmaya cesaret
edebildim. Bu kisisel detaylarin yeri burasi degilmis gibi gorunse de ben aksini dusunuyorum. Bir bilim
insani yetistirmek icin o insani sadece teknik bilgilerle donatmak yeterli degil. Yaptigi buluslarin sadece
kucuk bir topluluk tarafindan anlasildigi veya toplumdan ayrismis bir figur olmak gorevimi eksik yapmis
hissettirirdi. Toplumu ileri tasima hayaliyle uygulamali bilimleri secip, kapsayici ve esitlikci isler yapma
gayesi tasiyan bir bilim insani olmamin temelleri size ait.
Last but not least, I would like to express my deepest gratitude to my family. I am indebted to my
mother Necla Cinay and my father Sami Cinay for providing me with the tools and the space to be myself.
You have consistently supported me and kept believing in me. Knowing that you will always be there for
me is the reason for the inner peace I have today, which I am deeply grateful for. Mother, by telling me
“Turkiye will be proud of you one day!”, you have given me a North Star. Now, I look up North when I
feel lost. The numerous plays, movies, and museums you took me to by holding my little hands, made me
the versatile and sensitive person I am today. I feel lucky to be this aware. Father, you are the architect of
the discipline, balance, and hope in my life. You are my soft but strong side. You are the reason for the
moments I display endearment over harshness. When I look back, I realize that I was able to break out of
my shell and put down roots far away from Istanbul by following your footsteps. These personal details
may seem irrelevant to mention here, but I believe that the making of a scientist is not simply equipping a
person with technical knowledge. You have laid the foundations for me to be a scientist who has chosen
applied sciences with the dream of moving society forward, and who aims to do inclusive and equitable
work by raising me as a versatile person.
Without you all, this work would not have been possible.

ix
Table of Contents
Epigraph ......................................................................................................................................... ii
Dedication ..................................................................................................................................... iv
Acknowledgements ........................................................................................................................ v
List of Tables .............................................................................................................................. xiii
List of Figures ............................................................................................................................. xiv
Abstract ....................................................................................................................................... xvi
Chapter 1: General Introduction .................................................................................................... 1
1.1 Cancer Therapies ............................................................................................................... 1
1.2 Adoptive Cell Therapy ....................................................................................................... 3
1.3 CAR T Cells ....................................................................................................................... 4
1.3.1 FDA-approved CAR T cell therapies ....................................................................... 5
1.3.2 Challenges of CAR T cell therapies .......................................................................... 6
1.4 Natural Killer (NK) Cells ................................................................................................... 8
1.4.1 Sources of NK cells for cancer therapy .................................................................. 11
1.4.2 CAR.NK cells ......................................................................................................... 12
1.4.3 CAR engineering of primary NK cells ................................................................... 14
1.4.4 CAR engineering of NK cell lines .......................................................................... 14
1.4.5 CAR NK cells in clinical trials ............................................................................... 15
1.4.6 NK cell combination therapy .................................................................................. 16
1.5 CAR-engineered T and NK Cell Clinical Trials with Solid Tumors ............................... 16
1.6 Challenges of CAR-engineered Immune Cells Therapies in Solid Tumors .................... 17
1.6.1 Tumor antigen heterogeneity .................................................................................. 17
1.6.2 Trafficking and infiltration into tumor tissue .......................................................... 18
1.6.3 Immunosuppressive tumor microenvironment ....................................................... 19
1.7 Special Topics 1 ............................................................................................................... 22
1.7.1 Platelets ................................................................................................................... 22
1.7.2 Platelet-derived microparticles ............................................................................... 24
1.7.3 Different roles of platelets ...................................................................................... 26
1.7.4 Role of platelets in cancer ....................................................................................... 29
1.8 Special Topics 2 ............................................................................................................... 32
1.8.1 Adenosine in the tumor microenvironment ............................................................ 32
1.8.2 Adenosine receptors ................................................................................................ 33
1.8.3 A2aR blockade ........................................................................................................ 35
1.8.4 SCH-58261 ............................................................................................................. 36
1.8.5 Drug delivery .......................................................................................................... 37
1.8.6 Vascularization in tumor area ................................................................................. 38
1.8.7 Nanoparticles .......................................................................................................... 39

x
Chapter 2: Adnectin-Based Design of Chimeric Antigen Receptor for T Cell Engineering ....... 42
2.1 Abstract ............................................................................................................................ 42
2.2 Introduction ...................................................................................................................... 43
2.3 Materials and Methods ..................................................................................................... 45
2.3.1 Construction of plasmids ........................................................................................ 45
2.3.2 Cell lines and culture media .................................................................................... 46
2.3.3 Retroviral vector production ................................................................................... 46
2.3.4 T cell transduction and expansion ........................................................................... 47
2.3.5 Surface immunostaining and flow cytometry ......................................................... 47
2.3.6 EGFR surface staining and quantification .............................................................. 47
2.3.7 Intracellular cytokine staining ................................................................................. 48
2.3.8 Degranulation assay ................................................................................................ 48
2.3.9 Cytotoxicity assay ................................................................................................... 48
2.3.10 Antitumor efficacy of CAR T cells in a non-small cell lung cancer
xenograft mouse model .................................................................................................... 49
2.3.11 Statistical analysis ................................................................................................. 49
2.4 Results .............................................................................................................................. 49
2.4.1 Design and generation of EGFR-specific CAR constructs ..................................... 49
2.4.2 Evaluation of adnectin-based CARs ....................................................................... 51
2.4.3 E3-CAR displayed lower binding affinity toward EGFR compared to Cetux-
CAR ................................................................................................................................. 52
2.4.4 E3-CAR T cells displayed higher selectivity against EGFR overexpressing
cancer cells from lower EGFR-expressing cells .............................................................. 53
2.4.5 E3-CAR had comparable reactivity against H292 lung cancer cells to that of
Cetux-CAR ...................................................................................................................... 56
2.4.6 E3-CAR T cells showed similar antitumor efficacy to Cetux-CAR T cells in
vivo ................................................................................................................................... 57
2.5 Discussion ........................................................................................................................ 59

Chapter 3: Engineering CAR-Expressing Natural Killer Cells with Cytokine Signaling
and Synthetic Switch for an Off-the-Shelf Cell-Based Cancer Immunotherapy ......................... 63
3.1 Abstract ............................................................................................................................ 63
3.2 Introduction ...................................................................................................................... 63
3.3 Materials and Methods ..................................................................................................... 66
3.3.1 Cell culture .............................................................................................................. 66
3.3.2 Virus production ..................................................................................................... 66
3.3.3 Transduction of NK-92 and SKOV3 cells .............................................................. 66
3.3.4 CAR detection on NK cell surface .......................................................................... 67
3.3.5 Cytokine release assay ............................................................................................ 67
3.3.6 Cytotoxicity assay ................................................................................................... 68
3.3.7 Chemical inducer of dimerization sensitivity assay ................................................ 68
3.3.8 Xenograft tumor model ........................................................................................... 68
3.3.9 Ex vivo NK cell staining ......................................................................................... 69
3.3.10 Statistical analysis ................................................................................................. 69
3.4 Results and Discussion .................................................................................................... 69
3.5 Conclusions ...................................................................................................................... 76

xi

Chapter 4: CAR T Cell-Platelet Complexation for Enhanced Tumor Homing and
Antitumor Efficacy in Solid Tumors ........................................................................................... 78
4.1 Abstract ............................................................................................................................ 78
4.2 Introduction ...................................................................................................................... 79
4.3 Methods............................................................................................................................ 81
4.3.1 Cell lines ................................................................................................................. 81
4.3.2 Plasmid construction ............................................................................................... 82
4.3.3 Lentiviral and retroviral vector preparation and T cell transduction ...................... 82
4.3.4 CAR detection on T cell surface ............................................................................. 84
4.3.5 Freezing and thawing PLTs .................................................................................... 84
4.3.6 PLT activation and PMP formation ........................................................................ 85
4.3.7 CAR T-PLT/PMP complex formation .................................................................... 85
4.3.8 Cytokine release assay ............................................................................................ 86
4.3.9 Degranulation assay ................................................................................................ 87
4.3.10 Transmigration assay ............................................................................................ 87
4.3.11 Trans-Matrigel cell migration assay ..................................................................... 87
4.3.12 Cytotoxicity assay ................................................................................................. 88
4.3.13 In vivo antitumor activity ...................................................................................... 89
4.3.14 Ex vivo tissue analysis ........................................................................................... 90
4.4 Results .............................................................................................................................. 91
4.4.1 Anti-CD19 CAR is expressed in human T cells ..................................................... 91
4.4.2 PLTs and PMPs form stable complexes with CAR T cells .................................... 92
4.4.3 IFN-𝛾 secretion of CAR T cells is unaffected by complex formation
between CAR T cells and PLTs/PMPs in vitro ............................................................... 97
4.4.4 Degranulation of CAR T cells is unaffected by complex formation between
CAR T cells and PLTs/PMPs in vitro .............................................................................. 99
4.4.5 Migration ability of CAR T cells towards a chemoattractant is unaffected by
complex formation between CAR T cells and PLTs/PMPs in vitro ................................ 99
4.4.6 CAR T cell-PLT complexes enhance 3D Trans-Matrigel migration ability of
CAR T cells towards a chemoattractant in vitro ............................................................ 100
4.4.7 Cytotoxicity of CAR T cells is unaffected by complex formation between
CAR T cells and PLTs/PMPs in vitro ............................................................................ 102
4.4.8 CAR T cell-PLT complexes enhance antitumor efficacy compared to CAR
T cells in a mouse xenograft model ............................................................................... 104
4.4.9 CAR T cells in CAR T cell-PLT complexes have greater T cell infiltration
into the tumor and persistence in blood compared to PLT-free CAR T cells ................ 108
4.5 Discussion ...................................................................................................................... 111

Chapter 5: CAR-Engineered Natural Killer Cells as a Carrier of Drug-Encapsulated
Nanoparticles Targeting Adenosine Receptors in Solid Tumors ............................................... 119
5.1 Abstract .......................................................................................................................... 119
5.2 Introduction .................................................................................................................... 120
5.3 Methods.......................................................................................................................... 124
5.3.1 Cell lines and reagents .......................................................................................... 124
5.3.2 Lentiviral and retroviral vector preparation .......................................................... 124

xii
5.3.3 Transduction of NK-92 and SKOV3 cells ............................................................ 125
5.3.4 CAR detection on NK cell surface ........................................................................ 125
5.3.5 Preparation of cMLVs ........................................................................................... 126
5.3.6 Nanoparticle conjugation with cells and in situ PEGylation ................................ 127
5.3.7 Quantification of cell-bound cMLVs .................................................................... 127
5.3.8 NK cell viability assay .......................................................................................... 127
5.3.9 NK cell proliferation assay ................................................................................... 128
5.3.10 Transmigration assay .......................................................................................... 128
5.3.11 Cytokine release assay ........................................................................................ 129
5.3.12 Cytotoxicity assay ............................................................................................... 129
5.3.13 In vivo antitumor activity .................................................................................... 129
5.4 Results ............................................................................................................................ 131
5.4.1 Anti-mesothelin CARs with or without membrane-bound IL-15/IL-15Rα
complex are expressed in NK-92 Cells .......................................................................... 131
5.4.2 cMLVs are stably conjugated to the surface of NK cells ..................................... 133
5.4.3 In vitro CAR.NK cell viability is unaffected by drug-loaded nanoparticle
conjugation onto cell surface ......................................................................................... 136
5.4.4 In vitro CAR.NK cell expansion is unaffected by drug-loaded nanoparticle
conjugation onto cell surface ......................................................................................... 137
5.4.5 In vitro CAR.NK cell migration ability is unaffected by drug-loaded
nanoparticle conjugation onto cell surface .................................................................... 138
5.4.6 In vitro CAR.NK cell IFN-γ secretion is unaffected by drug-loaded
nanoparticle conjugation onto cell surface .................................................................... 140
5.4.7 In vitro CAR.NK cell cytotoxicity is unaffected by drug-loaded nanoparticle
conjugation onto cell surface ......................................................................................... 142
5.4.8 Presence of free SCH or cMLV(SCH) does not exert cytotoxicity on
SKOV3.meso cells or NK cells in vitro ......................................................................... 144
5.4.9 CAR.NK cells conjugated with SCH-loaded nanoparticles have greater
antitumor efficacy compared to cMLV(SCH)-free CAR.NK cells ............................... 145
5.4.10 ameso.CAR.NK cells conjugated with SCH-loaded nanoparticles elongated
mice survival .................................................................................................................. 151
5.5 Discussion ...................................................................................................................... 152

Chapter 6: Conclusions and Future Perspectives ....................................................................... 164

References .................................................................................................................................. 179

xiii
List of Tables
Table 5.1 Statistical analysis summary of the tumor growth curve at each time point. .............. 148
Table 5.2 Statistical analysis summary of the percent change in initial mice body weight at each
time point. .................................................................................................................... 149
Table 6.1 Existing issue and future recommendations for the implementation of affordable CAR
T cell therapy in the US. .............................................................................................. 176

xiv
List of Figures
Figure 1.1 Adoptive CAR T cell therapy. ........................................................................................ 3
Figure 1.2 Chimeric antigen receptor structure. .............................................................................. 4
Figure 1.3 Limitations to durable remissions after CAR T cell therapy. ......................................... 7
Figure 1.4 Various functions and characteristics of CAR.NK cells. ............................................... 9
Figure 1.5 Immunosuppressive tumor microenvironment. ............................................................ 20
Figure 1.6 Extracellular adenosine and its effects on immune cells. ............................................. 32
Figure 2.1 Engineering T cells with scFv-based CAR and adnectin-based CAR to target tumor. 50
Figure 2.2 Evaluation of adnectin-CARs based on their expression and functional activity in
human T cells. .............................................................................................................. 51
Figure 2.3 Comparison of binding affinity of Cetux-CAR and E3-CAR. ..................................... 53
Figure 2.4 Activation of E3-CAR T cells is positively correlated with EGFR density on target
cells. ............................................................................................................................. 55
Figure 2.5 Activity of Cetux-CAR and E3-CAR against H292 lung cancer cells. ........................ 56
Figure 2.6 Antitumor efficacy of CAR T cells in a human lung cancer xenograft model. ............ 58
Figure 2.7 Trafficking of CAR T cells in the human lung cancer xenograft NSG mice model. ... 59
Figure 3.1 CAR expression on NK-92 cells and proliferation of CAR.NK cells. ......................... 70
Figure 3.2 Evaluation of interferon-γ secretion and cytotoxicity of CAR.NK cells. ..................... 71
Figure 3.3 iCas9 expression on NK-92 cells and viability of iCas9-expressing CAR.NK cells in
the presence of CID. .................................................................................................... 73
Figure 3.4 Evaluation of tumor volumes of treated mice. ............................................................. 74
Figure 3.5 Evaluation of NK cells in tail blood and bone marrow. ............................................... 75
Figure 3.6 Evaluation of NK cells in tail blood and bone marrow after PBS or CID
administration. ............................................................................................................. 76
Figure 4.1 Rationale design of CAR T cell-PLT complexes to improve CAR T cell homing into
solid tumors and evaluation of CAR expression on T cells. ........................................ 92
Figure 4.2 Evaluation of complex formation between CAR T cells and PLTs or PMPs. ............. 95
Figure 4.3 Complexation of CAR T cells with PLTs and PMPs does not alter CAR T cell
function. ....................................................................................................................... 98
Figure 4.4 Complexation of CAR T cells with PLTs and PMPs does not alter in vitro cytotoxicity
of CAR T cells. .......................................................................................................... 104
Figure 4.5 CAR T cells in complex with PLTs demonstrated superior in vivo antitumor efficacy
compared to CAR T cells alone in a human lung cancer xenograft model. .............. 105
Figure 4.6 Supplementary data for the evaluation of tumor volume after CAR T-PLT and CAR T-
PMP treatment and swimmer plot analysis for T cell treatments. ............................. 108
Figure 4.7 CAR T cells in complex with PLTs exhibited superior accumulation in tumors and
persistence in blood compared to CAR T cells alone in a human lung cancer xenograft
model. ......................................................................................................................... 109
Figure 5.1 CAR.NK cells-to-cMLV conjugation is successful and drug-loaded cMLV conjugation
does not negatively affect CAR.NK cell viability and proliferation. ......................... 132
Figure 5.2 CAR.NK cells conjugated with nanoparticles show similar levels of cell migration and
IFN-γ secretion compared to nanoparticle free CAR.NK cells. ................................ 139
Figure 5.3 CAR.NK cells conjugated with nanoparticles show similar or elevated levels of
cytotoxicity compared to nanoparticle free CAR.NK cells. ...................................... 143
Figure 5.4 CAR.NK cells conjugated with SCH-loaded nanoparticles enhance antitumor efficacy
compared to nanoparticle free CAR.NK cells in a human ovarian cancer xenograft
model. ......................................................................................................................... 146
Figure 5.5 Swimmer plot analysis of mice treated with CAR.NK cell transfers. ........................ 150

xv
Figure 5.6 ameso.CAR.NK cells conjugated with SCH-loaded nanoparticles elongated mice
survival compared to nanoparticle free CAR.NK cells in a human ovarian cancer
xenograft model. ........................................................................................................ 151

xvi
Abstract
This dissertation is a compilation of four projects. The projects outlined herein utilize synthetic
biology and combination therapy approaches with the goal of improving the antitumor efficacy of CAR-
engineered immune cell therapies, especially in solid tumors. Chapter 1 gives a general introduction to
cancer immunotherapy field, CAR-engineering of immune cells and challenges of using CAR-modified
immune cells in the treatment of solid tumors. Chapter 1 also provides a more detailed look into the special
topics that would help the reader to have an in-dept understanding of the background for the projects
discussed in this dissertation. In Chapter 2, the study called Adnectin-Based Design of Chimeric Antigen
Receptor for T Cell Engineering showed the successful construction of adnectin-based CARs, as an
alternative to using single chain variable fragments (scFv) as extracellular antigen recognition domains.
The results demonstrate that bearing equivalent potency to traditional CARs, adnectin-based CARs may
benefit from reduced immunogenicity, increased tumor selectivity, and improved safety profile due to
optimal affinity tuning. In Chapter 3, the study named Engineering CAR-expressing Natural Killer Cells
with Cytokine Signaling and Synthetic Switch for an Off-the-shelf Cell-based Cancer Immunotherapy
demonstrated the capacity at which the off-the-shelf candidate NK-92 cells can be engineered with synthetic
biology tools for enhanced tumor targeting capability, better proliferative potential, and treatment
regulation. In Chapter 4, a study named CAR T Cell-Platelet Complexation for Enhanced Tumor Homing
and Antitumor Efficacy in Solid Tumors is presented. Inspired by the clinical success of CAR T cells as
well as the intrinsic tendency of platelets (PLTs) to interact with and assist T lymphocytes to penetrate into
tumors by degrading tumor basal membrane and therefore accumulate in tumor sites, we present a cell-only
combinational “living-drug” strategy, CAR T cell-PLT complexes, to enhance the antitumor effects of CAR
T cells in their site of action. In Chapter 5, in the study named CAR-Engineered Natural Killer Cells as a
Carrier of Drug-Encapsulated Nanoparticles Targeting Adenosine Receptors in Solid Tumors, we
combined immunotherapy with nanomedicine to better treat solid tumors. We used target-specific CAR-
engineered NK cells as active carriers of nanoparticles loaded with an immunomodulatory drug to deliver

xvii
the drug to solid tumor sites. We aimed to inhibit immunosuppressive effects of the tumor
microenvironment and maintain CAR.NK cell effector functions in solid tumors, with minimal toxicity.
Finally, Chapter 6 concludes this dissertation with future perspectives for CAR T cell therapies.

1
1 Chapter 1: General Introduction
1.1 Cancer Therapies
Cancer is one of the leading causes of morbidity and mortality worldwide and urgently requires
effective treatments. For many years, researchers and cancer physicians have relied only on surgery,
radiation therapy and chemotherapy as primary treatment modalities
1,2
. In 1990s, the forth pillar of cancer
therapy, cancer immunotherapy has emerged
1
. Immunotherapies strengthen the patients’ immune system
to attack tumors
3
. Cancer immunotherapy has several subgroups, including checkpoint inhibitors,
cytokines, co-stimulatory receptor agonists, cancer vaccines, other emerging approaches such as oncolytic
viruses and bispecific antibodies, and cell-based cancer therapies
4
.
Immune checkpoints maintain physiological immune responses and protect healthy cells from
immune attack
5
. The most common checkpoint inhibition strategies are PD-1/PD-L1 blockade and CTLA4
inhibition which were awarded the 2018 Nobel Prize in Physiology or Medicine
4,6
. Researchers are
currently investigating ways to reduce the severe side effects of systemically administered checkpoint
inhibitors and the reasons of the lack of response in many patients who were treated with checkpoint
inhibitors
4
.
Injected lymphocyte-promoting cytokines, such as interferons and interleukins, directly stimulate
the proliferation and activation of immune cells
4
. Interleukins stimulate the activity and maturation of
numerous immune cells including NK cells
7
and CD8
+
T cells
8
. Moreover, cytokine therapy using IL-15
and IL-21 has been shown to have some advantages over IL-2 in terms of causing less autoimmune
reactions while promoting CD8
+
T cell proliferation
9,10
. Current research and clinical trials have been
devoted to investigating the use of cytokines in combination treatment strategies with other cytokines,
checkpoint inhibitors or chemotherapies to reduce the high treatment doses when used in independent
therapies in order to reduce the adverse effects of these treatments
11
.

2
Agonistic antibodies against co-stimulatory receptors, such as CD28 and 4-1BB, specifically bind
to T cell receptors and trigger intracellular cell signaling which promote T cell growth, persistence and
anticancer activity
12,13
. These antibodies are not approved by the FDA yet and their toxicity is being
evaluated in clinical trials
4
.
There are various classes of cancer vaccines including tumor cell lysate, nucleic acids, neo-antigens
or dendritic cells
14
. Dendritic cell vaccines are the most commonly studied cell-based vaccine class
15
.
Nucleic acid therapeutics rely on the intracellular delivery of exogenous nucleic acids into APCs, their
translation to induce antigen expression and presentation of the targeted antigens to T cells to activate T
cells against corresponding antigen-expressing tumor cells
16
. Neoantigens are tumor-specific antigens
which originate from somatic DNA alterations in cancer cells. There are several advantages of using
neoantigens. One of them is their limited off-target adverse effects due to their sole presence in cancer cells.
Another one is their potential in treating heterogeneous cancers due to the possibility of generating vaccines
with a large number of neoantigens in them
17
.
Cell-based cancer therapies have demonstrated unprecedented clinical success for some advanced
cancer patients
3
and adoptive cell transfers are discussed in detail in the next section.
Despite promising advances, the clinical use of immunotherapies present challenges related to their
safety and efficacy. In terms of safety related challenges, while exerting cytotoxicity against cancer tissues,
immunotherapy may attack healthy tissues. In addition, immunotherapies may lead to major adverse side
effects such as cytokine release syndrome and vascular leak syndrome, which cause severe hypotension,
fever and renal dysfunction
18–20
. In terms of efficacy related issues, majority of immunotherapies have been
investigated in hematological cancers and their efficacy is limited in solid cancer tumors, due to delivery
barriers and immune-suppressive tumor microenvironments in solid tumors, which are further discussed
below
21
. Moreover, only a small subset of patients respond to immunotherapies
22
.

3
1.2 Adoptive Cell Therapy
Adoptive cell therapy (ACT) is a treatment that uses cancer patients’ own immune cells with
antitumor activity, following the ex vivo stimulation and expansion of autologous or allogeneic
lymphocytes, and reinfusion of the expanded lymphocyte population back into the cancer patient (Figure
1.1)
23
.

Figure 1.1 Adoptive CAR T cell therapy.
CAR T cell therapy is a type of cancer treatment where a patient’s T cells are modified to gain specificity towards
cancer cells. Figure adapted from National Cancer Institute. https://www.cancer.gov/about-
cancer/treatment/types/immunotherapy/t-cell-transfer-therapy.

In the ACT, there exist two main approaches. In the first approach, autologous tumor-reactive T
cells are expanded from tumor biopsies and reinfused to the patient, as in the case of tumor infiltrating
lymphocyte (TIL) therapy
24,25
. Alternatively, peripheral blood T cells are genetically modified to express a
tumor-specific T cell receptor (TCR) or a so-called chimeric antigen receptor (CAR; a fusion protein that
links scFv-mediated tumor antigen-binding with intracellular endodomains associated with T cell

4
activation)
26
. Alternatively, another type of ACT, CAR-engineered natural killer (NK) cells, has recently
been investigated in clinical settings and has shown promising results.
ACT of tumor-specific lymphocytes have demonstrated great clinical success for the treatment of
some human hematological malignancies, including leukemia and lymphoma
27–31
. Researchers’ ability to
genetically modify human lymphocytes and use them to mediate cancer regression have opened possibilities
for the utilization of ACT immunotherapy as a promising new approach to cancer treatment in a wide
variety of cancer types. Thus, efforts to engineer T cells with enhanced tumor-specificity presented itself
as an area of intense research.
1.3 CAR T Cells
One approach to engineer immune cells is to modify them to express chimeric antigen receptors
(CARs)
32
. CARs are artificial receptors that can redirect T cells to tumor targets by genetic modification of
the autologous or allogeneic lymphocytes
33
. Recently, the development of combinatorial strategies using
CAR-engineered immune cells together with other drugs and delivery vehicles has become an important
area of research, with the aim to improve the antitumor efficacy of this therapy in various tumor types
34
.

Figure 1.2 Chimeric antigen receptor structure.
Chimeric antigen receptor (CAR) structure includes an extracellular antigen binding domain (single chain variable
fragment; scFv) fused to intracellular co-stimulatory domains (i.e., CD28 or 4-1BB) and signaling domains (i.e.
CD3z). Figure adapted from Lim, Wendell A. and June, Carl H. The Principles of Engineering Immune Cells to Treat
Cancer. Cell. 2017 February 09; 168(4): 724–740.
CARs consist of an external recognition domain, called single-chain variable fragment (scFv) fused
to a transmembrane domain followed by one or more cytoplasmic signaling domains (Figure 1.2). The

5
antibody-derived scFv domain is responsible for antigen recognition, while the endodomains are
responsible for cell activation
34
.
The extracellular binding domain of CARs are mostly derived from a monoclonal antibody
fragment (scFv). The scFv binding to a tumor antigen activates the T cell in a major histocompatibility
(MHC)-independent manner and induces cytotoxic response
35
. Novel CAR constructs have been developed
by altering targeting domains with different molecules, such as nanobodies, designed ankyrin repeat
proteins (DARPins), ligands or receptors instead of scFvs
36–39
. CAR design follows a modular design in
which CAR constructs can be independently created and feasibly used in different systems, such as
combining any scFv with any signaling/co-signaling domain. CAR classification has been done based on
the alternations in intracellular signaling domains. First-generation CARs do not possess any intracellular
signaling region and only include signaling motifs which are derived from a T cell receptor (TCR) CD3 (ζ
or γ chain)
40,41
and an scFv region. These initial CARs comprising the scFv and the CD3ζ signaling domain
have shown to give T cells transient activation and cytotoxicity
42
, yet CAR-engineered cells become
inactivated by tumors leading them to anergy
41
. In order to eliminate this issue and increase CAR
functionality, second-generation CARs have been designed by incorporating additional costimulatory
domains (i.e. CD28, 4-1BB). 2
nd
generation CAR design has enhanced T cell expansion and achieved
persistent T cell activation
43
. Later, by the addition of one more costimulatory domain, 2
nd
generation CARs
have evolved to 3
rd
generation CARs which improved the efficacy of CAR T cells by enhancing their
activation, survival and proliferation
44
. Research indicates that CARs with CD3ζ and CD28 domains and/or
additional 4-1BB domains demonstrate enhanced killing upon binding to the complimentary tumor
associated antigens (TAAs)
45,46
.
1.3.1 FDA-approved CAR T cell therapies
With the exceptional success of adoptive CAR T cell transfer for treating hematological
malignancies
47
, four CD19-targeted and one BCMA-targeted CAR T cell products received approval from

6
the US Food and Drug Administration (FDA). The FDA approved the first CAR T cell therapy, Novartis’
Kymriah™ (tisagenlecleucel) in August 2017, calling it “a historic action ushering in a new approach to
the treatment of cancer and other serious and life-threatening diseases”
48
. Kymriah™ is approved for B-cell
precursor acute lymphoblastic leukemia (ALL) patients up to age 25 that resisted therapies or relapsed
48
.
Subsequently, Kymriah™ and Kite’s Yescarta™ (axicabtagene ciloleucel) are approved to treat adults with
large B-cell lymphoma (the most common type of non-Hodgkin lymphoma) that has resisted two or more
treatments or relapsed. Later, Yescarta™ is approved for adults with follicular lymphoma (FL) that resisted
treatment or relapsed after two other therapies
49
. Next, the FDA approved Kite’s second CAR T cell therapy
Tecartus™ (brexucabtagene autoleucel) to treat adults with mantle cell lymphoma that resisted treatment
or came back
50,51
. In February 2021, Juno Therapeutics’ (a Bristol-Myers Squibb Company) Breyanzi™
(lisocabtagene maraleucel) is approved for its use in adults with relapsed or refractory large B-cell
lymphoma
52,53
. More recently in March 2021, the FDA approved bluebird bio and Bristol-Myers Squibb’s
Abecma™ (idecabtagene vicleucel) as the first B-cell maturation antigen (BCMA)-directed CAR T cell
immunotherapy to treat adults with treatment-resistant or relapsed multiple myeloma after four or more
prior lines of therapy
54
.
1.3.2 Challenges of CAR T cell therapies
Although promising, the use of CAR T cells in cancer treatment have had limitations such as
achieving durable remissions, disease relapse after therapy, CAR T cell related toxicity and issues with
moving beyond hematologic malignancies (Figure 1.3)
55
.
CAR T cell failures in durable remissions may have several causes. In case of some patients, the
CAR T cells cannot be manufactured successfully or the CAR T cell product do not expand sufficiently
either during in vitro manufacturing or after in vivo administration
56–58
. A significant obstacle with CAR T
cell-based therapies is the need to isolate and use autologous cells
55
. Moreover, in some patients, CAR T
cells demonstrate limited persistence in vivo which may cause disease relapse
47,59,60
. One potential way to
improve durability is to attack multiple antigens simultaneously. However, in other patients, T cells have

7
been shown to persist for months to years after cell infusions
61
. T cell persistence after engineered-T cell
infusions has been demonstrated to cause chronic on-target-off-tumor effects (i.e. B-cell aplasia with the
aCD19-CAR T cells in clinical trials
62
). Thus, patients treated with adoptive CAR T cell transfer should be
monitored for the degree and side effects of T cell persistence. Furthermore, the production of CAR T cells
is technically complex, time intensive and thus, expensive, which are major considerations for the
widespread implementation of CAR T cell therapy
63
.

Figure 1.3 Limitations to durable remissions after CAR T cell therapy.
Limitations to durable remissions after CAR T cell therapy includes CAR T cell failures, antigen modulation, therapy
related toxicities and the unmet need for the clinical use of CAR T cell therapies beyond hematological malignancies.
Figure adapted from Shah, N.N., Fry, T.J. Mechanisms of resistance to CAR-T cell therapy. Nat Rev Clin
Oncol 16, 372–385 (2019).

Another mechanism which leads to resistance in CAR T cell therapy by antigen escape is antigen
modulation, specifically by the loss or downregulation of CD19 and/or CD22 on malignant B cells, which
can also be a problem in non-B cell malignancies, including solid tumors
64,65
.
Another concern for CAR Therapy is safety related, regarding the use of polyclonal T cells and
subsequent increase in the number of activated lymphocytes mediating tumor cell death. The possible
complication to polyclonal T cell use is the development of cytokine release syndrome (CRS), which is

8
briefly the rapid and massive release of proinflammatory cytokines, such as IFN-𝛾, TNF-𝛼, and IL-6
66
.
Other major side effects of CAR T cell therapy are B cell aplasia
67
, which is a mass die off of normal CD19-
expressing B cells, neurotoxicity (i.e. seizures and confusion) and cerebral edema
68
. B-cell aplasia can be
compensated by immunoglobulin therapy
55
. CRS can also be managed with standard supportive therapies
however the effect of these therapeutic interventions on the durability of CAR remission is unknow and
requires more data. Given the limitations and toxicity of CAR T cell therapy, researchers have started to
explore alternative approaches including using other cell populations in CAR-related therapies, such as
engineering NK cells which were documented to have roles in tumor immunosurveillance and tumor cell
killing
69–71
.
Finally, the most recent ongoing research efforts have been in optimization of the clinical use of
CAR T cell therapies in beyond hematological malignancies, such as lymphoma subtypes and other disease
contexts including central nervous system CNS diseases. Critically, CAR T cell therapy has limited efficacy
in patients with solid tumors
55
.
1.4 Natural Killer (NK) Cells
Natural killer (NK) cells are peripheral blood lymphocytes that mediate immune surveillance as a
response to virus infected malignant cells
72,73
. Different from T cells, NK cells bridge the innate and
adaptive immune systems (Figure 1.4). NK cells are characterized by the absence of CD3 and expression
of CD56 and CD16 surface antigens. NK cells do not express antigen-specific receptors (i.e. TCRs on T
cells). Instead, they receive signals by their germ-line–encoded receptors and generate effector functions.
They monitor autologous cells’ surfaces for aberrant expression of major histocompatibility complex class
I (MHC-I) molecules and cell stress markers, and protect the body from pathogen invasion and malignant
cell transformation. There have since been numerous studies which demonstrate that NK cells can
contribute to cancer immunosurveillance and target tumors by distinguishing malignant cells from
healthy
74
.

9

Figure 1.4 Various functions and characteristics of CAR.NK cells.
Chimeric antigen receptor engineered natural killer cells can have various origins. NK cell in vivo persistence can be
enhanced by interleukins. NK cell innate cancer surveillance abilities can be directed towards tumor associated
antigens by CAR-engineering of these cells. They perform cytotoxicity by releasing various perforins and granzymes.
CAR-NK cell therapies allow multiple infusions and can serve as off-the-shelf products which are ready to use for
allogeneic cell transfers. Figure adapted from Sivori S, Meazza R, Quintarelli C, Carlomagno S, Della Chiesa M,
Falco M, Moretta L, Locatelli F, Pende D. NK Cell-Based Immunotherapy for Hematological Malignancies. J Clin
Med. 2019 Oct 16;8(10):1702.

Different from T cells, NK cells give rapid responses to virus infections or malignant cells without
having been previously sensitized to them and without a need for HLA matching
75
. NK cells have roles in
early disease detection and cell killing which are mediated by several mechanisms including inflammatory
cytokine secretion [e.g., IFN-γ, TNF-α, IL-10], tumor cell apoptosis by receptor ligand binding (e.g., Fas

10
ligand and TRAIL)
76
, and release of cytoplasmic granule toxins (e.g., perforin, granzyme A, granzyme
B)
77,78
as a result of antibody-dependent cellular cytotoxicity (ADCC) (Figure 1.4)
79
. NK cells can control
tumor cells via direct exocytosis of cytotoxic granules or indirectly by the production of IFN-γ which
activates M1 and TH1 immune responses
80
.
NK cell activity and function are regulated by the integration of signals received from an array of
activating and inhibiting receptors. Recognition of virus infected or malignant cells occurs by means of
activating cell surface receptors such as NKG2D and natural cytotoxicity receptors (NKp30, NKp46, and
NKp44)
81
, or CD16 for the ADCC
82–84
. The activity of NK cells is dominated by their expressed inhibitory
receptor for molecules of MHC-I, such as KIRs, LAIR-1, CD161, CD328, and receptor complex
CD94/NKG2C
40,85
.
All NK cells individually express both inhibitory and activating receptors
85
. In the resting state,
signals from activating and inhibitory ligands are balanced. NK cells operate by detecting missing
information on their target which is known as the “missing self-hypothesis”. According to this phenomenon,
NK cell cytotoxicity inversely correlates with the target expression of MHC-I
73
. Binding of self MHC-I is
proposed as a major mechanism for the tolerance of NK cells to self-tissue
86
. After adaption to self MHC-
I environment, NK cells respond to ligands for activating receptors resulting in active elimination of
susceptible targets. While the presence of self MHC-I shows inhibitory response, the lack of constitutive
self MHC-I results in cytokine production, or perforin and granzymes release
40
. The presence of these
activating and inhibitory receptors suggests that NK cells perform constant surveillance of tissues for
normal levels of surface MHC class I molecules
87
. As opposed to healthy cells, virus infected or malignant
cells show reduced MHC-I expression profiles, and evade antitumor T cell recognition. Therefore, when
NK cells encounter cells lacking MHC-I molecules, the activating signals cannot be balanced by inhibitory
signaling, which leads to cytokine secretion and target cell killing
82
.
However, tumors develop mechanisms to escape from immune system to protect themselves from
NK cell attack. The soluble immunosuppressive molecules in the tumor microenvironment dampen NK cell
cytotoxicity by various mechanisms, i.e. reduction of NKG2D expression, which leads to decrease in

11
recognition of target cells. To overcome this inhibitory effect, new strategies have been developed. One
example to these strategies is the use of cell modifications such as vector transduction
88
.
1.4.1 Sources of NK cells for cancer therapy
NK cells are potent cytotoxic effector cells for cancer therapy. Researchers have made numerous
attempts to obtain sufficient levels of NK cells from patients’ blood, however technical challenges still
remain. The possible sources for isolating NK cells have been shown as bone marrow, peripheral blood
(PB) or umbilical cord blood (CB), however NK cell populations constitute only 10 % and 20 % of cells in
PB and CB, respectively. In addition, these patient isolated NK cells are often dysfunctional.
The source of cellular therapy with NK cells can be the patient (autologous) or a healthy donor
(allogeneic) peripheral blood (Figure 1.4). Autologous NK cells are mostly compromised by the effects of
cancer and its treatments. Also, the function of autologous NK cells becomes silenced when they encounter
self-MHC antigens. While autologous NK cells present these limitations, allogeneic NK cell infusions carry
the risk of graft-versus-host GvH reactions
89
. NK cells obtained from healthy donors, requires the depletion
of allogeneic T cells to restrain GvH reactions. Moreover, the invasive and repeated leukaphereses to collect
sufficient numbers of NK cells is a major inconvenience for donors and patients. Furthermore, the low
transfection efficiency of ex vivo expanded blood NK cells present another challenge in the use of blood-
derived NK cells for therapeutic purposes. In order to increase the number of NK cells for multiple infusion
in treatments, isolated NK cells have been cultured in the presence of cytokines and with feeder layers. As
an alternative NK cell source, hematopoietic stem cells (HSCs) have been differentiated into NK cells and
induced pluripotent stem cell (iPSC) technology has been utilized for obtaining high number of NK cells
90
.
One other issue about the clinical use of NK cells is the need for the improvement of NK cell
activation techniques for NK cell proliferation, persistence and function. Examples to these techniques are
NK cell priming and addition of cytokines. The recent research question that still demands an answer is
whether these stimulations should be done pre- or post-injection for the favorable outcome.

12
Given the disadvantages of obtaining and using primary NK cells for immune cell therapies,
researchers have generated stable cell lines from blood-NK cells and clonal NK cell lymphoma (Figure
1.4). NK cell lymphoma is a rare disease, and also the clonal growth of a cell line is also relatively rare.
The immortalized clonal NK cell lines easily proliferate and expand in culture. Among the established NK
cell lines
91–97
, NK-92 is the most prominent with its consistently and reproducibly shown high antitumor
cytotoxicity, as well as its ability to be genetically manipulated easily. NK-92 is established from a patient
with non-Hodgkin’s lymphoma and has displayed a robust and high cytotoxicity to cancer targets
93
. It is
important to note that NK-92 cells have completed phase I trials in cancer patients
98,99
in clinical trials
NCT00900809 and NCT00990717.
NK-92 cells are irradiated before infusion for safety. After irradiation with 1000 cGy, the killing
and cytokine production ability of NK-92 cells were maintained and their in vivo proliferation was
completely abrogated
100
. Data from phase I studies in different countries (US, Canada, and Germany) in
which patients have been treated with repeated infusions of irradiated NK-92 cells confirmed that infusions
are safe without any short or long term complications and unexpected side effects and regardless of the cell
numbers injected
98,99
. Despite treatment-resistant advanced cancer, clinically significant responses were
reported in patients with lung and kidney cancer, and other solid tumors.
1.4.2 CAR.NK cells
It is important to note that NK-92 cells do not express the FcγRIIIa receptor (CD16) and thus,
cannot mediate antibody-dependent cellular cytotoxicity (ADCC)
101
.
For autologous NK cell transfer, cells are isolated from the patients, activated, and expanded in
vitro in the presence of cytokines, followed by the reinfusion of cells back to the patient. For allogeneic
transfer, NK cells are obtained from HLA-matched or half-matched donors. Donor T cells are removed
before reinfusion in order to eliminate the risk of GvHD
102
.
An important aspect of using NK-92 cells for treatment is their high transfection efficiency in order
to express specific receptors targeting malignant cells. Another huge advantage of using NK-92 cell line

13
for treatment is its “off-the-shelf” availability. The lack of long term and labor-intensive culture systems,
as in the case of engineered T cells, would mean significantly lowered cost implications for NK-92 cell
therapies. With the optimized and streamlined techniques, NK cell immunotherapy would possibly be
adopted as a clinical therapy in the future. The preparation and administration costs of NK-92 cells are
significantly lower than autologous or allogeneic NK cells, as well as CAR T cells. Currently, treatment
with CAR T cells is highly patient specific and costly ($250,000 or more). In contrast, treatments with
engineered NK-92 cells are believed to cost on the order of $20,000, with the option of repeated infusions.
Among the advantages of NK cells are their limited lifespan of several weeks or months
73,103
and the lack
of memory cell formation - on contrary to CAR T cells -, which could present the possibility of multiple
administration of CAR NK cells to patients.
Furthermore, the donor-independent NK-92 cells can be frozen as a clinical grade stock and shipped
to the treatment site. The NK-92 cell line based off-the-shelf CAR technology allows to generate large
numbers of CAR.NK cells which are specific to the TAAs (Figure 1.4). With the expression of a CAR, NK
cells get to the site of tumor and target the malignant cells among the healthy ones
101
.
CAR T cells have been widely studied, however more discoveries are awaiting CAR-engineered-
NK cells as alternative effector cells. For example, although the effects of CD28 and 4-1BB on cytotoxicity
and cytokine production of NK cells have been shown to differ from T cells, both domains are used in CAR
constructs for NK cells
104–111
. CAR-engineered allogeneic NK cells have gained significant attention due to
their ‘‘off the shelf’’ potential
112
. For CAR engineering, both primary NK cells and NK cell lines have been
investigated.
Both primary NK cells and NK cell lines (i.e. NK-92) can be utilized as allogeneic “off-the-shelf”
products. Allogeneic NK cells are a promising option for ACT. These cells do not express antigen-specific
receptors and thus do not carry the risk of inducing GvHD
113
. In addition, researchers have demonstrated
that allogeneic NK cells eliminate alloreactive T cells in the transplant setting, indicating that the risk of
rejection is lower for CAR-engineered NK cells
114
. Moreover, CAR-modified NK cells could possibly
remain active against tumor antigens even if the targeted tumor associated antigen is down-regulated on

14
tumor cells (i.e. during tumor progression), because of the ability of NK cells to recognize tumor cells by
their native receptors and initiate lytic activity
115
.
1.4.3 CAR engineering of primary NK cells
Primary NK cells do not require irradiation before patient administration and therefore they can be
expanded in vivo, which was demonstrated to correlate with increased effectiveness in an acute myeloid
leukemia clinical trial
73
. CAR-engineered primary NK cells are advantageous over CAR T cells in term of
their ability to kill their targets via CAR-specific mechanism as well as by triggering target cell death
independent of the tumor antigens. Thus, even target antigens are down-regulated on tumor cells as an
attempt to evade immune detection, NK cells are still capable of staying effective, as opposed to hindered
cell function of CAR T cells. Moreover, primary human NK cells produce cytokines, such as IL-3 which
are different from the proinflammatory cytokines produced by T cells. Therefore, NK cells are not
associated with the onset of cytokine release syndrome
112,116
. Despite these advantages, primary human NK
cells are difficult to isolate and expand due to the cell source related limitations and donor dependent
variable yield. For this reason, NK cell lines have gained significant interest for CAR NK cell therapy.
1.4.4 CAR engineering of NK cell lines
Compared to primary NK cells, NK cell lines present a more homogenous and well-defined
population. NK cell line-based immunotherapy does not require isolation from donors. However, before
patient administration, NK cell lines must be irradiated.
In addition to the success of parental NK-92 cells in eliminating AML
117
, myeloma
118
, and
melanoma
119
, preclinical studies in SCID mice with CAR-modified NK-92 have been demonstrated to
eliminate different malignancies such as lymphoma (CD19.CAR
120
), prostate cancer (EpCAM.CAR
121
),
breast cancer (Her2.CAR
122
), neuroblastoma (GD2.CAR
123
) and glioblastoma (EGFR.CAR
124
). Most of the
studies for CAR.NK-92 cells have used first-generation CARs against TAAs such as CD19 and CD20 for

15
B-cell lymphoma
120,125,126
, ErbB2 for breast, ovarian and squamous cell carcinoma
122,127,128
, GD2 for
neuroblastoma
129
and CD138 for multiple myeloma
130
.
The second-generation CAR NK cells from the NK-92 cell line have used CD28 or 4-1BB
intracellular signaling domains and both constructs have been shown to improve killing against breast
cancer, compared to their first-generation counterparts
122
. When CD3ζ was used in combination with 4-
1BB, DAP10, or 2B4, strong efficacy was observed based on upregulation of the PI3K/AKT pathway. On
the other hand, CD3ζ constructs that were used with CD28 resulted in decreased cytotoxicity. Other
antigens targeted with second-generation CAR NK cells include EpCAM for multiple carcinomas
121
, HLA-
A2 EBNA3 complex for Epstein-Barr virus
131
, CS1 for multiple myeloma
130
, and ErbB2 for HER2-positive
epithelial cancers
122,128
. Moreover, high efficacy levels were revealed by third-generation CARs (CD28/4-
1BB/CD3ζ)
104,132
. The research in CAR T cells suggests that CAR constructs containing 4-1BB can be
superior
133
, but such an investigation has not been attempted for CAR NK cells yet. Also, a third-generation
NK-92 CAR comprised of anti-CD5 scFv with CD3ζ, CD28, and 4-1BB intracellular signaling domains
showed specific antitumor activity against various T cell leukemia and lymphoma cell lines and primary
tumor cells. Also, these anti-CD5.CAR.NK cells were able to inhibit disease progression in xenograft
mouse models of T cell acute lymphoblastic leukemia (ALL) cell lines and other primary tumor cells
134
.
1.4.5 CAR NK cells in clinical trials
As opposed to the high number of clinical trials using CAR T cells for cancer treatment, CAR NK
cell clinical studies are relatively few. In the first CAR NK cell clinical study, NK cells redirected against
CD19 was utilized. These antiCD19-4-1BB-CD3ζ-CAR NK cells were administered to patients with B-
ALL (NCT 00995137; NCT 01974479) but results have not been published yet. In the second trial IL-2-
activated haploidentical NK cells were expanded in culture before administering to pediatric and adult
patients. In a third study (NCT 03056339), NK cells derived from umbilical cord blood (CB) were
genetically modified to express antiCD19-CD28-CD3ζ-CARs, with a iCasp9 safety switch and IL-15
expression. These cells were administered to patients suffering from relapsed and/or refractory B-cell

16
lymphoma or leukemia. Although in their early clinical trials, CAR-expressing NK cells has the potential
to offer enhanced effector cell function with increased specificity. In the case of CAR T cell therapy, the
important concerns include GvHD, on target/off tumor effects, and tumor lysis syndrome. However,
allogeneic CAR.NK cells are expected to induce antitumor effects and dissipate after a few days, which
makes them favorable for ACT
135
.
1.4.6 NK cell combination therapy
It is clear that the NK-92 platform is proven to be effective even in its unmodified, monotherapy
form. By various combination therapy approaches, NK cell therapies could be used to further augment
immune responses. Additional improvements could be achieved through genetic modifications, i.e. CAR
modification, on NK-92 parental cells, or conjugation of NK cells with other agents including checkpoint
inhibitors, immunomodulatory drugs, and chemotherapy drugs. Given the universal administration and off-
the-shelf therapy advantage of NK-92 platform, it can also provide the possibility of personalized treatment
regimens to target specific patient needs with combinatorial therapy approaches.
1.5 CAR-engineered T and NK Cell Clinical Trials with Solid Tumors
To date, CAR T cell based therapies have shown great promise in eradicating hematological
malignancies, such as using CD19 CARs in lymphomas and leukemias
27–31
. CAR therapy has an emerging
role for its use against solid tumors. As the pharmaceutical industry has become more involved in CAR-
engineered T cell and NK cell therapies, the number of clinical trials that test these cells has expanded
significantly. As of May 2021, according to the data in clinicaltrials.gov, the number of clinical trials with
CAR-engineered T cells worldwide is 1373 and with CAR-engineered NK cells is 34. It is important to
mention that no CAR T cell Phase III trials has returned results yet. China and the United States are the
most active countries in the field of clinical research for CAR T cells (38% and 37% of the clinical trials in
the USA and China, respectively) and CAR NK cells (24% and 62% of the clinical trials in the USA and
China, respectively). Out of these clinical trials, only 55 of them aim to treat solid cancer malignancies with

17
CAR-engineered T cells and 3 with CAR-engineered NK cells. Early reports from CAR T cell trials with
solid tumors have not reported the same success that’s been seen with hematologic cancers. Thus far, CAR
T cells have successfully treated few solid tumors, such as mesothelin overexpressing pancreatic and lung
cancers, and epidermal growth factor receptor variant III (EGFRvIII)-expressing glioblastoma
136
. In these
examples, the expression of target antigens on tumor cells were dramatically high compared to normal cells.
1.6 Challenges of CAR-engineered Immune Cells Therapies in Solid Tumors
The clinical studies with CAR T cell therapy for solid tumors has been less promising due to non-
optimal antitumor efficacy and non-durable clinical response
137,138
. There are various hurdles that
researchers need to overcome for helping CAR T cells surmount in the solid tumor microenvironment and
extrapolating CAR T cells’ success in treating hematological malignancies to solid tumors. The major
roadblocks in CAR T cell therapy against solid tumors are tumor antigen heterogeneity, trafficking and
infiltration into tumor tissue, and immunosuppressive tumor microenvironment
34
.
1.6.1 Tumor antigen heterogeneity
One of the impediments to the validity of CAR-engineered immune cell therapy against solid
tumors is antigen heterogeneity. The differential expression of tumor-associated antigens (TAAs), which
are the most useful targets for CAR engineering, on different tumor cells, impairs the detection of target
antigen expressing cancer cells by CAR T cells and decreases the efficacy of CAR T cell therapy. Moreover,
the level of antigen expression at different tumor sites varies and combined with the cancer cell antigen
diversity, CAR T cell function is impaired
139
. Among the methods that have been used to target multiple
TAAs by CAR T cells, the most common are co-expression of several CARs on one T cell and expression
of a CAR including two or more antigen recognition domains
140
. As another approach, researchers have
investigated targeting cancer stems cells, instead of cancer cells, to eliminate tumor heterogeneity. For
instance, a tumor stem cell marker CD133 is overexpressed in many solid tumors and is considered a target

18
tumor marker for CAR T cells
141
. The membrane protein CD19 is widely and homogenously expressed on
almost all B-cell hematologic cancers. This is one of the important reasons of the outstanding success of
CD19-targeted CAR T cell therapy
142
. However, the success of CAR T cell therapy is limited in solid tumors
due to the tumor antigen heterogeneity and limited knowledge on tumor-specific antigens
143
.
1.6.2 Trafficking and infiltration into tumor tissue
Even for the case of CAR T cells where the target antigen for a solid tumor is identified, cells first
need to reach the tumor site to exploit their functions
34,144
. In hematological cancers, CAR T cells start to
circulate in the bloodstream upon their adoptive transfer, which indeed is their destination and site of action.
In solid tumors, however, there are multiple barriers that a CAR T cell must overcome to reach the tumor
site. It is necessary for migrating lymphocytes to follow a chemotactic gradient, or bind to a surface
expressed molecules such as selectins on endothelial cells that bind the circulating lymphocytes. Upon
binding, signaling cascades would be induced for the following extravasation into sites of inflammation.
As another barrier to T cell entry, abnormal vasculature at the tumor site can block migration of T cells into
tumor tissues
145
.
Researchers have demonstarted the importance of chemokines in lymphocyte migration and
homing, and hence studied various designs to deliver chemokines intratumorally to attract TILs. In one
study, the design of mesothelin targeting CAR T cells that constitutively expressed the cytokine IL-7 and
the chemokine CCL19 demonstrated complete tumor regression and prolonged survival in a solid tumor
mouse model suggesting the effect of IL-7 and CCL19 on recruiting endogenous antitumor TILs
146
. In
another study, mesothelin CAR/CCR2 T cells that express both the CAR and chemokine receptors showed
12-fold increased homing and tumor regression in subcutaneous human MPM tumors in vivo
147
.
Another promising way to augment CAR T cell infiltration into tumor sites is the development of
FAP (fibroblast activation protein) targeting CARs. FAPs are expressed on multiple types of stromal cells
which are associated with most epithelial tumors
148
. In one study, FAP CAR T cells efficiently killed FAP-

19
positive human MPM tumor cells in vitro, and inhibited tumor growth and elongated the survival of mice
in a xenograft model with intraperitoneal (IP) tumors
149
.
One other intuitive method of bypassing the difficulties of suboptimal T cell homing is
regional/local CAR T cell administration. This method has already been tested in xenograft mouse models
and patients with solid tumors, resulting in varying degrees of success. For instance, in a xenograft mouse
model of human breast cancer metastatic to the brain, HER2-CAR T cells were administered via intracranial
and intra-tumoral routes and yielded improved antitumor activity with complete response and 100%
survival even after tumor re-challenge, compared with intravenous delivery
150
.
In addition, the physical barriers in the TME have multiple effects on the biological behaviors of
tumors and interfere with infused CAR T cell activity. The extracellular matrix (ECM) in stroma-rich solid
tumors, including proteoglycans and glycopeptidases effectively inhibits the infiltration and accumulation
of T cells. Studies have shown that the nonstructural matrix proteins in the TME, such as heparan sulfate
proteoglycans (HSPGs) have roles in maintenance of proliferation and migration of tumor cells
151,152
.
HSPGs in stroma- rich solid tumors present a physical barrier for T cells and the low penetration and
aggregation of T cells in tumor areas result in below optimal antitumor activity
153
. Strategies to improve
the capacity of CAR T cells to degrade the components of the ECM while maintaining T cell effector
functions would improve their antitumor efficacy.
1.6.3 Immunosuppressive tumor microenvironment
The battle of CAR T cells is far from over when they find their way into the tumor. The tumor
microenvironment (TME) has been reported to comprise a variety of cell types and characterized to be
hostile for T cells (Figure 1.5)
145,154,155
.
The intratumoral microenvironment is associated with hypoxia (poor oxygen levels) and lack of
nutrients
156–158
, due to high activity of constitutively active of growth-promoting pathways and their
unsaturated fatty acid catabolism
159
. This causes elevated lactic acid and carbonic acid generation, and
therefore an acidic microenvironment
160
. Additionally, cancer cells have been demonstrated to prefer

20
glycolytic metabolism which help maintain the hypoxic and acidic TME that is prone to oxidative
stress
137,154
. There exist various strategies to protect T cells from the hostile effects of the TME. One
approach is to protect them from the oxidative stress caused by ROS, by a CAR design which allows the
co-expression of catalase, an enzyme which converts hydrogen peroxide to water and oxygen. As a results,
authors reported that CAR T cells demonstrated a reduced oxidative state and their proliferation and
cytotoxicity profile improved in vitro
161
.

Figure 1.5 Immunosuppressive tumor microenvironment.
Solid tumors are characterized to be immunosuppressive due to various cells and soluble factors, as well as the
physically hostile microenvironment. Figure adapted from Zhang E, Gu J, Xu H. Prospects for chimeric antigen
receptor-modified T cell therapy for solid tumors. Mol Cancer. 2018 Jan 12;17(1):7.

In order to protect against tissue damage from an overexuberant inflammatory response, the
immune system has evolved a broad array of regulatory mechanisms including inhibitory receptors and
their related signaling networks, known as “immune checkpoint pathways”. These upregulated inhibitory
receptors on lymphocytes provide a negative feedback mechanism that is crucial for immunoregulation
162
.

21
Although these checkpoint pathways are critical in modulating excessive inflammation, they provide
tumors with a means of immune evasion. Several clinical trials have confirmed that blockade of CTLA-4
and PD-1 receptor-mediated immune checkpoint pathways led to unprecedented responses in multiple solid
tumor types and long-term remissions
163,164
. Activated T cells also express other immune checkpoints which
rapidly neutralize the antitumor activities, such as LAG3, TIM3, and VISTA
139
. One other targetable
checkpoint pathway that is active in the tumor microenvironment is adenosine signaling via the A2a
receptor
5,165–168
.
Moreover, in the TME immune suppression is mainly mediated by immunosuppressive immune
cells, such as myeloid-derived suppressor cells (MDSCs), tumor associated macrophages (TAMs), tumor
associated neutrophils (TANs), mast cells, and regulatory T cells (Tregs)
145
, and stromal cells like cancer
associated fibroblasts (CAFs). Immunosuppressive cells also release cytokines, such as transforming
growth factor-β (TGF-β) and interleukin (IL) 10, inside solid tumors significantly dampen the cytotoxic
function and efficacy of infused CAR T cells
169
.
Another concern in the TME is characterized as T cell exhaustion
170
which is caused by chronic
antigen exposure and which leads to the loss of effector and memory phenotypes, decreased production of
cytokines such as IFN-γ, TNF-α, and IL-2, and upregulated expression of inhibitory receptors that further
shut down T cell effector functions upon binding to inhibitory ligands or soluble factors in the TME
154
.
Tumor cells as well as these other cells secrete soluble factors like vascular endothelial growth
factor (VEGF) and TGF-β, which contribute to abnormal tumor vasculature
171
and also produce reactive
oxygen species (ROS) and molecules like adenosine, prostaglandin E2 (PGE2) and soluble Fas, which
contribute to the suppression of the T cell immune response
172,173
. In addition, other immunosuppressive
factors, such as CD47, which is overexpressed on cancer cells, allows tumor cells to bypass immune
surveillance mostly through CD47-SIRPα-mediated antiphagocytic signaling
174,175
. Similarly, indoleamine
2,3-dioxygenase (IDO) is an immunosuppressive factor which creates a favorable environment for tumor
cells by the degradation of tryptophan which increases Treg-mediated immunosuppression and suppress T
cell cytotoxicity
176,177
. As another important example, the elevated cellular activity in the TME causes an

22
increase in the concentration of extracellular adenosine, which is generated from extracellular AMP by the
activity of ectoenzyme CD73
178
. High adenosine levels result in a decrease in adoptively transferred CAR
T cell effector functions and thus lower their antitumor efficacy in solid tumors
166,179
.
1.7 Special Topics 1
In this subsection, you will find more in-depth information about the projects presented in this
dissertation. The subsections related to platelets will provide a literature review of platelets and their
immune roles in order to provide the rationale for the project in Chapter 4 where we utilized CAR T cell-
platelet complexes with the goal to enhance CAR T cell infiltration into solid tumors. Similarly, subsections
related to adenosine in the TME, adenosine receptors, SCH-58261 and nanoparticles constitute the
groundwork for the project presented in Chapter 5 in which we conjugate drug loaded liposomes onto CAR-
engineered NK cells as an active carrier of an immunomodulatory drug to the TME.
1.7.1 Platelets
Platelets (PLTs) are anucleate cells which participate in blood clotting, known since late 19th
century
180
. For the first time in 1882, Italian pathologist Giulio Bizzozero discovered the platelets and
described their hemostatic function and demonstrated their adhesion, aggregation and subsequent fibrin
formation
181
. For a long time, platelet research revolved around the traditional function of platelets, i)
hemostasis, which is the process of ending bleeding on an injured blood vessel, and ii) thrombosis, which
is formation of blood clotting within blood vessels
180
.
In blood, platelets circulate in a quiescent state. Vascular injury causes multiple subsequent platelet
receptor-ligand interactions and mediate stable adhesion of platelets to the endothelial cells. Following the
initiation of a cascade of intracellular responses that results in amplification of activation signals, such as
by the release of platelet agonist thrombin, platelets get activated. Next, they degranulate and upregulate
the expression of surface receptors that leads to further platelet aggregation and platelet recruitment to the
sites of tissue damage or infection
182
.

23
PLTs sense their environment and respond to physiological changes by using their receptors. While
many receptors on PLT surface support hemostasis, many others assist the interaction with immune cells
and aid in immune function. Several receptors that participate in immune function are constitutively
expressed on the surface of PLTs, whereas others are stored in PLT granules and surface expressed upon
PLT activation
180
.
Platelets are the smallest blood cells around 2-5 µm in diameter
183
. They are anucleate
hematopoietic cells and are produced in the bone marrow through the budding process of megakaryocytes
and released into the bloodstream
183
. Their average life span in circulation is ~7–10 days before being
cleared by resident phagocytes in the liver and spleen
184
.
The number of circulating platelets is around 150,000-400,000 cells per µL in healthy humans
185
.
In peripheral blood, platelets constitute the second most abundant cell population after red blood cells and
they exceed monocytes in quantity by almost 100-fold
182,183
.
Although PLTs are anucleate cells, they contain pre-stored proteins on their membranes and in their
granules. In addition, they still hold the ability to synthesize proteins. They contain messenger RNAs
(mRNAs) and are able to synthesize proteins upon activation. For example, activated PLTs were shown to
synthesize interleukin-1β, which promotes inflammatory responses in leukocytes and endothelial
cells
186,187
.
1.7.1.1 Freezing platelets
Traditionally, human PLTs have been preserved in liquid at 22°C up to 5 days. We received the
human PLTs from the vendors in the PLT-rich plasma. With the intention to store human PLTs for longer
durations with similar life span values to liquid preserved PLTs, we followed a previously published
protocol
188
. The lifespan of these cryopreserved human PLTs is 7 days. It was demonstrated in this previous
work that the allogeneic human PLTs collected from stable thrombocytopenic patients, when frozen with
this method, could be stored at −80°C for at least 2 years with ~75% recovery values in vitro and ~40%
recovery after transfusions
188
.

