University of Southern California Dissertations and Theses

Genomic, transcriptomic and immunologic landscapes of human cancers

(USC Thesis Other)

Genomic, transcriptomic and immunologic landscapes of human cancers

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
1
“Genomic, Transcriptomic and Immunologic Landscapes of Human
Cancers”
by
Eleni Marina (Marilena) Melas

A Dissertation Submitted to the
Faculty of the USC Graduate School at
The University of Southern California
In Partial Fulfillment of the Requirements for the
Degree of Doctor of Philosophy in
Cancer Biology and Genomics
August 2019

Doctoral Committee
Professor Michael F. Press (chair)
Professor Stephen B. Gruber (mentor)
Professor Timothy J. Triche
Associate Professor Siamak Daneshmand

2

Statement of Originality
The content of this dissertation has not previously been submitted, and to the best
of my knowledge contains no material previously published or written by another
person except where due acknowledgement is made in the report itself.
This work has been performed at the USC Norris Comprehensive Cancer Center,
which is a part of the USC Keck School of Medicine at University of Southern
California.
Eleni Marina (Marilena) Melas, 2019

© Eleni Marina (Marilena) Melas, 2019

3

I would like to dedicate this PhD dissertation to my lovely parents, Vassilis
and Aggeliki, who are my life’s greatest blessing and all of us scientists, who
are fearless in the pursuit of what sets our heart on fire.
The only real voyage of discovery consists not in seeking new landscapes, but in
having new eyes.
“Fight on” always!

4
Acknowledgements
First and foremost, I would like to express my deepest gratitude to my mentor, Dr. Stephen B. Gruber for
his remarkable supervision and support during my PhD studies. He has contributed in a fundamental way
to my doctoral studies and has been a role model in science for me with his expert guidance, teaching,
enthusiasm and motivation. He has taught me to build inner strengths following the five ideal attributes of
a Trojan: faithful, scholarly, skillful, courageous and ambitious.
My special thanks go to Dr. Kevin J. McDonnell who has been a great co-mentor to me, above and
beyond his responsibilities. His excellent guidance, insightful advice and professional help since the first
day I joined the lab, helped me tremendously to unveil my full scientific potential. He taught me that
success is the sum of small efforts repeated day in and day out. I learned to be curious, patient,
resourceful, passionate and grateful. I also want to thank Dr. Gregory Idos for his constant support,
positive attitude and friendly demeanor towards me. I owe my existence in the Gruber lab to Drs.
McDonnell and Idos who were the first to select me among other candidates and as a result I was initially
hired as a Lab Specialist by Dr. Gruber before I got accepted as a direct admit to continue my PhD studies
in his laboratory. Thanks to the three of them, the Gruber lab soon became my home away from home
when I landed in Los Angeles in March 2013 with a suitcase full of dreams.
I am very grateful to my USC professors and advisors including Dr. Ite Offringa, Dr. Baruch Frenkel, Dr.
Joyce Perez and Mrs. Bami Andrada. Also, I want to thank the USC Norris Cancer Center Facilities
Manager, Alfred Ascencio, who has always been very helpful and supportive. I am very grateful to my
PhD dissertation committee members, Dr. Michael Press, Dr. Timothy Triche and Dr. Siamak
Daneshmand for their constructive criticism and useful insights which helped me evolve as a scientist. I
am also very thankful to USC, for having received the USC student recognition award, Order of Arete,
which represents the highest honor accorded graduate students upon completion of their academic
programs.

I would also like to thank many current and past colleagues at the Gruber lab including Christopher
Walker, Chase Bowen, Dr. Harris Lazaris, Dr. Stephanie Schmit, Dr. Joe Bonner, Dr. Chenxu Qu, Dr.
Gad Rennert, Dr. Phil Boonstra and Dr. Victor Moreno who have played a fundamental role in the
completion of this dissertation. Also, many thanks to everyone who have contributed to my PhD studies
including genetic counselors (Julie Culver, Ilana Solomon, Duveen Sturgeon and Charite Ricker), support
staff (Christine Hong and the ORIEN team), collaborators (at the University of Michigan and Children’s
Hospital of Philadelphia), study participants (from Israel and Los Angeles). It would not have been
possible to get through this process without the long hours of team work, intellectual conversation and
unconditional support from my friends and colleagues at USC Norris Comprehensive Cancer Center. Two
of the professors who I owe my deepest gratitude is Dr. David W. Craig and Dr. Brooke E. Hjelm. They
gave me the chance to learn Bioinformatics and also made me realize my passion for Medical Genetics.
Through their mentorship and guidance, I got accepted to the Clinical Laboratory Genetics and Genomics
fellowship program at Nationwide Children’s Hospital starting in August 2019.
Last, but not least, I would like to thank my role models in life, my parents Vassilis and Aggeliki, to
whom I owe everything I have achieved so far being an inspiration to me. Their unconditional love,
patience and constant support during my PhD studies and throughout my life give me the strength to keep
chasing my dreams. I will try to make them proud every single day. In addition, my partner in life and
science, Harris, who taught me to “shoot for the moon” and with whom I am very fortunate to share many
life moments of personal and professional happiness and success.

5
TABLE OF CONTENTS
Acknowledgements.......................................................................................4

List of Tables..............................................................................................11

List of Figures.............................................................................................12

CHAPTER 1 - Introduction........................................................................14

1.1 History of Cancer Genomics....................................................................14
1.2 Human Genome Reference......................................................................16
1.3 Challenges of Cancer Genome Analysis......................................................17
1.4 Immunogenomics: The Parallel Study of Cancer Genomics and Immune Cell......18
1.5 Quantification of the Host Immune Response...............................................19
1.6 Personalized Medicine in different types of cancer.........................................20
1.6.1 Colorectal Cancer (CRC) ..................................................................20
1.6.2 Renal Cell Carcinoma (RCC) .............................................................22

CHAPTER 2 - Tumor Infiltrating Lymphocytes, ImmunoSEQ, and
Consensus Molecular Subtypes (CMS) classification in Colorectal
Cancer.....................................................................................................................24

2.1 Abstract..............................................................................................24

6
2.2 Introduction.........................................................................................25
2.2.1 Biology of Colorectal Cancer (CRC) ......................................................25
2.2.2 Prognostic Factors of CRC...................................................................29
2.2.3 The Molecular Epidemiology of Colorectal Cancer (MECC) Study...................30
2.2.4 Consensus Molecular Subtypes (CMS) classification ...................................31

2.3 Materials and
Methods...................................................................................................32
2.3.1 Tumor Infiltrating Lymphocytes (TIL) infiltration measurement by pathologist ....32
2.3.2 TILs infiltration measurement by ultrasequencing techniques (immunoSEQ) .......33
2.3.3 Gene Expression/Transcriptomic Profiles and CMS classification of 371 Primary
Colorectal Tumors ...................................................................................34
2.4 Results................................................................................................35
2.5 Discussion............................................................................................39
2.6 Future Directions...................................................................................42

CHAPTER 3 - Characterization of the Lynch Syndrome germline
MSH2*c.705delA private mutation in a Druze population............................45
3.1 Abstract..............................................................................................45

7
3.2 Introduction.........................................................................................47
3.3 Materials and Methods............................................................................49
3.3.1 Population.......................................................................................49
3.3.2 MSH2 Sanger Sequencing and Mutation Analysis........................................49
3.3.3 Microsatellite Short Tandem Repeat (STR) analysis.....................................50
3.3.4 Single Nucleotide Polymorphism (SNP) Array Genotyping.............................50
3.3.5 Haplotype Analysis............................................................................51
3.3.6 Somatic Mutational Analysis through Paired Tumor / Normal Whole Exome
Sequencing............................................................................................52
3.3.7 Pathologic Features and Molecular features of Cancers arising in MSH2 c.705delA
carriers.................................................................................................53
3.4 Results................................................................................................56
3.4.1 Germline genomic sequencing of the MSH2 exome identifies a private mutation,
c.705delA..............................................................................................56
3.4.2 Microsatellite marker genotyping in Druze and Christian Arabs.......................58
3.4.3 Detection of Identity By Descent (IBD) relationships....................................58
3.4.4 Haplotype Analysis............................................................................59
3.4.5 Molecular Features of Colorectal Cancer arising in Druze MSH2*c.705delA
carriers.................................................................................................61

8
3.4.6 Pathologic features of Colorectal Cancers Arising in Druze MSH2*705delA carriers
..........................................................................................................61
3.5 Discussion ...........................................................................................62

Chapter 4 - De novo Palindromic Adenine Thymidine Rich Repeat (PATRR) –
Mediated Constitutional Balanced t(3;8) Translocation Associated with Clear
Cell Renal Cell Carcinoma…………………………………............................65
4.1 Abstract..............................................................................................65
4.2 Introduction.........................................................................................66
4.2.1 Palindrome-Mediated Translocations in Humans.........................................66
4.2.2 Factors that influence PATRR-mediated translocations..................................69
4.2.3 Parental origin of de novo Balanced Translocations......................................70
4.2.4 Molecular Genetics of Renal Cell Carcinoma (RCC) ....................................72
4.2.5 Case Presentation..............................................................................74
4.2.5.1 History of Present Illness.............................................................74
4.2.5.2 Clinical Workup .......................................................................74
4.3 Materials and Methods............................................................................75
4.3.1 Lymphocytes isolation........................................................................75

9
4.3.2 Spectral Karyotyping (SKY) technique.....................................................76
4.3.2.1 Cell Pretreatment and metaphase preparation......................................76
4.3.2.2 Slide pretreatment (Pepsin digestion)...............................................77

4.3.2.3 Chromosome and probe denaturation and hybridization..........................78
4.3.2.4 Fluorescent Probe Detection..........................................................79
4.3.2.5 Image acquisition and analysis.......................................................79
4.3.3 Breakpoint Identification.........................................................................80
4.3.4 cDNA synthesis and Fusion Transcript Expression..........................................80
4.3.5 Isolation of Tumor and Germline DNA........................................................81
4.3.6 Haplotype Analysis using B Allele Frequency (BAF) ......................................81
4.3.7 Determination of the Parental of Origin of the t(3;8) translocation........................82
4.3.8 Whole exome sequencing (WES) analysis.....................................................82
4.3.9 Transcriptomic Analysis..........................................................................83
4.3.10 Generation of t(3;8) cell lines..................................................................84
4.3.10.1 Tissue collection........................................................................84
4.3.10.2 Renal Cell Carcinoma Cell Isolation.................................................84
4.3.10.3 Cytospin technique.....................................................................85
4.3.10.4 Cell immortalization...................................................................86

10

4.4 Results................................................................................................86
4.4.1 Cytogenetics....................................................................................86
4.4.2 Breakpoint Sequencing.......................................................................87
4.4.3 Fusion Transcript Expression................................................................88
4.4.4 Allele-specific Haplotype Transmission....................................................89
4.4.5 Next generation sequencing analysis........................................................91
4.4.6 Cell line establishment........................................................................92
4.4.7 Quantifying the Host Immune Response...................................................93
4.5 Discussion............................................................................................97

Chapter 5 - Concluding remarks..............................................................102

Bibliography............................................................................................105

11
List of Tables
Table 2.1 Proposed taxonomy of colorectal cancer, reflecting significant biological differences
in the gene expression-based molecular subtypes.......................................................31
Table 2.2 The demographic features of the CRC cases (n=2,193) ..................................36
Table 2.3 Pathologic and Molecular Features of Colorectal Cancers by CMS in MECC (n=371
with expression profiling) .................................................................................37
Table 3.1 Clinical Characteristics of the Druze MSH2*c.705delA Carriers........................57
Table 3.2 Short Tandem Repeat (STR) analysis of 5 markers flanking MSH2 c.705delA in Druze
carriers ........................................................................................................58
Table 3.3 Molecular Features of Colorectal Cancer arising in Druze MSH2*c.705delA
carriers.........................................................................................................61
Table 3.4 Pathologic features of Colorectal Cancers Arising in Druze MSH2*705delA
carriers.........................................................................................................61
Table 4.1 Primer design spanning the predicted fusion transcript....................................80
Table 4.2 Differential expression (tumors VS normal) for TRC8 and FHIT genes................81

12
List of Figures
Figure 2.1 CRC progression models and therapeutic targets in MSI and MSS CRC..............27
Figure 2.2 A) Pearson Correlation between TILs/hpf and TCRs/cell (n=2,092 CRC tested)
B) Pearson Correlation between TILs/hpf and TCRs/cell upon exclusion of the statistical
outliers.........................................................................................................37

Figure 2.3 Pathologic and Molecular Features of Colorectal Cancers by CMS in MECC........38

Figure 2.4 A) Pearson correlations between TILs/hpf and FOXP3, CD3, CD8 gene expression
B) Pearson correlations between TCRs/cell and FOXP3, CD3, CD8 gene expression............38

Figure 3.1 A selected region of exon 4 is shown from Sanger sequencing of germline DNA
derived from lymphocytes, with a) MSH2 wild type sequence, and b) MSH2*c.705delA........57

Figure 3.2 IBD comparison of all pairs of Druze in the combined data.............................59
Figure 3.3 A 13 Mb haplotype common to all MSH2*c.705delA mutation carriers...............60

Figure 4.1 Pedigree of the proband with the de novo constitutional PATRR mediated t(3;8)
balanced translocation.......................................................................................75

Figure 4.2 Whole exome sequencing analysis workflow..............................................83

Figure 4.3 The de novo constitutional balanced t(3;8)(p14.2;q24.1) translocation was validated
using Spectral Karyotyping (SKY) technique...........................................................87

Figure 4.4 PCR determines the breakpoint region of the t(3;8) translocation in the
proband........................................................................................................88

Figure 4.5 cDNA Sequence of the Fusion Transcript..................................................88

Figure 4.6a B allele frequency (BAF) plots for the proband’s germline (yellow)
and four of the tumor DNA samples (red) for chromosome 3.........................................89

Figure 4.6b B allele frequency (BAF) plots for the proband’s germline (yellow)
and four of the tumor DNA samples (red) for chromosome 8.........................................90

Figure 4.7a Allele specific copy number plots for each of the tumor samples for chromosome 3
using the normal (germline) sample as a reference .....................................................91

13
Figure 4.7b Allele specific copy number plots for each of the tumor samples for chromosome 8
using the normal (germline) sample as a reference.....................................................91

Figure 4.8 Volcano plot of gene expression data ......................................................92

Figure 4.9a H&E of the t(3;8) tumor.....................................................................93

Figure 4.9b H&E of the cell line t38tumK/cytospin...................................................93

Figure 4.9c H&E of the cell line t38norK/cytospin....................................................94

Figure 4.9d Cytokeratin stain of the cell line t38tumK/cytospin.....................................94
Figure 4.9e PAX8 stain of the cell line t38tumK/cytospin............................................95
Figure 4.9f CD10 stain of the cell line t38tumK/cytospin............................................95

Figure 4.10: a) Shared T-Cell Receptor β Amino Acid Sequences among the five ccRCC tumors.
b) Frequency of the single clone which is shared among all five ccRCC tumors...................96

14
Chapter 1
Introduction

1.1 History of Cancer Genomics

Cancers are highly complex and versatile diseases of the genome. Essentially all cancers are the
phenotypic result of numerous genomic and/or epigenomic alterations that have accumulated
within cells, and the interactions between the altered cells and the stromal components in a
unique host microenvironment. Over the past several decades, scientific approaches towards
cancer prevention, diagnosis, and treatment have radically shifted from organ-based to
morphology-based and most recently, to genomic and immunologic-based. Personalized or
precision medicine (tailoring a treatment for a patient’s particular disease at a precise time point
based on their unique molecular and genomics profile) is becoming the standard of care at
different levels in the clinical setting (Carpten & Mardis, 2018).

In the early 70’s, Drs. Janet Rowley, Peter Nowell, and Alfred Knudson studying leukemia cell
chromosomes under the microscope suggested that a specific chromosomal translocation and the
formation of the BCR-ABL fusion oncogene led to Chronic Myelogenous Leukemia (CML),
laying the foundation of cancer genomics (Nowell, Rowley, & Knudson, 1998). Oncogenic
proteins consequently became a focus of therapeutic development suppressing their aberrant
functions to inhibit tumor progression (Druker et al., 2000) ; (Rossari et al., 2018). A second
example of this paradigm is the breast cancer subtype known as “HER2 positive,” in which a
segment of chromosome 17 harboring the ERBB2 or HER2 gene is amplified, generating extra
copies. The resulting overabundance of HER2 protein, a tyrosine kinase, leads to the
development of an aggressive subtype of breast cancer (Slamon et al., 1987). Several HER2

15
targeted (Slamon et al., 2001) and BCR-ABL targeted therapies (Jabbour et al., 2008) are
currently FDA-approved due to their ability to prolong survival in patients whose cancers have
these specific driver alterations.

With the completion of the Human Genome Project in 2003, cancer researchers could identify
tumor-specific somatic alterations in cancers. This provided the opportunity to explore the
advent of precision drug development in the context of specific targeted therapies such as
tyrosine kinase inhibitors. Polymerase Chain Reaction (PCR) method and Sanger sequencing
were used to characterize the mutational status of genes in cancer patients who either did or did
not respond to treatment with these new small-molecule inhibitors (Lynch et al., 2004) ; ( Paez et
al., 2004) ; (Pao et al., 2004). These and other studies provide evidence of interactions between
genes, specific alterations, and therapeutics which led to novel paradigms that involved DNA
sequencing.

The next major milestone in the field of oncogenomics occurred with the advent next-generation
sequencing (NGS) methods and their application to the study of cancer genomes (Ley et al.,
2008) ; (Mardis, 2008). These methods have allowed genomic alterations from tens of thousands
of cancers to be characterized at single-nucleotide resolution and have enhanced our
understanding of how each cancer genome is created from different combinations of somatic
alterations. The sequencing of matched normal genomes from cancer patients, have enabled us to
better understand the scope of inherited and de novo germline alterations that result in cancer
susceptibility. Large studies have now illustrated that the genetic susceptibility to both adult and
pediatric cancers is high (Susswein et al., 2016) ; (Zhang et al., 2016).

16
This new understanding of cancers as diseases of the genome coupled with our ability to
comprehensively interrogate tumors enable us to exploit cancer genomic vulnerabilities using
target-specific therapies. As a result, a “precision” approach to cancer treatment has emerged,
based upon the use of NGS, advanced analytics, and databases meant to predict therapeutic
targets for individual patients based on their unique molecular and genomic profile and give us
better understanding of the mutational landscape present in many cancers (Koboldt et al., 2013).

1.2 Human Genome Reference

The human genome reference sequence is the keystone for the interpretation of NGS sequencing
read data. The alignment of reads to the human reference sequence is the first step to identify
variation of all types. Tumor and normal DNA are compared to the reference, subtracting the
differences between the two in order to decode what is uniquely somatic and what is
constitutional/inherited (Gullapalli et al., 2012). This approach elucidates each patient’s
constitutional and tumor genomes at the level of single nucleotides and large scale events such as
amplifications, copy number variations and translocations that define the drive and passenger
mutations characterizing the cancer in that individual patient.

RNA can also add huge amounts of information and insight into the pathogenesis of cancers.
Alignment and analysis of RNA sequence data provides information about gene expression and
detection of fusion transcripts which are frequently associated with cancer and are often drug
targets for cancer treatment (Serratì et al., 2016). Isoform and allele-specific expression can also
be informative. Certain genes, which are shared across tissue types, can be altered among
different cancer types and in many different ways, such as point mutations, deletions,
amplifications, translocations, inversions, changes in methylation, or even changes in terms of

17
the histone packaging. Therefore, rather than focusing on individual genes, many analytic
strategies focus on specific cellular pathways that are perturbed by the combined somatic and
germline alterations (Cancer Genome Atlas Research Network et al., 2013).

1.3 Challenges of Cancer Genome Analysis
In solid tumors, normal cells (stroma, immune cells, etc) are present to different degrees within
the complex tapestry of tumors, and malignant cells can be quantified by measures of tumor
cellularity. Certain tumor types are quite diffuse (eg. prostate, pancreas, and leukemias) and may
require specific tumor cell isolation by Laser Capture Microdissection (LCM) or flow sorting to
enhance the tumor nuclei that are being used to isolate DNA for further study. The downside of
using these methods is that these techniques can lead to very limited numbers of tumor cells and
as a result not as much DNA or RNA to use in downstream experiments. These challenges limit
preparation of adequate libraries with sufficient coverage and high quality. In addition, nucleic
acids derived from formalin fixed paraffin embedded (FFPE) preparation from pathology leads
to DNA/RNA degradation which is elevated with the age of the sample. Aneuploidy and
amplification of chromosomal segments impacts the coverage model.
Another factor that makes cancer genome analysis rather challenging is the cellular
heterogeneity; not all cells within a tumor contain all mutations and actionable mutations may be
present at only low levels (in a minor subclone). In blood or “liquid” tumors, a skin biopsy is
taken for the normal but may contain high circulating tumor cells. A cheek swab or mouthwash
normal can be used, but contamination by microbial genomes can present challenges (Valencia
& Hidalgo, 2012) ; (Vazquez, de la Torre, & Valencia, 2012).

18

1.4 Immunogenomics: The Parallel Study of Cancer Genomics and Immune Cells

Although germline and somatic alterations in cancer cells have been analyzed in depth for many
cancer sites, molecular changes in immune cells associated with disease such as cancer, have
only been more recently characterized (Sanmamed & Chen, 2018). Our immune system plays a
critical role in various biological and pathological conditions. The recent development of drugs
modulating the immune checkpoint molecules in the field of cancer immunotherapy, has clearly
demonstrated the importance of host immune cells in cancer treatments. However, the molecular
mechanisms by which these new therapeutic agents recognize and kill the cancer cells are still
not fully understood. The fields of immunogenomics and immunopharmacogenomics has been
recently emerged studying the role of newly developed tools such as NGS in the characterization
of both cancer cells and host immune cells (Zewde et al., 2018).

The observation that individual cancers contain many mutant genes which are absent in normal
tissues ignited considerable interest to elucidate the cancer epitope landscape. To further
understand such effects, in 2008 James Allison and Bert Vogelstein applied in silico-
based epitope prediction algorithms and high throughput analysis using breast and colorectal
sequencing data to identify candidate tumor antigens hypothesizing that these cancers
accumulate unique HLA epitopes (Segal et al., 2008). They proposed that as cancer is a process
where transformed tissues accumulate genetic alterations over time, all cancers would contain
mutations with a potential to form epitopes detectable by the immune system. Subsequent studies
have additionally demonstrated that cancers contain a remarkable number of mutations that could
form epitopes (Alexandrov et al., 2013).

19
In the last few years, NGS and advanced computational analysis has been utilized in order to
interpret these mutations in the context of each patient’s immune system to identify the most
potent antigenic novel tumor specific peptides that should induce an immune response. This is a
new frontier in how we interpret the genomics of the tumor, focusing on the mutational
landscape of the patient. Overall, the prediction of neoantigen load for a specific tumor can be
achieved by using NGS data in order to compare cancer and normal exomes and identify
mutations, RNA sequencing data from the cancer cells which tells us which of the mutations
identified are being expressed, the HLA haplotypes (class I and II) for each patient so we can
computationally evaluate the potential or binding these neoantigens with the immune molecules
in that patient’s system (Hundal et al., 2016).

1.5 Quantification of the Host Immune Response
The cellular adaptive immune system generates a remarkable breadth of diversity in antigen-
specific T cell receptors (TCRs). T and B cell receptor loci undergo combinatorial
rearrangement, generating a diverse immune receptor repertoire that is important for recognition
of potential antigen (Carlson et al., 2013). The TCR is composed of two peptide chains, encoded
by the TCRA and TCRB genes, respectively. The antigenic specificity of αβ T lymphocytes is in
mainly determined by the amino acid sequence in the hypervariable complementarity
determining region 3 (CDR3) regions of the TCRs (Arstila et al., 1999). The nucleotide sequence
that encodes the CDR3 loops are generated by V(D)J recombination: non-contiguous variable
(Vβ), diversity (Dβ) and joining (Jβ) genes are rearranged to form a β chain, while Vα and Jα
genes rearrange to form an α chain.

20
The existence of multiple Vα and Jα gene segments in the TCRA locus and multiple Vβ, Dβ, and
Jβ gene segments in the TCRB locus permits a large combinatorial diversity in receptor
composition, and template-independent deletion or insertion of nucleotides at the Vα-Jα, Vβ-Dβ,
and Dβ-Jβ junctions further adds to the potential diversity of receptors (Davis & Bjorkman,
1988). This rearrangement of the CDR3 region in germline DNA occurs after cell lineage
commitment, and rearranged CDR3 chains are not present in cell types other than lymphocytes.
A healthy adult has approximately 10 million different TCRB chains contained within their 10
12

circulating T cells (Robins et al., 2010).
To date, the high variability and nature of standard quantification and characterization methods
for TILS have prevented their implementation as a clinically useful prognostic marker. In
addition, the primary hallmark of a tumor-specific immune response is a large oligoclonal
expansion of T cells within a tumor, a feature which standard pathology-based methods cannot
assess. To respond to the current barriers to usage of immune response as a critical prognostic
factor, a novel method that amplifies rearranged TCRB CDR3 sequences (Robins et al., 2009)
was developed called immunoSEQ assay; it exploits the capacity of high-throughput sequencing
technology to sequence tens of thousands of TCRB CDR3 chains simultaneously.

1.6 Personalized Medicine in different types of cancer
1.6.1 Colorectal Cancer (CRC)
CRC is the third most common cancer worldwide, with more than 1.4 million new cases
diagnosed annually (Arnold et al., 2017). It is expected to cause about 51,020 deaths during 2019
in the United States. While substantial efforts have been made to reduce the population burden of

21
CRC in the US, largely attributable to increasing rates of early screening and better risk
stratification, CRC remains a public health priority and challenge. A remarkable feature of CRC
is the difference in prognosis of the early and late stages of the disease: stage I and II have low to
moderate risk of relapse after surgical resection, whereas patients with stage III have a higher
chance of recurrence (Siegel et al., 2017). Stage I and II tumors are often curable by surgical
excision, but 10% of Stage I and 20% of Stage II patients recur. Up to 73% of stage III cases are
curable by surgery combined with adjuvant chemotherapy but stage IV tumors are usually
incurable (André et al., 2015). The median survival for newly diagnosed patients with metastatic
disease is only 32.0 months, even with somatic mutational profiling and state-of-the-art
chemotherapy (Siegel et al., 2017).

CRC is a genetically heterogeneous disease with at least 80% of cases considered to be sporadic,
without any significant family history. The main genetic pathways in sporadic CRC with
significant overlap include: the chromosomal instability (CI) pathway, the microsatellite
instability (MSI) pathway and the CpG island methylator phenotype pathway (Vilar et al., 2010).
Recently, the colorectal cancer (CRC) subtyping consortium has unified six independent
molecular classification systems, based on gene expression data, into a single consensus system
with four distinct groups, known as the consensus molecular subtypes (CMS) (Guinney et al.,
2015). This reflects one emerging system of classification that is gaining traction, although not
yet firmly established.

The CMS classification system has been defined and correlated with epigenomic, transcriptomic,
microenvironmental, genetic, prognostic and clinical characteristics. The CMS1 subtype is
immunogenic and hypermutated, enriched in BRAF positive tumors. CMS2 tumors are activated
by the WNT-β-catenin pathway and have the highest overall survival. CMS3, which is

22
characterized by KRAS positive tumors, feature a metabolic cancer phenotype and finally CMS4
cancers have the worst survival and have a strong stromal gene signature. It is likely that in the
future this molecular characterization by CMS classification will enable improved
prognostication and targeted therapy in order to deliver more personalized treatment for CRC.
Novel targeted therapeutic strategies, such as immune checkpoint blockade and metabolic
normalization can be applied in highly individualized treatment regimens to improve life
expectancy even in advanced cases of colorectal carcinoma. (Thanki et al., 2017) ; (Fischer et al.,
2019).

1.6.2. Renal Cell Carcinoma (RCC)
The genomic landscape of renal cell carcinoma (RCC) has been greatly clarifie in the last few
years. Clinicians are used to classify RCC based on tumor histology, distinguishing the most
frequent clear cell RCC type from the other RCC subtypes, which are simplistically grouped as
non-clear cell RCC. To date, the only validated systems for prognostically stratifying patients
with metastatic renal cell carcinoma rely on the evaluation of clinical factors, since no molecular
biomarkers with a prognostic or predictive value have been identified so far. Efforts are made in
order to delineate signaling pathways underlying clear cell and non-clear cell RCC
carcinogenesis, possibly identifying driven-mutations as potential targets for therapy.

The motivation behind the current PhD dissertation is to describe the genomic, transcriptomic
and immunologic landscapes of human cancers, focusing on CRC and ccRCC.
The second chapter applies next-generation sequencing to quantify and characterize the T-cell
receptor (TCR) repertoire of individual CRC cancers from a large, population-based study.
Subtypes of CRC cases have distinct pathologic and molecular features that can be distinguished

23
by expression profiles.
In chapter three, the constitutional MSH2*c.705delA among a Druze population with Lynch
Syndrome is established as a private mutation and the challenges of statistical genetic analyses in
highly endogamous populations are illustrated.
Finally, in my fourth chapter, I elucidate the pathobiological consequences of a constitutional
de novo Palindromic Adenine Thymidine Rich Repeat (PATRR)- Mediated balanced t(3;8)
translocation associated with ccRCC.

24

CHAPTER 2
Tumor Infiltrating Lymphocytes, ImmunoSEQ, and
Consensus Molecular Subtypes (CMS) classification in Colorectal Cancer

2.1 Abstract
Tumor Infiltrating Lymphocytes (TILs) are prognostic and predictive biomarkers in colorectal
cancer (CRC) and are associated with improved prognosis and response to immunotherapy.
While TILs are routinely assessed by pathologists, a standardized technique (immunoSEQ,
Adaptive Biotechnologies) that leverages targeted next-generation sequencing can also be used
to quantify and characterize the T-cell receptor (TCR) repertoire of individual colorectal cancers.
In a population-based study of incident colorectal cancer (n=2139), the host immune responses
were measured by an expert pathologist and immunoSEQ to quantify and characterize the T-cell
receptor (TCR) repertoire and understand the relationship between TILs, TCRs/cell and specific
subgroups of CRC. Incident cases of adenocarcinoma of the colon or rectum from the Molecular
Epidemiology of Colorectal Cancer (MECC) study included 2,193 cancers that were uniformly
evaluated for TILs and other histopathologic features by one pathologist, Dr. Joel K. Greenson
(J.K.G). FFPE-derived DNA from macrodissected tumor tissue was extracted and sequenced
using immunoSEQ analysis for the same 2,193 individuals. A resulting quantitative metric from
this assay includes TCRs/cell, a measure of rearranged T cell quantity relative to all nucleated
cells in a tumor sample. Correlation between TILs/high power field and TCRs/cell was
quantified using Pearson Correlation Coefficient (n=2,092 cases with both measures available).
Gene expression profiles were measured for 371 primary tumors from mRNA derived from

25
snap-frozen tissue (n=337) using Affymetrix Human Genome U133 Arrays (U133A and U133
Plus2.0) and FFPE-derived mRNA quantified by RNA-Seq (n=34). Consensus Molecular
Subtypes (CMS) classification was performed using the R package 3.5.1, CMS classifier,
randomForest 4.6-14. Statistical analyses of pathologic and molecular features across the CMS
categories were performed by χ
2
tests and ANOVA. TILs/hpf and TCRs/cell were significantly
correlated (r=0.3, p<0.0001); both were found to be associated with the CMS classification and
are most enriched in the CMS1 group. Positive correlations between tumor cell infiltration and
known immunological biomarkers CD3, CD8 and FOXP3 were observed. Subtypes of MECC
CRC cases have distinct pathologic and molecular features that can be distinguished by
expression profiles.

2.2 Introduction
2.2.1. Biology of Colorectal Cancer (CRC)
Cancer is a condition of aberrant genetic programming, where changes in the genomic sequence
can potentially alter the structure, function, and expression of proteins that control cellular
processes such as growth and differentiation (Hanahan & Weinberg, 2011). This dysregulation
can lead to cellular transformation, and ultimately, tumor formation. Genetic variation in a DNA
sequence may be inherited in the germline or somatically acquired. Although inherited mutations
may lead to familial forms of cancer, the majority of neoplastic disease is sporadic and arises
from the progressive gain of somatic alterations. These somatic alterations often occur on a
background of heritable susceptibility alleles, leading to an increased risk of developing cancer
(Knudson, 1993). Each tumor may have several alterations. Driver mutations are mutations that
confer a selective growth advantage, therefore promoting cancer development (Vogelstein et al.,

26
2013). Passenger mutations are those which do not provide a growth advantage (Vogelstein et
al., 2013). The molecular events that drive the initiation, promotion, and progression of CRC
occur on many levels. This dynamic process involves interactions among environmental
influences, germline factors dictating individual cancer susceptibility, and accumulated somatic
changes in the colorectal epithelium.

Most colorectal cancers (CRC) begin as a growth of the epithelial lining of the colon or rectum.
These growths are called polyps. Some types of polyps can change into cancer over time but not
all polyps lead to cancer. The chance of a polyp changing into cancer depends on the nature of
the polyp. There are two main types of polyps: a) hyperplastic polyps and inflammatory polyps
which are more common but in general they are not pre-cancerous, b) adenomatous polyps
(adenomas) which are pre-cancerous and have greater potential for malignant transformation.
Adenomatous polyps can develop into an advanced adenoma with high-grade dysplasia and
subsequently progresses to an invasive cancer. Moreover, a premalignant lesion of the colon, is
the sessile serrated adenoma (SSA) predominantly seen in the cecum and ascending colon.
Multiple SSAs may be part of the serrated polyposis syndrome (Kinzler & Vogelstein, n.d.)
(Walther et al., 2009). Genetic models for colorectal carcinogenesis have previously been well
described (Figure 2.1) based on molecular, genetic, and epigenetic patterns, including the
chromosomally instable (CIN), microsatellite instability (MSI) and CpG island methylator
phenotype (CIMP) pathways (Vogelstein et al., 1988) ; (Vilar & Gruber, 2010).

27
Figure 2.1: CRC progression models and therapeutic targets in MSI and MSS CRC. Molecular
CRC groups based on a) chromosomal instability (MSS in blue) and b) the mutator phenotype (MSI in
grey). The genetic models for CRC tumorigenesis are presented in parallel for each pathway of tumor
development. Targeted therapies based on molecular events are also presented for MSI tumors.
Abbreviations: CRC, colorectal cancer; MSI, microsatellite instability; MSS, microsatellite stable. Figure
adapted from (Vilar & Gruber, 2010).
Mutations in key genes in these pathways have been identified in a small number of rare, highly
penetrant familial syndromes, accounting for less than 5% of all CRC (Goss & Groden, 2000)
(Kemp et al., 2004) Marra & Boland, 1995). Major CRC genetic syndromes include familial
adenomatous polyposis (FAP), MUTYH-associated polyposis (MAP), and Lynch syndrome
(hereditary nonpolyposis colorectal cancer or HNPCC) (Kohlmann & Gruber, 1993) ; (Lynch et

28
al., 2009). MUTYH-associated polyposis (MAP) is an autosomal recessive disorder
characterized by adenomatous polyps of the colon and rectum and a very high risk of CRC. MAP
is associated with biallelic mutations in the MUTYH gene located on chromosome 1p, which
encodes a protein in the DNA base excision repair pathway whose impaired function leads to
increased G:C to T:A transversions (Al-Tassan et al., 2002). McDonnell et al., recently published
a study on a MUTYH variant, p.C306W (c.918C>G), with a tryptophan residue in place of
native cysteine, that ligates the [4Fe4S] cluster in a patient with colonic polyposis and family
history of early colon cancer (McDonnell et al., 2018) illustrating that pathogenic mutations can
be related to disrupted electron transfer mediating defective DNA repair.

Family history of non-syndromic CRC is associated with a two-fold increase in risk (Carstensen
et al., 1996) and is estimated to account for ~35% of CRC (Tenesa & Dunlop, 2009). Familial
non-syndromic CRC is thought to be attributed to more common, low- to moderate- penetrance
mutations, mainly based on the identification of variants through candidate gene studies.
Examples of these genetic variants include the I1307K mutation in the APC gene (Laken et al.,
1997) and variation in TGFBR1 (de Jong et al., 2002). Genome-Wide Association Study
(GWAS) designs have now been widely implemented in colorectal cancer, resulting in the
identification of many new loci associated with CRC such as 8q23.3, 8q24, 10p14, 11q23,
15q13, and 18q21 (Broderick et al., 2007) ; (Gruber et al., 2007) ; (Broderick et al., 2007) ;
(COGENT Study et al., 2008) ; (Tenesa & Dunlop, 2009), (Zanke et al., 2007). Indeed, recently
published GWAS findings now describe more than 100 novel common genetic susceptibility loci
for colorectal cancer (Schumacher et al., 2015) ; (Schmit et al., 2019) ; (Huyghe et al., 2019).
Although many common genetic variants have now been statistically associated with CRC, it

29
appears that the majority common variants explain only a small fraction of the genetic risk
(Peters et al., 2012).

2.2.2 Prognostic Factors of CRC
Prognosis for CRC is highly dependent on the stage at diagnosis, treatment, and a limited
number of well characterized host and tumor features, including MSI and immune host response.
A number of factors have definitively proven to be of prognostic importance, including the
AJCC / TNM staging that incorporates features including vascular or lymphatic invasion,
perineural invasion, extranodal deposits, residual tumor following surgery, and preoperative
elevation of carcinoembryonic antigen. Other factors examined in multiple investigations that
have not yet been incorporated into the AJCC staging system include a high degree of
microsatellite instability (MSI-H), host lymphoid response to the tumor, and specific mutational
profiles (Phipps et al., 2015) ; (Walther et al., 2009). One of the most important recently
identified prognostic factors in CRC is the host immune response triggered by abnormal surface
proteins expressed on tumor cells, as classically measured by expert pathologists with
histopathologic review (Greenson et al., 2009) ; (Rozek et al., 2016).

Tumor lymphocyte infiltration has been shown to be an important prognostic factor for overall
and disease-specific survival in CRC and other cancers (Gooden et al., 2011) ; (Huh, Lee, &
Kim, 2012) ; (Lee et al., 2010) ; (Ohtani, 2007) ; (Smyrk et al., 2001). In CRC, the presence of
TILs is more common in MSI-H than MSS colorectal cancers (Greenson et al., 2003, 2009). For
tumors which exhibit a high number of TILs (high T-cell infiltrate), the prognosis for patients
with strong tumor expression of major histocompatibility antigen class I (MHC 1) is more
favorable than for those who have weak MHC 1 expression, likely attributable to the hypothesis

30
that the immune response mounted against the tumor cells is weaker in these patients (Simpson
et al., 2010).

2.2.3 The Molecular Epidemiology of Colorectal Cancer (MECC) Study

The Molecular Epidemiology of Colorectal Cancer Study (MECC) is a population-based case-
control study of pathologically-confirmed, incident cases of CRC recruited from a
geographically-defined region of northern Israel (Poynter et al., 2005). The MECC study uses the
modified staging system of the 7th edition of the AJCC Cancer Staging manual, with uniform
pathologic review and standardized staging. All newly diagnosed CRC cases from March 31,
1998, to June 30, 2017 who signed an IRB-approved informed consent, were invited to be
interviewed, provide a venous blood sample and gave permission to retrieve and analyze their
paraffin embedded tumor tissue. All cases (and controls) provided extensive information about
their personal characteristics, detailed cancer family history, personal medical history and their
exposure to a variety of risk factors. Individuals previously diagnosed with CRC are not eligible
to participate. Written, informed consent was obtained according to Institutional Review Board-
approved protocols at Carmel Medical Center in Haifa and the University of Southern California
(HS-12-00324, HS-12-00672, and HS-08-00378). A total of 6,252 cases were recruited into the
MECC study and were abstracted by physician reviewers to record all treatment details,
including extent of surgery, use of adjuvant radiation therapy (primarily for rectal cancers), and
specific chemotherapy regimens (including doses, dose modifications, and therapeutic changes
for progressive disease) for a subset of cases. Electronic prescription records were used to verify
self-reported medication use as previously described.

31
2.2.4 Consensus Molecular Subtypes (CMS) classification
The characterization of the molecular basis of CRC has led to the identification of favorable and
non-favorable immunologic features associated with clinical outcome (Roelands et al., 2017). In
order to resolve inconsistencies among the reported gene expression-based CRC classifications
and improve clinical translation, an international consortium has been formed dedicated to large-
scale data sharing across experts. This consortium classified CRC into four consensus molecular
subtypes (CMS) with unique molecular features (Table 2.1) (Guinney et al., 2015). CMS groups
are considered one of the most robust classification systems currently available for CRC, with
clear biological interpretability and the basis for subtype-based future targeted clinical
interventions.
Characteristic CMS1
(MSI Immune)
CMS2
(Canonical)
CMS3
(Metabolic)
CMS4
(Mesenchymal)
Frequency 14% 37% 13% 23%
Mutational
Profile
MSI, CIMP
high,
Hypermutation
SCNA high Mixed MSI
status, SCNA
low, CIMP low
SCNA high
Molecular
Features
Immune
Infiltration and
Activation
WNT and MYC
Activation
Metabolic
Deregulation
Stromal
Infiltration,
TGF-β
Activation,
Angiogenesis
Outcomes Worse survival
after relapse
Highest
Survival
Worse relapse-
free and overall
survival

Table 2.1: Proposed taxonomy of colorectal cancer, reflecting significant biological
differences in the gene expression-based molecular subtypes. CIMP, CpG island methylator
phenotype; MSI, microsatellite instability; SCNA, somatic copy number alterations (Guinney et al., 2015).

32
2.3 Materials and Methods
In the MECC population-based study of incident colorectal adenocarcinoma in Israel (n=2,193)
host immune responses were measured by an expert pathologist (TILs/high power field) and
immunoSEQ (TCRs/cell) to quantify and characterize the T-cell receptor (TCR) repertoire and
understand the relationship between TILs/hpf, TCRs/cell and specific subgroups of CRC. FFPE
derived DNA from macrodissected tumor tissue was extracted (AllPrep Qiagen) and sequenced
using immunoSeq. Correlation between TILs/hpf and TCRs/cell was quantified using the
Pearson correlation coefficient (n=2,092 cases with both measures) in RStudio. Expression
profiles for 371 primary tumors from mRNA were derived from snap-frozen tissue (n=337) using
Affymetrix Human Genome Arrays (U133A and U133 Plus2.0), & FFPE-derived mRNA
quantified by RNASeq (n=34). CMS classification was performed using R package 3.5.1, CMS
classifier, randomForest 4.6. Correlations between gene expression of known immunological
biomarkers such as CD3, CD8, FOXP3 with TILs/hpf and TCRs/cell were quantified using
Pearson correlation coefficient (n=33 CRC tested). Statistical analyses of pathologic and
molecular features across the CMS categories were performed by 𝛘
2
tests and one-way ANOV A
using RStudio and GraphPad Prism8 with p> 0.05 denoting no significant statistical difference

2.3.1 Tumor Infiltrating Lymphocytes (TIL) infiltration measurement by
pathologist
The MECC study took advantage of uniform histopathologic review by a single pathologist, Dr.
Joel K. Greenson (J.K.G). Methods and procedures for pathologic evaluation have been
previously described (Rozek et al., 2016). In brief, tumor infiltrating lymphocytes (TILs) were
identified on hematoxylin and eosin (H&E) -stained sections as small blue mononuclear cells

33
typically surrounded by a halo. Only cells infiltrating between tumor cells (inta-tumoral) were
counted. Apoptotic cells were excluded from the counting process. H&E stained slides were
scored for histology, grade, TILs cells per HPF, and crohn’s like reaction (CLR), among other
features. Analyses were restricted to cases with pathologically confirmed adenocarcinoma
(n=2,092). The tumor was scanned at low power to look for the area with the majority of TILs.
Once this area was identified, five consecutive 40× fields of an Olympus BX40 microscope with
a UPlanFl objective (Olympus America Inc., Melville, NY, U.S.A.) were counted (total area
equal to 0.94 mm
2
). The mean TIL/high power field for each tumor was then calculated by
dividing the total number of TIL by 5.

2.3.2 TILs infiltration measurement by ultrasequencing techniques
(immunoSEQ)

In brief, using a commercial vendor (Adaptive Biotechnologies, Seattle, WA), a multiplex PCR
system was used to amplify CDR3β sequences from DNA samples. The immunoSEQ assay can
amplify all 49 V segments in 32 gene segment families, 8 pseudogenes segments in 7 gene
families, 10 orphan segments in 10 gene families, both D genes and the 13 functional J segments.
This approach generates an 87 base-pair fragment capable of identifying the VDJ region
spanning each unique CDR3β. The primers were designed to amplify the shortest feasible
fragment containing the CDR3 region, to ensure that degraded DNA from an FFPE sample is
still able to harbor an amplifiable fragment. Amplicons were sequenced using the Illumina HiSeq
platform. Using a baseline developed from synthetic templates, primer concentrations and
computational/statistical analyses were applied to correct for the primer bias common to

34
multiplex PCR reactions. Raw sequence data was filtered based on the TCRβ V, D and J gene
definitions provided by the IMGT database (www.imgt.org) and binned using a modified
nearest-neighbor algorithm to merge closely related sequences and remove both PCR and
sequencing artifacts. Sequencing results were normalized based on an assay targeting
housekeeping genes with the same length amplicons, so that TIL fraction in a sample can be
measured independent of the level of degradation. This approach enables the detection of 1 in
200,000 cells (Robins et al., 2009). One of the resulting quantitative metrics from immunoSEQ
assay which was used in our analyses includes TCRs/cell, a measure of rearranged T cell
quantity relative to all nucleated cells in a tumor sample.

2.3.3. Gene Expression/Transcriptomic Profiles and CMS classification of 371
Primary Colorectal Tumors

Gene expression data were obtained for 371 primary colorectal tumors (Sanz-Pamplona et al.,
2011) . Affymetrix Human Genome Arrays (U133A and U133 Plus2.0) were used to measure the
expression profiles for 337 tumors from mRNA derived from snap-frozen tissue. In addition,
FFPE-derived mRNA (AllPrep Qiagen) for 34 tumors was quantified by RNA-seq at 20M. For
tumor RNA-seq libraries preparation, TruSeq RNA Exome Kit was used (Illumina). CMS
classification was performed using R package 3.5.1, CMS classifier, randomForest 4.6 (Guinney
et al., 2015). Correlations between the gene expression of known immunological biomarkers
such as CD3, CD8, FOXP3 with TILs/hpf and TCRs/cell were quantified using Pearson
correlation coefficient (n=33 CRC tested).

35
2.4 Results
The demographic features of the CRC cases in our population based study (n=2,193) are
summarized in Table 2.2. In brief, the mean age of patients with adenocarcinoma was 69.6
years; amongst them, 52.1% were male and the rest female. Regarding their ethnicity, the
majority were Ashkenazi Jews (63.0%), followed by Sephardi (19.5%), Arab (12.1%) and others
(5.4%). The Pearson Correlation coefficient test for the 2,092 CRC cases with both TILs/hpf and
TCRs/cell metrics, showed a statistically significant moderate positive correlation (r=0.3,
p<0.0001) (Figure 2.2). The categorization of 371 CRC cases in MECC with expression
profiling data into the CMS groups together with their pathologic and molecular features by
CMS are summarized in the Table 2.3 and Figure 2.3 (for easier visualization of the results
using bar plots). The majority of the CRC cases tested were classified in the CMS2 group
(144/371, 39%) followed by the CMS4 group (90/371, 24%), the CMS3 group (60/371, 16%),
the CMS1 group (54/371, 15%) and finally the group which contains mixed features (23/371,
6%). The mean TILs/hpf in the 371 CRC cases with gene expression data was 2.2 with SD 5.3.
The mean TCRs/cell in the same cohort was 0.12 with SD 0.09.

TILs/hpf and TCRs/cell were both associated with the CMS classification and were most
enriched in the CMS1 group. In particular, the mean TILs/hpf was statistically significant higher
in the CMS1 group (6.3, p<0.0001) and similarly the mean TCRs/cell higher in the same group
(0.12, p=0.04). MSI-H tumors were most frequently observed within CMS1 cancers (43% of
CMS1 were MSI-H, p<0.0001), followed by CMS3 group (18.8% of CMS3 were MSI-H), the
CMS4 group (10.1% of CMS4 were MSI-H), the group with mixed features (9% of the mixed
features group were MSI-H) and finally the CMS2 group (5.6% of CMS2 group were MSI-H).