24
It is important to note that cryopreserved PLTs have increased hemostatic functions and reduce
blood loss easier, when compared to fresh, liquid-stored PLTs
189
. This improved in vivo effect is attributed
to altered surface receptor phenotype and release of procoagulant mediators
189–191
. Moreover, PLT
microparticle counts isolated from the supernatant of cryopreserved PLTs were founds to be 15-fold higher
than in fresh PLTs
192
. Notably, these PLT microparticles formed by cryopreservation are functional and
express the PLT and microparticle markers
192
.
Recently, much interest has been devoted to cellular microparticles which shed from the cell
membrane of cells undergoing apoptosis or being subjected to various stimuli or stress
180
.
1.7.2 Platelet-derived microparticles
Even though the PLTs are anucleate cells and are incapable of replicating themselves, they
effectively respond to external stimuli. For this, they utilize the pre-synthesized proteins within the PLT
granules. These proteins can be secreted extracellularly or expressed on the surface of PLTs following their
activation. Surprisingly, one proteomic study indicated the presence of 827 proteins in PLT granules
193
.
Through the secretion and surface expression of these proteins, PLTs interact with various cell types,
including immune cells, and hence play vital roles in various physiological functions.
Among the factors that generate platelet-derived microparticles (PMPs), the most known ones are
platelet activation, high shear forces and apoptosis
194
. Upon platelet activation, intracellular calcium
concentration increases which activates calpain, cytoskeleton proteins and Rho kinase, resulting in
membrane blebbing
195
. Different from this, resting platelets can release PMPs by αIIβ3 integrin-mediated
destabilization of the actin cytoskeleton
196
.
Blood plasma contains other plasma microparticles. Though other cell types including endothelial
cells, leukocytes, or cancer cells release microparticles, majority of them are derived from platelets
197
. The
physiological roles of PMPs have not received attention until recent years. These heterogeneous group of
vesicles initially thought to be bystander components in the circulation, yet decades of research identified
that they differ in composition and mediate various roles from hemostasis to cellular signaling. For example,

25
PMPs promote the expression of various adhesion molecules
198
and regulate the release of immunologically
important cytokines IL-17 and IFN-γ
199
. The tumor-promoting capacity of PMPs was also reported and this
capacity was associated with their VEGF, PDGF and TGF-β release.
Microparticles are characterized by their surface antigens. One surface molecule expressed on both
PLTs and PMPs is integrin αIIbβ3 (Glycoprotein GPIIb/IIIa), which is also named as CD41
200
.
1.7.2.1 Types of platelet microparticles
Mature platelets carry three types of cytoplasmic storage compartments: i) alpha (α) granules, ii)
dense (δ) granules, and iii) lysosomal (λ) granules. All of them possess a vast array of bioactive molecules
that can be translocated to the platelet surface or released into the blood circulation upon platelet
activation
180
.
α-granules are the most abundant granule type with granule numbers of 50-80 per PLT and contains
hundreds of different proteins. Their size is between 200–500 nm in diameter
201
. Alpha granules contain
adhesive molecules, such as fibronectin, fibrinogen, GPIb and integrin αIIbß3, coagulation factors, growth
factors, cytokines and chemokines. They are responsible for the hemostasis and thrombosis function of
PLTs. They contain proteins like fibrinogen, von Willebrand factor (VWF), vitronectin, fibronectin,
thrombospondin, factor V, factor VIII, and cell adhesion molecules like integrins αIIbβ3 (GPIIb/IIIa) and
αvβ3, and mediate platelet adhesion, aggregation and coagulation. α-granules also contain proteins and
peptides that recruit, localize, and activate immune cells, and modulate inflammatory or immune function
as a result. For example, by the chemokines secreted from α-granules, PLTs can recruit and activate
neutrophils (chemokines: PF4, CXCL4 and CXCL-5)
202,203
and monocytes (chemokines: CCL3 CCL5
(RANTES) and CXCL)
204,205
. It should be noted that P-selectin (CD62) on activated platelets can bind to
its ligand P-selectin glycoprotein ligand-1 (PSGL-1) expressed on neutrophils and monocytes and activate
these leukocytes
206,207
. Moreover, α-granules influence innate immunity with the anti-bacterial and anti-
fungal actions of their proteins
208
.

26
Dense granules appear as dense bodies under electron microscope due to high calcium and
phosphate within their structure. They are ten-fold less abundant than α-granules and approximately 150
nm in diameter
209
. The contents of dense granules are released into the extracellular milieu upon activation
of platelets. Dense granules store nucleotides ATP and ADP, bioactive amines (histamine and serotonin)
and ions (calcium and phosphate)
210
. They modulate inflammatory pathways by activating dendritic cells,
and provide a feedback mechanism which activates platelets, respectively
211
. ADP, ATP and serotonin are
also known as platelet agonists.
Lysosomal granules are around 200–250 nm in diameter
212
. They contain lysosomal enzymes, like
acid phosphatase or arylsulfatases
212
, proteases like carboxylpeptidases, and lysosomal-associated
membrane proteins. LAMP1, LAMP2 and LAMP-3 (CD63)
211
.
1.7.3 Different roles of platelets
In recent years, increasing evidence suggested that platelets participate in inflammatory and
immune responses. In addition, platelets were shown to be critically involved in tumor metastasis,
angiogenesis, and tissue regeneration. Platelets also get activated in case of traumas or pathologies like
cancer.
1.7.3.1 Inflammatory functions of platelets
Platelets also have vital roles in inflammation. Important surface receptors, such as P-selectin,
integrins and FcγRIIa increase upon platelet activation, facilitating interactions between activated platelets
and leukocytes
213,214
.
1.7.3.2 Immune functions of platelets
The immunoregulatory roles of platelets (PLTs) have been recently recognized
182,215
. Upon
activation, PLTs express activation molecules on their membrane, such as P-selectin (CD62P) or CD40L,
and release PLT-derived microparticles (PMPs), and various cytokines, chemokines, and growth factors
216
.

27
Research indicates that T cell activation increases platelet aggregation by mechanisms in which
both T-cytolytic and T-helper cells are involved
217
. Throughput the studies in which platelet-leukocyte
interactions have been studied in thrombosis, researchers blocked certain receptors that are known to
mediate these interactions by competitive antibodies. They reported that P-selectin, integrin aIIbb3,and
CD40 are the main mediators of platelet-leukocyte interactions
218
. Platelets interact with immune cells by
two major ways; direct interactions via adhesive mechanisms (contact-dependent mechanism) or indirect
interactions via soluble factors (independent mechanism). Currently, little is known about the effect of
platelets on immune cells functions and it requires further research.
1.7.3.3 Adhesive mechanisms of platelet-leukocyte interactions
Several important adhesive molecules known to participate in platelet-leukocyte interactions are
the following: the interaction between P-selectin on platelets to P-selectin glycoprotein ligand-1 (PSGL-1)
on leukocytes, which mediates neutrophil rolling on platelets and endothelial cells in the
microcirculation
219
, as well as leukocyte capture and rolling
180
, ii) platelet glycoprotein Ibα (GPIbα; CD42b)
located on platelets and Mac-1 (CD11b/CD18; α Mβ 2 integrin) on leukocytes to promote microvascular
inflammation and thrombosis
220
, and iii) CD40L (CD154; gp39) on platelets and CD40 on leukocytes.
CD40L has shown to be present both on activated T cells
221
and in resting platelets. Upon PLT activation,
CD40L is translocated to the membrane, cleaved and released in its soluble form, sCD40L, in a CD40
dependent manner
222,223
. Besides, CD40 is expressed on a variety of cells, including B cells, endothelial
cells
224
, neutrophils
225
and dendritic cells (DCs)
226
. CD40 has also been described in numerous cancer types,
including ovarian, colon, lung and breast carcinomas and in melanoma
227,228
. Through the activation of DCs,
platelets also contribute to adaptive immunity. DC-PLT interactions occur through the CD40-CD40L axis.
Upon their activation, DCs present antigens to T cells
182
.

28
1.7.3.4 Soluble mechanisms of platelet-leukocyte interactions
In addition to the adhesive mechanisms of platelet-leukocyte interactions, platelets also talk to
leukocytes via soluble mechanisms. By releasing soluble factors, platelets activate, modulate, or assist with
capture of circulating leukocytes. Hundreds of biologically active molecules in platelet granules are capable
of various effects. Soluble platelet released factors also have immune modulatory roles, such as chemokine
PF4 (platelet factor 4; CXCL4) inhibiting Treg proliferation
229
, promoting Treg differentiation and IL-10
production, and limiting Th17 cell expansion and differentiation
230
. PF4 also activates neutrophils
229
. In
addition, CD4
+
T lymphocytes bound with PLTs demonstrating decreased IFN-𝛾 and IL-17 production and
lowered proliferation of CD4
+

cells by the effect of TGF-β, compared to the condition lack of PLT
binding
231–233
. Soluble CD40L (sCD40L) released by PLTs can reduce the inflammatory cytokines
produced by CD4
+
T cells
234,235
. Moreover, both PLTs and PMPs can bind to lymphocytes using several
molecules
199,217,232,236
. For instance, CD62P on PLTs interacts with P-selectin glycoprotein ligand 1 (PSGL-
1) on lymphocytes for this binding
217,232,233
. Furthermore, PMPs released upon the activation of PLTs
decrease IFN-𝛾 and IL-17 production by IL-17
+
Tregs in a P-selectin-dependent manner rather than by
soluble factors
199
. Also, it has been demonstrated that resting PLTs suppressed cytotoxicity of mouse CD8
+
T cells
237
.
Platelets also store chemokines within their α-granules, and release these mediators upon platelet
activation
238,239
. Chemokines such as β-thromboglobulin recruit neutrophils and suppress neutrophil
apoptosis
202
, CXCL-5 modulates neutrophil chemotaxis
203
. Secretion of molecules such as serotonin and
RANTES mediate T cell activation and differentiation
240,241
. For example, serotonin force naive T cells to
activate and proliferate
240
. Similarly, the release of PF4 (platelet factor 4; CXCL4) impacts T cell function
by limiting Th17 cell expansion and differentiation
230
. It is obvious that activated platelets contribute to the
immune function by releasing soluble factors and chemokines that recruit, localize, or activate immune
cells.

29
It has been indicated that T cell activation increases platelet aggregation by mechanisms in which
both T-cytolytic and T-helper cells are involved
217
. Several recent studies have shown that platelets bind to
CD4 T cells via multiple receptors, including integrins (αIIbβ3), P-selectin (CD62P), CD40L, and
lymphocyte CD11b
217
. It has been also suggested that platelets can facilitate the recruitment of lymphocytes
at a site of inflammation or infection, which is known as a central step in T cell trafficking.
Direct interactions between PLTs and T cells in the context of cancer have raised questions and
needs to be explored more. Further research is necessary to understand the complex role of PLT-leukocyte
interactions in the development of cancer-associated thrombosis and facilitation of cancer metastasis. It is
important to note that PLT-lymphocyte interactions would benefit from further research in cancer
immunotherapy framework
233
.
1.7.4 Role of platelets in cancer
Platelet involvement in cancer growth and metastasis is a widely-known and longstanding concept.
It has been shown in various articles that both PLTs and PMPs infiltrate into the tumor microenvironment
to directly interact with cancer cells
242
.
In recent years, experimental and clinical studies have evidenced that platelets play several roles in
the progression of cancer and in cancer-associated thrombosis which is correlated with shortened survival
and poor prognosis
243–245
in various types of cancer. Platelets and platelet released factors play several roles
in cancer progression including immune tumor evasion, tumor growth and metastasis
246
. Given the short
life span of circulating platelets, healthy adults must produce approximately 100 x 10
9
platelets daily to
maintain a platelet count within normal range
242
. Cancer can influence the platelet count and activation
state. Several clinical studies have shown elevated platelet count or thrombocytosis (> 400 × 10
9
/L) in
cancer patients. Correlations between high platelet counts and shorter disease-specific survival have been
described for lung, colon, breast, pancreatic, kidney, and gynecologic cancers
242
. In addition, different
studies have demonstrated that the plasma levels of soluble P-selectin (sP-sel), soluble CD40 ligand
(sCD40L), thrombospondin-1 (TSP1) and ß-thromboglobulin, which are released upon activation of

30
platelets, were significantly higher in cancer patients than in healthy controls, suggesting platelet activation
during cancer progression
234,247–249
. Also, platelet counts have effects on disease burden and treatment
efficacy in cancer patients. In addition, platelets have been shown to protect cancer cells against
chemotherapy-induced apoptosis. For example, in vitro experiments showed that platelets help colon and
ovarian cancer cell lines increase resistance to 5-fluorouracil and paclitaxel treatments
250,251
. Also, high
platelet counts were indicated to be associated with a poor response to chemotherapy in cancer patients and
mouse models of cancer
252,253
.
Platelets are shown to induce a transitory epithelial-mesenchymal transition (EMT) by physically
contacting with blood-borne cancer cells, facilitating extravasation and initiation of metastases
145
.The
interactions between platelets and various cell types in the TME, such as epithelial cells, endothelial cells
and fibroblasts were also shown to contribute to cancer cell metastasis and angiogenesis in tumors
254
.
The abilities of tumor cells to aggregate and activate platelets give them many advantages in the
bloodstream. Recent studies have demonstrated that cancer can educate platelets (tumor-educated platelets).
Cancer cells can activate the platelets and the coagulation system by direct PLT interactions or indirectly
by PMPs and secreted factors/cytokines. Activated PLTs and PMPs adhere to tumor cells by integrins and
selectins
255–257
. They facilitate cancer progression and metastasis by different ways. For example, i) they
form aggregates with tumor cells; ii) induce tumor growth, epithelial-mesenchymal transition, and invasion;
iii) shield circulating tumor cells (CTCs) from immune surveillance and killing; iv) facilitate tethering and
arrest of CTCs; and v) promote angiogenesis and tumor cell establishment at distant sites
258
.
In the case of tumor shielding effect of platelets, platelets have been shown to protect cancer cells
by creating a shield to block immune cell infiltration into tumors. The literature on platelet-NK cell
relationship is the most studied in regards to platelet-leukocyte interactions. Platelets shield tumor cells
from NK cell lysis. NK cells are known as the primary killers of metastasizing tumor cells
259,260
. Platelets
interfere with NK cell binding to tumor cells by the formation of aggregates on tumor cell surface
261
. They
both sterically hinder their binding to tumor cells and inhibit NK cell cytolytic function
262,263
. For the
inhibition of NK cytotoxicity, they release soluble factors (i.e. TGF-β) that downregulates NK cell

31
immunoreceptors and eventually inhibit NK cell functions such as IFN-γ production, cytotoxicity, and
granule mobilization
264
.
In addition, PMPs promote angiogenesis
265
by stimulating the formation of capillary tube and
network formation
266,267
. According to a clinical study, PMPs collected from cancer patients express tissue
factor which causes the further activation of PLTs and stimulation of the angiogenic cascade which in turn
support metastatic tumor growth
268
. Furthermore, platelets help tumors maintain the integrity of tumor
vasculature and participate in cancer metastasis. It is shown that high platelet counts correlate with
metastasis
269
and poor treatment outcomes in multiple types of cancer
244,253,270–273
.
Platelet-tumor cell aggregates form through the binding of platelet integrin α IIbβ 3 to tumor cell
integrin α vβ 3 via RGD-containing proteins including fibrinogen, von Willebrand factor, and fibronectin
256
.
Once activated, platelets can then bind to tumor cells via P-selectin
274–276
and glycoproteins
277,278
and
directly induce tumor growth and metastasis by the release of pro-tumor angiogenic and growth factors. In
one study, platelets were genetically modified to express the tumor necrosis factor related apoptosis-
inducing ligand (TRAIL) and they were shown to induce the apoptosis of cancer cells in vitro and
significantly reduce metastasis in vivo
185
.
Although the enthusiasm for cancer management protocols reliant on anti-platelet strategies has
lowered in recent years, many drugs have been developed and clinically used that target PLT receptors,
interfere with PMP release, or inhibit the action of PLT-specific enzymes. Several of these drugs showed
survival advantage in cancer patients. However, the use of these drugs raises many questions due to our
current limited understanding of the consequences of interfering with the normal functions of PLTs in
hemostasis, inflammation and immunity. For example, use of these anti-platelet reagents resulted in life-
threatening bleeding complications
242
.
Moreover, increased efficacy of adoptive T cell transfer (ACT) has been reported by the
combination of ACT with anti-platelet agents aspirin and clopidogrel in a mouse model, although it is not
clear whether this result is due to the effect of aspirin on T cell infiltration or its inhibitory effect on

32
platelets
279
. For now, our knowledge of the mechanisms by which platelets contribute to the efficacy of
adoptive T cell transfer and the resulting tumor growth retention is very limited.
Conflicting observations that have been made about the influence of PLTs on cancer incidence,
survival and metastasis could be explained by the off-target effects of, and variable response to anti-platelet
agents. Here, we aim to bring a novel perspective on utilizing platelets in cancer treatment and add to our
growing potential to succeed in treating solid tumors by immune-based cancer therapies.
1.8 Special Topics 2
1.8.1 Adenosine in the tumor microenvironment

Figure 1.6 Extracellular adenosine and its effects on immune cells.
Elevated extracellular adenosine levels in the tumor microenvironment lead to the suppression of multiple immune
subsets. Arrows indicate increased expression or activation. T bars indicate inhibition or reduced activity. Figure
adapted from Sek K, Mølck C, Stewart GD, Kats L, Darcy PK, Beavis PA. Targeting Adenosine Receptor Signaling
in Cancer Immunotherapy. Int J Mol Sci. 2018 Dec 2;19(12):3837.

33
Under normal physiologic conditions, the cytosolic concentration of ATP is in the range of 1-10
mM
280
and its extracellular level is negligible
281
. However, this ATP gradient can rapidly change upon
disturbance of the plasma membrane which can be induced by apoptosis, necrosis and mechanical stress.
In the extracellular space, ATP undergoes rapid dephosphorylation by the tandem activity of
ectonucleotidases CD39 and CD73
282,283
. CD39 converts ATP to ADP and ADP to AMP, whereas CD73
dephosphorylates AMP to adenosine
284,285
. Primarily due to hypoxia and incessant inflammation, high
expression of CD39 and CD73 has been detected on the surface of tumor cells (Figure 1.6)
286
. CD39 and
CD73 nucleotidase overexpression in solid tumors is strongly associated with a poor clinical outcome for
patients suffering a variety of cancer types
287–289
.
In healthy tissues under homeostatic conditions, extracellular level of adenosine is balanced by
rapid cellular uptake
290
. Extracellular adenosine levels are critically involved in immune regulation and are
tightly regulated under homeostatic conditions
291–294
. However, due to necrosis, apoptosis, hypoxia and
persistent inflammation
284,295
ATP levels are highly elevated in the TME. Coupled with the overexpression
of CD39 and CD73 which contributes to ATP catabolism, intra-tumoral adenosine levels are significantly
elevated in solid tumors, as opposed to its negligible level within healthy tissues
296–299
. It has been reported
that extracellular adenosine levels in solid tumors are 10-20 times higher than adjacent tissues, which
hinders the function of activated cytotoxic T lymphocytes (CTLs)
298
.
1.8.2 Adenosine receptors
A1R, A2aR, A2bR, and A3R have been identified as the four adenosine receptors (ARs)
300
. High
affinity ARs (A1, A2a, and A3) are activated in the presence of low concentrations of adenosine with an
EC 50 of 0.1-0.7 μM, while A2bR is considered as low affinity with its EC 50 in the range of 15–25 μM which
may be found in the TME.
Human T cells express all four ARs and upon T cell activation
301–303
, and levels of A2aR
168,304–308
,
A2bR
168,301,305
, and A3R
302,305
increase. Specifically, T cells are primarily affected by the predominantly
expressed A2aR
303,304,309
.

34
1.8.2.1 A2aR
Adenosine binds to A2a receptor or A2b receptor to increase cAMP. The accumulation of
intracellular cAMP within immune cells inhibits immune response
310,311
. This inhibition effect has been
observed in natural killer cells, dendritic cells, and T cells
166,312–314
. The elevated levels of cAMP has a
broad range of immunosuppressive effects
315
, such as increased production of immunosuppressive
cytokines (e.g., TGF-β)
168
and upregulation of alternate immune checkpoint pathway receptors (e.g., PD-
1, LAG-3)
168,316
. Researchers have also shown that high levels of adenosine in the TME can potentially
diminish the differentiation, proliferation and effector activities of CD4
+
and CD8
+
T cells predominantly
through PKA activation, while promoting the differentiation toward Tregs, by inducing the accumulation
of intracellular cAMP
310
.
Enhancing T cell effector function is an important aspect of the A2aR blockade mode of action.
Adenosine signaling through the A2aR has suppressive effects on both CD4
+
and CD8
+
effector T cells,
including: decreased production of IFN-γ, IL-2, and TNF-α; reduced cytotoxicity of CTLs; and reduced
TCR signaling
168,315,316
. In preclinical studies researchers have verified that A2aR inhibition has the ability
to enhance effector function during an immune response
166,168,294
. Given these properties, it is expected that
the combination of adoptive T cell therapy with A2aR blockade would generate enhanced T cell function
and extended duration of cytotoxic response. Several groups have demonstrated that combination therapy
of ACT and A2aR antagonist administration gives superior results compared to single treatments in terms
of decreasing tumor growth
317,318
and improving survival
317–320
. It is important to note that Powell lab
demonstrated that A2aR blockade during initial T cell activation can greatly enhance T cell expansion and
generation of memory phenotypes
162
. In another study it has been shown that pharmacologic blockade of
A2aR resulted in a significant increase in tumor rejecting capacity of adoptively transferred and tumor-
specific CD8
+
T cells in a mouse sarcoma model
166
. The mechanism of the A2aR antagonism has shown to
increase intra-tumoral presence of adoptively transferred T cells
318
and elevate their activation status.

35
Along with T cells, many other cell types in the TME including NK cells, cancer cells, tumor
associated fibroblasts and endothelial cells also express ARs
287,321–326
. In brief, by engaging A2aR and
A2bR, adenosine dampens the activity of T cells, NK cells, DCs, and macrophages
315,327–330
, while boosting
the inhibitory capacity of immunosuppressive subsets, including Tregs. Adenosine binding to A2aR on the
surface of NK cells restricts their proliferation
331,332
, maturation
331
, IFN-γ
333,334
and TNF-α
332,335
production,
target cell killing
332,333,336,337
and capacity for stimulus-induced CD69 upregulation
333,334
.
A2aR stimulation by adenosine also has a profound impact on tumor cell biology. For example, the
triggering of A2aR induces a variety of cellular responses that augment cancer cell survival
338–340
,
proliferation
341,342
, motility
343
, and promotes angiogenesis.
1.8.3 A2aR blockade
When the A2a receptor is inhibited by an A2aR antagonist, the immunosuppressive effect of
adenosine- A2aR binding can be inhibited and the antitumor effect of immune cells can be maintained
166,312–
314,344,345
. A2aR blockade holds great potential as the next generation of immune checkpoint inhibition in
cancer immunotherapy. Remarkably, it was shown that tumors derived from A2aR antagonist-treated mice
were more heavily infiltrated by CD8
+
T cells and NK cells, and comprised less Tregs
346–348
. Also, in vivo
A2aR antagonism was shown to lead to increased expression of CD69
348
and elevated production of IFN-γ
and TNF-α
318,347
by intra-tumoral CD8
+
T cells. Moreover, in vivo A2aR antagonism elicited elevated levels
of intra-tumoral NK cells producing GzB
333
and decreased the expression of PD-1, LAG3, FoxP3 and A2aR
by tumor-infiltrating Tregs
318,347
. It is important to note that ZM-241385 and SCH-58261, two A2aR
antagonists, exhibited the capacity to halt primary tumor growth in a T cell-independent manner
338
.
Given the importance of adenosinergic signaling in tumor-promotion and immunosuppression in
the TME, approaches to both blocking the generation of adenosine and inhibiting its binding to adenosine
receptors have gained significant research interest. Administration of small molecules or mAbs for A2aR
blockade has markedly enhanced antitumor immunity in various pre-clinical tumor models. Furthermore,
it has been demonstrated in several preclinical studies that A2aR inhibition shows high efficacy in

36
promoting tumor regression. When combined with other approaches to immunotherapy, A2aR blockade
has demonstrated to potentiate additive effects on tumor control
349
.
1.8.4 SCH-58261
5-Amino-7-(2-phenylethyl)-2-(2-furyl)-pyrazolo(4,3-e)-1,2,4-triazolo(1,5-c) pyrimidine (SCH
58261) is an adenosine A2aR antagonist
350
. This non-xanthine heterocyclic compound is a potent and
selective A2aR antagonist
351
. Among other adenosine receptor subtypes, SCH-58261 particularly binds to
the A2a receptor
351,352
. Compound SCH-58261 shows a binding affinity of 0.6 nM for A2aR, with 478,
8351, and above 10,000 fold selectivity against A1aR, A2bR, and A 3AR, respectively
352
. SCH-58261 is
potent against cAMP production and binding
351,353,354
. SCH-58261 showed an IC 50 of 17 nM in the G
protein-mediated cAMP functional assay
351
. According to an experiment in which Chinese hamster ovary
cells stably expressing the human A2A adenosine receptor were pre-treated with adenosine deaminase and
then incubated for 30 min, 25℃, pH 7.4 with a [3H]-SCH 58261 concentration ranging from 0.0625 to 64
nM, the specific binding of [3H]-SCH 58261 was rapid, saturable, and increased linearly with respect to
protein concentration over the range of 25 ± 250 mg of protein/assay. Binding equilibrium was achieved
within 5 min and stable for at least 4 h. A Kd value of 2.3 nM was determined for the A2a adenosine
receptors
355
.
Given that A2aR is the dominant receptor for adenosine in its immunosuppressive roles
294,326
,
blocking A2aR signaling would create the largest impact for immunomodulation of the adenosine-rich
TME. In one study, blocking A2aR signaling with the antagonist SCH-58261 improved IFN-gamma levels
and cytotoxicity of CD8
+
T cells, leading to tumor suppression
333
. Similarly, in CD73
+
mouse tumor
models, SCH-58261 enhanced tumor immunotherapy and suppress metastases
333,356
.
Since its development, SCH-58261 has been widely used as a reference drug in the development
of other A2aR antagonists
292,357–360
and the majority of the research about the application of SCH-58261 is
on neurology and psychiatry, specifically in Parkinson’s disease
350
.

37
However, there is still little information known about its absorption, distribution, metabolism, and
excretion (ADME) and pharmacokinetics (PK) in humans
350
. The compound is known with its poor
solubility when orally dosed
355
. For example, in a study where SCH-58261 was orally administered to rats,
it was shown that approximately 28% of SCH-58261 was excreted in the first 24h without absorption, which
was attributed to the low bioavailability of SCH-58261, likely due to no/little GI absorption, and its high
metabolization by the liver in vivo
350
.
SCH-58261 is not water-soluble. It is soluble in DMSO; however administering SCH-58261
intravenously in a DMSO solution would be considered unethical due to DMSO’s irritating properties
361
.
Encapsulating SCH-58261 in a nanoparticle with lipophilic properties before its in vivo administration
could be beneficial.
1.8.5 Drug delivery
Permeation and accumulation of drugs in a tumor mass is limited due to the physical barrier of the
tumor stroma, rapid metabolism of the tumors and concomitant clearance of pharmaceuticals
362
. For
example, intravenously administered doxorubicin was found to permeate tumor cells that are only in close
proximity to blood vessels in breast cancer patients
363
. To overcome the poor permeation of drugs into the
tumor mass and achieve therapeutically relevant drug concentrations in tumors, the general tendency is to
increase the dose of drugs which cause systemic toxicity. Increasing the drug dosage, however, causes
systemic toxicity. In order to overcome the drug permeation bottleneck, and improve accumulation of drug
payloads in tumors, researchers have engineered nanoparticles by utilizing modified materials and surface
chemistries
364–366
. Numerous studies have shown that distribution profiles of anticancer drugs in tissues can
be controlled by their entrapment in nanoparticle systems. The rationale behind this approach is to increase
antitumor efficacy while simultaneously reducing systemic side effects.

38
1.8.6 Vascularization in tumor area
In tumor sites, tumor vascularization is heterogeneous due to the presence of necrosis or
hemorrhage regions. Also, angiogenesis is frequently observed in some areas in tumor, in order to provide
rapid growth advantage to tumors by oxygen and nutrient supply through dense vasculature
367
.
The tumor’s interstitial compartment, sited between the capillaries and the cells, is mainly
composed of a network of collagen and elastic fiber
368
. The cross-linked structure of the interstitium is made
up of the interstitial fluid and macromolecular elements, such as hyaluronate and proteoglycans, which
establish a hydrophilic gel
368
. Different from most normal tissues, the interstitium is attributed to high
interstitial pressure resulting in an outward directed convective (interstitial fluid) flow
367,368
. Thus,
physiological properties like pressure, as well as physicochemical properties such as composition, charge,
and structure of the interstitium, govern the transport of anticancer drugs in the interstitium. Additionally,
the physicochemical properties of anticancer drugs, such as the hydrophobicity, size and configuration of
the molecule determine their distribution
368
.
There are three barriers to delivery of therapeutic agents to tumor cells, The first is non-cellular
drug resistance mechanisms. These resistance mechanisms produce resistance at the tumor level and are
formed due to physiological barriers. Examples of these physiological barriers are (i) poor vascularization
in tumor areas which reduces drug entry to the tumor, (ii) acidic tumor environment which presents a
resistance mechanism against alkaline drugs, and (iii) high interstitial pressure, coupled with low
microvascular pressure which causes outward convective interstitial fluid flow and limited extravasation of
therapeutic agents
369
. The second barrier to delivery is cellular mechanisms which result in drug resistance
at the cellular level. Biochemistry of tumor cells becomes altered due to changes in the activity of enzyme
systems, regulation of apoptosis and cellular transport mechanisms
369,370
. Biodistribution, transformation,
and clearance of therapeutic agents in the body constitute the third barrier to the delivery of anticancer
drugs. The distribution of toxic drugs to both tumor and normal cells tend to result in side-effects and
decreases the efficacy of chemotherapy.

39
To overcome these non-cellular and cellular barriers to delivery of therapeutic agents to tumor cells,
and increasing the selectivity of drugs to target cells, researchers have turned to colloidal nanoparticles.
Specific delivery of anticancer agents to the tumor site through utilizing nanoparticles has been shown to
enhance the clinical effectiveness of the drugs and reduce their systemic toxic side effects
371,372
.
1.8.7 Nanoparticles
Nanoparticles can be defined as microscopic particles with at least one dimension being less than
100 nm
373
. Although primarily made of polymers, nanoparticles can be composed of various materials such
as biomaterials. Also, by various processes, different nanoparticles can be prepared, such as nanospheres
or nanocapsules
373
. Scientific effort has focused on nanoparticle research that utilizes nanoparticles in wide
variety of potential applications in biomedical and electronic fields. For example, lipophilic drugs can be
incorporated within liposomal nanoparticles. The drugs are then be released in a controlled manner. This
feature of nanoparticles makes them promising carriers for drugs with poor water solubility
374,375
.
When designed appropriately, nanoparticles can be used as drug delivery vehicles targeting tumor
tissues, while protecting the cargo from premature inactivation during its transport.
It is a widely studied phenomenon in nanoparticle field is that intravenously injected nanoparticles
accumulate in tumor sites due to passive diffusion in leaky, hyperpermeable tumor vasculature
376
. Research
indicates that after intravenous injection, nanoparticles predominantly accumulate in a tumor due to the
enhanced permeability and retention (EPR) effect. This phenomenon, in which the surface targeting group
is disregarded, is known as passive targeting
377,378
. The EPR effect is most useful when the blood circulation
time is prolonged
379
which is primarily achieved by coating nanoparticles with polymers like PEG thereby
diminishing their uptake by mononuclear phagocytic systems (MPSs)
380,381
. In solid tumors with leaky
vasculature, nanoparticle accumulation in tumor tissues is more likely to occur
379
. However, leaky
vasculature is very heterogeneous and is often completely absent in some tumors resulting in nanoparticles
being trapped outside tumor vessels
382
. Thus, there is a need for active targeting of drugs to disseminated
tumors. Alternatively, the drug uptake can be performed in an ‘active’ manner, as a result of a specific

40
recognition of tumor cells
383
. The conjugation of drug-loaded nanoparticles onto lymphocyte plasma
membrane allows for the delivery of the cargo to tumor sites by the recognition of nearby target cell surface
receptors by lymphocytes
379
.
Size and surface characteristics, such as surface charge and hydrophobicity, are the major cause of
biodistribution and clearance kinetics of colloidal particles. Due to high volumes of blood supply, intense
vasculature, and the high number of mononuclear phagocytic system (MPS) cells, the majority of
intravenously administrated particles (< 5µm) will accumulate in liver, spleen and other MPS organs
384–386
.
MPS, a part of the immune system that consists of phagocytic cells that spread throughout the body, is
known to affect the biodistribution profiles of drugs carried by surface non-modified conventional carriers.
Cytotoxic drugs were demonstrated to accumulate in the MPS organs, such as bone marrow, spleen, liver
and lungs, following their introduction to the bloodstream, rapid opsonization, and clearance by
macrophages of MPS organs
387
.
Despite the accumulation of drug carrying conventional nanoparticles in MPS organs, compared to
free therapeutic agents, conventional nanoparticles have been preferred due to their better safety profiles
on normal tissues. For instance, doxorubicin-loaded nanoparticles showed low levels of cardiac
accumulation which offers great promise considering the cardiotoxicity side effect of this cancer
therapeutic
388
.
Drug-loaded conventional nanoparticles have been used to target tumors which are localized in the
MPS area. However, their accumulation in tumor tissues was shown to be limited due to their short half-
life (approximately 3-5 minutes) after intravenous injection. Because conventional nanoparticles were
captured by macrophages of the MPS leading to subtherapeutic drug concentrations in tumor areas,
researchers modified the surface of conventional nanoparticles to increase their half-life and antitumor
efficacy
380,381
. A breakthrough in the application of colloidal particles was using steric stabilization to
increase particle stability in blood circulation, specifically by surface attachment of hydrophilic, nonionic
polymers onto the particles. In the early seventies, researchers demonstrated that hydrophilic particles

41
remain in the bloodstream for long durations whereas hydrophobic ones are removed from the circulation
rapidly [98].
Coated nanoparticles show enhanced localization in bone marrow
389,390
, spleen filtration
391,392
and
lung accumulation
393
. In addition, comprehensive biodistribution studies demonstrated increased uptake of
coated particles by macrophage poor organs, in particular heart, brain and kidneys, by as a result of their
adherence to the local capillary endothelium instead of the effect of MPS
394,395
. However, it must be
emphasized that the liver and spleen are the main sites where nanoparticles lodge.
By modifying the nanoparticle surface with a hydrophilic polymer, such as poly-ethylene glycol
(PEG), a hydrophilic layer is created on the particle surface, which repels plasma proteins, decreases the
uptake of circulating nanoparticles by the MPS and increases the plasma half-life of nanoparticles
396,397
.
Reported half-lives for PEG-liposomes are typically approximately 20 hours
398,399
.
When amphiphilic polymers, such as PEG, are used for coating nanoparticles, the hydrophobic
block of the polymer anchors to the polymer surface and the hydrophilic block serves as a protective layer
around the particle. By a combination of mechanisms, such as (i) surface charge (repulsive interactions)
and hydrophilicity of PEG-coated liposomes and most importantly (ii) flexibility of PEG polymer (free
rotation of individual polymer units around inter-unit linkages); coated liposomes escape recognition by
cells and become sterically hindered from blood components
397
. PEG is the most commonly used polymer
for MPS prevention and circulation half-life extension of liposomes. PEG’s other advantages include
biocompatibility, low immunogenicity and antigenicity, and lack of toxicity
396
. CAR-modified NK cells
which are conjugated with PEGylated liposomes show prolonged circulation times with a better chance for
target site binding. We hypothesized that a platform with engineered cells and drug loaded liposomes may
provide a means of the TME modifying drug, SCH, to be delivered to tumor sites with a high degree of
specificity. I will discuss the details of our hypothesis in Chapter 5.

42

2 Chapter 2: Adnectin-Based Design of Chimeric Antigen
Receptor for T Cell Engineering
Portions of this chapter are adapted from: Han X, Cinay GE, Zhao Y, Guo Y, Zhang X, Wang P. Adnectin-
Based Design of Chimeric Antigen Receptor for T Cell Engineering. Mol Ther. 2017 Nov 1;25(11):2466-
2476. doi: 10.1016/j.ymthe.2017.07.009.
2.1 Abstract
Although chimeric antigen receptor (CAR)-engineered T cell therapy has achieved encouraging
clinical trial results for treating hematological cancers, further optimization can likely expand this
therapeutic success to more patients and other cancer types. Most CAR constructs used in clinical trials
incorporate single chain variable fragment (scFv) as the extracellular antigen recognition domain. The
immunogenicity of nonhuman scFv could cause host rejection against CAR T cells and compromise their
persistence and efficacy. The limited availability of scFvs and slow discovery of new monoclonal antibodies
also limit the development of novel CAR constructs. Adnectin, a class of affinity molecules derived from
the tenth type III domain of human fibronectin, can be an alternative to scFv as an antigen-binding moiety
in the design of CAR molecules. We constructed adnectin-based CARs targeting epithelial growth factor
receptor (EGFR) and found that compared to scFv-based CAR, T cells engineered with adnectin-based
CARs exhibited equivalent cell-killing activity against target H292 lung cancer cells in vitro and had
comparable antitumor efficacy in xenograft tumor-bearing mice in vivo. In addition, with optimal affinity
tuning, adnectin- based CAR showed higher selectivity on target cells with high EGFR expression than on
those with low expression. This new design of adnectin CARs can potentially facilitate the development of
T cell immunotherapy for cancer and other diseases.

43
2.2 Introduction
Chimeric antigen receptor (CAR) is a synthetic chimeric receptor composed of an antigen
recognition domain, typically an extracellular single chain variable fragment (scFv) derived from
monoclonal antibody (mAb), a hinge, a transmembrane domain, intracellular signaling, and costimulatory
domains. Adoptive transferred CAR engineered T (CAR T) cells can specifically bind to their targets, such
as tumor-associated antigens (TAAs), via the antigen recognition domain and thus mediate cell-killing
activity toward the target cells
400
. CAR T cell therapy has achieved notable success and has shown promise
in cancer treatment, especially for treating hematological malignancies. In recent years, multiple clinical
trials testing CD19-targeting CAR T cells in adult and pediatric patients with relapsed and refractory B cell
acute lymphoblastic leukemia (ALL) have exhibited a high rate of remissions (70%–90%)
47,60,67,401–404
.
CD19 CAR T cells have shown a very encouraging clinical outcome in chronic lymphocytic leukemia
(CLL) and lymphoma patients as well
405,406
; the overall response rate in CLL patients is not as high, yet the
responding patients have achieved durable remissions for years
61,407,408
. Inspired by this impressive success
of CAR T cells in B cell malignancies, researchers in the field have been actively broadening indications
for existing CARs, developing new CAR constructs and exploring new targets for this modality in order to
treat a broader range of cancers. There are over 100 ongoing clinical trials worldwide evaluating CAR T
cell therapy in both liquid tumors and solid tumors
409
, and the number of trials is still increasing.
However, because the most widely used CAR design strategy has some considerable limitations, it
is necessary to develop an effective method for novel CAR construct design to meet fast-growing needs.
CARs are typically designed to incorporate scFv as the ectodomain to confer specificity against target
antigens. Due to the availability of existing well-characterized mouse monoclonal antibodies, many CAR
constructs used in clinical trials contain murine-derived scFv
410
. The immunogenicity of the xenogeneic
regions in CAR molecules increases the risk of developing an undesired immune response against the CAR
T cells in host. Previous clinical experience has shown responses, such as human anti-mouse antibody
(HAMA), and even one death caused by life-threatening anaphylaxis after multiple infusions of murine

44
scFv-derived CAR T cells targeting mesothelin
411
. The immune responses also resulted in limited
persistence and compromised efficacy of CAR T cells in the clinic
412,413
. One strategy to reduce
immunogenicity of CAR T cells is to humanize the scFv. However, despite its time-consuming procedure,
the humanization of scFv by CDR grafting alone might not completely preclude the potential
immunogenicity
410,413
. Rather than nonhuman scFvs, human-derived proteins could be advantageous in
preventing the unwanted immune responses to CARs. scFv has several other limitations as a key domain
in CAR molecules, such as impaired stability
414
and the potential to cause tonic signaling and exhaustion
of CAR T cells
415
. Alternative affinity molecules with a simple and stable structure may be of advantage to
substitute scFvs in CARs and overcome their limitations.
Adnectin is derived from a single domain scaffold, the tenth type III domain of human fibronectin
(10Fn3). The 10Fn3 domain is a member of the immunoglobulin (Ig) superfamily and contains a beta
sandwich protein fold, which resembles an Ig variable domain but has no disulfide bonds
416,417
. The protein
fold contains three loops, BC, DE, and FG, which are structurally analogous to antibody complementarity-
determining regions (CDRs)
418
. Introduction of mutations in these loops can confer different binding
capacities in the 10Fn3-based variants
419
. Hence, the 10Fn3 domain can be modified to mimic scFv to bind
to proteins of interest other than its natural target integrin. By mRNA, phage, or yeast display, adnectins
can be efficiently adapted to bind the target of interest with high affinity and specificity
419
. Adnectins
typically have a smaller size and more stable structure than scFv and are monomeric without disulfide
bonds, which may mitigate the basal level CAR activation caused by random crosslinking or dimerization
of scFv-derived extracellular domains
415
. Adnectins originate from human fibronectin and therefore have
minimal immunogenic potential. Because of these stated properties, adnectin has multiple advantages over
scFv and may be suitable as an alternate antigen binding domain of the CAR molecule.
We designed both scFv-based and adnectin-based CARs targeting human wild-type epidermal
growth factor receptor and compared their functions in this study. Epithelial growth factor receptor (EGFR)
is one of the most attractive targets for cancer therapy. It is widely overexpressed in a variety of cancers,
and high level of EGFR expression is correlated with poor prognosis for many cancer types
420–422
. The scFv-

45
based CAR was derived from the scFv sequence of Cetuximab, a Food and Drug Administration (FDA)-
approved chimeric monoclonal antibody against human EGFR, which has achieved considerable success
in the clinic
423
. We exploited the previous work of Emanuel et al.
424
, in which they have evaluated different
adnectin clones with high affinity to human EGFR to construct four adnectin-based EGFR CARs.
2.3 Materials and Methods
2.3.1 Construction of plasmids
The third generation of Cetuximab scFv.28BBz CAR was cloned into the MP71 retroviral vector
as previously described
425
. The amino acid sequences of adnectins targeting EGFR were published
previously
424
and shown as follows. The bolded sequences in each clone represent different BC, DE, and
FG binding loops. The adnectin CAR constructs were constructed as follows. The corresponding adnectin
DNA sequences were codon optimized (see below) and synthesized by Integrated DNA Technologies.
Then, PCR-amplified DNA sequences were assembled together with all the other fragments encoding
CD28, 4-1BB, CD3z, and MP71 backbone vector using the Gibson Assembly Cloning Kit (New England
Biolabs).
DNA Sequence of various adnectins:
E1:
ATGGGAGTGTCTGATGTGCCAAGAGACCTGGAGGTGGTAGCCGCCACGCCGACAAGTCTCTT
GATCTCATGGGACAGCGGGAGAGGTTCCTACCAATACTATCGAATCACGTACGGAGAAACA
GGCGGAAACTCCCCTGTTCAGGAGTTCACAGTGCCCGGTCCCGTGCACACTGCAACCATCAG
TGGCCTTAAGCCGGGTGTAGACTATACCATCACAGTGTATGCAGTAACTGATCATAAGCCTC
ACGCAGACGGCCCCCACACCTACCATGAGTCTCCCATTTCTATTAATTACAGAACTGAGATC
GACAAG
E2:
ATGGGAGTGTCTGATGTGCCAAGAGACCTGGAGGTGGTAGCTGCTACCCCTACGTCCTTGCT
CATCTCTTGGTTGCCCGGCAAATTGCGATACCAGTATTATCGGATTACCTATGGCGAGACCG
GGGGGAACTCCCCCGTGCAGGAGTTTACAGTACCTCATGACCTGAGAACAGCAACTATAAG
CGGCCTTAAGCCCGGTGTGGACTATACTATAACGGTGTACGCAGTGACTAACATGATGCATG
TCGAATACAGCGAGTATCCCATTTCTATTAATTACAGAACTGAGATCGACAAG
E3:
ATGGGAGTGTCTGATGTGCCAAGAGACCTGGAGGTGGTCGCCGCAACACCGACCAGCCTGC
TGATCAGTTGGGTCGCAGGAGCAGAAGACTATCAGTATTACAGGATCACCTACGGTGAAAC
GGGGGGAAATTCCCCTGTGCAGGAGTTTACTGTGCCTCATGACCTCGTAACCGCGACCATCT

46
CTGGCCTGAAGCCTGGGGTGGACTACACCATTACCGTGTACGCGGTCACGGACATGATGCAC
GTGGAATACACCGAGCACCCCATTTCTATTAATTACAGAACTGAGATCGACAAG
E4:
ATGGGAGTGTCTGATGTGCCAAGAGACCTGGAGGTGGTTGCCGCAACACCTACATCACTTCT
CATCTCCTGGTGGGCCCCTGTGGATAGGTATCAGTACTACCGGATCACATACGGTGAAACTG
GCGGAAATTCCCCCGTTCAGGAGTTCACCGTGCCCAGGGACGTGTACACCGCCACTATCAGC
GGTCTTAAACCAGGAGTCGATTACACGATCACGGTGTACGCTGTCACCGATTATAAACCCCA
CGCCGATGGGCCACATACCTACCATGAATCCCCCATTTCTATTAATTACAGAACTGAGATCG
ACAAG

E1(105aa):MGVSDVPRDLEVVAATPTSLLISWDSGRGSYQYYRITYGETGGNSPVQEFTVPGPVH
TATISGLKPGVDYTITVYAVTDHKPHADGPHTYHESPISINYRTEIDK;

E2(100aa):MGVSDVPRDLEVVAATPTSLLISWLPGKLRYQYYRITYGETGGNSPVQEFTVPHDLR
TATISGLKPGVDYTITVYAVTNMMHVEYSEYPISINYRTEIDK;

E3(100aa):MGVSDVPRDLEVVAATPTSLLISWVAGAEDYQYYRITYGETGGNSPVQEFTVPHDLV
TATISGLKPGVDYTITVYAVTDMMHVEYTEHPISINYRTEIDK;

E4(105aa):MGVSDVPRDLEVVAATPTSLLISWWAPVDRYQYYRITYGETGGNSPVQEFTVPRDV
YTATISGLKPGVDYTITVYAVTDYKPHADGPHTYHESPISINYRTEIDK

2.3.2 Cell lines and culture media
Cell lines 293T, MDA-MB-231, and U87 were cultured in D10 medium consisting of DMEM
medium supplemented with 10% fetal bovine serum (FBS), 2 mM L-glutamine, 100 U/mL penicillin, and
100 mg/mL streptomycin. H292 cells were cultured in R10 medium consisting of RPMI-1640 medium
supplemented with 10% FBS, 2 mM L-glutamine, 100 U/mL penicillin, 100 mg/mL streptomycin, and 10
mM HEPES. The above culture media and supplements were purchased from Hyclone. Human PBMCs
from healthy donors were obtained from AllCells. PBMCs were cultured in T cell medium (TCM)
consisting of X-Vivo 15 (Lonza) supplemented with 5% human AB serum (GemCell), 1% HEPES (Gibco),
1% Pen-Strep (Gibco), 1% GlutaMax (Gibco), and 0.2% N-acetyl cysteine (Sigma- Aldrich).
2.3.3 Retroviral vector production
As previously described, retroviral vectors were prepared by transient transfection of 293T cells
using a standard calcium phosphate precipitation protocol
426
. Fresh supernatant-containing retroviral
vectors were collected 48 h after transfection and used to transduce activated T cells.

47
2.3.4 T cell transduction and expansion
Frozen human PBMCs were thawed in TCM and rested overnight. PBMCs were activated by
culturing with Dynabeads Human T-Activator CD3/CD28 (Thermo Fisher Scientific) at a bead-to cell ratio
of 3:1 and 10 ng/mL recombinant human IL-7 and IL-15 (PeproTech). After 2 days, activated T cells were
added onto retroviral vector loaded non-tissue-culture-treated 12-well plates as previously described
425
and
spun at 1,000 x g at 32°C for 10 min and incubated overnight. On the following day, transduced T cells
were harvested in fresh TCM and the same transduction procedure was repeated to enhance the transduction
rate. During transduction and ex vivo expansion, culture medium was supplemented with 10 ng/mL IL-7
and IL-15 and replenished every 2 days. Cell density was adjusted to 0.5 x 10
6
cells/mL for optimal T cell
growth.
2.3.5 Surface immunostaining and flow cytometry
To detect CAR expression on the cell surface, 1 x 10
6
cells were harvested and washed three times
with fluorescence-activated cell sorting (FACS) buffer (PBS containing 4% bovine serum albumin fraction
V) and then stained with recombinant human EGFR-Fc (R&D Systems) in FACS buffer at 4°C for 30 min.
After two washes, cells were stained with phycoerythrin (PE) AffiniPure F(ab’)2 fragment of goat
antihuman IgG (Jackson ImmunoResearch) in FACS buffer at 4°C for 30 min. Cells were washed twice
and resuspended in PBS. Fluorescence was assessed using a Miltenyi Biotec flow cytometer, and the FACS
data were analyzed with FlowJo software.
2.3.6 EGFR surface staining and quantification
To detect EGFR expression on the cell surface, 1 x 10
6
cells from different cell lines (K562, 293T,
U87, MDA-MB-231, and H292) were harvested, washed, and then stained with PE anti-human EGFR
antibody (BioLegend) or mouse IgG1-PE (BioLegend) as the isotype control. The EGFR molecules were
quantified based on the mean fluorescence intensity of stained cells. The calibration was performed with
Sphero Rainbow Calibration Particles (Spherotech) following the manufacturer’s instructions.

48
2.3.7 Intracellular cytokine staining
1 x 10
6
T cells were cocultured with target cells at a ratio of 1:1 for 6 h at 37°C and 5% CO 2 with
GolgiPlug (BD Biosciences) in 96-well round-bottom plates. PE-Cy5.5 anti-CD3 antibody, APC-Cy7 anti-
CD4 antibody, Pacific Blue anti-CD8 antibody, and PE anti-IFN-γ antibody were used for immunostaining.
All the antibodies were purchased from BioLegend. Cytofix/Cytoperm Fixation and Permeabilization Kit
(BD Biosciences) was used to permeabilize the cell membrane and perform intracellular staining according
to the manufacturer’s instructions.
2.3.8 Degranulation assay
0.5 x 10
6
T cells were cocultured with target cells at a ratio of 1:1 for 4 h at 37°C and 5% CO 2 with
GolgiStop (BD Biosciences) and FITC anti-CD107a antibody in 96-well round-bottom plates. PerCP/Cy5.5
anti-CD4 antibody and Pacific Blue anti-CD8 antibody were used for immunostaining of the T cell surface
marker. All the antibodies were purchased from BioLegend.
2.3.9 Cytotoxicity assay
Target cells H292 were resuspended at the concentration of 1 x 10
6
cells/mL and labeled with 5
mM fluorescent dye CFSE in PBS+0.1% BSA. After a 30-min incubation at 37°C, the same volume of FBS
was added into the cell suspension and incubated for 2 min at room temperature to stop the labeling reaction.
The labeled target cells were then washed twice and suspended in fresh R10 medium. Cocultures were set
up in round-bottom 96-well plates in triplicates at the following effector-to-target ratios: 1:1, 3:1, and 10:1,
and each well had 5 x 10
4
target cells. After an 18-h incubation at 37°C, the suspended cells were directly
harvested, whereas the attached cells were obtained by trypsinization. All the cells were stained with 7-
AAD and then flow cytometric analysis was performed to quantify remaining live (7-AAD negative) target
cells. The cytotoxicity was calculated as 100%, the percentage of alive target cells/alive target cells in
control wells without effectors. The statistics were presented in mean ± SEM.

49
2.3.10 Antitumor efficacy of CAR T cells in a non-small cell lung cancer xenograft mouse
model
The animal experiments were conducted according to the animal protocol approved by USC
Institutional Animal Care and Use Committee (IACUC). 6- to 8-week-old female NSG mice (Jackson
Laboratory) were used in this study. On day 0, 4 x 10
6
H292 cells were injected into the right flank of NSG
mice in PBS. When the average tumor size reached 120 mm
3
on day 19, all the mice were randomized
based on the tumor size and assigned into four groups (n = 8). Mice were treated with four million CAR T
cells via tail vein injection on day 19 and day 33. CAR expression was normalized to 30% in all the CAR
groups by the addition of donor matched non-transduced T cells. Tumor size was monitored twice every
week by calipers and calculated by the following formula: L x W x H/2. Mice were euthanized when they
displayed obvious weight loss, ulceration of tumors, or tumor size larger than 1,000 mm
3
.
2.3.11 Statistical analysis
Statistical analysis was performed in GraphPad Prism, version 5.01. One-way ANOVA with
Tukey’s multiple comparison was performed to assess the differences among different groups in the
cytotoxicity assays. Tumor growth curve was analyzed using one-way ANOVA with repeated measures
(Sidak’s multiple comparison method). A p value less than 0.05 was considered statistically significant.
Significance of findings was defined as ns, not significant, p > 0.05; *p < 0.05.
2.4 Results
2.4.1 Design and generation of EGFR-specific CAR constructs
We have constructed several third-generation CARs targeting human wild-type EGFR. The
conventional scFv-based CAR incorporates the scFv derived from cetuximab as the antigen-recognition
domain and hence is designated as Cetux-CAR. The adnectin-CARs use EGFR targeting adnectin
sequences that were previously published by Emanuel et al. to substitute the scFv sequence
424
. Using an

50
mRNA display, they generated four adnectin clones (E1, E2, E3, and E4) targeting the human wild-type
EGFR extracellular domain with a different affinity (KD = 0.7, 3.4, 9.92, and 0.13 nM)
424
. The
corresponding CARs are designated as E1-CAR, E2-CAR, E3-CAR, and E4-CAR. The above scFv-based
and adnectin-based ectodomains are linked through CD8a hinge and transmembrane domains, CD28 and
4-1BB costimulatory domains, and CD3z T cell receptor signaling domain (Figure 2.1).

Figure 2.1 Engineering T cells with scFv-based CAR and adnectin-based CAR to target tumor.
(A) Schematic representation of adnectin-based CAR structure. Adnectin can be used as the antigen-recognition
domain in the CAR construct instead of scFv. (B) CAR construct targeting human wild-type EGFR. For scFv-based
CAR, the scFv region of Cetuximab was fused in frame with the CD8a hinge and transmembrane domain, followed
by the CD28/4-1BB/CD3z signaling domains. For adnectin-CAR, different clones of EGFR-targeting adnectins
(E1/E2/E3/E4) were cloned upstream of the hinge domain to replace the Cetuximab scFv sequence.

51
2.4.2 Evaluation of adnectin-based CARs
Human peripheral blood mononuclear cells (PBMCs) were activated for 2 days and transduced
with viral vectors encoding different adnectin-CARs. All four groups of adnectin-CAR T cells
demonstrated expected surface expression of CARs and a similar CAR-positive percentage (approximately
50%–60%), measured by binding of recombinant human EGFR (Figure 2.2A). When cocultured with
EGFR-overexpressing breast cancer cells MDA-MB-231
427
, these four adnectin-CARs displayed different
levels of reactivity against target cells. E3-CAR T cells had the highest percentage of interferon gamma
(IFN-γ)-secreting CD8 T cells, measured by intracellular cytokine staining (Figure 2.2B).

Figure 2.2 Evaluation of adnectin-CARs based on their expression and functional activity in human T cells.
(A) Different groups of adnectin-CAR T cells and non-transduced T cells (UT) were incubated with recombinant
human EGFR-Fc protein followed by staining with PE conjugated goat anti-human Fc antibody to assess the CAR
expression via antigen binding. (B) All four groups of adnectin-CAR T cells were cocultured with EGFR
overexpressing MDA-MB-231 breast cancer cells in the presence of GolgiPlug inhibitor for 6 h to assess the CAR T
cell activity. Non-transduced T cells were used as a negative control. Interferon-gamma (IFN-γ)-producing CD8
+

T
cells were detected by intracellular staining with anti-IFN-γ antibody and their percentage over the total CD8
+
T cell
population is shown in each panel.

It indicates that although E3 adnectin has the lowest affinity toward EGFR among the four clones,
E3-CAR exhibits the highest reactivity against the target MDA-MB-231 cells
424
. The differences among
these adnectin clones may be attributed to their different binding sites and hence distinct accessibility to the

52
target antigen in the tumor cell surface. In the following experiments, E3-CAR was chosen as the adnectin-
CAR to compare with Cetux-CAR.
2.4.3 E3-CAR displayed lower binding affinity toward EGFR compared to Cetux-CAR
To compare scFv-CAR and adnectin-CAR, we activated and transduced human T cells with viral
vectors encoding Cetux-CAR or E3-CAR in parallel and measured the CAR expression on the cell surface.
E3-CAR had a higher rate of expression than did Cetux-CAR (52.9% versus 38.2%), whereas the CAR
expression level was lower than that of Cetux-CAR. The E3-CAR
+
population had a relatively lower median
fluorescence intensity (MFI 600) than that of the Cetux-CAR
+
population (MFI 973). Overall, their CAR
expression in human primary T cells was comparable (Figure 2.3A).
We next determined whether the difference of Cetuximab scFv (KD = 0.39 nM)
428
and adnectin E3
(KD = 9.92 nM)
424
in their binding affinity toward EGFR remained in the CAR structure. The two groups
of CAR T cells were stained with recombinant human EGFR-Fc at serially diluted concentrations. As
expected, Cetux-CAR exhibited about a 30-fold higher binding affinity to the target antigen compared to
E3-CAR. Cetux-CAR achieved 50% of normalized antigen binding at 0.41 nM of target antigen EGFR,
whereas E3-CAR reached a concentration of 12.21 nM (Figure 2.3B). The relatively lower binding affinity
of E3 could be exploited for better distinction between high and low EGFR-expressing target cells.

53

Figure 2.3 Comparison of binding affinity of Cetux-CAR and E3-CAR.
(A) Comparison of Cetux-CAR and E3-CAR expression in human T cells. (B) The binding affinity of Cetux-CAR
and E3-CAR to their target antigen EGFR was assessed. CAR T cells were stained with recombinant human EGFR-
Fc at different concentrations and subsequently with goat anti-human Fc antibody. The fluorescence was measured
by flow cytometry and normalized into percentage of antigen binding.

2.4.4 E3-CAR T cells displayed higher selectivity against EGFR overexpressing cancer
cells from lower EGFR-expressing cells
Affinity-tuned CARs bearing lower affinity scFvs have been demonstrated to be beneficial for
enhanced selectivity against cells with a different target antigen density
429,430
. To evaluate the selectivity of
E3-CAR toward different target cells, we first collected multiple EGFR-positive cell lines and one EGFR-
negative cell line and quantified their EGFR expression levels. The EGFR-positive cell lines (293T, U87,

54
MDA-MB-231, and H292) all express wild-type EGFR on the cell surface, but the expression levels are
very different, ranging from 10
3
to 10
5
molecules per cell. Based on the quantification results, H292 and
293T cells had the highest and lowest EGFR expression levels, respectively, and MDA-MB-231 cells
exhibited a medium-level EGFR expression. As expected, the EGFR-negative cell line K562 did not display
detectable EGFR molecules on the surface (Figure 2.4A and Figure 2.4B).
Cetux-CAR displayed similar levels of reactivity against all the EGFR positive cell lines,
presumably due to the high affinity of Cetuximab scFv (Figure 2.4C). In contrast, E3-CAR T cells could
distinguish target cells with a different EGFR density. E3-CAR T cells had the upregulated degranulation
marker CD107a in the presence of EGFR-overexpressing cancer cells, such as U87, MDA-MB-231, and
H292, and the percentage of degranulation with H292 was significantly higher than that of Cetux-CAR (p
< 0.001). On the other hand, E3-CAR T cells did not show reactivity toward 293T cells, which only had
low-level endogenous EGFR expression
431
.

55

Figure 2.4 Activation of E3-CAR T cells is positively correlated with EGFR density on target cells.
(A) Different cell lines (K562, 293T, U87, MDA-MB-231, and H292) were stained with PE anti-human EGFR
antibody (dark gray histograms) and mouse IgG1-PE as isotype control (light gray histograms). The representative
histograms from triplicates are shown above. (B) EGFR expression was measured and quantified by flow cytometry.
Summarized statistics from triplicates are shown in the bar graph and table. (C) On day 10 post-activation, Cetux-
CAR T, E3-CAR T, and non-transduced T cells were cocultured with H292 cells at a 1:1 ratio for 4 h with GolgiStop
inhibitor and FITC-conjugated CD107a antibody against the degranulation marker CD107a. Non-stimulated CAR T
cells were used as negative control, whereas anti-CD3 and anti-CD28 antibody stimulated CAR T cells were used as
positive control. The upregulated degranulation marker CD107a was identified, and the CD107a
+
CD8
+
T cell
population was gated and its percentage over total CD8
+
T cells is shown in each scatterplot. This degranulation assay
was repeated with CAR T cells derived from three different donors, and the summarized statistics are shown in bar
graphs (n = 3, mean ± SEM; ns, not significant; *p < 0.05; **p < 0.01; ***p < 0.001, one-way ANOVA with Tukey’s
multiple comparison).

56
2.4.5 E3-CAR had comparable reactivity against H292 lung cancer cells to that of Cetux-
CAR

Figure 2.5 Activity of Cetux-CAR and E3-CAR against H292 lung cancer cells.
(A) On day 10 after activation and expansion ex vivo, Cetux-CAR T, E3-CAR T, and non-transduced T cells were
cocultured with H292 cells with GolgiPlug inhibitor for 6 h. Intracellular cytokine staining of CAR T cells was
performed, and the gated IFN-γ
+
CD8
+
T cells are shown in each scatterplot. (B) Cetux-CAR T, E3-CAR T, or non-
transduced T cells were cocultured with H292 cells for 18 h at different effector-to-target ratios of 1:1, 3:1, or 10:1.
Cytotoxicity of CAR T cells against target cells was measured. The summarized statistics were shown as mean ± SEM
(n = 3; ns, not significant; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; one-way ANOVA with Tukey’s
multiple comparison).

To further determine the function of E3-CAR T cells, we performed an intracellular staining assay
and cytotoxicity assay with the lung cancer cell line H292, which has a high expression level of wildtype
EGFR on the cell surface
407
. In the presence of the GolgiPlug inhibitor, Cetux-CAR, E3-CAR, and non-

57
transduced T cells were cocultured with H292 cells for 6 h and then the production of inflammatory
cytokine IFN-γ in the CD8
+
T cell population was measured and shown in the scatterplot. All the CD8
+
T
cell populations in different groups only displayed a background level of IFN-γ in the absence of
stimulation. Upon stimulation of H292 target cells, Cetux-CAR T cells displayed 12.2% IFN-γ
+
CD8 T
cells, whereas E3-CAR T cells exhibited higher activity (20.0% IFN-γ
+
CD8 T cells) (Figure 2.5A). This
observed higher activity of E3-CAR T cells is consistent with the previous finding of the degranulation
assay (Figure 2.4C).
To test the cell killing activity of both CAR T cells, they were cocultured with H292 cells at
different effector to target ratios (E:T) (1:1, 3:1, and 10:1) for 18 h and the specific target cell lysis was
calculated. Both Cetux-CAR and E3-CAR T cells had significantly higher cytotoxicity toward H292 than
did non-transduced T cells at all three E:T ratios (Figure 2.5B). There was no significant difference in the
cytotoxicity between Cetux-CAR and E3-CAR T cells.
2.4.6 E3-CAR T cells showed similar antitumor efficacy to Cetux-CAR T cells in vivo
Given the comparable cytotoxicity of E3-CAR T cells toward H292 cells in vitro, we next sought
to investigate the antitumor efficacy in vivo. We created a subcutaneous human lung cancer xenograft model
in immunodeficient NOD.Cg-PrkdcscidIL2Rgtm1Wjl/Sz (NSG) mice. On day 0, NCI-H292 cells were
injected into the right flank of NSG mice. When the average tumor size reached around 120 mm
3
on day
19 after tumor inoculation, all the tumor-bearing mice were randomized into tumor size rank matched
cohorts (n = 8 per treatment group), and then CAR T treatment was started. Mice were treated with four
million CAR T cells through intravenous injection on day 19 and 33, and tumor growth was monitored
(Figure 2.6A).

58

Figure 2.6 Antitumor efficacy of CAR T cells in a human lung cancer xenograft model.
(A) The CAR T treatment timeline. H292 cells were injected into the right flank of NSG mice on day 0. On day 19,
when the xenografted tumors were established, mice were randomized into three groups (n = 8 each group) and treated
with four million Cetux-CAR T, E3-CAR T, or non-transduced T cells on day 19 and day 33. Tumor size was
measured by caliper twice every week. (B) Tumor growth curve in each group was shown as mean ± SEM (ns, not
significant; *p < 0.05). (C) The average body weight change of each group was shown as mean ± SEM (ns, not
significant; *p < 0.05).

Although all groups of animals showed tumor progression, mice receiving Cetux-CAR or E3-CAR
T cells had significant tumor growth inhibition compared to those receiving non-transduced T cells (one-
way ANOVA: p < 0.05, Cetux-CAR versus UT; p < 0.05, E3-CAR versus UT) (Figure 2.6B). The average
body weight change throughout the study was not significantly different in all three groups (Figure 2.6C).
On day 42, the tumor tissues were harvested and tumor infiltrating T cells were enumerated. The Cetux-
CAR group had a significantly higher percentage of T cells trafficking to the tumor site than did the control
group, whereas the E3 CAR group was not significantly different from either the control or Cetux-CAR
group (Figure 2.7).

59

Figure 2.7 Trafficking of CAR T cells in the human lung cancer xenograft NSG mice model.
NCI-H292 cells were inoculated into the right flank of NSG mice on day 0. When the average tumor size reached
around 120 mm3 on day 19 after tumor inoculation, the tumor-bearing mice were randomized into tumor size rank
matched cohorts (n = 8 per treatment group), and then treated with 4 million CAR T cells through intravenous injection
on day 19 and 33. On day 42, the tumor tissues were harvested and processed into single cell suspension, and then
stained with T cell marker antibodies for enumeration by flow cytometry. Mean ± SEM of CD3+ CD8+ T cell number
was shown. Student’s t-test was employed to assess the differences among groups (ns = not significant, P > 0.05; *, P
< 0.05).

2.5 Discussion
Adoptive transfer of CAR T cells has achieved considerable success in multiple clinical trials;
however, the inherent limitations of conventional CAR design and the lack of a more efficient method for
novel construct design have curbed the development of this promising therapy. Our study has shown that
compared to scFv-based CARs, CARs using adnectin as the antigen recognition domain are equally
effective to kill tumor cells. It indicates that the novel method to derive CAR constructs from adnectin
sequences is feasible and has potential value for future CAR development.
This method certainly offers some advantages and opportunities for CAR design. The flexibility of
adnectin selection could allow optimal affinity tuning of the antigen-binding moiety to enhance CAR
specificity to tumors. Affinity tuning is critical for CAR to discriminate tumor cells that overexpress target
antigens from normal tissues that express target antigens at physiological levels. Traditionally, CARs are
mostly designed to incorporate high-affinity scFvs derived from monoclonal antibodies. However, previous

60
studies have demonstrated that the activation threshold of CARs is inversely correlated with the scFv
affinity and that high-affinity CARs show a poor discrimination power among target cells with different
levels of antigen expression
432,433
. Different from high-affinity therapeutic antibodies, high-affinity CARs
may result in much more serious on-target off-tumor toxicity in clinical trials
434
. This is presumably caused
by the higher sensitivity of CAR T cells to cells with low target expression than antibody-based therapy
435
.
Previous studies have provided evidence that by tuning down the affinity of CARs toward the target antigen
via low-affinity scFvs, both anti-EGFR or anti-human epidermal growth factor receptor 2 (HER2) CARs
could distinguish tumor cells from normal tissues and only recognize and eradicate tumor cells with high
expression levels of target antigens
429,430,432
. One example is the nimotuzumab scFv-derived CAR-targeting
EGFR designed by Caruso et al.
430
. Nimotuzumab has a 10-fold lower Kd than cetuximab, resulting from
a 59-fold reduced on-rate of binding, which imparts a requirement for (at least) bivalent binding to EGFR
and restricts the binding to cells expressing high-density EGFR
424
. It should be pointed out that
nimotuzumab-derived CAR bears a similar level of affinity against EGFR as that of E3 adnectin (2-fold
difference). Both nimotuzumab-CAR and E3-CAR display a similar level of biological function in vitro as
that of high-affinity scFv CARs, and their relatively low affinity leads to the enhanced selectivity toward
EGFR-overexpressing tumor cells over normal cells with the endogenous level of EGFR expression.
Therefore, developing low-affinity adnectin-CARs with enhanced tumor selectivity might be a promising
strategy to improve the safety profile of CAR T therapy. Paradoxically, low-affinity adnectins may have
disadvantages in terms of T cell persistence and proliferation. E3-CAR had a lower percentage of tumor-
infiltrating T cells compared to Cetux-CAR, although the difference was not statistically significant, and it
may compromise the antitumor efficacy of E3-CAR T cells in vivo. It highlights the importance to optimize
the affinity range for the antigen-recognition domain of CAR to achieve a balance of both efficacy and
safety.
In addition, the human-derived sequences of adnectin render relatively low immunogenicity and
could potentially allow longer persistence and higher efficacy of adnectin-CAR T cells. It has been more
well-known that the routine lymphodepletion procedure before adoptive transfer of engineered T cells is

61
one of the key factors leading to clinical success of CAR T therapy
435
. In the short time window provided
by lymphodepletion for infused cells to evade host rejection, a less immunogenic CAR might not be more
advantageous; however, in the long run, the construct may be more durable; this requires further clinical
test to be demonstrated though. The newly formed epitopes in adnection-binding loops and junction sites
in CARs might still cause immune responses, but we expect to see a much lower chance for life threatening
responses to occur. Other antigen-binding moieties, such as naturally occurring receptor ligands, have been
used to design less immunogenic CARs
436,437
before and there are some successful examples in the clinic,
such as interleukin-13 (IL-13)(E13Y)-zetakine CAR-targeting IL-13Ra2 in glioblastoma treatment
438
. Yet
the availability of candidate ligands is very limited and cannot meet a broad variety of needs for recognizing
tumor cells, and the ligand-based CARs may bind to other receptors with lower affinity and cause undesired
off-target toxicity
437
.
Multiple clinical trials of CAR T cells have shown antigen escape of cancer cells and relapses in
patients due to the heterogeneous target antigen expression in cancer cells
67,403
. Previous study has
demonstrated that combinational antigen recognition by bispecific OR-Gate CAR may provide a safeguard
against antigen escape
439
. With similar rationale, multi-domain adnectin, such as EGFR and insulin-like
growth factor 1 receptor (IGF-IR) bispecific adnectin in a previous report, can also be used to enhance the
efficacy of CAR
424
. The small size of adnectin and its native structure derived from fibronectin also make
it very adaptable to develop a multi-domain adnectin that is multi-specific to different targets
424
. It provides
a new means to explore adnectin CAR design.
Despite the great potential of adnectin-CAR, there are also some limitations and concerns of this
design method. It remains unknown whether it is a widely applicable strategy to design CAR constructs
based on adnectin sequences. Although we have demonstrated that EGFR-targeting adnectin-CAR has
equivalent efficacy against target cancer cells both in vitro and in vivo, it still requires further evaluation of
its efficacy and extensive study of its reactivity to normal tissues in a clinical setting. It is especially crucial
for low-affinity adnectins because they may bear less specificity at the same time. In addition, many existing
therapeutic monoclonal antibodies and their targeted epitopes in the antigens are extensively studied and

62
the understanding of their structure and potential side effects can facilitate rational design of CARs
440,441
.
In contrast, although a number of adnectins binding to different targets, such as EGFR, IGF-IR, tumor
necrosis factor alpha (TNF- α), and vascular endothelial growth factor receptor 2 (VEGFR-2), have already
been developed
424,442–444
, little is known about the structural information of adnectins and their interactions
with target antigens
445
. The unpredictability is a double-edged sword. If the soluble extracellular domain of
a membrane-bound target protein is used as the antigen during selection, it is possible that the resultant
adnectin may not recognize the target in the physiological condition. It may also cause unexpected toxicity.
On the other hand, it offers opportunities to select out targeting sites that are only exposed in transformed
cells but not in normal cells. Therefore, a preliminary screening of multiple candidate adnectin-CARs,
based on their potential to react to tumor cells and normal tissues, may be necessary.
In summary, CAR T therapy is rapidly evolving and our study has provided a novel strategy to
develop CAR molecules from adnectins rather than the conventional CAR design from scFvs. The results
demonstrate that bearing equivalent potency to traditional CARs, adnectin-based CARs may benefit from
reduced immunogenicity, increased tumor selectivity, and improved safety profile due to optimal affinity
tuning. The method to design CAR molecules based on the adnectin sequence may help expand the
applicability of CAR therapy to a wide range of tumor types.

63
3 Chapter 3: Engineering CAR-Expressing Natural Killer Cells
with Cytokine Signaling and Synthetic Switch for an Off-the-
Shelf Cell-Based Cancer Immunotherapy
Portions of this chapter are adapted from: Qu, Y., Siegler, E., Cheng, C., Liu, J., Cinay, G., Bagrodia, N.,
& Wang, P. (2019). Engineering CAR-expressing natural killer cells with cytokine signaling and synthetic
switch for an off-the-shelf cell-based cancer immunotherapy. MRS Communications, 9(2), 433-440.
doi:10.1557/mrc.2019.31
3.1 Abstract
Immune cells can be genetically engineered with a synthetic chimeric antigen receptor (CAR) to
eliminate cancer cells, but clinical efficacy in solid tumors has been disappointing due in part to the
immunosuppressive tumor microenvironment (TME). Additionally, the cost and logistical issues of
personalized medicine necessitate the creation of an off-the-shelf CAR therapy. Synthetic biology tools
were implemented in addressing these problems: an anti-mesothelin CAR, membrane-bound IL-15/IL-
15Rα complex, and inducible caspase 9 “kill switch” were expressed in natural killer cells for tumor-
targeting capabilities, immunostimulatory effects, and safety in treating a preclinical model of ovarian
cancer with a renewable, allogenic cell therapy.
3.2 Introduction
Synthetic biology initially focused on the use of molecular biology tools to engineer bacterial hosts,
but has also made considerable progress in the engineering of mammalian cells with potential therapeutic
applications
446,447
. The strategies of synthetic biology have been used to engineer immune cells to treat
cancer using chimeric antigen receptors (CARs). CARs are synthetic receptors that can improve the cancer-
targeting capabilities of immune cells by equipping them to recognize tumor-associated antigens. CAR-

64
engineered autologous T cells have shown remarkable success in treating hematologic malignancies, but
have had disappointing clinical results in targeting solid tumor due to the immunosuppressive tumor
microenvironment (TME). Additionally, CARs targeting tumor-associated antigens that are also expressed
on normal tissue may result in “on-target, off-tumor” toxicity and have caused significant safety
concerns
400
. These unresolved issues of CAR-related treatments necessitate better CAR design. A number
of tools and approaches can program the therapeutic CAR-modified immune cells to improve persistence
as well as safety in cancer treatment.
Genetic engineering and synthetic biology strategies have been developed to improve CAR T cell
proliferation, persistence, cytotoxic function, and safety in vivo. For example, T cells can be engineered to
overexpress cytokine transgenes such as IL-12 and IL-15 that work to promote T cell proliferation and
cytotoxic functions
448,449
. Control systems through the addition of a small molecule or biologic circuit to the
CAR design have been shown to increase the safety of CAR therapy. One strategy involves the use of a
split CAR system containing drug-inducible heterodimerization domains that remain inactive but can be
turned on by the addition of a heterodimering drug
450
. T cells have also been engineered to express an
inducible suicide gene caspase 9 (iCAS9) as a safety switch which responded to a chemical inducer of
dimerization (CID) and resulted in dose-dependent levels of apoptosis
451,452
. This suicide switch is a tool
for physicians to regulate treatment when complications including “on-target, off-tumor” toxicity or
cytokine release syndrome occur
410
.
Among the candidates of immune cells to express CARs for the cellular immunotherapy, the well-
characterized cytotoxic natural killer (NK) cell line, NK-92 cell line, is a promising one due to its lack of
immunogenicity and availability as an “off-the-shelf” product
453–455
. The standardized NK-92 cell line is
thus an ideal target for a synthetic biology approach which equips cells with CARs or other engineered
structures for improved antitumor efficacy.
NK cells typically rely on interleukin (IL)-2 for growth and cytotoxic function, so during animal
studies or clinical trials, they are often co-administered with exogenous IL-2, which may cause acute
toxicity in high doses
453
. Instead, lentiviral vectors encoding IL-15 can transduce NK-92 cells to stably

65
express the cytokine and sustain continued proliferation and cytotoxic capabilities without exogenous IL-
2
121,456
. Moreover, the receptor subunit IL-15Rα can form a stable complex with IL-15 on cell surfaces and
enables autocrine stimulation of NK cells by membrane-bound IL-15 in cis presentation
457
. This membrane
bound IL-15/IL-15Rα complex has shown to contribute to the long survival of CD8 memory T cells and
the persistence of T cells engineered to co-express CAR and the complex in a tethered form
449,458
. We
hypothesized that NK-92 cells engineered to co-express CAR and the membrane-bound IL-15/IL-15Rα
complex would also survive and proliferate without exogenous IL-2.
For both preclinical animal studies and clinical trials using NK-92 cells to treat cancer, the cells are
typically irradiated prior to infusion for safety concerns to prevent engrafting cells from in vivo
tumorigenicity
453,454
. As a result, the infused NK-92 cells experience limited expansion without long-term
persistence. For NK-92 cells engineered to express membrane bound IL-15/IL-15Rα complex, irradiation
prior to infusion would greatly defeat the purpose of the complex to improve NK-92 proliferation and
persistence in vivo. The iCAS9 suicide gene, when transduced into the CAR-NK cells with membrane
bound IL-15/IL-15Rα complex, serves as a safety mechanism other than irradiation without affecting the
proliferation and persistence of infused NK-92 cells.
Mesothelin is a cell-surface molecule overexpressed in many carcinomas and is an attractive target
for CAR-engineered immune therapies
459
. A third-generation anti-mesothelin CAR consisting of a single-
chain variable fragment domain targeting mesothelin and two co-stimulatory domains, CD28 and CD137
(4-1BB), transduced into T cells has shown to eradicate large, established tumors overexpressing
mesothelin antigen
460
. Therefore, we utilized synthetic biology technology and engineered NK-92 cells to
express anti-mesothelin CAR and membrane-bound IL-15/IL-15Rα complex for specific killing and better
proliferation in an ovarian cancer mouse model. We further transduced the engineered CAR-NK cells with
the iCAS9 suicide gene as a safety switch to eliminate infused CAR-NK cells in case of adverse side effects
or posttreatment, in order to mitigate the risk of tumorigenicity from the infused cells.

66
3.3 Materials and Methods
3.3.1 Cell culture
SKOV3 (ATCC HTB-77) tumor cell lines were maintained in a 5% CO 2 environment in RPMI
1640 (Gibco) media supplemented with 10% FBS, 1% pen-strep, and 2 mM L-glutamine. NK-92 cells (Dr.
Jihane Khalife, Children’s Hospital Los Angeles; ATCC CRL-2407) were maintained in MEM-α (Gibco)
supplemented with 10% FBS, 10% horse serum, 1% NEAA, 1% pen-strep, 1% sodiumpyruvate, 0.1 mM
2-β mercaptoethanol, 0.2 mM myo-inositol, and 2.5 μM folic acid. Mesothelin
+
SKOV3 (SKOV.meso)
cells were generated by transducing SKOV3 cells with lentivirus containing mesothelin cDNA and sorting
mesothelin
+
cells with fluorescence-activated cell sorting.
3.3.2 Virus production
The CAR constructs consisted of ss1, the murine-derived scFv anti-mesothelin antigen-binding
domain; CD8 hinge and CD28 transmembrane regions; and CD28, 4-1BB, and CD3ζ cytoplasmic regions
in the retroviral MP71 vector. The membrane-bound IL-15/IL-15Rα complex was included in one of the
CAR constructs following a 2A linker. The iCAS9 suicide switch with GFP marker, pMSCV-F-del
Casp9.IRES.GFP, was purchased from Addgene. The plasmid for mesothelin consisted of human
mesothelin cDNA cloned into a lentiviral FUW backbone. These plasmids were used to transfect HEK
293T cells in 30 mL plates using CaCl 2 precipitation methods. Fresh media (high glucose DMEM
supplemented with 10% FBS and 1% pen-strep) was plated onto the cells 4 h after initial transfection.
Supernatants were harvested and filtered (0.45 μm) 48 h later and used fresh.
3.3.3 Transduction of NK-92 and SKOV3 cells
NK-92 cells were transduced with anti-mesothelin CAR and iCAS9 retroviral vectors. Non-tissue
culture-treated 12-well plates were coated overnight with 25 μg RetroNectin per well (Clontech
Laboratories, Mountain View, CA). Retroviral vectors were subsequently spin-loaded onto the plates by

67
centrifuging at 2000 x g for 2 h at 32°C. NK-92 cells were resuspended at a concentration of 5 x 10
5
/mL
with fresh media complete with 200 U/mL human IL-2 and added to the vector-coated plates. The plates
were centrifuged at 600 x g for 30 min at 32°C and incubated overnight at 37°C and 5% CO 2. CAR
+
or
GFP
+
NK-92 cells were sorted using fluorescence-activated cell sorting. SKOV3 cells were similarly
transduced to express human mesothelin via a lentiviral vector. One million cells were added to 2 mL of
fresh viral supernatant and centrifuged at 1050 x g for 90 min at room temperature and resuspended in fresh
media. Mesothelin
+
cells were sorted using fluorescence-activated cell sorting.
3.3.4 CAR detection on NK cell surface
After transduction and cell sorting, 1 x 10
5
NK-92 cells were incubated with recombinant human
mesothelin-Fc chimera (R&D Systems, Minneapolis, MN) at a volume ratio of 1:50 (2 μg/mL) in PBS at
4°C for 30 min and rinsed with PBS. The cells were subsequently incubated with PE-labeled goat antihuman
Fc (Jackson ImmunoResearch) at a volume ratio of 1:150 in PBS at 4°C for 10 min, rinsed, and read using
flow cytometry. Non-transduced NK-92 cells served as a negative control.
3.3.5 Cytokine release assay
The 1 x 10
5
NK cells per well were coincubated with target cells in 96-well round-bottom plates at
a 1:1 ratio for 6 h at 37°C. One microgram Brefeldin-A (Sigma, St. Louis, MO) was added to each well to
prevent protein transport. At the end of the incubation, cells were permeabilized using the
CytoFix/CytoPerm kit (BD Biosciences, San Jose, CA) and stained for CD45 and interferon-γ (IFN-γ) using
Pacific Blueconjugated anti-human CD45 (Biolegend, San Diego, CA) and PE-conjugated anti-human IFN-
γ (Biolegend). NK cells without target cells were used as a negative control. Results were read using flow
cytometry. The data were determined in triplicate and presented as the mean ± SEM.

68
3.3.6 Cytotoxicity assay
The 2 x 10
4
target cells were labeled with 5 μM carboxyfluorescein succinimidyl ester (CFSE, Life
Technologies, Waltham, MA) as previously described
425
and coincubated with NK cells at various ratios in
96-well round-bottom plates for 24 h at 37°C. The cells were then incubated in 7-AAD (Life Technologies)
in PBS (1:1000 dilution) for 10 min at room temperature and analyzed via flow cytometry. Percentages of
killed cells were calculated as [CFSE
+
7-AAD
+
cells/ (CFSE
+
7-AAD
−
+ CFSE
+
7-AAD
+
)] cells, with
live/dead gates based on control wells of target cells only to account for spontaneous cell death. The
cytotoxicity was determined in triplicate and presented as the mean ± SEM.
3.3.7 Chemical inducer of dimerization sensitivity assay
CIDAP20187 (B/B Homodimerizer, Takara, Mountain View, CA) was diluted in complete NK cell
media at a concentration of 10 nM. The 5 x 10
4
NK cells were incubated in media or 10 nM CID for 24 h
and analyzed for viability. Number of viable cells were counted under a light microscope via trypan blue
exclusion and percentage of viable cells was calculated using AnnexinV/7-AAD staining (BD Biosciences)
per the manufacturer’s instructions.
3.3.8 Xenograft tumor model
All animal experiments were conducted according to the animal protocol approved by the
University of Southern California Institutional Animal Care and Use Committee (IACUC). Eight-week-old
female NOD.Cg-Prkdc
scid
IL2Rγ
tm1Wj1
/SZ (NSG) mice (Jackson Laboratories) were inoculated
subcutaneously with 3.5 x 10
6
SKOV.meso cells, and treatment began when tumors grew to 50–70 mm
3
.
The mice were injected intravenously through the tail vein with 10 x 10
6
NK cells twice a week, three times
total. Tumor growth and body weight of the remaining mice were recorded until the end of the experiment.
The tumor length and width were measured with a fine caliper three times a week, and tumor volume was
calculated as ½ x (length) x (width)
2
. The survival endpoint was determined to be when the tumor reached
1000 mm
3
or became ulcerated.

69
3.3.9 Ex vivo NK cell staining
One week after the final NK cell injection, a subset of mice were sacrificed and organs were
harvested. Each tissue sample was made into a single-cell suspension through a 70 μm strainer, washed
with PBS, centrifuged, and resuspended in 3 mL TAC buffer for 10 min at room temperature. The samples
were then washed and resuspended with PBS. Fluorescent-conjugated antibodies (Biolegend) were used to
stain the samples in 100 μL PBS at a 1:100 volume ratio for 15 min at 4°C in a 96-well plate. Samples were
analyzed using flow cytometry in triplicate and presented as the mean ± SEM.
3.3.10 Statistical analysis
Statistical analysis was performed in Microsoft Excel and GraphPad Prism 6. One-way analysis of
variance with Tukey’s multiple comparison was used to assess the data between different groups. P values
of <0.05 were considered significant and denoted as follows: not significant (NS) P > 0.05; *P < 0.05; **P
< 0.01; ***P < 0.001.
3.4 Results and Discussion
The addition of a membrane-bound cytokine to the CAR construct is one method to program
immune cells to be more therapeutically effective. Anti-mesothelin CARs with or without membrane-bound
IL-15/IL-15Rα complex were transduced efficiently in NK-92 cells. Retroviral vectors as described in
“Materials and methods” section were used to transduce NK-92 cells with a third-generation anti-
mesothelin CAR, either without (NK.αmeso) or with (NK.αmeso.mbIL15) membrane bound IL-15/IL-
15Rα complex as shown in the schematic representation in Figure 3.1a. Genetically engineering CAR-NK
cells with membrane-bound IL-15/IL-15Rα complex allowed CAR-NK cells to expand in the absence of
exogenous IL-2 in vitro. Unsorted transduced NK.αmeso and NK.αmeso.mbIL15 cells were cultured in
complete media without IL-2 supplementation over 10 days. NK.αmeso cells were mostly dead by day 5,
and the CAR-expressing population of the live cells also decreased over time, while NK.αmeso.mbIL15
increased the purity of the CAR-expressing population over time in the absence of IL-2 (Figure 3.1b).

70
Although CAR
+
cells ultimately were sorted using flow cytometry, this demonstrates that starving unsorted
cells of IL-2 can serve to enrich the transduced population. After selection for transduced NK-92 cells via
fluorescence-activated cell sorting, the cells displayed high CAR expression (Figure 3.1c). Non-transduced
NK-92 cells were used as a negative control. Previous research indicates that the membrane-bound IL-
15/IL-15Rα complex acts in cis instead of trans, behaving as an autocrine stimulation by binding to the β
and γ IL-15R subunits expressed on the transduced cells instead of those on surrounding immune cells
457
.
This data supports that finding, as CAR
+
populations selectively proliferated, not the entire cell
population—cells were stimulated by the membrane-bound IL-15/IL-15Rα complex only if the cells
already expressed that transgene.

Figure 3.1 CAR expression on NK-92 cells and proliferation of CAR.NK cells.
(a) Schematic representation of CAR constructs. (b) CAR expression over time in the absence of exogenous IL-2. (c)
CAR expression of sorted cell populations. (d) Expansion of NK cells with or without exogenous IL-2.

It was further demonstrated that NK.αmeso.mbIL15 cells could survive and proliferate in the
absence of exogenous IL-2 by observing the expansion over 10 days of sorted CAR-NK cells in IL-2-free
media (Figure 3.1d). All cells expanded well in media with IL-2 supplementation, but only the
NK.αmeso.mbIL15 cells expanded without IL-2—non-transduced NK-92 cells and NK.αmeso cells died

71
by day 5 in media without IL-2 (P < 0.001). NK.αmeso.mbIL15 cells also expanded significantly more than
non-transduced NK-92 cells (P < 0.01).
CAR-NK cells with membrane-bound IL-15/IL-15Rα complex secreted cytokines and exhibited
cytotoxicity to mesothelin-expressing target cells. In vitro functionality of the CAR-NK cell lines was
assessed by performing IFN-γ release and cytotoxicity assays. NK cells incubated without target cells or
with SKOV3 cells served as a negative control, as they do not express mesothelin, while SKOV.meso cells
transduced to overexpress mesothelin were used as the desired target cells. In the IFN-γ assay, flow
cytometry was used to analyze the IFN-γ
+
NK cell populations after coincubation with each of the target
cell lines (Figure 3.2a). None of the NK cell lines reacted to SKOV3 cells, indicating specificity toward
mesothelin. Non-transduced NK-92 cells did not release IFN-γ in response to SKOV.meso cells,
demonstrating the necessity of an anti-mesothelin CAR to generate an immune response. NK.αmeso cells
had a significantly larger IFN-γ
+
population when coincubated with SKOV.meso cells when compared with
non-transduced NK cells (P < 0.001), but NK.αmeso.mbIL15 cells had significantly higher percentages of
IFN-γ
+
cells after exposure to SKOV.meso cells compared with both non-transduced (P < 0.001) and
NK.αmeso cells (P < 0.001). These data are in accordance with previous work showing that IL-15 aids in
greater production of IFN-γ in NK cells
461–463
.

Figure 3.2 Evaluation of interferon-γ secretion and cytotoxicity of CAR.NK cells.
(a) Interferon-γ secretion by NK cells coincubated with mesothelin
+
or mesothelin
−
target cells. (b) Cytotoxicity of
NK cells coincubated with mesothelin
+
or mesothelin
−
target cells. Significance stars compare non-transduced NK
cells to NK.αmeso and non-transduced NK cells to NK.αmeso.mbIL15 (*P < 0.05, **P < 0.01, ***P < 0.001).

Cytotoxicity assays were performed using the above-mentioned target and effector cells in different
ratios. NK.αmeso.mbIL15 cells killed lower percentages of SKOV.meso target cells compared with

72
NK.αmeso cells, perhaps due to slightly lower CAR expression on the cell surface. Both CAR-NK cell lines
were significantly more cytotoxic to SKOV.meso cells compared with non-transduced NK-92 cells at all
effector-to-target ratios (Figure 3.2b; significance stars are comparing non-transduced NK cells to
NK.αmeso and non-transduced NK cells to NK.αmeso.mbIL15].
The iCAS9 suicide gene harnesses technology from genetic engineering and small molecule drugs.
This kill switch was efficiently transduced in NK-92 cells (NK-92 without CAR, NK.αmeso, and
NK.αmeso.mbIL15) using the commercially available retroviral vector described in “Materials and
methods” section. After fluorescence-activated cell sorting, all cell lines showed similar high expression
levels of iCAS9 (Figure 3.3a; NK-92 at 98.5% iCAS9
+
, NK.αmeso at 97.8%, and NK.αmeso.mbIL15 at
98.3%]. Uniform transgene expression across all groups is important, as higher iCAS9 expression leads to
greater sensitivity to the CID AP20187 used when triggering apoptosis
452
.
NK-92 cells expressing iCAS9 were highly sensitive to low concentrations of the CID AP20187,
an otherwise non-toxic small molecule dimerizer. To test the capability of the suicide transgene to induce
apoptosis in cells after exposure to the AP20187, viable cell populations were observed using either trypan
blue exclusion or AnnexinV/7-AAD staining. Six cell lines were tested—non-transduced NK-92,
NK.αmeso, and NK.αmeso.mbIL15 cells, each either with or without iCAS9 transgene. None of the cells
without iCAS9 displayed any sensitivity to the CID after 24 h coincubation, while cells expressing iCAS9
showed markedly decreased viability after CID exposure (P < 0.001) in both trypan blue exclusion and
AnnexinV/7-AAD staining (Figure 3.3b). These experiments demonstrate that if CAR-NK therapy needs
to be stopped, either due to unwanted toxicity or simply at the end of treatment, CID can be administered
and quickly trigger apoptosis only in cells expressing the iCAS9 transgene. However, a small percentage
of cells remained viable after CID treatment, likely due to iCAS9
−
NK cells which remained in the sorted
population or which gradually lost iCAS9 expression over time. This emphasizes the importance of
complete iCAS9 expression if the cells are to be used in a clinical setting; these issues can be addressed by
obtaining the therapeutic cell product from a single cell clone and carefully monitoring for transgene
downregulation or silencing prior to treatment.

73

Figure 3.3 iCas9 expression on NK-92 cells and viability of iCas9-expressing CAR.NK cells in the presence of
CID.
(a) iCAS9 expression of sorted NK cell lines. (b) Viability of NK cells, analyzed by both trypan blue staining and 7-
AAD/AnnexinV flow cytometry, after incubation in either plain media or 10 nM CID for 24 h.

To evaluate the antitumor efficacy of NK.αmeso.mbIL15 cells in vivo, 10 million cells were
injected intravenously into NSG mice with established SKOV.meso tumors twice a week for 2 weeks. The
data demonstrate that NK.αmeso.mbIL15 cells slowed tumor growth and exhibited better tumor growth
control than NK.αmeso cells (Figure 3.4). The treatment with NK.αmeso cells did not show better antitumor
efficacy than that of NK-92 or PBS only. This was unexpected given the cytokine secretion and cytotoxicity
of NK.αmeso cells against target cells in vitro, but immune cell exhaustion and lack of persistence in the
TME is not faithfully replicated in in vitro conditions.

74

Figure 3.4 Evaluation of tumor volumes of treated mice.
Tumor volume of mice treated with PBS, NK cells, NK.αmeso cells or NK.αmeso.mbIL15 cells were measured.
Representative measurements from Day 9 and Day 20 are presented as a bar graph.

Infused NK.αmeso.mbIL15 cells proliferated more in vivo compared with NK.αmeso cells. Tumor-
bearing mice were separated into two groups receiving NK.αmeso.mbIL15 cells and NK.αmeso treatment,
respectively, on days 0, 5, and 9. The mice were tail bled three times a week following the first NK injection.
By day 12, the group receiving NK.αmeso.mbIL15 cells had significantly higher NK cell populations in
tail blood (Figure 3.5a). One week after the last injection of NK cells, the mice were euthanized, and organs
were harvested and analyzed for NK cell populations. Interestingly, the group receiving NK.αmeso.mbIL15
cells treatment exhibited significantly higher populations of NK cells homing to the bone marrow, while
NK.αmeso cells had few NK cells in the bone marrow. Representative flow cytometry plots are shown in
Figure 3.5b, and quantitative data are displayed in Figure 3.5c NK-92 cells used for adoptive cell therapy
has not been reported in the literature to be able to home to bone marrow, making NK-92 cells expressing
membrane-bound IL-15/IL-15Rα a potential therapy for malignancies which heavily metastasize to the
bone marrow
120
.

75

Figure 3.5 Evaluation of NK cells in tail blood and bone marrow.
(a) Percentage of NK cells in mouse peripheral blood over time. (b) Representative FACS plots of NK cells in mouse
bone marrow 7 days after the final NK injection. (c) Percentage of NK cells in mouse bone marrow.

AP20187 injected into the mice induced apoptosis of NK cells expressing the iCAS9 transgene.
Tumor-bearing mice received three injections of 10 million NK.αmeso.mbIL15 cells as described above on
days 0, 5, and 9. The mice were tail bled on days 2, 7, 12, and 16 and were analyzed for NK cell populations
in tail blood. One group of mice received 50 ug CID encapsulated in a previously reported drug-delivery
liposomal nanoparticle
464
through intravenous injection, and another group of mice received PBS-only
injections. The fold changes of the percentage of CD45
+
cells in tail blood on day 16 over the percentage
of CD45
+
cells in tail blood on day 12 were calculated for each mouse and averaged for each group. Mice
without any CID treatment had nearly doubled NK cell populations in tail blood while mice receiving CID
treatment had less than half of the NK cell populations remaining (Figure 3.6a). On day 16, the mice were
euthanized, and their organs were analyzed for NK cell populations. Mice treated with CID had minimal

76
NK cell populations in the bone marrow, while mice without CID treatment retained high levels of NK cell
populations that homed to the bone marrow (Figure 3.6b). Overall, the addition of an engineered cytokine
and suicide switch allows for the optimization of biologic functions in CAR-NK cells.

Figure 3.6 Evaluation of NK cells in tail blood and bone marrow after PBS or CID administration.
(a) Fold-change of NK cells in mouse peripheral blood comparing before and after CID or PBS administration. (b)
Percentages of NK cells in mouse bone marrow after CID or PBS administration.

3.5 Conclusions
CAR-engineered immune cells have been successful in treating a variety of hematologic cancers,
but that measure of clinical success in treating solid tumors is lacking. CARs are a clinically relevant
application of synthetic biology, and their modular design offers the potential to target a variety of diseases
simply by switching out the antigen-binding domain. CARs can be further improved by genetically
modifying human immune cells with additional engineered receptors. This research presents two more
synthetic molecules in addition to the anti-mesothelin CAR: a membrane-bound cytokine, IL-15/IL-15Rα,
and a suicide switch, iCAS9, for immune stimulation and safety mechanisms. Instead of autologous T cells,
which are used in most CAR studies, the NK-92 cell line has proven effective in executing CAR-mediated
antitumor functions in preclinical settings and provides a renewable, off-the-shelf source of engineered
cells. A homogenous, allogenic cellular product has the potential for more precise modulation of biologic
outputs than autologous cells, which vary widely in function from patient to patient. There are many more
applications of synthetic biology for CARs beyond the scope of this research letter, including logic

77
gates
439,465
and synthetic notch receptors
466
. CARs require fine-tuning through molecular engineering to
have the precision and persistence needed to eradicate solid tumors, and additional artificial receptors
modulating the immune environment and cell response are key to more effective CARs in the future.

78
4 Chapter 4: CAR T Cell-Platelet Complexation for Enhanced
Tumor Homing and Antitumor Efficacy in Solid Tumors
4.1 Abstract
There has been an explosion of preclinical and clinical efforts to use adoptive T cell transfer,
specifically chimeric antigen receptor (CAR)-engineered T cells, as an alternative to solve the therapeutic
limitations of conventional cancer therapies, such as tumor relapse and metastasis. There is a greater interest
in moving CAR T cell therapy into the treatment landscape for patients with solid tumors. However, clinical
trials with CAR T cells in solid cancers have been disappointing due in part to the inability of T cells to
efficiently traffic to tumor sites and penetrate physical barriers to reach to cancer cells. Combining CAR
T cell therapy with other strategies can provide an alternative to defeat the challenges that limit the success
of CAR T cells in solid tumor treatments. Inspired by the clinical success of CAR T cells as well as the
intrinsic tendency of platelets (PLTs) to interact with and assist T lymphocytes to penetrate into tumors by
degrading tumor basal membrane and therefore accumulate in tumor sites, we present here a cell-only
combinational “living-drug” strategy, CAR T cell-PLT complexes, to enhance the antitumor effects of CAR
T cells in their site of action. We established CAR T cell-PLT complexes by self-complexation via one day
co-incubation, which stay stable over 2 days. We demonstrated in vitro that complex formation between
CAR T cells and PLTs or platelet microparticles, which are released upon the activation of platelets, does
not interfere with the target specificity and effector functions of CAR T cells, and that CAR T cell invasion
is improved when in complexes with PLTs. Using an immunocompetent human lung cancer xenograft
model, we found that the intravenous injection of CAR T cell-PLT complexes significantly reduced the
tumor growth by effective accumulation of CAR T cells into tumors. Our findings suggest that CAR T cell-
PLT complexes can facilitate the enhanced trafficking and infiltration of CAR-engineered T cells to the
tumor site and improve the antitumor immune response, thereby potentially overcome the hurdles when
treating solid cancers with CAR T cell.

79
4.2 Introduction
After relying only on surgery, radiation therapy and chemotherapy for cancer treatment for many
years, the emergence of the forth pillar of cancer therapy, cancer immunotherapy, has attracted significant
excitement
2
. Specifically cell-based cancer immunotherapies have demonstrated outstanding clinical
success for the treatment of patients with advanced cancer
3
. Chimeric antigen receptor (CAR) T cells are
genetically engineered T cells that express a receptor which is specific to a cancer surface antigen for
increased cancer-targeted cytotoxic efficacy and reduced off-target toxicity
33
. With the exceptional success
of adoptive CAR T cell transfer for treating hematological malignancies
47
, four CD19-targeted CAR T cell
products, Kite’s Tecartus™ and Yescarta™, Novartis’ Kymriah™ and Breyanzi™, as well the most recent
and BCMA-targeted CAR T cell product, Abecma™ received approval from the US Food and Drug
Administration (FDA)
48–54
. However, the clinical studies with CAR T cell therapy for solid tumors has been
less promising due to non-optimal antitumor efficacy and non-durable clinical response
137,138
. The major
roadblocks in CAR T cell therapy against solid tumors are tumor antigen heterogeneity, trafficking and
infiltration into tumor tissue, and immunosuppressive tumor microenvironment
34
. Specifically, trafficking
and accumulation of infused CAR T cells into solid tumors remains a significant challenge and strategies
to overcome solid tumor barriers need to be investigated
34,144
.
To attract CAR T lymphocytes to solid tumors the following steps are necessary: arrest of T cells
from bloodstream, T cell rolling along the vascular wall through the adhesive interaction between T cells
and vascular endothelium, initiation of trans-endothelial migration (TEM), and subsequent chemokine
receptor engagement
467
. After the TEM is initiated by the engagement of cell adhesion molecules (CAMs)
by activated T cells, a secondary wave of more universal CAMs on endothelial cells are sensed by T cells,
which allows to reach the adhesion threshold needed for T cell capture from bloodstream
468
. Specifically,
the CAMs that are known to mediate T cell rolling on endothelial cells are L-selectin, E-selectin, and P-
selectin which are expressed on leukocytes, endothelial cells and platelets, respectively
469
.

80
While the role of platelets in blood clotting have been long appreciated
180
, the larger impact of the
platelets on immune response by the interaction they form with leukocytes
218
, is increasingly evident, but
perhaps still not fully appreciated. Activated platelets release lipid membrane vesicles called platelet-
derived-microparticles (PMPs)
194,216
and contribute to immune function by releasing soluble factors and
chemokines that recruit, localize, or activate immune cells
182,215
. It is important to note that PLT-lymphocyte
interactions in cancer immunotherapy framework would benefit from further research in order to determine
the impact of platelets on immune cells in the TME, including T cells
233
. Several recent studies have shown
that platelets bind to T cells via multiple receptors, including integrins (αIIbβ3), P-selectin (CD62P),
CD40L, and lymphocyte CD11b
217
. It has been also suggested that platelets can facilitate the recruitment
of lymphocytes at a site of inflammation or infection, which is known as a central step in T cell
trafficking
182,199,215,217,232,236
.
For infiltration, T cells attacking solid tumors need to actively degrade the main components of
sub-endothelial basement membrane, including heparan sulphate proteoglycans (HSPGs) by releasing
heparanase to access tumor cells
470
. It is recently shown that heparanase deficiency in in vitro-engineered
and cultured CAR-redirected T lymphocytes can restrict their antitumor activity in stroma-rich solid
tumors
471
. In addition, CAR T cells engineered to express heparanase showed improved capacity to degrade
the ECM, which enhanced tumor T cell infiltration and antitumor activity
471
. It should also be noted that
PLTs express heparanase
472
and activated platelets release acid hydrolases including heparanase from their
lysosomes
473
.
Poor homing of CAR T cells to tumors is a major impediment to effective T cell therapy
474
.
Intravenous administration of CAR T cells alone in patients with solid tumors gives rise to suboptimal
results. An alternate strategy, intra-tumoral administration of CAR T cells is shown to elicit a desired
response with increased bioavailability of the cells; however this approach is invasive and cannot be used
in all organs/all parts of the body, and thus has limited applicability. Here, we present another strategy,
CAR T cells in complex with PLTs to achieve improved antitumor efficacy in CAR T treatment of solid
tumors. We speculate that with the presence of PLTs in close vicinity to T cells, the ineffective T cell–

81
cancer endothelial cell interaction can be transformed. The T cell-PLT interaction can aid in the capture of
T cells from bloodstream and heparanase introduced by PLTs can assist T cell infiltration into the tumor
stroma by HSPG degradation, which would enhance CAR T cell accumulation in tumors.
As an attempt to overcome the limited CAR T cell trafficking and infiltration into tumor tissue, we
present here a novel approach, CAR T cell-PLT complexes, that uses the intrinsic tendency of PLTs to
interact and capture intravenously injected T cells from bloodstream, and the ability of PLTs to express and
release HSPG-degrading enzyme heparanase to disintegrate the basal membrane of tumor tissues.
Therefore, PLTs in CAR T-PLT complexes can assist CAR T cells to accumulate in tumor sites where they
exhibit their target-specific cytotoxic effect and improve the therapeutic effect of CAR T cells in solid
tumors.
4.3 Methods
4.3.1 Cell lines
The human ovarian cancer cell line SKOV3 (ATCC HTB-77) and human embryonic kidney cell
line HEK 293T (ATCC CRL-3216) were purchased and authenticated from ATCC. The lung cancer line
NCI-H292 was kindly provided by Dr. Ite Laird-Offringa (University of Southern California, Los Angeles,
CA) and was used without further authentication. The H292.CD19 and SKOV3.CD19 cell lines were
generated by the transduction of parental NCI-H292 and SKOV3 cells with a lentiviral FUW plasmid
encoding the cDNA of human CD19. CD19
+
cells were then sorted to yield a population of CD19
overexpressing cells, with fluorescence-activated cell sorting (FACS)
475
.
All cells were maintained at 37°C in a 5% CO 2 environment and were routinely tested for potential
mycoplasma contamination using the MycoSensor qPCR Assay Kit (Agilent Technologies, Santa Clara
CA). HEK 293T cell culture was maintained in DMEM supplemented with 10% FBS, 2mM L-glutamine,
penicillin (100 U/mL) and streptomycin (100 μg/mL). Parental and CD19-expressing SKOV3 and H292
cells were maintained in RPMI-1640 with 10% FBS, 2mM L-glutamine, 10mM HEPES, penicillin (100

82
U/mL), and streptomycin (100 μg/mL). All above cell culture reagents were purchased from Hyclone
(Logan, UT). Human peripheral blood mononuclear cells (PBMCs) were cultured in T cell media (TCM),
which is composed of X-Vivo15 serum-free media (Lonza, Allendale NJ), 5% (v/v) GemCell human serum
antibody AB (Gemini Bio-Products, West Sacramento CA), 1% (v/v) Glutamax- 100X (Gibco Life
Technologies, Grand Island NY), 10mM HEPES buffer (Corning, Corning NY), 1% (v/v)
penicillin/streptomycin (Corning, Corning NY) and 12.25mM N-Acetyl- L-cysteine (Sigma-Aldrich, St.
Louis MO).
4.3.2 Plasmid construction
The lentiviral vector encoding the same CD19 CAR sequence was constructed based on the CD19
CAR previously reported
133
. The sequence consisted of CD19 single-chain fragment variable (scFv)
sequence, followed by the human CD8 hinge region and the intracellular domain of human CD3ζ was
inserted downstream of the human ubiquitin-C promoter in the lentiviral plasmid pFUW using Gibson
assembly
475
. The retroviral vector encoding anti-CD19 chimeric antigen receptor (CAR) was constructed
on the basis of the MP71 retroviral vector kindly provided by Prof. Wolfgang Uckert, as described
previously
476
. The CAR constructs consisted of anti-CD19 scFv antigen-binding domain; CD8 hinge and
CD28 transmembrane regions; and CD28, 4-1BB and CD3ζ cytoplasmic regions in the retroviral MP71
vector. The cDNAs of CD19 CAR and truncated EGFR (tEGFR) linked by a 2A sequence were inserted
into the MP71 vector to yield the retroviral vectors for making CART.tEGFR cells.
4.3.3 Lentiviral and retroviral vector preparation and T cell transduction
Lentiviral vectors were prepared by transient transfection of HEK 293T cells using a standard
calcium phosphate precipitation protocol
475
. HEK 293T cells were seeded at 18 x 10
6
cells per 15 cm tissue
culture dish (BD Biosciences, San Jose CA) and were transiently transfected with 40μg of the lentiviral
backbone plasmid, 20μg of the VSV-G-encoding envelope plasmid, and 20μg of the packaging plasmids
pMDLg/pRRE and pRSV-Rev. Fresh media (high glucose Dulbecco’s modified Eagle’s medium

83
supplemented with 10% FBS and 1% pen-strep) was plated onto the cells 4h after initial transfection. The
viral supernatants were harvested 48 hours post-transfection and filtered through a 0.45μm filter (Corning,
Corning, NY). The supernatant was then ultra-centrifugated at 25,000 rpm for 90 minutes, using an Optima
L-90 K preparative ultracentrifuge and an SW28 rotor. The pellets were resuspended in an appropriate
volume of cold Hank’s balanced salt solution (Hyclone) and frozen at – 80°C until later use.
SKOV3 and H292 cells were transduced to express human CD19 via corresponding lentiviral
vector. 1 x 10
6
cells/well were added to concentrated virus in 24-well plates and centrifuged at 1050 x g for
90min at room temperature. CD19
+
cells were sorted using fluorescence-activated cell sorting and cultured
as described for SKOV3 and H292 cells.
Similarly, retroviral vectors were prepared by transient transfection of HEK 293T cells using the
same standard calcium phosphate precipitation protocol. Confluent HEK 293T cells cultured in 15cm tissue
culture dishes were transfected with 37.5μg of the retroviral backbone plasmid, along with 18.75μg of the
envelope plasmid pRD114 and 30μg of the packaging plasmid encoding gag-pol.
Activated human T cells were transduced and expanded as previously described
477
. Briefly, human
peripheral blood mononuclear cells (PBMCs) from healthy donors were thawed and cultured in T cell
media. The culture was supplemented with human IL-2 (10 ng/mL; Peprotech, Rocky Hill, NJ). Thawed
PBMCs were rested overnight. Next, PBMCs were activated and expanded using Dynabeads® human T-
expander CD3/CD28 (Invitrogen, Carlsbad CA) at a bead:PBMC ratio of 3:1. Activated PBMCs were
transduced with viral vectors 48 hours after activation.
For the transduction of activated human T cells, non-tissue culture-treated 12-well plates were first
coated overnight with 25μg RetroNectin per well (Clontech Laboratories, Mountain View, CA). Retroviral
vectors were subsequently spin-loaded onto the plates by centrifuging at 2000 x g for 2h at 32°C. Activated
T cells were resuspended at a concentration of 5 x 10
5
cells/mL with fresh TCM completed with
recombinant human IL-2 (10 ng/mL; Peprotech, Rocky Hill, NJ) and added to the vector-coated plates. The
plates were centrifuged at 600 x g for 30 min at 32°C. Cell were then incubated at 37°C and 5% CO 2.
During ex vivo expansion, culture media was supplemented with 10 ng/mL recombinant human IL-2 and

84
replenished every 2 days. Cell density was maintained between 0.5 x 10
6
to 1 x 10
6
cells/mL for optimal T
cell growth. CAR T cell cultures were expanded up to 10 days before freezing.
4.3.4 CAR detection on T cell surface
To detect CAR expression on the cell surface 3-5 days after transduction, retrovirus transduced T
cells (1 x 10
6
) were stained with APC-conjugated anti-hEGFR (Biolegend) for tEGFR detection. Non-
transduced cells served as negative control. Fluorescence was assessed using a MACSquant cytometer
(Miltenyi Biotec, San Diego, CA), and all the FACS data were analyzed using FlowJo software version
9.3.2 (Tree Star, Ashland, OR).
4.3.5 Freezing and thawing PLTs
We received the human PLTs from the vendors in the PLT-rich plasma. Fresh platelets (PLTs) can
only be stored in room temperature for about 6h, according to the vendor’s descriptions (AllCells, Alameda,
CA). After receiving the human platelets, we froze the platelets and store them in -80°C, as described
previously
188
. Briefly, we used the same concentration of cells for freezing as mentioned in the article,
specifically 4 x 10
11
PLTs in 400-600mL PLT freezing solution. We first prepared the PLT freezing solution
containing 27% DMSO in 0.9% NaCl solution. We added the freezing solution into the PLTs which were
suspended in platelet rich plasma (PRP) (as cells comes in PRP upon purchase) to achieve a final
concentration of 6% DMSO. Cells were span down at 1250 x g for 10min. Most of the supernatant was
removed. PLTs were aliquoted in vials for desired cell count, frozen in this residual amount of DMSO and
stored in -80°C up to 2 years. For PLT thawing, frozen vials were gently thawed in 37°C water bath and
diluted with drop-wise addition of 0.9% NaCl solution at 1:1 volume ratio. PLTs were centrifuged at 1250
x g for 10min. Supernatant was removed and cells were resuspended in pH 7.4 Tyrode’s buffer (134 mM
NaCl, 12 mM NaHCO 3, 2.9 mM KCl, 1mM MgCl 2, 10 mM HEPES, 0.34 mM Na 2HPO 4). In this study, we
froze PLT at 0.5 x 10
9
cells per vial, and yielded approximately 75% cell viability after freeze and thaw.
The lifespan of these cryopreserved human PLTs is 7 days. It was demonstrated in this previous work that

85
the allogeneic human PLTs collected from stable thrombocytopenic patients, when frozen with this method,
could be stored at −80°C for at least 2 years with ~75% recovery values in vitro and ~40% recovery after
transfusions
188
.
4.3.6 PLT activation and PMP formation
Platelet microparticles (PMPs) were prepared from PLTs as previously described
478,479
.

PLTs were
activated by Thrombin (2 U/mL) in 2.5mM CaCl 2 solution for at 37°C with occasional shaking. After
incubation, cells were centrifuged at 800 x g for 20min, and PMP enriched supernatants were collected to
be used in the following steps. Similarly, the pellet which was containing activated PLTs was suspended in
TCM for the next steps.
4.3.7 CAR T-PLT/PMP complex formation
For all cocultures, we used retrovirus transduced T cells expressing anti-CD19 CAR and tEGFR
and (CART.tEGFR). To form CAR T-PLT or CAR T-PMP complexes, CART.tEGFR cells were first
incubated with PLTs or PMPs for 24h. These complexes were then used for further in vitro tests and in
vivo injections. To determine the optimal CAR T:PLT/PMP ratio for complex formation, CAR-engineered
T cells were either cocultured with PLTs or PMPs in different ratios. We cocultured the same number of T
cells with different ratios of PLTs and PMPs and formed CAR T-PLT and CAR T-PMP complexes at two
T cell-to-PLT/PMP ratios; 1:20 and 1:80. We used activated PLTs in CAR T-PLT complexes. Similarly,
in CAR T-PMP complexes, we used PMPs formed by the activation of PLTs. Although we did not
determine the number of PMPs in PMP suspensions after PLT activation, in CAR T-PMP complexes, we
used PMP suspensions which is collected from the activation of the same number of PLTs which is used in
CAR T-PLT complexes at the same T cell-to-PLT/PMP ratio.
For complex formation, T cells (0.2 x10
4
cells/well) were mixed with different ratios of PLTs or
PMPs in T cell media supplemented with IL-2 in 96-well plates. CART.tEGFR-PLT/PMP cocultures were
incubated for 6 h, 1 day or 2 days in 37°C incubator. Following the coculture of PLTs/PMPs with CAR T

86
cells, to remove unbound PLTs or PMPs, CAR T cell-PLT/PMP cocultures were washed with PBS. The
complex formation efficiency and stability were examined using immunostaining and flow cytometry. CAR
T-PLT/PMP complexes were stained for T cell markers and PLT/PMP markers. PE/Cy5-conjugated anti-
human CD3 and Pacific Blue-conjugated anti-human CD8 were used as T cell markers, whereas FITC-
conjugated anti-human CD41 was used as the PLT/PMP marker. All antibodies were purchased from
BioLegend. Non-transduced (NT) T cells, CAR T cells and PMP/PLT only groups were used as controls.
4.3.8 Cytokine release assay
T cells (0.1 x x10
6
cells/well) were co-incubated with target cells in round bottom 96-well plates at
a 1:1 ratio for 6 h at 37°C and 5% CO 2 with GolgiPlug (BD Biosciences, San Jose, CA). Cytokine release
of non-transduced T cells, CAR T cells, and CAR T cells in complex with PLT (CAR T-PLT) or PMPs
(CAR T-PMP) was compared. Mostly, cytokines are secreted proteins and for their analysis they must
first be trapped inside the cell by using a protein transport inhibitor. BD GolgiPlug (brefeldin) works with
accumulation of protein at the Golgi complex by inhibiting the protein transport via redistribution of
intracellularly produced proteins from the cis/medial Golgi complex to the endoplasmic reticulum
480
. At
the end of the incubation, Cytofix/Cytoperm Fixation and Permeabilization Kit (BD Biosciences) was used
to permeabilize the cell membrane and perform intracellular staining according to the manufacturer’s
instructions. Cells were stained for CD3, CD8, CD41 and IFN-γ using PE/Cy5-conjugated anti-human CD3,
Pacific Blue-conjugated anti-human CD8, FITC-conjugated anti-human CD41 and PE-conjugated anti-
human IFN-γ, respectively. All antibodies were purchased from BioLegend. Unstimulated cells served as
a negative control. CAR T cells stimulated with soluble OKT3/CD28 antibodies were used as a positive
control. Results were read using flow cytometry. The data were determined in triplicates and presented as
mean ± SD.

87
4.3.9 Degranulation assay
0.2 x 10
6
T cells were cocultured with target cells at a ratio of 1:1 for 6 h at 37°C and 5% CO 2 with
GolgiStop (BD Biosciences) at 0.67 µL/mL final concentration and FITC anti-CD107a antibody in R10 in
96-well round-bottom plates. BD GolgiStop (monensin) prevents protein secretion by accumulation of
protein at the endoplasmic reticulum (ER) stage, by interacting with the Golgi transmembrane
Na
+2
/H
+
transport
480
. Degranulation in non-transduced T cells, CAR T cells, and CAR T cells in complex
with PLTs (CAR T-PLT) or PMPs (CAR T-PMP) was compared. PE/Cy5 anti-CD3 antibody and Pacific
Blue anti-CD8 antibody were used as T cell surface markers, and FITC anti-CD107a antibody was used as
the degranulation marker. All antibodies were purchased from BioLegend. CAR T cells stimulated with
soluble OKT3/CD28 antibodies were used as positive control. NT cells cocultured with target cells were
used as negative controls for FACS gating.
4.3.10 Transmigration assay
T cell transmigration assays were performed in 24-mm diameter, 3-μm pore size Transwell plates
(Costar). Non-transduced T cells, CAR T cells or CAR T cells in complex with PLT or PMPs (0.5 x 10
6
T
cells/well at 1:80 CAR T:PLT/PMP ratio) were plated on the upper wells in 0.1 mL TCM (without IL-2
supplementation). 0.5 mL TCM was added to the lower wells. The T cell chemoattractant CXCL9 (100
ng/mL; Peprotech, Rocky Hill NJ) was added to the lower wells. Control groups did not contain the
chemoattractant whereas others contained CXCL9. After incubation at 37°C, the number of T cells which
migrated into the lower chamber was counted under light microscope at 2 h and 6 h time points. Percent
transmigration was calculated as (number of migrated cells in lower chamber of Transwell plates for that
condition) / (initial cell number in upper chamber before assay begin) x 100.
4.3.11 Trans-Matrigel cell migration assay
T cell Trans-Matrigel migration was studied in 24-mm diameter, 8-μm pore size Transwell plates
(Costar) covered with standard Matrigel (STD MG) (Corning) or Growth Factor Reduced Matrigel (GFR

88
MG) (Corning). Non-Matrigel coated wells were used as control groups. Matrigel was first diluted to
4mg/mL (according to the manufacturer’s instructions) in serum free Ex-vivo medium. Transwell plate
inserts were covered with STD MG or GFR MG overnight. 50µL of 4 mg/mL Matrigel solution was used
to cover each insert, yielding at 200mg Matrigel coverage per insert. The excess unsolidified liquid was
removed from the Matrigel-covered inserts. Lower wells of Transwell plates were filled with 0.5 mL TCM
(without IL-2 supplementation) with CXCL9 chemoattractant (100 ng/mL). In upper wells, 0.5 x 10
6
CAR
T cells and CAR T-PLT/PMP complexes at a ratio of 1:80 were suspended in CXCL9-free TCM and
introduced to the non-coated or Matrigel-coated inserts. CXCL9 chemoattractant was used in all lower
wells. T cells were incubated at 37°C for 36h and Matrigel-invaded cell in the lower wells were counted
under microscope at 8 h, 20 h, and 36 h time points. Percent Trans-Matrigel migration was calculated as
(number of cells invading through the Matrigel coated chamber membrane for that condition) / (number of
cells migrating through the uncoated insert membrane) x 100.
4.3.12 Cytotoxicity assay
Lysis of target cells was measured by a method as previously described
425
. Both parental SKOV3
and H292 cells as well as CD19-overexpressing SKOV3.meso and H292.meso target cells were
fluorescently labeled by suspending them in PBS+0.1% BSA with 5 mM of fluorescent dye CFSE (Life
Technologies, Waltham, MA) at a concentration of 1 x 10
6
cells/mL. The cells were incubated for 30 min
at 37°C with occasional shaking. To stop labeling reaction, the same volume of FBS was added into the
cell suspension and then incubated for 2 min at room temperature. The cells were then washed twice and
suspended in fresh R10 medium. CAR T-PLT and CAR T-PMP complexes were used in target cell-T cell
cocultures. Cocultures were set up in round bottom 96-well plates in triplicates at the following effector-
to-target (E:T) ratios: 1:1, 5:1, 10:1 and 20:1. Each well contained 2 x 10
4
cancer cells. The T cell-cancer
cell cocultures were incubated at 37°C for 6 h and 18 h, with SKOV3/ SKOV3.CD19 and H292/
H292.CD19 cells, respectively. After incubation, the suspended cells were directly harvested, whereas the
attached cells were obtained by trypsinization. All the cells were stained with 7-AAD (Life Technologies)

89
in PBS (1:1000 dilution) for 10 min at room temperature. Cytotoxicity of non-transduced T cells, CAR T
cells, and CAR T cells in complex with PLTs (CAR T-PLT) or PMPs (CAR T-PMP) was compared. The
fluorescence was analyzed by flow cytometry. Percentages of killed cells were calculated as CFSE
+
7-AAD
+

cells / (CFSE
+
7-AAD
-
+ CFSE
+
7-AAD
+
) cells, with live/dead gates based on control wells of target cells
to account for spontaneous cell death. The statistics were presented in mean ± SD.
4.3.13 In vivo antitumor activity
All animal experiments will be conducted according to the guidelines set by the NIH and the animal
protocol approved by the University of Southern California Institutional Animal Care and Use Committee
(IACUC). All mice will be hold under specific pathogen-reduced conditions in the University of Southern
California animal facility (Los Angeles, CA, USA). Six-to-eight-week-old female NOD.Cg-
PrkdcscidIL2Rγtm1Wj1/SZ (NSG) mice (Jackson Laboratories, (Bar Harbor, ME, USA) will be used for
all in vivo experiments.
To establish tumors, NSG mice were inoculated subcutaneously with 2 x 10
6
H292.CD19 cells.
After inoculation, mice were observed every day and tumor size was measured using a fine caliper. Tumor
volume was calculated as 1/2 (length) x (width)
2
. As tumor size reached to 40-70 mm
3
, all mice were
randomized into tumor size rank matched cohorts and assigned into 5 groups (n = 5 per treatment group).
Then, treatment formulations were intravenously injected through tail vein in the following groups: Group
1: PLT; Group 2: non-transduced (NT) T cells (NT); Group 3: aCD19.CAR T cells (CAR T); Group 4: NT
T cell-PLT (NT-PLT); Group 5: aCD19.CAR T cell-PLT (CAR T-PLT). Adoptive cell treatments were
administered only once. For all treatment groups, 3 x 10
6
non-transduced T cells or CAR
+
T cells were used
per group. For T cell treatment groups with PLTs (Group 3 and 5), 3 x 10
6
T cells were co-incubated with
240 x 10
6
PLTs corresponding to 1:80 CAR T-to-PLT ratio for one day to establish CAR T-PLT complexes
with PLTs. Physical states of the mice were monitored daily, and tumor growth and body weight of the
mice were measured and recorded every other day until the sacrifice point. Change in body weight was
represented as percent initial weight and calculated as the average of (body weight of individual mouse at

90
measurement day) / (pre-therapy body weight of individual mouse) x 100, per treatment group. Mice were
euthanized when they display obvious weight loss, tumors start to ulcerate, or tumor size reached 1,000
mm
3
and more. The survival rates were shown in Kaplan−Meier curves. The survival curves of individual
groups were compared by a log-rank test.
4.3.14 Ex vivo tissue analysis
To establish tumors, NSG mice were inoculated subcutaneously with 2 x 10
6
H292.CD19 cells.
After inoculation, mice were observed every day and tumor size was measured using a fine caliper. Tumor
volume was calculated as 1/2 (length) x (width)
2
. As tumor size reached to 40-70 mm
3
, all mice were
randomized into tumor size rank matched cohorts and assigned into 2 groups (n = 3 per treatment group).
Then, treatment formulations were intravenously injected through tail vein in the following groups: Group
1: aCD19.CAR T cells (CAR T) and Group 2: aCD19.CAR T cell-PLT (CAR T-PLT). Adoptive cell
treatments were administered only once. For all treatment groups, 3 x 10
6
CAR
+
T cells were used per
group. For CAR T-PLT treatment groups, 3 x 10
6
T cells were co-incubated with 240 x 10
6
PLTs
corresponding to 1:80 CAR T-to-PLT ratio for one day to establish CAR T-PLT complexes. Physical states
of the mice were monitored daily until sacrifice day for ex vivo analyses.
12 days after treatment, all mice were sacrificed. The blood, bone marrow, spleen and tumor
samples were collected for ex vivo analyses. Tumors from each mouse were harvested and weighed. To
obtain single cell suspensions for further analysis, each tissue sample were processed by manual mincing
and filtering them through a 70 μm nylon mesh to obtain a single-cell suspension. Red blood cells in the
samples were lysed in a lysis buffer (155 mM NH 4Cl, 12 mM NaHCO 3, and 0.1 mM EDTA in MilliQ
water) at room temperature for 10 minutes and single cell suspension were washed with PBS once. Cells
were centrifuged at 2000 x g for 5 minutes at 4°C, counted and resuspended in PBS for further staining and
FACS analysis. Fluorescent-conjugated antibodies (Biolegend) were used to stain the samples in 100 μL
PBS at a 1:50 volume ratio for 15 min at 4°C in a 96-well plate. Samples were analyzed using flow
cytometry in triplicates and results were presented as the mean ± SD. For the staining of tumor and blood

91
samples; PE/Cy5 anti-CD3, Pacific Blue anti-CD8, FITC anti-CD41, and APC/Cy7 anti-CD4 antibodies
were used for immunostaining. For the staining of spleen and bone marrow samples; PE/Cy5 anti-CD3,
Pacific Blue anti-CD8, FITC anti-CD4, and PE anti-CD45 antibodies were used for immunostaining.
4.4 Results
4.4.1 Anti-CD19 CAR is expressed in human T cells
The previously constructed third-generation CAR which targets human CD19 and expresses
tEGFR (for labeling purposes) was used to transduce activated T cells
477
. The retroviral vector MP71
encoding anti-CD19 CAR was generously provided by Dr. Wolfgang Uckert
476
. The CAR construct is
composed of a scFv-based anti-CD19 region which is linked to CD8 hinge and CD28 transmembrane
domains, intracellular CD28 and 4-1BB costimulatory domains, and intracellular CD3ζ signaling domain.
The cDNAs of CD19 CAR and truncated EGFR (tEGFR) linked by a 2A sequence were inserted into the
MP71 vector (Figure 4.1b). The corresponding CAR is designated as CART.tEGFR.
Human peripheral blood mononuclear cells (PBMCs) were activated by CD3/CD28 beads for 2
days. Activated human PBMCs were transduced with retroviral vector MP71 encoding aCD19 expressing
CAR. To verify CAR expression, transduced T cells were stained with APC-conjugated anti-hEGFR for
tEGFR detection. FACS analysis of surface CAR expression showed 76.5% transduction efficiency
(representative data shown in Figure 4.1c), indicating the efficient expression of CD19-specific CARs in
human T cells. Similarly, H292.CD19 and SKOV3.CD19 cells, which stably overexpress human CD19
antigen, were generated by transducing parental NCI-H292 and SKOV3 cells with lentiviral vector
encoding the cDNA of human CD19
481
. As seen in Figure 4.1d, we did not detect CAR
+
signal in PLTs and
non-transduced T cells. We also tested PLT complexation by staining CAR T-PLT complexes for PLT
marker CD41. As expected, only CAR T-PLT complexes but not non-transduced cells and CAR T cells
alone showed CD41 presence. We showed that ~75% of CAR T cells in CAR T-PLT group were in complex
with PLTs after co-incubation and wash steps.