36
BRAF positive tumors more frequently observed within the CMS1 group (16%, p =0.003).
KRAS positive tumors more frequently observed within the CMS3 group (49%, p<0.0001).

We observed a statistically significant positive correlation between TILs/hpf, and known
immunological biomarkers including FOXP3, CD3 and CD8 (r=0.5 with p=0.008, r=0.6 with
p=0.0003 and r=0.8 with p<0.0001 respectively) (Figure 2.4A). Interestingly, a similar
association pattern was observed between gene expression of the same biomarkers and TCRs/cell
(r=0.6 with p<0.0001, r=0.5 with p=0.02 and r=0.7 with p<0.0001 respectively) (Figure 2.4B).

Table 2.2: The demographic features of the CRC cases (n=2,193)
Demographic Characteristics Mean (SD) or Frequency (%)
Age (years) 69.6 (12.1)
Sex
Male 1,141 (52.1%)
Female 1,052 (47.9%)
Ethnicity
Ashkenazi 1,380 (63.0%)
Sephardi 428 (19.5%)
Arab 266 (12.1%)
Other 119 (5.4 %)
Total 2,193

37

Figure 2.2: A) Pearson Correlation between TILs/hpf and TCRs/cell (n=2,092 CRC tested)
B) Pearson Correlation between TILs/hpf and TCRs/cell upon exclusion of the
statistical outliers.

Table 2.3: Pathologic and Molecular Features of Colorectal Cancers by CMS in MECC
(n=371 with expression profiling). MSI, BRAF and KRAS data missing for 51, 36 and 26 cases
respectively

All
Samples
(n=371)
CMS1 CMS2 CMS3 CMS4 Mixed
Features
p value
TILs/hpf
Mean (SD)
2.2 (5.3) 6.3 1.5 2.2 1.8 1.7 <0.0001
TCRs/cell
Mean (SD)
0.12
(0.09)
0.12 0.08 0.11 0.098 0.087 0.04
MSI-H
(%)
47/371
(12.6%)
21/49
(43%)
7/123
(5.6%)
9/48
(18.8%)
8/79
(10.1%)
2/22
(9%)
<0.0001
BRAF +
(%)
18/371
(4.84%)
8/50
(16%)
2/129
(1.6%)
4/51
(7.8%)
3/83
(3.6%)
1/23
(4.4%)
0.003
KRAS +
(%)
108/371
(29%)
14/50
(28%)
35/137
(25.5%)
25/51
(49%)
26/85
(31%)
8/23
(34.8%)
0.04
Total (N) 371 54
(15%)
144
(39%)
60
(16%)
90
(24%)
23
(6%)

38

Figure 2.3: Pathologic and Molecular Features of Colorectal Cancers by CMS in MECC
(n=371 with expression profiling).

0
5
10
15
20
25
0 5 10 15 20 25 30 35
FOXP3 gene expression
TILs/hpf
r=0.5
p=0.008
0
5
10
15
20
25
0 10 20 30 40 50 60 70 80
r=0.6
p=0.0003
CD3 gene expression
0
5
10
15
20
25
30
0 20 40 60 80 100
r=0.8
p<0.0001
CD8 gene expression
0
0.05
0.1
0.15
0.2
0 5 10 15 20 25 30 35
FOXP3 gene expression
TCRs/cell
r=0.6
p<0.0001
0
0.05
0.1
0.15
0.2
0.25
0.3
0 10 20 30 40 50 60 70 80
CD3 gene expression
r= 0.5
p=0.02
0
0.05
0.1
0.15
0.2
0.25
0.3
0 10 20 30 40
r=0.7
p<0.0001
CD8 gene expression
n=33 n=33
n=33

Figure 2.4: A) Pearson correlations between TILs/hpf and FOXP3, CD3, CD8 gene
expression
B) Pearson correlations between TCRs/cell and FOXP3, CD3, CD8 gene
expression

A
B

39
2.5 Discussion
CRC is driven by numerous genetic and epigenetic abnormalities. The evolution of CRC depends
not only on the progressive accumulation of these abnormalities (Fearon & Vogelstein, 1990)
(Moyret-Lalle et al., 2008) but also on the complex interactions among different cell sub-
populations within the tumor microenvironment (cancer cells, normal stromal cells, TILs and the
soluble factors they produce) (McAllister & Weinberg, 2010). Numerous studies have described
the importance of cell surface molecules which mediate intercellular adhesion and
communication between malignant and normal cells (Leth-Larsen, Lund, & Ditzel, 2010) and the
significance of signaling pathways involved in CRC progression (Baldus et al., 2010) ; (Fang &
Richardson, 2005) ; (Markowitz & Bertagnolli, 2009) ; (Saif & Chu, 2010) ; (Sancho, Batlle, &
Clevers, 2004) .

The immune response is increasingly recognized as critical to the pathogenesis and treatement of
subsets of colorectal cancer. Studies have demonstrated the importance of components of the
immune system in the prognosis of CRC including tumor-infiltrating cells such as lymphocytes,
natural killer cells and macrophages (Deschoolmeester, Baay, Lardon, Pauwels, & Peeters,
2011). Furthermore analyses of the type, density and location of tumor-infiltrating immune cells
within CRC tumor samples have revealed that, in addition to genetic mutations and
tumor/node/metastasis staging, immunological data are independent predictors of patient survival
(Galon et al., 2006) ; (Kocián et al., 2011).

Our data shows that expression profiling can define distinct pathologic and molecular features
characteristic of CRC subtypes. First, our work clearly validates the CMS classification system

40
proposed by Quinney et al. Significantly, our work compares expert histological determination
of TILs/hpf by an expert pathologist (Rozek et al., 2016) with a novel NGS-based method which
provides a quantitative molecular summary of the T-cell repertoire in CRC.

MECC data analysis shows that MSI-H tumors are characteristically CMS1 group with
statistically significant higher mean values of both TILs/hpf and TCRs/cell. As our studies
employed FFPE-derived DNA from enriched tumor areas and not microscopically isolated tumor
cells TCR measurement by ultra-sequencing technology (immunoSEQ) could not discriminate
between epithelial and stromal lymphocytes.

Other investigators have employed a variety of other methods for the quantification of TILs and
their subtypes, including the pathology-based approaches of Immunoscore (a scoring system
based on type, location, and density of T lymphocytes) (Fridman et al., 2012) and anti-CD3
antibody immunohistochemistry. However, these standard technologies for assessing TILs are
time-consuming, semi-quantitative or qualitative, and subjective, introducing assay variability.

CD3, CD8 and FOXP3 are established T cell biomarkers with prognostic significance in CRC. In
particular expression of these molecules correlate with better prognosis and survival outcome in
CRC (Lavotshkin et al., 2015) ; (Gabrielson et al., 2016) ; (Belov, Zhou, & Christopherson,
2010).
We observed a statistically significant strong correlation between expression of all three
biomarkers and T cell infiltration measured by both TILs/hpf and TCRs/cell. FOXP3 has an
important role in the development of tumor immunology. In addition, previous studies have
reported that Foxp3 expression in tumor cells has been shown to play an important role in the
prognosis of many cancers (Kashimura et al., 2012) ; (Winerdal et al., 2011). It has also been

41
shown that FOXP3 has an immune suppression effect in colorectal cancer cells and higher
FOXP3 expression is associated with improved disease-free survival and overall survival in
colorectal cancer (Sun et al., 2017). However, this result is in contrast to other studies in which
expression of FOXP3 correlates with disease progression in patients with colorectal cancer (Kim
et al., 2013). The main reason for this may be differences in sample size, experimental methods,
and statistical analysis. The biological mechanism of the FOXP3 action in colorectal cancer is
not fully elucidated. It has been shown that FOXP3 can have an anti-tumor immune effect by
inhibiting proto-oncogenes and activating the transcription of tumor suppressor genes (Triulzi et
al., 2013).
SKP2, a proto-oncogene expressed by many tumors, regulates cell division and proliferation in
the G2/M phase of the cell cycle via SKP2-p27-CDK1/CDK2 (Nakayama & Nakayama, 2006).
FOXP3 is a transcriptional repressor of SKP2 and inhibits SKP2 expression via interaction with
the promoter of SKP2 (Li et al., 2011). In many tumors, lack of FOXP3 expression leads to
overexpression of SKP2 which causes loss of inhibition of cell proliferation promoting
tumorigenesis (Pagano, 2004). Similarly, it has also been proved that higher expression of
SPARC and FOXP33 have been associated with good prognosis in stage II colorectal cancer
(Chew et al., 2011), suggesting that Foxp3 may be a prognostic indicator in cancer. Novel anti-
tumor immunotherapies could be developed to help advance the field of anti-tumor therapy.
In the field of precision medicine, the recent consensus on molecular classification of CRC is
that this system is a useful biological and computational construct which might pave the way to a
more personalized approach in the treatment of this disease, especially in the metastatic setting
although it is not used for treatment decisions and is not FDA approved yet. Patients with CMS1
subtypes CRC (mainly MSI-H) seem to be the best candidate for immunotherapy, with clinical

42
trials demonstrating very promising results that led to regulatory approval of pembrolizumab and
nivolumab (Le et al., 2015) ; (Lipson et al., 2013). However, some clinical challenges need to be
addressed in the near future in the treatment of MSI-H CRC. Firstly, we need to better
understand why some patients are primarily resistant to these drugs and the molecular
mechanisms of the development of secondary resistance. Secondly, it might be crucial to explore
the role of immunotherapy in other settings, such as in the prevention of CRC, in the conversion
therapy of potentially resectable liver metastases, in the adjuvant treatment of early-stage disease
or in the neoadjuvant treatment of locally advanced rectal cancer.
Regarding the rapidly advanced era of immunotherapy biomarkers, a few concerns need to be
taken into account. First, heterogeneity within and between tumors; in particular tumor
mutational burden (TMB) is a measurement of mutations carried by tumor cells and is a
predictive biomarker being studied to evaluate its association with response to immune-oncology
therapy (Alexandrov et al., 2013) ; (Yuan et al., 2016). Tumor cells with high TMB may have
more neoantigens, with an associated increase in cancer-fighting T cells in the tumor
microenvironment and periphery. These neoantigens can be recognized by T cells, inciting an
anti-tumor response. Second, there is a complex interaction between immunity, tumor and host
factors creating challenges for identifying predictive biomarkers for combinations.
2.6 Future Directions
Gene expression analysis will be performed using RNA-seq data in additional 265 FFPE-derived
RNA tumor samples from our MECC study. CMS classification will be performed as described
in the methods section, increasing our current cohort from n=371 to n=636 total tumor samples
with gene expression data. In addition, the larger sample size will give us insights in regards to

43
correlations between tumor infiltrating lymphocytes as measured by our pathologist (TILs/hpf)
and as quantified using immunoSEQ (TCRs/cell). Our findings will be also correlated with
survival analysis data within the context of the GWAS study. The corresponding GWAS study
will identify germline mutations, genetic variants, or SNPs that are also related to a favorable
clinical outcome. Combining genetic variables with quantitative and qualitative metrics of the
host immune response in the tumor may reveal novel molecular and cellular signatures
associated with immune-mediated, tissue-specific destruction. This work has positioned us well
for future mechanistic studies. For patients with restricted and over-represented TCRβ usage in
the tumor, a subset of existing, fresh frozen matching tumor specimens can be subjected to RNA-
seq analysis to characterize and quantify individual transcriptomes in order to identify potential
tumor associated proteins that are overexpressed and that might be recognized by infiltrating T
cells.

One of the immunological markers worth exploring is PD-L1 (Valentini et al., 2018) ; (Pardoll,
2012) ; (Song et al., 2013) ; (Droeser et al., 2013). The inhibition of PD-1/ PD-L1 interaction has
been proposed as a therapeutic target and PD-1 and PD-L1 specific monoclonal antibodies have
been successfully developed and tested in phase I clinical trials. (Brahmer et al., 2012) ;
(Topalian et al., 2012). However, there are some published studies showing contradictory results
(Droeser et al., 2013) underlining the specificities of tumor immune system interaction in CRC.
In that context, it will be interesting to test correlations between PD-1, PD-L1 gene expression
and tumor infiltrating lymphocytes in our MECC study elucidating their prognostic
significance/value in CRC in more depth.

44
Although this work is outside of the scope of the current proposal, matching these tumor
associated proteins to samples with limited TCR variable chain usage with HLA type will allow
the prediction of which peptides of these tumor associated proteins would be candidates for
presentation to T cells by these HLA molecules through widely available HLA-peptide binding
computer prediction programs. The candidate peptides can then be used in in vitro immunization
assays with T cells obtained from the CRC patients in order to identify the protein(s) that tumor
associated antigen specific T cells recognize and which recognition results in favorable clinical
outcomes. Such proteins would be used in the design of future therapeutic cancer vaccines that
would induce favorable T cell responses in CRC patients.

45
CHAPTER 3

Characterization of the Lynch Syndrome germline MSH2*c.705delA private
mutation in a Druze population

3.1 Abstract
Lynch Syndrome, originally described as hereditary non-polyposis colorectal cancer (HNPCC),
is an autosomal dominant cancer genetic syndrome resulting from mutations in the DNA
mismatch repair genes MLH1, MSH2, MSH6, PMS2 and EPCAM. It confers high risk for colon,
rectum, endometrial and other cancers, typically at an early age. A number of founder mutations
in Lynch-causing genes are helpful in identifying individuals at risk within specific populations.
The Druze are a unique religious and ethnic population that are thought to originate 11,000 years
ago in the Arabian Peninsula and are religiously discouraged from marrying outside of their
communities; therefore, endogamy rates are relatively high. In the present study we report an
MSH2*c.705delA mutation in twelve Druze individuals from northern Israel with Lynch
Syndrome. To establish whether this variant represents a founder or a private mutation we
performed haplotype analysis using SNP genotyping data from these 12 Druze carrying the
MSH2*c.705delA mutation (p.Asp236Thrfs) and 47 Druze individuals from the Human Genome
Diversity Project (HGDP) with chromosome 2 phasing. The length of the shared haplotype was
initially determined by Short Tandem Repeat (STR) analysis employing five highly polymorphic
microsatellite markers near the MSH2 gene, augmented by SNP genotyping. Haplotype phasing
showed a 13 Mb shared haplotype common to all MSH2*c.705delA mutation carriers. Attempts
to determine the age of the mutation were confounded by the high level of consanguinity among

46
these Druze individuals, emphasizing the difficulty of estimating the age of a mutation that
depends on the assumption of independent ascertainment. In summary, the present study
establishes the MSH2*c.705delA among the Druze population with Lynch Syndrome as a private
mutation and illustrates the challenges of statistical genetic analyses in highly endogamous
populations.

47
3.2 Introduction
Analysis of mutations identified in relatively high frequency individual alleles that are particular
to a specific population is an informative approach to recognizing the genetic anthropology and
clinical features of Mendelian syndromes in isolated populations. The concept of a founder effect
was originally informed by the observations of Sewall Wright (Wright, 1931), and likely first
formally named by Ernst Mayr in 1942 (Provine, 2004). The founder principle proposes that an
isolated population established by a small number of people, who carry a subset of the genetic
variability relative to the parent population, would lack overall genetic diversity after expansion
with new genetic variation occurring spontaneously or migrating in from other populations
(Templeton, 1980).

Luca Cavalli-Sforza and Walter Bodmer elaborated these principles in relationship to HLA and
other diseases in their seminal work, The Genetics of Human Populations (L. Cavalli-Sforza &
Bodmer, 1971), setting the stage for the study of specific diseases in ethnic and ancestral
populations. In certain ethnic groups, founder mutations explain a substantial fraction of Lynch
Syndrome. For example, a number of MLH1 heterozygous mutations or deletions have been
reported in United States, Finnish, Swiss and Chinese populations (Chan et al., 2001) ;
(Lagerstedt-Robinson et al., 2016) ; (von Salomé et al., 2017) ; (Tomsic et al., 2012). In addition,
our group has previously described an MSH2 1906G>C heterozygous mutation in Ashkenazi
Jews (Foulkes et al., 2002) and other founder mutations (Raskin et al., 2011). More than 170
MSH2 mutations have been identified; about 40% of Lynch Syndrome cases are attributed to this
gene (Goldberg et al., 2014). https://www.insight-database.org/

48
Another well-known example of a founder mutation in Israeli populations is the APC I1307K
allele identified in 6% of Ashkenazi Jews and 1-2% of Sephardi Jews that confers a two-fold risk
of colorectal cancer in carriers (Laken et al., 1997) ; (Poynter et al., 2006). The APC founder
mutation was traced to a Most Recent Common Ancestor (MRCA) estimated to live between
2,200-2,950 years (87.9-118 generations) ago (Niell et al., 2003) ; (Rozen et al., 2002).

The Druze are a middle-eastern group that migrated from Arabian Peninsula 500 to 1,000 years
ago and currently live mostly in Israel, Jordan, Syria and Lebanon. They represent a unique
religious and ethnic population with high endogamy rates. For nearly 1,000 years, the Druze
population has strictly adhered to its beliefs and traditions, including prohibiting marriages to
non-Druze individuals, and an inability to convert to a Druze believer. These religious beliefs,
combined with the fact that like many minority groups, most Druze communities have remained
relatively isolated from their neighboring non-Druze communities, make the Druze community
truly a genetic and social isolate. Moreover, it is traditional in the Druze community to marry
within a person’s extended family, resulting in a high rate of consanguineous marriages (Vardi-
Saliternik, Friedlander, & Cohen, 2002). Marriage outside the Druze faith is culturally
discouraged and rare (Marshall et al., 2016). In the present study we report and characterize an
MSH2 mutation c.705delA (p.Asp236Thfrs) recurrently identified in a group of Druze
individuals from North Israel with Lynch syndrome.

49
3.3 Materials and Methods
3.3.1 Population
Mutation carriers were identified from among all Druze participants in the Molecular
Epidemiology of Colorectal Cancer (MECC Study). The MECC study is a population-based
case-control study of all newly diagnosed cases of colorectal cancer in Northern Israel diagnosed
between March 1998 and June 2016, and age-sex-religion-residence matched controls without
colorectal cancer (Poynter et al., 2005) ; (Rozek et al., 2016). The study protocol included
detailed family history and epidemiologic interviews, blood collection, and retrieval of formalin-
fixed, paraffin-embedded tumor samples with assessment for Microsatellite instability (MSI-H).
Written informed consent was obtained from all study participants with IRB-approval from all
relevant institutions (University of Southern California HS-12-00324 and Carmel Medical
Center, Haifa, Israel). MSI-H cases were further studied for mutations in the Mismatch Repair
(MMR) genes (MLH1, MSH2, MSH6, PMS2).

3.3.2 MSH2 Sanger Sequencing and Mutation Analysis
Sequencing of the complete MSH2 coding region (1-16 exons) was carried out from germline
DNA from 43 Israeli Druze individuals diagnosed with Lynch Syndrome; 26 Druze individuals
without any cancer diagnosis were sequenced as controls. Total genomic DNA was isolated from
peripheral blood lymphocytes by protein salting out and nucleic acid precipitation (Miller,
Dykes, & Polesky, 1988) . The quantity and quality of the DNA samples was determined with
NanoDrop 8000 spectrophotometer (Thermo Fisher Scientific, MA, USA).

50
Exons 1-16 of the MSH2 gene (NCBI Reference Sequence: NC_000002.12) were amplified by
polymerase chain reaction (PCR). Primer pairs were designed with "Primer 3" software. The
PCR products were purified with the QIAquick PCR Purification Kit Protocol (Qiagen, USA).
Sanger sequencing of amplified fragments in forward and reverse directions was performed by
GENEWIZ (La Jolla, CA). Electropherogram results were analyzed and mutations were
identified using Mutation Surveyor 2.61 (SoftGenetics, PA).

3.3.3 Microsatellite Short Tandem Repeat (STR) analysis
Microsatellite markers (loci), also known as short tandem repeats (STRs), are highly
polymorphic DNA loci consisting of a repeated nucleotide sequence. During a microsatellite
analysis, microsatellite loci are amplified by PCR using fluorescently labeled (FAM) forward
primers and unlabeled reverse primers. The PCR amplicons are separated by size using
electrophoresis. Five STR markers flanking MSH2 were selected for genotyping in order to
provide data for informative haplotyping, based on their distance from the MSH2 deletion and
their GC content. Fragment sizes were analyzed using GeneMarker software (SoftGenetics, PA).

3.3.4 Single Nucleotide Polymorphism (SNP) Array Genotyping
All germline samples included from the Molecular Epidemiology of Colorectal Cancer Study
were genotyped on high-density SNP arrays as part of an NIH funded study, the Colorectal
Transdisciplinary Study, (CORECT) on Illumina 1M, 1M-Duo genotyping platforms using
germline DNA (Schmit et al., 2019). In brief, quality control filters were applied at both the
individual subject and single nucleotide polymorphism (SNP) levels. Genotype data passing
quality control were imputed to the 1,000 Genomes Project Phase 1 multiethnic reference panel

51
(March 2012 release, n=1,092) using SHAPE-IT/IMPUTE2. Genotype data for cases and
controls were phased and imputed together by platform to avoid differential imputation errors
between cases and controls. Stringent imputation quality (info≥0.7, certainty≥0.9,
concordance≥0.9 and minor allele frequency filters (MAF≥1%) were imposed on variants prior
to the analysis phase.

3.3.5 Haplotype Analysis
Haplotype analysis was performed using data derived from SNP genotyping array and
genotyping of Short Tandem Repeats (STRs). SNP genotyping data derived from twelve Druze
mutation carriers (Infinium HumanCore BeadChip) and Human Genome Diversity Project
(HGDP) including 1,043 individuals (47 Druze and 996 non-Druze). ShapeIT2 software
(https://mathgen.stats.ox.ac.uk/genetics_software/shapeit/shapeit.html) [16-20] was used for the
haplotype analysis of SNP genotyping data. Six highly polymorphic microsatellite markers
(D2S367, D2S2238, D2S391, D2S1248, D2S378, and D2S370) surrounding the MSH2 mutation
locus were genotyped in 12 Druze carriers, 9 Druze non-carriers and 8 Christian Arab non-
carriers using PCR amplification and fragment analysis at GENEWIZ (La Jolla, CA). Christian
Arab individuals from northern Israel were included in our analysis as additional non-Jewish,
non-Druze controls.

Fragment sizes of STRs were determined using GeneMarker software (Softgenetics, State
College, PA). Fifty-nine Druze individuals including the twelve mutation carriers and 47 Druze
from the Human Genome Diversity Project (HGDP) were used for the pairwise IBD estimation
using PLINK (https://www.cog-genomics.org/plink/1.9/ibd) (Delaneau et al., 2013) ; (Delaneau,

52
Marchini, 1000 Genomes Project Consortium, & 1000 Genomes Project Consortium, 2014) ;
(Delaneau, Marchini, & Zagury, 2011) ; (O’Connell et al., 2014) ; (Purcell et al., 2007).

3.3.6 Somatic Mutational Analysis through Paired Tumor / Normal Whole Exome
Sequencing
In an exploratory analysis given the limited tissue availability of Druze carriers, archived
formalin-fixed paraffin-embedded (FFPE) tumor blocks were retrieved if available, and
macrodissection, nucleic acid extraction, paired tumor-normal whole exome sequencing.

Tumor DNA and RNA were purified from macrodissected, unstained, recut slides of paraffin-
embedded tumors with a single-use razor blade and the samples were transferred to
nonsiliconized tubes. The areas for macrodissection were circled by one pathologist, Dr. Joel K.
Greenson (J.K.G) and the Hematoxylin and Eosin stained slide was used as a template.
Following dissection from slides, 1ml of xylene was added to remove paraffin and the DNA was
precipitated with 1ml ethanol. Following centrifugation, the supernatant was discarded and the
pellet was lyophilized. The pellet was resuspended in 150 ul Buffer PKD and 10 ul Proteinase K
following by 15 min incubation at 56°C and a 3 mins incubation on ice. The samples were then
centrifuged for 15 mins at 20,000 x g. The supernatant was transferred to a new RNAase free
tube (for RNA purification). The DNA-containing pellet was resuspended in 180 ul ATL buffer
and 40 ul proteinase and was incubated at 56°C for 1 hour and then denatured at 90°C for 2
hours (QIAamp all Prep FFPE Tissue kit, Qiagen). Both the DNA and RNA purifications were
completed using QIAcube robotic protocols according to the manufacturer’s instructions
(Qiagen). Tumor Whole Exome sequencing was performed at 100X using KAPA HyperPlus
library prep kit and IDTxGen whole exome kit (Illumina, La Jolla, Ca). Tumoe RNASeq (50M
reads) was performed using TruSeq RNA Access Library Prep Kit (Illumina, La Jolla, CA).

53

3.3.7 Pathologic Features and Molecular features of Cancers arising in MSH2 c.705delA
carriers
To further characterize the pathologic and molecular features of the tumors arising in MSH2
c.705delA carriers, uniform histopathologic review was performed by a single expert GI
pathologist Joel K. Greenson (J.K.G) (Greenson et al., 2003, 2009).

Tumor grade:
Tumors of each of the Druze mutation carriers were given a single grade of differentiation (well,
moderate or poor) (Greenson et al., 2009) ; (Jass et al., 2002). The worst grade of tumor detected
was used for the overall tumor grade, unless the worst area was a small focus (<10%) at the
advancing margin of the tumor.

Mucinous differentiation:
Tumors with greater than 50% area showing extracellular mucin were classified as mucinous.
Tumor with less than 50% area showing extracellular mucin were classified as having focal
mucinous differentiation.

Tumor necrosis:
Dirty or garland necrosis is a common feature of colorectal cancers, and is routinely scored by
the expert pathologist of the MECC study. This histologic feature reflects nuclear debris that is
observed within cystic spaces of the epithelium. Tumors were classified as negative if only a rare
focus of necrosis was present (<10%).

Histologic heterogeneity:
Tumors with at least two distinct growth patterns were classified as histologically heterogeneous.

54
Mucinous and non-mucinous areas were excluded unless there were other differences in patterns
such as tumor grade or architecture.

Signet ring cells:
Tumor cells with intracytoplasmic mucin vacuoles were designated as signet ring cells. Tumors
with signet ring cells in greater than 50% of their area were classified as signer ring cell
carcinomas. Tumors were classified as having focal signet cell differentiation, if less than 50% of
their area showed signet cell differentiation.

Growth pattern:
The advancing edge of the tumor was examined at low power to determine whether the tumor
grew with an expansile or infiltrative pattern ((Jass et al., 2002).

Prominent Crohn’s-like Host Response
The advancing edge of the tumor was assessed for the presence of a Crohn’s-like inflammatory
response. For the reaction to be considered prominent, a minimum of three lymphoid aggregates
was required per section. If the advancing edge of the tumor were not present, this field was
graded as unknown.
Tumor Infiltrating Lymphocytes (TIL) infiltration quantification by pathology
Methods and procedures for pathologic evaluation have been previously described (Rozek et al.,
2016). In brief, Tumor infiltrating lymphocytes (TILs) were identified on hematoxylin and eosin
(H&E) -stained sections as small blue mononuclear cells typically surrounded by a halo. Only
cells infiltrating between tumor cells were counted. Apoptotic cells were excluded from the
counting process. H&E stained slides were scored for histology, grade, TILs cells per HPF, and

55
Crohn’s Like Reaction (CLR), among other features. Analyses were restricted to cases with
pathologically confirmed adenocarcinoma. The tumor was scanned at low power to look for the
area with the majority of TILs. Once this area was identified, five consecutive 40× fields of an
Olympus BX40 microscope with a UPlanFl objective (Olympus America Inc., Melville, NY,
U.S.A.) were counted (total area equal to 0.94 mm
2
). The mean TIL/high power field for each
tumor was then calculated by dividing the total number of TIL by 5.

TILs infiltration measurement by ultrasequencing techniques (ImmunoSEQ)
DNA extraction from macrodissected tumor tissue was performed using the QIAamp DNA FFPE
Tissue Kit as previously described. CDR3β regions were amplified and sequenced by Adaptive
Biotechnologies Corp (Seattle, WA) using the ImmunoSEQ assay. In brief, a multiplex PCR
system was used to amplify CDR3β sequences from DNA samples. The ImmunoSEQ assay can
amplify all 49 V segments in 32 gene segment families, 8 pseudogenes segments in 7 gene
families, 10 orphan segments in 10 gene families, both D genes and the 13 functional J segments.
This approach generates an 87 base-pair fragment capable of identifying the VDJ region
spanning each unique CDR3β. The primers were designed to amplify the shortest feasible
fragment containing the CDR3 region, to ensure that degraded DNA from an FFPE sample is
still able to harbor an amplifiable fragment. Amplicons were sequenced using the Illumina HiSeq
platform. Using a baseline developed from a suite of synthetic templates, primer concentrations
and computational corrections were used to correct for the primer bias common to multiplex
PCR reactions. Raw sequence data was filtered based on the TCRβ V, D and J gene definitions
provided by the IMGT database (www.imgt.org) and binned using a modified nearest-neighbor
algorithm to merging closely related sequences and remove both PCR and sequencing errors.

56
Sequencing results were normalized based on an assay targeting housekeeping genes with the
same length amplicons, so that TIL fraction in a sample can be measured independent of the
extent of degradation. This approach enables the detection of 1 in 200,000 cells (Robins et al.,
2009). One of the resulting quantitative metrics from this assay includes, TCRs/cell, a measure
of rearranged T cell quantity relative to all nucleated cells in a tumor sample. We were able to
get a TCRs/cell value for 9 out of 12 Druze mutation carriers for whom tumor DNA was
available.
Microsatellite Instability (MSI) testing
Areas for macrodissection were selected by one pathologist (J.K.G) and circled on hematoxylin
and eosin-stained slide for a microdissection template. DNA was extracted as previously
described (QIAamp all Prep FFPE Tissue kit, Qiagen). Paired normal and tumor samples were
evaluated using a consensus panel of ten fluorescently labelled microsatellite markers (BAT25-
FAM, BAT26-FAM, BAT40-FAM, TGF-b-RII-FAM, D18S58-FAM, D17S250-FAM,
D10S197-FAM, D2S123-FAM, D5S346-HEX, Beta-Catenin-HEX).Consistent with consensus
guidelines (Boland et al., 1998), the threshold for MSI-high (MSI-H) is >= 30% of markers
demonstrating instability.

3.4 Results
3.4.1 Germline genomic sequencing of the MSH2 exome identifies a private mutation,
c.705delA
Among the tested cases with Lynch Syndrome, 12 were heterozygous for c.705delA of the 4th
exon of MSH2 (chr 2: 47,639,611) (Figure 3.1).

57

Figure 3.1: A selected region of exon 4 is shown from Sanger sequencing of germline DNA
derived from lymphocytes, with a) MSH2 wild type sequence, and b) MSH2*c.705delA.

This deletion was identified in none of the 26 Druze healthy controls tested. The clinical
characteristic of the 12 Druze MSH2*c.705delA Carriers are summarized below (Table 3.1).

Table 3.1: Clinical Characteristics of the Druze MSH2*c.705delA Carriers

58
3.4.2 Microsatellite marker genotyping in Druze and Christian Arabs
The STR analysis results across all Druze carriers tested, showing the common haplotype shared
across the five microsatellite markers used, are shown in table 3.2. These results are organized
by position relative to the MSH2 mutation locus. Fragment sizes are shown as called by
GeneMarker software (SoftGenetics, PA).
Table 3.2: Short Tandem Repeat (STR) analysis of 5 markers flanking MSH2 c.705delA in
Druze carriers

3.4.3 Detection of Identity By Descent (IBD) relationships
To improve haplotype phasing and chromosome modeling, 1043 samples (including 47 Druze
individuals) from the Human Genome Diversity Project (HGDP) (L. L. Cavalli-Sforza, 2005) ;
(J. Z. Li et al., 2008) were merged with the 12 Druze carriers. The standard pairwise Identity By
Descent (IBD) estimation provided by PLINK resulted in 1711 possible unique pairings of the 59
Druze in individuals. Several close relationships including 2 parent-child pairs, 1 set of full
siblings and 5 sets of 2nd degree relationships were identified among mutation carriers (Figure

59
3.2). The relationships among the Druze carriers detected by IBD were consistent with the
pedigrees information.

Figure 3.2: IBD comparison of all pairs of Druze in the combined data (47 HGDP Druze and 12
Druze carriers). PC= Parent-Child, FS=Full Sibling, 2
0
=second degree, 3
0
=third degree

3.4.4 Haplotype Analysis
Haplotype phasing using ShapeIT2 based on the HGDP + Lynch data (1,043+12=1,055
samples) identified a 13 Mb haplotype common to all MSH2*c.705delA mutation carriers
(Figure 3.3). The figure shows the resulting 24 haplotypes from the 12 Druze mutation carriers.
This common haplotype region was not found in Druze non-carriers nor was it found in 47 Druze
individuals from the HGDP.

60
Other Chr. Chr. w/ Deletion
13782
19245
14005
17795
14479
15577
14147
15002
15387
17939
18515
18710
13782
19245
14005
17795
14479
15577
14147
15002
15387
17939
18515
18710
35.9mb 45.3mb 58.2mb 71.2mb
705delA

Figure 3.3: A 13 Mb haplotype common to all MSH2*c.705delA mutation carriers. The 24
copies of chromosome 2 from the 12 Druze carriers are shown. The shading is relative to the
deletion chromosome of the sample with ID 13782, which is used as a reference: white
corresponds to SNPs where the chromosomes match and black where they do not. The 705delA
is marked by the solid line. The largest common region that bounds the deletion is marked by
dashed lines (58.2mb-45.3mb=13mb).

61
3.4.5 Molecular Features of Colorectal Cancer arising in Druze MSH2*c.705delA carriers
Table 3.3: Molecular Features of Colorectal Cancer arising in Druze MSH2*c.705delA carriers

3.4.6 Pathologic features of Colorectal Cancers Arising in Druze MSH2*705delA carriers
Table 3.4: Pathologic features of Colorectal Cancers Arising in Druze MSH2*705delA carriers

62
3.5 Discussion

Founder effects arise when a small group of individuals becomes genetically isolated from a
larger population. The new population will resemble only the individuals that founded the
smaller, distinct population. The founder effect is a form of genetic drift due to the randomness
that accompanies selecting a small group from a larger population. The smaller the population,
the higher the chance that the small population does not represent the larger population. If the
few organisms that migrate or get separated from the parent population do not carry the same
frequency of alleles as the main population, the resulting founder effect will cause the population
that separated to become genetically distinct from the original population reducing allelic
variation. This may lead to a new subspecies of organisms, or even entirely new species given
enough time.

In the present study, a bona fide germline truncating mutation in the MSH2 gene detected in 12
Druze individuals with Lynch Syndrome from Northern Israel was characterized. This is one of
the few mutations to have been described in a colon cancer susceptibility gene in Druze
individuals. Given the unique social structure, their history, religious beliefs, and marriage
patterns of the Druze population, this population behaves as a relative genetic isolate. Therefore,
it is possible that this germline mutation in MSH2 may be a recognizable instance of a founder
mutation. However, it is exceptionally difficult to distinguish a private mutation from a founder
mutation, especially when populations are small and many carriers are shown to be related to one
another. The length of the shared 13 Mb haplotype common to all MSH2*c. 705delA mutation
carriers, was initially determined by Short Tandem Repeat (STR) analysis employing five highly
polymorphic microsatellite markers near the MSH2 gene, and further refined by SNP genotyping.

63
Different methods have been proposed to estimate the age of rare mutations, (Rannala &
Bertorelle, 2001) ; (Slatkin & Rannala, 2000) which are based either on allele frequencies

(Kimura & Ohta, 1973)

or on intra-allelic variability and the pattern of linkage disequilibrium at
closely linked marker loci

(Rannala & Reeve, 2001) ; (Slatkin & Rannala, 1997)

with extensions
to the analysis of multilocus data.(Guo & Xiong, 1997) ; (McPeek & Strahs, 1999) ; (Reeve &
Rannala, 2002). However, these latter approaches are dedicated more to the fine mapping of the
mutation, and may not be appropriate for estimating the age of rare mutant alleles that are only
found in very few affected individuals.

In our study, attempts to determine the age of the MSH2* 705delA mutation were complicated
by the high level of consanguinity among Druze and violation of the assumption of independent
ascertainment. Estimating the age of the Most Recent Common Ancestor (MRCA) harboring a
private mutation is challenging for relatively rare events in an endogamous population. In
practice, the identification of a repeated, mutation in a meaningful proportion of the Druze who
participated in this analysis suggests that Druze patients diagnosed with Lynch Syndrome may
benefit from genetic testing for this point mutation. Carrier identification has the potential to
efficiently screen individuals at increased risk of Lynch Syndrome, especially given the
interventions which clearly reduce the burden of disease in a cost-effective manner (Dinh et al.,
2011), while minimizing unnecessary screening in individuals at average population-risk.

Regarding the molecular and pathologic features of the tumors in the Druze MSH2*705delA
carriers tested, the cancers were representative of the Lynch-associated cancers reported
previously in the literature. The presence of increased numbers of TIL cells as a reliable marker
of MSI-H in colorectal cancers has been noted in many studies (Alexander et al., 2001) ;
(Dolcetti et al., 1999) ; (Gruber, 2006) ; (Ionov et al., 1993) ; (Jass, 2000) ; (Peltomäki et al.,

64
1993) ; (Smyrk et al., 2001).
.
Although these studies have used different methods to identify and
count TIL cells, the overall results have been consistent. The majority of TIL cells have been
shown to be activated T lymphocytes (Dolcetti et al., 1999) ; (Gruber, 2006).

The results of our study suggest that long regions (13Mb) of identity by descent (IBD) sharing in
the Druze population include at least one instance of a disease-causing mutation that is likely a
private mutation. More Druze MSH2*705delA carriers with Lynch Syndrome are needed for
further analysis. These data permit even greater specificity for understanding the distribution of
Lynch Syndrome in global populations and the clinical, pathologic and molecular features of
Lynch Syndrome attributable to MSH2*705delA in Druze.

65
Chapter 4
De novo Palindromic Adenine Thymidine Rich Repeat (PATRR) – Mediated
Constitutional Balanced t(3;8) Translocation Associated with Clear Cell Renal
Cell Carcinoma

4.1 Abstract
The specific class of chromosomal translocations, designated Palindromic Adenine and
Thymine Rich Repeat (PATRR)-mediated chromosomal translocations, entails exchange across
chromosomes at sites enriched in palindromic repeats of the nucleotides adenine (A) and
thymine (T). Although a variety of PATRR mediated translocation events have been documented
in disease including cancer, the associated pathobiology remains incompletely described. In this
study we addressed the parental origin underlying the constitutional de novo PATRR-mediated
t(3;8) chromosomal translocation and the pathobiologic consequences of this genetic event
which underlie its association with clear cell renal cell carcinoma (ccRCC). We pursued a
mechanistic interrogation of the pathobiology of a single patient diagnosed with ccRCC (8
primary renal tumors) found to harbor a germline de novo PATRR-mediated balanced
translocation involving chromosomes 3 and 8 [t(3;8)] as validated by spectral karyotyping
(SKY). By performing translocation specific PCR and DNA sequencing it was determined that
the chromosome 3 breakpoint is located in an AT rich palindromic sequence in the third intron of
the FHIT gene (chr3p14.2); and the chromosome 8 breakpoint lies in an AT-rich palindromic
sequence in the first intron of the RNF139 (a.k.a TRC8) gene (chr8q24.1). A hypothesized causal

66
association between the t(3;8) translocation and the development of ccRCC is suggested not only
by the clinical presentation of the individual in our study, but also by previous reports in the
literature in which ccRCC segregates with the t(3;8) balanced translocation (Glover et al., 1988) ;
(Kato et al., 2014) ; (Valle et al., 2005). The observations and experimental systems established
here offer further insights into pathobiological sequelae of this t(3:8) constitutional
rearrangement. It was determined that this de novo PATRR-mediated t(3;8) balanced
translocation arose during spermatogenesis (paternally derived). In addition, we characterized the
pathobiology of this constitutional phenomenon by employing advanced next generation
sequencing (NGS) technology and detailed characterization of the clonal host immune responses
within the tumors.

4.2 Introduction
4.2.1 Palindrome-Mediated Translocations in Humans
PATRR-mediated t(3;8) is an example of reciprocal chromosomal translocations that involve the
rearrangement of genetic material across nonhomologous chromosomes (Nambiar et al., 2008) ;
(Kato et al., 2014). Reciprocal translocations are the most frequent type of chromosomal
abnormalities, occurring in about 1 in 500 live births (Hook et al., 1977). Chromosomal
translocations may either be balanced, entailing a reciprocal exchange without an overall loss of
genetic material, or non-balanced with a nonreciprocal event leading to a net loss of genetic
material (Rabbitts et al., 1994). Balanced reciprocal non-Robertsonian translocations detected by
conventional karyotyping are found in about 1:1,000 newborns (Gardner and Sutherland, 2003).
In most of these cases one of the parents carries the same translocation. In contrast, de novo

67
balanced reciprocal non-Robertsonian translocations are rare, but their investigation might help
to improve our understanding of the parental origin and formation of constitutional chromosomal
rearrangements. Often, individuals with a balanced translocation are asymptomatic with no
discernible aberrant phenotype; however, when a balanced translocation disrupts a
physiologically crucial gene a pathologic phenotype may ensue including a predispotition to
cancer (Nambiar et al., 2008) ; (Rabbitts et al., 1994).
Previous research has identified one specific class of chromosomal translocations designated
Palindromic Adenine and Thymine Rich Repeat (PATRR)-mediated chromosomal
translocations; and, as the name suggests, these translocations involve exchange across
chromosomes at sites enriched in palindromic repeats of the nucleotides adenine (A) and
thymine (T) (Inagaki et al., 2016) ; (Kurahashi et al., 2006a, 2006b). Palindromic DNA
sequences are distributed across the human genome, can form secondary structures and are a
recognized mechanism responsible for examples of genomic instability.
DNA palindromes consist of two units of identical sequences which are connected in an inverted
position with respect to each other. In palindromes, the sequences on the complementary strands
read the same in either direction. The complementary sequence appears in the same strand in an
inverted orientation. Palindromic DNA can consequently form specific tertiary structures,
namely, single-stranded “hairpin” or double-stranded “cruciform” DNA. Such unusual DNA
tertiary structures are called non-B DNA structures (Wang & Vasquez, 2014). These non-B
DNA structures are generated in a cell under specific situations, although their in vivo existence
is still a controversial subject (Kurahashi et al., 2006) ; (Kato et al., 2012).

68
Hairpin structures can be formed when the double helix DNA is dissociated into single-stranded
DNA molecules at the palindrome. Such single-stranded DNA might occur during DNA or RNA
synthesis during replication or transcription. Cruciform formation starts from unwinding of the
center of the double-stranded palindromic DNA, followed by extrusion at the center of the
palindrome to form an intra-strand base-paring of each strand. As the DNA unwinds, the
cruciform increases in size. Cruciform formation requires an under-twisted state, that is, negative
superhelicity, of the DNA. Such unusual DNA structure itself could have an impact on DNA
replication, repair, transcription, and other important biological pathways (Inagaki et al., 2013).
The DNA regions that potentially form non-B DNA structures are often characterized by
genomic instability that induces gross chromosomal rearrangements (Pearson, Nichol Edamura,
& Cleary, 2005) ; (Tanaka et al., 2005) ; (Maizels, 2006) ; (Raghavan & Lieber, 2006) ; (Mirkin
et al., 2007) ; (Tanaka et al., 2010).
The best studied example of palindrome-induced genomic instability in humans is the
constitutional balanced translocation between chromosomes 11 and 22. The breakpoints of
numerous unrelated t(11;22) cases have been consistently shown to be located within
palindromic AT-rich repeats (PATRRs) on 11q23 and 22q11 (PATRR11 and PATRR22)
(Koduru & Chaganti, 1989; Shaikh, Budarf, Celle, Zackai, & Emanuel, 1999; Tapia-Páez et al.,
2001). The majority of the breakpoints have been localized at the center of the PATRRs,
suggesting that the center of the palindrome is susceptible to Double Strand Breaks (DSBs)
inducing chromosomal rearrangement. The t(11;22)(q23;q11) is the most common recurrent non-
Robertsonian translocation in humans (Kurahashi et al., 2010). Balanced carriers of the t(11;22)
are at high risk of having offspring with the derivative 22 syndrome, also known as Emanuel
Syndrome, due to 3:1 meiotic non-disjunction event (Zackai & Emanuel, 1980). PATRR22 has

69
been known to be a hotspot for translocation breakpoints (Kurahashi et al., 2010) ; (Kurahashi et
al., 2006b). Findings of PATRR-like sequences at the translocation breakpoints of other
chromosome 22 partner chromosomes supports the conclusion that palindrome-mediated
chromosomal translocation appears to be one of the universal pathways for human genomic
rearrangements (Nimmakayalu et al., 2003) ; (Gotter et al., 2004).

4.2.2 Factors that influence PATRR-mediated translocations
The spatial organization of chromosomes is non-random, and is a contributing factor in the
formation of specific somatic translocations (Misteli, 2004) ; (Roix et al., 2003) ; (Parada et al.,
2004). The spatial proximity of chromosome 11 and 22 and their corresponding breakpoint
regions in meiosis play a role in generating the translocation as shown by FISH analysis during
male and female meiosis. The proximity between 11q23 and 22q11 is closer than that of a
control region and 22q11, suggesting that spatial proximity during meiosis might play a role in
the generation of recurrent translocations (Ashley et al., 2006). Thus, using various
methodologies and approaches, attempts have been made to clarify the mechanism behind
PATRR-mediated translocations. Nonetheless, despite numerous speculations and experiments,
the mechanism of PATRR-mediated translocations remains elusive. Therefore, additional studies
will be required to determine the enzymatic pathway(s) and the timing involved in formation of
these translocations in gametogenesis.

Despite the obvious AT richness at the t(11;22) breakpoint, no significant homology is apparent
between PATRR11 and PATRR22, suggesting that t(11;22) events result from double-strand
break repair occurring by a non-homologous end joining mechanism (Kato et al., 2008). In
contrast to translocation formation, deletions within the PATRRs appear to utilize more

70
extensive microhomology (Kato et al., 2008). Moreover, deletions, but not translocations, are
induced by the inhibition or suppression of DNA replication (Kurahashi et al., 2009). These
results imply that the mechanism for the generation of a palindrome-mediated rearrangement
differs between deletions and translocations.

4.2.3 Parental origin of de novo Balanced Translocations
Structural abnormalities in humans often demonstrate a gender bias with respect to parental
origin (Hassold & Hunt, 2001). Chromosomal translocations mainly occur in the paternal
germline (Buwe, Guttenbach, & Schmid, 2005). One would expect that if translocations take
place by a replication error, a positive correlation between paternal age and de novo
translocations would be expected (Crow, 2000). Nevertheless, the de novo t(11;22) frequency
does not appear to increase with increasing paternal age (Kato et al., 2007). This data suggests
that translocations do not occur via a pre-meiotic replication-dependent mechanism.

So far, the knowledge of the parental origin and mechanisms of formation of de novo reciprocal
translocations in humans is limited. Due to mutagenesis studies in Drosophila and mice, it has
been postulated that structural chromosomal aberrations arise more often in spermatogenesis,
whereas numerical chromosomal aberrations are more often of maternal origin (Chandley, 1991).
In humans, the former has been documented for the more frequent microdeletions and the latter
for the common trisomies 13, 18, and 21 (Thomas et al., 2006). Robertsonian translocations are
the most frequent balanced translocations in humans. Most de novo Robertsonian translocations
arise during oogenesis and have breakpoints within a consistent region (Bandyopadhyay et al.,
2002).