92

Figure 4.1 Rationale design of CAR T cell-PLT complexes to improve CAR T cell homing into solid tumors and
evaluation of CAR expression on T cells.
(a) Schematics of the rationale design of CAR T cell-PLT complexes to improve CAR T cell homing into solid tumors.
PLT-assisted trafficking of CAR T cells to tumor sites can be achieved by adoptive transfer of CAR T-PLT complexes,
which can potentially increase the antitumor efficacy of CAR T cell treatments in solid tumors. (b) CAR construct
targeting human anti-CD19 antigen. The human anti-CD19 single-chain variable fragment (scFv) was cloned into an
MP71 retroviral expression vector containing a CD8α transmembrane-CD28-4-1BB-CD3ζ signaling motif followed by
a 2A linker and truncated EGFR (tEGFR). The corresponding construct encoded the anti-CD19 CAR.tEGFR. (c) Stable
anti-CD19 CAR expression on T cell surface. Human peripheral blood mononuclear cells (PBMCs) were activated and
transduced with retroviral vectors encoding aCD19.CAR.tEGFR. Retrovirus transduced T cells were stained with APC-
conjugated anti-hEGFR for tEGFR detection. Surface anti-CD19 CAR.tEGFR expression was evaluated by flow
cytometry. Non-transduced (NT) T cells were used as a negative control. Representative flow cytometry data was shown.
(d) PLT marker CD41 expression in CAR T-PLT complexes. Anti-CD19 CAR T cells were cocultured with PLTs at 1:80
CAR T:PLT ratios for 1 day. Representative flow cytometry analysis demonstrating CD41
+
expression in CAR T-PLT
complexes, and not in PLT-free NT or CAR T cells was exhibited.

4.4.2 PLTs and PMPs form stable complexes with CAR T cells
As described in detail in Methods section, human platelets (PLTs) were purchased from ATCC and
frozen upon arrival. It has been known that frozen PLTs induce production of more platelet microparticles
(PMPs) due to the physical stress caused by freeze-thaw
482
. Given the short lifetime of freshly isolated PLTs
in platelet rich plasma in room temperature (~ 6h), and to induce and collect more PMPs, we used PLTs
after freeze-thaw, in all experiments
482
. CaCl 2 is shown to promote PMP formation, along with the
activation by thrombin. To collect high number of PMPs, we activated PLTs by thrombin in a calcium

93
chloride (CaCl 2) solution
478
. Upon activation, PLTs form lipid membrane vesicles called platelet-derived
microparticles (PMPs)
182
. Following PLT activation and PMP formation, the PLT-PMP mixture was
centrifuged. PLTs in pellet and PMPs in supernatant were collected. The dynamic light scattering (DLS)
measured PMP size and size distribution results agree with the literature with particle size between 100-
500nm (data not shown)
201,209,212
.
In order to find the optimal duration for complex formation, T cells were cocultured with PLTs and
PMPs for different durations (6h, 1 day and 2 days). In addition, to determine the optimal CAR T-to-
PLT/PMP (CAR T:PLT/PMP) ratio for complex formation, we cocultured the same number of CAR T cells
at 1:20 and 1:80 CAR T:PLT/PMP ratios. The rationale behind the selection of these CAR T:PLT/PMP
ratios (1:20 and 1:80) is our theoretical calculation for the maximum number of PLTs that could cover CAR
T cell surface and the T lymphocyte:PLT ratios used in T cell-PLT cocultures in previous literature. By
assuming CAR T cell diameter as 8µm and activated PLT diameter as 2µm, and that all T cell surface is
covered with PLTs, we calculated that one CAR T cell could be covered with maximum 64 PLTs,
corresponding to a CAR T:PLT ratio of 1:64. Also, the 1:80 CAR T:PLT ratio takes into consideration the
possible loss of PLTs in CAR T cell-PLT complexes throughout the wash steps. In the literature, we found
one study showing that a platelet to CD4
+
T cell ratio of 15 to 1 or less in cocultures did not stimulate Tregs,
whereas 100 to 1 platelet:CD4
+
T cell ratio resulted in a significant difference
233
. In an earlier study, Treg
stimulation was achieved by a high platelet to CD4
+
T cell ratio (250 to 1) in cocultures
483
. The CAR
T:PLT/PMP ratios we use in this study, 1:20 and 1:80, are in agreement with the theoretical calculation and
in the same range of what was used in previous studies, yet not high enough to stimulate Tregs which could
cause immune suppression
233,483
.
The fact that platelet-derived TGFβ1 is able to stimulate Tregs only when platelets are present in
high numbers, is further discussed in a review paper
484
. It is important to mention that TGFβ1 is found in
platelet α-granules in relatively high concentrations
485
, which questions the effect of high PLT ratios in
cocultures. However, a likely explanation is that TGFβ1 is secreted by platelets into blood serum where it
is mostly present in its biologically inactive form
486
, indicating the necessity of certain cytokines to be in

94
their active form to be effective. Thus, we also want to emphasize in this study that the number of PMPs in
the CAR T-PMP cocultures is not the main concern for the purpose of this work. We used the PMPs
collected from the activation of the same number of activated PLTs that were used in CAR T-PLT
cocultures, to be able to make a scientific comparison between these groups, as also performed
previously
487
. However, it is not the number of PLTs or PMPs, yet the amount of activated surface
expressed receptors or released cytokines that creates certain outcomes in CAR T-PLT and CAR T-PMP
cocultures.
Among many antigens that have been used to identify PLTs and PMPs, we chose the most abundant
integrin Glycoprotein (GP) IIb/IIIa (also known as integrin α IIbβ 3 and CD41 antigen), as the marker of PLTs
and PMPs in our study
488
. Integrins are a family of surface adhesion molecules that are expressed on many
cell types and they mediate direct cell-cell and cell-matrix interactions
489
. PLTs express several integrins
which are formed as a heterodimer consisting of two chains as α- and β-chain. Mostly, PLTs express
integrins of the β 1- and β 3-subfamily, including α IIbβ 3
490
. Integrin α IIbβ 3 is the most abundant platelet
membrane receptor with around 80,000 copies per platelet
488
. It is an integrin complex which is also a
receptor for fibrinogen and von Willebrand factor, plays a major role in the regulation of platelet
aggregation and activation. On resting platelets, GPIIb/IIIa is maintained in its inactive form and serves as
a low-affinity adhesion receptor for fibrinogen. When activated, both the α- and β-subunits of GPIIb/IIIa
receptors change their conformation, and thus allow the binding to fibrinogen, von Willebrand factor,
thrombospondin, fibronectin, and vitronectin, leading to platelet aggregation and endothelial adherence
491
.
This change from low affinity conformation to high affinity one only occurs through the activation of
platelets
492
. With their ability to transduce activating signals into the cell, integrins initiate an intracellular
signaling cascade and trigger further platelet activation or degranulation upon ligand binding
490
.
The complex formation efficiency and stability were examined using immunostaining and flow
cytometry. CD8 and CD41 were used as the cytotoxic T cell and PLT/PMP markers, respectively. The
CD8
+
CD41
+
cell population in each coculture representing the cytotoxic T cells (CD8
+
) that are in complex
with PLT/PMPs (CD41
+
) were shown in Figure 4.2.

95

Figure 4.2 Evaluation of complex formation between CAR T cells and PLTs or PMPs.
(a) Percentage of T cells in complex with PLTs or PMPs. CAR-engineered T cells were cocultured with PLTs and
PMPs at 1:20 and 1:80 CAR T:PLT/PMP ratios for either 6h, 1 day or 2 days. Unbound PLTs or PMPs were washed
out and all groups were stained with anti-human anti-CD8 and anti-CD41 antibodies. CD41
+
PLTs in CD8
+
cytotoxic
T cell population were gated. Representative flow cytometry analysis of CAR T cells and CAR T-PLT/PMP
complexes were exhibited. (b) Bar graph of summarized statistics of triplicates for CD8
+
CD41
+
cells representing
CAR T-PLT/PMP complexes in CD8
+
T cells (n = 3; mean ± SD; ns: not significant; *p<0.05; **p<0.01; ***p<0.001;
****p<0.0001, two-way ANOVA with Tukey’s multiple comparison).

As expected, CAR T cells without PLT or PMP addition did not show CD41 expression. This group
served as negative control. Notably, the percentage of CD41 expression did not increase with CAR T-
PLT/PMP co-incubation duration in any group, suggesting that 6h co-culture duration was sufficient to
form stable complexes. In addition, the high percentage of CD41 expression in 2-day cocultures of CAR
T-PLT/PMP complexes demonstrates that complexes which are formed in 6 hours remain stable for 2 days.
The CD8
+
CD41
+
cell percentage in CAR T-PMP complexes was not significantly different from
CAR T cells alone for all three time points (Figure 4.2, ns). This indicates that either PMPs did not form
stable complexes with CAR T cells, or the formed complexes disintegrated during wash steps and PMPs
were washed out. Mainly, the CD8
+
CD41
+
cell percentage in CAR T-PLT cocultures were significantly
higher than CAR T-PMP cocultures at both CAR T:PLT/PMP ratios. For instance, the CD8
+
CD41
+
cell
percentage for all time points were approximately 45% and 80% for the CAR T-PLT complexes with 1:20
and 1:80 CAR T:PLT/PMP ratios, respectively; whereas the CD8
+
CD41
+
cells were between 5-13% and
around 15% for corresponding CAR T-PMP complexes (Figure 4.2). The non-proportional CD8
+
CD41
+

signal levels of 45% for CAR T-PLT complexes at 1:20, and 80% for CAR T-PLT complexes at 1:80 CAR

96
T:PLT/PMP ratio suggests that CAR T-PLT complex formation reached to a saturation point between these
two CAR T:PLT ratios. This saturation was expected based on the theoretical calculation of the maximum
number of PLTs that could cover the CAR T cell surface area, which is 1:64 CAR T cell-to-PLT ratio.
Similarly, there is no proportional increase in the percentage of CD8
+
CD41
+
cells with the increase in PMPs
in CAR T-PMP cocultures with 1:20 and 1:80 CAR T:PMP ratios, with 5-13% and 15% CD8
+
CD41
+
cells,
respectively (Figure 4.2), which can perhaps be explained by the weak binding of PMPs to T cells which
resulted in physical detachment of PMPs from CAR T cells likely after wash steps. This finding also
indicates that CAR T cell-PLT complexes are more resistant to mechanical detachment. Although integrins
and selectins are known for their role in the interaction between T cells and PLTs or PMPs
217
, to our
knowledge, there is no published research on the affinity of PLTs and PMPs to T cells or other lymphocytes,
which would benefit from further research.
To gain more insight on complex formation efficiency, we further calculated the percent ratio of
CAR T-PLT/PMP complexes (CD8
+
CD41
+
cells) within the CAR T cell population (CD8
+
cells) as
represented in Figure 4.2b. Not surprisingly, when compared to CAR T cells, all other groups demonstrated
significantly higher levels of CD8
+
CD41
+
to CD8
+
signal ratio, regardless of the coculture duration (Figure
4.2b, p<0.0001). Similarly, when CAR T-PLT complexes were compared for their complex formation
efficiency at 1:20 and 1:80 CAR T-to-PLT ratios, we observed significantly more complex formation with
the increased PLT number in the coculture, regardless of the coculture duration. With this finding and
considering ease in experiment design, we decided to continue the rest of the study with CAR T cell-
PLT/PMP complexes that are formed in 1 day cocultures.
We next tested whether complex formation with PLTs or PMPs could impact key cellular functions
of CAR T cells, such as cell cytokine secretion, degranulation, cytotoxicity, and migration.

97
4.4.3 IFN-𝜸 secretion of CAR T cells is unaffected by complex formation between CAR T
cells and PLTs/PMPs in vitro
To test if antigen-specific activation of anti-CD19 CAR T cells could be achieved when CAR T
cells are in complex with PLTs or PMPs, we performed an intracellular cytokine staining assay with the
CD19-overexpressing non-small cell lung cancer cell line H292.CD19.
CAR T cells were co-incubated with PLTs or PMPs for 1 day to form CAR T cell-PLT/PMP
complexes. Next, non-transduced (NT) T cells and CAR T cell-PLT/PMP complexes were cocultured with
target antigen expressing H292.CD19 cells with GolgiPlug inhibitors for 6 h and proinflammatory cytokine
interferon-gamma (IFN-𝛾) production was measured by intracellular staining and flow cytometry (Figure
4.3a). We also investigated the effect of CAR T cell-to-PLT/PMP ratio (1:20 and 1:80) on IFN-𝛾 secretion
of T cells.
aCD19-expressing CART.tEGFR T cells displayed reactivity against CD19-overexpressing
H292.CD19 cells. In the absence of stimulation, both NT T cells and CAR T cells displayed a low level
IFN-𝛾 secretion around 5000 MFI. In contrast, the expression of IFN-𝛾 was upregulated when CAR T cells
were stimulated by soluble anti-CD3/anti-CD28 antibodies, which served as positive control. As
anticipated, when cocultured with target antigen-expressing H292.CD19 cells, anti-CD19.CAR T cells
yielded significantly higher levels of IFN-𝛾 secretion, as opposed to NT T cells (Figure 4.3a, 6035 MFI for
NT T cells and 16615 MFI for CAR T cells, p<0.0001), demonstrating the necessity of an anti-CD19 CAR
to generate a target-specific immune response. Moreover, upon stimulation by H292.CD19 cells, CAR T
cell-PLT and CAR T cell-PMP complexes responded with a similar range of IFN-𝛾 release, regardless of
the PLT or PMP ratios that were used to establish the CAR T-PLT/PMP complexes. In addition, the
duration of cocultures (1-day vs 2-day coculture) for CAR T cell - PLT/PMP complex formation did not
have a significant effect on IFN-𝛾 secretion of T cells (data not shown). We proved that the complexation
of CAR T cells with either PLTs or PMPs does not interfere with the target antigen specific activation and
IFN-𝛾 release abilities of the CAR T cells.

98

Figure 4.3 Complexation of CAR T cells with PLTs and PMPs does not alter CAR T cell function.
(a) IFN-γ release assay. Non-transduced T cells (NT), CAR T cells and CAR T-PLT/PMP complexes at 1:20 and 1:80
CAR T:PLT/PMP ratios were cocultured with H292.CD19 cancer cells in the presence of GolgiPlug inhibitor for 6 h
to detect IFN-γ release by intracellular staining. Unstimulated CAR T cells were used as negative control, whereas
CAR T cells stimulated with soluble anti-CD3 and anti-CD28 antibodies were used as positive control. Mean
fluorescence intensity (MFI) levels for IFN-𝛾 were determined and the summarized statistics of triplicates were
represented in bar graphs (n = 3, mean ± SD; ns, not significant; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001,
one-way ANOVA with Tukey’s multiple comparison). (b) Degranulation assay. NT cells, CAR T cells and CAR T
cell-PLT or CAR T cell-PMP complexes at 1:80 CAR T:PLT ratio were cocultured with H292.CD19 cells at a 1:1
ratio for 4 h with GolgiStop inhibitor and stained for degranulation marker CD107a. Unstimulated CAR T cells were
used as negative control, whereas CAR T cells stimulated with soluble anti-CD3 and anti-CD28 antibodies were used
as positive control. The CD107a
+
cell population was gated in CD3
+
CD8
+
cell population. Summarized statistics of
CD107a
+
cell percentages within CD3
+
CD8
+
T cell populations in triplicates are shown in bar graphs (n = 3, mean ±
SD; ns, not significant; *p<0.05; **p< 0.01; ***p<0.001; ****p<0.0001, one-way ANOVA with Tukey’s multiple
comparison). (c) Cell migration assay. CAR T cells and CAR T cell-PLT or CAR T cell-PMP complexes at 1:80 CAR
T:PLT ratio were plated in the upper chambers of Transwell plates. Negative controls had plain media in the lower
chambers and CXCL9 was used as a chemoattractant in the lower chambers of non-control groups. After 2 and 6 hours
of incubation, media from the lower chambers were collected and CAR T cells were counted under light microscope.
Summarized statistics of triplicates are displayed in bar graphs (n = 3, mean ± SD; ns, not significant; *p<0.05; **p<
0.01; ***p<0.001; ****p<0.0001, two-way ANOVA with Tukey’s multiple comparison). (d) Trans-Matrigel cell
migration assay. Inserts in the upper chambers of Transwell plates were coated with Standard Matrigel (STD MG) or
Growth Factor Reduced Matrigel (GFR MG). NT cells, CAR T cells and CAR T cell-PLT or CAR T cell-PMP
complexes at 1:80 CAR T:PLT ratio were plated onto the MG coverage. The lower chambers were loaded with
CXCL9 added T cell medium. After 8h, 20h and 36 hours of incubation, media from the lower chambers were
collected and T cells were counted under light microscope. Summarized statistics are displayed in bar graphs (n = 3
samples per group, mean ± SD; ns: not significant; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001, two-way
ANOVA with Tukey’s multiple comparison).

Given that CAR T:PLT/PMP ratio did not alter CAR T cell therapeutic behavior (Figure 4.3a), and
that 1:80 CAR T:PLT/PMP ratio showed higher complex formation efficiency (Figure 4.2), we used CAR
T-PLT and CAR T-PMP complexes which are formed by co-incubating CAR T cells with PLTs or PMPs

99
at ratio of 1:80 for 1-day, as the most successful candidate to study the effect of PLTs/PMPs on CAR T
cells in vitro and in vivo effector functions in this study.
4.4.4 Degranulation of CAR T cells is unaffected by complex formation between CAR T
cells and PLTs/PMPs in vitro
To further assess the effector function of CAR T cell-PLT and CAR T cell-PMP complexes, we
tested the ability of CAR T cells to degranulate upon antigen-specific stimulation via degranulation assay
(Figure 4.3b). Non-transduced T (NT) cells and CAR T cell-PLT/PMP complexes were cocultured with
target antigen over-expressing H292.CD19 cells, and degranulation of the cytotoxic T cells were detected
by CD107a expression via flow cytometry. Unstimulated cells and cell complexes displayed CD107a
expression at levels close to zero with no significant difference between them. When co-incubated with
H292.CD19 cells, NT T cells showed similar percentage of degranulation marker expression as compared
to non-activated NT T cells (ns), indicating the necessity of target associated antigen (TAA) induced
activation. Upon stimulation by soluble anti-CD3/CD28 antibodies, CAR T cells exhibited above 40%
CD107a expression (serving as positive control), which was comparable to but lower than H292.CD19
activated group. In the presence of CD19-overexpressing H292 cells, all CAR T cell groups significantly
upregulated the expression of degranulation marker CD107a in contrast to NT T cells (Figure 4.3b,
p<0.0001). On the other hand, both CAR T-PLT and CAR T-PMP groups displayed similar levels of
reactivity against H292.CD19 cells compared to CAR T cells alone (ns). This finding was consistent with
the IFN-𝛾 expression data confirming that the functionality of CAR T cells was not affected negatively by
the PLT or PMPs that they formed complexes with.
4.4.5 Migration ability of CAR T cells towards a chemoattractant is unaffected by
complex formation between CAR T cells and PLTs/PMPs in vitro
We next investigated the effect of CAR T cell complexation with PLTs or PMPs on CAR T cell
migration towards a chemoattractant. CAR T cells and CAR T cells in complexes with PLTs (CAR T-PLT)

100
or PMPs (CAR T-PMP) in the upper chambers of Transwell plates migrated to the CXCL9 containing lower
chambers and we counted the migrated cell number at 2h and 6h time points. Both in 2h and 6h cell counts,
we showed impaired cell migration in the lack of the CXCL9, indicating a chemoattractant mediated cell
migration (Figure 4.3c). With no chemoattractant stimulus, CAR T cells in all groups migrated poorly to
lower wells, and in response to CXCL9. CAR T cell migration increased in response to time, as all groups
showed higher numbers of migrated cells at the end of 6h compared to 2h cell count. In either early or later
time points, there was no significant difference in the total number of migrated cells between CAR T cell
alone, CAR T-PLT complexes, and CAR T-PMP complexes (Figure 4.3c). Overall, complexes formed with
PLT or PMPs did not have a significant effect on CAR T cell migration towards a chemoattractant in a 2-
dimentional (2D) setting at any time point.
4.4.6 CAR T cell-PLT complexes enhance 3D Trans-Matrigel migration ability of CAR T
cells towards a chemoattractant in vitro
We next examined the motility of CAR T cells through a Matrigel (MG) layer via a 3-dimentional
(3D) Trans-Matrigel migration assay. As described in Methods section, inserts in the upper chamber of
Transwell plates were covered either with Standard Matrigel (STD MG) or Growth Factor Reduced
Matrigel (GFR MG). Matrigel (MG) is extracted from the Engelbreth-Holm-Swarm (EHS) mouse sarcoma,
a tumor rich in extracellular matrix proteins. The composition of isolated Corning STD MG is consisted of
around 60% laminin, 30% collagen IV, 8% nidogen, and heparan sulfate proteoglycans (HSPG)
493,494
. On
the other hand, Corning GFR MG was developed by the purification of STD MG to remove a variety of
growth factors. Specifically, the level of heparan sulfate proteoglycan in STD MG is reduced by 40-50%.
For the purposes of this study, GFR MG is designated as HSPG-reduced MG in the rest of this work.
It is important to note that PLTs express heparanase
472
and activated platelets release acid hydrolases
including heparanase from their lysosomes
473
. In one study, in which the role of PMPs in angiogenesis was
investigated, the stimulatory effect of PMPs on endothelial cell (EC) movement was demonstrated through

101
a Trans-Matrigel migration assay, in which the inhibition of heparanase markedly reduced the number of
invaded ECs
495
.
In this assay, we compared the 3D migration ability of CAR T cells alone and CAR T-PLT/PMP
complexes in STD MG with high HSPG composition and HSPG-reduced MG, by the additional effect of
heparanase introduced by PLTs or PMPs in CAR T-PLT/PMP complexes. 8h, 20h and 36h after cells
migrate through STD MG or HSPG-reduced MG covered inserts into CXCL9 added lower chambers, the
percentage of migrated T cells was quantified and possible time-dependent effect of PLTs and PMPs on
migration was also investigated. The migration durations used in this study are in accordance with the
literature
496
. Figure 4.3d represents the percent T cell invasion through STD MG and HSPG-reduced MG.
After 8h, all groups in STD MG cell migration assay displayed low levels of migration (< 5%
migration in all groups) with no significant difference (Figure 4.3d). In addition, non-transduced T cells
showed significantly lower levels of cell migration compared to all CAR T cell groups at all time points.
At 20h time point, the percentage of Trans-Matrigel migration of CAR T cells in CAR T cell alone and
CAR T-PLT complex groups were 16.03±2.91% and 19.63±4.07%, respectively, with no significant
difference. On the contrary, CAR T cell migration in CAR T-PMP group was significantly lower (9.64%)
compared to both CAR T cells (p<0.001) and CAR T-PLT complexes (p<0.0001). Within the last 16 hours
of the migration experiment, the percentage of CAR T cell invasion in CAR T cells alone did not
significantly change, whereas a significant increase was observed both in CAR T cell-PLT and CAR T cell-
PMP complexes (p<0.0001 for both groups). Hence, at the end of 36h, the percentage of CAR T cell
migration in CAR T cells alone and CAR T cell-PMP complexes were similar, with percent Trans-Matrigel
cell migration around 17.5%, whereas CAR T cell invasion in CAR T cell-PLT complexes was significantly
higher (31.83%, p<0.0001).
Similar to the cell invasion results in STD MG cell migration assay, NT T cell migration was
significantly lower compared to all other CAR T cell groups in HSPG-reduced MG environment (Figure
4.3d). At the end of the first 8 hours of the assay, only CAR T cell-PMP complexes displayed higher levels
of CAR T cell migration through HSPG-reduced MG (18.09 %, p<0.0001) compared to CAR T cells alone

102
and CAR T cell-PLT complexes (4.35% and 4.18%, respectively). In the following 12 hours, CAR T cells
in complexes with PMPs exhibited significantly increased cell migration through HSPG-reduced MG,
reaching to 30.96%, p<0.0001), whereas the percentage of 3D migration in CAR T cells alone and CAR T
cell-PLT complexes did not increase significantly. At the end of 36h observation, the percentage of cell
migration in HSPG-reduced MG was similar in CAR T cells alone and CAR T-PLT complexes with 16.71%
(p<0.01) and 15.19% (p<0.05), respectively; which was expected given the lower HSPG content in the
Matrigel, in which PLT-released heparanase did not create a cell invasion advantage through degradation
of HSPGs. Meanwhile, very interestingly PMPs in CAR T-PMP complexes facilitated increased migration
in HSPG-reduced MG (above 42%), compared to CAR T-PLT and CAR T cell groups. Although we do
not know the underlying mechanisms for this migration advantage, it is worth investigating in future studies.
The difference in the higher migration profiles of CAR T cell-PLT and CAR T cell-PMP complexes
compared to free CAR T cells in two different matrices points out that PLT and PMPs aid in T cell migration
by different mechanisms.
Overall, the cell migration of CAR T cells in CAR T-PLT complexes through STD MG was
significantly higher than other groups, indicating the degradation of HSPGs through the action of PLT-
introduced heparanase in the environment which created a cell migration advantage. STD MG represents
the true composition of tumor basal membranes. Thus, the contribution of PLTs in CAR T-PLT complexes
to the degradation of HSPGs in STD MG which enabled T cell movement in a 3D barrier presents a
promising strategy to improve T cell migration in tumor mass in vivo.
4.4.7 Cytotoxicity of CAR T cells is unaffected by complex formation between CAR T
cells and PLTs/PMPs in vitro
The ability of CAR T cells to trigger cytotoxicity against antigen-expressing target cells when in
complexes with PLTs or PMPs was tested by repeated cytotoxicity assays in which different cancer cell
lines and effector-to-target (E:T) ratios were tested. After establishing CAR T-PLT and CAR T-PMP
complexes at 1:80 CAR T:PLT/PMP ratio, we designed different T cell-cancer cell cocultures in which NT

103
T cells, CAR T cells and CAR T-PLT/PMP complexes were incubated with cancer cells expressing or non-
expressing the cognate antigen for specific CAR T cell activation. We compared the CAR T cell cytolytic
functions between H292-H292.CD19 and SKOV3-SKOV3.CD19 cell lines. In addition, we compared 1:1,
5:1, 10:1, and 20:1 E:T ratios to test dose dependent cancer cell killing. As seen in Figure 4.4a and Figure
4.4b, non-transduced T cells had no cytolytic activity against cancer cells. Similar to IFN-𝛾 secretion
results, CAR T cells showed low levels of cell killing when cocultured with target cells without the cognate
antigen (Figure 4.4). On the other hand, anti-CD19 CAR T cells co-incubated with the antigen-expressing
H292.CD19 and SKOV3.CD19 target cells demonstrated significantly greater percentages of cancer cell
killing (p<0.0001 for both groups), confirming that cytolysis is distinctly mediated through the engagement
of CD19 by CAR molecules on T cells. CAR T cells showed greater cytotoxicity towards H292.CD19 cells,
which reached to saturation at 5:1 E:T ratio, compared to SKOV3.CD19 cells (Figure 4.4b and Figure 4.4d),
with 87% vs 47% cytotoxicity at E:T of 20:1, respectively. In addition, cytotoxicity against both
SKOV3.CD19 and H292.CD19 cells significantly increased with increasing effector-to-target ratios.
CAR T cells in complex with PLTs or PMPs showed similar cytotoxicity profiles compared to
CAR T cells alone (Figure 4.4e and Figure 4.4f). For example, there is no significant difference in
H292.CD19 cell killing between CAR T cells and CAR T cell-PLT/PMP complexes, as well as among
CAR T-PLT and CAR T-PMP complexes at any E:T ratio (Figure 4.4f, ns). One exception is CAR T cell
groups cocultured with SKOV3.CD19 cells at 20:1 E:T ratio, as CAR T cells (46.87 ± 3.99%) showed
higher level of cytotoxicity compared to CAR T cell-PMP (42.40 ± 1.37%, p<0.01) and CAR T cell-PLT
groups (38.53 ± 1.45%), p < 0.0001 ).
Taken together, these findings indicated that CD19-specific CAR T cell effector functions were not
hindered by the complexation of CAR T cells with PLTs or PMPs.

104

Figure 4.4 Complexation of CAR T cells with PLTs and PMPs does not alter in vitro cytotoxicity of CAR T
cells.
(a-d) In vitro cell cytotoxicity assays. CAR T cells and CAR T cell-PLT/PMP complexes were either cocultured with
(a and c) SKOV3 and SKOV3.CD19 cells, and with (b and d) H292 and H292.CD19 cells for 6h, at different effector-
to-target (E:T) ratios of 1:1, 5:1, 10:1, or 20:1. Cytotoxicity of CAR T cells against target cells was measured. Non-
transduced T cells (NT) were used as negative controls. (c and d) Summary of percent cytotoxicity in triplicates was
represented in bar graphs (n = 3, mean ± SD; ns, not significant; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001,
two-way ANOVA with Tukey’s multiple comparison).

4.4.8 CAR T cell-PLT complexes enhance antitumor efficacy compared to CAR T cells in
a mouse xenograft model
After demonstrating that CAR T cells in CAR T cell-PLT and CAR T cell-PMP complexes perform
T cell effector functions as well as their PLT-free counterparts in vitro, we tested the antitumor efficacy of
these groups in a lung cancer xenograft model. NSG mice were injected with human H292.CD19 cells and
when xenografted tumors were established, mice were treated either with CAR T cells, CAR T-PLT
complexes or CAR T-PMP complexes. As seen in Figure 4.6a, CAR T-PLT treatment group showed
statistically significant tumor growth control, whereas the tumor growth in CAR T-PMP group yielded in
similar tumor size with CAR T treatment only group. Similar tumor size in mice treated with CAR T cells
and CAR T-PMP complexes was likely due to the low number of PMPs in CAR T-PMP complexes (as

105
proved with their low complexation efficiency in vitro) which would not generate a strong effect to make
a difference in CAR T cell homing and infiltration into tumors and improve tumor growth inhibition. Given
the most distinguished tumor size control achieved by CAR T-PLT treatment in this first set of experiments,
we decided to focus on CAR T-PLT complexes, but not their PMP counterpart, in the rest of the study.
To test whether the complex formation between CAR T cells with PLTs could improve antitumor
efficacy of CAR T cells, we treated tumor bearing mice with various T cell treatment modalities in a human
lung cancer xenograft model. H292.CD19 tumor-bearing mice were assigned into five different groups as
shown in Figure 4.5a.

Figure 4.5 CAR T cells in complex with PLTs demonstrated superior in vivo antitumor efficacy compared to
CAR T cells alone in a human lung cancer xenograft model.
(a) Schematic representation of the experimental procedure for tumor challenge and T cell adoptive transfer. 2 x 10
6

H292.CD19 cells were injected into the right flank of NSG mice. The xenografted tumors were established 15 days
later and mice were randomized into five groups (n = 5 mice per group). Mice were treated with 3 x 10
6
T cells (non-
transduced or CAR modified) or the same number of CAR T cells in complex with 240 x 10
6
activated PLTs. (b)
Change in body weight. Mice were weighed in every 2 days until Day 33. The percentage of change in mouse body
weight was calculated in comparison to initial weight of individual mouse. (n = 5, mean ± SD; ns, not significant;
*p<0.05; **p<0.01; ***p<0.001; ****p<0.0001, two-way ANOVA with Tukey’s multiple comparison). The
summarized statistics for Day 33 were shown on the figure legend. (c) Tumor growth analysis. Tumor size was
measured by a digital caliper in every 2 days until Day 33 (n = 5, mean ± SD; ns, not significant; *p<0.05; **p<0.01;
***p<0.001; ****p<0.0001, two-way ANOVA with Tukey’s multiple comparison). The summarized statistics for
Day 33 were shown on the figure legend. (d) Tail bleeding assay and CD41
+
PLTs in peripheral blood. On day 2, 6
and 10 after adoptive T cell transfer, peripheral blood of tumor-bearing mice was collected and the percentage of
CD41
+
PLTs were determined by flow cytometry. All data are in triplicates (n = 3, mean ± SD; ns: not significant;
*p<0.05; **p<0.01; ***p<0.001, ****p<0.0001, one-way ANOVA with Tukey’s multiple comparison).

106
Animals in all treatment groups showed tumor progression. As expected, there was no significant
difference in tumor growth control between PLT and NT, NT and NT-PLT, and PLT and NT-PLT
treatments. In particular, CAR T-PLT treatment group showed statistically significant tumor growth
control, compared to its PLT-free counterpart, with p=0.0003. As expected, mice treated with CAR T-PLT
cell complex also significantly prolonged tumor growth inhibition compared to NT-PLT (p<0.0001), non-
transduced T cells alone (p<0.0001), and PLTs alone (p<0.01) treatments, due to the TAA-targeted
specificity demonstrated by CAR T cells. One important result to discuss is the fact that the tumor growth
control between CAR T and NT, and between CAR T and PLT treatments groups was not significant. This
result was unexpected given the target-specific cancer cell killing effect of CAR T cells. This effect could
be explained by taking a closer look at the swimmer plot analysis (Figure 4.6b). One explanation to this can
be the early termination of the study due to euthanasia of the mice with tumor ulceration in all mice in PLT
group and majority of the mice in CAR T group, which did not allow the observation of a meaningful tumor
growth control difference between these groups created with the target-specific cytotoxicity of CAR T cell
treatment. Another explanation to this could be the difference between the tumor size of these two treatment
groups in the beginning of the experiment. In this in vivo study, the random assignment of mice into
treatment groups resulted in different tumor size average prior to adoptive cell treatments. For example, the
average tumor size on Day 15 was recorded as 48.66 ±18.13 for CAR T-PLT treatment group, whereas the
initial tumor size for PLT group was 25.39 ±12.59. This initial difference in tumor size impacted the tumor
growth profiles as well as the survival of mice. In addition, all animals in CAR T-PLT treatment group
were euthanized before size endpoint, when the average tumor size was around 400 mm
3
. Notably, one
animal in CAR T-PLT treatment group showed a partial response which signifies a 30% or greater decrease
in tumor size, before it started to continue growing again (Figure 4.6b). This suggests that multiple
injections of CAR T-PLT complexes could assist a prolonged tumor growth control.
Throughout the in vivo experiment, we also recorded the body weight of individual mice and
calculated the change in body weight compared to their pre-treatment weight. The cumulative weight loss

107
was observed in all treatment groups and it was within the range approved by the IACUC, as weight loss is
an expected result in mice with tumor burden over time. As seen in Figure 4.5b, no significant difference
was observed in body weight between mice treated with CAR T and CAR T-PLT, or NT and NT-PLT over
the course of the study. The difference in body weight was less in NT and NT-PLT groups, compared to
their CAR-engineered counterparts. However, this difference was not significant due to the variations in
body weight for individual mice in all groups. This finding indicates that PLTs do not cause toxicity when
injected intravenously in complex with T cells at 1:80 CAR T:PLT ratio and that the animals responded
well to the injected free and T cell bound PLTs.
In addition, we performed a tail bleeding assay 2, 6 and 10 days post-injection to investigate the
persistence of PLTs in circulating blood. Blood samples were collected at three time points and cells were
stained for PLT marker CD41. As displayed in Figure 4.5d, the percentage of CD41
+
cells significantly
increased over time. Although we did not inject platelet microparticles (PMPs) into the mice, it is known
that activated platelets form PMPs by exocytosis, and thus CD41
+
signal likely represents presence of both
PLTs or PMPs. PMP formation can also be induced as a result of physical stress, such as continuous
circulation of blood in mice. Therefore, the increase in CD41
+
cells over time, as opposed to an expected
decrease in signal given the relatively short lifetime of PLTs (7-10 days), could be attributed to the
formation of PMPs as time progress. Particularly, the persistence of CD41 signal suggests the lasting effect
of PLTs in animals. In addition, the percentage of CD41
+
cells was similar in blood of mice treated with
NT-PLT and CAR T-PLT complexes or free PLTs, suggesting that PLT complexation status does not affect
the persistence of PLTs in blood over time.

108

Figure 4.6 Supplementary data for the evaluation of tumor volume after CAR T-PLT and CAR T-PMP
treatment and swimmer plot analysis for T cell treatments.
(a) Tumor growth analysis. 2 x 10
6
H292.CD19 cells were injected into the right flank of NSG mice. After the
xenografted tumors were established, mice were randomized into three groups (n = 5 mice per group) and treated with
3 x 10
6
T cells (non-transduced or CAR modified) or the same number of CAR T cells in complex with 240 x 10
6

activated PLTs or PMPs derived from the activation of 240 x 10
6
PLTs. Tumor size was measured by caliper in every
2 days. Tumor growth curve in each group was displayed. (b) Swimmer plot analysis. Responses of individual mice
to T cell treatments are represented by horizontal bars. The length of the bars represents the duration of survival, and
the symbols at the end of each bar (explained in legend) indicates the reason of their euthanasia. Mice in different
treatment groups are grouped and color-coded. Dotted line represents the day in which T cell treatments were
administered (Day 15).

4.4.9 CAR T cells in CAR T cell-PLT complexes have greater T cell infiltration into the
tumor and persistence in blood compared to PLT-free CAR T cells
We performed ex vivo tissue analysis in the same lung cancer mouse xenograft tumor model to
support our hypothesis that PLTs in CAR T-PLT complexes aid in CAR T cell accumulation in tumor site.
12 days post-treatment, two groups of mice either treated with CAR T cells alone or with CAR T-PLT
complexes were euthanized. Their tumor, spleen, bone marrow and blood were harvested and analyzed for
different CAR T cell populations (Figure 4.7a).
In accordance with the tumor growth curve in Figure 4.5c, the average weight of excised tumors in
CAR T-PLT treatment group was significantly less compared to CAR T cell alone treated mice (Figure
4.7b, p<0.001).
High percentage of CD3
+
T cells were detected in blood and tumor, compared to spleen and bone
marrow as expected (Figure 4.7d). Number of blood circulating CD3
+
T cells in mice treated with CAR T-
PLT complexes were significantly higher compared to mice treated with CAR T cells alone (Figure 4.7d,

109
p<0.01), which demonstrates the successful engraftment and in vivo persistence of adoptively transferred
CAR T cells, and suggests that PLTs in complex with CAR T cells increased the engraftment and prolonged
persistence of CAR T cells.

Figure 4.7 CAR T cells in complex with PLTs exhibited superior accumulation in tumors and persistence in
blood compared to CAR T cells alone in a human lung cancer xenograft model.
(a) Schematic representation of the experimental procedure for tumor challenge and CAR T cell adoptive transfer. 2
x 10
6
H292.CD19 cells were injected into the right flank of NSG mice. The xenografted tumors were established 15
days later and mice were randomized into two groups (n = 3 mice per group). Mice were treated with 3 x 10
6
T cells
(non-transduced or CAR modified) or the same number of CAR T cells in complex with 240 x 10
6
activated PLTs.
On day 34, mice were euthanized and tumor, spleen, bone marrow and blood samples were collected. Tissues were
processed for ex vivo analyses. The percentage of different human T cell populations in the blood, spleen, bone marrow
and tumor of H292.CD19 tumor-bearing mice that were adoptively transferred with anti-CD19 CAR T cells and anti-
CD19 CAR T-PLT complexes were investigated by flow cytometry. (b) Tumor weight. Tumors were weighed after
harvest. The percentage of tumor weight relative to body weight of mice for each treatment group were shown in bar
graphs (c) Percentage of stable CAR T-PLT complexes in mouse blood and tumor. The percentage of CD41+ PLT
population in CD3
+
CD8
+
cytotoxic T cell populations were gated in blood and tumor samples. Percentage of
CD3
+
CD8
+
CD41
+
cell population representing stable CAR T-PLT complexes was exhibited in bar graphs. (c-e)
Percentage of human (c) CD3
+
T cells, (d) CD3
+
CD8
+
cytotoxic T cells, and (e) CD3
+
CD4
+
helper T cells in mouse
blood, spleen, bone marrow and tumor. All data are in triplicates (n = 3, mean ± SD; ns: not significant; *p<0.05;
**p<0.01; ***p<0.001, ****p<0.0001, one-way ANOVA with Tukey’s multiple comparison).

Markedly, the mice receiving CAR T-PLT complexes exhibited significantly higher presence of
CD3
+
T cells in the tumors (p<0.001), while CAR T cell treated mice had fewer T cells in the tumors (Figure
4.7d), indicating PLT-assisted homing of CAR T cells to tumor sites.
In order to investigate the percentage of cytotoxic and helper T cells in CD3
+
T cell population, we
stained the cells from harvested tissues for CD8

and CD4, respectively. We showed that CD3
+
CD4
+
T helper
cells mostly accumulated in spleen and bone marrow (Figure 4.7f).The number of CD3
+
CD4
+
T cells in

110
spleen and bone marrow was significantly higher in CAR T cell treated mice compared to CAR T-PLT
group, whereas helper T cell percentage in tumor and blood was low with no significant difference in these
two treatment groups. On the other hand, the majority of CD3
+
CD8
+
cytotoxic T cells were in blood and
tumors (Figure 4.7e). Mice treated with CAR T-PLT complexes had significantly more cytotoxic T cells in
circulating blood (p<0.001) compared to CAR T treatment (Figure 4.7e). Prominently, CAR T-PLT treated
mice showed significantly higher (p<0.05) numbers of CD8
+
cytotoxic T cells in tumors, compared to CAR
T cell treated group (p<0.05). Higher numbers of CD3
+
T cells, and specifically CD8
+
cytotoxic T cells in
tumors of mice treated with CAR T-PLT complexes compared to mice treated with CAR T cells alone,
signifies that the remarkable tumor growth control in CAR T-PLT treated mice was a result of improved
accumulation of CAR T cells to tumor sites with the aid of PLTs.
Next, we investigated if CAR T-PLT complexes stayed stable in blood and tumor or if PLTs
detached from CAR T cells, 12 days post-treatment. To detect the percentage of cytotoxic T cells in
complex with PLTs, we stained the cells in tumor and blood samples for CD3, CD8 and CD41 and
quantified CD3
+
CD8
+
CD41
+
cells, representing cytotoxic T cells that are in complex with PLTs, by
cytometry. In blood, the percentage of CD3
+
CD8
+
CD41
+
cells were significantly higher in CAR T-PLT
group, with an average of 46.5% (p<0.0001), compared to CAR T group with a negligible background level
signal of 4.7% (Figure 4.7c). This finding shows that almost half of the CAR T-PLT complexes stay intact
in blood 12 days after adoptive cell transfer. On the other hand, we did not detect any signal for
CD3
+
CD8
+
CD41
+
cells in tumors of mice injected with CAR T-PLT or CAR T treatment groups, indicating
that the cytotoxic T cells in tumors were not in complex form with PLTs (Figure 4.7c). When evaluated in
combination with the findings in Figure 4.5c, we speculate that PLTs which are in complex with CAR T
cells help CAR T cells traffic and infiltrate into tumors, and then detach from T cells and leave the tumor
site, mostly due to the mechanical stress in tumor stroma. Previously, platelets were shown to contribute to
tumor metastasis
269
. When we excise various organs from mice for ex vivo tissue analysis, we looked for
metastasis in other organs and we did not detect any metastasized tumor in mice injected with any PLT-
consisting treatment. This result suggests that our approach presents an alternative to enhance the antitumor

111
efficacy of CAR T cell therapy in solid tumors, in which we achieved by increased homing and
accumulation, and perhaps retention, of CAR T cells in tumor areas by the use of PLTs while not allowing
the possible tumor metastasis side effect of PLTs because they leave the tumor area after delivering their
cargo.
Overall, anti-CD19 CAR T cells in CAR T-PLT complexes successfully infiltrated into CD19 over-
expressing lung tumors after intravenous injection and exhibited improved antitumor activity while
maintaining a favorable safety profile.
4.5 Discussion
In this study, we employed platelets in complex with CAR T cells to augment CAR T cell
infiltration into tumor sites in a xenograft mouse model of lung cancer. We demonstrate how CAR T-PLT
complexes harness the power of the preexisting/intrinsic tendency of PLTs to traffic towards and infiltrate
into tumors, which transforms the inadequate infiltration of T cell in tumors to a permissive one. Overall,
compared with CAR T cell treatment, CAR T cells complexed with PLTs (CAR T-PLT) markedly
improved tumor growth inhibition, indicating higher efficiency in homing of CAR T cells into the tumor
areas where they performed targeted cytotoxic effects.
We formed cell-only “living drug” complexes with two different cell types, anti-CD19 CAR T cells
and PLTs. We formed cell-cell complexes which consist of CAR T T cells and activated PLTs, or CAR T
cells and PMPs (released by activated PLTs) at different complexation ratios and durations. Among many
antigens that have been used to identify PLTs and PMPs, we chose the most abundant integrin Glycoprotein
(GP) IIb/IIIa, also known as α IIbβ 3 and CD41 antigen, as the marker of PLTs and PMPs in our study
488
.
Stable CAR T-PLT/PMP complexes were formed in a coincubation duration as short as 6 hours. However,
for the ease of experiment design, CAR T-PLT/PMP complexes used in this work were prepared in 1-day
cocultures. Additionally, we demonstrated that CAR T-PLT and CAR T-PMP complexes remain stable for
2 days, evident in high CD8
+
CD41
+
signal observed in 2-day cocultures.

112
CAR T-PLT cocultures showed significantly higher CD8
+
CD41
+
cell percentage compared to CAR
T-PMP cocultures at both 1:20 and 1:80 CAR T:PLT/PMP ratios. This difference implies that PMPs
showed weaker binding to CAR T cells, in comparison to PLTs, and they were washed out by the wash step
following CAR T-PLT/PMP co-incubation, and that CAR T cell-PLT complexes are more resistant to
mechanical detachment. This could be attributed to possible differential affinity of PLTs and PMPs to CAR
T cells in CAR T-PLT/PMP complexes. Although it is known that interaction between T cells and
PLTs/PMPs are integrin-mediated
217
, there is no literature on the affinity of PLTs and PMPs to T cells or
other lymphocytes, which would benefit from further research.
Also, CAR T-PLT complexes prepared at 1:80 CAR T:PLT ratio showed significantly higher
presence of PLTs compared to CAR T-PLT complexes at 1:20 CAR T:PLT ratio. After verifying that CAR
T-PLT complexes at both CAR T-to-PLT ratios exhibited comparable IFN-𝛾 release profiles upon target-
specific stimulation, which are similar to PLT-free CAR T cell, we decided to use CAR T-PLT complexes
prepared at 1:80 CAR T-to-PLT ratio for the rest of this work, in order to observe the effect of PLTs more
clearly in in vitro and in vivo experiments. However, it is important to note that what creates certain
outcomes in CAR T-PLT and CAR T-PMP cocultures is not the number of PLTs or PMPs, yet the amount
of activated surface expressed receptors or released cytokines
484–486
.
According to intracellular staining results for IFN-𝛾 in CAR T-PLT/PMP complexes, only anti-
CD19 CAR T cells but not non-transduced T cells, efficiently released IFN-𝛾 upon stimulation by CD19-
overexpressing target cells. Similar to CAR T cells, CAR T-PLT complexes at 1:20 and 1:80 CAR T:PLT
ratio exhibited comparable IFN-𝛾 release profiles. Consistent with target-specific IFN-𝛾 secretion and
degranulation results, a target-specific cell killing trend was observed when CAR T cells and CAR T cell-
PLT/PMP complexes were cocultured with SKOV3.CD19 and H292.CD19 target cells. In addition,
cytotoxicity against target cells was significantly increased with the increase in E:T ratios, suggesting a
positive correlation that saturates at higher E:T ratios. Several articles have shown that PLT-T cell binding
suppresses T cell functionality
231,237
. However, the results of our in vitro IFN-𝛾 release, degranulation and
cytotoxicity assays have shown similar levels of T cell functionality upon PLT or PMP complexation. We

113
believe that CAR expression and thus, TAA recognition-induced IFN-𝛾 release, degranulation and cancer
cell killing compensate the probable effector-function diminishing effect of PLT complexation on T cells.
CAR T cells may not be able to penetrate into tumors through the vascular endothelium
497
due to a
collection of mechanisms in solid tumor tissues. The reduced secretion of vascular-related factors in tumor
tissue prevents T cell escape from blood vessels
498
, and reduced expression of chemokines that are involved
in T cells migration in tumor tissue
499
as well as the presence of dense fibrotic matrix in solid tumors, reduce
T cell migration and invasion into tumors
500
.
Leukocyte extravasation requires the solubilization of basal membrane (BM) through a
collaborative and sequential process mainly between leukocytes, platelets and endothelial cells, and
degradative enzymes expressed by them
501
. For example, when T lymphocytes bind to endothelium and
extravasate into tissues, there has been three ECM degradative enzymes associated with different cellular
participants for this migration
472
. First, ECM gets degraded by proteases, which are majorly expressed by
endothelial cells (HUVECs) and lymphocytes, resulting in intact or partially cleaved heparan sulfate
proteoglycans (HSPGs). Then, HSPGs are further cleaved into heparan sulfate side chains through the
action of heparanases which are majorly expressed by platelets. As the final stage of BM degradation,
sulfatases, which are expressed by HUVECs, degrade HSPGs yielding in free sulfate [ 35SO
4
]
472
. It has been
demonstrated in another study that long term ex vivo expanded T lymphocytes lack expression of
heparanase and hence lose their ability to degrade BM HSPGs
471
. In one study, in which the role of PMPs
in angiogenesis was investigated, the stimulatory effect of PMPs on endothelial cell (EC) movement was
demonstrated by a Matrigel invasion assay. The same study showed that the inhibition of heparanase
markedly was shown to reduce the number of invaded ECs
495
.
BM is known as a specialized ECM consisting of laminin, collagen type IV, nidogen, fibronectin
and HSPGs
501
, which is very similar to the composition of standard Matrigel (STD MG), a gelatinous cell
culture matrix commonly used in in vitro and in vivo experiments. Here, we showed in the in vitro trans-
Matrigel cell migration assays that CAR T cells which are in complex with PLTs showed higher migration
levels compared to the condition lack of PLTs in Standard Matrigel with high HSPG content, whereas the

114
migration levels of two groups were similar in HSPG-reduced Matrigel, as well as in 2D transmigration
with no Matrigel layer, indicating the degradation of HSPGs through the action of PLT-introduced
heparanase in the environment which created a cell migration advantage. STD MG represents the true
composition of tumor basal membranes. Thus, the contribution of PLTs in CAR T-PLT complexes to the
degradation of HSPGs in STD MG which enabled T cell movement in a 3D barrier presents a promising
strategy to improve T cell migration in tumor mass in vivo. Given this finding as well as the lack of in vivo
antitumor efficacy enhancement in mice treated with CAR T-PMP complexes, we decided to test the most
successful candidate, CAR T-PLT complexes at 1:80 CAR T:PLT ratio, for further in vivo antitumor
efficacy investigation.
Here, we investigated the antitumor efficacy of CAR T cell-PLT treatment in a xenograft model of
non-small cell lung cancer, established by the inoculation of H292 mucoepidermoid carcinoma cells. The
metastatic potential of H292 cells was shown in a study in which an extracellular heparan sulfate
endosulfatase regulate Wnt signaling by acting on HSPGs
502
. It is also important to mention here that the
heparanase content of tumor cells is related to their metastatic potential
503,504
. Direct interactions between
PLTs and T cells in the context of cancer have raised questions and needs further investigation. Platelets
and platelet derived factors were shown to influence metastasis; however their presence and amount may
not be sufficient for the activation of metastasis related signaling pathways and may require additional
players in the TME
487
. The utilization of PLTs for therapeutic purposes in the case of cancer treatment is
potentially debatable due to their reported effect on tumor growth and metastasis. However, we postulate
that this finding is not a concern in the scope of our work. Considering the relatively small number of
injected PLTs/PMPs in complex with CAR T cells in this study (1:80 CAR T-to-PLT ratio as opposed to
1:100, 1:250, and 1:500 T cell/PBMC-to-PLT ratios in literature
483,505
and the limited lifetime (~7-10
days
184
) of PLTs in circulation, we argue that the injection of PLTs/PMPs in complex form with CAR T
cells would not induce tumor growth or metastasis. Indeed, we justified the opposite with the findings of
our in vivo study, in which the tumor growth was significantly decelerated in the CAR T-PLT treatment
group compared to CAR T group. When we excise various organs from mice for ex vivo tissue analysis, we

115
looked for metastasis in other organs. We did not detect any metastasized tumor in mice injected with any
PLT-consisting treatment, suggesting the lack of metastasis-inducing effect of PLTs in CAR T-PLT
complexes although a longer observation may be necessary.
Intravenous administration of CD19-directed CAR T cell-PLT complexes into mice bearing human
H292.CD19 lung tumors demonstrated significantly increased levels of T cell accumulation in tumors and
improved antitumor efficacy. Specifically, more T cells were detected in tumors of mice treated with CAR
T-PLT complexes compared to CAR T cells alone. Given this finding and the improved T cell invasion in
Matrigel with high HSPG-content in CAR T-PLT group, we argue that the enhanced infiltration of CAR T
cells to tumor mass is likely a result of the degradation of the basal membrane HSPGs by PLT introduced
heparanases in CAR T-PLT complexes. We believe that the heparanase secreted from PLTs in CAR T-PLT
complexes was sufficient to increase permeability across the BM to allow CAR T cell accumulation in solid
lung cancer tumor tissue and not in high enough to stimulate the release of growth factors and cytokines
leading to metastasis and tumor development via angiogenesis. By ex vivo tumor analysis, we detected no
CD8
+
CD41
+
signal in the tumor mass, indicating that all CD8
+
T cells in tumors were PLT free. This finding
suggests that the self-formed complexes between CAR T cells and PLTs also self-disintegrate, likely when
PLTs complete their lifespan. We also showed that there is no significant change in mouse body weight
among different groups, indicating that free or T cell bound PLTs do not cause toxicity when used at CAR
T-to-PLT ratio of 1:80.
It is important to note that how platelets influence the TME-resident cells (leukocytes, endothelial
cells, fibroblasts, pericytes) may be contextually dependent on the composition of the TME which varies
depending on organ/tissue type, and the anatomical location and degree of intratumor vascularity of tumors.
In the future, this system can be tested in various cancer models including metastatic cancers and the
complex outcomes of PLT-T cell interactions can be studied.
We also want to emphasize that CAR T cells and PLTs which were used in this study were from
different donors. Cell-based combination therapy strategies with autologous cells can provide an alternative
to improve cancer immunotherapy with reduced possibility of an immunological reaction against the

116
donor’s cells and of graft-versus-host disease (GvHD). The two cell types that would be used in this
combination therapy strategy; CAR T cells for the targeted tumor cell killing and platelets for the enhanced
homing and infiltration of T cells into tumors, can be obtained from the same cancer patient by single
apheresis. Isolation of the mononuclear cell (MNC) layer can provide the circulating lymphocytes to be
used in CAR T cell manufacturing, whereas platelets can be isolated from the less dense plasma
506
. It is
important to note that we used frozen platelets in this study. Given the 7-10 day lifetime of platelets and 2-
3 week time requirement for CAR-engineering and expansion of T cells, platelets can be stored in freezer
after their isolation, and cocultured with CAR T cells when CAR T cell manufacturing is complete.
In clinic, prior to adoptive cell transfer, the hosts’ immune system needs to be conditioned to create
an appropriate environment that is depleted from regulatory mechanisms, such as regulatory T cells and
other competing cell populations. Thus, lymphodepleting conditioning enables the hosts to accommodate
transferred antitumor effector T lymphocytes, and as a result, support the robust long-term persistence and
function of T cells and might improve the treatment outcome
507
. A common phenomenon after
lymphodepletion is patients’ low platelet counts which can stay low for weeks and necessitate frequent
platelet transfusions
508
. Thus, given this highly platelet-depleted condition after CAR T cell adoptive
transfers, we believe that the occurrence of spontaneous complex formation between CAR T cells and PLTs
would likely be very low. Taking this minimal spontaneous CAR T cell-PLT complex formation, we posit
that complexing CAR T cells with PLTs beforehand presents a meaningful strategy.
The extravasation and inefficient traffic of CAR T cells into solid tumors are among the major
hurdles in their use in immunotherapies for solid tumors
474
. To overcome the extravasation limitation,
inspired by the interaction between chemokine receptors and their ligands, ECM-directed CAR T cells
which target components of the ECM, such as αvβ6 integrin, were engineered, shown to destruct the
architecture of tumor neo-vessels and likely limit the need for T cells to penetrate tumors
509
. It is important
to note that the tumor endothelium constitutes a barrier against T cell infiltration. In one study, in order to
improve the capacity of T lymphocytes to degrade the BM and thus promote their tumor infiltration and
antitumor activity, researchers engineered CAR T cells to express heparanase
471
. Inspired by these studies,

117
as well as the recent spotlight on platelets’ effect on immune response, we investigated the effect of CAR
T cell-platelet complexes in improved accumulation and antitumor efficacy of CAR T cells in solid tumors.
Instead of genetically engineering T cells to target a specific component of the ECM or to express one
ECM-destructive enzyme (i.e. heparanase), by virtue of using activated PLTs in this study, we target
multiple components of the ECM and introduce various adhesion molecules and enzymes which are
expressed and released by PLTs. With the presence of PLTs in injected CAR T-PLT complexes, a collection
of released molecules with roles in the capture of T cells from blood stream, as well as extravasation and
infiltration, we achieved an improved CAR T cell accumulation in tumors. To give our CAR T-PLT
complex system a more specific migration ability into the tumor bed, CAR T cells directed towards TAAs
can be further modified to express chemokine-specific receptors depending on differing chemokines ligands
in different tumors to support their efficient contact with tumor cells
510
. Alternatively, if the risk of
metastasis with platelet presence is a concern and to potentially eliminate this, CAR T-PLT complex
strategy can be combined with an anti-metastatic drug which would target the potential signaling pathways
that platelet-tumor cell interactions would trigger. Similarly, our T cell homing strategy with CAR T-PLT
complexes can be paired with other strategies to improve the antitumor efficacy of T cells. For instance,
our system can be combined with strategies targeting the immune-suppressive TME, such as co-injecting
or decorating CAR T cells with a small molecule immunomodulator- or checkpoint inhibitor-loaded
nanoparticles.
Anti-CD19 CAR has served as a proof-of-concept for utilizing PLTs to improve CAR T cell
accumulation in CD19-overexpressing lung cancer model. However, T cells in CAR T-PLT complexes can
be engineered to express CARs targeting various tumor associated antigens and tumors with different levels
of vascularization, metastasis potential and for cancer types in various locations of the body, to evaluate the
potential of CAR T-PLT complexes in different solid-organ malignancies.
To the best of our knowledge, the existing literature on lymphocyte-PLT interactions has majorly
focused on PBMCs and helper T cells, either with isolated PLTs or in whole blood samples. In our study,
T cell activation and genetic modification, as well as PLT activation were performed separately, and the

118
subsequent complex formation between CAR T cell-PLTs and CAR T cell-PMPs was achieved by co-
incubation of these two components. To our knowledge, this present study is the first in literature that
utilizes CAR-engineered T cell-PLT complexes as a strategy to enhance the CAR T cell accumulation in
tumors to overcome the limited antitumor efficacy of CAR T cells in solid tumor treatments.
The precise mechanism of platelets’ effect on CAR T cell homing in the form of CAR T-PLT
complex remains elusive and requires further investigations. This study however presents CAR T-PLT
complexes as a successful strategy to improve tumor infiltration and antitumor efficacy of CAR T cells.

119
5 Chapter 5: CAR-Engineered Natural Killer Cells as a Carrier of
Drug-Encapsulated Nanoparticles Targeting Adenosine
Receptors in Solid Tumors
5.1 Abstract
Traditional cancer treatments often fail to eradicate solid tumors resulting in tumor recurrence and
metastasis, and immunotherapy is a promising alternative. The immunosuppressive tumor
microenvironment (TME) plays a key role in failed eradication of solid tumors as adoptively transferred
immune cells lose their effector cell functions. One of the underlying mechanisms responsible for the
progressive loss of immune cell function in the adenosine rich TME is the A2a receptor (A2aR) inhibitory
pathway which is triggered by the binding of adenosine to A2aRs expressed on immune cells, which in turn
causes immune cell hypofunction. This suppression can be blocked by SCH-58261 (SCH), a potent and
selective A2aR antagonist drug, whose suboptimal solubility and pharmacokinetic profile could be
improved with a carrier. Nanoparticle-based delivery of immunomodulators to tumor sites requires
improvements, including target-specificity and longer retention time. Recently, immune cell-mediated
nanoparticle drug delivery to tumor sites has shown to be promising which can be further improved by
chimeric antigen receptor (CAR)-engineering of immune cells to specifically target tumor associated
antigens (TAAs) while abstaining from healthy tissues. Natural killer (NK) cells, which are potent cytotoxic
effector cells, have recently gained attention as an alternative “off-the-shelf” cellular immunotherapy which
would potentially solve expansion, cost and storage issues of immunotherapy. CAR-engineering of the
human NK cell line, NK-92, has been shown to have promising antitumor efficacy in vivo. We engineered
NK-92 cells with two different anti-mesothelin expressing CARs (with and without membrane-bound IL-
15/IL-15Rα complex) and covalently attached these CAR.NK cells with SCH-loaded cross-linked
multilamellar liposomal vesicles (cMLVs) for the targeted and active delivery of the small molecule

120
immunomodulator SCH to tumor sites to combat the immunosuppressive TME. We next verified that
cMLV conjugation and SCH encapsulation in cMLVs did not have adverse effects on CAR.NK cell
viability, proliferation, and effector functions of CAR.NK cells, including migration towards a
chemoattractant and cognate antigen-specific IFN-γ secretion and cytolytic activity. We demonstrated that
SCH-loaded nanoparticle conjugation to both CAR.NK cell types improved tumor growth control and
prolonged survival in an ovarian cancer xenograft mouse model. This study demonstrates that our
combinational therapy strategy can be used to enhance the antitumor efficacy of adoptive CAR.NK cell
therapy in solid tumors by interfering with the adenosine axis in the TME and preserving the effector
functions of NK cells via efficient delivery of an A2aR antagonist drug in NK cell conjugated liposomal
nanoparticles.
5.2 Introduction
For many years, researchers and cancer physicians have relied only on surgery, radiation therapy
and chemotherapy as primary cancer treatment modalities
2
. Even though these direct tumor interference
methods eliminate the majority of tumors, tumor relapse in primary tumor locations and metastatic
recurrence is very common
511
. Recently, the forth pillar of cancer therapy, immunotherapy has emerged as
an alternative solution to these obstacles in cancer treatment. Immune-based cancer therapies have
demonstrated unprecedented clinical success for advanced cancers
512
. Immune cells can be genetically
engineered with synthetic chimeric antigen receptors (CARs) that can improve their cancer targeting
capabilities by equipping them to recognize tumor-associated antigens (TAAs) with extracellular antigen-
binding domains derived from antibody fragments
60
. Specifically, adoptive cell transfer (ACT) of CAR-
engineered T (CAR T) cells have shown remarkable success in treating hematologic malignancies
47
. More
recently, among the candidates of immune cells to express CARs to be used in cellular immunotherapy,
ACT and CAR-engineering strategies have started to incorporate natural killer (NK) cells
122,124,128,513
. NK
cells bridge the innate and adaptive immune systems by giving rapid responses to virus infections or
malignant cell transformations without having been previously sensitized to them, upon receival of signals

121
by their germ-line–encoded receptors when monitoring autologous cells’ surfaces for aberrant expression
of MHC-I molecules and cell stress markers, although they do not express antigen-specific receptors
74,75
.
There have been numerous studies which demonstrate that NK cells contribute to cancer
immunosurveillance and killing that are mediated by several mechanisms including inflammatory cytokine
secretion [e.g., IFN-γ, TNF-α, IL-10], release of cytoplasmic granule toxins (e.g., perforin, granzyme A,
granzyme B)
77,78
as a result of antibody-dependent cellular cytotoxicity (ADCC)
79
.
Notably, the cytotoxic natural killer cell line NK-92 has been used extensively in preclinical and
clinical studies against various cancers, and is promising for its use in ACT due to its lack of
immunogenicity and availability as an “off-the-shelf” product which would potentially solve expansion,
cost, shipment and storage issues of cell-based immunotherapy
455
. The specificity and efficacy of tumor
killing of NK-92 cells can be enhanced by CAR engineering. Mesothelin is a cell-surface protein which has
limited distribution on normal tissues yet overexpressed in many cancers (including ovarian cancers), and
thus is an attractive target for tumor-specific therapies
459
. Recently, another preferred method to program
immune cells to be more therapeutically effective is the manipulation of CAR-engineered immune cells by
the addition of a membrane-bound cytokine to the CAR construct. Membrane-bound cytokines on T cells,
NK cells and antigen presenting cells have shown to help increase antitumor efficacy of CAR T cells by
recapitulating the effects of soluble cytokines in culture
514
, and increase NK cell proliferation and
expression of IFN-γ
515
. In a previous study of our group, a third-generation anti-mesothelin CAR consisting
of a single-chain variable fragment domain targeting mesothelin, and CD28 and 4-1BB co-stimulatory
domains was transduced into NK-92 cells (ameso.CAR.NK). In addition, the same CAR was modified to
express a membrane-bound IL-15/IL-15Rα complex and was efficiently transduced into NK-92 cells
(ameso.mbIL15.CAR.NK). The advantages of ameso.mbIL15.CAR.NK cells were shown as their
expansion ability in the absence of exogenous IL-2 and higher specific cell killing profiles against ovarian
cancer
516
.
As opposed to the outstanding success of CAR T cells in treating hematologic cancers, their use in
solid tumor treatments have had disappointing clinical results due to non-optimal antitumor efficacy and

122
non-durable response. One of the major roadblocks in CAR-engineered immune cell therapy against solid
tumors is the immunosuppressive tumor microenvironment
517
. To be effective, immune cells must
efficiently infiltrate into the solid tumor mass and have persistence in vivo
518
. The immunosuppressive TME
plays a key role in tumor recurrence, as resident immune cells become hypofunctional and lose their cancer
surveillance abilities. In addition, the TME contains a variety of pro-tumorigenic factors that prevent
cancer-killing immune cells from entering and persisting in the tumor area
519
. One approach to combat the
immune suppressive TME is to use immunomodulatory drugs to suppress the immune cell inhibitory
mechanisms, and thus prevent immune cell hypofunction in their cite of action. Nanoparticles are
considered promising drug delivery vehicles for cancer therapy based on their ability to increase drug
accumulation at tumor sites via the enhanced permeability and retention (EPR) effect. However, guiding
nanoparticles to poorly vascularized tumor tissues requires new strategies. Recently, numerous
nanoparticle-based active targeting approaches have emerged to enhance the accumulation of drugs around
the tumor; however, efficient delivery of these systems to the tumor site while sparing healthy tissue
remains elusive
511
. In a self-complementing way, immune cells are an attractive option as active carriers
for nanoparticles carrying a drug payload, such as immune modulators, because of their natural ability to
traffic and extravasate into the tumor and inflammatory sites
520
. NK-92 cells engineered with chimeric
antigen receptors recognize and traffic to cells expressing specific TAAs
60
and these cells can provide more
targeted immunomodulatory drug delivery to the tumor site when conjugated to drug-loaded nanoparticles.
Our lab has previously shown that CAR.NK cells conjugated to drug-encapsulated crosslinked
multilamellar liposomal vesicles (cMLVs) containing a chemotherapeutic drug were shown to greatly
improve drug delivery to the tumor site and enhance antitumor efficacy in vivo
521
.
One of the underlying mechanisms responsible for the progressive loss of adoptively transferred
immune cell function in the TME is the A2a adenosine receptor (A2aR) inhibitory pathway. A2aRs are
expressed on the surface of activated T and NK cells. High concentrations of the extracellular adenosine in
the TME, triggers the A2aR pathway, which suppresses T and NK cell proliferation and exhausts their
antitumor effects
319,331
. Given the importance of adenosinergic signaling in tumor-promotion and

123
immunosuppression in the TME, approaches to both blocking the generation of adenosine and inhibiting
its binding to adenosine receptors have gained significant research interest. Preclinical studies utilizing
pharmacologic A2aR blockade as its own or in combination with cell-based immunotherapy approaches
have shown great promise
349
. A small molecule immunomodulatory drug, SCH-58261 (SCH), is one of the
most selective and potent antagonists of A2aR
522
. The clinical application of SCH has been hindered by the
drug’s poor solubility and suboptimal in vivo pharmacokinetic profile which could be improved with a
carrier capable of regulating drug circulation time in vivo and efficiently delivering the drug to adoptively
transferred cells in tumors
523
. In a recent study of our group, CAR-engineered T cells were used as active
delivery vehicles to transport SCH-58261-loaded cMLVs into tumor areas to prevent or rescue the
emergence of hypofunctional CAR T cells within the TME
477
.
Targeted drug delivery mediated by human immune cells presents a promising method of
efficiently delivering small molecule drugs to the tumor site. Our system combines nanoparticle-based drug
delivery with immunotherapy to produce a cell-mediated, active targeting strategy. To improve the efficacy
of CAR-engineered NK cell therapy in solid tumor malignancies, we used NK-92 cells engineered with
chimeric antigen receptors (CAR.NK) as an active drug delivery vehicle carrying nanoparticles loaded with
a small drug A2aR antagonist, SCH-58261. While CAR.NK cells would have a target-specific antitumor
effect on the cancer cells themselves, we hypothesized that the delivery of SCH within the nanoparticles
conjugated onto cells, would potentially reverse immune suppression in the tumor microenvironment
(TME) due to A2aR antagonism and improve the antitumor immune response against solid tumor cancers
with minimal off-target toxicity to the patient. We tested our hypothesis by anti-mesothelin CAR.NK cells
(with or without membrane-bound IL-15/IL-15Rα complex expression) surface-conjugated to SCH-loaded
liposomal nanoparticles referred to as cMLVs (cMLV(SCH)) in a mesothelin-overexpressing ovarian
cancer xenograft model.

124
5.3 Methods
5.3.1 Cell lines and reagents
All lipids were obtained from NOF Corporation (Japan): 1,2-dioleoyl-sn-glycero-3-
phosphocholine (DOPC), 1,2-dioleoyl-sn-glycero-3-phospho-(10-rac-glycerol) (DOPG), and 1,2-dioleoyl-
sn-glycero-3-phosphoethanolamine-N-[4-(p-maleimidophenyl) butyramide (maleimide-headgroup lipid,
MPB-PE). SCH-58261 was purchased from Sigma-Aldrich (St. Louis, MO).
HEK 293T cell culture was maintained in DMEM supplemented with 10% FBS, 2 mM L-
glutamine, penicillin (100 U/mL) and streptomycin (100 μg/mL). Parental SKOV3 tumor cell line (ATCC
HTB-77) and mesothelin-expressing SKOV3 cells were maintained in RPMI 1640 (GIBCO) media
supplemented with 10% fetal bovine serum (FBS), 1% penicillin-streptomycin (pen-strep), and 2 mM L-
glutamine. NK-92 cells (ATCC CRL-2407; Dr. Jihane Khalife, Children’s Hospital Los Angeles) and their
CAR-engineered forms were maintained in MEM-𝛼 (GIBCO) supplemented with 10% FBS, 10% horse
serum, 1% non-essential amino acid (NEAA), 1% pen-strep, 1% sodium pyruvate, 0.1mM 2-𝛽-
mercaptoethanol, 0.2mM myo-inositol, and 2.5mM folic acid. All cells were maintained at 37℃ in a 5%
CO 2 environment.
5.3.2 Lentiviral and retroviral vector preparation
The CAR constructs consisted of ss1, the murine-derived scFv anti-mesothelin antigen-binding
domain; CD8 hinge and CD28 transmembrane regions; and CD28, 4-1BB, and CD3ζ cytoplasmic regions
in the retroviral MP71 vector. The membrane-bound IL-15/IL-15Rα complex was included in one of the
CAR constructs following a 2A linker. The plasmid for mesothelin consisted of human mesothelin cDNA
was cloned into a lentiviral FUW backbone.
Lentiviral and retroviral vectors were prepared by transient transfection of HEK 293T cells using
a standard calcium phosphate precipitation protocol
475
. HEK 293T cells were seeded at 18 x 10
6
cells per
15-cm tissue culture dish (BD Biosciences, San Jose CA). For lentivirus, HEK 293Ts were co-transfected

125
with 40μg of the lentiviral backbone plasmid, 20μg of the VSV-G-encoding envelope plasmid, and 20μg
of the packaging plasmids pMDLg/pRRE and pRSV-Rev. For retrovirus, HEK 293Ts were co-transfected
with 37.5μg of the retroviral backbone plasmid, 18.75μg of the envelope plasmid pRD114, and 30μg of the
packaging plasmid encoding gag-pol. Fresh media (high glucose DMEM supplemented with 10% FBS and
1% pen-strep) was plated onto the cells 4h after initial transfection. The viral supernatants were harvested
48 hours post-transfection and filtered through a 0.45μm filter (Corning, Corning, NY).
5.3.3 Transduction of NK-92 and SKOV3 cells
NK-92 cells were transduced with anti-mesothelin CAR retroviral vector as previously reported
516
.
Briefly, non-tissue culture-treated 12-well plates were coated overnight with 25μg RetroNectin per well
(Clontech Laboratories, Mountain View, CA). Fresh retroviral vectors were subsequently spin-loaded onto
the plates by centrifuging at 2000 x g for 2h at 32°C. NK-92 cells were resuspended at a concentration of 5
x 10
5
cells/mL with fresh media supplemented with 200 U/mL human IL-2 and added to the vector-coated
plates. The plates were centrifuged at 600 x g for 30min at 32°C and incubated overnight at 37°C and 5%
CO 2. CAR
+
NK-92 cells were sorted using fluorescence-activated cell sorting (FACS).
SKOV3 cells were similarly transduced to express human mesothelin via a lentivirus containing
ameso cDNA. Lentiviral supernatant was concentrated at 25,000 rpm using an Optima L-90 K preparative
ultracentrifuge and an SW28 rotor (90min at 4°C), resuspended in HBSS, and frozen at -80°C until later
use. Cells suspended in lentiviral supernatant were centrifuged at 1050 x g for 90min at room temperature
and resuspended in fresh media. Mesothelin
+
cells were sorted using FACS.
5.3.4 CAR detection on NK cell surface
After transduction and cell sorting, 1 x 10
5
NK-92 cells were incubated with recombinant human
mesothelin-Fc chimera (R&D Systems, Minneapolis, MN) at a volume ratio of 1:50 (2 μg/mL) in PBS at
4°C for 30min and rinsed with PBS. The cells were subsequently incubated with PE-labeled goat antihuman

126
Fc (Jackson ImmunoResearch) at a volume ratio of 1:150 in PBS at 4°C for 10min, rinsed, and read using
flow cytometry.
As an alternative method, three days after transduction, anti-mesothelin CAR.NK cells (1 x 10
5
)
were incubated with biotinylated Protein L (PeproTech) at a volume ratio of 1:50 in PBS + 4% FBS at 4°C
for 45 min and rinsed with PBS. The cells were subsequently incubated with streptavidin conjugated to
FITC (BioLegend) at a volume ratio of 1:500 in PBS + 4% FBS at 4°C for 10min, rinsed twice, and read
using flow cytometry. Non-transduced NK-92 cells served as a negative control. Fluorescence was assessed
using a MACSquant cytometer (Miltenyi Biotec, San Diego, CA), and all the FACS data were analyzed
using FlowJo software version 9.3.2 (Tree Star, Ashland, OR).
5.3.5 Preparation of cMLVs
Liposomes were prepared based on the conventional dehydration–rehydration method
524,525
.
Briefly, cMLVs were prepared from 1.5μmol of lipids at the molar ratio of the lipid composition of
DOPC:DOPG:MPB-PE = 4:1:5. Then, the lipids were mixed in chloroform and evaporated under argon
gas before drying under vacuum overnight to form dried thin lipid films. To encapsulate SCH-58261 into
cMLVs (cMLV(SCH)), 1mg of SCH in organic solvent was mixed with the lipid mixture to form dried thin
lipid films. The lipid film was rehydrated in 10mM Bis-Tris propane at pH 7.0. After the lipid was mixed
through vigorous vortexing every 10 minutes for 1 hour, they underwent three cycles of 15-second
sonication (Misonix Microson XL2000, Farmingdale, NY) and rested on ice at 1-minute intervals after each
cycle. A final concentration of 10mM MgCl 2 was added to induce divalent-triggered vesicle fusion. The
crosslinking of multilamellar vesicles (cMLVs) was performed by addition of dithiothreitol (DTT; Sigma
Aldrich) at a final concentration of 1.5mM for 1 hour at 37°C. The cMLVs were collected by centrifugation
at 14,000 x g for 5 minutes and washed twice with PBS. The particles were suspended in filtered water.
The particles in filtered water were vortexed and sonicated prior to analysis. Morphology of multilamellar
structure of the vesicles was analyzed and confirmed by cryo-electron microscopy in previous studies
524
.

127
5.3.6 Nanoparticle conjugation with cells and in situ PEGylation
Chemical conjugation of nanoparticles to NK cells was performed based on a method reported in
previous studies
266,520,526
. NK cells were resuspended in serum-free MEM-𝛼 (GIBCO) medium at the
concentration 10 x 10
6
cells/mL. An equal volume of nanoparticles in nuclease-free water were added to
the cells at 1000:1 cMLV-to-NK cell conjugation ratio and incubated at 37°C. The cells and nanoparticles
were mixed every 10 min for 30 min. After a PBS wash to remove unbound cMLVs, cells were further
incubated with 1mmol of 2 kDa thiol-terminated polyethylene glycol (PEG-SH) (Laysan Bio Inc. Arab,
AL) at 37°C for 1h in complete media to quench residual maleimide groups on cell-bound particles. To
remove unbound PEG, the particles were then centrifuged and washed twice with PBS. The final products
were stored in PBS at 4°C.
5.3.7 Quantification of cell-bound cMLVs
For quantification of cell-bound particles, nanoparticles were labeled with the lipid-like fluorescent
dye DiR at a 0.01:1 molar ratio (DiR:lipids) before conjugation. Fluorescence signal was assessed using
the MACSquant cytometer (Miltenyi Biotec, San Diego, CA).
5.3.8 NK cell viability assay
NK cells (5 x 10
4
cells/well) were incubated in media in a 96-well plate for 24h and analyzed for
viability. AnnexinV/7-AAD staining (BD Biosciences) was used according to the manufacturer’s
instructions. Fluorescent signal was read by flow cytometry and percentage of viable, early apoptotic and
late apoptotic cells were calculated. The data were collected in triplicates and presented as the mean ± SD.
To test the cytotoxicity of SCH-58261 against NK cells and SKOV3.meso cells, NK cells or ovarian
cancer cells (5 x 10
4
cells/well) were incubated in corresponding culture media in a 96-well plate for 24h
and analyzed for viability. Briefly, free SCH and SCH-loaded cMLVs were incubated with these cells. For
cMLV(SCH)-mixed or -conjugated subsets, we used 1000:1 cMLV:cell ratio. For example, 5ug SCH was
encapsulated in 5 x 10
7
cMLVs to be either mixed or conjugated to 5 x 10
4
cells. For free SCH subset, 5µg

128
free SCH was added in each well. Cells with no SCH or liposome addition served as control. AnnexinV/7-
AAD staining (BD Biosciences) was used according to the manufacturer’s instructions. Fluorescent signal
was read by flow cytometry and percentage of AnnexinV
+
/7AAD
+
cells were calculated. The data were
collected in triplicates and presented as the mean ± SD.
5.3.9 NK cell proliferation assay
To test the effect of IL-2 supplementation in cell proliferation, cells were suspended in NK cell
media in 0.5 x 10
6
cell/mL in 24-well plate and supplemented with fresh media (with or without IL-2
supplementation) in every 2 days. 200 U/mL human IL-2 was used in IL-2 supplemented group. Cells were
counted under light microscope with Trypan Blue exclusion. Cell expansion was recorded on Day 2, 5, 9,
and 13. The data was shown as fold increase in comparison to the initial cell number on Day 0 (0.5 x 10
6
cell/mL). Representative data collected on Day 2 and Day 13 were provided. The data were collected in
triplicates and presented as the mean ± SD.
To evaluate the effect of cMLV conjugation and SCH-58261 (SCH) on non-transduced and CAR-
engineered NK cell expansion, we cultured 5 x 10
4
NK cells per well in 24-well plates in IL-2 supplemented
media. For cMLV(SCH)-mixed or -conjugated subsets, we used 1000:1 cMLV:cell ratio, and the previously
explained amount of SCH. Cells with no SCH or liposome addition served as control. Cells were counted
under light microscope with Trypan Blue exclusion. Cell expansion was recorded on Day 2, 5, 9, and 13.
The data was shown as fold increase in comparison to the initial cell number on Day 0 (0.5 x 10
6
cell/mL).
Representative data collected on Day 2 and Day 13 were provided. The data were collected in triplicates
and presented as the mean ± SD.
5.3.10 Transmigration assay
As previously described
516
, NK cell transmigration assays were performed in 24-mm diameter, 5-
μm pore size Transwell plates (Costar). cMLV-conjugated and unconjugated NK cells (0.5 x 10
6
cells/well)
were plated on the upper wells in 0.1mL TCM. 0.5mL TCM with or without the CXCL9 chemoattractant

129
(200 ng/mL; Peprotech, Rocky Hill, NJ) was added to the lower wells. After incubation at 37°C for 8 hours,
NK cells that migrated into the lower chamber were counted under light microscope. The data were
collected in triplicates and presented as the mean ± SD.
5.3.11 Cytokine release assay
NK cells (1 x 10
5
cells/well) were coincubated with target cells in 96-well round-bottom plates
(Corning, Corning NY) at a 1:1 ratio for 6h at 37°C. GolgiPlug (BD Biosciences) was added to each well
to prevent protein transport. At the end of the incubation, Cytofix/Cytoperm Fixation and Permeabilization
Kit (BD Biosciences) was used to permeabilize cell membrane and intracellular staining was performed
according to the manufacturer’s instructions. Cells were stained for CD45 and IFN- 𝛾 using fluorophore-
conjugated anti-human CD45 and anti-human IFN-𝛾 (BioLegend, San Diego, CA). The data were
determined in triplicates and presented as the mean ± SD.
5.3.12 Cytotoxicity assay
The target cells (2 x 10
4
) were labeled with 5μM carboxyfluorescein succinimidyl ester (CFSE,
Life Technologies, Waltham, MA) as previously described
425
and coincubated with NK cells at effector-to-
target (E:T) cell ratios of 20:1, 10:1, 5:1, and 1:1. The cytotoxicity assay was performed in 96-well round-
bottom plates (Corning, Corning NY) for 7h at 37°C. The cells were then incubated in 7-AAD (Life
Technologies) in PBS (1:1000 dilution) for 10 min at room temperature and the fluorescence was analyzed
via flow cytometry. Percentages of killed cells were calculated as [CFSE
+
7-AAD
+
cells/ (CFSE
+
7-AAD
−
+
CFSE
+
7-AAD
+
)] cells, with live/dead gates based on control wells of target cells only to account for
spontaneous cell death. The cytotoxicity was determined in triplicates and presented as the mean ± SD.
5.3.13 In vivo antitumor activity
All animal experiments were conducted according to the guidelines set by the NIH and the animal
protocol approved by the University of Southern California Institutional Animal Care and Use Committee

130
(IACUC). All mice were hold under specific pathogen-reduced conditions in the University of Southern
California animal facility (Los Angeles, CA, USA). Six-to-eight-week-old female NOD.Cg-
PrkdcscidIL2Rγtm1Wj1/SZ (NSG) mice (Jackson Laboratories, (Bar Harbor, ME, USA) were used for all
in vivo experiments.
2 x 10
6
mesothelin over-expressing ovarian cancer (SKOV3.meso) cells which were in logarithmic
growth phase from cell culture, were subcutaneously inoculated into the right flank of NSG mice, on Day
-24. Tumors were allowed to grow to 40-70 mm
3
and mice were randomly divided into six groups of five
mice each. On Day 0, we started treatments according to their group description. The groups were
cMLV(SCH), non-transduced NK cells, ameso.CAR.NK cells, ameso.mbIL15.CAR.NK cells,
ameso.CAR.NK cells conjugated with SCH loaded liposomes (ameso.CAR.NK.cMLV(SCH)), and
ameso.mbIL15.CAR.NK cells conjugated with SCH loaded liposomes
(ameso.mbIL15.CAR.NK.cMLV(SCH)). The intravenous injections were conducted through the tail vein
on days 0, 4, 8, and 12. For all 4 injections, 5 x 10
6
NK cells (non-transduced or CAR modified) were used
per treatment. For CAR.NK cell groups with drug loaded cMLV conjugation, the same number of NK cells
were conjugated with SCH loaded liposomes at 1000:1 cMLV-to-cell ratio. For cMLV(SCH) group, the
same number of SCH loaded cMLVs (5 x 10
9
cMLV(SCH)/mouse) were suspended in PBS and injected to
mice. Physical states of the mice were observed. Tumor growth and body weight of the mice were measured
and recorded in every 2 days until the sacrifice point. The tumor length and width were measured with a
fine caliper, and tumor volume was calculated as 1/2 x (length) x (width)
2
. Mice were euthanized when they
displayed obvious weight loss accompanied with decreased mobility, when tumors started to ulcerate, or
tumor size reached 1,000 mm
3
and more. The survival rates were shown in Kaplan−Meier curves. The
survival curves of individual groups were compared by a log-rank test.

131
5.4 Results
5.4.1 Anti-mesothelin CARs with or without membrane-bound IL-15/IL-15Rα complex
are expressed in NK-92 Cells
To improve the efficacy of CAR-engineered NK cell therapy, we used CAR.NK cells as a drug
delivery vehicle carrying nanoparticles loaded with SCH-58261 (Figure 5.1a). Manipulation of CAR T cells
by the addition of a membrane-bound cytokine to the CAR construct is a recently preferred method to
program immune cells to be more therapeutically effective. Membrane-bound cytokines on T cells, NK
cells and antigen presenting cells have shown to help increase antitumor efficacy of CAR T cells by
recapitulating the effects of soluble cytokines in culture
514
, and increasing NK cell proliferation and
expression of IFN-γ and TNF-α
515
.
Considering the attested anti-apoptotic effects of 4-1BB and effective cytotoxicity of CD28
costimulatory domains for desired CAR function, two types of 3
rd
generation CAR molecules targeting
human mesothelin that were constructed in a previous study of our group were used for this study presented
in Chapter 3
415,527
.
Two CAR constructs which were consisted of a murine-derived single-chain variable fragment
(scFv)-derived antigen binding domain against mesothelin antigen, CD8 hinge and CD28 transmembrane
regions, CD28 and 4-1BB costimulatory domains, and CD3ζ T cell receptor signaling domain were
designed in a previous study
516
. One method to program immune cells to be more therapeutically effective
is engineering CAR constructs by the addition of a membrane-bound cytokine. The membrane-bound IL-
15/IL-15Rα complex is included in one of the CAR constructs following a 2A linker, for anti-
mesothelin.mbIL15 CAR. Both of these CARs were cloned into a retroviral vector for delivery to human
NK-92 cells. The schematic representation of the retroviral vector constructs used for transducing the NK-
92 cells in this study is shown in Figure 5.1b. The retroviral vector encoding anti-mesothelin CAR was
designated as ameso.CAR. The second CAR used in this study with its anti-mesothelin targeting domain
and membrane-bound IL-15/IL-15Rα complex encoding region was designated as ameso.mbIL15.CAR.

132

Figure 5.1 CAR.NK cells-to-cMLV conjugation is successful and drug-loaded cMLV conjugation does not
negatively affect CAR.NK cell viability and proliferation.
(a) Schematic illustration of the delivery of an A2 adenosine receptor (A2aR) antagonist drug (SCH-58261; SCH) to
the adenosine-rich immunosuppressive TME by conjugation of SCH-loaded liposomal nanoparticles, named as
crosslinked multilamellar liposomal vesicles (cMLVs), onto CAR-engineered NK cell surface. The A2aR antagonist
SCH blocks the binding of adenosine to A2aR on NK cells, which inhibits the immune-suppressive effects of
adenosine. Thus, CAR.NK cell effector functions can remain in desired levels, which can potentially increase the
antitumor efficacy of CAR NK cell treatments in solid tumors. (b) CAR construct targeting human mesothelin. The
anti-mesothelin scFv region was fused in frame with the CD8 hinge and CD28 transmembrane domain, followed by
the CD28/4-1BB/CD3ζ signaling domains. The cDNA of membrane-bound IL-15/IL-15Rα complex was linked by a
2A sequence. (c) Liposomal nanoparticle conjugation on CD45
+
NK-92 cells. Liposomal nanoparticles (cMLVs) were
labeled with DiR lipophilic dyes and conjugated to NK cells at 1000:1 cMLV:NK cell ratio. DiR fluorescence was
analyzed to calculate the percentage of cMLV conjugation by flow cytometry. cMLV-free NK cells were used as a
negative control. Representative flow cytometry data was shown. (d) Viability and apoptosis stages of NK cells.
Viability and apoptosis stages of NK cells were analyzed after 24h culture by using AnnexinV/7-AAD staining
followed by flow cytometry analysis. The percentage of NK cells (non-transduced NK cells (NK), ameso.CAR.NK
cells, and ameso.mbIL15.CAR.NK cells in no cMLV, empty cMLV and drug-loaded cMLV conjugation conditions)
with different staining results were shown in bar graphs. All data are representative of three independent experiments.
(e) NK cell expansion with or without exogenous IL-2 over time. Non-transduced NK cells (NK), ameso.CAR.NK
cells and ameso.mbIL15.CAR.NK cells, each either not conjugated with cMLVs (Cells), conjugated with empty
cMLVs (Cells+cMLV(empty)), or conjugated with SCH-loaded cMLVs (Cells+cMLV(SCH)) were cultured in
complete NK cell media with or without exogenous IL-2 supplementation. The cell numbers were counted under light
microscope in every 2-3 days for 13 days and fresh media was added in cell culture. The fold increase in cell number
was calculated compared to the initial cell number. Summarized statistics of triplicates were represented in bar graphs
(n = 3, mean ± SD; ns, not significant; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001, two-way ANOVA with
Tukey’s multiple comparison).

Both anti-mesothelin CAR.NK (ameso.CAR.NK) cells and anti-mesothelin CAR.NK cells with
membrane-bound IL-15/IL-15Rα complex (ameso.mbIL15.CAR.NK) were generated with retroviral
transduction of human NK-92 cells, using the above mentioned MP71 vector. Transduced NK-92 cells with
each constructs were sorted using fluorescence-activated cell sorting (FACS) to further increase the

133
percentage of CAR-positive cells. After transduction and cell sorting, to detect CAR expression on NK cell
surface, transduced cells were incubated with recombinant human mesothelin-Fc chimera and stained with
PE-labeled goat antihuman Fc. Fluorescent signal was read using flow cytometry. We confirmed the ability
of NK-92 cells to express anti-mesothelin CARs with or without membrane-bound IL-15/IL-15Rα
complex. As also shown in the previous work of our group, both CAR.NK cell groups displayed high levels
of CAR expression, with slightly lower CAR percentage in ameso.mbIL15.CAR.NK cell surface
516
. CAR
expression for both anti-mesothelin CAR, either without (ameso.CAR.NK) or with
(ameso.mbIL15.CAR.NK) membrane bound IL-15/IL-15Rα complex was shown to stay stable months
after the transduction and sorting (data not shown). We also generated a target cell line, SKOV3.meso, to
stably overexpress the human mesothelin. SKOV3 cells were transduced to express human mesothelin via
a lentiviral vector and mesothelin
+
cells were sorted using FACS.
Previous research of our group indicates that all cells - non-transduced (NT) NK-92 cells,
ameso.CAR.NK cells and ameso.mbIL15CAR.NK cells - expanded well in media with IL-2
supplementation, while ameso.mbIL15.CAR.NK cells also survived and proliferated in media without IL-
2 over 10 days and NT NK-92 cells and ameso.CAR.NK cells died by day 5 in the absence of exogenous
IL-2
516
. In the same article, we demonstrated that unsorted ameso.CAR.NK cells with membrane-bound IL-
15/IL-15Rα survived and expanded in the absence of exogenous IL-2 in vitro as well
516
.
5.4.2 cMLVs are stably conjugated to the surface of NK cells
To advance the efficacy of CAR-engineered NK cell therapy, we utilized CAR.NK cells as carriers
of nanoparticles loaded with SCH-58261 which is a selective and potent adenosine receptor A2A
antagonist. Nanoparticles were named as crosslinked multilamellar liposomal vesicles (cMLVs). These
liposomes were conjugated to the surface of CAR.NK cells. Previous studies of our group have shown that
cMLVs were successfully incorporated with both hydrophobic and hydrophilic drugs, such as
chemotherapy drugs paclitaxel and doxorubicin, as well as SCH-58261
266,464,477,521,528
.

134
cMLVs were prepared based on the conventional dehydration-rehydration method. Liposomal
vesicles were synthesized by covalently crosslinking functionalized headgroups of adjacent lipid bilayers
with a vesicle formation process triggered by the addition of MgCl 2, as previously exhibited
464,524
. Robust
and stable vesicles were generated by interbilayer crosslinking across the opposing sides of lipid bilayers
by the reactive headgroups with dithiothreitol (DTT). Previously, it was also shown that cMLVs restored
their stability in PBS at 4℃ over two weeks with no significant change in their size or size distribution
524
.
In previous studies, morphology of multilamellar structure of the vesicles was confirmed by cryo-
electron microscopy
464,524
. The images demonstrated that the cMLVs exhibited multilayered vesicle
formation with thick walls, implying the stability of multilayered structure of the liposomes generated by
the covalent linkage between adjacent bilayers. Also, the hydrodynamic size and size distribution of cMLVs
were measured by dynamic light scattering (DLS)
464
. The hydrodynamic size of empty cMLVs was reported
as 220±14.99 nm (mean diameter), whereas the cryo-electron microscopy-estimated mean diameter was
around 160 nm
524
. Moreover, researchers indicated that cMLV particles exhibited a narrow size distribution
with polydispersity of 0.101±0.0082, suggesting non-significant particle aggregation during the
crosslinking process
524
.
Synthesized cMLVs were stably conjugated to the CAR.NK cell surface. As shown in literature,
high amounts of free thiols have been detected on the surface of lymphocytes
529
. The stable conjugation of
cMLVs to NK cells took place between the reduced thiol groups on the surface of NK cells and the thiol-
reactive maleimide head groups on the lipid bilayer surface. The cMLV conjugation was performed in two
steps. In the first step, NK cells and cMLVs were coincubated for the initial coupling reaction of maleimide-
functionalized lipids and free thiols on the cell membrane. As the final step, residual thiol-reactive groups
on the cell-conjugated cMLVs were quenched by thiol-terminated PEG. PEGylation further improve vesicle
stability and extend blood circulation half-life
530,531
.
As determined by a recent study of our group, the cMLV-to-cell ratio of 1000:1 was chosen as
optimal for the conjugation of cMLVs to NK cells, which yielded an average conjugation of approximately
150 nanoparticles per NK cell
521
. To determine the maximum numbers of particles that could be conjugated

135
per NK cell, they performed a serial dilution of the conjugation at different fluorescent-labeled cMLVs-to-
cell ratios (2000:1, 1000:1, 500:1, 100:1, and 10:1). The optimal ratio to use was determined as 1000:1,
because further increasing the cMLV number did not cause an increase in the number of conjugated cMLVs
on the cell surface, as determined by the plateaued fluorescent signal
521
. In this study, we also used cMLV
to cell ratio of 1000:1 for all in vitro and in vivo experiments.
In this work, for the quantification of cell-bound particles, liposome particles were labeled with
DiR lipophilic dyes before conjugation. DiR-loaded liposomes were attached to CAR.NK cells and DiR
signal was detected by FACS analysis assuring the stable attachment of cMLVs to CAR.NK cells. As seen
in Figure 5.1c, the conjugation efficiency of the nanoparticles on the NK-cell population was determined
around 35%.
Next, we incorporated the hydrophobic drug SCH into the lipid membranes and conjugated these
drug loaded cMLVs to NK cells following the aforementioned steps. In a previous study of our group, the
surface conjugation of DiD-labeled cMLVs on carboxyfluorescein diacetate succinimide ester (CFSE)-
labeled CAR T cells was visualized using confocal microscopy. Single-cell imaging and 3D reconstruction
of CAR T.cMLVs demonstrated that the nanoparticles were distributed in several clusters on the CAR T
cell surface
477
. Similarly, in a previous study of our group, CAR.NK cells were conjugated with cMLVs.
For quantification of cell-bound particles, cMLVs were labeled with DiD and particle fluorescence was
detected with flow cytometry and a fluorescent microplate reader. In addition, the dual staining of cMLVs
with DiD and CAR.NK cells with CFSE allowed the visualization of cMLV-conjugation to NK cells by
confocal microscopy
521
. Because we use a very similar system in this work, we decided not to perform a
confocal microscopy imaging. We instead confirmed the conjugation of cMLVs to the NK cell surface
through aforementioned DiR labeling of cMLVs and the following FACS analysis.
It is also important to mention that elongated surface retention of nanoparticles on the surface of
carrier cells has major advantages, such as (i) the prevention from immediate particle degradation due to
internalization of nanoparticles into degradative intracellular compartments; and (ii) the preserved
continuous drug release from the nanoparticle-conjugated cells which enables drug targeting to tumor

136
cells
520,526
. Several researchers have shown that nanoparticles can be endocytosed by a variety of cells, such
as endothelial cells and macrophages
532–535
. For the purpose of our study, longer retention of drug loaded
cMLVs on CAR.NK cell surface is also preferred. Thus, researchers from our lab have previously quantified
cMLV internalization by NK cells
521
. They concluded that the attachment of cMLVs to NK cells did not
trigger endocytosis of liposomes and that cMLVs bound to NK cells remained at the cell surface.
5.4.3 In vitro CAR.NK cell viability is unaffected by drug-loaded nanoparticle
conjugation onto cell surface
To test the effect of empty or SCH-loaded liposome conjugation on the viability of non-transduced
or CAR-engineered NK cells, we performed AnnexinV/7-AAD staining 24h after setting cell cultures. We
then analyzed the FACS data to ascertain whether the cells are at the early stages of apoptosis (AnnexinV
+
7-AAD
-
), later stages of apoptosis and actively undergoing apoptosis (AnnexinV
+
7-AAD
+
) or double
negative healthy cells (AnnexinV
-
7-AAD
-
).
As seen in Figure 5.1d, AnnexinV
+
7-AAD
+
necrotic cells constituted 7%, 8.5%, and 13% of the
total cell population in overall non-transduced NK cells, ameso.CAR-engineered NK (ameso.CAR.NK)
cells and ameso.mbIL15.CAR-engineered NK (ameso.mbIL15.CAR.NK) cells, respectively, with no
significant difference between groups regardless of liposome binding status. The percentage of cells at the
terminal stage of cell death significantly differed between ameso.CAR.NK and ameso.mbIL15.CAR.NK
cells, all in their cMLV-unconjugated (8.54±0.36% and 12.93±0.76%, p<0.001), empty cMLV-
conjugated (8.44±0.06% and 14.97±4.36%, p<0.0001), and SCH-loaded cMLV-conjugated (7.83±0.38%
and 11.30±0.40%, p<0.01) forms, respectively. Similarly, the percentage of the cells starting to undergo
apoptosis (AnnexinV
+
7-AAD
-
) were significantly higher in ameso.mbIL15.CAR.NK cells compared to
ameso.CAR.NK cells. Furthermore, the cMLV conjugation status of both CAR-engineered NK cell groups
did not have any effects on the amount of cells at early apoptosis stage.
Non-transduced NK cells had the highest percentage of healthy cells in cell culture with 83% cell
viability, followed by ameso.CAR.NK cells with 78%, and ameso.mbIL15.CAR.NK cells with 66.5%. The

137
percentage of healthy cells did not significantly change either in ameso.CAR.NK cells or
ameso.mbIL15.CAR.NK cells with the conjugation of empty or drug loaded cMLVs. Regardless of their
liposome conjugation status, ameso.CAR.NK cells showed markedly high viability compared to anti-
mesothelin CAR.NK cells with membrane-bound IL-15/IL-15Rα complex (ameso.mbIL15.CAR.NK). The
percentage of cells in early apoptosis stage was significantly higher than the percentage of cells in late
apoptosis stage for both CAR-engineered NK cell groups and for all liposome conjugation conditions.
These experiments demonstrated that the conjugation of empty or drug-loaded cMLVs did not have
any adverse effects on the viability of NK cells. We also anticipate that the higher percentages of apoptotic
cells (early and late) in ameso.mbIL15.CAR.NK cell group was likely due to the possible nutrient
deprivation in the culture media of these cells as a result of their high duplication rates, as well as their
tendency to form big cell clumps that we observed under microscope (data not shown).
5.4.4 In vitro CAR.NK cell expansion is unaffected by drug-loaded nanoparticle
conjugation onto cell surface
To evaluate the effect of IL-2 on NK cell expansion in vitro, we analyzed the fold change in NK
cell number with or without exogenous IL-2 supplementation over 13 days (Figure 5.1e). In the presence
of exogenous IL-2, all NK cell types expanded 7-10 fold in 13 days. In the absence of IL-2 cytokine which
would stimulate NK cell survival and proliferation, non-transduced and ameso.CAR engineered NK cells
did not show cell expansion. In fact, a majority of them died by Day 13. On the other hand, genetically
engineered CAR.NK cells with membrane-bound IL-15/IL-15Rα complex allowed cells to expand in the
absence of exogenous IL-2. Both on Day 2 and Day 13, no significant difference was observed between the
expansion profiles of ameso.mbIL15.CAR.NK cells cultured with or without exogenous IL-2
supplementation.
We also analyzed whether cMLV (empty or drug-loaded) conjugation onto NK cell surface have
an effect on cell expansion over time (Figure 5.1e). In the absence of IL-2, no significant difference was
recorded in NK cell number (for both non-transduced and CAR-engineered NK cell data sets), among any

138
conjugation status, both in Day 2 and Day 13. In the presence of exogenous IL-2, the expansion of empty
cMLV-conjugated ameso.mbIL15.CAR.NK cells was lower than its SCH-loaded liposome conjugated
counterpart (p<0.01) on Day 2, which did not repeat itself on Day 13. In addition, no significant difference
was observed between the expansion of liposome free and cMLV(SCH)-conjugated
ameso.mbIL15.CAR.NK cells either on Day 2 or Day 13. On Day 13, ameso.CAR.NK cells expanded
similarly when liposome free and drug-loaded liposome conjugated subsets were compared, whereas the
expansion of empty cMLV-conjugated ameso.CAR.NK cell were lower (p<0.05). It is important to note
that we observed no significant difference between the expansion profiles of NK, ameso.CAR.NK and
ameso.mbIL15.CAR.NK cells either in liposome free or cMLV(SCH)-conjugated conditions on Day 13.
Overall, the cell expansion analysis results indicate that neither CAR modification nor drug loaded
nanoparticle conjugation onto NK cell surface interfere with proliferation kinetics in vitro over two weeks.
We next tested whether surface-bound cMLVs could impact key cellular functions of CAR.NK
cells, such as cell migration, cytokine secretion and cytotoxicity.
5.4.5 In vitro CAR.NK cell migration ability is unaffected by drug-loaded nanoparticle
conjugation onto cell surface
As an indicator of NK cell effector function, we investigated NK cell migration with or without
cMLV conjugation. NK cells must respond to chemoattractants in tumor areas and migrate into tumors to
generate anticancer immune response
536
.

139

Figure 5.2 CAR.NK cells conjugated with nanoparticles show similar levels of cell migration and IFN-γ
secretion compared to nanoparticle free CAR.NK cells.
(a) Cell migration assay. 0.5 x 10
1
NK/CAR.NK cells or their cMLV conjugated counterparts were plated in the upper
chambers of Transwell plates with chemoattractant CXCL9 addition to the lower chambers. After 4h or 8h incubation,
media from the lower chambers was collected and NK cells were counted under light microscope. Summarized
statistics are displayed in the bar graphs (n = 3, mean ± SD; ns: not significant; *p<0.05; **p<0.01; ***p<0.001, two-
way ANOVA with Tukey’s multiple comparison. (b) IFN-γ release assay. Non-transduced NK.92 cells (NK),
ameso.CAR.NK cells and ameso.mbIL15.CAR.NK cells, each either not conjugated with cMLVs, conjugated with
empty cMLVs, or conjugated with SCH-loaded cMLVs, were prepared. All NK cell groups were either cultured
without any stimulation or cocultured with mesothelin+ or mesothelin- SKOV3 cells at 1:1 ratio in the presence of
GolgiPlug inhibitor. After 6h, intracellular cytokine staining was performed. IFN-γ secreting CD45+ NK cell were
detected by flow cytometry. Unstimulated CAR.NK cells were used as negative controls. The percentage of
CD45+IFN-γ+ NK cells under different conditions were represented in a bar graph (n = 3, mean ± SD; ns, not
significant; *p<0.05; **p<0.01; ***p<0.001, ****p<0.0001, two-way ANOVA with Tukey’s multiple comparison).

In order to confirm that cMLV conjugation to NK cell surface did not impede with cell migration,
we performed an NK cell transmigration assay using a Boyden chamber. We observed the migration of NK
cells from the upper chamber to lower wells towards the chemoattractant CXCL9, which is known to induce
chemotaxis of leukocytes. The percentage of migrated ameso.CAR.NK cells (18.83%, p<0.0001) and
ameso.mbIL15.CAR.NK cells (12.05%, p<0.05) were significantly higher than non-transduced NK cells
(4%) (Figure 5.2a) after 4h incubation. Notably, although there is no significant difference between the
migration levels of ameso.CAR.NK and ameso.mbIL15.CAR.NK cells in the first 4h of the 2D-
transmigration assay, significantly more (p<0.0001) ameso.CAR.NK cells migrated towards the
chemoattractant within the following 4 hours. At 8h time point, non-transduced NK cell migration was
significantly less compared to ameso.CAR.NK (p<0.0001) and ameso.mbIL15.CAR.NK cells (p<0.01).
Also, although there is no significant difference in the number of migrated NK cells or
ameso.mbIL15.CAR.NK cells between 4h and 8h time points, significantly higher ameso.CAR.NK cell

140
migration (p<0.0001) was recorded at 8h cell count compared to 4h. In a previous study of our group,
transmigration of aCD19.CAR-expressing NK-92 cells was studied using the same chemoattractant
521
. The
reported number of migrated cells was significantly lower compared to this present study, indicating that
ameso.CAR.NK and ameso.mbIL15.CAR.NK cells show superior migration abilities.
Markedly, there was no significant difference between the migration levels of cMLV conjugated
and unconjugated NK cell groups either at 4h or 8h time points (Figure 5.2a), indicating that conjugation
of cMLVs to the cell surface did not interfere with NK cell migratory abilities over time and CAR.NK cells
maintain their transmigration capabilities.
5.4.6 In vitro CAR.NK cell IFN-𝛄 secretion is unaffected by drug-loaded nanoparticle
conjugation onto cell surface
Next, we investigated whether CAR.NK cells have altered their therapeutic behavior with liposome
conjugation compared to liposome-free CAR.NK cells. For this purpose, we designed in vitro experiments
testing inflammatory cytokine IFN-γ secretion of NK cells. We transduced parental SKOV3 cells with
mesothelin lentivirus to express mesothelin (SKOV3.meso) to serve as target cells for our anti-mesothelin
CAR.NK cells. Figure 5.2b represents the IFN-γ secretion data of CD45
+
non-transduced (NT) and CAR-
engineered NK cells which are either not conjugated with cMLVs, conjugated with empty cMLVs or
conjugated with SCH-loaded cMLVs. All NK cell groups were either not stimulated with any soluble
markers or cell-surface molecules, or cocultured wild type or mesothelin-expressing SKOV3 cells for 6 h,
in the presence of a GolgiPlug inhibitor. Next, intracellular cytokine staining was performed for the
inflammatory cytokine IFN-γ and the production of IFN-γ in CD45
+
NK cell population was measured and
representative data were presented in bar graphs.
In the absence of stimulation, NK cells displayed a background level of IFN-γ secretion (~1-2%).
Similarly, none of the non-transduced NK cells or CAR.NK cells reacted to SKOV3 cells which do not
express the target antigen mesothelin. For instance, this low level non-specific IFN-γ release was recorded
as approximately 5% and 7% for ameso.CAR.NK cells and ameso.mbIL15.CAR.NK cells, respectively

141
(Figure 5.2b). As expected, non-transduced NK cells, regardless of their nanoparticle conjugation status,
displayed low levels of IFN-γ secretion when co-incubated with SKOV3.meso cells, showing the necessity
of an anti-mesothelin CAR to generate an immune response when cocultured with mesothelin
+
target cells.
In contrast, ameso.CAR.NK cells and their empty liposome conjugated (ameso.CAR.NK.cMLV(empty))
and SCH-loaded liposome conjugated (ameso.CAR.NK.cMLV(SCH)) counterparts were activated and
showed significantly higher levels of IFN-γ secretion compared to their non-transduced NK cell
counterparts when cocultured with target antigen mesothelin-expressing SKOV3 cells (SKOV3.meso)
(p<0.0001) (Figure 5.2b). Similarly, NK cells which were modified with anti-mesothelin CARs with
membrane-bound IL-15/IL-15Rα complex and their liposome conjugated counterparts demonstrated the
same trend when cocultured with SKOV3.meso target cells (Figure 5.2b). These results indicate antigen-
specific activation and specificity toward the appropriate TAA, mesothelin. Upon stimulation with
SKOV3.meso target cells, ameso.CAR.NK cells exhibited approximately 18% IFN-γ
+
CD45
+
NK cells,
whereas ameso.mbIL15.CAR.NK cells yielded greater activity (~35% IFN-γ
+
CD45
+
NK cells, p<0.0001)
(Figure 5.2b). These data are in accordance with previous work showing that IL-15 results in increased
production of IFN-γ in NK cells
461–463
.
Consistent with the previous studies in which drug-loaded cMLV conjugation did not negatively
affect the cytokine secretion of anti-CD19.CAR.NK cells
521
, both empty and drug-loaded cMLV conjugated
CAR.NK cells responded with a similar range of IFN-γ release percentages for both CAR.NK cells types.
For instance, all ameso.CAR.NK, ameso.CAR.NK.cMLV(empty) and ameso.CAR.NK.cMLV(SCH)
groups cocultured with SKOV3.meso cells showed similar levels of IFN-γ secretion with no significant
difference, as 17.73±4.65%, 19.63±4.03%, and 17.36±2.8%, respectively (Figure 5.2b). Similarly, when
the ameso.mbIL15.CAR.NK cells were conjugated to either empty cMLVs or SCH-loaded cMLVs, IFN-γ
release was not significantly different from that of unconjugated CAR.NK cells (Figure 5.2b), recorded as
32.17±1.39% for ameso.mbIL15.CAR.NK cells, 37.03±5.84% for ameso.mbIL15.CAR.NK-
cMLV(empty), and 36.47±6.20% for ameso.mbIL15.CAR.NK-cMLV(SCH). Taken together, these data

142
indicated that the mesothelin-specific NK cell function was not impaired by the conjugation of empty or
SCH-loaded liposomes, yielding similar levels of IFN-γ secretion efficiency.
5.4.7 In vitro CAR.NK cell cytotoxicity is unaffected by drug-loaded nanoparticle
conjugation onto cell surface
The ability of liposome-free CAR.NK cells and liposome conjugated CAR.NK cells to trigger
cancer cell lysis in the presence of target antigen mesothelin was assessed by a cytotoxicity assay. NK cells
engineered with anti-mesothelin CARs with or without membrane-bound IL-15/IL-15Rα complex were
cocultured with either wild-type SKOV3 cells or mesothelin-overexpressing SKOV3.meso cells. In
addition, different effector-to-target (E:T) ratios were analyzed in order to investigate the dose dependent
cell killing. After 7 h of NK cell-cancer cell coculture, the percentage of dead cancer cells was analyzed by
flow cytometry. As expected, non-transduced NK cells exhibited no killing activity toward SKOV3 or
SKOV3.meso cells, with unspecific cell killing that is smaller than 5% at all E:T cell ratios (Figure 5.3a
and Figure 5.3b).
Similarly, coculture of CAR.NK cells with SKOV3 cells which did not express the cognate antigen
generated a low percentage of cytotoxicity, indicating the necessity of the TAA for the activation and
functionality of CAR.NK cells (Figure 5.3a and Figure 5.3b). In contrast, due to CAR-mediated cell killing,
both ameso.CAR.NK cells and ameso.mbIL15.CAR.NK cells displayed significantly greater cytotoxicity
against SKOV3.meso compared to non-transduced NK cells, at high E:T ratios of 5:1 and 10:1 (p < 0.0001
for both CAR.NK cell models in Figure 5.3a and Figure 5.3b). As anticipated, cytotoxicity of CAR.NK
cells increased with increasing E:T ratios, suggesting a positive correlation. For instance, in the case of
ameso.CAR.NK cells, we observed 2.35%, 9.59%, and 12.83% SKOV3.meso cell killing at 1:1, 5:1, and
10:1 E:T ratios, respectively (Figure 5.3a and Figure 5.3b).

143

Figure 5.3 CAR.NK cells conjugated with nanoparticles show similar or elevated levels of cytotoxicity
compared to nanoparticle free CAR.NK cells.
(a-b) Non-transduced NK cells (NK), ameso.CAR.NK cells and ameso.mbIL15.CAR.NK cells were either conjugated
with empty or SCH-loaded cMLVs, or tested without cMLV conjugation. NK cell were either co-incubated with
SKOV3 cells or SKOV3.meso cells for 7 h at different effector-to-target (E:T) ratios of (1:1, 5:1, and 10:1). NT cells
were used as negative controls. (a) Cytotoxicity of NK cells against target cells was measured. (b) Summary of percent
cytotoxicity in triplicates was represented in bar graphs (n = 3, mean ± SD; ns, not significant; *p<0.05; **p<0.01;
***p<0.001; ****p<0.0001, two-way ANOVA with Tukey’s multiple comparison). (c) Cytotoxicity of free SCH and
cMLV(SCH) against SKOV3 and SKOV3.meso human ovarian cancer cells in vitro. Wild type (SKOV3) and
mesothelin-overexpressing (SKOV3.meso) human ovarian cancer cells, each either as cells only (Cells), with free
SCH addition (Cells+SCH), or mixed with SCH-loaded cMLVs (Cells+cMLV(SCH)) were cultured in complete
RPMI medium. The same amount of SCH was used for Cells+SCH and Cells+cMLV(SCH) groups. Viability of
cancer cells was analyzed after 24h culture by using AnnexinV/7-AAD staining followed by flow cytometry analysis.
The percentage of AnnexinV+7-AAD+ cells was shown in bar graphs. Summarized statistics of triplicates were
represented in bar graphs (n = 3, mean ± SD; ns, not significant; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001,
two-way ANOVA with Tukey’s multiple comparison). (d) Cytotoxicity of free SCH and cMLV(SCH) against NK
cells. Non-transduced NK cells (NK), ameso.CAR.NK cells and ameso.mbIL15.CAR.NK cells, each either not
conjugated with cMLVs (Cells), with free SCH addition (Cells+SCH), mixed with SCH-loaded cMLVs
(Cells+cMLV(SCH)) or conjugated with SCH-loaded cMLVs (Cells-cMLV(SCH)) were cultured in IL-2
supplemented NK cell media. The same amount of SCH was used for Cells+SCH group and cMLV(SCH) including
groups. Viability of NK cells was analyzed after 24h culture by using AnnexinV/7-AAD staining followed by flow
cytometry analysis. The percentage of AnnexinV+7-AAD+ NK cells was shown in bar graphs. Summarized statistics
of triplicates were represented in bar graphs (n = 3, mean ± SD; ns, not significant; *p<0.05; **p<0.01; ***p<0.001;
****p<0.0001, two-way ANOVA with Tukey’s multiple comparison).

When we compare SKOV.meso target cell killing abilities of ameso.CAR.NK cells and
ameso.mbIL15.CAR.NK cells at different E:T ratios, ameso.CAR.NK cells yielded in slightly higher
cytotoxicity at 5:1 E:T ratio, whereas cytotoxicity of ameso.mbIL15.CAR.NK cells were higher at 10:1 E:T

144
ratio (Figure 5.3b). With SKOV3.meso target cells, CAR.NK cells conjugated with SCH-loaded liposomes
showed the greatest killing ability at each E:T ratio, except the low E:T ratio of 1:1 (Figure 5.3b). For
example, conjugation of SCH loaded liposomes onto CAR.NK cell surface significantly increased cell
killing percentages of both ameso.CAR.NK cells (21.23±1.34%, p<0.0001) and ameso.mbIL15.CAR.NK
cells (20.30±3.69% p<0.05), compared to their liposome-free counterparts (12.83±1.36% for
ameso.CAR.NK cells and 15.97±2.82% for ameso.mbIL15.CAR.NK cells), at 10:1 effector-to-target ratio,
whereas empty cMLV conjugation did not have any significant effect on cell killing ability of NK cells
(Figure 5.3b). On the other hand, it is important to note that conjugation of SCH-loaded cMLVs did not
have a major effect on cell killing in non-transduced NK cells, as well as at 1:1 E:T ratio in all groups,
indicating that SCH does not boost cytotoxicity in the lack of CAR signaling (Figure 5.3b). All together
these data indicated that although empty cMLVs do not affect CAR.NK cell function, the release of SCH
from cMLVs in proximity to the target cells helped CAR.NK cells increase their cytotoxic effects which
were induced by the CAR-mediated response against mesothelin
+
target cells.
Taken all together, the conjugation of cMLVs to CAR.NK cell surface does not hinder the recognition of
target cells, their IFN-γ secretion, degranulation, cancer cell killing abilities, and migration.
5.4.8 Presence of free SCH or cMLV(SCH) does not exert cytotoxicity on SKOV3.meso
cells or NK cells in vitro
We performed a cell viability assay to test whether SCH induces toxicity on NK cells themselves
or SKOV3 and SKOV3.meso cancer cells. Briefly, free SCH and SCH-loaded cMLVs were incubated with
these cells and cell viability was measured by AnnexinV/7-AAD staining.
In Figure 5.3c, we studied the in vitro cytotoxicity of free or cMLV-loaded SCH in ovarian cancer
cells. Either free SCH or the presence of cMLV(SCH) in the media did not induce SKOV3 cell death. In
SKOV3.meso group, the presence of cMLV(SCH) had no significant effect on cell viability compared to
cells in SCH-free media. Also, the addition of SCH into cell culture media, unexpectedly lowered the
SKOV3.meso cell death (p<0.0001) in comparison to SKOV3.meso cells in no SCH added media. Although

145
SKOV3.meso cells displayed higher AnnexinV
+
7-AAD
+
percentage (2.36%, p<0.0001) compared to
SKOV3 cells (0.28%), these percentages were very low. This result confirms that SKOV3.meso cells were
not directly affected by SCH, indicating that SKOV3.meso tumor killing was mainly achieved by the effect
of SCH on CAR-engineered NK cells in in vitro cytotoxicity assays.
In Figure 5.3d, we examined the effect of free SCH, cell-unbound cMLV-loaded SCH and NK cell-
conjugated cMLV(SCH) on NK cell viability in vitro. No significant difference in non-transduced NK and
ameso.mbIL15.CAR.NK cell viability was observed with SCH exposure. In ameso.CAR.NK group,
AnnexinV
+
7-AAD
+
necrotic cell population was, unexpectedly, smaller with SCH addition (p<0.05)
compared to cells alone. On the other hand, ameso.mbIL15.CAR.NK cells mixed and conjugated with
cMLV(SCH) displayed significantly higher percentages of cell death (p<0.001 and p<0.0001, respectively),
which was in accordance with the results of Figure 5.1d. Overall, the percentage of AnnexinV
+
7-AAD
+

dead cells for all conditions were very low with values lower than 5%, indicating that the presence of the
adenosine A2A receptor antagonist SCH-58261 did not have any major effects on NK cell viability in vitro.
5.4.9 CAR.NK cells conjugated with SCH-loaded nanoparticles have greater antitumor
efficacy compared to cMLV(SCH)-free CAR.NK cells
To explore whether the conjugation of anti-mesothelin CAR.NK cells with A2a receptor
antagonist-loaded nanoparticles could improve antitumor efficacy of CAR.NK cells, we established a solid
tumor mouse xenograft model and treated tumor bearing mice. To establish tumors, NSG mice were
subcutaneously inoculated with human mesothelin overexpressing ovarian cancer cells (SKOV3.meso).
After inoculation, mice were observed every day and tumor size was measured. 24 days after tumor
inoculation, mice were randomized based on their tumor size and assigned into 6 groups of 5 mice each.
Then, intravenous injection of NK cells were performed through the tail vein in every 4 days and 4 times
total. Either only (1) cMLV(SCH), (2) non-transduced NK cells, (3) ameso.CAR.NK cells, (4)
ameso.mbIL15.CAR.NK cells, or (5) ameso.CAR.NK cells conjugated with SCH loaded liposomes
(ameso.CAR.NK.cMLV(SCH)), and (6) ameso.mbIL15.CAR.NK cells conjugated with SCH loaded

146
liposomes (ameso.mbIL15.CAR.NK.cMLV(SCH)) were injected intravenously on days 0, 4, 8, and 12
(Figure 5.4a). For all NK cell injections, 5 x 10
6
NK cells (non-transduced or CAR modified) were used
per treatment. For injections with cmLV(SCH)-conjugated CAR.NK cells, liposomes were conjugated to
CAR.NK cells at a 1000:1 cMLV-to-NK cell ratio, following the protocol provided earlier. Physical states,
body weight, tumor growth and survival of the mice were observed.

Figure 5.4 CAR.NK cells conjugated with SCH-loaded nanoparticles enhance antitumor efficacy compared to
nanoparticle free CAR.NK cells in a human ovarian cancer xenograft model.
(a) Schematic representation of the experimental procedure for tumor challenge and NK cell adoptive transfer. 2 x
106 SKOV3.meso cells were injected into the right flank of NSG mice. 24 days later, mice were randomized into six
groups (n = 5 mice per group) and treated with 4 injections of 5 x 106 NK cells (non-transduced or CAR modified) or
the same number of NK cells conjugated with SCH loaded liposomes in 1000:1 cMLV-to-cell ratio. (b) Tumor growth
analysis. Tumor size was measured in every 2 days until 33 days after first adoptive NK cell transfer (n = 5, mean ±
SD; ns, not significant; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001, two-way ANOVA with Tukey’s multiple
comparison). The summarized statistics for Day 31 and Day 33 were shown on the right and left side of the figure
legend, respectively. (c) Representative tumor growth data for Day 19, 25 and 31 after first ACT was represented in
bar graphs. (d) Change in body weight. Mice were weighed in every 2 days until 33 days after first adoptive NK cell
transfer. The percentage (%) of change in mouse body weight was calculated in comparison to the initial weight of
each mouse (n = 5, mean ± SD; ns, not significant; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001, one-way ANOVA
with Tukey’s multiple comparison). The summarized statistics for Day 31 and Day 33 were shown on the right and
left side of the figure legend, respectively.