71
De Gregori et al., 2007 found paternal origin in 5 de novo reciprocal and 11 de novo complex
chromosomal rearrangements, all associated with an abnormal phenotype (De Gregori et al.,
2007). By linkage analysis subsequent to flow sorting of the derivative chromosomes Thomas et
al. [2010] investigated 27 patients with de novo reciprocal translocations (Thomas et al., 2010).
In only one of them the origin was maternal. Twenty-one of them were associated with an
abnormal phenotype. Furthermore, Grossmann et al. [2010] found exclusively paternal origin in
5 de novo balanced complex chromosomal rearrangements (Grossmann et al., 2010).
De novo formation in paternal meiosis of the PATRR-mediated t(11;22)(q23;q11) translocation,
and may be relevant to other examples of PATRR-mediated translocations, was deduced from
analysis of few cases and also particularly from detection in sperm of men who were not carriers
of this translocation. A mechanism of non-homologous end joining of palindromic AT-rich
repeats on 11q23 and 22q11 was assumed (Ohye et al., 2010). Segregation of this translocation
has been reported in several hundred families, but a de novo translocation event has been
identified in only 8 cases and every time the translocation originated in paternal germ-line
chromosomes. In addition, de novo t(11;22) rearrangements have been detected in the sperm of
healthy individuals, leading to the hypothesis that it occurs long the meiosis-spermatogenesis
pathways. However, a recent report indicates that the translocation occurred post-fertilization,
providing the first evidence of a de novo t(11;22)(q23;q11.2) occurring in a maternal mitotic
environment (Correll-Tash et al., 2018).

In general, the occurrence of chromosomal rearrangements is linked to complex features of the
genomic architecture, with predisposing DNA structures such as palindromic AT-rich repeats,
pericentromeric repeats, low-copy repeats, and sometimes repetitive sequences. These DNA
structures predispose to non-allelic homologous recombination, non-homologous end joining,

72
and fork stalling and template switching, which are considered to be the major mechanisms of
formation of recurrent and non- recurrent structural rearrangements, respectively. Although all
these mechanisms are considered to occur with equal frequency in male and female meiosis,
postnatally observed frequencies are different. Thomas et al. [2010] suggested that this might be
the consequence of the higher number of premeiotic cell divisions in older men (Thomas et al.,
2010). Confirmation would be possible by linkage with grandparental haplotypes. If formed
premeiotically, the translocation chromosomes should be a mix of grandpaternal and
grandmaternal chromosomes in at least some cases. In summary, it has been shown that (a) the
majority of de novo cytogenetically balanced reciprocal translocations are of paternal origin, and
that (b) the preponderance and the paternal age effect do not appear to be as obvious as
previously thought.

4.2.4 Molecular Genetics of Renal Cell Carcinoma (RCC)

Renal cell carcinoma (RCC) consists of a highly heterogeneous group of malignant neoplasms
that have characteristic histopathologic features, cytogenetic abnormalities, biologic behavior,
and characteristic imaging findings constituted by a number of different types of cancer.
Sporadic, nonfamilial kidney cancer includes clear cell kidney cancer (75%), type 1 papillary
kidney cancer (10%), papillary type 2 kidney cancer (including collecting duct and medullary
RCC) (5%), the microphalmia-associated transcription (MiT) family translocation kidney
cancers (TFE3, TFEB, and MITF), chromophobe kidney cancer (5%), and oncocytoma (5%).
Each has a distinct histology, a different clinical course, responds differently to therapy, and is
caused by mutation in a different gene.

73
Familial RCC is far less common characterized by an early age of diagnosis and it is often
bilateral or with multiple synchronous tumors (Franksson et al., 1972) ; (Goldman et al., 1979).
Von Hippel-Lindau (VHL) Syndrome is the most common familial RCC syndrome characterized
by autosomal dominant inheritance and by alterations in the VHL gene which is located at
chromosome 3p25 (Gnarra et al., 1994) ; (Clifford et al., 1998) ; (Woodward et al., 2008) ;
(Kondo et al., 2002). VHL is a E3 ubiquitin protein ligase involved in the ubiquitination and
subsequent proteasomal degradation via the von Hippel-Lindau ubiquitination complex. It acts as
target recruitment subunit in the E3 ubiquitin ligase complex and recruits hydroxylated hypoxia-
inducible factor (HIF) under normoxic conditions which is degraded through ubiquitin-mediated
pathway. A mutated VHL stabilizes HIF and leads to the up-regulation of many pro-angiogenic
factors including GLUT-I, VEGF, PDGF-β, erythropoietin and TGF-α (Kanno et al., 2014).

Genomic studies identifying the genes for familial kidney cancer, including the MET, FLCN, FH
(Ahvenainen et al., 2008), SDH family of genes (Raymond et al., 2012), TSC1, TSC2,
and TFE3 genes, have significantly altered the ways in which patients with kidney cancer are
managed. While a number of FDA-approved agents that target the VHL pathway have been
approved for the treatment of patients with advanced kidney cancer, further genomic studies,
such as whole exome sequencing, gene expression patterns, and gene copy number, will be
required to gain a complete understanding of the genetic basis of kidney cancer, of the kidney
cancer gene pathways and, most importantly, to provide the foundation for the development of
effective forms of therapy for patients with this disease (Linehan, 2012).

74
4.2.5 Case Presentation
4.2.5.1 History of Present Illness
In September of 2011, a 32 year old man (proband) with a history of Type II diabetes managed
with metformin, presented for an interval evaluation which included an abdominal ultrasound.
Multiple bilateral kidney lesion (total of 7) were identified, warranting further workup. In
October 2011, CT-guided needle biopsy revealed bilateral ccRCC. In November of the same
year and January of 2012, the patient underwent staged nephron-sparing, minimally invasive
robotic nephrectomies. Four tumors were resected from the right and three additional tumors
from the left. In May 2018, he developed another tumor and underwent right redo partial
nephrectomy. From the surgical tissue cell lines were established from normal and tumor tissue
renal parenchyma (see methods section).

4.2.5.2 Clinical Workup
The initial suspected clinical diagnosis was Von-Hippel Lindau (VHL) Syndrome. However, a
formal clinical cancer genetics workup did not reveal any of the VHL syndrome associated
phenotypic characteristics such as retinal angiomas, cerebellar or spinal hemangioblastomas,
acoustic endolymphatic sac tumors or epidydymal cysts. In addition, genetic testing did not
reveal any germline VHL pathogenic variants. Pedigree analysis revealed that the only history of
cancer in the family was a maternal grandmother who was diagnosed with breast cancer at the
age of 70, and a history of two previous miscarriages. The products of conception (POCs) had
been previously analyzed by conventional karyotyping and were found to carry a balanced t(3;8)
translocation. The parents were then screened and our patient found to be the carrier of this
constitutional translocation. To establish whether this represented a de novo genetic event
occurring during meiosis, the proband’s parents and brother were evaluated as part of the present

75
study (Figure 4.1). Karyotyping analysis of family members confirmed that the t(3;8)
translocation was de novo.

4.3 Materials and Methods
4.3.1 Lymphocytes isolation

For the lymphocyte isolation, the following reagents were used: lymphocyte separation medium
(LSM) (ICN Biomedicals Catalog #50494), PBS, cell lysis solution (Puregene-Gentra Systems
Catalog #D-50K2), protein precipitation solution (Puregene-Gentra Systems Catalog #D-50K3),
isopropanol (ice cold) and 70% ethanol. In brief, 4ml of LSM was pipetted into a 15ml conical
tube and 4ml of PBS to another conical tube. 4ml of the proband’s blood was added to the PBS
and we mixed using the pipette. The blood/PBS mixture was carefully layered into the LSM and
was centrifuged for 30min at 400 g (1350 rpm, GH 3.8 rotor). The plasma layer was removed to
within a few mm of the lymphocyte layer with a pasteur pipette. The lymphocyte layer was
pipetted off and transferred to a new 15ml conical tube. Equal volume of PBS was added, we
Figure 4.1: Pedigree of the proband with the de
novo constitutional PATRR mediated t(3;8)
balanced translocation. Arrow shows the proband
diagnosed with ccRCC at the age of 32, and a history
of two miscarriages which initially led to a genetic
work-up. His parents and brother are healthy and none
harbor the t(3;8) translocation. In addition, there is no
history of kidney cancer in the family.

76
mixed by pipetting and then centrifuged at 200 g (1000 rpm, GH 3.8 rotor) for 10 min, high
brake. This process was repeated twice. Finally, the supernatant was removed using a pasteur
pipette leaving behind the lymphocyte pellet.

4.3.2 Spectral Karyotyping (SKY) technique

The clinical finding of the conventional karyotyping prompted us to do a more in depth analysis
and study the pathological consequences of this phenomenon. The reported chromosomal
abnormality was validated performing Spectral Karyotyping (SKY) technique using the
proband’s isolated lymphocytes in conjunction with the City of Hope cytogenetic core facility.
SKY is a multicolor fluorescence in-situ hybridization (FISH) technique used to detect
metaphase chromosomes with spectral microscope. The SKY protocol consisted of five steps:
cell pretreatment and metaphase preparation, slide pretreatment (pepsin digestion), chromosome
and probe denaturation and hybridization, fluorescent probe detection and image acquisition and
analysis.

4.3.2.1 Cell Pretreatment and metaphase preparation
In brief, the lymphocytes were resuspended in Dulbecco's Modification of Eagle's Medium
(DMEM) containing 10-15% fetal bovine serum (FBS) and 1% penicillin/streptomycin and were
treated with colcemid solution at 0.05 µg/ml for 30-60 min. The medium containing any floating
cells was collected in 50ml sterile falcon centrifuge tubes. The cells were rinsed with sterile 1X
PBS. Rinse the cells on the plate with sterile1X-PBS. After incubation with sterile trypsin for 1-2
min, the remaining cells were harvested and collected into the same tube. The tubes were spined
at 1000 rpm for 5 mins, we aspirated the supernatant leaving 0.5 ml and loosened the pellet by
flicking with finger only. 10ml of hypotonic solution of 0.56% KCl was added in dH
2
O and the

77
suspension was incubated at 37°C for 30-45 min. One drop of methanol/acetic acid (3:1 vol/vol)
per ml of hypotonic cell suspension was added and the tube was gently inverted for mixing.

We centrifuged at 1200 rpm for 5 min and collected the pellet. 5 ml of fresh methanol/acetic acid
(3:1 vol/vol) fixative solution was added dropwise while flicking the pellet continuously. We
centrifuged again at 1200 rpm for 5 min and added 10 ml of ice-cold methanol/acetic acid
fixative along the wall of the tube. At this stage, if needed, the cells can be stored in fixative
solution in a tightened and sealed tube at -20°C for short term or -80°C for long term use. Slides
were cleaned in absolute ethanol, and dipped in dH
2
O for approximately10X in order to form a
sheath of water on the surface of the slide. The slide was placed on a glass plate and dropped 15-
20µl of cell suspension from 10" above the slide. The slide was placed in a water bath at 65-70°C
for 1-2 min and allowed to dry. The slide was then checked under a light microscope using 10X
and 40X dry objectives making sure there is metaphase chromosomes and the spreads are evenly
spaced. We checked for cytoplasm presence surrounding the chromosomes. Because cytoplasm
was present we proceeded with slide pretreatment (pepsin digestion).

4.3.2.2 Slide pretreatment (Pepsin digestion)

We applied 120 µl of 1:200 RNase solution (20 mg/ml) dissolved in 2X salt, sodium citrate
(SSC) onto a 24mm x 60 mm microscope coverglass and inverted the metaphase slide face down
on the coverglass. Then the metaphase slide was gently inverted face up and incubated at 37°C
for 45 min. The coverglass was carefully removed without scratching the slide and washed in
2X-SSC buffer in a coplin jar every 5 min for 15 min with shaking. 5-15 µl of pepsin stock
solution (100 mg/ml in dH
2
O) was added into a clean beaker and then added 100 ml of
prewarmed (37°C) 0.01M HCl. The slide was incubated in a coplin jar containing the HCl/persin

78
solution at 37°C for 3-5 min. This was a critical step as too much digestion would cause the
chromosomes to be overdigested and too little digestion would leave the cytoplasm undigested
which might have led to non-specific binding of the probe and interfere with the hybridization
signal.

The slide was then washed in a coplin jar containing 100ml 1X-PBS for 5 min twice at room
temperature (RT). The slide was placed in a coplin jar containing 100ml 1% formaldehyde for 10
min at RT and then the slide was washed again with 1X-PBS for 5min. The slide was then
observed under a light microscope to make sure that the slides are properly digested without any
cytoplasm present and the chromosome morphology is maintained.
4.3.2.3 Chromosome and probe denaturation and hybridization

Fresh denaturing solution (70% formamide/2X SSC, pH 7.0) was prepared and prewarmed to
80°C. The slide was placed in a coplin jar containing the denaturing solution in a water bath at
80°C for 1 min. The slide was then immediately placed in ice-cold 70% ethanol for 3 min
followed by 80% and 100% ethanol for 3 min each and air dry. The SkyPaint probe was warmed
at 37°C with shaking for 20 min, followed by vortex and brief centrifuge at 1000 rpm for a few
seconds. The probe was denatured in a thermocycler for a two-step cycle at 85°C for 5 min cycle
followed by 37°C for 60 min to allow labeled-probe DNA for preannealing. 10 µl of the
denatured probe was applied onto the area of hybridization and covered with a 22mm x 40 mm
coverglass avoiding air bubbles. The edges of the coverglass were sealed with rubber cement and
was incubated in a humidified chamber 37°C for 48-72 hrs.

79
4.3.2.4 Fluorescent Probe Detection
The coverglass was removed carefully and the slide was placed in a coplin jar containing
prewarmed (45°C) washing solution I (freshly prepared 50% formamide in 2X SSC). We washed
three times for 5 min at 45°C in a shaking water bath at 45 rpm. The slide was washed using
washing solution II (1X SSC) at 45°C for 5 min twice with shaking followed by washing
solution III (4X SSC/0.1% Tween 20) for 5 min at 45°C with shaking. 80 µl of blocking reagent
was applied, coverslip was placed and an incubation at 37°C for 30 min was followed. The slide
was removed and the fluid was allowed to drain. 80 80 µl of Cy5 staining reagent was added, a
coverglass was applied and an incubation at 37°C for 40 min was followed. The slide was
washed with washing solution III at 45°C for 5 min three times with shaking. 80 µl of Cy5.5
staining reagent was added, a coverglass was placed and an incubation at 37°C for 40 min was
followed. The slide was finally washed with washing solution III at 45°C for 5 min three times
with shaking. Finally, the slide was tilted and allowed the fluid to drain. 20 µl of the anti-fade
DAPI reagent was applied and a 24 mm x 60 mm microscope coverglass was placed.
4.3.2.5 Image acquisition and analysis

Metaphase slides were viewed using an Olympus microscope equipped with a 60X oil
immersion lens, a Spectral cube, a DAPI filter and a sagnac interferometer module with a CCD
camera. Spectral Karyotypes were performed using SKY View software (Applied spectral
imaging Version 1.62) following the user's manual. Upon analyzing the images, the
chromosomes were viewed as color images (with specific fluorescent colors), pseudo color
images (with colors for classification) and inverted DAPI images.

80
4.3.3 Breakpoint Identification
In collaboration with Dr. Beverly Emanuel’s group at Children’s Hospital of Philadelphia, by
performing PCR we were able to confirm the t(3;8) translocation breakpoints on chromosomes 3
and 8 previously published by the Emanuel group (Kato et al., 2014) (see results section for more
details).

4.3.4 cDNA synthesis and Fusion Transcript Expression
For the cDNA synthesis from proband’s RNA the following protocol was followed (NEB
#M0277). The RNA sample with primer d(T)
23
VN were mixed in a sterile RNase-free microfuge
tube. The sample RNA/primer was denatured for 5 min at 65°C. A brief spin was followed and
we put promptly on ice. Afterwards, the following components were added: 2 µl 10X AMV
buffer, 0.2 to 2 µl AMV RT (10 U/µl), 0.2 µl RNase inhibitor (40 U/µl) and nuclease-free water
to a total volume of 20 µl. The 20 µl cDNA synthesis reaction was incubated at 42
o
C for one
hour. The enzyme was inactivated at 80
o
C for 5 min. The cDNA product was stored at -20°C
until further use. Table 4.1 shows the primer design used spanning the predicted fusion
transcript.

Table 4.1: Primer design spanning the predicted fusion transcript

Primer Design Spanning the Predicted Fusion Transcript
Forward primer, TRC8, exon 1 CGACGCCATCTTCAACTCCT
Reverse primer, FHIT, exon 4 CCACTGAGGACTCCGAA
Expected size of the fusion product 150 bp

81
4.3.5 Isolation of Tumor and Germline DNA
Tumor DNA was obtained from macrodissected formalin-fixed, paraffin-embedded (FFPE)-
derived tumor sections from six of the seven primary renal tumors. Each of the tumors represents
a discrete well circumscribed neoplasm. DNA and RNA were extracted individually from each
tumor with a pathologist-verified tumor content exceeding 70 percent (AllPrep Qiagen kit). The
proband’s germline DNA was isolated from peripheral blood lymphocytes using Omega Bio-
tek’s Mag-Bind® Blood & Tissue DNA HDQ Kit (https://www.omegabiotek.com/product/mag-
bind-hdq-blood-dna-96-kit/).

4.3.6 Haplotype Analysis using B Allele Frequency (BAF)
Genotyping was performed employing a high density custom exomechip array (Multi Ethnic
Genotyping Array-MEGAarray by Illumina) in four of the seven renal tumors due to DNA
quality and quantity. It was determined which haplotypes are associated with the translocation by
measuring B-allele frequencies (BAF) in the parental germline DNA and the proband’s germline
DNA. B Allele Frequency (BAF) is used to identify copy number variation across the genome
and can be useful in haplotype analysis. It was determined which haplotypes are associated with
the translocation by measuring BAF in the parental germline DNA and the proband’s germline
DNA. BAF measures the relative frequency of the B allele at a particular locus. It is the fraction
of all chromosomes that carry that allele. In a normal sample discrete BAFs of 0.0, 0.5 and 1 are
expected for each locus representing AA, AB and BB. Deviations from this expectation are
indicative of aberrant copy number.

82
4.3.7 Determination of the Parental of Origin of the t(3;8) translocation

In order to determine the parental origin of the t(3;8) translocation, allele-specific copy number
plots for four of the seven tumors were generated using the data analysis software Partek. Allele
specific copy number (ASCN) is a simple, effective method to identify duplicated alleles in
heterozygous samples. By measuring the quantity of both alleles at heterozygous loci,
genotyping arrays allow the estimation of the copy numbers of each allele. The Log R ratio
(LRR) represents the log ratio of observed probe intensity to expected intensity. Any deviation
from zero in this metric are evidence for copy number change.

4.3.8 Whole exome sequencing (WES) analysis
To characterize the pathobiology of the t(3;8) balanced translocation associated with ccRCC, we
performed next generation sequencing experiments and subsequent bioinformatics analysis. We
performed 60X Whole Exome Sequencing (WES) of the germline DNA of the proband together
with 300X whole exome sequencing of six of the seven the primary tumors. The data analysis
workflow is shown in Figure 4.2. Through WES, we also calculated the tumor mutational load
or tumor mutational burden; one biomarker which often informs decision making as it has been
associated with response to immunotherapy in multiple cancer types (Yarchoan, Hopkins, &
Jaffee, 2017).

83

Figure 4.2: Whole exome sequencing analysis workflow. The NGS Data were downloaded to the
University of Southern California High Performance Cluster. The alignment was performed using BWA-
MEM using the FASTQ files. Variant Calling was performed using GATK and MuTect2 to subtract
germline from somatic calls (Cibulskis et al., 2013). Manual curation and data integrity assessment was
performed using Integrative Genome Viewer (IGV). Finally the biological interpretation and possible
drug matching were discussed during our weekly Precision Oncology Tumor Board.

4.3.9 Transcriptomic Analysis
The transcriptomic profile of the renal tumors and matched normal tissues was assessed
employing next generation RNA-Seq technology at a depth of 50 million reads (TruSeq RNA
Exome Kit, by Illumina). Because FFPE-derived RNA can be highly degraded we initially
employed barcoded probe hybridization technology (NanoString) to assess the transcriptomic
profile of the renal tumors and matched normal tissues. We used the PanCancer pathway panel
which includes 13 canonical pathways and 770 genes including the VHL. Raw data were
normalized relative to a panel of housekeeping genes and positive controls. The ratios were
generated by dividing the mean of all the tumor samples by the mean of the normal samples. A t-
test was performed with log2 transformed normalized data. The threshold used was Log2FC
(fold change) =1.

84

Despite the possibility of significant RNA degradation, we have had consistent success
generating highly informative RNA-Seq libraries with recently optimized protocols (TruSeq
RNA Exome Kit, by Illumina). Therefore, we also performed next generation RNA-Seq
technology at a depth of 50 million paired reads to validate NanoString findings and obtain a
more comprehensive transcriptomic analysis. The RNA- Seq analysis was performed in a
commercial lab and the data were aligned using the STAR aligner and analyzed by DESeq2 for
the differential expression analysis.

4.3.10 Generation of t(3;8) cell lines
4.3.10.1 Tissue collection
Human kidney tissue samples were obtained from the proband in May 2018 when he underwent
right redo partial nephrectomy. Tissue samples were collected from areas macroscopically
identified as normal renal parenchyma and tumoral immediately after the specimen extraction,
by an expert uropathologist. Tissues were placed in separated sterile 50-mL tubes with ice-cold
culture medium and then transferred to a cell culture laboratory on ice.

4.3.10.2 Renal Cell Carcinoma Cell Isolation

Renal cell carcinoma cell isolation took place within 30 minutes of renal tissue collection. All
subsequent procedures were performed in a tissue culture flow hood, under sterile conditions. To
avoid cell cross contamination, the normal and tumor specimens were handled in two individual
tissue-culture hoods. The tissue was transferred to a 60 mm Petri dish. The fibrous capsule and
adjacent medulla from the cortical tissue together with any fat and blood from the tumor tissue

85
were dissected off using forceps and scissors. The tissue sample was cut into small pieces with
scalpels. The tissue fragments were transferred to a sterile 50 ml centrifuge tube, were vigorously
rinsed with ice cold han’s balanced salt solution (HBSS) and the supernatant was discarded. The
fragments were transferred to a clean 60 mm petri dish and the tissue was minced into 1 mm
3
pieces with scalpels.

The small fragments were resuspended in 25ml pre-warmed nonsupplemented culture medium
which was combined with the collagenase solution (1 mg/mL final concentration) in the
incubation vessel. An incubation for 20 min at 37
o
C with gentle stirring was followed. The
digested tissue was passed onto the first sieve (100 mm) into a 50 ml centrifuge tube. The same
procedure was the applied to the following sieves (70 and 40 mm). The sieved cells were washed
by centrifugation (400 g, 5 min at 4
o
C) and the pellet was resuspended in HBSS. This process
was repeated two more times, and the cell pellet was resuspended in culture medium with
supplements (DMEM-F12). The cell number and viability was determined using a Newbauer
hemocytometer using trypan blue solution.

4.3.10.3 Cytospin technique
The cells were prepared for microscopy using cytospin applying the following protocol: The
centrifuge tube was filled with 40ml of the specimen. The specimen was centrifuged at 2500rpm
for 10 minutes. The supernatant was removed using a pipette with 0.5ml to be left at the end. The
cell pellet was resuspended. The resuspended fluid was used to make cytospin slides. Two drops
of each cytospin slide was used at 1250 rpm for 3 minutes. One cytospin slide was fixed in 95%
alcohol for Pap stain. Another slide was left dry for Diff Quick stain. 10% formalin was added to
fix the rest of the specimen in the tube for 15 minutes and we it was centrifuged at 2500rpm for

86
10mins. The supernatant was discarded and 10ml of 95% alcohol was added. The cells were
resuspended and were centrifuged at 2500 rpm for 10 mins. The supernatant was discarded. The
pellet was removed with a spatula, was placed on a piece of tissue paper, folded the tissue paper
around the pellet, placed it in a cassette and closed the lid. Finally, cassette was placed in 95%
alcohol and was transferred to the histology lab for processing.

4.3.10.4 Cell immortalization
Renal cells were immortalized using the hTERT Antigen Cell Immortalization Kit (by
ALSTEM). The target cells were plates in one well of 6-well plate at density of 1-2 × 105
cells/well. The next day, the concentrated recombinant lentivirus was thawed in a a 37° C water
bath and was removed from the bath immediately when thawed. The target cells were infected in
a 6-well plate with 4-20 µl/well supernatant in the presence of 4 µl TransPlus reagent (ALSTEM,
Cat. # V050). The following day, the viral supernatant was removed and the appropriate
complete growth medium (DMEM-F12) was added to the cells and were incubated at 37° C.
After 72 hours of incubation, the cells were subcultured into 2 x 100 mm dishes and 1-10ug/ml
puromycin was added for stable cell-line generation. Note: After thawing, the supernatant was
not frozen again for future use since the virus-titer would decrease significantly.