To test whether the pharmacological inhibition of A2aR would prevent ameso.CAR.NK and
ameso.mbIL15.CAR.NK cell hypofunction in the adenosine-rich tumor microenvironment, we conjugated
SCH-loaded liposomes onto CAR.NK cell surface and injected these cell-liposome conjugates into NSG
mice with established SKOV3.meso tumors in every 4 days for 4 times total (Figure 5.4a). We monitored
tumor growth and mice survival rates for 33 days after the first ACT.

147
As seen in Figure 5.4b, animals in all treatment groups showed tumor progression. No significant
difference between tumor size was observed until 17 days after the first adoptive cell transfer. On Day 17,
we recorded a significantly large tumor size in mice treated with SCH-loaded liposomes (cMLV(SCH)) in
comparison to ameso.mbIL15.CAR.NK cell-treated mice (p<0.05). In the following two weeks, we
observed a maintained tumor growth control in all CAR.NK groups regardless of their drug-loaded
liposome conjugation status, whereas mice treated with cMLV(SCH) or non-transduced NK cells alone
showed significantly larger tumors (The detailed statistics can be found in Table 5.1). 15 days after the last
ACT (on day 31), the mice treated with ameso.CAR.NK.cMLV(SCH) had significantly small tumors
compared with mice treated with liposome-free ameso.CAR.NK cells. Overall, the greatest tumor growth
inhibition was achieved by SCH-loaded liposome conjugated anti-mesothelin CAR engineered NK cell
(ameso.CAR.NK.cMLV(SCH)) treatment. This result suggests that the A2AR antagonist drug which was
carried into the tumor areas via NK cells achieved an immune regulatory effect in the TME and helped
preserve effector functions of ameso.CAR.NK cells, which produced strong and prolonged tumor growth
control.
Notably, we observed a difference in tumor growth control between two CAR.NK cell types in this
study. As seen in Figure 5.4c, until 19 and 25 days after the first NK cell transfer, both CAR.NK cell types
exhibited similar antitumor effects with no significant difference among them. In fact, on day 19, the most
significant tumor growth control was achieved by ameso.mbIL5.CAR.NK and
ameso.mbIL5.CAR.NK.cMLV(SCH) treatments evident in significantly lower tumor sizes compared to
cMLV(SCH) (p<0.001 for ameso.mbIL5.CAR.NK and p<0.01 for ameso.mbIL5.CAR.NK.cMLV(SCH))
and non-transduced NK cell alone (p<0.05 for ameso.mbIL5.CAR.NK) treatments. In contrast,
ameso.CAR.NK treatment did not start showing its tumor growth inhibition effect on day 19 (no significant
difference in tumor size when compared to cMLV(SCH) alone and non-transduced NK cells alone). On
day 25, ameso.CAR.NK and ameso.CAR.NK.cMLV(SCH) treatments started to show tumor growth
control resulting in significantly smaller tumors compared to tumors in mice treated with cMLV(SCH)
(p<0.05 for ameso.CAR.NK and p<0.01 for ameso.CAR.NK.cMLV(SCH)). However, the tumor growth

148
inhibitory effect of ameso.mbIL5.CAR.NK and ameso.mbIL5.CAR.NK.cMLV(SCH) treatments were
similar or higher (p<0.01 for ameso.mbIL5.CAR.NK and p<0.01 for
ameso.mbIL5.CAR.NK.cMLV(SCH)). However, towards the end the study, as represented with tumor
size data on day 31, the tumor growth control effect of ameso.CAR.NK.cMLV(SCH) treatment got ahead
of the results of ameso.CAR.NK treatment with, and both cMLV(SCH) conjugated and unconjugated
ameso.mbIL5.CAR.NK treatments. This result was unexpected given the high in vitro cytokine secretion
and cytotoxicity of ameso.mbIL15.CAR.NK cells against target cells, but lack of immune cell persistence
in the TME is not replicated in in vitro conditions with great accuracy. In order to further explain the reasons
of lower antitumor efficacy of ameso.mbIL15.CAR.NK cells, the mouse body weight data and details about
the physical health states of mice in this group becomes informative.
Table 5.1 Statistical analysis summary of the tumor growth curve at each time point.

Statistics: The tumor growth curve was analyzed using two-way ANOVA with Tukey’s multiple comparisons test.
Significance of findings was defined as: ns: not significant; *p<0.05; **p<0.01; ***p<0.001, ****p<0.0001.

Throughout the in vivo experiment, we recorded the body weight of individual mice and calculated
the change in body weight compared to their pre-treatment weight as seen in Figure 5.4d. We observed a
significant drop in body weight of mice in both liposome free and cMLV(SCH)-conjugated

149
ameso.mbIL15.CAR.NK treatment groups. On Day 31, the drop in mice body weight in these groups were
significantly higher (p<0.01 for ameso.mbIL15.CAR.NK and p<0.05 for ameso.mbIL15.CAR.NK-
cMLV(SCH)) compared to ameso.CAR.NK. This finding indicate that animals responded to
ameso.mbIL15.CAR.NK cell treatment negatively, regardless of the cMLV(SCH) conjugation status. Also,
the change in mice body weight became more pronounced towards the end of the study. This time dependent
weight loss could be attributed to an accumulated toxicity in the body of mice treated with
ameso.mbIL15.CAR.NK cells, given that body weight assessments are a widely used way to observe
toxicity. The detailed statistics of the percent change in initial mice body weight can be found in Table 5.2.
Table 5.2 Statistical analysis summary of the percent change in initial mice body weight at each time point.

Statistics: The mice body weight change was analyzed using two-way ANOVA with Tukey’s multiple comparisons
test. Significance of findings was defined as: ns: not significant; *p<0.05; **p<0.01; ***p<0.001, ****p<0.0001.

Another key consideration was that during the course of the in vivo study, the physical health of
the mice in ameso.mbIL15.CAR.NK and ameso.mbIL15.CAR.NK-cMLV(SCH) treatment groups
deteriorated, which resulted in sudden deaths (Figure 5.5). We had found a total of 6 (out of 12) mice dead
in their cages (4 mice in ameso.mbIL15.CAR.NK group and 2 mice in ameso.mbIL15.CAR.NK-
cMLV(SCH) group), with no cases of oversized or ulcerated tumors recorded. We also euthanized 2

150
additional mice, which were ‘unhealthy in physical appearance’ in ameso.mbIL15.CAR.NK-cMLV(SCH)
group upon veterinarians recommendations. Considering that each experiment group had 6 mice, majority
of the mice in both ameso.mbIL15.CAR.NK (4/6) and ameso.mbIL15.CAR.NK-cMLV(SCH) (4/6) groups
were not in good health condition, which is also evident in their weight loss profiles. This unexpected
decline in physical health which eventually threatened the life of the animals was only observed in
ameso.mbIL15.CAR.NK and ameso.mbIL15.CAR.NK-cMLV(SCH) treatment groups.

Figure 5.5 Swimmer plot analysis of mice treated with CAR.NK cell transfers.
The response of individual mice to NK cell treatments are represented by horizontal bars. The length of the bars
represents the duration of survival, and the symbols at the end of each bar (explained in legend) indicates the reason
of the termination of observation. Mice in different treatment groups are grouped and color-coded. Dotted lines
represent the days in which NK cell treatments were administered (Day 0, 4, 8, and 12).

Given the weight loss and poor physical health profiles were both and only in
ameso.mbIL15.CAR.NK and ameso.mbIL15.CAR.NK-cMLV(SCH) cell treated mice, we believe that the
toxicity was not caused by the presence of cMLV nanoparticles or SCH, but it can be attributed to adverse
events caused by the expression of membrane-bound IL-15/IL-15Rα complex or its mediated effects in
mouse body. We believe that the limited number of mice in ameso.mbIL15.CAR.NK and
ameso.mbIL15.CAR.NK-cMLV(SCH) treatment groups affected the tumor growth data in later time points

151
of this experiment, by not allowing the observation of the cumulative tumor growth control effect of these
treatment groups.
As seen in the swimmer plot (Figure 5.5), majority of the mice were euthanized according to the
size end point. Aside from the size end point euthanasia, 3 mice out of 12 were euthanized in
ameso.CAR.NK and ameso.CAR.NK-cMLV(SCH) groups due to ulceration developed in their tumors.
Mice treated with ameso.CAR.NK-cMLV(SCH) group survived the longest, with 40 days, whereas mice
treated with ameso.mbIL15.CAR.NK-cMLV(SCH) treated mice survived the shortest, with 31 days.
5.4.10 ameso.CAR.NK cells conjugated with SCH-loaded nanoparticles elongated mice
survival

Figure 5.6 ameso.CAR.NK cells conjugated with SCH-loaded nanoparticles elongated mice survival compared
to nanoparticle free CAR.NK cells in a human ovarian cancer xenograft model.
Mice survival. The survival of SKOV3.meso ovarian tumor bearing mouse was monitored until Day 47 after first
adoptive NK cell transfer. Mouse survival curves for six different treatment groups (n = 5 mice per group) were
calculated using the Kaplan-Meier method and the survival curves of individual groups were compared by Log Rank
(Mantel-Cox) test.

As a consequence of the most distinguished tumor growth control achieved by
ameso.CAR.NK.cMLV(SCH) treatment, the survival of mice in this group was markedly improved
compared to ameso.mbIL15.CAR.NK.cMLV(SCH) (p<0.01), non-transduced NK (p<0.05), and
cMLV(SCH) (p<0.05) treatment groups (Figure 5.6). Although not significantly different, the mice treated
with cMLV(SCH) conjugated ameso.CAR.NK cells survived up to 40 days, whereas mice treated with
ameso.CAR.NK cells survived up to 35 days. Mice treated with ameso.mbIL15.CAR.NK,

152
ameso.mbIL15.CAR.NK.cMLV(SCH), ameso.CAR.NK and ameso.CAR.NK.cMLV(SCH) had a median
survival of 35, 31, 35 and 40 days after the first treatment, respectively. Both cMLV(SCH) and non-
transduced NK treatment had median survival of 33 days. Only ameso.CAR.NK.cMLV(SCH) treatment
significantly increased the duration of survival compared to cMLV(SCH) and non-transduced NK groups
(p<0.05 for both groups).
Overall, conjugation of cMLV(SCH) to ameso.CAR.NK cells remarkably extended tumor growth
inhibition compared with free cMLV(SCH) treatment, indicating higher efficiency in blocking the A2AR
pathway and preventing CAR.NK cell hypofunction.
5.5 Discussion
By integrating cell-based immunotherapy and nanomedicine, our CAR.NK-cMLV(SCH) system
targets the ovarian cancer TME using nanoparticles loaded with an A2a receptor antagonist delivered by
anti-mesothelin CAR.NK cells, with minimal toxicity.
In this study, we used NK cells engineered with two different third-generation CAR molecules
targeting human mesothelin that were constructed in a previous study of our group
516
. We confirmed the
stable CAR expression for both anti-mesothelin CARs, either without (ameso.CAR.NK) or with
(ameso.mbIL15.CAR.NK) membrane bound IL-15/IL-15Rα complex, after one cycle of freeze and thaw.
The multilamellar structure of the cMLV vesicles was previously confirmed by cryo-electron
microscopy
464,524
and dynamic light scattering-measured hydrodynamic size of empty cMLVs was reported
as 220±14.99 nm (mean diameter)
524
. Synthesized cMLVs were stably conjugated to CAR.NK cell surface
at an optimal conjugation ratio of 1000 cMLVs per NK cell. This conjugation was previously shown to
yield an average conjugation of approximately 150 nanoparticles per NK cell
521
. Here, we showed that the
conjugation efficiency in total cell population is 35%. Furthermore, researchers from our lab have
previously quantified internalization of cMLVs by NK cells, and concluded that the attachment of cMLVs
to NK cells did not trigger endocytosis of cMLVs and that NK cell-bound cMLVs remained at the cell

153
surface
521
. By PEGylation of cMLVs, vesicle stability was further improved, which also aids in extending
their blood circulation half-life
530,531
.
We verified that cMLV conjugation and SCH encapsulation in cMLVs did not have an adverse
effects on NK cell viability. Notably, ameso.CAR.NK cells showed markedly high viability compared to
ameso.mbIL15.CAR.NK cells. We anticipate that the higher percentages of early and late stage apoptotic
cells in ameso.mbIL15.CAR.NK cell group was possibly due to the nutrient deprivation in culture media
of these cells, as a result of their higher duplication rates.
In the absence of IL-2 cytokine which would stimulate NK cell survival and proliferation, non-
transduced and ameso.CAR-engineered NK cells did not increase their number, whereas membrane-bound
IL-15/IL-15Rα complex expressed on ameso.mbIL15.CAR.NK cells allowed these cells to expand.
Notably, the cell expansion analysis also indicated that drug loaded nanoparticle conjugation onto NK cell
surface did not interfere with cell proliferation kinetics in vitro over two weeks.
NK cells must respond to chemoattractants in tumor areas and infiltrate into tumors to generate an
anticancer immune response
536
. We demonstrated that CAR.NK cell migration towards chemoattractant
CXCL9 is higher compared to non-transduced NK cells, and that the conjugation of cMLVs to the cell
surface did not impede NK cell migratory abilities.
Although many CAR.NK studies present cytotoxicity analysis but not cytokine release
results
122,125,128,537,538
, here we demonstrated that CAR.NK cells secreted IFN-γ by TAA-specific activation.
Only the coculture of CAR.NK cells with target cells which expressed the cognate antigen, but not
cocultures with non-transduced NK cells or TAA
-
cancer cells showed IFN-γ release. Also, consistent with
a previous study of our group in which paclitaxel-loaded cMLV conjugation did not impair the cytokine
secretion of anti-CD19.CAR.NK cells
521
, anti-mesothelin CAR.NK cells yielded similar levels of IFN-γ
secretion with empty or drug-loaded nanoparticle conjugation. It is important to note that
ameso.mbIL15.CAR.NK cells exhibited significantly greater IFN-γ secretion compared to ameso.CAR.NK
cells in response to target cells. These findings are in accordance with another study demonstrating that IL-
15 serves in higher production and release of IFN-γ in NK cells
461–463,516
.

154
Similar to intracellular cytokine staining results for IFN-γ, only the coculture of anti-mesothelin
CAR.NK cells with mesothelin
+
target cells generated a high percentage of cytotoxicity, indicating the
necessity of the TAAs for the activation and functionality of CAR.NK cells. While comparable,
ameso.CAR.NK cells and ameso.mbIL15.CAR.NK cells showed different levels of cytotoxicity at different
effector-to-target (E:T) ratios. ameso.CAR.NK cells exhibited higher cytotoxicity at 5:1 E:T ratio, whereas
cytotoxicity of both ameso.CAR.NK and ameso.mbIL15.CAR.NK cells reached to its highest level at 10:1
E:T ratio, prompting the possible necessity of dose optimization in clinical use of these CAR.NK cells. It
is also important to restate that as an immune activating cytokine, IFN-γ release from NK cells initiate
broader immune responses by signaling neighboring lymphocytes, which can presumably be higher in
ameso.mbIL15.CAR.NK cells given their proliferation advantage over time
80
. Overall, this study agree with
the literature showing that the IFN-γ produced by activated CAR.NK cells promote target cell lysis. It is
important to note that free SCH and cell-unbound cMLV(SCH) were found not to adversely affect NK cell
viability or proliferation in vitro, which indicates that the drug-loaded cMLVs which could disconnect from
NK cell surface or the SCH that is released from nanoparticles over time did not exert any cytotoxicity on
NK cells when used at this dose. Moreover, either free SCH or the presence of cMLV(SCH) in the media
did not induce SKOV3 or SKO3.meso cell death in vitro. This result confirms that SKOV3.meso cell killing
in in vitro cytotoxicity assays was mainly achieved by CAR.NK cell-induced target cell lysis.
We also showed that while empty cMLVs did not have adverse effects in CAR.NK cell function,
the release of SCH from cMLVs in close proximity to NK cells target cells helped CAR.NK cells maintain
and possibly increase their effector cell functions in vitro. Our in vivo data also proved that CAR.NK cells
conjugated with drug-loaded nanoparticles slowed tumor growth more effectively than any other treatment
group, which provides evidence that CAR.NK cells mediated the delivery of the small molecule inhibitor
of A2a receptor, SCH-58261, to the tumor mass, which help combat the immunosuppressive TME.
The findings presented in this article are in agreement with previous research demonstrating that
liposomal nanoparticles with thiol-reactive maleimide headgroups can be conjugated to the thiol-rich

155
surface of immune cells successfully and that this can be achieved without altering effector functions and
transmigration capabilities of these cells
477,520,521,526
.
In order to achieve the maximum action of immune cell regulatory drug, SCH-58261, on NK cells
in vivo, (1) the drug-loaded nanoparticles must be able to reach the tumor areas, (2) remain in circulation
for longer durations, and (3) the drug must be released from the nanoparticles with high efficiency. It is
known that the conjugation of nanoparticles onto cells increase their circulation time and delays their
clearance from the body. In a previous study of our group, in vivo biodistribution investigation of
fluorescently-labeled cMLV nanoparticles that were conjugated onto anti-CD19 CAR.NK cells showed that
CAR.NK cells enhanced nanoparticle accumulation within CD19-overexpressing ovarian tumors and
achieved longer retainment of cell conjugated nanoparticles in circulation
521
. The elongated surface
retention of nanoparticles on the carrier cells allowed the sustained release of the chemotherapeutic drug,
paclitaxel, from nanoparticles, and hence achieved a longer therapeutic effect
520,521,526
. In another work of
our group, CAR T cell-conjugated cMLVs demonstrated sustained SCH release over five days in vitro
477
.
In addition, CAR T cell-conjugated cMLV(SCH) treatment led to a higher concentration of SCH in tumor
tissue compared to SCH delivery with free cMLV(SCH)
477
. The same investigation was not repeated in this
study given the use of identical cMLV nanoparticles and the same mechanism of TAA-directed tumor
recognition by mesothelin-directed CAR.NK cells, which was attested by TAA-specific IFN-γ secretion
and tumor killing in vitro. In the light of the prior findings of our group, we are confident that our
combination therapy strategy with the CAR.NK-cMLV drug delivery system which aims to carry the A2aR
antagonist SCH-58261 to adenosine rich tumor areas, meet all above mentioned conditions for exploiting
the maximal SCH-58261 drug action in vivo, and we proved this argument in antitumor efficacy
experiments in vivo.
The small molecule immunomodulatory drug, SCH-58261 (SCH), is one of the most selective and
potent antagonists of A2aR. The clinical application of SCH has been hindered by the drug’s poor solubility
and suboptimal in vivo pharmacokinetic profile, which results in low bioavailability of SCH. For example,
in a study where SCH-58261 was orally administered to rats, it was shown that approximately 28% of SCH-

156
58261 was excreted in the first 24h without absorption, likely due to no/little GI absorption, and its high
metabolization by the liver in vivo
350
. Examples of reported dosage of SCH-58261 in in vivo mouse studies
can be two studies investigating the efficacy of the A2AR inhibition effect of SCH-58261 in Parkinson’s
disease and in pulmonary adenocarcinoma in xenograft models, where both studies used 2 mg/kg SCH-
58261 per intraperitoneal administration daily in a 20 day treatment regimen
338,361
. As another example of
the in vivo use of SCH-58261, rats were dosed with 0.1, 1, 3, and 10 mg/kg of SCH-58261 through
intraperitoneal injection
355
. In this present work, we only used a total of 2 mg SCH-58261 per mouse. Given
the conjugation yield of 150 nanoparticles per NK cell at 1000:1 cMLV-to-cell ratio
521
, 35% conjugation
efficiency in total cell population, and the gross 2 mg SCH used in preparation of 20 x 10
6
NK-cMLV
conjugates injected per mouse in total of four injections, we calculated the total amount of SCH received
by each mouse as 0.105 mg. In this present study, this SCH amount was sufficient to exert its A2aR
blockage effect while not introducing any side effects. Taking the detractive effect of drug loading and drug
release efficiency, as well as the biodistribution and clearance of SCH in mouse body into consideration,
the amount that reaches to tumor areas is even less. Given the high amounts of SCH used in in vivo
experiments in the literature
338,361
, our system allows the use of a minimal amount of SCH for desired effect.
For example, in one study where 40 mg/kg SCH was used in total of 20 i.p. injections, which corresponds
to 0.8 mg SCH per mouse (assuming 20 mg mouse weight), our system successfully used only 13% of this
reported amount. We also showed that the low dose of SCH-58261 used in this study did not kill NK cells
or decrease their proliferation, while providing the intended inhibition of adenosine binding to A2aR on
NK cells which helped preserve NK cell effector functions in vivo. This indicates that encapsulation of the
A2aR antagonist SCH-58261 in cMLVs help drug exert its maximal effect at a lower dose, presumably by
improving its solubility and pharmacokinetics profiles by directly delivering the drug to its place of action
in a protected manner.
We demonstrated that within the two weeks after first adoptive cell transfer, CAR.NK cells
exhibited a maintained tumor growth control, whereas mice treated only with cMLV(SCH) or non-
transduced NK cells showed significantly larger tumors. In the following two weeks, CAR.NK cell

157
treatment groups showed similar levels of tumor growth inhibition with a time dependence. The more
significant tumor growth control effect was achieved with ameso.mbIL15.CAR.NK cell treatments
regardless of cMLV(SCH) conjugation status in the early time points of our observation (e.g. day 19 after
first ACT). In later time points, the tumor growth inhibition effect of ameso.CAR.NK groups reached to a
similar levels with ameso.mbIL15.CAR.NK treatments, with higher antitumor efficacy achieved by
cMLV(SCH) conjugation for both CAR.NK groups. Towards the end of the experiment, for example on
day 31 after first ACT, the mice treated with SCH-loaded nanoparticle conjugated ameso.CAR.NK cells
showed dramatically slower tumor growth compared with the mice treated with liposome-free
ameso.CAR.NK cells. The success of ameso.CAR.NK group was also evident in their higher migration
levels, as well as the achieved high toxicity even with lower effector-to-target ratios in vitro. Overall, the
greatest tumor growth inhibition was achieved by SCH-loaded liposome conjugated anti-mesothelin CAR
NK cell treatment, suggesting that the A2aR antagonist drug which was carried into the tumor areas via NK
cells demonstrated a pharmacological inhibition of A2aR and prevented ameso.CAR.NK cell hypofunction
in the adenosine-rich tumor microenvironment. In addition, as a consequence of the most distinguished
tumor growth control achieved by ameso.CAR.NK.cMLV(SCH) treatment, the survival of mice in this
group was markedly improved, which further proves higher efficiency in blocking the A2aR pathway and
prevention of CAR.NK cell hypofunction. Another major advantage of blocking the A2aR pathway with
our CAR.NK-cMLV(SCH) therapeutic system is the ability of the A2aR antagonist drug to exert its effect
both on carrier CAR.NK cells themselves as well as on other endogenous NK and T cells in the tumor site,
and show its effect in the TME in a more comprehensive way.
Also, in the early time points of our tumor growth observation, both CAR.NK treatment group
performed in a level similar to that of non-transduced NK group. This result was unexpected given the in
vitro cytokine secretion and cytotoxicity of CAR.NK cells against target cells, but lack of immune cell
persistence in the TME is not replicated in in vitro conditions with great accuracy. Also, it is also known
that NK cells, as innate immune cells, have major roles in cancer immune surveillance and are able to
eliminate tumor cells without prior sensitization. They form immune synapses with target cells, release

158
preformed cytolytic granules and induce cell lysis
539
. Thus, although not as strong as TAA-associated cell
lysis, a low level cell killing is typical in non-transduced NK cells.
It is important to point out that mice in both liposome free and cMLV(SCH)-conjugated
ameso.mbIL15.CAR.NK cell treatment groups, compared to ameso.CAR.NK cell treated mice, displayed
a significant weight loss in addition to a decline in their physical health which resulted in sudden deaths.
Given that the weight loss and poor physical health profiles were only observed in ameso.mbIL15.CAR.NK
and ameso.mbIL15.CAR.NK-cMLV(SCH) cell treated mice, we believe that the accumulated toxicity is
not caused by the presence of cMLV nanoparticles or SCH. This toxicity can be attributed to adverse events
caused by the expression of membrane-bound IL-15/IL-15Rα complex or its mediated effects in mouse
body, although the mechanism of action would benefit from further investigation.
In the previous work of our group in which ameso.mbIL15.CAR construct was first prepared and
the antitumor efficacy of ameso.mbIL15.CAR-engineered NK cells were tested in vivo, we administered a
total of 30 x 10
6
NK cells into tumor bearing mice in three injections within 9 days
516
. Although not
published, sudden animal deaths were observed in ameso.mbIL15.CAR.NK treated mice alone, similar to
what we redemonstrated here, which implies a treatment induced toxicity, presumably due to the reported
high degree of CAR.NK cell accumulation in bone marrow
516
. In this work, we used the same number of
CAR.NK cells for ACT, which was administered in 4 injections in 12 days (5 x 10
6
cells per injection).
Given the controlled tumor growth achieved in these two studies, we are confident that the treatment
regimen and dose (i.v. injection of 20 x 10
6
CAR.NK cells in 4 injections and 30 x 10
6
CAR.NK cells in 3
injections or 3-4 days apart) is effective to achieve the desired response. It should be noted that 20 x 10
6

cells instead of 30 x 10
6
cells that was used in the first study was sufficient to achieve the desired tumor
growth control and the antitumor effect of the lower number of CAR.NK cells was also enhanced by the
surface conjugated cMLVs loaded with the A2aR antagonist SCH, which aids in the durable cytotoxic
response. This finding also emphasizes the importance of further toxicity analysis if the
ameso.mbIL15.CAR.NK cells are to be used in clinic. Given the difference in the dose dependent in vitro
cytotoxic effect between ameso.CAR.NK and ameso.mbIL15.CAR.NK cells (greater target cell lysis with

159
ameso.mbIL15.CAR.NK group in lower effector-to-target ratio compared to ameso.CAR.NK group), one
explanation to the sudden animal deaths in ameso.mbIL15.CAR.NK-treated mice could be a dose-
dependent toxicity which likely precipitated other side effects in mouse body. The ameso.mbIL15.CAR.NK
cell treatment-induced toxicity can be addressed by methodically changing the dose and frequency of
CAR.NK cell injections and carefully monitoring the health of the patients. On the other hand, these
findings indicate that as a response to declined patient health or unwanted toxicity, CAR.NK therapy may
be required to be stopped. This would be possible with a CAR design which has the ability to trigger the
apoptosis of injected CAR.NK cells in case of severe toxicity and therefore improve the safety in cancer
treatment. A great alternative could be using ameso.mbIL15.CAR.NK cells which also express an inducible
suicide gene caspase 9 (iCAS9) as a safety switch which responds to a chemical inducer of dimerization
(CID) and trigger apoptosis in cells expressing the iCAS9 transgene
516
.
Also, the feasibility and efficacy of this platform in cancer treatment can be investigated in other
tumor models. In this study, SKOV3 ovarian cancer model was selected due to high expression of CD39
and CD73 as well as high production of adenosine. For example, the antitumor efficacy of this platform
can be investigated in other human solid tumor models in which high adenosine levels were reported, such
as A549 lung and, T-84 and TH-29 colon cancer models
298,540
. Similarly, the effect of the SCH-58261
carried in CAR.NK cell conjugated cMLVs can be tested in breast, prostate and brain cancers which were
reported for high levels of CD73 expression
173,541
.
Surface modification of immune cells has enabled the delivery of various therapeutic agents to the
tumor site. To the best of our knowledge, this present study is the first in literature that utilizes CAR-
engineered NK cells as active carriers for the delivery of an immune regulatory drug to the TME. We
believe that this novel platform which combines nanomedicine with immunotherapy, is widely applicable
and its use could be expanded to different clinical settings with further modifications and improvements.
By making changes in different components of this system which consists of the cargo molecule, drug
carrying nanoparticles and CAR-engineered NK cells, various cancer types can be treated with various
intervention strategies. For instance, various cancers with different TAAs could be targeted by changing

160
the tumor targeting scFv region of CARs on NK cells. In addition, by the use of different therapeutic cells,
such as CAR T cells as tumor specific active-carriers of drug loaded cMLVs, improved antitumor effects
can be achieved by the maximal effect of both the effector immune cells and carried-cargo, such as
cytokines, small-molecule inhibitors, or chemotherapeutics
520
.
We believe that this system can be used for the co-delivery of other immunomodulators and
chemotherapeutic agents aiming dual immune regulatory and tumor-killing effect in the tumor
microenvironment. As an example, in order to increase the antitumor efficacy of this tumor-targeted
CAR.NK-cMLV(SCH) therapeutic system and given that A2aR signaling can be inhibited by various drugs
interfering with other components of adenosine axis, such as drugs for blockade of adenosine production
and A2aR/A2bR antagonism, multiple drugs can be codelivered to the adenosine-rich TME in cMLVs with
potential synergistic effects. Similarly, different immune modulators targeting both A2aR signaling and
other immunosuppressive pathways can be codelivered in cMLVs in order to generate a sufficiently strong
response to reduce the immune suppressing effects of the TME. Also, immune modulators can be delivered
in combination with immune checkpoint inhibitor antibodies, such as anti-programmed cell death protein-
1 (PD-1), to further improve antitumor response
317,333
. For example, it is known that various clinical studies
have demonstrated positive outcomes in treatment of ovarian and lung cancers when small-molecule
inhibitors of VEGF and EGFR were used in combination with anti–PD-1 and anti–PD-L1
542
. Furthermore,
as shown in various models, chemotherapeutics, such as doxorubicin and paclitaxel can also be actively
delivered to tumor sites with a cell-mediated delivery platform
464,521
. Also, the potential of RNA molecules
(e.g., siRNA, microRNA, and mRNA) for immunomodulation and cancer immunotherapy due to their
ability to silence or upregulate immune-related genes, has been shown recently. Correspondingly, various
nanomaterials, including liposomes and lipid based nanoparticles, have been developed to improve RNA
drug delivery to the tumor and immune cells for the induction of antitumor immune responses
543
. Given
that the platform presented in this work, with its active-delivery capability of both hydrophilic and
hydrophobic drugs to tumors in lipid based multilamellar cMLV nanoparticles, can be investigated for its
use to deliver immune-modulatory RNA therapeutics to improve cancer treatments. Alternatively, various

161
forms of IL-15 which are encapsulated in different nanoparticle designs have been shown to have high
therapeutic potential with their better tumor control by T cell expansion and reduced tumor cell proliferation
through activation of the host immune system or T cell carriers. Some examples are the delivery of IL-15
in calcium alginate nanoparticles, IL-15-encoding mRNA in a protamine/liposome system, or IL-15 super-
agonist complex (IL-15Sa) in nanoparticles attached to T cells via anti-CD45 antibodies, to tumor areas
544–
546
. The in vivo toxicity that was displayed in mice treated with membrane bound IL15/IL15a complex
expressing ameso.CAR.NK cells (ameso.mbIL15.CAR.NK) could potentially be solved by the
encapsulation of IL-15/IL-15a complex in cMLVs.
NK cells are potent cytotoxic effector cells for cancer therapy. Primary human NK cells are difficult
to isolate and expand due to cell source related limitations and donor dependent variable yield. Thus, easily
expanding, stable and immortalized clonal NK cell lines have been generated. Specifically, NK-92 cell line
is the most prominent with its robust high antitumor cytotoxicity, ability to be genetically manipulated
easily
93
and promising safety results of phase I cancer treatment trials NCT00900809 and
NCT00990717
98,99
. Also, NK-92 cell line has been found to be safe in patients upon irradiation
98,99
. Another
huge advantage of using NK-92 cell line for treatment is its “off-the-shelf” availability. The lack of long
term and labor-intensive culture systems, as standard in the case of engineered T cells, would mean
significantly lowered cost implications for NK cell therapies. With the optimized and streamlined
techniques, and the flexibility of NK-92 cells to be frozen and shipped to the treatment site, NK cell
immunotherapy would be adopted as a clinical therapy in the future. The preparation and administration
costs of NK-92 cells are significantly lower than autologous or allogeneic NK cells, as well as CAR T cells.
According to a recent estimation of the total cost of care associated with the administration of CAR T cell
therapy in April 2020, ranges between $420,000 and $455,000
547
. In contrast, treatments with engineered
NK-92 cells are believed to cost on the order of $20,000, with the option of repeated infusions
101
. Recently,
NK cell lines have gained significant interest for CAR NK cell therapy as alternative effector cells. The
NK-92 cell line based “off-the-shelf” CAR technology allows to generate large numbers of CAR.NK cells
which are donor-independent and specific to the TAAs
101
. Preclinical studies in xenograft mice models with

162
CAR-modified NK-92 have been demonstrated to eliminate different malignancies such as lymphoma
(CD19.CAR
120
), breast cancer (Her2.CAR
122
), neuroblastoma (GD2.CAR
123
), and glioblastoma
(EGFR.CAR
124
), among many. Most of the studies for CAR.NK-92 cells have used first-generation
CARs
120,122,128,130
, although second-generation CAR-NK cells are not uncommon
121,122,128,130
. Third-
generation NK-92 CARs comprised of CD28, 4-1BB, and CD3ζ intracellular signaling domains resulted in
high efficacy levels
104,132
and inhibition of disease progression in xenograft mouse models
134
.
Cellular immunotherapy has attracted much attention as fourth pillar of cancer treatment. As
opposed to the high number of clinical trials using CAR T cells for hematological malignancies, only a few
clinical trials are using CAR T cells to treat patients with relapsed and refractory solid tumors, such as
ovarian, lung, pancreatic and prostate cancers
548
. CAR.NK cell clinical studies in solid malignancies are
even more rare with eight studies focusing on ovarian, brain, lung, prostate and pancreatic cancer, to date
549
.
In one study (NCT 03056339), NK cells derived from umbilical cord blood (CB) were genetically modified
to express antiCD19-CD28-CD3ζ-CARs, with a iCasp9 safety switch and IL-15 expression. These cells
were administered to patients suffering from relapsed and/or refractory B-cell lymphoma or leukemia.
Although primary NK cells are a preferred cell source due to safety concerns, and that CAR-expressing NK
cells are in their early clinical trials, we believe that CAR.NK-92 cells are strong candidates that would
offer an alternative “off-the-shelf” cellular immunotherapy with increased specificity towards TAAs in
solid tumors, which would also potentially solve expansion, cost, shipment and storage issues of
immunotherapy. It is clear that the NK-92 platform has been proven to be effective even in its unmodified,
monotherapy form. By various combination therapy approaches, NK cell therapies could be used to further
augment immune responses. Additional improvements could be achieved through genetic modifications,
i.e. CAR modification on NK-92 parental cells, or conjugation of NK cells with other agents such as
checkpoint inhibitors, immunomodulatory drugs and chemotherapy drugs. Given the universal
administration and off-the-shelf therapy advantage of NK-92 platform, it can also provide the possibility to
provide personalized treatment regimens to target specific patient needs with combinatorial therapy
approaches.

163
Overall, this study demonstrates a promising combination of CAR-targeted immunotherapy and
drug delivery for enhanced antitumor treatment by means of overcoming the immunosuppressive TME via
interfering with adenosine axis.

164
6 Chapter 6: Conclusions and Future Perspectives
Due to the exceptional success of adoptive CAR T cell therapies for treating hematological
malignancies
47
, four CD19-targeted (Kymriah™, Yescarta™, Tecartus™, Breyanzi™) and one BCMA-
targeted (Abecma™) CAR T cell products received approval from the US Food and Drug Administration
(FDA)
48–50,52,54
. These advancements are expected to help patients with relapsed and/or refractory B-cell
lymphoma, acute lymphoblastic leukemia, follicular lymphoma, mantle cell lymphoma, and multiple
myeloma. Researchers have been working to reinstate the remarkable success of adoptive transfer of tumor-
specific lymphocytes in the treatment of malignancies beyond lymphoma, leukemia and myeloma. To
achieve this goal, the CAR-engineered immune cell therapy field has been active in eliminating the
limitations of these therapies such as failures in durable remissions, resistance in therapy by antigen escape,
and treatment related toxicities
55
.
The use of CAR-engineered T cell therapy in solid tumors has had poor clinical results due to
various reasons. Researchers have been working on identifying the reasons why CAR T cell therapy fails
in solid tumors and finding intelligent solutions. As detailed in Chapter 1, the major challenges in CAR T
cell therapy against solid tumors are tumor antigen heterogeneity, T cell trafficking and infiltration into
tumor tissue, and the immunosuppressive tumor microenvironment
34
. In the projects presented in this
dissertation, my and my colleagues’ goal was to find innovative solutions to overcome the challenges of
CAR-engineered immune cell therapies in solid tumors.
In the project detailed in Chapter 2 Adnectin-Based Design of Chimeric Antigen Receptor for T
Cell Engineering, we constructed adnectin-based CARs, as an alternative to using single chain variable
fragments (scFv) as extracellular antigen recognition domains. Adnectin, a class of affinity molecules
derived from human fibronectin, was used to construct four adnectin-based epithelial growth factor receptor
(EGFR)-targeting CARs, with different adnectin sequences. In vitro, the EGFR-targeting adnectin-based
CAR-engineered T cells had equivalent efficacy against target cancer cells compared to the scFv-based
CAR T cells, in which the scFv sequence was derived from an FDA-approved chimeric monoclonal

165
antibody against human EGFR, Cetuximab. Also, the adnectin-based CAR T cells had comparable
antitumor efficacy to their scFv-based counterpart in xenograft tumor-bearing mice in vivo. Very
importantly, with optimal affinity tuning, adnectin- based CAR showed higher selectivity on target cells
with high EGFR expression than on those with low expression.
The overarching goal of this study was to overcome the inherent limitations of conventional CAR
design and address the lack of a more efficient method for novel construct design. Our study has presented
a novel and feasible method to derive CAR constructs from adnectin sequences which has potential value
for future CAR development. This study served as a proof-of-concept that CAR constructs are not limited
to existing scFvs, many of which are murine-derived and may pose a risk of high immunogenicity in
patients
410,412,413
. Therefore, using human-derived proteins instead of nonhuman scFV as the ectodomain in
CAR designs offers an opportunity in preventing the unwanted immune responses to CARs.
In addition, traditional CARs with high-affinity scFvs derived from monoclonal antibodies show a
poor discrimination power among target cells with different levels of antigen expression
432,433
. The poor
discrimination power among target cells with different levels of antigen expression may result in on-target
off-tumor toxicity in clinical trials
434
. Previous studies have provided evidence that by tuning down the
affinity of CARs toward the target antigen via low-affinity scFvs, CARs could distinguish tumor cells from
normal tissues and only recognize and eradicate tumor cells with high expression levels of target
antigens
429,430,432
. Out of the four adnectin-based EGFR CARs constructed in this study, E3-CAR displayed
a similar level of biological function in vitro as that of high-affinity scFv CARs, and its relatively low
affinity led to enhanced selectivity toward EGFR-overexpressing tumor cells over normal cells with
endogenous level of EGFR expression. Therefore, our study demonstrated that developing low-affinity
adnectin-CARs with enhanced tumor selectivity might be a promising strategy to improve the safety profile
of CAR T therapy. It is important to note that E3-CAR had a lower percentage of tumor-infiltrating T cells
compared to Cetux-CAR, although the difference was not statistically significant, and it may compromise
the antitumor efficacy of E3-CAR T cells in vivo. This finding highlights the importance to optimize the

166
affinity range for the antigen-recognition moiety of CAR to enhance CAR specificity to tumors while
eliminating on-target-off-tumor toxicity.
Another contribution of these adnectin-based CAR constructs could be their potential utility in
combinational antigen recognition by bispecific CARs which may provide a safeguard against antigen
escape of cancer cells and the resulting relapses in patients after CAR T cell therapy. Bispecific CARs, such
as the multi-domain adnectin bispecific to EGFR and insulin-like growth factor 1 receptor (IGF-IR) in a
previous report
424
, may successfully target heterogeneous antigen expressing-cancer cells and enhance the
efficacy of CARs. Importantly, the small size of adnectin and its native structure derived from fibronectin
make it very adaptable to develop a multi-domain adnectin that is multi-specific to different targets
424
. In
the future, it will be necessary to evaluate the reactivity of adnectin-based CAR T cells to normal tissues in
a clinical setting. To achieve a balance between the efficacy and safety of adnectin-based CAR constructs,
their affinity range should be optimized and the therapeutic efficacy of low-affinity adnectins must be
verified in the clinic.
In Chapter 3, the study named Engineering CAR-expressing Natural Killer Cells with Cytokine
Signaling and Safety Switch used synthetic biology to include additional components alongside the CAR
itself. An anti-mesothelin CAR, a membrane-bound IL-15/IL-15Rα complex, and an inducible caspase 9
“kill switch” were expressed in natural killer cells for tumor-targeting capabilities, immunostimulatory
effects, and safety in treating a preclinical model of ovarian cancer with a renewable, allogenic cell therapy.
The addition of a membrane-bound cytokine to the CAR construct is one method to program immune cells
to be more therapeutically effective by improving their survival and proliferation. Natural killer (NK) cells
typically rely on interleukin (IL)-2 for growth and cytotoxic function; however, the co-administration of
exogenous IL-2 with NK cells in animal studies or clinical trials may cause acute toxicity in high doses
453
.
Alternatively, viral vectors encoding IL-15 can transduce NK-92 cells to stably express the cytokine. These
cells sustain continued proliferation and cytotoxic capabilities without exogenous IL-2
121,456
. Moreover, the
stable complex formed between receptor subunit IL-15Rα and membrane-bound IL-15 enables autocrine
stimulation of NK cells
457
. In this study, anti-mesothelin CARs, with or without membrane-bound IL-15/IL-

167
15Rα complex were transduced efficiently in NK-92 cells. In the absence of exogenous IL-2, genetic
engineering of CAR-NK cells with membrane-bound IL-15/IL-15Rα complex allowed CAR-NK cells to
survive and proliferate in vitro. In addition, CAR-NK cells with IL-15/IL-15Rα had more IFN-γ
+
cells in
response to mesothelin-expressing target cells. However, NK.αmeso.mbIL15 cells killed lower percentages
of target cells compared with NK.αmeso cells. In vivo, CAR-NK cells with IL-15/IL-15Rα controlled tumor
growth more successfully and also proliferated more in vivo compared to NK.αmeso cells. Interestingly,
NK.αmeso.mbIL15 cells homed to the bone marrow more than NK or CAR-NK cells. Specific
accumulation of membrane-bound IL-15/IL-15Rα-expressing NK-92 cells might make these cells a
potential therapy for malignancies which heavily metastasize to the bone marrow
120
.
It has been previously demonstrated that T cells which were engineered to express an inducible
suicide gene caspase 9 (iCAS9) as a safety switch responded to a chemical inducer of dimerization (CID)
and showed dose-dependent apoptosis
451,452
. This suicide switch is a tool for physicians to regulate treatment
when complications including “on-target, off-tumor” toxicity or cytokine release syndrome occur
410
. In this
study, we further transduced the engineered CAR.NK cells with the iCAS9 suicide gene. We showed that
iCAS9, caused NK cell ablation in response to CID both in vitro and in vivo. In addition, mice treated with
CID had minimal NK cell populations in the bone marrow. It is important to mention that a small percentage
of cells remained viable after in vitro CID treatment, likely due to iCAS9
−
NK cells which remained in the
sorted population or which gradually lost iCAS9 expression over time. This emphasizes the importance of
complete iCAS9 expression if the cells are to be used in a clinical setting.
Synthetic biology can be used not only to fine-tune the CAR itself, as described in Chapter 2, but
also to build additional components into the system. Additionally, synthetic biology can increase
persistence through autocrine signaling of immunostimulatory cytokines and to create a safety mechanism
in case adverse side effects necessitate the termination of the therapy.
The well-characterized cytotoxic NK cell line, NK-92, is promising due to: (i) its lack of
immunogenicity, (ii) robust high antitumor cytotoxicity, (iii) ability to be genetically manipulated easily,
and (iv) availability as an “off-the-shelf” product
93,98,99
. It is also proven as safe in phase I cancer treatment

168
trials. Another huge advantage of using NK-92 cell line for treatment is its “off-the-shelf” availability
453–
455
. The lack of long term and labor-intensive culture systems, as standard in the case of engineered T cells,
would mean significantly lowered cost implications for NK cell therapies. With the optimized and
streamlined techniques, and the flexibility of NK-92 cells to be frozen and shipped to the treatment site,
NK cell immunotherapy would likely be adopted as a clinical therapy in the future
453–455
.
In Chapter 4, in the study named CAR T Cell-Platelet Complexation for Enhanced Tumor Homing
and Antitumor Efficacy in Solid Tumors, we employed platelets in complex with CAR T cells to augment
CAR T cell infiltration into tumor sites in a xenograft mouse model. We demonstrated how CAR T-PLT
complexes harness the power of the intrinsic tendency of PLTs to traffic towards and infiltrate into tumors,
which transforms the inadequate infiltration of T cell in tumors to a permissive one.
The role of platelets in blood clotting have been long appreciated
180
. However, while the larger
impact of platelets on immune response by the interaction they form with leukocytes
218
is increasingly
evident, it is perhaps not yet fully appreciated. Activated platelets release lipid membrane vesicles called
platelet-derived-microparticles (PMPs)
194,216
and contribute to immune function by releasing soluble factors
and chemokines that recruit, localize, or activate immune cells
182,215
. It has been also suggested that platelets
can facilitate the recruitment of lymphocytes at a site of inflammation or infection, which is known as a
central step in T cell trafficking
182,199,215,217,232,236
. We utilized platelets in complex with CAR T cells to
enhance the trafficking and accumulation of infused CAR T cells into solid tumors.
The cell-cell complexes we formed consisted of CAR T cells and activated PLTs, or CAR T cells
and PMPs. To ensure there was no interference with the IFN-𝛾 release, degranulation and lytic activity of
CAR T cells against target cells, we verified that complex formation between CAR T cells and PLTs or
PMPs.
CAR T cells may be prevented from penetrating tumors through the vascular endothelium
497
due
to a collection of mechanisms in solid tumor tissues
498–500
. For infiltration, T cells attacking solid tumors
need to actively degrade the main components of sub-endothelial basement membrane, including heparan
sulphate proteoglycans (HSPGs) by releasing heparanase to access tumor cells
470
. PLTs express

169
heparanase
472
and activated platelets release acid hydrolases including heparanase from their lysosomes
473
.
In this study, we demonstrated that in vitro CAR T cell invasion through a Matrigel with high HSPG content
was improved in CAR T-PLTs group. This finding suggests the HSPG-degrading effect of heparanase
introduced by PLTs in CAR T-PLT complexes.
Using an immunocompetent human lung cancer xenograft model, we found that the intravenous
injection of CAR T cell-PLT complexes significantly reduced tumor growth by effective accumulation of
CAR T cells into tumors. Given this finding and the improved T cell invasion in Matrigel with high HSPG-
content in CAR T-PLT group, we argue that the enhanced infiltration of CAR T cells to tumor mass is
likely a result of the degradation of the basal membrane HSPGs by PLT introduced heparanases in CAR T-
PLT complexes. Our findings suggest that CAR T cell-PLT complexes can facilitate the enhanced
trafficking and infiltration of CAR-engineered T cells to the tumor site and improve the antitumor immune
response when potentially treating solid cancers with CAR T cells.
Although PMPs have similar functionalities to PLTs, CAR T-PMP complexes did not generate an
in vitro cell invasion or in vivo tumor growth control advantage in comparison to CAR T cells alone and
CAR T cell-PLT complexes. This finding suggests that PMPs were washed out by the wash step following
CAR T-PLT/PMP co-incubation, and that CAR T cell-PLT complexes are more resistant to mechanical
detachment. This could be attributed to the possible differential affinity of PLTs and PMPs to CAR T cells
in CAR T-PLT/PMP complexes. Although it is known that interaction between T cells and PLTs/PMPs are
integrin-mediated
217
, there is no literature on the affinity of PLTs and PMPs to T cells or other lymphocytes,
which requires further investigation. In addition, the precise mechanism of platelets’ effect on CAR T cell
homing in the form of CAR T-PLT complex remains elusive and would benefit from further research.
Moreover, PLT-lymphocyte interactions need to be further studied in cancer immunotherapy framework to
determine the impact of platelets on immune cells in the TME, including but not limited to T cells
233
.
It is important to note that how platelets influence the TME-resident cells (leukocytes, endothelial
cells, fibroblasts, pericytes) may be contextually dependent on the composition of the TME which varies
depending on organ/tissue type, and the anatomical location and degree of intratumor vascularity of tumors.

170
In the future, this system can be tested in various cancer models including metastatic cancers and the
complex outcomes of PLT-T cell interactions can be studied.
We also want to emphasize that the two cell types that would be used in this combination therapy
strategy can be obtained from the same cancer patient by a single apheresis. Isolation of the mononuclear
cell (MNC) layer can provide the circulating lymphocytes to be used in CAR T cell manufacturing, whereas
platelets can be isolated from the less dense plasma
506
.
Lastly, the anti-CD19 CAR has served as a proof-of-concept for utilizing PLTs to improve CAR T
cell accumulation in solid tumor. In the future, T cells in this CAR T cell-PLT platform can be engineered
to be redirected towards various TAAs for the targeting of various solid tumor malignancies. In addition,
to give this CAR T-PLT complex system a more specific migration ability into the tumor bed, CAR T cells
directed towards TAAs can be further modified to express chemokine-specific receptors depending on
different chemokine ligands in various tumors to support T cells’ efficient contact with tumor cells
510
. Given
the resurgence of research interest regarding the roles platelets play in the immune system and promising
experimental results in recent literature, we believe that our study brings a novel perspective on the use of
platelets for adoptive T cell therapy.
In Chapter 5, in the study named CAR-Engineered Natural Killer Cells as a Carrier of Drug-
Encapsulated Nanoparticles Targeting Adenosine Receptors in Solid Tumors, we combined
immunotherapy with nanomedicine to better treat solid tumors. We used target-specific CAR-engineered
NK cells as an active carrier of nanoparticles loaded with an immunomodulatory drug to deliver the drug
to solid tumor sites. We aimed to inhibit immunosuppressive effects of the tumor microenvironment and
maintain CAR.NK cell effector functions in solid tumors, with minimal toxicity. Also, NK-92 cells used in
this study provide a potential off-the-shelf allogeneic CAR-based therapy circumventing the high cost and
GvHD concerns that arise with autologous CAR T cell therapy.
In this study, we used NK-92 cells engineered with two different third-generation CAR molecules
targeting human mesothelin that were constructed in the study detailed in Chapter 3. These two different
types of CAR.NK cells, anti-mesothelin CARs without (ameso.CAR.NK) or with membrane bound IL-

171
15/IL-15Rα complex (ameso.mbIL15.CAR.NK), were successfully conjugated to cross-linked
multilamellar liposomal vesicles (cMLVs) containing an A2aR antagonist drug, SCH-58261. We
demonstrated that drug-loaded nanoparticle conjugation did not interfere with NK cell effector functions in
vitro. Specifically, membrane-bound IL-15/IL-15Rα complex expressed on ameso.mbIL15.CAR.NK cells
allowed these cells to expand in the absence of exogenous IL-2 cytokine, whereas non-transduced NK-92
cells and ameso.CAR-engineered NK cells did not increase their number. Our in vivo data also proved that
CAR.NK cells conjugated with SCH-58261-loaded nanoparticles slowed tumor growth more effectively
than any other treatment group. This finding provides evidence that CAR.NK cells mediated the delivery
of A2aR antagonist drug to the tumor mass, which helps combat the immunosuppressive TME. Both
ameso.CAR.NK and ameso.mbIL15.CAR.NK cells conjugated with drug-loaded liposomal nanoparticles
showed superior tumor growth control in vivo. However, the greatest tumor growth inhibition was achieved
by ameso.CAR.NK-cMLV(SCH) treatment according to the data collected towards the end of the study.
Importantly, mice treated with ameso.mbIL15.CAR.NK cells showed marked decline in health by the end
of the in vivo experiments. This toxicity can be attributed to adverse events caused by the expression of
membrane-bound IL-15/IL-15Rα complex or its mediated effects in mouse body, although the mechanism
of action requires further investigation. This finding also emphasizes the importance of further toxicity
analysis if the ameso.mbIL15.CAR.NK cells are to be used in clinic. On the other hand, as a response to
declined patient health or unwanted toxicity, CAR.NK therapy may be required being stopped. This would
be possible with a CAR design which has the ability to trigger the apoptosis of injected CAR.NK cells in
case of severe toxicity and therefore improve the safety in cancer treatment. A desirable alternative could
be using ameso.mbIL15.CAR.NK cells which also express an inducible suicide gene caspase 9 (iCAS9) as
a safety switch which responds to a chemical inducer of dimerization (CID) and trigger apoptosis in cells
expressing the iCAS9 transgene, as exemplified in the study presented in Chapter 3.
The small molecule immunomodulatory drug, SCH-58261 (SCH), is one of the most selective and
potent antagonists of A2aR. The clinical application of SCH has been hindered by the drug’s poor solubility
and suboptimal in vivo pharmacokinetic profile. Given the high amounts of SCH used in in vivo experiments

172
in the literature
338,361
, our system allows the use of a minimal amount of SCH to exert its A2aR blockage
effect while not introducing side effects. This indicates that encapsulation of the A2aR antagonist SCH-
58261 in cMLVs help drug exert its maximal effect at a lower dose, presumably by improving its solubility
and pharmacokinetics profiles by directly delivering the drug to its place of action in a protected manner.
Another major advantage of blocking the A2aR pathway with our CAR.NK-cMLV(SCH)
therapeutic system is the ability of the A2aR antagonist drug to exert its effect both on carrier CAR.NK
cells themselves, as well as on other endogenous NK and T cells in the tumor site, in addition to showing
its effect in the TME in a more comprehensive way. The effect of A2aR pathway inhibition on endogenous
NK and T cells in the tumor site is worth investigating in the future.
Conducting research to determine the degree of intratumoral drug accumulation to verify the
efficacy of active-drug delivery to tumor sites by this combinational therapy platform is also important.
Additionally, conducting a biodistribution study to investigate the degree of nanoparticle trafficking to the
tumor site would benefit from completion. The applicability of this novel platform can be expanded to
different clinical settings with further modifications and improvements. By making changes in different
components of this system, consisting of the cargo molecule, drug carrying nanoparticles and CAR-
engineered NK cells, various cancer types can be treated with various intervention strategies. This system
can also be used for the co-delivery of other immunomodulators and chemotherapeutic agents aiming dual
immune regulatory and tumor-killing effect in the tumor microenvironment. Furthermore, the multilamellar
liposomal vesicles (cMLVs) used in this study have the ability to efficiently load both hydrophilic and
hydrophobic drugs in the aqueous core of liposomal vesicles and in the lipid membranes, respectively.
It is clear that the NK-92 platform has been proven to be effective even in its unmodified,
monotherapy form. By applying various combination therapy approaches, NK cell therapies could be used
to further augment immune responses. Additional improvements could be achieved through genetic
modifications, i.e. CAR modification on NK-92 parental cells, or conjugation of NK cells with other agents
such as checkpoint inhibitors, immunomodulatory drugs and chemotherapy drugs. Given the universal
administration and off-the-shelf therapy advantage of NK-92 cells, this platform can also provide the

173
possibility for personalized treatment regimens that target specific patient needs with combinatorial therapy
approaches.
With the knowledge acquired from pre-clinical and clinical studies in which various methods have
been used to interfere with adenosine axis, including blockade of adenosine production and A2aR/A2bR
antagonism, we anticipate that more attention will be given to the development of new agents. The new
agents will target the immunosuppression mediated by adenosine. Additionally, improved treatment of
cancer patients can be achieved by reclaiming existing agents in combinatorial strategies with adenosine
receptor blockade.
An additional perspective to consider, beyond what is presented in the projects in this dissertation,
is the use of synthetic biology to include human-derived affinity proteins, suicide switches, membrane-
bound cytokines, immunostimulatory ligands, and immune checkpoint inhibitors, as well as the
combination with nanoparticle-formulated chemotherapeutics or immunomodulatory drugs, to further the
ultimate goal of durable remissions with minimal side effects from CAR-based cancer immunotherapy.
Other issues related to CAR T cell manufacturing are the barriers to making CAR T cell therapy a prevalent
cancer treatment modality. Specifically, the strict requirement of isolation and the use of autologous cells
for CAR T cell manufacturing, unsuccessful expansion of T cells in case of some patients, and the
technically complex and time intensive production of CAR T cells are among the main considerations for
the widespread implementation of CAR T cell therapy
63
.
For CAR-based therapies to be used as a front-line treatment, allogenic, off-the-shelf cells need to
take the place of autologous CAR T cells currently used in the clinic. Instead of autologous T cells, which
are used in most CAR studies, researchers have created allogenic CAR T cells in which the native TCR
gene is disrupted to mitigate concerns about graft-vs-host disease
550,551
and several clinical trials with donor
CAR T cells are ongoing
552
. NK cells are another option for standardized CAR therapy, as they have
reduced risk of cytokine release syndrome and graft-vs-host disease
553
. Different sources of NK cells, such
as NK cells derived from induced pluripotent stem cells, have proven to be suitable immune cells for CARs
in preclinical trials
554
. As another alternative source of NK cells, NK-92 cells have been discussed

174
extensively in the previous chapters. A homogenous, allogenic cellular product has the potential for more
precise modulation of biologic outputs than autologous cells, which vary widely in function from patient to
patient. NK-92 cell line based “off-the-shelf” CAR technology supports generating large numbers of
CAR.NK cells which are donor-independent and specific to the TAAs
101
. The NK-92 cell line has proven
effective in executing CAR-mediated antitumor functions in preclinical settings. Studies in xenograft mice
models with CAR-modified NK-92 have been shown to eliminate different malignancies, such as
lymphoma, breast cancer, neuroblastoma and glioblastoma, amongst others
120,122–124
. Although CAR-
expressing NK cells are being evaluated in early-stage clinical trials, CAR.NK-92 cells are strong
candidates for offering an alternative “off-the-shelf” cellular immunotherapy with increased specificity
towards TAAs in solid tumors, which would also potentially solve the expansion, cost, shipment and storage
issues currently associated with immunotherapy.
With the recent FDA approval of CAR T cell products, a major concern under discussion is the
affordability of this therapy. According to a recent study, the estimated total cost of care associated with
the administration of CAR T cell therapy ranges between $420,000 and $455,000
547
, with unclear guidelines
as to what insurance will cover. In contrast, the preparation and administration costs of NK-92 cells are
significantly lower than autologous or allogeneic NK cells, as well as CAR T cells. Treatments with
engineered NK-92 cells are believed to cost approximately $20,000, with the option of repeated infusions
101
.
The increasing number of FDA-approved commercial cancer therapies with CAR T cells initiated
a debate about the significant financial burden that these therapies impose on public and private payers.
The first two CAR T cell products produced, tisagenlecleucel (Kymriah, Novartis) and axicabtagene
ciloleucel (Yescarta, Kite), received domestic regulatory approval from the US regulatory authorities as
well as international approval in Canada, Australia, England and the European Union
555–560
.
The market price for a single infusion of CAR T cell therapies is $373,000 for axicabtagene
ciloleucel and $475,000 for tisagenlecleucel
561
. Currently, CAR T cell therapies are administered at a
limited number of cancer centers and are primarily delivered in an inpatient setting
547
. According to a recent
study, which took into account the incidence of key adverse events (i.e. cytokine release syndrome and

175
neurological toxicities), the estimated total cost of care associated with the administration of CAR T cell
therapy is calculated to be $454,611 (95% CI, $452 466-$458 267) in the academic hospital inpatient setting
and $421,624 (95% CI, $417 204-$422 325) in the nonacademic specialty oncology network setting
547
.
Given the high market price and therapy costs of CAR T cell therapies, serious concerns are raised over its
value and affordability. Therefore, it is necessary to discuss the key issues concerning the cost-effectiveness
of CAR T cell therapies and propose solutions that may increase the affordability of CAR T cell therapy
561
.
Many cancer survivors face serious financial burdens related to cancer treatment
562
. The financial
burden of oncologic treatments vary between different demographic groups and younger age is an
established risk factor for financial burden among cancer survivors
562
. Financial burdens also impact
treatment decisions which can negatively impact treatment outcomes
563,564
. The high cost of treatment can
cause changes in survivors’ financial circumstances through treatment-driven debt which result in a poorer
quality of life
565–567
. Patients experience several additional forms of financial hardship, including having
insurance companies refuse claims, denial of loans or insurance due to cancer history, and experiencing
significant indebtedness (i.e. facing large bills, insolvency, or bankruptcy)
561
. While the out-of-pocket
expenses for CAR T cell therapy have not yet been reported, they are predicted to be high due to high
insurance deductibles, potential co-pays for the therapy itself as well as infection prophylaxis, and the
necessity to transfer and lodge non-local patients due to the low number of administration centers
565
.
Previous studies indicate that manufacturing times and the cost of CAR T cell therapy, as well as
the financial burdens imposed by the price of CAR T cell therapy, could be decreased by in-house
manufacturing of CAR T cell products in academic centers using entirely enclosed good manufacturing
compliant (cGMP) platforms
568,569
. Research indicates that these cGMP platforms can reduce hospital costs
of therapy to US$20,000 without compromising patient outcomes
570
.
It should be noted that Novartis currently offers an outcomes-based reimbursement model in which
pediatric patients who do not achieve a complete response (CR) within a month are not required to pay for
tisagenlecleucel
571
. Although this type of supplier and payer-focused payment initiative may improve the
affordability of CAR T cell therapy, these initiatives should be structured in a way that a more diverse

176
demographic and a larger number of patients would benefit from them. Currently, Novartis’ reimbursement
scenario only applies to a small section of patients (i.e. it does not account for approximately 50% of patients
with B-ALL who relapse following CR)
556
.
The summary of problems related to the value and affordability of CAR T cell therapy in the US
and potential solutions offered by Fiorenza et al. are shown in Table 6.1
561
.
Table 6.1 Existing issue and future recommendations for the implementation of affordable CAR T cell therapy in
the US.
EXISTING ISSUES FUTURE RECOMMENDATIONS
Efficacy evaluations
• Single-arm, nonrandomized trials with short
follow-up
duration
• Randomized head-to-head trials of CAR T cells
versus modern therapies (e.g. inotuzumab) and
tracked over longer
durations
Value assessments
• Static QALY measurements
• Ambiguous US willingness-to-pay (WTP)
thresholds
• Cohort-level data averaged across trials
• Incomplete cost assessments focused on the
payer perspective
• Base case scenario not specified
• Noncontemporaneous comparators
• Reliance on ex ante estimates of survival
• Dynamic QALY measurements
• Develop explicit WTP thresholds
• Patient-level data
• Societal perspective cost assessments including
patient out-of-pocket costs and patient and
caregiver time-use opportunity costs
• Detailed definition of base case scenarios
• Recalculating assessments from upcoming
randomized trials with long-term follow-up
Insurance coverage considerations
• Vague coverage criteria • Definitive coverage criteria from all payers

Budget impact analyses
• Lack of published analyses • Peer-reviewed budget impact analyses
Pricing
• Nontransparent pricing
• Noncompetitive pricing
• High cost of manufacture
• Pricing transparencies along entire
pharmaceutical supply chain
• Legislated negotiation with suppliers
• In-house CAR T cell manufacturing (with
compulsory licensing)
Payment
• Varied payment mechanisms and programs that
do not
address high cost and potentially curative novel
technologies that can impact patients over long-
term
horizons in which the change health plans
• Increased payment supplements for new
technologies
• Bundled payments for care improvement
• Outcomes based reimbursement
• Cost recuperation for insurers
Clinical and scientific advances
• Relapse in a large proportion of patients
• Considerable toxicity profiles
• Appropriate patient selection
• Lower toxicity therapies
Table adapted from Fiorenza, S., Ritchie, D.S., Ramsey, S.D. et al. Value and affordability of CAR T-cell therapy in the
United States. Bone Marrow Transplant 55, 1706–1715 (2020). https://doi.org/10.1038/s41409-020-0956-8

177
Due to the presence of strong research and high number of clinical trials in the US, as well as a
commercial base for CAR T cell therapies, North America has dominated the T cell therapy market. China
has also emerged as a potential market for CAR T therapies due to the high number of CAR T cell clinical
trials and government investment in these therapies. According to a February 2021 report on T cell therapy
market analysis, the global T cell therapy market size was valued at USD 4.6 billion in 2020 and is expected
to expand at a compound annual growth rate (CAGR) of 20.2% from 2021 to 2028
572
.
As of May 2021, according to the data in clinicaltrials.gov, the number of clinical trials with CAR-
engineered T cells worldwide is 1373 and with CAR-engineered NK cells is 34, as China and the United
States being the most active countries. For instance, 38% of the clinical trials for CAR T cells is being
conducted in the USA and 37% is being conducted in China. Similarly, 62% of the clinical trials for CAR
NK cells is being conducted in China, following by the USA with 24%. Close to 10,000 patients worldwide
have received CAR T cell therapy in a clinical trial. However, only a very small number of the patients
have been from low- and middle-income countries. Currently, South America, Africa, and India have not
registered any CAR T cell therapy trials
573
. Turkey, has only one registered study in clinicaltrials.gov, and
it is in recruiting status.
With its high cost of hundreds of thousands of USD, CAR T cell therapy cannot be considered as
an essential treatment for emerging economies. In low-income countries, CAR T cell therapy would more
likely be left to the control of privately funded institutions which would further hamper the accessibility of
this cancer therapy by less well-off patients
573
. One study suggests the employment of existing compulsory
licensing frameworks (to avoid patent infringements) as a potential way to increase the affordability of
these treatments in low- and middle-income countries
574
.
It would make a meaningful difference if under-developed and developing countries invest in the
training of highly specialized oncologists and clinical care specialists for the implementation of this therapy
in hospitals. Also, supporting the training of a workforce of highly-skilled engineers and scientists would
be a visionary move for the possible future low-cost and in-house manufacturing of CAR T cell products
in academic centers/university hospitals using cGMP platforms.