4.4 Results
4.4.1 Cytogenetics
While the cytogenetic analysis of the products of conception from the proband’s offspring
clearly demonstrated the presence of the t(3;8)(p14.2;q24.1) balanced translocation, spectral
karyotyping of lymphocytes from the proband unequivocally confirmed the presence of the
constitutional translocation (Figure 4.3). This confirms the clinical diagnosis of hereditary renal

87
cell carcinoma due to PATRR-mediated balanced translocation (OMIM #144700).

Figure 4.3: The de novo constitutional balanced t(3;8)(p14.2;q24.1) translocation was validated
using Spectral Karyotyping (SKY) technique.

4.4.2 Breakpoint Sequencing
The chromosome 3 breakpoint was determined to be located in an AT-rich palindromic sequence
in the third intron of the FHIT gene (chr3p14.2) and the chromosome 8 breakpoint to lie in an
AT-rich palindromic sequence in the first intron of the RNF39 gene (a.k.a TRC8 gene)
(chr8q24.1) (Figure 4.4).

88

Figure 4.4: PCR determines the breakpoint region of the t(3;8) translocation in the
proband. a) Primer locations are shown as arrows above the chromosomes. The black bar indicates
chromosome 8q, the red bar the chromosome 8 PATRR, the white bar indicates the chromosome 3
PATRR and the grey bar chromosome 3p (Adapted from Kato et al., 2014). b) Agarose gel showing 1kb
ladder and PCR DNA products from: “Lane –”, representing a negative control patient DNA without
t(3;8), “Lane +”, positive controls with t(3;8) (courtesy Emanuel Lab); and “arrow Lane”, t(3;8) proband.
4.4.3 Fusion Transcript Expression
Sanger Sequencing of the cDNA and subsequent validation through RNA-Seq, revealed that a
fusion transcript between exon of TRC8 and exon 4 of FHIT is expressed as a result of the
translocation (expected size 150 bp) (Figure 4.5)

Figure 4.5: cDNA Sequence of the Fusion Transcript

89
4.4.4 Allele-specific Haplotype Transmission
All of the proband’s tumors which were tested demonstrated a consistent loss of the entire
aberrant chromosome 8, der(8), which includes the segment 3p14.2. The mosaic proportion of
the normal chromosome 8 varies among the tumor samples as evidenced by the separation
between the two central bands (Figure 4.6). None of the other tumor chromosomes exhibited this
distinguishing pattern. The variable tumor cellularity evident in the B allele-frequency plots we
interpret as deriving from admixture of normal, non-cancerous cells and/or intra-tumoral
heterogeneity.

6a) chr.3 FHIT

90

6b) chr.8 TRC8
Figure 4.6: B allele frequency (BAF) plots for the proband’s germline (yellow) and four of the
tumor DNA samples (red) for chromosomes 3 (panel a) and 8 (panel b).
https://www.pagestudy.org/index.php/multi-ethnic-genotyping-array.

For the regions on chromosome 3 and chromosome 8, it was determined whether the allele with
the lower frequency (min allele) originated from the father or from the mother for each of the
tumors tested. A loss of the derivative (aberrant) chromosome 8 in the tumors was detected as
shown in figure 6. Therefore, the inherited allele with the lower frequency (min allele) should
represent the chromosome that is lost (Figure 4.7). Hence, for SNPs in the regions where father
and mother are both homozygous but discordant, and where proband germline is heterozygous,
paternal and maternal allelic inheritance were ascribed. The regions were defined as 1kb
upstream of FHIT gene on chromosome 3, and 1kb upstream of TRC8 gene on chromosome 8.

91
BAF in the tumor sample was required to be either >0.6 or <0.4. The alleles showing the loss in
the trisomic regions, derive from the father.
0.0MBps 49.5MBps 99.0MBps 148.5MBps 198.0MBps
chr3
7
Allele Specific Copy Number - Tumor (6) Max Allele Min Allele
0.50
1.00
1.50
Allele Specific Copy Number - Tumor (5) Max Allele Min Allele
0.50
1.00
1.50
Allele Specific Copy Number - Tumor (2T) Max Allele Min Allele
0.50
1.00
1.50
Allele Specific Copy Number - Tumor (1) Max Allele Min Allele
0.50
1.00
1.50

0.0MBps 36.6MBps 73.2MBps 109.8MBps 146.4MBps
chr8
6
Allele Specific Copy Number - Tumor (6) Max Allele Min Allele
0.50
1.00
1.50
Allele Specific Copy Number - Tumor (5) Max Allele Min Allele
0.50
1.00
1.50
Allele Specific Copy Number - Tumor (2T) Max Allele Min Allele
0.50
1.00
1.50
Allele Specific Copy Number - Tumor (1) Max Allele Min Allele
0.50
1.00
1.50

7a) chr.3 FHIT 7b) chr.8 TRC8
Figure 4.7: Allele specific copy number plots for each of the tumor samples for chromosome 3
(figure 7a) and chromosome 8 (figure 7b) using the normal (germline) sample as a reference
(generated using the data analysis software Partek).

4.4.5 Next generation sequencing analysis
No deleterious mutation in VHL gene was detected in any of the tumors tested. The total
mutational load (mutations/megabase) for two of the renal tumors tested was very low (0.2 and
0.9 respectively). Anything less than 5 is considered low mutational burden. In regards to the
transcriptomic analysis, significantly no difference in VHL gene expression between normal and
renal tumor was detected. Nanostring tumor expression profiles demonstrated that in the t(3;8)
ccRCCs, the well-established DNA repair, tumor suppressor genes BRCA1 and BRIP-1 and the
anti-tumor cytokines TNF and INFγ were down-regulated; conversely, the oncogenes, FOS,
JUN, HES1, and PGF, all previously associated with VHL-mutated ccRCC were up-regulated

92
(Figure 8). In addition, RNA-Seq analysis confirmed our Nanostring results and also showed
that TRC8 and FHIT expression is not statistically significant different in the tumors compared
to the normal (Table 4.2).

Gene p-value Log2 Fold Change Total reads normalized
TRC8 0.9 0.05 2,793.6
FHIT 0.6 0.3 497.2

4.4.6 Cell line establishment
Figure 4.9 shows the Hematoxylin and Eosin (H&E) histopathology of the t(3;8) renal cell
carcinoma and the newly established cell lines. In particular, figure 9a shows the H&E of the
latest primary renal tumor (tumor number 8) which was used for the generation of the cell lines.
Figure 9b shows the H&E of the cell line t38tumK/cytospin. Figure 9c, shows the H&E of the
cell line t38norK/cytospin. The special stains of the cell line t38tumK/cytospin with cytokeratin,
PAX8 and CD10 as shown in figures 9d, figure 9e and figure 9f respectively.
Figure 4.8: Volcano plot of gene expression data

Red dots represent the genes with significantly
differential overexpression.

Blue dots represent the genes with significantly
differential underexpression.

Grey dots represent genes with no significant difference

Table 4.2: Differential expression (Tumors VS normal) for TRC8 and FHIT genes.

93

Figure 4.9a: H&E of the t(3;8) tumor (Olympus BX46, Tokyo, Japan, x100)

Figure 4.9b: H&E of the cell line t38tumK/cytospin (Olympus BX46, Tokyo, Japan, x100)

94

Figure 4.9c: H&E of the cell line t38norK/cytospin (Olympus BX46, Tokyo, Japan, x200)

Figure 4.9d: Cytokeratin stain of the cell line t38tumK/cytospin (Olympus BX46, Tokyo, Japan,
x100)

95

Figure 4.9e: PAX8 stain of the cell line t38tumK/cytospin (Olympus BX46, Tokyo, Japan,
x100)

Figure 4.9f: CD10 stain of the cell line t38tumK/cytospin (Olympus BX46, Tokyo, Japan, x100)

96
4.4.7 Quantifying the Host Immune Response
The host immune response within each of five primary renal carcinomas was qualitatively and
quantitatively assessed using ImmunoSEQ by PCR amplification of the TCR β receptor,
demonstrating a similar host response in each of the tumors. A single clone was shared among
all five tumors with informative data, and another single clone shared in four tumors. Four
individual lymphocyte clones were shared among three tumors and finally four additional clones
were shared among two tumors (Figure 4.10a). Interestingly, the distributional landscape of the
single clone shared among all five tumors varies proportionally between the five primary cancers
from the same patient, and represents a relatively low fraction of the total TCRs measured within
the tumors (range 0.5% to 4%), (Figure 4.10b). These data provide evidence of a low-level of
shared immune activity across all t(3;8) ccRCC tumors from the same patient, suggesting the
possibility that a shared neoantigen driven by the expressed fusion transcript might be
responsible for this lymphocytic clone. Other shared clones seen in some, but not all cancers,
may also represent a host response targeted to the t(3;8) fusion.

Figure 4.10: a) Shared T-Cell Receptor β Amino Acid Sequences among the five ccRCC tumors.
b) Frequency of the single clone which is shared among all five ccRCC tumors.

97
4.5 Discussion
In the current study, we reported a patient who harbors a de novo constitutional PATRR-
mediated t(3;8)(p14.2;q24.1) balanced translocation associated with bilateral clear cell renal
cell carcinoma. The derivation of this constitutional rearrangement were characterized by both
SNP arrays, Sanger sequencing, and next generation sequencing, clearly demonstrating that this
translocation is paternally derived and arose during spermatogenesis. Furthermore we show that
this truly represents a balanced translocation, as there is no loss of chromosomal material at the
site of the breakpoint. The first report of this constitutional translocation was described in a
family with the classical features of hereditary renal cell carcinoma (RCC), including autosomal
dominant inheritance and early onset, multiple primary cancers (Cohen et al., 1979).

The translocation co-segregated with the RCC phenotype, and a follow-up analysis reported the
occurrence of thyroid cancer in two translocation carriers with RCC (F. P. Li et al., 1993).
Additional published series of the t(3;8) translocation in unrelated families confirmed the clinical
features of this syndrome. In particular, the current case brings the total number of reported,
unrelated families to four (R M Gemmill et al., 1998; Meléndez et al., 2003; Poland et al., 2007;
Rodríguez-Perales et al., 2004).The t(3;8)(p14.2;q24.1) is the first PATRR-mediated
translocation described that does not involve the chromosome 22 as the translocation partner
(Kato et al., 2014). The chromosome 3 breakpoint lies at the center of an AT-rich rich
palindromic sequence in the third intron of the FHIT gene (chr3p14.2). The chromosome 8
breakpoint lies at the center of an AT-rich palindromic sequence in the first intro on the RNF139

98
gene (a.k.a TRC8). The 3p14.2 breakpoint interrupts FHIT in its 5-prime non-coding region.
FHIT gene comprises 10 exons and encodes for a 140aa protein. It is a member of the histidine
triad gene family and encodes a diadenosine P1,P3-bis(5'-adenosyl)-triphosphate
adenylohydrolase involved in purine metabolism. The gene encompasses the common fragile site
FRA3B on chromosome 3 (Glover, Wilson, & Arlt, 2017), where carcinogen-induced damage
can lead to translocations and aberrant transcripts of this gene. Though the exact molecular
function of FHIT is still partially unclear, the gene works as a tumor suppressor gene as it has
been demonstrated in animal studies (Zanesi et al., 2001). Furthermore FHIT has been shown to
synergize with VHL in protecting against chemically - induced lung cancer (Zanesi et al., 2005).

TRC8 gene comprises 2 exons, encodes a predicted 664-aa multi-membrane spanning protein
located in the endoplasmic reticulum (ER), and contains a C-terminal RING-H2 domain with E3-
ubiquitin ligase activity (Brauweiler et al., 2007) ; (Gemmill et al., 2002, 2005). In addition to
the RING domain, TRC8 contains a predicted sterol-sensing domain (SSD) in its N-terminus,
suggesting that sterols may influence some aspect of TRC8 function (Gemmill et al., 1998).
Interestingly, the clear-cell phenotype in RCC is caused by abundant inclusions of endogenously
synthesized cholesterol esters and other lipids, suggesting that lipid homeostasis may be
deregulated in this disease (Gebhard et al., 1987) ; (Tosi & Tugnoli, 2005). The 8q24.1
breakpoint disrupts TRC8 within its sterol sensing domain.

In the present study, no deleterious germline or somatic mutations in VHL gene contribute to the
t(3;8) phenotype. No difference in VHL expression between normal and renal tumor tissue was

99
detected in any of the tumors. In addition, no deleterious mutations in the breakpoint genes FHIT
and TRC8 were detected via WES.

Analysis of RNA-Seq data demonstrated that the expression of each of the breakpoint genes does
not differ significantly between the ccRCC tumors compared to the adjacent normal. Tumor
expression profiles contributed to the appreciation of the genes that are dysregulated in t(3;8)
associated ccRCC, including the downregulation of well-known tumor suppressor genes
(BRCA1, BRIP1, TNF, IFNγ), and upregulation of well-known oncogenes such as FOS, JUN,
HES1, and PGF. For future studies, we can compare our RNA-Seq results with publicly
available expression data generated using VHL-mutated, non-t(3;8) ccRCC cases from TCGA
and see what pathways are dysregulated or activated as a consequence of the PATRR-mediated
t(3;8) translocation.

Cell lines were established from normal renal parenchyma and ccRCC, and these constitute an
exceptional experimental model for testing candidate therapies directed specifically against the
expressed TRC8/FHIT fusion protein. While it is important to understand that this constitutional
translocation is obviously present in every cell within the patient, and therefore one has to be
attentive to the possibility of toxicity to tissues other than ccRCC, several therapies could be
imagined that could offer a potential therapeutic advantage (for example siRNA, anti-sense,
computationally predicted designer drugs that match the predicted 3-dimensional structure of the
fusion protein, or even high-throughput candidate drug screen) (Graham et al., 2017)(Adams et
al., 2018), FDA news release (August 10
th
, 2018) FDA approves first-of-its kind targeted RNA-
based therapy to treat a rare disease retrieved from: https://www.fda.gov/news-events/press-
announcements/fda-approves-first-its-kind-targeted-rna-based-therapy-treat-rare-disease

100
The balanced translocation leads to the expression of a fusion transcript, TRC8/FHIT. The
presence of an expressed fusion transcript raises the possibility that a neoantigen that is specific
to the t(3;8) TRC8/FHIT fusion may induce a host immune response, and that this specific host
response could be manipulated for therapeutic or preventive purposes in a single patient. The
data presented here suggest that the CASSKKGSGYGQPQHF amino acid sequence from the
hypervariable region of TCR-B that is shared across all five informative tumors designates one
of these candidate TIL clones. Indeed, this hypothesis could be tested using both observational
and experimental approaches. For example, this patient routinely develops new primary ccRCCs,
and applying immunoSEQ sequentially over time to liquid biopsies of blood, both before and
after resection of a tumor, could provide evidence that a single clone varies with the tumor
burden that is evident pre- and post op. Experimentally, one could test an experimentally
designed neoantigen by synthesizing a protein fragment corresponding to the fusion transcript
and introducing this into cultured lymphocytes from the patient.

Here we demonstrate for the first time that the constitutional t(3;8)(p14.2;q24.1) balanced
translocation is indeed paternally derived and arose de novo during spermatogenesis, as one
would expect for PATRR-mediated translocations. The establishment of novel cell lines from
normal renal parenchyma and clear cell renal carcinoma arising on the basis of the fusion
transcript also adds new models for the scientific community to interrogate this specific
translocation. In addition, the current in-depth single case study has provided further
characterization of the deleterious biological consequences of the PATRR-mediated t(3;8)
constitutional rearrangement giving rise to ccRCC. These studies collectively advance our
understanding of the pathogenesis of a subset of renal carcinomas, and position the scientific

101
community for the development of new therapeutic strategies directed against t(3;8)-associated
ccRCC.

102

Chapter 5

Concluding remarks

In this dissertation, we elucidated the genomic, transcriptomic and immunologic landscapes of
human cancers, through laboratory studies, statistical genetics analyses, and bioinformatics
methods, focusing on colorectal cancer and clear cell renal cell carcinoma.

The first project studied tumor infiltrating lymphocytes (TILs) and clonal analysis of the
hypervariable region of T-Cell β receptors in order to quantify the TCR repertoire of individual
colorectal cancers from a large, population-based study of colorectal cancer in Israel. Using
clinical, epidemiologic, genomic and transcriptomic data from the Molecular Epidemiology of
Colorectal Cancer (MECC) Study, colorectal cancers were classified according to consensus
molecular subtypes (CMS) where gene expression data were available, and interpreted from the
perspective of the host immune response.

The second project focused on a subset of Druze individuals with Lynch syndrome from Israel
who harbor a germline MSH2*c.705delA mutation. In this study, a recurrent mutation in MSH2
was studied as a potential founder mutation by characterizing the allele sharing identical-by-
descent among Druze carriers. The genetic closeness of this community in conjunction with the
cryptic familial relationships of independently ascertained families limits the statistical analysis
and interpretation of this recurrent mutation as a “founder” mutation. Therefore, this private
mutation that is known to be an important cause of Lynch Syndrome in the Druze population was
characterized from a clinical, immunologic and pathologic perspective.

103
For the third project, we elucidated the pathobiological consequences and the parental origin of a
de novo constitutional PATRR-mediated t(3;8)(p14.2;q24.1) balanced translocation. For the first
time, this dissertation research proved that the t(3;8) PATRR-mediated translocation is paternally
derived and arose during spermatogenesis, as one would expect for PATRR-mediated
translocations.. This definitive conclusion is made possible by the dissertation case-study
describing the first t(3;8) proven to arise de novo. In addition, it has provided further
characterization of the deleterious biological consequences of the PATRR-mediated t(3;8)
constitutional rearrangement giving rise to ccRCC and has positioned us well for future
functional studies which will provide the basis for the development of new therapeutic strategies
directed against t(3;8) associated ccRCC.

From a clinical point of view, research has entered a new era in cancer treatment. The growth of
knowledge through cancer genomics and immunology within the last few years has already
altered our view of cancer therapies. Data from this dissertation begin to close the gap between
fundamental basic genomic science and its clinical application, and points towards a future of
well-designed, personalized medicine trials, and the corresponding translational laboratory
studies to inform new approaches to human cancers.

Recent advances in genome technologies and the cascade of genomic information have
accelerated the convergence of discovery science and clinical medicine. Successful examples
of translating cancer genomics into therapeutics and diagnostics reinforce its potential to
make possible personalized cancer medicine. However, the bottlenecks along the path of
converting a genome discovery into a tangible clinical endpoint are numerous. This
dissertation emphasizes the importance of establishing the biological relevance of a cancer

104
genomic discovery in realizing its clinical potential in order to move from the bench to the
bedside.

105
Bibliography

Adams, D., Gonzalez-Duarte, A., O’Riordan, W. D., Yang, C.-C., Ueda, M., Kristen, A. V.,
Tournev, I., et al. (2018). Patisiran, an rnai therapeutic, for hereditary transthyretin
amyloidosis. The New England Journal of Medicine, 379(1), 11–21.
Ahvenainen, T., Lehtonen, H. J., Lehtonen, R., Vahteristo, P., Aittomäki, K., Baynam, G.,
Dommering, C., et al. (2008). Mutation screening of fumarate hydratase by multiplex
ligation-dependent probe amplification: detection of exonic deletion in a patient with
leiomyomatosis and renal cell cancer. Cancer Genetics and Cytogenetics, 183(2), 83–88.
Alexander, J., Watanabe, T., Wu, T. T., Rashid, A., Li, S., & Hamilton, S. R. (2001).
Histopathological identification of colon cancer with microsatellite instability. The
American Journal of Pathology, 158(2), 527–535.
Alexandrov, L. B., Nik-Zainal, S., Wedge, D. C., Aparicio, S. A. J. R., Behjati, S., Biankin, A.
V., Bignell, G. R., et al. (2013). Signatures of mutational processes in human cancer.
Nature, 500(7463), 415–421.
Al-Tassan, N., Chmiel, N. H., Maynard, J., Fleming, N., Livingston, A. L., Williams, G. T.,
Hodges, A. K., et al. (2002). Inherited variants of MYH associated with somatic G:C--
>T:A mutations in colorectal tumors. Nature Genetics, 30(2), 227–232.
André, T., de Gramont, A., Vernerey, D., Chibaudel, B., Bonnetain, F., Tijeras-Raballand, A.,
Scriva, A., et al. (2015). Adjuvant Fluorouracil, Leucovorin, and Oxaliplatin in Stage II
to III Colon Cancer: Updated 10-Year Survival and Outcomes According to BRAF
Mutation and Mismatch Repair Status of the MOSAIC Study. Journal of Clinical
Oncology, 33(35), 4176–4187.
Arstila, T. P., Casrouge, A., Baron, V., Even, J., Kanellopoulos, J., & Kourilsky, P. (1999). A
direct estimate of the human alphabeta T cell receptor diversity. Science, 286(5441),
958–961.
Ashley, T., Gaeth, A. P., Inagaki, H., Seftel, A., Cohen, M. M., Anderson, L. K., Kurahashi, H.,
et al. (2006). Meiotic recombination and spatial proximity in the etiology of the recurrent
t(11;22). American Journal of Human Genetics, 79(3), 524–538.
Baldus, S. E., Schaefer, K.-L., Engers, R., Hartleb, D., Stoecklein, N. H., & Gabbert, H. E.
(2010). Prevalence and heterogeneity of KRAS, BRAF, and PIK3CA mutations in
primary colorectal adenocarcinomas and their corresponding metastases. Clinical Cancer
Research, 16(3), 790–799.
Bandyopadhyay, R., Heller, A., Knox-DuBois, C., McCaskill, C., Berend, S. A., Page, S. L., &
Shaffer, L. G. (2002). Parental origin and timing of de novo Robertsonian translocation
formation. American Journal of Human Genetics, 71(6), 1456–1462.
Belov, L., Zhou, J., & Christopherson, R. I. (2010). Cell surface markers in colorectal cancer
prognosis. International Journal of Molecular Sciences, 12(1), 78–113.
Boland, C. R., Thibodeau, S. N., Hamilton, S. R., Sidransky, D., Eshleman, J. R., Burt, R. W.,
Meltzer, S. J., et al. (1998). A National Cancer Institute Workshop on Microsatellite
Instability for cancer detection and familial predisposition: development of international
criteria for the determination of microsatellite instability in colorectal cancer. Cancer
Research, 58(22), 5248–5257.
Brahmer, J. R., Tykodi, S. S., Chow, L. Q. M., Hwu, W.-J., Topalian, S. L., Hwu, P., Drake, C.
G., et al. (2012). Safety and activity of anti-PD-L1 antibody in patients with advanced
cancer. The New England Journal of Medicine, 366(26), 2455–2465.

106
Brauweiler, A., Lorick, K. L., Lee, J. P., Tsai, Y. C., Chan, D., Weissman, A. M., Drabkin, H.
A., et al. (2007). RING-dependent tumor suppression and G2/M arrest induced by the
TRC8 hereditary kidney cancer gene. Oncogene, 26(16), 2263–2271.
Broderick, P., Carvajal-Carmona, L., Pittman, A. M., Webb, E., Howarth, K., Rowan, A., Lubbe,
S., et al. (2007). A genome-wide association study shows that common alleles of SMAD7
influence colorectal cancer risk. Nature Genetics, 39(11), 1315–1317.
Buwe, A., Guttenbach, M., & Schmid, M. (2005). Effect of paternal age on the frequency of
cytogenetic abnormalities in human spermatozoa. Cytogenetic and Genome Research,
111(3–4), 213–228.
Cancer Genome Atlas Research Network, Weinstein, J. N., Collisson, E. A., Mills, G. B., Shaw,
K. R. M., Ozenberger, B. A., Ellrott, K., et al. (2013). The Cancer Genome Atlas Pan-
Cancer analysis project. Nature Genetics, 45(10), 1113–1120.
Carlson, C. S., Emerson, R. O., Sherwood, A. M., Desmarais, C., Chung, M.-W., Parsons, J. M.,
Steen, M. S., et al. (2013). Using synthetic templates to design an unbiased multiplex
PCR assay. Nature Communications, 4, 2680.
Carpten, J. C., & Mardis, E. R. (2018). The era of precision oncogenomics. Molecular Case
Studies, 4(2).
Carstensen, B., Soll-Johanning, H., Villadsen, E., Søndergaard, J. O., & Lynge, E. (1996).
Familial aggregation of colorectal cancer in the general population. International Journal
of Cancer, 68(4), 428–435.
Cavalli-Sforza, L., & Bodmer, W. (1971). The Genetics of Human Populations. San Francisco:
W. H. Freeman and Co.
Cavalli-Sforza, L. L. (2005). The Human Genome Diversity Project: past, present and future.
Nature Reviews. Genetics, 6(4), 333–340.
Chan, T. L., Yuen, S. T., Ho, J. W., Chan, A. S., Kwan, K., Chung, L. P., Lam, P. W., et al.
(2001). A novel germline 1.8-kb deletion of hMLH1 mimicking alternative splicing: a
founder mutation in the Chinese population. Oncogene, 20(23), 2976–2981.
Chandley, A. C. (1991). On the parental origin of de novo mutation in man. Journal of Medical
Genetics, 28(4), 217–223.
Chew, A., Salama, P., Robbshaw, A., Klopcic, B., Zeps, N., Platell, C., & Lawrance, I. C.
(2011). SPARC, FOXP3, CD8 and CD45 correlation with disease recurrence and long-
term disease-free survival in colorectal cancer. Plos One, 6(7), e22047.
COGENT Study, Houlston, R. S., Webb, E., Broderick, P., Pittman, A. M., Di Bernardo, M. C.,
Lubbe, S., et al. (2008). Meta-analysis of genome-wide association data identifies four
new susceptibility loci for colorectal cancer. Nature Genetics, 40(12), 1426–1435.
Cohen, A. J., Li, F. P., Berg, S., Marchetto, D. J., Tsai, S., Jacobs, S. C., & Brown, R. S. (1979).
Hereditary renal-cell carcinoma associated with a chromosomal translocation. The New
England Journal of Medicine, 301(11), 592–595.
Correll-Tash, S., Conlin, L., Mininger, B. A., Lilley, B., Mennuti, M. T., & Emanuel, B. S.
(2018). The Recurrent t(11;22)(q23;q11.2) Can Occur as a Post-Zygotic Event.
Cytogenetic and Genome Research, 156(4), 185–190.
Crow, J. F. (2000). The origins, patterns and implications of human spontaneous mutation.
Nature Reviews. Genetics, 1(1), 40–47.
Davis, M. M., & Bjorkman, P. J. (1988). T-cell antigen receptor genes and T-cell recognition.
Nature, 334(6181), 395–402.
De Gregori, M., Ciccone, R., Magini, P., Pramparo, T., Gimelli, S., Messa, J., Novara, F., et al.

107
(2007). Cryptic deletions are a common finding in “balanced” reciprocal and complex
chromosome rearrangements: a study of 59 patients. Journal of Medical Genetics, 44(12),
750–762.
Delaneau, O., Howie, B., Cox, A. J., Zagury, J.-F., & Marchini, J. (2013). Haplotype estimation
using sequencing reads. American Journal of Human Genetics, 93(4), 687–696.
Delaneau, O., Marchini, J., 1000 Genomes Project Consortium, & 1000 Genomes Project
Consortium. (2014). Integrating sequence and array data to create an improved 1000
Genomes Project haplotype reference panel. Nature Communications, 5(5), 3934.
Delaneau, O., Marchini, J., & Zagury, J.-F. (2011). A linear complexity phasing method for
thousands of genomes. Nature Methods, 9(2), 179–181.
Deschoolmeester, V., Baay, M., Lardon, F., Pauwels, P., & Peeters, M. (2011). Immune Cells in
Colorectal Cancer: Prognostic Relevance and Role of MSI. Cancer Microenvironment,
4(3), 377–392.
Dinh, T. A., Rosner, B. I., Atwood, J. C., Boland, C. R., Syngal, S., Vasen, H. F. A., Gruber, S.
B., et al. (2011). Health benefits and cost-effectiveness of primary genetic screening for
Lynch syndrome in the general population. Cancer Prevention Research, 4(1), 9–22.
Dolcetti, R., Viel, A., Doglioni, C., Russo, A., Guidoboni, M., Capozzi, E., Vecchiato, N., et al.
(1999). High prevalence of activated intraepithelial cytotoxic T lymphocytes and
increased neoplastic cell apoptosis in colorectal carcinomas with microsatellite
instability. The American Journal of Pathology, 154(6), 1805–1813.
Droeser, R. A., Hirt, C., Viehl, C. T., Frey, D. M., Nebiker, C., Huber, X., Zlobec, I., et al.
(2013). Clinical impact of programmed cell death ligand 1 expression in colorectal
cancer. European Journal of Cancer, 49(9), 2233–2242.
Druker, B. J., Lydon, N. B. Lessons learned from the development of an Abl tyrosine kinase
inhibitor for chronic myelogenous leukemia. (2000). The Journal of Clinical Investigation,
105(1),3-7.
Fang, J. Y., & Richardson, B. C. (2005). The MAPK signalling pathways and colorectal cancer.
The Lancet Oncology, 6(5), 322–327.
Fearon, E. R., & Vogelstein, B. (1990). A genetic model for colorectal tumorigenesis. Cell,
61(5), 759–767.
Fischer, J., Walker, L. C., Robinson, B. A., Frizelle, F. A., Church, J. M., & Eglinton, T. W.
(2019). Clinical implications of the genetics of sporadic colorectal cancer. ANZ Journal
of Surgery.
Foulkes, W. D., Thiffault, I., Gruber, S. B., Horwitz, M., Hamel, N., Lee, C., Shia, J., et al.
(2002). The founder mutation MSH2*1906G-->C is an important cause of hereditary
nonpolyposis colorectal cancer in the Ashkenazi Jewish population. American Journal of
Human Genetics, 71(6), 1395–1412.
Fridman, W. H., Pagès, F., Sautès-Fridman, C., & Galon, J. (2012). The immune contexture in
human tumours: impact on clinical outcome. Nature Reviews. Cancer, 12(4), 298–306.
Gabrielson, A., Wu, Y., Wang, H., Jiang, J., Kallakury, B., Gatalica, Z., Reddy, S., et al. (2016).
Intratumoral CD3 and CD8 T-cell Densities Associated with Relapse-Free Survival in
HCC. Cancer immunology research, 4(5), 419–430.
Galon, J., Costes, A., Sanchez-Cabo, F., Kirilovsky, A., Mlecnik, B., Lagorce-Pagès, C.,
Tosolini, M., et al. (2006). Type, density, and location of immune cells within human
colorectal tumors predict clinical outcome. Science, 313(5795), 1960–1964.
Gebhard, R. L., Clayman, R. V., Prigge, W. F., Figenshau, R., Staley, N. A., Reesey, C., & Bear,

108
A. (1987). Abnormal cholesterol metabolism in renal clear cell carcinoma. Journal of
Lipid Research, 28(10), 1177–1184.
Gemmill, R M, West, J. D., Boldog, F., Tanaka, N., Robinson, L. J., Smith, D. I., Li, F., et al.
(1998). The hereditary renal cell carcinoma 3;8 translocation fuses FHIT to a patched-
related gene, TRC8. Proceedings of the National Academy of Sciences of the United
States of America, 95(16), 9572–9577.
Gemmill, Robert M, Bemis, L. T., Lee, J. P., Sozen, M. A., Baron, A., Zeng, C., Erickson, P. F.,
et al. (2002). The TRC8 hereditary kidney cancer gene suppresses growth and functions
with VHL in a common pathway. Oncogene, 21(22), 3507–3516.
Gemmill, Robert M, Lee, J. P., Chamovitz, D. A., Segal, D., Hooper, J. E., & Drabkin, H. A.
(2005). Growth suppression induced by the TRC8 hereditary kidney cancer gene is
dependent upon JAB1/CSN5. Oncogene, 24(21), 3503–3511.
Glover, T W, Coyle-Morris, J. F., Li, F. P., Brown, R. S., Berger, C. S., Gemmill, R. M., &
Hecht, F. (1988). Translocation t(3;8)(p14.2;q24.1) in renal cell carcinoma affects
expression of the common fragile site at 3p14(FRA3B) in lymphocytes. Cancer Genetics
and Cytogenetics, 31(1), 69–73.
Glover, Thomas W, Wilson, T. E., & Arlt, M. F. (2017). Fragile sites in cancer: more than meets
the eye. Nature Reviews. Cancer, 17(8), 489–501.
Goldberg, Y., Kedar, I., Kariiv, R., Halpern, N., Plesser, M., Hubert, A., Kaduri, L., et al. (2014).
Lynch Syndrome in high risk Ashkenazi Jews in Israel. Familial Cancer, 13(1), 65–73.
Gooden, M. J. M., de Bock, G. H., Leffers, N., Daemen, T., & Nijman, H. W. (2011). The
prognostic influence of tumour-infiltrating lymphocytes in cancer: a systematic review
with meta-analysis. British Journal of Cancer, 105(1), 93–103.
Goss, K. H., & Groden, J. (2000). Biology of the adenomatous polyposis coli tumor suppressor.
Journal of Clinical Oncology, 18(9), 1967–1979.
Graham, M. J., Lee, R. G., Brandt, T. A., Tai, L.-J., Fu, W., Peralta, R., Yu, R., et al. (2017).
Cardiovascular and metabolic effects of ANGPTL3 antisense oligonucleotides. The New
England Journal of Medicine, 377(3), 222–232.
Greenson, J. K., Bonner, J. D., Ben-Yzhak, O., Cohen, H. I., Miselevich, I., Resnick, M. B.,
Trougouboff, P., et al. (2003). Phenotype of microsatellite unstable colorectal
carcinomas: Well-differentiated and focally mucinous tumors and the absence of dirty
necrosis correlate with microsatellite instability. The American Journal of Surgical
Pathology, 27(5), 563–570.
Greenson, J. K., Huang, S.-C., Herron, C., Moreno, V., Bonner, J. D., Tomsho, L. P., Ben-Izhak,
O., et al. (2009). Pathologic predictors of microsatellite instability in colorectal cancer.
The American Journal of Surgical Pathology, 33(1), 126–133.
Grossmann, V., Höckner, M., Karmous-Benailly, H., Liang, D., Puttinger, R., Quadrelli, R.,
Röthlisberger, B., et al. (2010). Parental origin of apparently balanced de novo complex
chromosomal rearrangements investigated by microdissection, whole genome
amplification, and microsatellite-mediated haplotype analysis. Clinical Genetics, 78(6),
548–553.
Gruber, S. B. (2006). New developments in Lynch syndrome (hereditary nonpolyposis colorectal
cancer) and mismatch repair gene testing. Gastroenterology, 130(2), 577–587.
Gruber, S. B., Moreno, V., Rozek, L. S., Rennerts, H. S., Lejbkowicz, F., Bonner, J. D.,
Greenson, J. K., et al. (2007). Genetic variation in 8q24 associated with risk of colorectal
cancer. Cancer Biology & Therapy, 6(7), 1143–1147.

109
Guinney, J., Dienstmann, R., Wang, X., de Reyniès, A., Schlicker, A., Soneson, C., Marisa, L., et
al. (2015). The consensus molecular subtypes of colorectal cancer. Nature Medicine,
21(11), 1350–1356.
Gullapalli, R. R., Desai, K. V., Santana-Santos, L., Kant, J. A., & Becich, M. J. (2012). Next
generation sequencing in clinical medicine: Challenges and lessons for pathology and
biomedical informatics. Journal of pathology informatics, 3, 40.
Guo, S. W., & Xiong, M. (1997). Estimating the age of mutant disease alleles based on linkage
disequilibrium. Human Heredity, 47(6), 315–337.
Hanahan, D., & Weinberg, R. A. (2011). Hallmarks of cancer: the next generation. Cell, 144(5),
646–674.
Hassold, T., & Hunt, P. (2001). To err (meiotically) is human: the genesis of human aneuploidy.
Nature Reviews. Genetics, 2(4), 280–291.
Huh, J. W., Lee, J. H., & Kim, H. R. (2012). Prognostic significance of tumor-infiltrating
lymphocytes for patients with colorectal cancer. Archives of Surgery, 147(4), 366–372.
Hundal, J., Miller, C. A., Griffith, M., Griffith, O. L., Walker, J., Kiwala, S., Graubert, A., et al.
(2016). Cancer immunogenomics: computational neoantigen identification and vaccine
design. Cold Spring Harbor Symposia on Quantitative Biology, 81, 105–111.
Huyghe, J. R., Bien, S. A., Harrison, T. A., Kang, H. M., Chen, S., Schmit, S. L., Conti, D. V., et
al. (2019). Discovery of common and rare genetic risk variants for colorectal cancer.
Nature Genetics, 51(1), 76–87.
Inagaki, H., Kato, T., Tsutsumi, M., Ouchi, Y., Ohye, T., & Kurahashi, H. (2016). Palindrome-
Mediated Translocations in Humans: A New Mechanistic Model for Gross Chromosomal
Rearrangements. Frontiers in genetics, 7, 125.
Ionov, Y., Peinado, M. A., Malkhosyan, S., Shibata, D., & Perucho, M. (1993). Ubiquitous
somatic mutations in simple repeated sequences reveal a new mechanism for colonic
carcinogenesis. Nature, 363(6429), 558–561.
Jabbour, E., Cortes, J. E., Ghanem, H., O’Brien, S., & Kantarjian, H. M. (2008). Targeted
therapy in chronic myeloid leukemia. Expert Review of Anticancer Therapy, 8(1), 99–
110.
Jass, J. R. (2000). Pathology of hereditary nonpolyposis colorectal cancer. Annals of the New
York Academy of Sciences, 910, 62–73; discussion 73.
Jass, J. R., Atkin, W. S., Cuzick, J., Bussey, H. J. R., Morson, B. C., Northover, J. M. A., &
Todd, I. P. (2002). The grading of rectal cancer: historical perspectives and a multivariate
analysis of 447 cases. Histopathology, 41(3A), 59–81.
de Jong, M. M., Nolte, I. M., te Meerman, G. J., van der Graaf, W. T. A., de Vries, E. G. E.,
Sijmons, R. H., Hofstra, R. M. W., et al. (2002). Low-penetrance genes and their
involvement in colorectal cancer susceptibility. Cancer Epidemiology, Biomarkers &
Prevention, 11(11), 1332–1352.
Kashimura, S., Saze, Z., Terashima, M., Soeta, N., Ohtani, S., Osuka, F., Kogure, M., et al.
(2012). CD83(+) dendritic cells and Foxp3(+) regulatory T cells in primary lesions and
regional lymph nodes are inversely correlated with prognosis of gastric cancer. Gastric
Cancer, 15(2), 144–153.
Kato, T., Franconi, C. P., Sheridan, M. B., Hacker, A. M., Inagakai, H., Glover, T. W., Arlt, M.
F., et al. (2014). Analysis of the t(3;8) of hereditary renal cell carcinoma: a palindrome-
mediated translocation. Cancer genetics, 207(4), 133–140.
Kato, T., Inagaki, H., Kogo, H., Ohye, T., Yamada, K., Emanuel, B. S., & Kurahashi, H. (2008).

110
Two different forms of palindrome resolution in the human genome: deletion or
translocation. Human Molecular Genetics, 17(8), 1184–1191.
Kato, T., Yamada, K., Inagaki, H., Kogo, H., Ohye, T., Emanuel, B. S., & Kurahashi, H. (2007).
Age has no effect on de novo constitutional t(11;22) translocation frequency in sperm.
Fertility and Sterility, 88(5), 1446–1448.
Kemp, Z., Thirlwell, C., Sieber, O., Silver, A., & Tomlinson, I. (2004). An update on the
genetics of colorectal cancer. Human Molecular Genetics, 13 Spec No 2, R177-85.
Kim, M., Grimmig, T., Grimm, M., Lazariotou, M., Meier, E., Rosenwald, A., Tsaur, I., et al.
(2013). Expression of Foxp3 in colorectal cancer but not in Treg cells correlates with
disease progression in patients with colorectal cancer. Plos One, 8(1), e53630.
Kimura, M., & Ohta, T. (1973). The age of a neutral mutant persisting in a finite population.
Genetics, 75(1), 199–212.
Kinzler, K. W., & Vogelstein, B. (n.d.). Colorectal tumors. In K. W. Kinzler & B. Vogelstein
(Eds.), The genetic basis of human cancer (2nd ed., pp. 583–612). New York: McGraw-
Hill.
Knudson, A. G. (1993). Introduction to the genetics of primary renal tumors in children. Medical
and Pediatric Oncology, 21(3), 193–198.
Koboldt, D. C., Steinberg, K. M., Larson, D. E., Wilson, R. K., & Mardis, E. R. (2013). The
next-generation sequencing revolution and its impact on genomics. Cell, 155(1), 27–38.
Kocián, P., Šedivcová, M., Drgáč, J., Cerná, K., Hoch, J., Kodet, R., Bartůňková, J., et al.
(2011). Tumor-infiltrating lymphocytes and dendritic cells in human colorectal cancer:
their relationship to KRAS mutational status and disease recurrence. Human
Immunology, 72(11), 1022–1028.
Koduru, P. R. K., & Chaganti, R. S. K. (1989). Meiotic chromosome segregation in human
t(11;22)(q23;q11) carriers: a theoretical consideration. Genome, 32(1), 24–29.
Kohlmann, W., & Gruber, S. B. (1993). Lynch Syndrome. In R. A. Pagon, M. P. Adam, H. H.
Ardinger, S. E. Wallace, A. Amemiya, L. J. Bean, T. D. Bird, et al. (Eds.),
GeneReviews(®). Seattle (WA): University of Washington, Seattle.
Kurahashi, H, Inagaki, H., Ohye, T., Kogo, H., Tsutsumi, M., Kato, T., Tong, M., et al. (2010).
The constitutional t(11;22): implications for a novel mechanism responsible for gross
chromosomal rearrangements. Clinical Genetics, 78(4), 299–309.
Kurahashi, Hiroki, Inagaki, H., Kato, T., Hosoba, E., Kogo, H., Ohye, T., Tsutsumi, M., et al.
(2009). Impaired DNA replication prompts deletions within palindromic sequences, but
does not induce translocations in human cells. Human Molecular Genetics, 18(18), 3397–
3406.
Kurahashi, Hiroki, Inagaki, H., Ohye, T., Kogo, H., Kato, T., & Emanuel, B. S. (2006a).
Palindrome-mediated chromosomal translocations in humans. DNA Repair, 5(9–10),
1136–1145.
Kurahashi, Hiroki, Inagaki, H., Ohye, T., Kogo, H., Kato, T., & Emanuel, B. S. (2006b).
Chromosomal translocations mediated by palindromic DNA. Cell Cycle, 5(12), 1297–
1303.
Lagerstedt-Robinson, K., Rohlin, A., Aravidis, C., Melin, B., Nordling, M., Stenmark-Askmalm,
M., Lindblom, A., et al. (2016). Mismatch repair gene mutation spectrum in the Swedish
Lynch syndrome population. Oncology Reports, 36(5), 2823–2835.
Laken, S. J., Petersen, G. M., Gruber, S. B., Oddoux, C., Ostrer, H., Giardiello, F. M., Hamilton,
S. R., et al. (1997). Familial colorectal cancer in Ashkenazim due to a hypermutable tract

111
in APC. Nature Genetics, 17(1), 79–83.
Lavotshkin, S., Jalas, J. R., Torisu-Itakura, H., Ozao-Choy, J., Lee, J. H., Sim, M. S.,
Stojadinovic, A., et al. (2015). Immunoprofiling for prognostic assessment of colon
cancer: a novel complement to ultrastaging. Journal of Gastrointestinal Surgery, 19(6),
999–1006.
Le, D. T., Uram, J. N., Wang, H., Bartlett, B. R., Kemberling, H., Eyring, A. D., Skora, A. D., et
al. (2015). PD-1 Blockade in Tumors with Mismatch-Repair Deficiency. The New
England Journal of Medicine, 372(26), 2509–2520.
Lee, W.-S., Park, S., Lee, W. Y., Yun, S. H., & Chun, H.-K. (2010). Clinical impact of tumor-
infiltrating lymphocytes for survival in stage II colon cancer. Cancer, 116(22), 5188–
5199.
Leth-Larsen, R., Lund, R. R., & Ditzel, H. J. (2010). Plasma membrane proteomics and its
application in clinical cancer biomarker discovery. Molecular & Cellular Proteomics,
9(7), 1369–1382.
Ley, T. J., Mardis, E. R., Ding, L., Fulton, B., McLellan, M. D., Chen, K., Dooling, D., et al.
(2008). DNA sequencing of a cytogenetically normal acute myeloid leukaemia genome.
Nature, 456(7218), 66–72.
Li, F. P., Decker, H. J., Zbar, B., Stanton, V. P., Kovacs, G., Seizinger, B. R., Aburatani, H., et
al. (1993). Clinical and genetic studies of renal cell carcinomas in a family with a
constitutional chromosome 3;8 translocation. Genetics of familial renal carcinoma.
Annals of Internal Medicine, 118(2), 106–111.
Li, J. Z., Absher, D. M., Tang, H., Southwick, A. M., Casto, A. M., Ramachandran, S., Cann, H.
M., et al. (2008). Worldwide human relationships inferred from genome-wide patterns of
variation. Science, 319(5866), 1100–1104.
Li, W., Wang, L., Katoh, H., Liu, R., Zheng, P., & Liu, Y. (2011). Identification of a tumor
suppressor relay between the FOXP3 and the Hippo pathways in breast and prostate
cancers. Cancer Research, 71(6), 2162–2171.
Linehan, W. M. (2012). Genetic basis of kidney cancer: role of genomics for the development of
disease-based therapeutics. Genome Research, 22(11), 2089–2100.
Lipson, E. J., Sharfman, W. H., Drake, C. G., Wollner, I., Taube, J. M., Anders, R. A., Xu, H., et
al. (2013). Durable cancer regression off-treatment and effective reinduction therapy with
an anti-PD-1 antibody. Clinical Cancer Research, 19(2), 462–468.
Lynch, T. J., Bell, D. W., Sordella, R., Gurubhagavatula, S., Okimoto, R. A., Brannigan, B. W.,
Harris, P. L., et al. (2004). Activating mutations in the epidermal growth factor receptor
underlying responsiveness of non-small-cell lung cancer to gefitinib. The New England
Journal of Medicine, 350(21), 2129–2139.
Maizels, N. (2006). Dynamic roles for G4 DNA in the biology of eukaryotic cells. Nature
Structural & Molecular Biology, 13(12), 1055–1059.
Mardis, E. R. (2008). Next-generation DNA sequencing methods. Annual Review of Genomics
and Human Genetics, 9, 387–402.
Markowitz, S. D., & Bertagnolli, M. M. (2009). Molecular origins of cancer: Molecular basis of
colorectal cancer. The New England Journal of Medicine, 361(25), 2449–2460.
Marra, G., & Boland, C. R. (1995). Hereditary nonpolyposis colorectal cancer: the syndrome, the
genes, and historical perspectives. Journal of the National Cancer Institute, 87(15),
1114–1125.
Marshall, S., Das, R., Pirooznia, M., & Elhaik, E. (2016). Reconstructing Druze population

112
history. Scientific reports, 6, 35837.
McAllister, S. S., & Weinberg, R. A. (2010). Tumor-host interactions: a far-reaching
relationship. Journal of Clinical Oncology, 28(26), 4022–4028.
McDonnell, K. J., Chemler, J. A., Bartels, P. L., O’Brien, E., Marvin, M. L., Ortega, J., Stern, R.
H., et al. (2018). A human MUTYH variant linking colonic polyposis to redox
degradation of the [4Fe4S]2+ cluster. Nature Chemistry, 10(8), 873–880.
McPeek, M. S., & Strahs, A. (1999). Assessment of linkage disequilibrium by the decay of
haplotype sharing, with application to fine-scale genetic mapping. American Journal of
Human Genetics, 65(3), 858–875.
Meléndez, B., Rodríguez-Perales, S., Martínez-Delgado, B., Otero, I., Robledo, M., Martínez-
Ramírez, A., Ruiz-Llorente, S., et al. (2003). Molecular study of a new family with
hereditary renal cell carcinoma and a translocation t(3;8)(p13;q24.1). Human Genetics,
112(2), 178–185.
Miller, S. A., Dykes, D. D., & Polesky, H. F. (1988). A simple salting out procedure for
extracting DNA from human nucleated cells. Nucleic Acids Research, 16(3), 1215.
Misteli, T. (2004). Spatial positioning; a new dimension in genome function. Cell, 119(2), 153–
156.
Moyret-Lalle, C., Falette, N., Grelier, G., & Puisieux, A. (2008). [Tumour genomics: an unstable
landscape]. Bulletin du Cancer, 95(10), 923–930.
Nakayama, K. I., & Nakayama, K. (2006). Ubiquitin ligases: cell-cycle control and cancer.
Nature Reviews. Cancer, 6(5), 369–381.
Niell, B. L., Long, J. C., Rennert, G., & Gruber, S. B. (2003). Genetic anthropology of the
colorectal cancer-susceptibility allele APC I1307K: evidence of genetic drift within the
Ashkenazim. American Journal of Human Genetics, 73(6), 1250–1260.
Nowell, P. C., Rowley, J. D., & Knudson, A. G. (1998). Cancer genetics, cytogenetics--defining
the enemy within. Nature Medicine, 4(10), 1107–1111.
O’Connell, J., Gurdasani, D., Delaneau, O., Pirastu, N., Ulivi, S., Cocca, M., Traglia, M., et al.
(2014). A general approach for haplotype phasing across the full spectrum of relatedness.
PLoS Genetics, 10(4), e1004234.
Ohtani, H. (2007). Focus on TILs: prognostic significance of tumor infiltrating lymphocytes in
human colorectal cancer. Cancer immunity : a journal of the Academy of Cancer
Immunology, 7, 4.
Ohye, T., Inagaki, H., Kogo, H., Tsutsumi, M., Kato, T., Tong, M., Macville, M. V. E., et al.
(2010). Paternal origin of the de novo constitutional t(11;22)(q23;q11). European Journal
of Human Genetics, 18(7), 783–787.
Paez, J. G., Jänne, P. A., Lee, J. C., Tracy, S., Greulich, H., Gabriel, S., Herman, P., et al. (2004).
EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy.
Science, 304(5676), 1497–1500.
Pagano, M. (2004). Control of DNA synthesis and mitosis by the Skp2-p27-Cdk1/2 axis.
Molecular Cell, 14(4), 414–416.
Pao, W., Miller, V., Zakowski, M., Doherty, J., Politi, K., Sarkaria, I., Singh, B., et al. (2004).
EGF receptor gene mutations are common in lung cancers from “never smokers” and are
associated with sensitivity of tumors to gefitinib and erlotinib. Proceedings of the
National Academy of Sciences of the United States of America, 101(36), 13306–13311.
Pardoll, D. M. (2012). The blockade of immune checkpoints in cancer immunotherapy. Nature
Reviews. Cancer, 12(4), 252–264.

113
Pearson, C. E., Nichol Edamura, K., & Cleary, J. D. (2005). Repeat instability: mechanisms of
dynamic mutations. Nature Reviews. Genetics, 6(10), 729–742.
Peltomäki, P., Aaltonen, L. A., Sistonen, P., Pylkkänen, L., Mecklin, J. P., Järvinen, H., Green, J.
S., et al. (1993). Genetic mapping of a locus predisposing to human colorectal cancer.
Science, 260(5109), 810–812.
Peters, U., Hutter, C. M., Hsu, L., Schumacher, F. R., Conti, D. V., Carlson, C. S., Edlund, C. K.,
et al. (2012). Meta-analysis of new genome-wide association studies of colorectal cancer
risk. Human Genetics, 131(2), 217–234.
Phipps, A. I., Limburg, P. J., Baron, J. A., Burnett-Hartman, A. N., Weisenberger, D. J., Laird, P.
W., Sinicrope, F. A., et al. (2015). Association between molecular subtypes of colorectal
cancer and patient survival. Gastroenterology, 148(1), 77–87.e2.
Poland, K. S., Azim, M., Folsom, M., Goldfarb, R., Naeem, R., Korch, C., Drabkin, H. A., et al.
(2007). A constitutional balanced t(3;8)(p14;q24.1) translocation results in disruption of
the TRC8 gene and predisposition to clear cell renal cell carcinoma. Genes,
Chromosomes & Cancer, 46(9), 805–812.
Poynter, J. N., Cooney, K. A., Bonner, J. D., White, K. A., Tomsho, L. P., Rennert, G., &
Gruber, S. B. (2006). APC I1307K and the risk of prostate cancer. Cancer Epidemiology,
Biomarkers & Prevention, 15(3), 468–473.
Poynter, J. N., Gruber, S. B., Higgins, P. D. R., Almog, R., Bonner, J. D., Rennert, H. S., Low,
M., et al. (2005). Statins and the risk of colorectal cancer. The New England Journal of
Medicine, 352(21), 2184–2192.
Provine, W. B. (2004). Ernst Mayr: Genetics and speciation. Genetics, 167(3), 1041–1046.
Purcell, S., Neale, B., Todd-Brown, K., Thomas, L., Ferreira, M. A. R., Bender, D., Maller, J., et
al. (2007). PLINK: a tool set for whole-genome association and population-based linkage
analyses. American Journal of Human Genetics, 81(3), 559–575.
Raghavan, S. C., & Lieber, M. R. (2006). DNA structures at chromosomal translocation sites.
Bioessays: News and Reviews in Molecular, Cellular and Developmental Biology, 28(5),
480–494.
Rannala, B., & Bertorelle, G. (2001). Using linked markers to infer the age of a mutation.
Human Mutation, 18(2), 87–100.
Rannala, B., & Reeve, J. P. (2001). High-resolution multipoint linkage-disequilibrium mapping
in the context of a human genome sequence. American Journal of Human Genetics,
69(1), 159–178.
Raskin, L., Schwenter, F., Freytsis, M., Tischkowitz, M., Wong, N., Chong, G., Narod, S. A., et
al. (2011). Characterization of two Ashkenazi Jewish founder mutations in MSH6 gene
causing Lynch syndrome. Clinical Genetics, 79(6), 512–522.
Raymond, V. M., Herron, C. M., Giordano, T. J., & Gruber, S. B. (2012). Familial renal cancer
as an indicator of hereditary leiomyomatosis and renal cell cancer syndrome. Familial
Cancer, 11(1), 115–121.
Reeve, J. P., & Rannala, B. (2002). DMLE+: Bayesian linkage disequilibrium gene mapping.
Bioinformatics, 18(6), 894–895.
Robins, H. S., Campregher, P. V., Srivastava, S. K., Wacher, A., Turtle, C. J., Kahsai, O.,
Riddell, S. R., et al. (2009). Comprehensive assessment of T-cell receptor beta-chain
diversity in alphabeta T cells. Blood, 114(19), 4099–4107.
Robins, H. S., Srivastava, S. K., Campregher, P. V., Turtle, C. J., Andriesen, J., Riddell, S. R.,
Carlson, C. S., et al. (2010). Overlap and effective size of the human CD8+ T cell

114
receptor repertoire. Science Translational Medicine, 2(47), 47ra64.
Rodríguez-Perales, S., Meléndez, B., Gribble, S. M., Valle, L., Carter, N. P., Santamaría, I.,
Conde, L., et al. (2004). Cloning of a new familial t(3;8) translocation associated with
conventional renal cell carcinoma reveals a 5 kb microdeletion and no gene involved in
the rearrangement. Human Molecular Genetics, 13(9), 983–990.
Roelands, J., Kuppen, P. J. K., Vermeulen, L., Maccalli, C., Decock, J., Wang, E., Marincola, F.
M., et al. (2017). Immunogenomic classification of colorectal cancer and therapeutic
implications. International Journal of Molecular Sciences, 18(10).
Roix, J. J., McQueen, P. G., Munson, P. J., Parada, L. A., & Misteli, T. (2003). Spatial proximity
of translocation-prone gene loci in human lymphomas. Nature Genetics, 34(3), 287–291.
Rossari, F., Minutolo, F., Orciulo, E. (2018). Past, present, and future of Bcr-Abl inhibitors: from
chemical development to clinical efficace. Journal and Hematology & Oncology, 11(14),
1-14.
Rozek, L. S., Schmit, S. L., Greenson, J. K., Tomsho, L. P., Rennert, H. S., Rennert, G., &
Gruber, S. B. (2016). Tumor-Infiltrating Lymphocytes, Crohn’s-Like Lymphoid
Reaction, and Survival From Colorectal Cancer. Journal of the National Cancer Institute,
108(8).
Rozen, P., Naiman, T., Strul, H., Taussky, P., Karminsky, N., Shomrat, R., Samuel, Z., et al.
(2002). Clinical and screening implications of the I1307K adenomatous polyposis coli
gene variant in Israeli Ashkenazi Jews with familial colorectal neoplasia. Cancer, 94(10),
2561–2568.
Saif, M. W., & Chu, E. (2010). Biology of colorectal cancer. Cancer Journal, 16(3), 196–201.
von Salomé, J., Liu, T., Keihäs, M., Morak, M., Holinski-Feder, E., Berry, I. R., Moilanen, J. S.,
et al. (2017). Haplotype analysis suggest that the MLH1 c.2059C > T mutation is a
Swedish founder mutation. Familial Cancer, 1–7.
Sancho, E., Batlle, E., & Clevers, H. (2004). Signaling pathways in intestinal development and
cancer. Annual Review of Cell and Developmental Biology, 20, 695–723.
Sanmamed, M. F., & Chen, L. (2018). A paradigm shift in cancer immunotherapy: from
enhancement to normalization. Cell, 175(2), 313–326.
Sanz-Pamplona, R., Cordero, D., Berenguer, A., Lejbkowicz, F., Rennert, H., Salazar, R.,
Biondo, S., et al. (2011). Gene expression differences between colon and rectum tumors.
Clinical Cancer Research, 17(23), 7303–7312.
Schmit, S. L., Edlund, C. K., Schumacher, F. R., Gong, J., Harrison, T. A., Huyghe, J. R., Qu, C.,
et al. (2019). Novel common genetic susceptibility loci for colorectal cancer. Journal of
the National Cancer Institute, 111(2), 146–157.
Schumacher, F. R., Schmit, S. L., Jiao, S., Edlund, C. K., Wang, H., Zhang, B., Hsu, L., et al.
(2015). Genome-wide association study of colorectal cancer identifies six new
susceptibility loci. Nature Communications, 6, 7138.
Segal, N.H., Parsons, D.W., Peggs, K.S., Velculescu, V., Kinzler, K.W., Vogelstein, B., Allison,
J.P. (2008). Epitope landscape in breast and colorectal cancer. Cancer Research, 68(3),
889-892.
Serratì, S., De Summa, S., Pilato, B., Petriella, D., Lacalamita, R., Tommasi, S., & Pinto, R.
(2016). Next-generation sequencing: advances and applications in cancer diagnosis.
OncoTargets and therapy, 9, 7355–7365.
Shaikh, T. H., Budarf, M. L., Celle, L., Zackai, E. H., & Emanuel, B. S. (1999). Clustered 11q23
and 22q11 breakpoints and 3:1 meiotic malsegregation in multiple unrelated t(11;22)

115
families. American Journal of Human Genetics, 65(6), 1595–1607.
Siegel, R. L., Miller, K, D., Fedewa, S. A., et al. (2017). Colorectal cancer statistics CA. Cancer
J. Clin. 67(3):177–193.
Simpson, J. A. D., Al-Attar, A., Watson, N. F. S., Scholefield, J. H., Ilyas, M., & Durrant, L. G.
(2010). Intratumoral T cell infiltration, MHC class I and STAT1 as biomarkers of good
prognosis in colorectal cancer. Gut, 59(7), 926–933.
Slamon, D. J., Clark, G. M., Wong, S. G., Levin, W. J., Ullrich, A., & McGuire, W. L. (1987).
Human breast cancer: correlation of relapse and survival with amplification of the HER-
2/neu oncogene. Science, 235(4785), 177–182.
Slamon, D. J., Leyland-Jones, B., Shak, S., Fuchs, H., Paton, V., Bajamonde, A., Fleming, T., et
al. (2001). Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic
breast cancer that overexpresses HER2. The New England Journal of Medicine, 344(11),
783–792.
Slatkin, M., & Rannala, B. (1997). Estimating the age of alleles by use of intraallelic variability.
American Journal of Human Genetics, 60(2), 447–458.
Slatkin, M., & Rannala, B. (2000). Estimating allele age. Annual Review of Genomics and
Human Genetics, 1, 225–249.
Smyrk, T. C., Watson, P., Kaul, K., & Lynch, H. T. (2001). Tumor-infiltrating lymphocytes are a
marker for microsatellite instability in colorectal carcinoma. Cancer, 91(12), 2417–2422.
Song, M., Chen, D., Lu, B., Wang, C., Zhang, J., Huang, L., Wang, X., et al. (2013). PTEN loss
increases PD-L1 protein expression and affects the correlation between PD-L1 expression
and clinical parameters in colorectal cancer. Plos One, 8(6), e65821.
Sun, X., Feng, Z., Wang, Y., Qu, Y., & Gai, Y. (2017). Expression of Foxp3 and its prognostic
significance in colorectal cancer. International Journal of Immunopathology and
Pharmacology, 30(2), 201–206.
Susswein, L. R., Marshall, M. L., Nusbaum, R., Vogel Postula, K. J., Weissman, S. M.,
Yackowski, L., Vaccari, E. M., et al. (2016). Pathogenic and likely pathogenic variant
prevalence among the first 10,000 patients referred for next-generation cancer panel
testing. Genetics in Medicine, 18(8), 823–832.
Tapia-Páez, I., Kost-Alimova, M., Hu, P., Roe, B. A., Blennow, E., Fedorova, L., Imreh, S., et al.
(2001). The position of t(11;22)(q23;q11) constitutional translocation breakpoint is
conserved among its carriers. Human Genetics, 109(2), 167–177.
Templeton, A. R. (1980). The theory of speciation via the founder principle. Genetics, 94(4),
1011–1038.
Tenesa, A., & Dunlop, M. G. (2009). New insights into the aetiology of colorectal cancer from
genome-wide association studies. Nature Reviews. Genetics, 10(6), 353–358.
Thanki, K., Nicholls, M. E., Gajjar, A., Senagore, A. J., Qiu, S., Szabo, C., Hellmich, M. R., et
al. (2017). Consensus Molecular Subtypes of Colorectal Cancer and their Clinical
Implications. International biological and biomedical journal, 3(3), 105–111.
Thomas, N. S., Durkie, M., Potts, G., Sandford, R., Van Zyl, B., Youings, S., Dennis, N. R., et
al. (2006). Parental and chromosomal origins of microdeletion and duplication syndromes
involving 7q11.23, 15q11-q13 and 22q11. European Journal of Human Genetics, 14(7),
831–837.
Thomas, N. S., Morris, J. K., Baptista, J., Ng, B. L., Crolla, J. A., & Jacobs, P. A. (2010). De
novo apparently balanced translocations in man are predominantly paternal in origin and
associated with a significant increase in paternal age. Journal of Medical Genetics, 47(2),

116
112–115.
Tomsic, J., Liyanarachchi, S., Hampel, H., Morak, M., Thomas, B. C., Raymond, V. M.,
Chittenden, A., et al. (2012). An American founder mutation in MLH1. International
Journal of Cancer, 130(9), 2088–2095.
Topalian, S. L., Hodi, F. S., Brahmer, J. R., Gettinger, S. N., Smith, D. C., McDermott, D. F.,
Powderly, J. D., et al. (2012). Safety, activity, and immune correlates of anti-PD-1
antibody in cancer. The New England Journal of Medicine, 366(26), 2443–2454.
Tosi, M. R., & Tugnoli, V. (2005). Cholesteryl esters in malignancy. Clinica Chimica Acta,
359(1–2), 27–45.
Triulzi, T., Tagliabue, E., Balsari, A., & Casalini, P. (2013). FOXP3 expression in tumor cells
and implications for cancer progression. Journal of Cellular Physiology, 228(1), 30–35.
Valencia, A., & Hidalgo, M. (2012). Getting personalized cancer genome analysis into the clinic:
the challenges in bioinformatics. Genome Medicine, 4(7), 61.
Valentini, A. M., Di Pinto, F., Cariola, F., Guerra, V., Giannelli, G., Caruso, M. L., & Pirrelli, M.
(2018). PD-L1 expression in colorectal cancer defines three subsets of tumor immune
microenvironments. Oncotarget, 9(9), 8584–8596.
Valle, L., Cascón, A., Melchor, L., Otero, I., Rodríguez-Perales, S., Sánchez, L., Cruz Cigudosa,
J., et al. (2005). About the origin and development of hereditary conventional renal cell
carcinoma in a four-generation t(3;8)(p14.1;q24.23) family. European Journal of Human
Genetics, 13(5), 570–578.
Vardi-Saliternik, R., Friedlander, Y., & Cohen, T. (2002). Consanguinity in a population sample
of Israeli Muslim Arabs, Christian Arabs and Druze. Annals of Human Biology, 29(4),
422–431.
Vazquez, M., de la Torre, V., & Valencia, A. (2012). Chapter 14: Cancer genome analysis. PLoS
Computational Biology, 8(12), e1002824.
Vilar, E., & Gruber, S. B. (2010). Microsatellite instability in colorectal cancer-the stable
evidence. Nature Reviews. Clinical Oncology, 7(3), 153–162.
Vogelstein, B, Fearon, E. R., Hamilton, S. R., Kern, S. E., Preisinger, A. C., Leppert, M.,
Nakamura, Y., et al. (1988). Genetic alterations during colorectal-tumor development.
The New England Journal of Medicine, 319(9), 525–532.
Vogelstein, Bert, Papadopoulos, N., Velculescu, V. E., Zhou, S., Diaz, L. A., & Kinzler, K. W.
(2013). Cancer genome landscapes. Science, 339(6127), 1546–1558.
Walther, A., Johnstone, E., Swanton, C., Midgley, R., Tomlinson, I., & Kerr, D. (2009). Genetic
prognostic and predictive markers in colorectal cancer. Nature Reviews. Cancer, 9(7),
489–499.
Wang, G., & Vasquez, K. M. (2014). Impact of alternative DNA structures on DNA damage,
DNA repair, and genetic instability. DNA Repair, 19, 143–151.
Winerdal, M. E., Marits, P., Winerdal, M., Hasan, M., Rosenblatt, R., Tolf, A., Selling, K., et al.
(2011). FOXP3 and survival in urinary bladder cancer. BJU International, 108(10),
1672–1678.
Wright, S. (1931). Evolution in Mendelian Populations. Genetics, 16(2), 97–159.
Yarchoan, M., Hopkins, A., & Jaffee, E. M. (2017). Tumor Mutational Burden and Response
Rate to PD-1 Inhibition. The New England Journal of Medicine, 377(25), 2500–2501.
Yuan, J., Hegde, P. S., Clynes, R., Foukas, P. G., Harari, A., Kleen, T. O., Kvistborg, P., et al.
(2016). Novel technologies and emerging biomarkers for personalized cancer
immunotherapy. Journal for immunotherapy of cancer, 4, 3.

117
Zackai, E. H., & Emanuel, B. S. (1980). Site-specific reciprocal translocation, t(11;22)
(q23;q11), in several unrelated families with 3:1 meiotic disjunction. American Journal
of Medical Genetics, 7(4), 507–521.
Zanesi, N, Fidanza, V., Fong, L. Y., Mancini, R., Druck, T., Valtieri, M., Rüdiger, T., et al.
(2001). The tumor spectrum in FHIT-deficient mice. Proceedings of the National
Academy of Sciences of the United States of America, 98(18), 10250–10255.
Zanesi, Nicola, Mancini, R., Sevignani, C., Vecchione, A., Kaou, M., Valtieri, M., Calin, G. A.,
et al. (2005). Lung cancer susceptibility in Fhit-deficient mice is increased by Vhl
haploinsufficiency. Cancer Research, 65(15), 6576–6582.
Zanke, B. W., Greenwood, C. M. T., Rangrej, J., Kustra, R., Tenesa, A., Farrington, S. M.,
Prendergast, J., et al. (2007). Genome-wide association scan identifies a colorectal cancer
susceptibility locus on chromosome 8q24. Nature Genetics, 39(8), 989–994.
Zewde, M., Kiyotani, K., Park, J.-H., Fang, H., Yap, K. L., Yew, P. Y., Alachkar, H., et al.
(2018). The era of immunogenomics/immunopharmacogenomics. Journal of Human
Genetics, 63(8), 865–875.
Zhang, J., Nichols, K. E., & Downing, J. R. (2016). Germline mutations in predisposition genes
in pediatric cancer. The New England Journal of Medicine, 374(14), 1391.

Abstract (if available)

Abstract In this dissertation, we elucidated the genomic, transcriptomic and immunologic landscapes of human cancers, through laboratory studies, statistical genetics analyses, and bioinformatics methods, focusing on colorectal cancer and clear cell renal cell carcinoma. The first project studied tumor infiltrating lymphocytes (TILs) and clonal analysis of the hypervariable region of T-Cell β receptors in order to quantify the TCR repertoire of individual colorectal cancers from a large, population-based study of colorectal cancer in Israel. Using clinical, epidemiologic, genomic and transcriptomic data from the Molecular Epidemiology of Colorectal Cancer (MECC) Study, colorectal cancers were classified according to consensus molecular subtypes (CMS) where gene expression data were available, and interpreted from the perspective of the host immune response. The second project focused on a subset of Druze individuals with Lynch syndrome from Israel who harbor a germline MSH2*c.705delA mutation. In this study, a recurrent mutation in MSH2 was studied as a potential founder mutation by characterizing the allele sharing identical-by-descent among Druze carriers. The genetic closeness of this community in conjunction with the cryptic familial relationships of independently ascertained families limits the statistical analysis and interpretation of this recurrent mutation as a “founder” mutation. Therefore, this private mutation that is known to be an important cause of Lynch Syndrome in the Druze population was characterized from a clinical, immunologic and pathologic perspective. For the third project, we elucidated the pathobiological consequences and the parental origin of a de novo constitutional PATRR-mediated t(3

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Integrative genomic and epigenomic analysis of human cancer

PDF

Applying multi-omics in cancer liquid biopsy for improved patient monitoring and biomarker discovery

PDF

Investigating the complexity of the tumor microenvironment's role in drug response

PDF

Molecular classification of breast cancer specimens from tissue morphology

PDF

Fish consumption and risk of colorectal cancer

PDF

A multi-center prospective cohort study of the diagnostic yield and patient experience of multiplex gene panel testing for hereditary cancer risk

PDF

Understanding prostate cancer genetic susceptibility and chromatin regulation

PDF

The accumulation of somatic mutations in humans with age

PDF

Genome sequencing and transcriptome analysis of the phenotypically plastic spadefoot toads

PDF

Identification of CBP/FOXM1 as a molecular target in triple negative breast cancer

PDF

Genetic studies of cancer in populations of African ancestry and Latinos

PDF

Molecular signature of aggressive disease and clonal diversity revealed by single-cell copy number analysis of prostate cancer cells across multiple disease states

PDF

Effects of chromatin regulators during carcinogenesis

PDF

Development of a colorectal cancer-on-chip to investigate the tumor microenvironment's role in cancer progression

PDF

Assessing the utility of wearable fitness trackers for objective capture and quantification of activity levels among patients undergoing radical cystectomy

PDF

Transposable element suppression in basal-like breast cancer

PDF

Understanding and controlling mitotic errors leading to aneuploidy in early ovarian cancer development

PDF

Investigating novel avenues for effective treatment of ovarian cancer

PDF

Characterization of the ZFX family of transcription factors that bind downstream of the start site of CpG island promoters

PDF

Targeting glioma cancer stem cells for the treatment of glioblastoma multiforme

Asset Metadata

Creator Melas, Eleni Marina (Marilena) (author)

Core Title Genomic, transcriptomic and immunologic landscapes of human cancers

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Cancer Biology and Genomics

Publication Date 08/05/2019

Defense Date 05/08/2019

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag cancer genomics,immunogenomics,OAI-PMH Harvest,personalized medicine,transcriptomics

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Press, Michael F. (committee chair), Daneshmand, Siamak (committee member), Gruber, Stephen B. (committee member), Triche, Timothy J. (committee member)

Creator Email marilenamela@yahoo.gr,melas@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-204928

Unique identifier UC11663032

Identifier etd-MelasEleni-7740.pdf (filename),usctheses-c89-204928 (legacy record id)

Legacy Identifier etd-MelasEleni-7740.pdf

Dmrecord 204928

Document Type Dissertation

Format application/pdf (imt)

Rights Melas, Eleni Marina (Marilena)

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 a...

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