178
It is important to recognize that the affordability of CAR T cell therapies cannot be increased solely
by scientific and engineering approaches. Efforts are being made to lower the manufacturing costs of CAR
T cell products and to minimize the CAR T cell therapy related toxicities. Both approaches would decrease
the cost of intensive care unit admissions and additional therapies for the treatment of adverse effects of
CAR T cell infusions, and are necessary steps but are insufficient in and of themselves for the enhanced
affordability of these therapies. Increasing the affordability of these therapies can only be possible by
combining scientific improvements with intentional governmental approaches. Although there are efforts
to lower drug costs by making changes in federal legislations in the US, the implementation of the
legislations may be limited due to pharmaceutical company lobbying
575
, especially given the market value
of T cell therapy market
572
.
Efforts must therefore be expanded by public and private entities to improve the affordability of
CAR T cell therapies. In the US and also in low- and middle-income countries, CAR T cell therapies must
be made available and affordable for the treatment of cancer patients who are eligible to receive these
treatments.

179
References
1. DeVita, V. T. & Chu, E. A history of cancer chemotherapy. Cancer Res. 68, 8643–8653 (2008).

2. Sudhakar, A. History of Cancer, Ancient and Modern Treatment Methods. J Cancer Sci Ther 1, 1–
4 (2009).

3. Seidel, J. A., Otsuka, A. & Kabashima, K. Anti-PD-1 and anti-CTLA-4 therapies in cancer:
Mechanisms of action, efficacy, and limitations. Front. Oncol. 8, 1–14 (2018).

4. Riley, R. S., June, C. H., Langer, R. & Mitchell, M. J. Delivery technologies for cancer
immunotherapy. Nat. Rev. Drug Discov. 18, 175–196 (2019).

5. Pardoll, D. M. The blockade of immune checkpoints in cancer immunotherapy. Nat. Rev. Cancer
12, 252–264 (2012).

6. Ellis, P. M., Vella, E. T. & Ung, Y. C. Immune Checkpoint Inhibitors for Patients With Advanced
Non–Small-Cell Lung Cancer: A Systematic Review. Clin. Lung Cancer 18, 444–459 (2017).

7. Müller, L., Aigner, P. & Stoiber, D. Type I interferons and natural killer cell regulation in cancer.
Front. Immunol. 8, 1–11 (2017).

8. Cox, M. A., Harrington, L. E. & Zajac, A. J. Cytokines and the Inception of CD8 T Cell
Responses The Pluripotency of Naïve CD8 T cells. Trends Immunol. 32, 180–186 (2011).

9. Hasan, A. N. et al. Soluble and membrane-bound interleukin (IL)-15 Rα/IL-15 complexes mediate
proliferation of high-avidity central memory CD8+ T cells for adoptive immunotherapy of cancer
and infections. Clin. Exp. Immunol. 186, 249–265 (2016).

10. Chapuis, A. G. et al. Combined IL-21-primed polyclonal CTL plus CTLA4 blockade controls
refractory metastatic melanoma in a patient. J. Exp. Med. 213, 1133–1139 (2016).

11. Sun, T. et al. Inhibition of Tumor Angiogenesis by Interferon-γ by Suppression of Tumor -
Associated Macrophage Differentiation. Oncol Res. 21, 227–35 (2014).

12. Peggs, K. S., Quezada, S. A. & Allison, J. P. Cancer immunotherapy: Co-stimulatory agonists and
co-inhibitory antagonists. Clin. Exp. Immunol. 157, 9–19 (2009).

13. Chester, C., Sanmamed, M. F., Wang, J. & Melero, I. Immunotherapy targeting 4-1BB:
Mechanistic rationale, clinical results, and future strategies. Blood 131, 49–57 (2018).

14. Guo, C. et al. Therapeutic cancer vaccines. Past, present, and future. Advances in Cancer
Research 119, (2013).

15. Garg, A. D., Coulie, P. G., Van den Eynde, B. J. & Agostinis, P. Integrating Next-Generation
Dendritic Cell Vaccines into the Current Cancer Immunotherapy Landscape. Trends Immunol. 38,
577–593 (2017).

180
16. Pardi, N., Hogan, M. J., Porter, F. W. & Weissman, D. mRNA vaccines-a new era in vaccinology.
Nat. Rev. Drug Discov. 17, 261–279 (2018).

17. Li, L., Goedegebuure, S. P. & Gillanders, W. E. Preclinical and clinical development of
neoantigen vaccines. Ann. Oncol. 28, xii11–xii17 (2017).

18. Lee, S. & Margolin, K. Cytokines in cancer immunotherapy. Cancers (Basel). 3, 3856–3893
(2011).

19. Milling, L., Zhang, Y. & Irvine, D. J. Delivering safer immunotherapies for cancer. Adv. Drug
Deliv. Rev. 114, 79–101 (2017).

20. June, C. H., Warshauer, J. T. & Bluestone, J. A. Is autoimmunity the Achilles’ heel of cancer
immunotherapy? Nat. Med. 23, 540–547 (2017).

21. Menon, S., Shin, S. & Dy, G. Advances in cancer immunotherapy in solid tumors. Cancers
(Basel). 8, 1–21 (2016).

22. Maleki Vareki, S., Garrigós, C. & Duran, I. Biomarkers of response to PD-1/PD-L1 inhibition.
Crit. Rev. Oncol. Hematol. 116, 116–124 (2017).

23. Lim, W. A. & June, C. H. The Principles of Engineering Immune Cells to Treat Cancer. Cell 168,
724–740 (2017).

24. Dudley, M. E. et al. Adoptive cell therapy for patients with metastatic melanoma: Evaluation of
intensive myeloablative chemoradiation preparative regimens. J. Clin. Oncol. 26, 5233–5239
(2008).

25. Rapoport, A. P. et al. Restoration of immunity in lymphopenic individuals with cancer by
vaccination and adoptive T-cell transfer. Nat. Med. 11, 1230–1237 (2005).

26. Feldman, S. A., Assadipour, Y., Kriley, I., Goff, S. L. & Rosenberg, S. A. Adoptive cell therapy—
tumor-infiltrating lymphocytes, T-cell receptors, and chimeric antigen receptors. Semin Oncol 42,
626–639 (2015).

27. Kalos, M. & June, C. H. Adoptive T Cell Transfer for Cancer Immunotherapy in the Era of
Synthetic Biology. Immunity 39, 49–60 (2013).

28. Restifo, N. P., Dudley, M. E. & Rosenberg, S. A. Adoptive immunotherapy for cancer: Harnessing
the T cell response. Nat. Rev. Immunol. 12, 269–281 (2012).

29. June, C. H., Riddell, S. R. & Schumacher, T. N. Adoptive cellular therapy: A race to the finish
line. Sci. Transl. Med. 7, 1–9 (2015).

30. Morrison, C. CAR-T field booms as next-generation platforms attract big players. Nat. Biotechnol.
33, 571–572 (2015).

31. Zhang, E. & Xu, H. A new insight in chimeric antigen receptor-engineered T cells for cancer
immunotherapy. J. Hematol. Oncol. 10, 1–11 (2017).

181
32. Gross, G., Waks, T. & Eshhar, Z. Expression of immunoglobulin-T-cell receptor chimeric
molecules as functional receptors with antibody-type specificity. Proc. Natl. Acad. Sci. U. S. A. 86,
10024–10028 (1989).

33. Curran, K. J., Pegram, H. J. & Brentjens, R. J. Chimeric antigen receptors for T cell
immunotherapy: Current understanding and future directions. J. Gene Med. 14, 405–415 (2012).

34. Marofi, F. et al. CAR T cells in solid tumors: challenges and opportunities. Stem Cell Res. Ther.
12, 1–16 (2021).

35. Hartmann, J., Schüßler‐Lenz, M., Bondanza, A. & Buchholz, C. J. Clinical development of CAR T
cells—challenges and opportunities in translating innovative treatment concepts. EMBO Mol.
Med. 9, 1183–1197 (2017).

36. Zhylko, A., Winiarska, M. & Graczyk-Jarzynka, A. The Great War of Today: Modifications of
CAR-T Cells to Effectively Combat Malignancies. Cancers (Basel). 12, 1–29 (2020).

37. You, F. et al. A novel CD7 chimeric antigen receptor-modified NK-92MI cell line targeting T-cell
acute lymphoblastic leukemia. Am. J. Cancer Res. 9, 64–78 (2019).

38. Duan, H., Huang, H. & Jing, G. An antibody fab fragment-based chimeric antigen receptor could
efficiently eliminate human thyroid cancer cells. J. Cancer 10, 1890–1895 (2019).

39. Balakrishnan, A. et al. Multispecific targeting with synthetic ankyrin repeat motif chimeric antigen
receptors. Clin. Cancer Res. 25, 7506–7516 (2019).

40. Long, E. O., Sik Kim, H., Liu, D., Peterson, M. E. & Rajagopalan, S. Controlling Natural Killer
Cell Responses: Integration of Signals for Activation and Inhibition. Annu. Rev. Immunol. 31,
227–258 (2013).

41. Marcus, A. & Eshhar, Z. Allogeneic chimeric antigen receptor-modified cells for adoptive cell
therapy of cancer. Expert Opin. Biol. Ther. 14, 947–954 (2014).

42. Thistlethwaite, F. C. et al. The clinical efficacy of first-generation carcinoembryonic antigen
(CEACAM5)-specific CAR T cells is limited by poor persistence and transient pre-conditioning-
dependent respiratory toxicity. Cancer Immunol. Immunother. 66, 1425–1436 (2017).

43. Savoldo, B. et al. CD28 costimulation improves expansion and persistence of chimeric antigen
receptor– modified T cells in lymphoma patients. J Clin Invest . 121, 1–5 (2011).

44. Zhong, X. S., Matsushita, M., Plotkin, J., Riviere, I. & Sadelain, M. Chimeric antigen receptors
combining 4-1BB and CD28 signaling domains augment PI 3 kinase/AKT/Bcl-X L activation and
CD8 T cell-mediated tumor eradication. Mol. Ther. 18, 413–420 (2010).

45. Benmebarek, M. R. et al. Killing mechanisms of chimeric antigen receptor (CAR) T cells. Int. J.
Mol. Sci. 20, (2019).

46. Zhao, L. & Cao, Y. J. Engineered T Cell Therapy for Cancer in the Clinic. Front. Immunol. 10,
(2019).

182
47. Davila, M. L. et al. Efficacy and toxicity management of 19-28z CAR T cell therapy in B cell
acute lymphoblastic leukemia. Sci. Transl. Med. 6, (2014).

48. U.S. Food and Drug Administration. FDA approval brings first gene therapy to the United States.
(2017). Available at: https://www.fda.gov/news-events/press-announcements/fda-approval-brings-
first-gene-therapy-united-states. (Accessed: 29th May 2021)

49. Papadouli, I. et al. EMA Review of Axicabtagene Ciloleucel (Yescarta) for the Treatment of
Diffuse Large B‐Cell Lymphoma. Oncologist 25, 894–902 (2020).

50. Administration, U. S. F. and D. TECARTUS (brexucabtagene autoleucel). (2021). Available at:
https://www.fda.gov/vaccines-blood-biologics/cellular-gene-therapy-products/tecartus-
brexucabtagene-autoleucel. (Accessed: 29th May 2021)

51. Wang, M. et al. KTE-X19 CAR T-Cell Therapy in Relapsed or Refractory Mantle-Cell
Lymphoma. N. Engl. J. Med. 382, 1331–1342 (2020).

52. U.S. Food and Drug Administration. FDA Approves New Treatment For Adults With Relapsed Or
Refractory Large-B-Cell Lymphoma. (2021). Available at: https://www.fda.gov/news-
events/press-announcements/fda-approves-new-treatment-adults-relapsed-or-refractory-large-b-
cell-lymphoma. (Accessed: 29th May 2021)

53. Mullard, A. FDA approves fourth CAR-T cell therapy. Nat. Rev. Drug Discov. 20, 166 (2021).

54. U.S. Food and Drug Administration. ABECMA (idecabtagene vicleucel). (2021). Available at:
https://www.fda.gov/vaccines-blood-biologics/abecma-idecabtagene-vicleucel. (Accessed: 29th
May 2021)

55. Shah, N. N. & Fry, T. J. Mechanisms of resistance to CAR T cell therapy. Nat. Rev. Clin. Oncol.
16, 372–385 (2019).

56. Stroncek, D. F. et al. Elutriated lymphocytes for manufacturing chimeric antigen receptor T cells.
J. Transl. Med. 15, 4–11 (2017).

57. Ceppi, F. et al. Lymphocyte apheresis for chimeric antigen receptor T-cell manufacturing in
children and young adults with leukemia and neuroblastoma. Transfusion 58, 1414–1420 (2018).

58. Feucht, J. et al. Calibration of CAR activation potential directs alternative T cell fates and
therapeutic potency. Nat. Med. 25, 82–88 (2019).

59. Mueller, K. T. et al. Cellular kinetics of CTL019 in relapsed/refractory B-cell acute lymphoblastic
leukemia and chronic lymphocytic leukemia. Blood 130, 2317–2325 (2017).

60. Lee, D. W. et al. T cells expressing CD19 chimeric antigen receptors for acute lymphoblastic
leukaemia in children and young adults: A phase 1 dose-escalation trial. Lancet 385, 517–528
(2015).

61. Porter, D. L. et al. Chimeric antigen receptor T cells persist and induce sustained remissions in
relapsed refractory chronic lymphocytic leukemia. Sci. Transl. Med. 7, 1–13 (2015).

183
62. Brentjens, R. J. et al. Safety and persistence of adoptively transferred autologous CD19-targeted T
cells in patients with relapsed or chemotherapy refractory B-cell leukemias. Blood 118, 4817–
4828 (2011).

63. Levine, B. L., Miskin, J., Wonnacott, K. & Keir, C. Global Manufacturing of CAR T Cell
Therapy. Mol. Ther. - Methods Clin. Dev. 4, 92–101 (2017).

64. Jacoby, E. et al. CD19 CAR immune pressure induces B-precursor acute lymphoblastic leukaemia
lineage switch exposing inherent leukaemic plasticity. Nat. Commun. 7, (2016).

65. Gardner, R. et al. Acquisition of a CD19-negative myeloid phenotype allows immune escape of
MLL-rearranged B-ALL from CD19 CAR-T-cell therapy. Blood 127, 2406–2410 (2016).

66. Michael, M. S. & Snook, A. Challenges to Chimeric Antigen Receptor (CAR)-T Cell Therapy for
Cancer. Discov Med 18, 265–271 (2014).

67. Maude, S. L. et al. Chimeric Antigen Receptor T Cells for Sustained Remissions in Leukemia. N.
Engl. J. Med. 371, 1507–1517 (2014).

68. Lowe, K. L. et al. Fludarabine and neurotoxicity in engineered T-cell therapy. Gene Ther. 25,
176–191 (2018).

69. Ingegnere, T. et al. Human CAR NK cells: A new non-viral method allowing high efficient
transfection and strong tumor cell killing. Front. Immunol. 10, 1–10 (2019).

70. Daher, M. & Rezvani, K. Next generation natural killer cells for cancer immunotherapy: the
promise of genetic engineering. Curr. Opin. Immunol. 51, 146–153 (2018).

71. Tang, X. et al. Erratum: First-in-man clinical trial of CAR NK-92 cells: safety test of CD33-CAR
NK-92 cells in patients with relapsed and refractory acute myeloid leukemia. Am. J. Cancer Res.
8, 1899 (2018).

72. Moretta, L. et al. Human NK cells: From surface receptors to clinical applications. Immunol. Lett.
178, 15–19 (2016).

73. Miller, J. S. et al. Successful adoptive transfer and in vivo expansion of human haploidentical NK
cells in patients with cancer. Blood 105, 3051–3057 (2005).

74. Imai, K., Matsuyama, S., Miyake, S., Suga, K. & Nakachi, K. Natural cytotoxic activity of
peripheral-blood lymphocytes and cancer incidence: An 11-year follow-up study of a general
population. Lancet 356, 1795–1799 (2000).

75. Vivier, E. et al. Innate or adaptive immunity? The example of natural killer cells. Science (80-. ).
331, 44–49 (2011).

76. Screpanti, V., Wallin, R. P. A., Ljunggren, H.-G. & Grandien, A. A Central Role for Death
Receptor-Mediated Apoptosis in the Rejection of Tumors by NK Cells. J. Immunol. 167, 2068–
2073 (2001).

77. Smyth, M. J. et al. Activation of NK cell cytotoxicity. Mol Immunol 42, 501–10 (2005).

184
78. Lanier, L. L. Up on the tightrope: natural killer cell activation and inhibition. Nat Immunol 9, 495–
502 (2008).

79. Becknell, B. & Caligiuri, M. A. Natural killer cells in innate immunity and cancer. J Immunother
31, 685–92 (2008).

80. Gross, E., Sunwoo, J. B. & Bui, J. D. Cancer Immunosurveillance and Immunoediting by Natural
Killer Cells. Cancer J. 19, 483–489 (2013).

81. Farag, S. S. & Caligiuri, M. A. Human natural killer cell development and biology. Blood Rev. 20,
123–137 (2006).

82. Lanier, L. L. NK Cell Recognition. Annu. Rev. Immunol. 23, 225–274 (2005).

83. Moretta, A. et al. Activating Receptors and Coreceptors Involved in Human Natural Killer Cell-
mediated Cytolysis. 19, 197–223 (2001).

84. Cooper, M. A., Fehniger, T. A. & Caligiuri, M. A. The biology of human natural killer-cell
subsets. Trends Immunol. 22, 633–640 (2001).

85. Watzl, C. & Long, E. O. Signal Transduction During Activation and Inhibition of Natural Killer
Cells. Curr Protoc Immunol. Chapter 11, (2010).

86. Raulet, D. H. & Vance, R. E. Self-tolerance of natural killer cells. Nat. Rev. Immunol. 6, 520–531
(2006).

87. Fauriat, C. Regulation of human NK-cell cytokine and chemokine production by target cell
recognition. Blood 115, 2167–2177 (2016).

88. Oberschmidt, O., Kloess, S. & Koehl, U. Redirected Primary Human Chimeric Antigen Receptor
natural Killer Cells As an “Off-the-Shelf immunotherapy” for improvement in Cancer Treatment.
Front. Immunol. 8, 654 (2017).

89. Shah, N. N. et al. Acute GVHD in patients receiving IL-15/4-1BBL activated NK cells following
T-cell-depleted stem cell transplantation. Blood 125, 784–792 (2015).

90. Becker, P. S. A. et al. Selection and expansion of natural killer cells for NK cell-based
immunotherapy. Cancer Immunol. Immunother. 65, 477–484 (2016).

91. Chen, I. M. et al. A new human natural killer leukemia cell line, IMC-1. A complex chromosomal
rearrangement defined by spectral karyotyping: Functional and cytogenetic characterization. Leuk.
Res. 28, 275–284 (2004).

92. Cheng, M. et al. Establishment, characterization, and successful adaptive therapy against human
tumors of NKG cell, a new human NK cell line. Cell Transplant. 20, 1731–1746 (2011).

93. JH, G., G, M. & HG, K. Characterization of a human cell line (NK-92) with phenotypical and
functional characteristics of activated natural killer cells. Leukemia. 8, 652–8 (1994).

185
94. Nagata, H. et al. Characterization of novel natural killer (NK)–cell and γδ T-cell lines established
from primary lesions of nasal T/NK-cell lymphomas associated with the Epstein-Barr virus. Blood
97, 708–713 (2001).

95. Robertson, M. et al. Characterization of a cell line, NKL, derived from an aggressive human
natural killer cell leukemia. Exp Hematol. 24, 406–15 (1996).

96. Tsuchiyama, J. et al. Characterization of a novel human natural killer-cell line (NK-YS)
established from natural killer cell lymphoma/leukemia associated with Epstein-Barr virus
infection. Blood 92, 1374–83 (1998).

97. Yagita, M. et al. A novel natural killer cell line (KHYG-1) from a patient with aggresive natural
killer cell leukemia carrying a p53 point mutation. Leukemia 14, 922–930 (2000).

98. Arai, S. et al. Infusion of the allogeneic cell line NK-92 in patients with advanced renal cell cancer
or melanoma: a phase I trial. Cytotherapy 10, 625–32 (2008).

99. Tonn, T. et al. Treatment of patients with advanced cancer with the natural killer cell line NK-92.
Cytotherapy 15, 1563–1570 (2013).

100. Klingemann, H., Wong, E. & Maki, G. A cytotoxic NK-cell line (NK-92) for ex vivo purging of
leukemia from blood. Biol Blood Marrow Transpl. 2, 68–75 (1996).

101. Klingemann, H., Boissel, L. & Toneguzzo, F. Natural killer cells for immunotherapy - Advantages
of the NK-92 cell line over blood NK cells. Front. Immunol. 7, 1–7 (2016).

102. Bishara, A. et al. The beneficial role of inhibitory KIR genes of HLA class I NK epitopes in
haploidentically mismatched stem cell allografts may be masked by residual donor-alloreactive T
cells causing GVHD. Tissue Antigens 63, 204–211 (2004).

103. Vivier, E., Tomasello, E., Baratin, M., Walzer, T. & Ugolini, S. Functions of natural killer cells.
Nat. Immunol. 9, 503–510 (2008).

104. Suerth, J. D. et al. Efficient generation of gene-modified human natural killer cells via
alpharetroviral vectors. J. Mol. Med. 94, 83–93 (2016).

105. Lang, S., Vujanovic, N. L., Wollenberg, B. & Whiteside, T. L. Absence of B7.1-CD28/CTLA-4-
mediated co- stimulation in human NK cells. Eur J Immunol 28, 780–786 (1998).

106. Kowolik, C. M. et al. CD28 costimulation provided through a CD19-specific chimeric antigen
receptor enhances in vivo persistence and antitumor efficacy of adoptively transferred T cells.
Cancer Res. 66, 10995–11004 (2006).

107. Kwon, B. S. & Weissman, S. M. cDNA sequences of two inducible T-cell genes. Proc. Natl.
Acad. Sci. U. S. A. 86, 1963–1967 (1989).

108. Habib-Agahi, M., Phan, T. T. & Searle, P. F. Co-stimulation with 4-1BB ligand allows extended
T-cell proliferation, synergizes with CD80/CD86 and can reactivate anergic T cells. Int. Immunol.
19, 1383–1394 (2007).

186
109. Navabi, S. sadat, Doroudchi, M., Tashnizi, A. H. & Habibagahi, M. Natural Killer Cell Functional
Activity After 4-1BB Costimulation. Inflammation 38, 1181–1190 (2015).

110. Fujisaki, H. et al. Expansion of highly cytotoxic human natural killer cells for cancer cell therapy.
Cancer Res. 69, 4010–4017 (2009).

111. Zhang, H. et al. Activating signals dominate inhibitory signals in CD137L/IL-15 activated natural
killer cells. J. Immunother. 34, 187–195 (2011).

112. Klingemann, H. Are natural killer cells superior CAR drivers? Oncoimmunology 3, (2014).

113. Liu, E. et al. Use of CAR-Transduced Natural Killer Cells in CD19-Positive Lymphoid Tumors.
N. Engl. J. Med. 382, 545–553 (2020).

114. Ruggeri, L. et al. Effectiveness of Donor Natural Killer Cell Alloreactivity in Mismatched
Hematopoietic Transplants. Science (80-. ). 295, 2097–2100 (2002).

115. Laskowski, T. & Rezvani, K. Adoptive cell therapy: Living drugs against cancer. J. Exp. Med.
217, 1–4 (2020).

116. Huenecke, S. et al. IL-2-driven regulation of NK cell receptors with regard to the distribution of
CD16+ and CD16-subpopulations and in Vivo influence after haploidentical NK cell infusion. J.
Immunother. 33, 200–210 (2010).

117. Yan, Y. et al. Antileukemia activity of a natural killer cell line against human leukemias. Clin.
Cancer Res. 4, 2859–2868 (1998).

118. Swift, B. E. et al. Natural killer cell lines preferentially kill clonogenic multiple myeloma cells and
decrease myeloma engraftment in a bioluminescent xenograft mouse model. Haematologica 97,
1020–1028 (2012).

119. Tam, Y. K., Miyagawa, B., Ho, V. C. & Klingemann, H.-G. Immunotherapy of malignant
melanoma in a SCID mouse model using the highly cytotoxic natural killer cell line NK-92. J
Hematother. 8, 281–90 (1999).

120. Boissel, L. et al. Retargeting NK-92 cells by means of CD19- and CD20-specific chimeric antigen
receptors compares favorably with antibody-dependent cellular cytotoxicity. Oncoimmunology 2,
e26527 (2013).

121. Sahm, C., Schönfeld, K. & Wels, W. S. Expression of IL-15 in NK cells results in rapid
enrichment and selective cytotoxicity of gene-modiWed eVectors that carry a tumor-speciWc
antigen receptor. Cancer Immunol. Immunother. 61, 1451–1461 (2012).

122. Schönfeld, K. et al. Selective inhibition of tumor growth by clonal NK cells expressing an
ErbB2/HER2-specific chimeric antigen receptor. Mol. Ther. 23, 330–338 (2015).

123. Seidel, D. et al. Disialoganglioside-specific human natural killer cells are effective against drug-
resistant neuroblastoma. Cancer Immunol. Immunother. 64, 621–634 (2015).

124. Han, J. et al. CAR-engineered NK cells targeting wild-type EGFR and EGFRvIII enhance killing
of glioblastoma and patient-derived glioblastoma stem cells. Sci Rep 5, 11483 (2015).

187
125. Müller, T. et al. Expression of a CD20-specific chimeric antigen receptor enhances cytotoxic
activity of NK cells and overcomes NK-resistance of lymphoma and leukemia cells. Cancer
Immunol Immunother (2008) 57, 411–423 (2008).

126. Boissel, L. et al. Comparison of mRNA and lentiviral based transfection of natural killer cells with
chimeric antigen receptors recognizing lymphoid antigens. Leuk. Lymphoma 53, 958–965 (2012).

127. Uherek, C. et al. Retargeting of natural killer-cell cytolytic activity to ErbB2-expressing cancer
cells results in efficient and selective tumor cell destruction. Blood 100, 1265–1273 (2002).

128. Liu, H. et al. Specific growth inhibition of ErbB2-expressing human breast cancer cells by
genetically modified NK-92 cells. Oncol. Rep. 33, 95–102 (2015).

129. Esser, R. et al. NK cells engineered to express a GD2-specific antigen receptor display built-in
ADCC-like activity against tumour cells of neuroectodermal origin. J. Cell. Mol. Med. 16, 569–
581 (2012).

130. Jiang, H. et al. Transfection of chimeric anti-CD138 gene enhances natural killer cell activation
and killing of multiple myeloma cells. Mol. Oncol. 8, 297–310 (2014).

131. Tassev, D. V., Cheng, M. & Cheung, N. K. Retargeting NK92 cells using an HLA-A2-restricted,
EBNA3C-specific chimeric antigen receptor. Cancer Gene Ther. 19, 84–100 (2012).

132. Kailayangiri, S. et al. Targeting ewing sarcoma with activated and GD2-specific chimeric antigen
receptor-engineered human NK cells induces upregulation of immune-inhibitory HLA-G.
Oncoimmunology 6, e1250050 (2017).

133. Milone, M. C. et al. Chimeric receptors containing CD137 signal transduction domains mediate
enhanced survival of T cells and increased antileukemic efficacy in vivo. Mol. Ther. 17, 1453–
1464 (2009).

134. Zhang, G. et al. Retargeting NK-92 for anti-melanoma activity by a TCR-like single-domain
antibody. Immunol. Cell Biol. 91, 615–624 (2013).

135. Klingemann, H. Challenges of cancer therapy with natural killer cells. Cytotherapy 17, 245–249
(2015).

136. O’Rourke, D. M. et al. A single dose of peripherally infused EGFRvIII-directed CAR T cells
mediates antigen loss and induces adaptive resistance in patients with recurrent glioblastoma. Sci
Transl Med. 9, (2017).

137. Newick, K., O’Brien, S., Moon, E. & Albelda, S. M. CAR T Cell Therapy for Solid Tumors. Annu
Rev Med 68, 139–152 (2017).

138. Yu, S. et al. Chimeric antigen receptor T cells: a novel therapy for solid tumors. J. Hematol.
Oncol. 10, 1–13 (2017).

139. Zhang, E., Gu, J. & Xu, H. Prospects for chimeric antigen receptor-modified T cell therapy for
solid tumors. Mol. Cancer 17, 1–12 (2018).

188
140. Khan, J. F., Khan, A. S. & Brentjens, R. J. Application of CAR T cells for the treatment of solid
tumors. Prog. Mol. Biol. Transl. Sci. 164, 293–327 (2019).

141. Bagley, S. J. & O’Rourke, D. M. Clinical investigation of CAR T cells for solid tumors: Lessons
learned and future directions. Pharmacol. Ther. 205, (2020).

142. Porter, D. L., Levine, B. L., Kalos, M., Bagg, A. & June, C. H. Chimeric Antigen Receptor–
Modified T Cells in Chronic Lymphoid Leukemia. N. Engl. J. Med. 365, 725–733 (2011).

143. Jamal-Hanjani, M., Quezada, S. A., Larkin, J. & Swanton, C. Translational implications of tumor
heterogeneity. Clin. Cancer Res. 21, 1258–1266 (2015).

144. Martinez, M. & Moon, E. K. CAR T cells for solid tumors: New strategies for finding, infiltrating,
and surviving in the tumor microenvironment. Front. Immunol. 10, 1–21 (2019).

145. Hanahan, D. & Coussens, L. M. Accessories to the Crime: Functions of Cells Recruited to the
Tumor Microenvironment. Cancer Cell 21, 309–322 (2012).

146. Adachi, K. et al. IL-7 and CCL19 expression in CAR-T cells improves immune cell infiltration
and CAR-T cell survival in the tumor. Nat. Biotechnol. 36, 346–351 (2018).

147. Moon, E. K. et al. Expression of a functional CCR2 receptor enhances tumor localization and
tumor eradication by retargeted human T cells expressing a mesothelin-specific chimeric antibody
receptor. Clin. Cancer Res. 17, 4719–4730 (2011).

148. Wang, L. C. S. et al. Targeting fibroblast activation protein in tumor stroma with chimeric antigen
receptor T cells can inhibit tumor growth and augment host immunity without severe toxicity.
Cancer Immunol. Res. 2, 154–166 (2014).

149. Schuberth, P. C. et al. Treatment of malignant pleural mesothelioma by fibroblast activation
protein-specific re-directed T cells. J. Transl. Med. 11, 1–11 (2013).

150. Priceman, S. J. et al. Regional delivery of chimeric antigen receptor-engineered T cells effectively
targets HER2+ breast cancer metastasis to the brain. Clin. Cancer Res. 24, 95–105 (2018).

151. Theocharis, A. D., Skandalis, S. S., Gialeli, C. & Karamanos, N. K. Extracellular matrix structure.
Adv. Drug Deliv. Rev. 97, 4–27 (2016).

152. Bernfield, M. et al. Functions of Cell Surface Heparan Sulfate Proteoglycans. Annu. Rev.
Biochem. 68, 729–777 (1999).

153. Digre, A. et al. Overexpression of heparanase enhances T lymphocyte activities and intensifies the
inflammatory response in a model of murine rheumatoid arthritis. Sci. Rep. 7, 1–12 (2017).

154. Renner, K. et al. Metabolic hallmarks of tumor and immune cells in the tumor microenvironment.
Front. Immunol. 8, 1–11 (2017).

155. Oliver, A. J. et al. Tissue-dependent tumor microenvironments and their impact on
immunotherapy responses. Front. Immunol. 9, 1–8 (2018).

189
156. Cairns, R. A., Harris, I. S. & Mak, T. W. Regulation of cancer cell metabolism. Nat. Rev. Cancer
11, 85–95 (2011).

157. Balkwill, F. R., Capasso, M. & Hagemann, T. The tumor microenvironment at a glance. J. Cell
Sci. 125, 5591–5596 (2012).

158. Hanahan, D. & Weinberg, R. A. Hallmarks of cancer: The next generation. Cell 144, 646–674
(2011).

159. Ackerman, D. & Simon, M. C. Hypoxia, lipids, and cancer: surviving the harsh tumor
microenvironment. Trends Cell Biol 24, 472–478 (2014).

160. Ganapathy, V., Thangaraju, M. & Prasad, P. D. Nutrient transporters in cancer: Relevance to
Warburg hypothesis and beyond. Pharmacol. Ther. 121, 29–40 (2009).

161. Ligtenberg, M. A. et al. Coexpressed Catalase Protects Chimeric Antigen Receptor–Redirected T
Cells as well as Bystander Cells from Oxidative Stress–Induced Loss of Antitumor Activity. J.
Immunol. 196, 759–766 (2016).

162. Leone, R. D., Lo, Y. C. & Powell, J. D. A2aR antagonists: Next generation checkpoint blockade
for cancer immunotherapy. Comput. Struct. Biotechnol. J. 13, 265–272 (2015).

163. Hodi, F. S. et al. Improved Survival with Ipilimumab in Patients with Metastatic Melanoma. N.
Engl. J. Med. 363, 1597–1607 (2010).

164. Lynch, T. J. et al. Ipilimumab in combination with paclitaxel and carboplatin as first-line
treatment in stage IIIB/IV non-small-cell lung cancer: Results from a randomized, double-blind,
multicenter phase II study. J. Clin. Oncol. 30, 2046–2054 (2012).

165. Keir, M. E., Butte, M. J., Freeman, G. J. & Sharpe, A. H. PD-1 and Its Ligands in Tolerance and
Immunity. Annu. Rev. Immunol. 26, 677–704 (2008).

166. Ohta, A. et al. A2A adenosine receptor protects tumors from antitumor T cells. Proc. Natl. Acad.
Sci. U. S. A. 103, 13132–13137 (2006).

167. Sitkovsky, M. V. et al. Hostile, hypoxia-A2-adenosinergic tumor biology as the next barrier to
overcome for tumor immunologists. Cancer Immunol. Res. 2, 598–605 (2014).

168. Zarek, P. E. et al. A2A receptor signaling promotes peripheral tolerance by inducing T-cell anergy
and the generation of adaptive regulatory T cells. Blood 111, 251–259 (2008).

169. Zou, W. Regulatory T cells, tumour immunity and immunotherapy. Nat. Rev. Immunol. 6, 295–
307 (2006).

170. Wherry, E. J. & Kurachi, M. Molecular and cellular insights into T cell exhaustion. Nat. Rev.
Immunol. 15, 486–499 (2015).

171. Poggi, A., Varesano, S. & Zocchi, M. R. How to hit mesenchymal stromal cells and make the
tumor microenvironment immunostimulant rather than immunosuppressive. Front. Immunol. 9, 1–
17 (2018).

190
172. Morrot, A. et al. Metabolic symbiosis and immunomodulation: How tumor cell-derived lactate
may disturb innate and adaptive immune responses. Front. Oncol. 8, 1–10 (2018).

173. Ohta, A. A metabolic immune checkpoint: Adenosine in Tumor Microenvironment. Front.
Immunol. 7, 1–11 (2016).

174. Sick, E. et al. CD47 update: A multifaceted actor in the tumour microenvironment of potential
therapeutic interest. Br. J. Pharmacol. 167, 1415–1430 (2012).

175. Chao, M. P., Weissman, I. L. & Majeti, R. The CD47-SIRPα Pathway in Cancer Immune Evasion
and Potential Therapeutic Implications. Curr Opin Immunol 24, 225–232 (2012).

176. Fallarino, F. et al. T cell apoptosis by tryptophan catabolism. Cell Death Differ. 9, 1069–1077
(2002).

177. Frumento, G. et al. Tryptophan-derived catabolites are responsible for inhibition of T and natural
killer cell proliferation induced by indoleamine 2,3-dioxygenase. J. Exp. Med. 196, 459–468
(2002).

178. Beavis, P. A., Stagg, J., Darcy, P. K. & Smyth, M. J. CD73: A potent suppressor of antitumor
immune responses. Trends Immunol. 33, 231–237 (2012).

179. Allard, B., Beavis, P. A., Darcy, P. K. & Stagg, J. Immunosuppressive activities of adenosine in
cancer. Curr. Opin. Pharmacol. 29, 7–16 (2016).

180. Lam, F. W., Vijayan, K. V. & Rumbaut, R. E. Platelets and their interactions with other immune
cells. Compr. Physiol. 5, 1265–1280 (2015).

181. Ribatti, D. & Crivellato, E. Giulio Bizzozero and the discovery of platelets. Leuk. Res. 31, 1339–
1341 (2007).

182. Semple, J. W., Italiano, J. E. & Freedman, J. Platelets and the immune continuum. Nat. Rev.
Immunol. 11, 264–274 (2011).

183. Thomas, M. R. & Storey, R. F. The role of platelets in inflammation. Thromb. Haemost. 1114,
449–458 (2015).

184. Sørensen, A. L. et al. Role of sialic acid for platelet life span: exposure of beta-galactose results in
the rapid clearance of platelets from the circulation by asialoglycoprotein receptor–expressing
liver macrophages and hepatocytes. Blood 114, 1645–1654 (2009).

185. Plantureux, L. et al. Impacts of Cancer on Platelet Production, Activation and Education and
Mechanisms of Cancer-Associated Thrombosis. Cancers (Basel) 10, 441 (2018).

186. Hottz, E. D. et al. Platelets mediate increased endothelium permeability in dengue through
NLRP3-inflammasome activation. Blood 122, 3405–3414 (2013).

187. Lindemann, S. et al. Activated platelets mediate inflammatory signaling by regulated interleukin
1β synthesis. J. Cell Biol. 154, 485–490 (2001).

191
188. Valeri, C. R., Ragno, G. & Khuri, S. Freezing human platelets with 6 percent dimethyl sulfoxide
with removal of the supernatant solution before freezing and storage at -80°C without postthaw
processing. Transfusion 45, 1890–1898 (2005).

189. Khuri, S. F. et al. Comparison of the effects of transfusions of cryopreserved and liquid- preserved
platelets on hemostasis and blood loss after cardiopulmonary bypass. J. Thorac. Cardiovasc. Surg.
117, 172–184 (1999).

190. Valeri, C. R., Macgregor, H. & Ragno, G. Correlation between in vitro aggregation and
thromboxane A2 production in fresh, liquid-preserved, and cryopreserved human platelets: effect
of agonists, pH, and plasma and saline resuspension. Transfusion 45, 596–603 (2005).

191. Barnard, M. R. et al. Fresh, liquid-preserved, and cryopreserved platelets-adhesive surface
receptors and membrane procoagulant activity. Transfusion 39, 880–888 (1999).

192. Raynel, S., Padula, M. P., Marks, D. C. & Johnson, L. Cryopreservation alters the membrane and
cytoskeletal protein profile of platelet microparticles. Transfusion 55, 2422–2432 (2015).

193. Zufferey, A. et al. Characterization of the platelet granule proteome : Evidence of the presence of
MHC1 in alpha-granules. J. Proteomics 101, 130–140 (2014).

194. Goubran, H. A., Burnouf, T., Stakiw, J. & Seghatchian, J. Platelet microparticle: A sensitive
physiological ‘fine tuning’ balancing factor in health and disease. Transfus. Apher. Sci. 52, 12–18
(2015).

195. Fox, J. E. B., Austin, C. D., Boyles, J. K. & Steffen, P. K. Role of the membrane skeleton in
preventing the shedding of procoagulant-rich microvesicles from the platelet plasma membrane. J.
Cell Biol. 111, 483–493 (1990).

196. Cauwenberghs, S. et al. Shedding of procoagulant microparticles from unstimulated platelets by
integrin-mediated destabilization of actin cytoskeleton. FEBS Lett. 580, 5313–5320 (2006).

197. Flaumenhaft, R. et al. Megakaryocyte-derived microparticles: direct visualization and distinction
from platelet-derived microparticles. Blood 113, 1112–1121 (2009).

198. Nomura, S. et al. High-shear-stress-induced activation of platelets and microparticles enhances
expression of cell adhesion molecules in THP-1 and endothelial cells_. Atherosclerosis 158, 277–
287 (2001).

199. Dinkla, S. et al. Platelet microparticles inhibit IL-17 production by regulatory T cells through P-
selectin. Blood 127, 1976–1986 (2016).

200. George, J. N., Thoi, L. L., Mcmanus, L. M. & Reimann, T. A. Isolation of human platelet
membrane microparticles from plasma and serum. October 60, 834–840 (1982).

201. Blair, P. & Flaumenhaft, R. Platelet a -granules : Basic biology and clinical correlates. Blood Rev.
23, 177–189 (2009).

202. Brandt, E. et al. The beta-thromboglobulins and platelet factor 4 : blood platelet-derived CXC
chemokines with divergent roles in early neutrophil regulation. J. Leukoc. Biol. 67, 471–478
(2000).

192
203. Mei, J. et al. CXCL5 Regulates Chemokine Scavenging and Pulmonary Host Defense to Bacterial
Infection. Immunity 33, 106–117 (2010).

204. Gear, A. R. L. & Camerini, D. Platelet Chemokines and Chemokine Receptors: Linking
Hemostasis, Inflammation, and Host Defense. Microcirculation 10, 335–350 (2003).

205. Hundelshausen, P. Von et al. RANTES Deposition by Platelets Triggers Monocyte Arrest on
Inflamed and Atherosclerotic Endothelium. Circulation 103, 1772–1777 (2001).

206. Moore, K. L. et al. Identification of a Specific Glycoprotein Ligand for P-selectin (CD62) on
Myeloid Cells. J. Cell Biol. 118, 445–456 (1992).

207. Zarbock, A., Polanowska-Grabowska, R. K. & Ley, K. Platelet-neutrophil-interactions: Linking
hemostasis and inflammation. Blood Rev. 21, 99–111 (2007).

208. Kwakman, P. H. S. et al. Native Thrombocidin-1 and Unfolded Thrombocidin-1 Exert
Antimicrobial Activity via Distinct Structural Elements. J. Biol. Chem. 286, 43506–43514 (2011).

209. Holmsen, H. Secretable storage pools in platelets. Annu Rev Med 30, 119–34 (1979).

210. Yeaman, M. R. Platelets: at the nexus of antimicrobial defence. Nat. Rev. Microbiol. 12, 426–437
(2014).

211. Rendu, F. & Brohard-Bohn, B. The platelet release reaction: Granules’ constituents, secretion and
functions. Platelets 12, 261–273 (2001).

212. Bentfeld-Barker, M. E. & Bainton, D. F. Identification of primary lysosomes in human
megakaryocytes and platelets. Blood 59, 472–481 (1982).

213. Kasirer-Friede, A., Kahn, M. L. & Shattil, S. J. Platelet integrins and immunoreceptors. Immunol.
Rev. 218, 247–264 (2007).

214. Bennett, J. S., Berger, B. W. & Billings, P. C. The structure and function of platelet integrins. J.
Thromb. Haemost. 7, 200–205 (2009).

215. Scharbert, G. et al. Comparison of patient intake of ticagrelor, prasugrel, or clopidogrel on
restoring platelet function by donor platelets. Transfusion 55, 1320–1326 (2015).

216. Zamora, C. et al. Binding of Platelets to Lymphocytes: A Potential Anti-Inflammatory Therapy in
Rheumatoid Arthritis. J. Immunol. 198, 3099–3108 (2017).

217. Li, N., Ji, Q. & Hjemdahl, P. Platelet-lymphocyte conjugation differs between lymphocyte
subpopulations. J. Thromb. Haemost. 4, 874–881 (2006).

218. Cerletti, C., Tamburrelli, C., Izzi, B., Gianfagna, F. & De Gaetano, G. Platelet-leukocyte
interactions in thrombosis. Thromb. Res. 129, 263–266 (2012).

219. Burns, A. R. et al. P-selectin mediates neutrophil adhesion to endothelial cell borders. J. Leukoc.
Biol. 65, 299–306 (1999).

193
220. Wang, Y. et al. Leukocyte engagement of platelet glycoprotein Ibα via the integrin Mac-1 is
critical for the biological response to vascular injury. Circulation 112, 2993–3000 (2005).

221. Armitage, R. J. et al. Molecular and biological characterization of a murine ligand for CD40.
Nature 357, 80–82 (1992).

222. Henn, V. et al. CD40 ligand on activated platelets triggers an inflammatory reaction of endothelial
cells. Nature 391, (1998).

223. Henn, V., Steinbach, S., Büchner, K., Presek, P. & Kroczek, R. A. The inflammatory action of
CD40 ligand (CD154) expressed on activated human platelets is temporally limited by
coexpressed CD40. Blood 98, 1047–1054 (2001).

224. Foy, T. M., Aruffo, A., Bajorath, J., Buhlmann, J. E. & Noelle, R. J. Immune Regulation By Cd40
and Its Ligand Gp39. Annu. Rev. Immunol. 14, 591–617 (1996).

225. Vanichakarn, P., Blair, P., Wu, C., Freedman, J. E. & Chakrabarti, S. Neutrophil CD40 enhances
platelet-mediated inflammation. Thromb. Res. 122, 346–358 (2008).

226. Duffau, P. et al. Platelet CD154 potentiates interferon-α secretion by plasmacytoid dendritic cells
in systemic lupus erythematosus. Sci. Transl. Med. 2, (2010).

227. Bussolati, B. et al. Expression of CD154 on renal cell carcinomas and effect on cell proliferation,
motility and platelet-activating factor synthesis. Int. J. Cancer 100, 654–661 (2002).

228. Ingersoll, S. B. et al. Deficiencies in the CD40 and CD154 receptor-ligand system reduce
experimental lung metastasis. Clin. Exp. Metastasis 26, 829–837 (2009).

229. Petersen, F., Bock, L., Flad, H. D. & Brandt, E. A chondroitin sulfate proteoglycan on human
neutrophils specifically binds platelet factor 4 and is involved in cell activation. J. Immunol. 161,
4347–4355 (1998).

230. Shi, G. et al. Platelet factor 4 limits Th17 differentiation and cardiac allograft rejection. J. Clin.
Invest. 124, 543–552 (2014).

231. Zhu, L., Huang, Z., Stålesen, R., Hansson, G. K. & Li, N. Platelets provoke distinct dynamics of
immune responses by differentially regulating CD4+ T-cell proliferation. J. Thromb. Haemost. 12,
1156–1165 (2014).

232. Zamora, C. et al. Functional consequences of platelet binding to T lymphocytes in inflammation.
J. Leukoc. Biol. 94, 521–529 (2013).

233. Starossom, S. C. et al. Platelets play differential role during the initiation and progression of
autoimmune neuroinflammation. Circ. Res. 117, 779–792 (2015).

234. Huang, J. et al. Elevated serum soluble CD40 ligand in cancer patients may play an
immunosuppressive role. Blood 120, 3030–3038 (2012).

235. Jenabian, M. A. et al. Soluble CD40-ligand (sCD40L, sCD154) plays an immunosuppressive role
via regulatory T cell expansion in HIV infection. Clin. Exp. Immunol. 178, 102–111 (2014).

194
236. Rinder, H. M., Bonan, J. L., Rinder, C. S., Ault, K. A. & Smith, B. R. Dynamics of leukocyte-
platelet adhesion in whole blood. Blood 78, 1730–1737 (1991).

237. Li, G. et al. Allogeneic platelet transfusions prevent murine T-cell-mediated immune
thrombocytopenia. Blood 123, 422–427 (2014).

238. Brandt, E., Ludwig, A., Petersen, F. & Flad, H.-D. Platelet-derived CXC chemokines: Old players
in new games. Immunol. Rev. 177, 204–216 (2000).

239. Boehlen, F. & Clemetson, K. J. Platelet chemokines and their receptors: What is their relevance to
platelet storage and transfusion practice? Transfus. Med. 11, 403–417 (2001).

240. León-Ponte, M., Ahern, G. P. & O’Connell, P. J. Serotonin provides an accessory signal to
enhance T-cell activation by signaling through the 5-HT7 receptor. Blood 109, 3139–3146 (2007).

241. Crawford, A., Angelosanto, J. M., Nadwodny, K. L., Blackburn, S. D. & Wherry, E. J. A role for
the chemokine RANTES in regulating CD8 T cell responses during chronic viral infection. PLoS
Pathog. 7, (2011).

242. Haemmerle, M., Stone, R. L., Menter, D. G., Afshar-Kharghan, V. & Sood, A. K. The Platelet
Lifeline to Cancer: Challenges and Opportunities. Cancer Cell 33, 965–983 (2018).

243. Mezouar, S. et al. Role of platelets in cancer and cancer-associated thrombosis: Experimental and
clinical evidences. Thromb. Res. 139, 65–76 (2016).

244. Ji, Y., Sheng, L., Du, X., Qiu, G. & Su, D. Elevated platelet count is a strong predictor of poor
prognosis in stage i non-small cell lung cancer patients. Platelets 26, 138–142 (2015).

245. Ikeda, M. et al. Poor Prognosis Associated With Thrombocytosis in Patients With Gastric Cancer.
Ann. Surg. Oncol. 9, 287–291 (2002).

246. Leader, A., Rozovski, U., Zelikson-Saporta, R. & Pereg, D. Clopidogrel Treatment Is Associated
with a Decrease in Cancer Incidence. Blood 126, 1124 (2015).

247. Suh, E. J. et al. Comparative profiling of plasma proteome from breast cancer patients reveals
thrombospondin-1 and BRWD3 as serological biomarkers. Exp. Mol. Med. 44, 36–44 (2012).

248. Starlinger, P. et al. Thrombospondin-1: A unique marker to identify in vitro platelet activation
when monitoring in vivo processes. J. Thromb. Haemost. 8, 1809–1819 (2010).

249. Ay, C. & Pabinger, I. PL-01 predictive potential of haemostatic biomarkers for venous
thromboembolism in cancer patients. Thromb. Res. 129, S6–S9 (2012).

250. Radziwon-Balicka, A. et al. Platelets increase survival of adenocarcinoma cells challenged with
anticancer drugs: Mechanisms and implications for chemoresistance. Br. J. Pharmacol. 167, 787–
804 (2012).

251. Demers, M., Ho-Tin-Noé, B., Schatzberg, D., Yang, J. J. & Wagner, D. D. Increased efficacy of
breast cancer chemotherapy in thrombocytopenic mice. Cancer Res. 71, 1540–1549 (2011).

195
252. Bottsford-Miller, J. et al. Differential platelet levels affect response to taxane-based therapy in
ovarian cancer. Clin. Cancer Res. 21, 602–610 (2015).

253. Kim, H. J. et al. Clinical Significance of Thrombocytosis Before Preoperative Chemoradiotherapy
in Rectal Cancer: Predicting Pathologic Tumor Response and Oncologic Outcome. Ann. Surg.
Oncol. 22, 513–519 (2015).

254. Yan, M. J. & Jurasz, P. The role of platelets in the tumor microenvironment: From solid tumors to
leukemia. Biochim. Biophys. Acta - Mol. Cell Res. 1863, 392–400 (2016).

255. Kitagawa, H. et al. Involvement of Platelet Membrane Glycoprotein Ib and Glycoprotein IIb/IIIa
Complex in Thrombin-dependent and -independent Platelet Aggregations Induced by Tumor
Cells. Cancer Res. 49, 537–541 (1989).

256. Felding-Habermann, B., Habermann, R., Saldívar, E. & Ruggeri, Z. M. Role of β3 integrins in
melanoma cell adhesion to activated platelets under flow. J. Biol. Chem. 271, 5892–5900 (1996).

257. Gay, L. J. & Felding-Habermann, B. Contribution of platelets to tumour metastasis. Nat. Rev.
Cancer 11, 123–134 (2011).

258. Meikle, C. K. S. et al. Cancer and thrombosis: The platelet perspective. Front. Cell Dev. Biol. 4,
1–10 (2017).

259. Wiltrout, R. H. et al. Role of organ-associated NK cells in decreased formation of experimental
metastases in lung and liver. J. Immunol. 134, 4267–4275 (1985).

260. Talmadge, J. E., Meyers, K. M., Prieur, D. J. & Starkey, J. R. Role of NK cells in tumour growth
and metastasis in beige mice. Nature 284, 622–4 (1980).

261. Nieswandt, B., Hafner, M., Echtenacher, B. & Männel, D. N. Lysis of tumor cells by natural killer
cells in mice is impeded by platelets. Cancer Res. 59, 1295–1300 (1999).

262. Philippe, C. et al. Protection from tumor necrosis factor-mediated cytolysis by platelets. Am. J.
Pathol. 143, 1713–1723 (1993).

263. Shau, H., Roth, M. & Golub, S. Regulation of natural killer function by nonlymphoid cells. Nat
Immun 12, 235–49 (1993).

264. Kopp, H. G., Placke, T. & Salih, H. R. Platelet-derived transforming growth factor-β down-
regulates NKG2D thereby inhibiting natural killer cell antitumor reactivity. Cancer Res. 69, 7775–
7783 (2009).

265. Martinez, M. C. & Andriantsitohaina, R. Microparticles in angiogenesis: Therapeutic potential.
Circ. Res. 109, 110–119 (2011).

266. Zhang, X. et al. Codelivery of carboplatn and paclitaxel via cross-linked multilamellar liposomes
for ovarian cancer treatment. RSC Adv. 7, 19685–19693 (2017).

267. Kim, H. K., Song, K. S., Chung, J. H., Lee, K. R. & Lee, S. N. Platelet microparticles induce
angiogenesis in vitro. Br. J. Haematol. 124, 376–384 (2004).

196
268. Hron, G. et al. Tissue factor-positive microparticles: cellular origin and association with
coagulation activation in patients with colorectal cancer. Thromb Haemost 97, 119–23 (2007).

269. Buergy, D., Wenz, F., Groden, C. & Brockmann, M. A. Tumor-platelet interaction in solid tumors.
Int. J. Cancer 130, 2747–2760 (2012).

270. Kim, M. et al. Preoperative thrombocytosis is a significant unfavorable prognostic factor for
patients with resectable non-small cell lung cancer. World J. Surg. Oncol. 12, 1–6 (2014).

271. Moschini, M. et al. Impact of preoperative thrombocytosis on pathological outcomes and survival
in patients treated with radical cystectomy for bladder carcinoma. Anticancer Res. 34, 3225–3230
(2014).

272. Voutsadakis, I. A. Thrombocytosis as a prognostic marker in gastrointestinal cancers. World J.
Gastrointest. Oncol. 6, 34 (2014).

273. Chadha, A. S. et al. Paraneoplastic thrombocytosis independently predicts poor prognosis in
patients with locally advanced pancreatic cancer. Acta Oncol. 54, 971–978 (2015).

274. Chen, M. & Geng, J. G. P-selectin mediates adhesion of leukocytes, platelets, and cancer cells in
inflammation, thrombosis, and cancer growth and metastasis. Arch. Immunol. Ther. Exp. 54, 75–
84 (2006).

275. Coupland, L. A., Chong, B. H. & Parish, C. R. Platelets and P-selectin control tumor cell
metastasis in an organ-specific manner and independently of NK cells. Cancer Res. 72, 4662–
4671 (2012).

276. Qi, C. et al. P-selectin-mediated platelet adhesion promotes tumor growth. Oncotarget 6, 6584–
6596 (2015).

277. Lonsdorf, A. S. et al. Engagement of αIIbβ3 (GPIIb/IIIa) with ανβ3 integrin mediates interaction
of melanoma cells with platelets: A connection to hematogenous metastasis. J. Biol. Chem. 287,
2168–2178 (2012).

278. Goubran, H. A., Burnouf, T., Radosevic, M. & El-Ekiaby, M. The platelet-cancer loop. Eur. J.
Intern. Med. 24, 393–400 (2013).

279. Rachidi, S. et al. Platelets Subvert T Cell Immunity Against Cancer via GARP-TGFβ Axis. Sci
Immunol. 2, (2017).

280. Milo, R., Jorgensen, P., Moran, U., Weber, G. & Springer, M. BioNumbers The database of key
numbers in molecular and cell biology. Nucleic Acids Res. 38, 750–753 (2009).

281. Falzoni, S., Donvito, G. & Di Virgilio, F. Detecting adenosine triphosphate in the pericellular
space. Interface Focus 3, (2013).

282. Zimmermann, H., Zebisch, M. & Sträter, N. Cellular function and molecular structure of ecto-
nucleotidases. Purinergic Signal. 8, 437–502 (2012).

283. Yegutkin, G. G. Nucleotide- and nucleoside-converting ectoenzymes: Important modulators of
purinergic signalling cascade. Biochim. Biophys. Acta - Mol. Cell Res. 1783, 673–694 (2008).

197
284. Allard, B., Longhi, M. S., Robson, S. C. & Stagg, J. The ectonucleotidases CD39 and CD73:
Novel checkpoint inhibitor targets. Immunol. Rev. 276, 121–144 (2017).

285. Zimmermann, H. Extracellular metabolism of ATP and other nucleotides. Naunyn.
Schmiedebergs. Arch. Pharmacol. 362, 299–309 (2000).

286. Stagg, J. & Smyth, M. J. Extracellular adenosine triphosphate and adenosine in cancer. Oncogene
29, 5346–5358 (2010).

287. Vijayan, D., Young, A., Teng, M. W. L. & Smyth, M. J. Targeting immunosuppressive adenosine
in cancer. Nat. Rev. Cancer 17, 709–724 (2017).

288. Jiang, T. et al. Comprehensive evaluation of NT5E/CD73 expression and its prognostic
significance in distinct types of cancers. BMC Cancer 18, 1–10 (2018).

289. Leone, R. D. & Emens, L. A. Targeting adenosine for cancer immunotherapy. J. Immunother.
Cancer 6, 1–9 (2018).

290. Morote-Garcia, J. C., Rosenberger, P., Nivillac, N. M. I., Coe, I. R. & Eltzschig, H. K. Hypoxia-
Inducible Factor-Dependent Repression of Equilibrative Nucleoside Transporter 2 Attenuates
Mucosal Inflammation During Intestinal Hypoxia. Gastroenterology 136, 607–618 (2009).

291. Eltzschig, H. K., Sitkovsky, M. V. & Robson, S. C. Purinergic signaling during inflammation. N.
Engl. J. Med. 367, 2322–2333 (2012).

292. Haskó, G., Linden, J., Cronstein, B. & Pacher, P. Adenosine receptors: Therapeutic aspects for
inflammatory and immune diseases. Nat. Rev. Drug Discov. 7, 759–770 (2008).

293. Idzko, M., Ferrari, D. & Eltzschig, H. K. Nucleotide signalling during inflammation. Nature 509,
310–317 (2014).

294. Ohta, A. & Sitkovsky, M. Role of G-protein-coupled adenosine receptors in downregulation of
inflammation and protection from tissue damage. Nature 414, (2001).

295. Di Virgilio, F., Sarti, A. C., Falzoni, S., De Marchi, E. & Adinolfi, E. Extracellular ATP and P2
purinergic signalling in the tumour microenvironment. Nat. Rev. Cancer 18, 601–618 (2018).

296. Lonnroth, P., Jansson, P. A., Fredholm, B. B. & Smith, U. Microdialysis of intercellular adenosine
concentration in subcutaneous tissue in humans. Am. J. Physiol. - Endocrinol. Metab. 256, (1989).

297. Hatfield, S. M. et al. Systemic oxygenation weakens the hypoxia and hypoxia inducible factor 1α-
dependent and extracellular adenosine-mediated tumor protection. J. Mol. Med. 92, 1283–1292
(2014).

298. Blay, J., White, T. D. & Hoskin, D. W. The extracellular fluid of solid carcinomas contains
immunosuppressive concentrations of adenosine. Cancer Res. 57, 2602–2605 (1997).

299. Pedata, F., Corsi, C., Melani, A., Bordoni, F. & Latini, S. Adenosine Extracellular Brain
Concentrations and Role of A2A Receptors in Ischemia. Ann. N. Y. Acad. Sci. 939, 74–84 (2006).

198
300. De Lera Ruiz, M., Lim, Y. H. & Zheng, J. Adenosine A2A receptor as a drug discovery target. J.
Med. Chem. 57, 3623–3650 (2014).

301. Mirabet, M. et al. Expression of A(2B) adenosine receptors in human lymphocytes: Their role in T
cell activation. J. Cell Sci. 112, 491–502 (1999).

302. Gessi, S. et al. Expression of A3 Adenosine Receptors in Human Lymphocytes: Up-Regulation in
T Cell Activation. Mol. Pharmacol. 65, 711–719 (2004).

303. Alam, M. et al. A2A adenosine receptor (AR) activation inhibits pro-inflammatory cytokine
production by human CD4+ helper T cells and regulates Helicobacter-induced gastritis and
bacterial persistence. Mucosal Immunol. 2, 232–242 (2009).

304. Cekic, C., Sag, D., Day, Y. J. & Linden, J. Extracellular adenosine regulates naive T cell
development and peripheral maintenance. J. Exp. Med. 210, 2693–2706 (2013).

305. Hoskin, D. W., Butler, J. J., Drapeau, D., Haeryfar, S. M. M. & Blay, J. Adenosine acts through an
A3 receptor to prevent the induction of murine anti-CD3-activated killer T cells. Int. J. Cancer 99,
386–395 (2002).

306. Koshiba, M., Rosin, D. L., Hayashi, N., Linden, J. & Sitkovsky, M. V. Patterns of A(2A)
extracellular adenosine receptor expression in different functional subsets of human peripheral T
cells. Flow cytometry studies with anti-A(2A) receptor monoclonal antibodies. Mol. Pharmacol.
55, 614–624 (1999).

307. Lappas, C. M., Rieger, J. M. & Linden, J. A2A Adenosine Receptor Induction Inhibits IFN-γ
Production in Murine CD4+ T Cells. J. Immunol. 174, 1073–1080 (2005).

308. Deaglio, S. et al. Adenosine generation catalyzed by CD39 and CD73 expressed on regulatory T
cells mediates immune suppression. J. Exp. Med. 204, 1257–1265 (2007).

309. Lukashev, D. E. et al. Analysis of A2a receptor-deficient mice reveals no significant
compensatory increases in the expression of A2b, A1, and A3 adenosine receptors in lymphoid
organs. Biochem. Pharmacol. 65, 2081–2090 (2003).

310. Cronstein, B. N. Adenosine, an endogenous anti-inflammatory agent. J Appl Physiol 76, 5–13
(1994).

311. Robeva, A. S. et al. Molecular characterization of recombinant human adenosine receptors. Drug
Dev. Res. 39, 243–252 (1996).

312. Sitkovsky, M. et al. Adenosine A2A receptor antagonists: Blockade of adenosinergic effects and T
regulatory cells. Br. J. Pharmacol. 153, 457–464 (2008).

313. Beavis, P. A. et al. Adenosine receptor 2A blockade increases the efficacy of anti-PD-1 through
enhanced antitumor T-cell responses. Cancer Immunol. Res. 3, 506–517 (2015).

314. Milne, G. R. & Palmer, T. M. Anti-inflammatory and Immunosuppressive Effects of the A2A
Adenosine Receptor. ScientificWorldJournal 11, 320–339 (2011).

199
315. Raskovalova, T. et al. Inhibition of cytokine production and cytotoxic activity of human
antimelanoma specific CD8+ and CD4+ T lymphocytes by adenosine-protein kinase A type I
signaling. Cancer Res. 67, 5949–5956 (2007).

316. Sevigny, C. P. et al. Activation of Adenosine 2A Receptors Attenuates Allograft Rejection and
Alloantigen Recognition. J. Immunol. 178, 4240–4249 (2007).

317. Beavis, P. A. et al. Targeting the adenosine 2A receptor enhances chimeric antigen receptor T cell
efficacy. J. Clin. Invest. 127, 929–941 (2017).

318. Leone, R. D. et al. Inhibition of the adenosine A2a receptor modulates expression of T cell
coinhibitory receptors and improves effector function for enhanced checkpoint blockade and ACT
in murine cancer models. Cancer Immunol. Immunother. 67, 1271–1284 (2018).

319. Jin, D. et al. CD73 on tumor cells impairs antitumor T-cell responses: A novel mechanism of
tumor-induced immune suppression. Cancer Res. 70, 2245–2255 (2010).

320. Kjaergaard, J., Hatfield, S., Jones, G., Ohta, A. & Sitkovsky, M. A2A Adenosine Receptor Gene
Deletion or Synthetic A 2A Antagonist Liberate Tumor-Reactive CD8 + T Cells from Tumor-
Induced Immunosuppression. J. Immunol. 201, 782–791 (2018).

321. Morello, S., Pinto, A., Blandizzi, C. & Antonioli, L. Myeloid cells in the tumor microenvironment:
Role of adenosine. Oncoimmunology 5, 1–8 (2016).

322. Haskó, G. & Pacher, P. Regulation of macrophage function by adenosine. Arterioscler. Thromb.
Vasc. Biol. 32, 865–869 (2012).

323. Hoskin, D. W., Mader, J. S., Furlong, S. J., Conrad, D. M. & Blay, J. Inhibition of T cell and
natural killer cell function by adenosine and its contribution to immune evasion by tumor cells
(Review). Int. J. Oncol. 32, 527–535 (2008).

324. Barletta, K. E., Ley, K. & Mehrad, B. Regulation of neutrophil function by adenosine.
Arterioscler. Thromb. Vasc. Biol. 32, 856–864 (2012).

325. Cronstein, B. N. & Sitkovsky, M. Adenosine and adenosine receptors in the pathogenesis and
treatment of rheumatic diseases. Nat. Rev. Rheumatol. 13, 41–51 (2017).

326. Kazemi, M. H. et al. Adenosine and adenosine receptors in the immunopathogenesis and treatment
of cancer. J. Cell. Physiol. 233, 2032–2057 (2018).

327. Novitskiy, S. V. et al. Adenosine receptors in regulation of dendritic cell differentiation and
function. Blood 112, 1822–1831 (2008).

328. Schnurr, M. et al. Role of adenosine receptors in regulating chemotaxis and cytokine production of
plasmacytoid dendritic cells. Blood 103, 1391–1397 (2004).

329. Visser, S. S., Theron, A. J., Ramafi, G., Ker, J. A. & Anderson, R. Apparent involvement of the
A(2A) subtype adenosine receptor in the anti-inflammatory interactions of CGS 21680,
cyclopentyladenosine, and IB- MECA with human neutrophils. Biochem. Pharmacol. 60, 993–999
(2000).

200
330. Scheibner, K. A. et al. The adenosine A2a receptor inhibits matrix-induced inflammation in a
novel fashion. Am. J. Respir. Cell Mol. Biol. 40, 251–259 (2009).

331. Young, A. et al. A2AR adenosine signaling suppresses natural killer cell maturation in the tumor
microenvironment. Cancer Res. 78, 1003–1016 (2018).

332. Miller, J. S., Cervenka, T., Lund, J., Okazaki, I. J. & Moss, J. Purine metabolites suppress
proliferation of human NK cells through a lineage-specific purine receptor. J. Immunol. 162,
7376–82 (1999).

333. Beavis, P. A. et al. Blockade of A2A receptors potently suppresses the metastasis of CD73+
tumors. Proc. Natl. Acad. Sci. U. S. A. 110, 14711–14716 (2013).

334. Hatfield, S. M. et al. Immunological mechanisms of the antitumor effects of supplemental
oxygenation. Sci. Transl. Med. 7, 1–13 (2015).

335. Lokshin, A. et al. Adenosine-mediated inhibition of cytotoxic activity and cytokine production by
IL-2/NKp46-Activated NK cells. Cancer Res. 66, 7758–7765 (2006).

336. Priebe, T., Nelson, J. A. & Platsoucas, C. D. Adenosine Receptors and Modulation of Natural
Killer Cell Activity by Purine Nucleosides. Cancer Res. 50, 4328–4331 (1990).

337. Williams, B. A., Manzer, A., Blay, J. & Hoskin, D. W. Adenosine acts through a novel
extracellular receptor to inhibit granule exocytosis by natural killer cells. Biochem. Biophys. Res.
Commun. 231, 264–269 (1997).

338. Mediavilla-Varela, M. et al. Antagonism of adenosine A2A receptor expressed by lung
adenocarcinoma tumor cells and cancer associated fibroblasts inhibits their growth. Cancer Biol.
Ther. 14, 860–868 (2013).

339. Sakowicz-Burkiewicz, M. et al. Differential effect of adenosine receptors on growth of human
colon cancer HCT 116 and HT-29 cell lines. Arch. Biochem. Biophys. 533, 47–54 (2013).

340. Kuzumaki, N. et al. Multiple Analyses of G-Protein Coupled Receptor (GPCR) Expression in the
Development of Gefitinib-Resistance in Transforming Non-Small-Cell Lung Cancer. PLoS One 7,
1–11 (2012).

341. Merighi, S. et al. Adenosine receptors as mediators of both cell proliferation and cell death of
cultured human melanoma cells. J. Invest. Dermatol. 119, 923–933 (2002).

342. Gessi, S. et al. Inhibition of A2A adenosine receptor signaling in cancer cells proliferation by the
novel antagonist TP455. Front. Pharmacol. 8, 1–13 (2017).

343. Zhou, J. et al. Differential impact of adenosine nucleotides released by osteocytes on breast cancer
growth and bone metastasis. Oncogene 34, 1831–1842 (2014).

344. Young, A. et al. Co-inhibition of CD73 and A2AR Adenosine Signaling Improves Anti-tumor
Immune Responses. Cancer Cell 30, 391–403 (2016).

201
345. Ohta, A. et al. The development and immunosuppressive functions of CD4+ CD25+ FoxP3+
regulatory T cells are under influence of the adenosine-A2A adenosine receptor pathway. Front.
Immunol. 3, 1–12 (2012).

346. Iannone, R., Miele, L., Maiolino, P., Pinto, A. & Morello, S. Adenosine limits the therapeutic
effectiveness of anti-CTLA4 mAb in a mouse melanoma model. Am. J. Cancer Res. 4, 172–81
(2014).

347. Ma, S. R. et al. Blockade of adenosine A2A receptor enhances CD8+ T cells response and
decreases regulatory T cells in head and neck squamous cell carcinoma. Mol. Cancer 16, 1–15
(2017).

348. Young, A. et al. Targeting adenosine in BRAF-mutant melanoma reduces tumor growth and
metastasis. Cancer Res. 77, 4684–4696 (2017).

349. Vigano, S. et al. Targeting Adenosine in Cancer Immunotherapy to Enhance T-Cell Function.
Front. Immunol. 10, 925 (2019).

350. Park, Y. et al. Assessment of pharmacokinetics and metabolism profiles of SCH 58261 in rats
using liquid chromatography-mass spectrometric method. Molecules 25, (2020).

351. Yang, M. et al. Characterization of the potency, selectivity, and pharmacokinetic profile for six
adenosine A2A receptor antagonists. Naunyn. Schmiedebergs. Arch. Pharmacol. 375, 133–144
(2007).

352. Ongini, E., Dionisotti, S., Gessi, S., Irenius, E. & Fredholm, B. B. Comparison of CGS 15943, ZM
241385 and SCH 58261 as antagonists at human adenosine receptors. Naunyn. Schmiedebergs.
Arch. Pharmacol. 359, 7–10 (1999).

353. Dionisotti, S. et al. Characterization of human A(2A) adenosine receptors with the antagonist
radioligand [3H]-SCH 58261. Br. J. Pharmacol. 121, 353–360 (1997).

354. Melani, A. et al. The selective A2A receptor antagonist SCH 58261 protects from neurological
deficit, brain damage and activation of p38 MAPK in rat focal cerebral ischemia. Brain Res.
1073–1074, 470–480 (2006).

355. U.S. Department of Health and Human Services. Inxight: Drugs: SCH-58261. National Center for
Advancing Translational Sciences Available at: https://drugs.ncats.io/substance/4309023MAH.
(Accessed: 29th May 2021)

356. Mittal, D. et al. Antimetastatic effects of blocking PD-1 and the adenosine A2A receptor. Cancer
Res. 74, 3652–3658 (2014).

357. Chen, J.-F. et al. Neuroprotection by caffeine and A(2A) adenosine receptor inactivation in a
model of Parkinson’s disease. J. Neurosci. 21, 2–7 (2001).

358. Cunha, G. M. A. et al. Adenosine A2A receptor blockade prevents memory dysfunction caused by
β-amyloid peptides but not by scopolamine or MK-801. Exp. Neurol. 210, 776–781 (2008).

202
359. Sušac, L., Eddy, M. T., Didenko, T., Stevens, R. C. & Wüthrich, K. A2A adenosine receptor
functional states characterized by 19F-NMR. Proc. Natl. Acad. Sci. U. S. A. 115, 12733–12738
(2018).

360. Eddy, M. T. et al. Allosteric Coupling of Drug Binding and Intracellular Signaling in the A2A
Adenosine Receptor. Cell 172, 68–80 (2018).

361. Popoli, P., Reggio, R. & Pèzzola, A. Effects of SCH 58261, an adenosine A(2A) receptor
antagonist, on quinpirole-induced turning in 6-hydroxydopamine-lesioned rats: Lack of tolerance
after chronic caffeine intake. Neuropsychopharmacology 22, 522–529 (2000).

362. Undevia, S. D., Gomez-Abuin, G. & Ratain, M. J. Pharmacokinetic variability of anticancer
agents. Nat. Rev. Cancer 5, 447–458 (2005).

363. Lankelma, J. et al. Doxorubicin gradients in human breast cancer. Clin. Cancer Res. 5, 1703–1707
(1999).

364. Hubbell, J. A. & Chilkoti, A. Nanomaterials for Drug Delivery. Science (80-. ). 337, 303–305
(2012).

365. Petros, R. A. & Desimone, J. M. Strategies in the design of nanoparticles for therapeutic
applications. Nat. Rev. Drug Discov. 9, 615–627 (2010).

366. Balmert, S. C. & Little, S. R. Biomimetic delivery with micro- and nanoparticles. Adv. Mater. 24,
3757–3778 (2012).

367. Jain, R. K. Delivery of molecular medicine to solid tumors: Lessons from in vivo imaging of gene
expression and function. J. Control. Release 74, 7–25 (2001).

368. Jain, R. K. Transport of Molecules in the Tumor Interstitium: A Review. Cancer Res. 47, 3039–
3051 (1987).

369. Krishna, R. & Mayer, L. D. Multidrug resistance (MDR) in cancerMechanisms, reversal using
modulators of MDR and the role of MDR modulators in influencing the pharmacokinetics of
anticancer drugs. Eur. J. Pharm. Sci. 11, 265–283 (2000).

370. Brown, R. & Links, M. Clinical relevance of the molecular mechanisms of resistance to anti-
cancer drugs. Expert Rev. Mol. Med. 1, 1–21 (1999).

371. Owens, D. E. & Peppas, N. A. Opsonization, biodistribution, and pharmacokinetics of polymeric
nanoparticles. Int. J. Pharm. 307, 93–102 (2006).

372. Gbadamosi, J. K., Hunter, A. C. & Moghimi, S. M. PEGylation of microspheres generates a
heterogeneous population of particles with differential surface characteristics and biological
performance. FEBS Lett. 532, 338–344 (2002).

373. Brigger, I., Dubernet, C. & Couvreur, P. Nanoparticles in cancer therapy and diagnosis. Adv. Drug
Deliv. Rev. 54, 631–651 (2002).

374. Zhang, L. et al. 10-Hydroxycamptothecin loaded nanoparticles: Preparation and antitumor activity
in mice. J. Control. Release 119, 153–162 (2007).

203
375. Fonseca, C., Simões, S. & Gaspar, R. Paclitaxel-loaded PLGA nanoparticles: Preparation,
physicochemical characterization and in vitro anti-tumoral activity. J. Control. Release 83, 273–
286 (2002).

376. Moghimi, S. M., Hunter, A. C. & Murray, J. C. Long-Circulating and Target-Specific
Nanoparticles: Theory to Practice. Pharmacol Rev 53, 283–318 (2001).

377. Kwon, G. et al. Enhanced tumor accumulation and prolonged circulation times of micelle-forming
poly (ethylene oxide-aspartate) block copolymer-adriamycin conjugates. J. Control. Release 29,
17–23 (1994).

378. Cabral, H., Nishiyama, N. & Kataoka, K. Optimization of (1,2-diamino-cyclohexane)platinum(II)-
loaded polymeric micelles directed to improved tumor targeting and enhanced antitumor activity.
J. Control. Release 121, 146–155 (2007).

379. Zhu, Z. et al. Paclitaxel-loaded poly(N-vinylpyrrolidone)-b-poly(ε-caprolactone) nanoparticles:
Preparation and antitumor activity in vivo. J. Control. Release 142, 438–446 (2010).

380. Kataoka, K., Harada, A. & Nagasaki, Y. Block copolymer micelles for drug delivery: Design,
characterization and biological significance. Adv. Drug Deliv. Rev. 64, 37–48 (2012).

381. Yamamoto, Y., Nagasaki, Y., Kato, Y., Sugiyama, Y. & Kataoka, K. Long-circulating
poly(ethylene glycol)-poly(D,L-lactide) block copolymer micelles with modulated surface charge.
J. Control. Release 77, 27–38 (2001).

382. Prabhakar, U. et al. Challenges and key considerations of the enhanced permeability and retention
effect for nanomedicine drug delivery in oncology. Cancer Res. 73, 2412–2417 (2013).

383. Yuan, F. Transvascular drug delivery in solid tumors. Semin Radiat Oncol 8, 164–75 (1998).

384. Juliano, R. L. Factors affecting the clearance kinetics and tissue distribution of liposomes,
microspheres and emulsions. Adv. Drug Deliv. Rev. 2, 31–54 (1988).

385. Helenius, A., Doxsey, S. & Mellman, I. Viruses as Tools in Drug Delivery. Ann. N. Y. Acad. Sci.
507, 1–6 (1987).

386. Senior, J. Fate and behavior of liposomes in vivo: a review of controlling factors. Crit Rev Ther
Drug Carr. Syst 3, 123–93 (1987).

387. Grislain, L. et al. Pharmacokinetics and distribution of a biodegradable drug-carrier. Int. J. Pharm.
15, 335–345 (1983).

388. Couvreur, P., Kante, B., Grislain, L., Roland, M. & Speiser, P. Toxicity of polyalkylcyanoacrylate
nanoparticles II: Doxorubicin‐loaded nanoparticles. J. Pharm. Sci. 71, 790–792 (1982).

389. Illum, L. & Davis, S. S. Targeting of colloidal particles to the bone marrow. Life Sci. 40, 1553–
1560 (1987).

390. Porter, C. J. H., Moghimi, S. M., Illum, L. & Davis, S. S. The polyoxyethylene/polyoxypropylene
block co-polymer Poloxamer-407 selectively redirects intravenously injected microspheres to
sinusoidal endothelial cells of rabbit bone marrow. FEBS Lett. 305, 62–66 (1992).

204
391. Litzinger, D. C. & Huang, L. Amphipathic poly(ethylene glycol) 5000-stabilized
dioleoylphosphatidylethanolamine liposomes accumulate in spleen. Biochim. Biophys. Acta 1127,
249–254 (1992).

392. Moghimi, S. M. Mechanisms of splenic clearance of blood cells and particles: towards
development of new splenotropic agents. Adv. Drug Deliv. Rev. 17, 103–115 (1995).

393. Kreuter, J., Täuber, U. & Illi, V. Distribution and elimination of poly(methyl‐2‐14C‐methacrylate)
nanoparticle radioactivity after injection in rats and mice. J. Pharm. Sci. 68, 1443–1447 (1979).

394. Tröster, S. D., Wallis, K. H., Müller, R. H. & Kreuter, J. Correlation of the surface hydrophobicity
of 14 C-poly(methyl methacrylate) nanoparticles to their body distribution. J. Control. Release 20,
247–260 (1992).

395. Tröster, S. D., Müller, U. & Kreuter, J. Modification of the body distribution of poly(methyl
methacrylate) nanoparticles in rats by coating with surfactants. Int. J. Pharm. 61, 85–100 (1990).

396. Storm, G., Belliot, S. O., Daemen, T. & Lasic, D. D. Surface modification of nanoparticles to
oppose uptake by the mononuclear phagocyte system. Adv. Drug Deliv. Rev. 17, 31–48 (1995).

397. Torchilin, V. P. & Trubetskoy, V. S. Which polymers can make nanoparticulate drug carriers
long-circulating? Adv. Drug Deliv. Rev. 16, 141–155 (1995).

398. Woodle, M. C., Newman, M. S. & Martin, F. J. Liposome leakage and blood circulation:
Comparison of adsorbed block copolymers with covalent attachment of PEG. Int. J. Pharm. 88,
327–334 (1992).

399. Bakker-Woudenberg, I. A. J. M., Storm, G. & Woodle, M. C. Liposomes in the treatment of
infections. J. Drug Target. 2, 363–371 (1994).

400. Sadelain, M., Brentjens, R. & Rivière, I. The basic principles of chimeric antigen receptor design.
Cancer Discov. 3, 388–398 (2013).

401. Grupp, S. A. et al. Durable Remissions in Children with Relapsed/Refractory ALL Treated with T
Cells Engineered with a CD19-Targeted Chimeric Antigen Receptor (CTL019). Blood 126, 681
(2015).

402. Brentjens, R. J. et al. CD19-targeted T cells rapidly induce molecular remissions in adults with
chemotherapy-refractory acute lymphoblastic leukemia. Sci. Transl. Med. 5, (2013).

403. Grupp, S. A. et al. Chimeric Antigen Receptor–Modified T Cells for Acute Lymphoid Leukemia.
N. Engl. J. Med. 368, 1509–1518 (2013).

404. Turtle, C. J. et al. CD19 CAR-T cells of defined CD4+:CD8+ composition in adult B cell ALL
patients. J. Clin. Invest. 126, 2123–2138 (2016).

405. Kochenderfer, J. N. et al. Lymphoma remissions caused by anti-CD19 chimeric antigen receptor T
cells are associated with high serum interleukin-15 levels. J. Clin. Oncol. 35, 1803–1813 (2017).

205
406. Schuster, S. J. et al. Sustained Remissions Following Chimeric Antigen Receptor Modified T
Cells Directed Against CD19 (CTL019) in Patients with Relapsed or Refractory CD19+
Lymphomas. Blood 126, 183 (2015).

407. Kochenderfer, J. N. et al. Chemotherapy-refractory diffuse large B-cell lymphoma and indolent B-
cell malignancies can be effectively treated with autologous T cells expressing an anti-CD19
chimeric antigen receptor. J. Clin. Oncol. 33, 540–549 (2015).

408. Kalos, M. et al. T cells with chimeric antigen receptors have potent antitumor effects and can
establish memory in patients with advanced leukemia. Sci Transl Med. 3, 1–21 (2011).

409. Holzinger, A., Barden, M. & Abken, H. The growing world of CAR T cell trials: a systematic
review. Cancer Immunol. Immunother. 65, 1433–1450 (2016).

410. Bonifant, C. L., Jackson, H. J., Brentjens, R. J. & Curran, K. J. Toxicity and management in CAR
T-cell therapy. Mol. Ther. - Oncolytics 3, 16011 (2016).

411. Maus, M. V. et al. T cells expressing chimeric antigen receptors can cause anaphylaxis in humans.
Cancer Immunol. Res. 1, 26–31 (2013).

412. Jensen, M. C. et al. Antitransgene rejection responses contribute to attenuated persistence of
adoptively transferred CD20/CD19-specific chimeric antigen receptor redirected T cells in
humans. Biol. Blood Marrow Transplant. 16, 1245–1256 (2010).

413. Lamers, C. H. J. et al. Immune responses to transgene and retroviral vector in patients treated with
ex vivo-engineered T cells. Blood 117, 72–82 (2011).

414. Wang, W., Singh, S., Zeng, D. L., King, K. & Nema, S. Antibody structure, instability, and
formulation. J. Pharm. Sci. 96, 1–26 (2007).

415. Long, A. H. et al. 4-1BB costimulation ameliorates T cell exhaustion induced by tonic signaling of
chimeric antigen receptors. Nat. Med. 21, (2015).

416. Koide, A. & Koide, S. Monobodies. in Protein Engineering Protocols. (Methods in Molecular
Biology) (ed. In: Arndt K.M., M. K. M.) (Humana Press., 2007). doi:10.1385/1-59745-187-8:95

417. Koide, A., Bailey, C. W., Huang, X. & Koide, S. The fibronectin type III domain as a scaffold for
novel binding proteins. J. Mol. Biol. 284, 1141–1151 (1998).

418. Hackel, B. J., Ackerman, M. E., Howland, S. W. & Wittrup, K. D. Stability and CDR Composition
Biases Enrich Binder Functionality Landscapes. J. Mol. Biol. 401, 84–96 (2010).

419. Weidle, U. H., Auer, J., Brinkmann, U., Georges, G. & Tiefenthaler, G. The emerging role of new
protein scaffold-based agents for treatment of cancer. Cancer Genomics and Proteomics 10, 155–
168 (2013).

420. Martinelli, E., De Palma, R., Orditura, M., De Vita, F. & Ciardiello, F. Anti-epidermal growth
factor receptor monoclonal antibodies in cancer therapy. Clin. Exp. Immunol. 158, 1–9 (2009).

421. Nicholson, R. I., Gee, J. M. W. & Harper, M. E. EGFR and cancer prognosis. Eur. J. Cancer 37, 9
(2001).

206
422. Sharma, S. V., Bell, D. W., Settleman, J. & Haber, D. A. Epidermal growth factor receptor
mutations in lung cancer. Nat. Rev. Cancer 7, 169–181 (2007).

423. Galizia, G. et al. Cetuximab, a chimeric human mouse anti-epidermal growth factor receptor
monoclonal antibody, in the treatment of human colorectal cancer. Oncogene 26, 3654–3660
(2007).

424. Emanuel, S. L. et al. A fibronectin scaffold approach to bispecific inhibitors of epidermal growth
factor receptor and insulin-like growth factor-I receptor. MAbs 3, 38–48 (2011).

425. Han, X. et al. Masked Chimeric Antigen Receptor for Tumor-Specific Activation. Mol. Ther. 25,
(2017).

426. Zhang, C., Hu, B., Xiao, L., Liu, Y. & Wang, P. Pseudotyping lentiviral vectors with lymphocytic
choriomeningitis virus glycoproteins for transduction of dendritic cells and in vivo immunization.
Hum. Gene Ther. Methods 25, 328–338 (2014).

427. Subik, K. et al. The expression patterns of ER, PR, HER2, CK5/6, EGFR, KI-67 and AR by
immunohistochemical analysis in breast cancer cell lines. Breast Cancer Basic Clin. Res. 4, 35–41
(2010).

428. Kim, G. P. & Grothey, A. Targeting colorectal cancer with human anti-EGFR monoclonocal
antibodies: Focus on panitumumab. Biol. Targets Ther. 2, 223–228 (2008).

429. Liu, X. et al. Affinity-tuned ErbB2 or EGFR chimeric antigen receptor T cells exhibit an increased
therapeutic index against tumors in mice. Cancer Res. 75, 3596–3607 (2015).

430. Caruso, H. G. et al. Tuning sensitivity of CAR to EGFR density limits recognition of normal
tissue while maintaining potent antitumor activity. Cancer Res. 75, 3505–3518 (2015).

431. Gan, H. K. et al. The epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor AG1478
increases the formation of inactive untethered EGFR dimers: Implications for combination therapy
with monoclonal antibody 806. J. Biol. Chem. 282, 2840–2850 (2007).

432. Chmielewski, M., Hombach, A., Heuser, C., Adams, G. P. & Abken, H. T Cell Activation by
Antibody-Like Immunoreceptors: Increase in Affinity of the Single-Chain Fragment Domain
above Threshold Does Not Increase T Cell Activation against Antigen-Positive Target Cells but
Decreases Selectivity. J. Immunol. 173, 7647–7653 (2004).

433. Hudecek, M. et al. Receptor affinity and extracellular domain modifications affect tumor
recognition by ROR1-specific chimeric antigen receptor T cells. Clin. Cancer Res. 19, 3153–3164
(2013).

434. Morgan, R. A. et al. Case report of a serious adverse event following the administration of t cells
transduced with a chimeric antigen receptor recognizing ERBB2. Mol. Ther. 18, 843–851 (2010).

435. Stone, J. D., Aggen, D. H., Schietinger, A., Schreiber, H. & Kranz, D. M. A sensitivity scale for
targeting T cells with chimeric antigen receptors (CARs) and bispecific T-cell Engagers (BiTEs).
Oncoimmunology 1, 863–873 (2012).

207
436. Zhang, T., Barber, A. & Sentman, C. L. Chimeric NKG2D-modified T cells inhibit systemic T-
cell lymphoma growth in a manner involving multiple cytokines and cytotoxic pathways. Cancer
Res. 67, 11029–11036 (2007).

437. Niederman, T. M. J. et al. Antitumor activity of cytotoxic T lymphocytes engineered to target
vascular endothelial growth factor receptors. Proc. Natl. Acad. Sci. U. S. A. 99, 7009–7014 (2002).

438. Brown, C. E. et al. Bioactivity and safety of IL13Rα2-redirected chimeric antigen receptor CD8+
T cells in patients with recurrent glioblastoma. Clin. Cancer Res. 21, 4062–4072 (2015).

439. Zah, E., Lin, M. Y., Anne, S. B., Jensen, M. C. & Chen, Y. Y. T cells expressing CD19/CD20
bispecific chimeric antigen receptors prevent antigen escape by malignant B cells. Cancer
Immunol. Res. 4, 498–508 (2016).

440. Voigt, M. et al. Functional Dissection of the Epidermal Growth Factor Receptor Epitopes
Targeted by Panitumumab and Cetuximab. Neoplasia 14, 1023–1031 (2012).

441. Yewale, C., Baradia, D., Vhora, I., Patil, S. & Misra, A. Epidermal growth factor receptor
targeting in cancer: A review of trends and strategies. Biomaterials 34, 8690–8707 (2013).

442. Xu, L. et al. Directed evolution of high-affinity antibody mimics using mRNA display. Chem.
Biol. 9, 933–942 (2002).

443. Getmanova, E. V. et al. Antagonists to Human and Mouse Vascular Endothelial Growth Factor
Receptor 2 Generated by Directed Protein Evolution In Vitro. Chem. Biol. 13, 549–556 (2006).

444. Parker, M. H. et al. Antibody mimics based on human fibronectin type three domain engineered
for thermostability and high-affinity binding to vascular endothelial growth factor receptor two.
Protein Eng. Des. Sel. 18, 435–444 (2005).

445. Ramamurthy, V. et al. Structures of adnectin/protein complexes reveal an expanded binding
footprint. Structure 20, 259–269 (2012).

446. Ho, P. & Chen, Y. Y. Mammalian synthetic biology in the age of genome editing and personalized
medicine. Curr. Opin. Chem. Biol. 40, 57–64 (2017).

447. Esensten, J. H., Bluestone, J. A. & Lim, W. A. Engineering therapeutic T cells:from synthetic
biology to clinical trials. Annu Rev Pathol 24, 305–330 (2017).

448. Koneru, M., Purdon, T. J., Spriggs, D., Koneru, S. & Brentjens, R. J. IL-12 secreting tumor-
targeted chimeric antigen receptor T cells eradicate ovarian tumors in vivo. Oncoimmunology 4,
1–11 (2015).

449. Hurton, L. V. et al. Tethered IL-15 augments antitumor activity and promotes a stem-cell memory
subset in tumor-specific T cells. Proc Natl Acad Sci U S A 113, E7788–E7797 (2016).

450. Wu, C.-Y., Roybal, K. T., Puchner, E. M., Onuffer, J. & Lim, W. A. Remote control of therapeutic
T cells through a small molecule-gated chimeric receptor. Science (80-. ). 350, aab4077 (2015).

451. Straathof, K. C. et al. An inducible caspase 9 safety switch for T-cell therapy. Blood 105, 4247–
4254 (2005).

208
452. Diaconu, I. et al. Inducible Caspase-9 Selectively Modulates the Toxicities of CD19-Specific
Chimeric Antigen Receptor-Modified T Cells. Mol. Ther. 25, 580–592 (2017).

453. Rezvani, K. & Rouce, R. H. The application of natural killer cell immunotherapy for the treatment
of cancer. Front. Immunol. 6, (2015).

454. Suck, G. et al. NK-92: an ‘off-the-shelf therapeutic’ for adoptive natural killer cell-based cancer
immunotherapy. Cancer Immunol. Immunother. 65, 485–492 (2016).

455. Zhang, C. et al. Chimeric antigen receptor-engineered NK-92 cells: An off-the-shelf cellular
therapeutic for targeted elimination of cancer cells and induction of protective antitumor
immunity. Front. Immunol. 8, (2017).

456. Zhang, J., Sun, R., Wei, H., Zhang, J. & Tian, Z. Characterization of interleukin-15 gene-modified
human natural killer cells: implications for adoptive cellular immunotherapy. Haematologica 89,
338–47 (2004).

457. Imamura, M. et al. Autonomous growth and increased cytotoxicity of natural killer cells
expressing membrane-bound interleukin-15. Blood 124, 1081–1088 (2014).

458. Sato, N., Patel, H. J., Waldmann, T. A. & Tagaya, Y. The IL-15/IL-15Rα on cell surfaces enables
sustained IL-15 activity and contributes to the long survival of CD8 memory T cells. Proc. Natl.
Acad. Sci. U. S. A. 104, 588–593 (2007).

459. Hassan, R., Bera, T. & Pastan, I. Mesothelin: A new target for immunotherapy. Clin. Cancer Res.
10, 3937–3942 (2004).

460. Carmine, C. et al. Control of large established tumor xenografts with genetically retargeted human
T cells containing CD28 and CD137 domains. Proc. Natl. Acad. Sci. U. S. A. 106, 3360–3365
(2009).

461. Tagaya, Y., Bamford, R. N., DeFilippis, A. P. & Waldmann, T. A. IL-15: A pleiotropic cytokine
with diverse receptor/signaling pathways whose expression is controlled at multiple levels.
Immunity 4, 329–336 (1996).

462. Guo, Y. et al. IL-15 Superagonist–Mediated Immunotoxicity: Role of NK Cells and IFN-γ. J.
Immunol. 195, 2353–2364 (2015).

463. Mah, A. Y. & Cooper, M. A. Metabolic regulation of natural killer cell IFN-γ production. Crit.
Rev. Immunol. 36, 131–147 (2016).

464. Liu, Y., Fang, J., Kim, Y. J., Wong, M. K. & Wang, P. Codelivery of doxorubicin and paclitaxel
by cross-linked multilamellar liposome enables synergistic antitumor activity. Mol. Pharm. 11,
1651–1661 (2014).

465. Kloss, C. C., Condomines, M., Cartellieri, M., Bachmann, M. & Sadelain, M. Combinatorial
antigen recognition with balanced signaling promotes selective tumor eradication by engineered T
cells. Nat. Biotechnol. 31, 71–75 (2013).

466. Roybal, K. T. et al. Engineering T Cells with Customized Therapeutic Response Programs Using
Synthetic Notch Receptors. Cell 167, 419–432 (2016).

209
467. Muller, W. A. Mechanisms of leukocyte transendothelial migration. Annu. Rev. Pathol. Mech. Dis.
6, 323–344 (2011).

468. Bullard, D. C. et al. Intercellular Adhesion Molecule-1 Expression Is Required on Multiple Cell
Types for the Development of Experimental Autoimmune Encephalomyelitis. J. Immunol. 178,
851–857 (2007).

469. Krieglstein, C. F. & Granger, D. N. Adhesion molecules and their role in vascular disease. Am. J.
Hypertens. 14, 44–54 (2001).

470. Stewart, M. D. & Sanderson, R. D. Heparan sulfate in the nucleus and its control of cellular
functions. Matrix Biol. 35, 56–59 (2014).

471. Caruana, I. et al. Heparanase promotes tumor infiltration and antitumor activity of CAR-redirected
T lymphocytes. Nat. Med. 21, 524–529 (2015).

472. Barlett, M., Underwood, P. & Parish, C. R. Comparative analysis of the ability of leucocytes,
endothelial cells and platelets to degrade the subendothelial basement membrane: Evidence for
cytokine dependence and detection of a novel sulfatase. Immunol. Cell Biol. 73, 113–124 (1995).

473. Golebiewska, E. M. & Poole, A. W. Platelet secretion: From haemostasis to wound healing and
beyond. Blood Rev. 29, 153–162 (2015).

474. D’Aloia, M. M., Zizzari, I. G., Sacchetti, B., Pierelli, L. & Alimandi, M. CAR-T cells: The long
and winding road to solid tumors review-article. Cell Death Dis. 9, (2018).

475. Dai, B., Xiao, L., Bryson, P. D., Fang, J. & Wang, P. PD-1/PD-L1 blockade can enhance HIV-1
Gag-specific T cell immunity elicited by dendritic cell-directed lentiviral vaccines. Mol. Ther. 20,
1800–9 (2012).

476. Engels, B. et al. Retroviral vectors for high-level transgene expression in T lymphocytes. Hum.
Gene Ther. 14, 1155–1168 (2003).

477. Siriwon, N. et al. CAR-T cells surface-engineered with drug-encapsulated nanoparticles can
ameliorate intratumoral T-cell hypofunction. Cancer Immunol. Res. 6, 812–824 (2018).

478. Mesri, M. & Altieri, D. C. Leukocyte microparticles stimulate endothelial cell cytokine release
and tissue factor induction in a JNK1 signaling pathway. J. Biol. Chem. 274, 23111–23118 (1999).

479. Forlow, S. B., McEver, R. P. & Nollert, M. U. Leukocyte-leukocyte interactions mediated by
platelet microparticles under flow. Blood 95, 1317–1323 (2000).

480. Schuerwegh, A. J., Stevens, W. J., Bridts, C. H. & De Clerck, L. S. Evaluation of monensin and
brefeldin a for flow cytometric determination of interleukin-1 beta, interleukin-6, and tumor
necrosis factor-alpha in monocytes. Commun. Clin. Cytom. 46, 172–176 (2001).

481. Li, S. et al. Enhanced cancer immunotherapy by chimeric antigen receptor–modified T cells
engineered to secrete checkpoint inhibitors. Clin. Cancer Res. 23, 6982–6992 (2017).

482. Johnson, L., Raynel, S., Seghatchian, J. & Marks, D. C. Platelet microparticles in cryopreserved
platelets: Potential mediators of haemostasis. Transfus. Apher. Sci. 53, 146–152 (2015).

210
483. Gerdes, N. et al. Platelets regulate CD4+ T-cell differentiation via multiple chemokines in
humans. Thromb Haemost 106, 353–362 (2011).

484. Ponomarev, E. D. Fresh evidence for platelets as neuronal and innate immune cells: Their role in
the activation, differentiation, and deactivation of Th1, Th17, and tregs during tissue
inflammation. Front. Immunol. 9, 1–7 (2018).

485. Meyer, A. et al. Platelet TGF-β1 contributions to plasma TGF-β1, cardiac fibrosis, and systolic
dysfunction in a mouse model of pressure overload. Blood 119, 1064–1074 (2012).

486. Grainger, D. J., Mosedale, D. E. & Metcalfe, J. C. TGF-β in blood: A complex problem. Cytokine
Growth Factor Rev. 11, 133–145 (2000).

487. Labelle, M., Begum, S. & Hynes, R. O. Direct Signaling between Platelets and Cancer Cells
Induces an Epithelial-Mesenchymal-Like Transition and Promotes Metastasis. Cancer Cell 20,
576–590 (2011).

488. Wagner, C. L. et al. Analysis of GPIIb/IIIa receptor number by quantification of 7E3 binding to
human platelets. Blood 88, 907–914 (1996).

489. Luo, B.-H., Carman, C. V. & Springer, T. A. Structural Basis of Integrin Regulation and
Signaling. Annu. Rev. Immunol. 25, 619–647 (2007).

490. Rossaint, J., Margraf, A. & Zarbock, A. Role of platelets in leukocyte recruitment and resolution
of inflammation. Front. Immunol. 9, 1–13 (2018).

491. Calvette, J. Clues for understanding the structure and function of a prototypic human integrin: the
platelet glycoprotein IIb/IIIa complex. Thromb Haemost 72, 1–15 (1994).

492. Springer, T. A. & Wang, J. H. The three-dimensional structure of integrins and their ligands, and
conformational regulation of cell adhesion. Adv. Protein Chem. 68, 29–63 (2004).

493. Kleinman, H. K. et al. Isolation and Characterization of Type IV Procollagen, Laminin, and
Heparan Sulfate Proteoglycan from the EHS Sarcoma. Biochemistry 21, 6188–6193 (1982).

494. Kleinman, H. K. et al. Basement Membrane Complexes with Biological Activity. Biochemistry
25, 312–318 (1986).

495. Brill, A., Dashevsky, O., Rivo, J., Gozal, Y. & Varon, D. Platelet-derived microparticles induce
angiogenesis and stimulate post-ischemic revascularization. Cardiovasc. Res. 67, 30–38 (2005).

496. Zheng, Y., Kong, Y. & Goetzl, E. J. Lysophosphatidic Acid Receptor-Selective Effects on Jurkat
T Cell Migration Through a Matrigel Model Basement Membrane. J. Immunol. 166, 2317–2322
(2001).

497. Salmon, H. et al. Matrix architecture defines the preferential localization and migration of T cells
into the stroma of human lung tumors. J. Clin. Invest. 122, 899–910 (2012).

498. Vedvyas, Y. et al. Manufacturing and preclinical validation of CAR T cells targeting ICAM-1 for
advanced thyroid cancer therapy. Sci. Rep. 9, 1–15 (2019).

211
499. Poznansky, M. C. et al. Thymocyte emigration is mediated by active movement away from
stroma-derived factors. J. Clin. Invest. 109, 1101–1110 (2002).

500. Li, J. et al. Chimeric antigen receptor T cell (CAR-T) immunotherapy for solid tumors: Lessons
learned and strategies for moving forward. J. Hematol. Oncol. 11, (2018).

501. Parish, C. R. et al. Treatment of central nervous system inflammation with inhibitors of basement
membrane degradation. Immunol. Cell Biol. 76, 104–113 (1998).

502. Lemjabbar-Alaoui, H. et al. Sulf-2, a heparan sulfate endosulfatase, promotes human lung
carcinogenesis. Oncogene 29, 635–646 (2010).

503. Nakajima, M., Irimura, T., Di Ferrante, D., Di Ferrante, N. & Nicolson, G. Heparan sulfate
degradation: relation to tumor invasive and metastatic properties of mouse B16 melanoma
sublines. Science (80-. ). 220, 611–3 (1983).

504. Vlodavsky, I., Fuks, Z., Bar-Ner, M., Ariav, Y. & Schirrmacher, V. Lymphoma Cell-Mediated
Degradation of Sulfated Proteoglycans in the Subendothelial Extracellular Matrix: Relationship to
Tumor Cell Metastasis. Cancer Res. 43, 2704–2711 (1983).

505. Polasky, C., Wendt, F., Pries, R. & Wollenberg, B. Platelet induced functional alteration of CD4+
and CD8+ T cells in HNSCC. Int. J. Mol. Sci. 21, 1–16 (2020).

506. Fesnak, A., Lin, C. Y., Siegel, D. L. & Maus, M. V. CAR-T Cell Therapies From the Transfusion
Medicine Perspective. Transfus. Med. Rev. 30, 139–145 (2016).

507. Muranski, P. et al. Increased intensity lymphodepletion and adoptive immunotherapy - How far
can we go? Nat. Clin. Pract. Oncol. 3, 668–681 (2006).

508. Qiu, L. et al. Idiopathic thrombocytopenic purpura treatment in a relapsed/refractory multiple
myeloma patient after chimeric antigen receptor T cell therapy. Regen. Ther. 14, 271–274 (2020).

509. Whilding, L. M., Vallath, S. & Maher, J. The integrin αvβ6: A novel target for CAR T-cell
immunotherapy? Biochem. Soc. Trans. 44, 349–355 (2016).

510. Irving, M., de Silly, R. V., Scholten, K., Dilek, N. & Coukos, G. Engineering chimeric antigen
receptor T-cells for racing in solid tumors: Don’t forget the fuel. Front. Immunol. 8, 1–19 (2017).

511. Gottesman, M. M. Mechanisms of cancer drug resistance. Annu. Rev. Med. 53, 615–27 (2002).

512. McCune, J. S. Rapid advances in immunotherapy to treat cancer. Clin. Pharmacol. Ther. 103,
540–544 (2018).

513. Glienke, W. et al. Advantages and applications of CAR-expressing natural killer cells. Front.
Pharmacol. 6, 1–7 (2015).

514. Singh, H. et al. Reprogramming CD19-specific T cells with IL-21 signaling can improve adoptive
immunotherapy of B-lineage malignancies. Cancer Res. 71, 3516–3527 (2011).

212
515. Lusty, E. et al. IL-18/IL-15/IL-12 synergy induces elevated and prolonged IFN-γ production by ex
vivo expanded NK cells which is not due to enhanced STAT4 activation. Mol. Immunol. 88, 138–
147 (2017).

516. Qu, Y. et al. Engineering CAR-expressing natural killer cells with cytokine signaling and
synthetic switch for an off-the-shelf cell-based cancer immunotherapy. MRS Commun. 9, 433–440
(2019).

517. Newick, K., Moon, E. & Albelda, S. M. Chimeric antigen receptor T-cell therapy for solid tumors.
Mol. Ther. - Oncolytics 3, 16006 (2016).

518. Dudley, M. E. et al. Cancer Regression and Autoimmunity in Patients After Clonal Repopulation
with Antitumor Lymphocytes. Science (80-. ). 298, 850–854 (2002).

519. Quail, D. F. & Joyce, J. A. Microenvironmental regulation of tumor progression and metastasis.
Nat Med 19, 1423–1437 (2013).

520. Stephan, M. T., Moon, J. J., Um, S. H., Bersthteyn, A. & Irvine, D. J. Therapeutic cell engineering
with surface-conjugated synthetic nanoparticles. Nat. Med. 16, 1035–1041 (2010).

521. Siegler, E. L. et al. Combination Cancer Therapy Using Chimeric Antigen Receptor-Engineered
Natural Killer Cells as Drug Carriers. Mol. Ther. 25, 2607–2619 (2017).

522. Fredholm, B. B., IJzerman, A. P., Jacobson, K. A., Klotz, K. N. & Linden, J. International Union
of Pharmacology. XXV. Nomenclature and classification of adenosine receptors. Pharmacol Rev
53, 527–52 (2001).

523. Cacciari, B., Pastorin, G. & Spalluto, G. Medicinal chemistry of A2A adenosine receptor
antagonists. Curr Top Med Chem 3, 403–11 (2003).

524. Joo, K. Il et al. Crosslinked multilamellar liposomes for controlled delivery of anticancer drugs.
Biomaterials 34, 3098–3109 (2013).

525. Moon, J. J. et al. Interbilayer-crosslinked multilamellar vesicles as synthetic vaccines for potent
humoral and cellular immune responses. Nat. Mater. 10, 243–251 (2011).

526. Huang, B. et al. Active targeting of chemotherapy to disseminated tumors using nanoparticle-
carrying T cells. Sci. Transl. Med. 7, (2015).

527. Stegen, S. J. C. Van Der, Hamieh, M. & Sadelain, M. The pharmacology of second- generation
chimeric antigen receptors. Nat. Rev. Drug Discov. 14, 499–509 (2015).

528. Kim, Y. J. et al. Co-Eradication of Breast Cancer Cells and Cancer Stem Cells by Cross-Linked
Multilamellar Liposomes Enhances Tumor Treatment. Mol. Pharm. 12, 2811–2822 (2015).

529. Mitchell, M. J. & King, M. R. Leukocytes as carriers for targeted cancer drug delivery. Expert
Opin. Drug Deliv. 12, 375–392 (2015).

530. Torchilin, V. P. Recent advances with liposomes as pharmaceutical carriers. Nat. Rev. Drug
Discov. 4, 145–160 (2005).

213
531. Dos Santos, N. et al. Influence of poly(ethylene glycol) grafting density and polymer length on
liposomes: Relating plasma circulation lifetimes to protein binding. Biochim. Biophys. Acta -
Biomembr. 1768, 1367–1377 (2007).

532. Yameen, B. et al. Insight into nanoparticle cellular uptake and intracellular targeting. J. Control.
Release 190, 485–499 (2014).

533. Choi, C. H. J., Hao, L., Narayan, S. P., Auyeung, E. & Mirkin, C. A. Mechanism for the
endocytosis of spherical nucleic acid nanoparticle conjugates. Proc. Natl. Acad. Sci. U. S. A. 110,
7625–7630 (2013).

534. Voigt, J., Christensen, J. & Shastri, V. P. Differential uptake of nanoparticles by endothelial cells
through polyelectrolytes with affinity for caveolae. Proc. Natl. Acad. Sci. U. S. A. 111, 2942–2947
(2014).

535. Champion, J. A. & Mitragotri, S. Shape induced inhibition of phagocytosis of polymer particles.
Pharm. Res. 26, 244–249 (2009).

536. Maghazachi, A. A. Role of chemokines in the biology of natural killer cells. Curr. Top. Microbiol.
Immunol. 341, 37–58 (2010).

537. Boissel, L., Betancur, M., Wels, W. S., Tuncer, H. & Klingemann, H. Transfection with mRNA
for CD19 specific chimeric antigen receptor restores NK cell mediated killing of CLL cells. Leuk.
Res. 33, 1255–1259 (2009).

538. Schirrmann, T. & Pecher, G. Specific targeting of CD33+ leukemia cells by a natural killer cell
line modified with a chimeric receptor. Leuk. Res. 29, 301–306 (2005).

539. Hu, W., Wang, G., Huang, D., Sui, M. & Xu, Y. Cancer immunotherapy based on natural killer
cells: Current progress and new opportunities. Front. Immunol. 10, 1–16 (2019).

540. Montalbán del Barrio, I. et al. Adenosine-generating ovarian cancer cells attract myeloid cells
which differentiate into adenosine-generating tumor associated macrophages - a self-amplifying,
CD39- and CD73-dependent mechanism for tumor immune escape. J. Immunother. Cancer 4, 1–
16 (2016).

541. Loi, S. et al. CD73 promotes anthracycline resistance and poor prognosis in triple negative breast
cancer. Proc. Natl. Acad. Sci. U. S. A. 110, 11091–11096 (2013).

542. Iwai, Y., Hamanishi, J., Chamoto, K. & Honjo, T. Cancer immunotherapies targeting the PD-1
signaling pathway. J. Biomed. Sci. 24, 1–11 (2017).

543. Lin, Y. X. et al. RNA nanotechnology-mediated cancer immunotherapy. Theranostics 10, 281–
299 (2020).

544. Huynh, M. T. 596. Effect of Chemokine/Cytokine Delivered by Nanoparticles on Tumor
Migration of Neuroblastoma Targeting Chimeric Antigen Receptor T Cells. Mol. Ther. 24, S236
(2016).

545. Lei, S. et al. Efficient Colorectal Cancer Gene Therapy with IL-15 mRNA Nanoformulation. Mol.
Pharm. 17, 3378–3391 (2020).

214
546. Tang, L. et al. Enhancing T cell therapy through TCR-signaling-responsive nanoparticle drug
delivery. Nat. Biotechnol. 36, (2018).

547. Lyman, G. H., Nguyen, A., Snyder, S., Gitlin, M. & Chung, K. C. Economic Evaluation of
Chimeric Antigen Receptor T-Cell Therapy by Site of Care among Patients with Relapsed or
Refractory Large B-Cell Lymphoma. JAMA Netw. Open 3, 1–15 (2020).

548. Fesnak, A. D., June, C. H. & Levine, B. L. Engineered T cells: The promise and challenges of
cancer immunotherapy. Nat. Rev. Cancer 16, 566–581 (2016).

549. Xie, G. et al. CAR-NK cells: A promising cellular immunotherapy for cancer. EBioMedicine 59,
(2020).

550. Philip, L. P. B. et al. Multiplex genome-edited T-cell manufacturing platform for ‘off-the-shelf’
adoptive T-cell immunotherapies. Cancer Res. 75, 3853–3864 (2015).

551. Berdien, B., Mock, U., Atanackovic, D. & Fehse, B. TALEN-mediated editing of endogenous T-
cell receptors facilitates efficient reprogramming of T lymphocytes by lentiviral gene transfer.
Gene Ther. 21, 539–548 (2014).

552. Yang, Y., Jacoby, E. & Fry, T. J. Challenges and Opportunities of Allogeneic Donor-Derived
CAR T cells. Curr Opin Hematol. 22, 509–515 (2015).

553. Rezvani, K., Rouce, R., Liu, E. & Shpall, E. Engineering Natural Killer Cells for Cancer
Immunotherapy. Mol. Ther. 25, 1769–1781 (2017).

554. Siegler, E. L., Zhu, Y., Wang, P. & Yang, L. Off-the-Shelf CAR-NK Cells for Cancer
Immunotherapy. Cell Stem Cell 23, 160–161 (2018).

555. Locke, F. L. Long-term safety and activity of axicabtagene ciloleucel in refractory large B-cell
lymphoma (ZUMA-1): a single-arm, multicentre, phase 1-2 trial. Lancet Oncol. 20, 31–42 (2019).

556. Maude, S. L. et al. Tisagenlecleucel in Children and Young Adults with B-Cell Lymphoblastic
Leukemia. N Engl J Med 378, 439–448 (2018).

557. Health Canada. Register of innovative drugs. Government of Canada (2019). Available at:
https://www.canada.ca/en/health-canada/services/drugs-health-products/drug-
products/applications-submissions/register-innovative-drugs.html. (Accessed: 31st May 2021)

558. O’Sullivan, G. M., Velickovic, Z. M., Keir, M. W., Macpherson, J. L. & Rasko, J. E. J. Cell and
gene therapy manufacturing capabilities in Australia and New Zealand. Cytotherapy 21, 1258–
1273 (2019).

559. European Medicines Agency. First two CAR-T cell medicines recommended for approval in the
European Union. Press Release (2018). Available at: https://www.ema.europa.eu/en/news/first-
two-car-t-cell-medicines-recommended-approval-european-union. (Accessed: 31st May 2021)

560. National Health Service England. CAR-T Therapy. National Health Service (2019). Available at:
https://www.england.nhs.uk/cancer/cdf/car-t-therapy/. (Accessed: 31st May 2021)

215
561. Fiorenza, S., Ritchie, D. S., Ramsey, S. D., Turtle, C. J. & Roth, J. A. Value and affordability of
CAR T-cell therapy in the United States. Bone Marrow Transplant. 55, 1706–1715 (2020).

562. Zafar, S. Y. et al. The Financial Toxicity of Cancer Treatment: A Pilot Study Assessing Out‐of‐
Pocket Expenses and the Insured Cancer Patient’s Experience. Oncologist 18, 381–390 (2013).

563. Azzani, M., Roslani, A. C. & Su, T. T. The perceived cancer-related financial hardship among
patients and their families: a systematic review. Support. Care Cancer 23, 889–898 (2015).

564. Markman, M. & Luce, R. Impact of the cost of cancer treatment: An internet-based survey. J.
Oncol. Pract. 6, 69–73 (2010).

565. Carrera, P. M., Kantarjian, H. M. & Blinder, V. S. The Financial Burden and Distress of Patients
with Cancer: Understanding and Stepping-Up Action on the Financial Toxicity of Cancer
Treatment. CA Cancer J Clin 68, 153–165 (2018).

566. Fenn, K. M. et al. Impact of financial burden of cancer on survivors’ quality of life. J. Oncol.
Pract. 10, 332–338 (2014).

567. Ell, K. et al. Economic stress among low-income women with cancer: Effects on quality of life.
Cancer 112, 616–625 (2008).

568. Castella, M. et al. Development of a Novel Anti-CD19 Chimeric Antigen Receptor: A Paradigm
for an Affordable CAR T Cell Production at Academic Institutions. Mol. Ther. - Methods Clin.
Dev. 12, 134–144 (2019).

569. Zhang, W., Jordan, K. R., Schulte, B. & Purev, E. Characterization of clinical grade CD19
chimeric antigen receptor T cells produced using automated CliniMACS Prodigy system. Drug
Des. Devel. Ther. 12, 3343–3356 (2018).

570. Ramanayake, S. et al. Low-cost generation of Good Manufacturing Practice-grade CD19-specific
chimeric antigen receptor-expressing T cells using piggyBac gene transfer and patient-derived
materials. Cytotherapy 17, 1251–1267 (2015).

571. Lin, J. K. et al. Cost effectiveness of chimeric antigen receptor T-cell therapy in relapsed or
refractory pediatric B-cell acute lymphoblastic leukemia. J. Clin. Oncol. 36, 3192–3202 (2018).

572. Grand View Research. T-cell Therapy Market Size, Share & Trends Analysis Report By Modality,
By Therapy (CAR T-cell, Tumor-Infiltrating Lymphocytes), By Indication (Hematologic
Malignancies, Solid Tumors), By Region, And Segment Forecasts, 2021 - 2028. (2021).

573. Burki, T. K. CAR T-cell therapy roll-out in low-income and middle-income countries. Lancet
Haematol. 8, e252–e253 (2021).

574. Son, K. B., Kim, C. yup & Lee, T. J. Understanding of for whom, under what conditions and how
the compulsory licensing of pharmaceuticals works in Brazil and Thailand: A realist synthesis.
Glob. Public Health 14, 122–134 (2019).

575. Sklar, T. & Robertson, C. Affordability Boards — The States’ New Fix for Drug Pricing. N. Engl.
J. Med. 381, 1301–1303 (2019).

Abstract (if available)

Abstract This dissertation is a compilation of four projects. The projects outlined herein utilize synthetic biology and combination therapy approaches with the goal of improving the antitumor efficacy of CAR-engineered immune cell therapies, especially in solid tumors. Chapter 1 gives a general introduction to cancer immunotherapy field, CAR-engineering of immune cells and challenges of using CAR-modified immune cells in the treatment of solid tumors. Chapter 1 also provides a more detailed look into the special topics that would help the reader to have an in-dept understanding of the background for the projects discussed in this dissertation. In Chapter 2, the study called 'Adnectin-Based Design of Chimeric Antigen Receptor for T Cell Engineering' showed the successful construction of adnectin-based CARs, as an alternative to using single chain variable fragments as extracellular antigen recognition domains. The results demonstrate that bearing equivalent potency to traditional CARs, adnectin-based CARs may benefit from reduced immunogenicity, increased tumor selectivity, and improved safety profile due to optimal affinity tuning. In Chapter 3, the study named 'Engineering CAR-expressing Natural Killer Cells with Cytokine Signaling and Synthetic Switch for an Off-the-shelf Cell-based Cancer Immunotherapy' demonstrated the capacity at which the off-the-shelf candidate NK-92 cells can be engineered with synthetic biology tools for enhanced tumor targeting capability, better proliferative potential, and treatment regulation. In Chapter 4, a study named 'CAR T Cell-Platelet Complexation for Enhanced Tumor Homing and Antitumor Efficacy in Solid Tumors' is presented. Inspired by the clinical success of CAR T cells as well as the intrinsic tendency of platelets (PLTs) to interact with and assist T lymphocytes to penetrate into tumors by degrading tumor basal membrane and therefore accumulate in tumor sites, we present a cell-only combinational “living-drug” strategy, CAR T cell-PLT complexes, to enhance the antitumor effects of CAR T cells in their site of action. In Chapter 5, in the study named 'CAR-Engineered Natural Killer Cells as a Carrier of Drug-Encapsulated Nanoparticles Targeting Adenosine Receptors in Solid Tumors', we combined immunotherapy with nanomedicine to better treat solid tumors. We used target-specific CAR-engineered NK cells as active carriers of nanoparticles loaded with an immunomodulatory drug to deliver the drug to solid tumor sites. We aimed to inhibit immunosuppressive effects of the tumor microenvironment and maintain CAR.NK cell effector functions in solid tumors, with minimal toxicity. Finally, Chapter 6 concludes this dissertation with future perspectives for CAR T cell therapies.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Engineering chimeric antigen receptor-directed immune cells for enhanced antitumor efficacy in solid tumors

PDF

Enhancing chimeric antigen receptor-engineered immune cell therapy with synthetic biology and nanomedicine

PDF

Novel design and combinatory therapy to enhance chimeric antigen receptor engineered T cells (CAR-T) efficacy against solid tumor

PDF

Self-secretion of checkpoint blockade enhances antitumor immunity by murine chimeric antigen receptor-engineered T cells

PDF

Engineering chimeric antigen receptor (CAR) -modified T cells for enhanced cancer immunotherapy

PDF

Engineering therapeutics for the improved antitumor efficacy of chimeric antigen receptor T cell therapy

PDF

Enhancing the specificity and cytotoxicity of chimeric antigen receptor Natural Killer cells

PDF

Development of TCR-like antibody and novel chimeric antigen receptor for cancer immunotherapy

PDF

Chimeric Antigen Receptor targeting Prostate Specific Membrane Antigen (PSMA)

PDF

Combination therapy for solid tumor

PDF

Engineered control of the CAR-T-tumor synapse using customized DNA linkers

PDF

Phospho-proteomic analysis of immune cell activation

PDF

Enhancing the anti-cancer specificity of chimeric antigen receptor T cells through targeting HLA loss

PDF

Generation and characterization of anti-CD138 chimeric antigen receptor T (CAR-T) cells for the treatment of hematologic malignancies

PDF

Lym-1 epitope targeted chimeric antigen receptor (CAR) T cells for the immunotherapy of cancer

PDF

Generation and characterization of fully human anti-CD19 chimeric antigen receptor T (CAR-T) cells for the treatment of hematologic malignancies

PDF

Therapeutic resistance in acute lymphoblastic leukemia: cIAP2 inhibition sensitizes B cell acute lymphoblastic leukemia to anti-CD19 chimeric antigen receptor T cell

PDF

Engineering immunotoxin and viral vectors for cancer therapy

PDF

Insights from a computational approach to natural killer cell activation

PDF

Optimization of nanomedicine based drug delivery systems for the treatment of solid tumors

Asset Metadata

Creator Cinay, Günce Ezgi (author)

Core Title Improving antitumor efficacy of chimeric antigen receptor-engineered immune cell therapy with synthetic biology and combination therapy approaches

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Biomedical Engineering

Degree Conferral Date 2021-12

Publication Date 11/03/2021

Defense Date 06/14/2021

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag A2aR blockade,adenosine,adenosine receptor,adnectin,adoptive cell therapy,cancer,cancer immunotherapy,CAR NK cell therapy,CAR T cell therapy,CD19,cell infiltration,cell trafficking,chimeric antigen receptor,Combination Therapy,drug delivery,EGFR,epithelial growth factor receptor,genetic engineering,genetic modification,iCas9,IL-15,immune cell therapy,immunosuppressive tumor microenvironment,immunotherapy,lentivirus vector,liposome,mesothelin,nanoparticle,natural killer cell,NK cell therapy,NK-cell therapy,OAI-PMH Harvest,platelet,retroviral vector,solid tumor,synthetic biology,T cell,T cell therapy,T-cell therapy,tumor microenvironment

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Wang, Pin (committee chair), Graham, Nicholas (committee member), McCain,Megan (committee member), Shen, Keyue (committee member)

Creator Email cinay@usc.edu,gcinay13@ku.edu.tr

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC16351960

Unique identifier UC16351960

Legacy Identifier etd-CinayGnceE-10201

Document Type Dissertation

Format application/pdf (imt)

Rights Cinay, Günce Ezgi

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu