Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Optimization of circulating tumor cells isolation for gene expression analysis
(USC Thesis Other)
Optimization of circulating tumor cells isolation for gene expression analysis
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
OPTIMIZATION OF CIRCULATING TUMOR CELLS ISOLATION FOR
GENE EXPRESSION ANALYSIS
By
Nita Elizabeth Jojo
A Thesis Presented to the
FACULTY OF THE USC KECK SCHOOL OF MEDICINE
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfilment of the
Requirements for the Degree
MASTER OF SCIENCE
(MOLECULAR MICROBIOLOGY AND IMMUNOLOGY)
August 2017
i
Acknowledgement
I would like to convey my sincere gratitude to my mentor and advisor, Dr.
Amir Goldkorn for giving me this wonderful opportunity of working in his lab
and to trust me with such an exciting and promising translational research project.
I would also like to thank him for all his support and guidance in completing this
thesis.
I would also like to thank my supervisor, Dr. Gareth Morrison for teaching
me patiently all the techniques for this thesis and for guiding me throughout the
entire project.
I would like to extend my gratitude to Dr. Daniel Zainfeld and Alexander
Cunha for consenting and collecting patient samples from the desired cohort.
I would also like to express my gratitude to all the members of the
Goldkorn laboratory: Dr. Tong Xu, Dr. Yucheng Xu, Emmanuelle Hodara. They
have all played such an integral part in the completion of this thesis and I am so
grateful for all of their help.
Table of Contents
Acknowledgement ......................................................................................................... i
Tables .............................................................................................................................. iii
Figures ............................................................................................................................ iv
Abstract ............................................................................................................................1
Chapter 1.0 Introduction .............................................................................................2
Chapter 2.0 Hypothesis ............................................................................................. 17
Chapter 3.0 Materials and Methods ...................................................................... 19
Chapter 4.0 Results .................................................................................................... 29
Chapter 5.0 Discussion ............................................................................................. 52
Chapter 6.0 References ............................................................................................. 57
iii
Tables
Table 3.1 List of primers used 26
Table 4.1 Performance of Double Separation method. 31
Table 4.2 Performance of single separation in Parsortix followed by
Negative depletion of the background WBC by Easy Sep Kit
33
Table 4.3 Performance of Rosette Sep depletion followed by Parsortix 35
Table 4.4 Range of detection of Rosette Sep depletion followed by Parsortix 38
Table 4.5 Comparing gene expression data for the specific panel designed by
Miyamoto across the most commonly used prostate cancer cell
lines
39
Table 4.6 A brief report of patient samples collected 47
iv
Figures
Figure 3.1 The Parsortix system and the technology used. 21
Figure 4.1 Methodological workflow of gene expression profiling of
enriched CTC fraction
29
Figure 4.2 Performance of Double Separation method. 31
Figure 4.3 Performance of single separation in Parsortix followed by
Negative depletion of the background WBC by Easy Sep Kit
33
Figure 4.4 Performance of Rosette Sep depletion followed by Parsortix 35
Figure 4.5 Range of detection of Rosette Sep depletion followed by
Parsortix by comparing the harvest rate.
37
Figure 4.6 Range of detection of Rosette Sep depletion followed by
Parsortix by comparing the background WBCs.
37
Figure 4.7 Comparing the the relative amount of transcripts in the pre-
amplified cDNA with a standard template in LnCap cells.
40
Figure 4.8 Comparing the the relative amount of transcripts in the pre-
amplified cDNA with a standard template in 22RV1 cells
41
Figure 4.9 Comparing the the relative amount of transcripts in the pre-
amplified cDNA after 18 and 14 cycles of pre-amplification
in LnCap cells.
42
Figure 4.10 Detection of gene expression using Rosette Sep depletion
followed by Parsortix protocol by qRT-PCR in spiked-in
samples
43
Figure 4.11 Detection of gene expression using Rosette Sep depletion
followed by Parsortix protocol by qRT-PCR in spiked-in
samples at time points 0, 3, 6 and 24 hours
44
Figure 4.12 Detection of gene expression using Rosette Sep depletion
followed by Parsortix protocol by qRT-PCR in spiked-in
samples at time points 0, 3 and 24 hours
45
Figure 4.13 Detection of gene expression using Rosette Sep depletion
followed by Parsortix protocol by qRT-PCR in patient
sample PS_1 at time points 0 and 24 hours
47
Figure 4.14 Expression of Miyamoto’s panel of genes and PBMC 48
v
specific genes in PBMC/WBC.
Figure 4.15 Detection of gene expression using Rosette Sep depletion
followed by Parsortix protocol by qRT-PCR in patient
sample PS_2 at time points 0 and 24 hours
49
Figure 4.16 Detection of gene expression using Rosette Sep depletion
followed by Parsortix protocol by qRT-PCR in patient
sample PS_3 at time points 0 and 24 hours
50
Figure 4.17 Detection of gene expression using Rosette Sep depletion
followed by Parsortix protocol by qRT-PCR in patient
sample PS_4.
51
1
Abstract
Circulating tumor cells (CTCs) are tumor cells that are disseminated from
the local site into the vasculature and thereby can be obtained by a simple blood
draw from the patients as a biopsy of the tumor. The technology of using CTCs as
a source to understand the current state of the tumor has been advancing and there
are many studies that have demonstrated the importance of characterizing these
cells. Gene expression profiling of CTCs has been a challenge since RNA
degrades faster when compared to genomic DNA. Even though there are studies
that have demonstrated single cell RNA-seq, the methods used are not robust and
reproducible and were also expensive for the large cohort study. Hence we
decided to use Parsortix (Angle. Inc, UK) which is a simple, cheap and robust
technology to enrich CTCs. One of the challenges that remained even after using
Parsrotix was the number of background WBCs that were along with the enriched
fraction. Therefore, we performed different methods to optimize the enrichment
fraction obtained using this technology and determined an efficient method by
using Rosette Sep Depletion followed by Parsortix. Therefore, we hypothesized
that using this technology in our enrichment process would enhance the ratio of
CTCs to WBC we obtain in the fraction and enable more reliable and high-
throughput sample analysis for detection of prostate cancer specific transcripts.
To test this hypothesis, we decided to use a specific panel of genes that are
prostate specific. Before using the method to identify signatures from patient
samples, the method optimized were tested using LnCap cells. Further, this assay
was validated using patient sample who were diagnosed with prostate cancer.
2
Chapter 1.0 Introduction
1.1. Precision medicine and its importance:
Precision medicine, known also as personalized therapy, is treatment
tailored to a specific individual that will optimize its efficiency or therapeutic
benefit, based on their molecular profile in the particular tumor. These somatic
mutations, that vary both between patients and within an individual’s tumor, can
be targeted. This has driven recently funded research into identifying these
specific genomic alterations or gene expression signature profiles that can be used
as targets (Collins & Varmus, 2015).
1.2. Biomarkers: their critical role in Precision medicine:
Cancer development and progression is driven by a multitude of genomic
alterations which include somatic mutations, rearrangements, amplifications,
deletions and epigenetic modulation of gene expression. Unfortunately, cancer is
also characterized by heterogeneity, making therapies targeted to a single
molecular site limited. Furthermore, therapy driven alterations often lead to
therapy resistance (Kelloff & Sigman, 2012). These changes are also due to the
variation in the tumor micro-environment and its influence on tumor, which
applies clonal selection through DNA damage, oxidative stress, mitotic stress,
pro-apoptotic stress and metabolic stress (Merlo, Pepper, Reid, & Maley, 2006).
These genomic alterations can be analyzed as biomarkers that could aid in the
selection of therapy.
3
The tissues collected during surgical resection or examination during
biopsy can be used to determine tumor profiling. But continuous biopsy from the
patient is not always possible as the procedure for biopsy is often painful and time
consuming and in certain cases, the location of the tumor is difficult to access.
Due to heterogeneity, genomic information gleaned from a single biopsy within a
patient’s tumor does not give the comprehensive genomic landscape for a given
tumor. (Bedard, Hansen, Ratain, & Siu, 2013; Marusyk, Almendro, & Polyak,
2012; Yap, Gerlinger, Futreal, Pusztai, & Swanton, 2012). As an
alternative/companion approach, liquid biopsies from patients could provide a
safer, cheaper and more robust way of collecting information about the current
state of the tumor. Liquid biopsy involves collection of peripheral blood from the
patient and isolation of circulating tumor cells (CTCs), or circulating tumor DNA
(ctDNA), or circulating tumor RNA (ctRNA) and / or extracellular vesicles
(EVs). Analysis of these isolated components can inform more about the current
state of the disease as it progresses under specific therapies (Siravegna, Marsoni,
Siena, & Bardelli, 2017). One advantage of using liquid biopsy is the multiple
blood draws can be performed repeatedly and non-invasively throughout disease
and treatment course. A second advantage is that ctDNA, ctRNA, and EVs
represent tumor cell throughout the body, rather than a single biopsy site.
4
1.2. Circulating Tumor Cells:
Circulating Tumor Cells (CTCs) are shed into the vasculature or
lymphatics from both primary and metastatic sites. The discovery of CTCs was
first reported by Ashworth when he detected cells in the blood of a cancer patient
and claimed that the cells are identical to the primary tumor (Ashworth, 1869).
1.2.1. Importance of CTCs:
CTCs are important due to the following reasons: i) Blood draw can be
performed easily when compared to biopsies and in certain metastatic cancer
types where the metastatic site is bone marrow; ii) Information from CTCs can be
used in predicting the prognosis of the patient; iii) Provides real-time tumor
information, which would also include certain mutations and alterations in the
tumor after chemotherapy and resection (West & JO, 2015).
1.2.2. Challenges faced in the study of CTC:
CTCs are rare and the isolation of CTCs from the blood stream is
extremely challenging: A standard blood draw (7.5ml) consists of red blood cells
(RBCs) (~ 40 billion), white blood cells (WBCs) (~40 million) and in the case of
a cancer patient, the blood will also consist of CTCs, which are extremely low in
number (~ 5 cells). Hence the isolation of CTCs requires the elimination of both
RBCs and WBCs (Alix-Panabieres & Pantel, 2014).
5
1.3. Rationale of liquid biopsy in Prostate cancer:
Worldwide, prostate cancer is the most common tumor among men where
1 man among 7 will be diagnosed (Siegel, Miller, & Jemal, 2016). Primary tumor
in prostate cancer patients are treated either by surgical resection or radiation.
These treatment decisions are based on pathological information from the prostate
biopsy, level of Prostate Specific Antigen (PSA), imaging and other factors.
Although PSA in blood has been used to detect the presence of prostate cancer in
men, this method has been less reliable with a large number of false positive
results (Crawford, Bennett, Andriole, Garnick, & Petrylak, 2013). Liquid biopsies
therefore have an important potential role in prostate cancer, as they would aid in
the tumor monitoring in the primary and metastatic state and also help in
determining the efficiency of the therapy that the patient is undergoing.
1.4 Different techniques to isolate CTCs:
CTCs can be isolated using different techniques that utilize either cancer
specific cell surface markers or differing size/deformability properties to capture
this rare population of cells. Different techniques that have been already
established to isolate CTCs are as follows:
6
1.4.1. Immunoaffinity:
This technique was developed on the principle of cell surface antigens that
could be targeted by antibodies. There are several approaches that have been
developed based on this technique:
1.4.1.1. Magnetic beads:
The FDA cleared CellSearch® was developed by Janssen Diagnostics,
(Raritan, NJ, USA) and was recently acquired by Menarini Silicon Biosystems
(Bologna, Italy)). It has been used clinically as a prognostic tool through isolation
of CTCs and subsequent enumeration. Antibodies targeted against epithelial cell
adhesion molecule (EpCAM) on the cell surface of CTCs are functionalized onto
magnetic beads, which are later isolated using a magnet cartridge holder. The
enumeration data from CellSearch® has been used as a prognostic marker and a
predictor of patient outcome in breast, prostate and colon cancer (Cristofanilli et
al., 2005; De Bono et al., 2008; Hu & Goldkorn, 2014; Huang, Wang, Xu, Huang,
& Zhang, 2013; Miller, Doyle, & Terstappen, 2010; Smerage & Hayes, 2006).
1.4.1.2. Microfluidic flow:
Ozkumur and Toner et al developed a CTC-iChip that uses an initial size
based enrichment step and hydrodynamic focusing followed by either positive
EpCAM CTC selection or leukocyte depletion. The efficiency of this device is
found to be much greater and purity of CTCs captured is higher too (Ozkumur et
al., 2013).
7
1.4.1.3. Leukocyte depletion:
Leukocyte depletion is a negative enrichment method that targets the
antigens expressed by the leukocytes using monoclonal antibodies and thereby
eliminate the cells using immunomagnetic separation (Lara et al., 2004; Yang et
al., 2009) or centrifugation with the Rosette Sep kit (Baccelli et al., 2013; He et
al., 2008). This approach leaves the CTCs free of magnetic beads and results in
high recovery rates but the purity of the sample is relatively low.
1.4.2. Physical properties:
The physical properties like density, size, deformability and di-electric
properties are different in CTCs when compared to the peripheral blood cells and
hence have been manipulated to allow the following approaches to be used to
effectively separate CTCs:
1.4.2.1. Density gradient Centrifugation:
Centrifugation with Ficoll or a porous barrier that separates cells based on
size would aid in the enrichment of CTCs from peripheral blood. There are many
methods that have been published based on this approach (Balic et al., 2005;
Fischer et al., 2013; Müller et al., 2005; Rosenberg et al., 2002). Successful RT-
PCR assays have been performed on these enriched fractions, but low level gene
expression of the targets in the WBCs has limited this approach. (Weitz, 1998).
8
1.4.2.2. Microfiltration:
CTCs are found to be larger in size when compared to the peripheral blood
cells and the technology of microfiltration uses this approach. There were several
microfilters that were developed by several groups (Chen et al., 2012; De Giorgi
et al., 2010; Hofman et al., 2011; Lecharpentier et al., 2011; Vona et al., 2004;
Zheng et al., 2011). The Goldkorn Lab developed a Parylene C filter in
collaboration with the Tai group from California Institute of Technology,
Pasadena. Isolated buffy coats were filtered through these slot filters and the
enriched fraction was used to determine telomerase activity using the TRAP
(Telomeric Repeat Amplification Protocol) assay (Xu, Lu, Tai, & Goldkorn,
2010).
1.4.2.3. Dielectrophoresis:
The tumor cells are influenced by positive DEP, when an electric field is
generated in a manner that is different from white and red blood cells. This
property of the tumor cells is being used in this technology and the capture
efficiency was 95%. ApoStream (Apocell, Texas) uses this technology to isolate
CTCs from the background WBCs and fractions of both the cell types are
obtained. The DEPArray (Silicon Biosystems) is another platform that isolates
single CTCs using a microfluidic cartridge from a pre-enriched sample.
9
1.4.2.4. High content scanning:
Recently, a platform to identify and scan all the materials and cells has
been developed and this approach has advanced identification of CTCs. Based on
immunofluorescent and morphologic features of whole blood smears Epic
Sciences (San Diego, USA) uses a high content scanning and CTC identification
in an algorithmic fashion. Similarly, Rarecyte (Seattle, USA) uses density
centrifugation in enriching nucleated cells and later creates smears for CTC
identification based on immunofluorescent and morphologic criteria using
automated multiplex scanner. Even though there are sophisticated methods for
CTC isolation that are being developed, the lack of universally applicable CTC
markers and complexity in isolating single cell for analysis has to be addressed.
1.5. Analytic assays of CTCs:
Successful CTC analysis depends on sample purity, sensitivity, cell
viability, and ability to recover cells from the enrichments process. The process
used for isolation of CTCs is critical as the process should be chosen based on the
downstream analysis desired. The characterization of the false-positive and false-
negative CTCs that could be identified using the respective enrichment process
and analysis tool should be determined by performing the experiments using cell
line models.
10
1.5.1. Immunophenotyping:
Immunostaining is the standard procedure used for both CTC detection
and enumeration. The criteria for identifying CTCs by Cell Search is as follows:
1) positive cytokeratin expression in CTCs; 2) negative expression of CD45,
which is an antigen commonly expressed in leukocytes; 3) positive DAPI staining
signifying presence of a nucleus.
In addition to immunostaining for enumeration, phenotyping of CTCs also
aids in the study of mechanisms of metastasis and therapy resistance. For
example, expression of androgen receptor (AR), estrogen receptor (ER),
progesterone receptor (PR), and epidermal growth factor receptor (EGFR) can be
used in phenotyping of CTCs in specific cancer types and used in the clinical
prognosis of patients (Lopez, Weiss, Robles, Fanta, & Ramamoorthy, 2016; Singh
et al., 2002; Subramanian & Simon, 2010).
1.5.2. Fluorescence In Situ Hybridization (FISH):
Copy number variation in the genes of a cancer cell can be very
informative and predict the prognosis of the patient and also therapeutic efficacy.
FISH assay on CTCs is used to determine the amplification/deletion of significant
genes for a particular type of cancer e.g., AR amplification in CRPC. These genes
may not be amplified/deleted in the primary tumor and hence the readout from the
CTCs could provide a better picture as disease progresses. Also, it could identify
the translocation of the gene or the rearrangement of genes. In the case of prostate
cancer, TMPRSS2-ERG translocation has been the most common translocation
11
and it has been shown that this can be identified using FISH analysis (Stott, Lee,
& Nagrath, 2010).
1.5.3. Oncogene/Tumor Suppressor gene mutation detection and genomic
analysis:
Detection and analysis of gain of function mutations in oncogenes and the
loss of function mutations in tumor suppressor genes is very important as there
are genomic changes that occur over time due to the Darwinian clonal evolution
of cancer cells caused by selective pressure like targeted therapy (Burrell,
McGranahan, Bartek, & Swanton, 2013). Using CTCs, the detection and analysis
of these genes can be performed, which is especially important in cases were the
primary tumor has been resected and is not available for analysis. Some of the
most common molecular events in prostate cancer include androgen receptor
(AR) mutation or amplification, gene rearrangements of ETS transcription factors,
loss of PTEN homolog (Roychowdhury & Chinnaiyan, 2013). Liu et al. showed
that the targeted next generation sequencing of CTCs could detect single
nucleotide variants (SNVs) in men with hormone sensitive prostate cancer
(HSPC), where SNVs were already shown to have a relation with the clinical
outcome of the prostate cancer patient (S. V Liu et al., 2013; Tsuchiya et al.,
2013). Steinestel et al. demonstrated that detection of key-resistance mediating
AR modifications, like AR-V7 splice variant and AR point mutations, in prostate
CTCs can be used to provide better cancer treatment (Steinestel et al., 2015). Loss
of function of tumor suppressor gene PTEN was detected in CTCs and the status
12
of PTEN was correlated to matched tumor samples, that could be used in the
investigation in CRPC clinical trials of PI3K/AKT targeted therapies
(Ferraldeschi et al., 2015)
1.5.4. Gene expression analysis of CTCs:
Gene expression profiling of CTCs can offer important information about
underlying cancer, such as: i) driver genes that are expressed at high levels or,
alternatively, repressed in tumors ; ii) gene translocations and alternative splice
variants; Antonarakis et al. studied the genomic alterations in AR and has recently
focused his work on the androgen receptor splice variant 7 (AR-V7) in CTCs
(Emmanuel S. Antonarakis., Changxue Lu., Hao Wang, et al., 2015). Alere Inc.,
(San Diego, CA, USA) developed the Alere
TM
CTC AdnaTest that was able to
detect the AR-V7 signature from the CTCs, since the presence of AR-V7
transcripts indicates a poor prognosis and eventual resistance when treated with
second line therapies abiraterone and enzalutamide.
1.6. Applications of the data from CTCs:
Enumeration or molecular analysis of CTCs are only useful if this
information can be used in the better treatment of cancer patients. The following
are the ways in which this information from the CTCs can be applied:
13
1.6.1. Prognostic marker:
The enumeration of CTCs can be used in the prediction of clinical
outcome of the patient. There are several studies that have found that in
Metastatic Prostate cancer patients, the CTC enumeration data from Cell Search
would provide prognostic information clinical outcome. A multi-centered clinical
trial based on CTC enumeration was first completed on metastatic breast cancer
patients by Cristofanilli et al. and patients having the 5 CTCs/7.5 mL of blood had
a worse prognosis than patients with less than 5 CTCs (Cristofanilli et al., 2005).
In prostate cancer, de Bono et al. studied that the overall survival of metastatic
castration resistant prostate cancer patients and predicted the survival rate in
response to therapy, based on the favorable outcome of <5 CTCs in 7.5 mL of
blood or unfavorable outcome (De Bono et al., 2008). Cohen et al. performed a
multi-centered study on metastatic colorectal cancer patients and predicted the
overall survival of the patients in response to the treatment by examining the
prognostic value of CTC enumeration to be favorable if there is <3 CTCs in 7.5
mL of blood and unfavorable if not (Cohen et al., 2008).
1.6.2. Studies on Predictive biomarkers:
CTCs aid in response monitoring, residual disease assessment and early
relapse detection. The CTCs are informative because the relapsed tumor are very
difficult to obtain and they are found to be very different from the primary tumor
like in Castration Resistant Prostate Cancer (CRPC). Darshan et al. studied the
14
location of AR in cytoplasm and nucleus in CTCs using immunofluorescence and
the presence of AR in either of the location was associated with the response to
the treatment (Darshan et al., 2011). Scher at al. confirmed the association
between expression of AR-V7 in CTCs and resistance to abiraterone or
enzmalutide hormonal therapies and showed that the presence of CTC AR-V7 in
CRPC patients predicts resistance to AR signaling inhibition and overall survival
rate of the patient can be improved on an alternative therapy (Scher, Lu, &
Schreiber, 2016).
1.7. Gene Expression Profiling of CTCs:
Gene Expression profiling is one of the most informative analyses that
could be performed with CTCs. There have been several studies that involved
gene expression profiling from CTCs. Most of the groups emphasize on single
CTC gene profiling (Miyamoto et al., 2015) or analysis from the whole blood
directly by lysing the whole blood in a preservative (Danila et al., 2014).
Miyamoto et al., focused on performing single cell RNA sequencing of CTCs
where they isolated CTCs after enrichment of the blood. Even in the process of
enrichment antibody staining against EPCAM is performed and the selection of
CTCs is based on the EPCAM staining. Also micromanipulation of single CTCs
would increase the time to isolate good quality RNA.
On the other hand, Danila et al., performed the gene expression profiling
of the whole blood sample where the whole blood was lysed and gene expression
15
profiling was performed. Most of the genes that are expressed by CTCs are also
expressed by WBCs even though in this study they used several bioinformatic
techniques to determine the genes that are expressed only in CTCs and not in
WBCs (Danila et al., 2014).
Gorges et al., 2016 has performed the enrichment thorugh Parsortix size-
based microfluidic platform (Angle UK), but late they isolated single cells from
the enriched fraction and inorder to identify single cells, they stained the cells for
EpCAM and Cytokeratin markers which is again anitbody staining for the
identification of CTCs (Gorges et al., 2016).
1.8. Limitations of current approaches to CTC gene expression profiling:
The above explained mehods and technologies to isolate CTCs and
perform a gene expression profiling has ceratin limitations. Scher et al. performed
a gene expression profile in whole blood by collecting 2.5 mL of blood in
PAXgene tubes (Danila et al., 2014). While this assay has been proved efficient,
there is definitely variability in the data obtained as the gene expression from the
nucleated background cells is more. The CTC-iChip is a microfluidic technology
that captures the CTCs without manipulating the tumor cells (Ozkumur et al.,
2013) and also effectively depleting the tumor cells would promise high-quality
RNA and also maintain the cells in a viable state, which is necessary for such
sensitive gene expression analysis (Miyamoto et al., 2015; Ting et al., 2014). But
CTC-iChip is not commercially available and also the background WBCs
16
obtained using this technology ranges up to 500 cells and hence the gene
expression profiling from such enriched fraction is cannot be completely reliable.
Also, other microfluidic technology like ClearCell FX (Clearbridge Biomedics,
Singapore) that is commercially available has a background hindrance of about
5000 cells.
17
Chapter 2.0 Hypothesis
Tumor transcriptome sequencing has provided a potential advantage of
spotting the drivers of cancer progression. Even though, RNA is relatively fragile
and the amount of RNA obtained from CTCs are low, there has been studies that
has completed analysis to the single cell level. RNA-seq of single prostate CTCs
showed that the progression of the disease was related to the non-canoncial Wnt
signlaing, which could also be a potential reason for the drug resistance
(Miyamoto et al., 2015). Miyamoto et al. performed this study by enriching the
CTCs using CTC-iChip and later isolating single cells using micro-manipulator.
The only other alternative method is to pick up single cells using DepArray
(Silicon Biosystems) using a microfluidic cartridge and then analyzing the
individual cells or using a micromanipulator to isolate single cells. But these
techniques again involve staining the CTCs and the time involved in picking up
these individual CTCs may alter the gene expression profiling and it may not
represent the expression from the cells from the patient. Hence, processing the
CTCs in a fast and robust method is required to obtain reliable gene expression
profile. On the other hand, whole blood RNA can be analyzed by using tubes with
preservatives (e.g. PAXgene, Qiagen; RNA Streck, Liquid Genomics). Danila et
al. performed a gene expression profile in whole blood by collecting 2.5 mL of
blood in PAXgene tubes (Danila et al., 2014). Liu et al detected the expression of
AR-V7 in whole blood samples of prostate cancer patients who were undergoing
second line therapies which had been previously shown in CTCs (X. Liu et al.,
2016). While this assay has been proved efficient, there is definitely variability in
18
the data obtained as the gene expression from the nucleated background cells is
more. Therefore, we sought an approach that could be more effective in large
cohorts and hence we decided to use the Parsortix (Angle. Inc, UK) technology.
Parsortix is a microfluidic technology that uses size and deformability for
enriching the rare cells and thereby the variability involved in the staining of the
cells is avoided. Using this technology in combination with other WBC depletion
methods can produce unprecedented enrichment of WBCs in a manner that is
relatively fast and requires little manipulation of the cells. Therefore, we
hypothesized that using this technology in our enrichment process would enhance
the ratio of CTCs to WBC we obtain in the fraction and enable more reliable and
high-throughput sample analysis for detection of prostate cancer specific
transcripts.
19
Chapter 3.0 Materials and Methods
3.1. Cell culture and its maintenance:
LnCap and 22RV1, human prostate cell lines were cultured in RPMI 1640
supplemented with 10% fetal bovine serum (Omega), penicillin (100 units/mL,
Invitrogen) and streptomycin (100µg/mL). The cells were maintained at 37°C, 5%
CO
2
. During sub-culture, the cells were trypsinized for 3 minutes at room
temperature.
3.2. Staining by CFDA-SE:
The cell lines were centrifuged to obtain a cell pellet and the supernatant
was aspirated. The cells were resuspended gently in 10mL of pre-warmed PBS
containing the 1mM of CFDA-SE reagent (Vybrant® CFDA SE Cell Tracer Kit,
Invitrogen). The cells were then incubated for 15 minutes at 37°C. The cells were
re-pelleted again and re-suspended in a fresh pre-warmed medium. The cells were
incubated for another 30 minutes and then pelleted again.
3.3. Spiking-in defined number of cells into healthy blood:
After staining the cells with CFDA-SE, the pellet was re-suspended in 1
mL of fresh pre-warmed medium. The cell concentration in the solution was
determined using a hemocytometer. The solution was diluted to obtain 10000
cells/mL using serial dilution. The cell number in 10 µL of a diluted solution was
20
counted again using an INCYTO disposable hemocytometer (DHC-N01,
Neubauer Improved). The wells of a 96-well plate were coated with 10%
Pluronic® F-68 solution (Poloxamer 188 solution, Sigma-Aldrich) so that the
cells would not adhere to the bottom of the well. Approximately 200, 100, 50 or
25 cells were transferred to the wells of a 96 well plate and 0.2% Pluronic in
1XPBS was added to make up a 100µL volume. The cells were allowed to settle
for 15 minutes on a level surface and then counted in one direction per counting
frame in a zig-zag manner for the whole well. In an EDTA vacutainer, 8 mL of
healthy blood was collected. The pipette tip was coated with 0.2% Pluronic in
1XPBS and it was used to transfer the 100µL of cell suspension from the well into
the blood. The leftover cells were counted under the microscope and the exact
number of cells spiked in was calculated.
3.4. Enumeration of the stained LnCap cells and background WBCs:
The stained LnCap cells were viewed using the FITC filter and
enumerated in one direction per counting frame in a zig-zag manner in wells. The
same technique was used for enumerating cells captured in the cassette. The
LnCap cells harvested from the Parsortix were enumerated and 1µL of Hoechst®
3342 dye (Sigma-Aldrich) was then added to the well and incubated for 5 minutes
to determine the total number of nucleated cells. After incubation, the cells were
viewed under DAPI filter. Images of different field of view under 10X
magnification was captured using the Zeiss Axio Observer. A1(Zeiss Microscope)
21
and the total number of cells in the well was estimated assuming the cells were
distributed uniformly in the well. By subtracting the number of LnCap cells
harvested, an estimation of the background nucleated WBCs was determined.
3.5. Parsortix and its features:
Parsortix system (Angle. Inc, UK) uses a patented microfluidic technology
in the form of a disposable cassette to a capture and then harvest CTCs from the
blood. The cassette captured CTCs based on their less deformable nature and
large size compared to other blood components. The disposable cassette was
placed in a clamp, and the patient sample was enriched for CTCs automatically.
CTCs were caught on a step that crisscrosses the microscope slide sized cassette.
Captured cells could be fixed and stained in the cassette to allow in-cassette
identification and enumeration or alternatively can be recovered (harvested) to
allow external staining and or genetic analyses such as quantitative Real-Time
Polymerase Chain Reaction (q RT-PCR) or sequencing.
Figure: 3.1.
Figure 3.1: The Parsortix system and the technology used.
22
3.6. Single Separation using Parsortix:
The single separation is the standard protocol for CTC enrichment using
Parsortix. Ethanol was used as a priming reagent and 1X PBS was used as a
running buffer. 8mL of whole blood in EDTA tube was spiked in with 200 LnCap
cells stained with CFDA-SE. The priming protocol PX2_PF was selected and
when prompted the desired separation cassette (6.5microns) was inserted. The
cassette was primed twice to ensure that there is no air bubble captured in the
separation cassette. The separation protocol PX2_S99F was selected. When
prompted, the existing vacutainer was rinsed and the vacutainer that contains the
stained cells with healthy blood was attached and the separation process was
initiated. The cells captured in the cassette were counted and !"#$%&' &"$' =
*+,-./ 01 2.334 2567+/.8 9: 7;. 2544.77. 517./ 4.65/5790:
*+,-./ 01 2.334 469<.8=9: 9:70 7;. ;.537;> -3008
. Cells were harvested using
the PX2_H protocol and collected in a well of 96 well plate. The cells were
allowed to settle for 15 minutes on a level surface and then counted. This helped
in calculating the ?"&@'A$ &"$' =
*+,-./ 01 2.334 ;5/B.47.8 1/0, 7;. C5/40/79D
*+,-./ 01 2.334 469<.8=9: 9:70 7;. ;.537;> -3008
and E'FG@'&H &"$' =
*+,-./ 01 2.334 ;5/B.47.8 1/0, 7;. C5/40/79D
*+,-./ 01 2.334 2567+/.8 9: 7;. 2544.77. 517./ 4.65/5790:
. The
residual nucleated white blood cells were also quantified.
3.7. Double Separation using Parsortix:
8 mL of whole blood in EDTA vacutainer was spiked in with 200 LnCap
cells stained with CFDA-SE was allowed to settle, to remove 1mL of plasma from
the blood sample using a pipette. The plasma was dispensed into a clean
23
vacutainer that has been rinsed with PBS to remove residual EDTA. 43% glycerol
(v/v) in PBS was used as a priming reagent and 1% BSA 2mM EDTA in PBS was
used as a running buffer. After priming the separation cassette of 6.5 microns
twice, the sample was attached to the system for separation. The cells captured
after the first separation was counted and noted as the First Capture Rate. The
cells were then harvested directly into 1 mL plasma collected earlier. The second
separation was performed after priming the 8 microns separation cassette twice
and using the same conditions as the first separation. The captured cells in the
cassette were enumerated after the separation and Second Capture rate was
calculated. Cells were harvested using the PX2_H protocol and collected in a well
of 96 well plate. The cells were allowed to settle for 15 minutes on a level surface
and then counted. The harvest rate and recovery rate was also calculated and the
residual nucleated WBCs were also quantified.
3.8. Single Separation in Parsortix followed by Negative depletion of the
background WBC by Easy Sep Kit:
8mL of whole blood in EDTA tube was spiked in with 200 LnCap cells
stained with CFDA-SE. The single separation protocol as described earlier was
performed and the output was harvested in a 0.25 mL Eppendorf tube. The Easy-
sep cocktail (Stemcell Technologies) was diluted to 1:10 dilution working
concentration (protocol optimized in the lab) and 5 µL of the diluted antibody was
added to the output obtained from the Parsortix and the mixture was incubated for
24
10 minutes. 10 µL of the magnetic particles was added to the mixture and
incubated for 10 minutes. The tubes were then placed on the 96 well plate on the
magnetic field and incubated for 10 minutes. Harvest material of 215 µL was
transferred to the well of a 96 well plate, leaving the magnetic particles behind.
The cells were allowed to settle for 15 minutes on a level surface and then
counted. The harvest rate and recovery rate was also calculated and the residual
nucleated WBCs were also quantified.
3.9. Rosette Sep depletion followed by Parsortix:
200 stained LnCap cells were spiked into 8mL of healthy blood and 50 µL
of Rosette Sep CD45 Depletion cocktail (Stemcell Technologies) was added per 1
mL of blood. It was incubated on a roller for 20 minutes at room temperature. In a
SepMate conical tube (Stemcell Technologies), the blood sample was layered
over 15 mL of Lymphoprep (Stemcell Technologies) using a serological pipette.
Care was taken to fill the lower chamber of the tube by pipetting through the hole
in the center of the tube. The layered sample was carefully transferred to the
centrifuge and spun at 1200xg for 20 minutes with the brake off. The sample was
removed carefully from the centrifuge and the plasma layer from the upper
chamber was poured into a standard 50 mL conical Falcon tube. This product was
then run on the Parsortix and the single separation protocol was followed. Cells
were captured and enumerated to calculate the Capture rate and also harvested
using the PX2_H protocol and collected in a well of 96 well plate. The cells were
25
allowed to settle for 15 minutes on a level surface and then counted. The harvest
rate and recovery rate was also calculated and the residual nucleated WBCs were
also quantified.
3.10. RNA extraction and cDNA synthesis:
The output from the Parsortix (200 µL) was directly collected into 600 µL
of RLT buffer (RNaeasy Microkit, Qiagen) with 0.1% β-Mercaptoethanol. 800
µL of 70% ethanol is added to the solution and mixed well by pipetting. This
mixture was then transferred into an Easy-Spin column and centrifuged for
8000*g for 15 seconds. The flow was then discarded. 350 µl of RW1 buffer
(RNaeasy Microkit, Qiagen) was then added into the Easy-Spin column (RNaeasy
Microkit, Qiagen) and centrifuged for 8000*g for 15 seconds. The flow was then
discarded. The RNaeasy MinElute spin column was then placed in a new 2 ml
collection tube and 500 µl Buffer RPE (RNaeasy Microkit, Qiagen) was added to
the column and centrifuged for 15 s at ≥8000 x g or (≥10,000 rpm) to wash the
spin column membrane. The flow was then discarded. 500 µl of 80% ethanol was
then added to the RNeasy MinElute spin column. The lid was closed gently and
centrifuged for 2 min at ≥8000 x g (≥10,000 rpm) to wash the spin column
membrane. The flow-through and collection tube was then discarded. The RNeasy
MinElute spin column was placed in a new 2 ml collection tube. The lid of the
spin column was opened and centrifuged at full speed for 5 min. The flow-
through and collection tube was discarded. The RNeasy MinElute spin column
26
was placed in a new 1.5 ml collection tube. 10 µl RNase-free water was added
directly to the center of the spin column membrane. The lid was closed gently and
centrifuged for 1 min at full speed to elute the RNA. The RNA was obtained in 8
µl volume.
For cDNA synthesis, 8 µl of RNase-free water was added into the reaction
tube along with the RNA extracted above. 4 µL of the cDNA SuperMix (Quanta
Biosciences) was then added to the mixture. The following was the reaction
conditions followed for cDNA synthesis: 22°C for 5 minutes (1 cycle), 42°C for
30 minutes (1 cycle), 85°C for 5 minutes (1 cycle) and 4 °C for ∞. After the
synthesis, the cDNA was stored at 4°C.
3.11. List of Primers used:
Table: 3.1.
Gene of
Interest
Accession
Number
Predict
ed size
of the
produc
t
Forward Primer Reverse Primer
PSA
(KLK3)
NM_001648.2 125 CGAGAAGCATTCCCAACC
CT
TCACGCTTTTGTTCCTGATG
C
KLK2 NM_005551.4 136 GCTGGGAGTGTGAGAAGC
AT
ACCTGGCTATTCTTCTTTAG
GCA
PSMA
(FOLH
1)
NM_004476.1 176 GAATGCCAGAGGGCGATC
TA
AATGACTCCTTTGGCCCCTG
UGT2B
15
NM_001076.3 170 CCTTCTTGGTCATCCCAA
AACCA
TGATGTCCACACTGAGGGC
T
AMAC
R
NM_014324.5 129 CGAGCTGCTGATCAAAGG
ACT
CACCACTCTGCCTTCGTCTT
AR NM_000044.3 156 ACTGCCAGGGACCATGTT
TT
CTTCTGTTTCCCTTCAGCGG
KRT7 NM_005556.3 73 GGTGCTGAAGAAGGATGT
GGA
ATTCAGGGCATCCACCTTGG
KRT8 NM_0012562
82.1
190 ACGAGGATATTGCCAACC
GC
GAAGCCCTCTGGCCTTTGA
KRT18 NM_000224.2 82 GACAATGCCCGCATCGTT
CT
CCAGCTCTGTCTCATACTTG
ACT
27
KRT19 NM_002276.4 87 GTGCCACCATTGAGAACT
CC
CCGTCTCAAACTTGGTTCGG
A
EPCA
M
NM_002354.2 100 GCTGGCCGTAAACTGCTT
TG
ACATTTGGCAGCCAGCTTTG
CDH1 NM_004360.4 153 TGGATGTGAATGAAGCCC
CC
TGTCTCTCCAAATCCGATAT
GTTAT
Beta-
Actin
NM_001101.3 184 AGAGCTACGAGCTGCCTG
AC
AGCACTGTGTTGGCGTACA
G
CD45 NM_080921.3 198 AGGATCCTCACTTGCTCC
TC
TCTGTCGCCTTAGCTTGACA
CD16 NM_0011275
92.1
70 CAGGTGCCAGACAAACCT
CT
CAACAGCCAGCCGATATGG
A
CD19 NM_001770 131 GGCCCGAGGAACCTCTAG
T
TAAGAAGGGTTTAAGCGGG
GA
Table 3.1: List of primers used.
3.12. Pre-amplification of cDNA:
A total of 9 primer pairs had been used in this Pre-Amplification process.
The primers were pooled together and a concentration of 180µM was used. 5 µL
of the cDNA synthesized was added to the reaction tube along with 5 µL of
pooled primers (180µM) and 10 µL of SybrGreen Mix (Quanta Biosciences). The
pre-amplification was performed using the following reaction condition: 95°C for
10 minutes (1cycle); 95°C for 15 seconds and 60°C for 4 minutes (14 or 18
cycles); 99°C for 10 minutes (1 cycle).
3.13. Quantitative Real-Time PCR:
The cDNA or the pre-amplified cDNA was used for real-time PCR
amplification with specific gene primers and Quanta B-R Syber Green qPCR
supermix (Bioscience) using Bio-Rad MyiQ single color Real-Time PCR
Detection System (Bio-Rad) and Bio-Rad iQ5 (Bio-Rad). β-actin was used as a
28
loading control and the other genes in the specific panel was tested. A list of
primer sequences can be found in the table above.
3.14. Statistical analysis:
Experiments were conducted in triplicates and the represented as means.
Error bars represent the standard deviation derived from both biological and
technical triplicates.
29
Chapter 4.0 Results
4.1. Experimental approach:
Figure: 4.1
Figure 4.1: Methodological workflow of gene expression profiling of enriched CTC fraction is shown
above. To enrich tumor cells from spiking samples, preliminary test was performed and the protocol that gave
the highest enrichment ratio was chosen to determine the gene expression profiling from the spiking sample
and from the patient samples and also for other characterization.
30
4.2. Single separation using Parsortix:
The enrichment using this protocol was tested by spiking in 200 LnCap
cells stained with CFDA-SE in 7.5 mL of healthy blood. After spiking, the sample
was processed using the Single Separation protocol. The number of cells captured
and harvested were enumerated and the capture and the harvest rate was
calculated.
The capture rate was found to be 56% and the harvest rate was found to be
37%. Based on the capture and the harvest, the recovery of the cells was found to
be 66%. The background cells were enumerated and found to be 2578 cells
(approximately). Even though the capture and the harvest efficiency using this
protocol was high enough to determine the gene expression profiling, the residual
background cells that was found was high and this would cause variability in the
expression profiling.
4.3. Double Separation using Parsortix:
The single separation using Parsortix seemed promising, but the
enrichment was not high enough to perform gene expression profiling from the
enriched fraction. Hence, there was a need to reduce the number of background
cells that is obtained as a residue along with our cells of interest. This need for a
better enrichment led to the Double Separation using Parsortix, which is obtaining
higher enrichment by using the Single Separation twice. The double separation
using Parsortix was tested by performing the spiking experiment in duplicates
31
with 200 LnCap cells spiked into 7.5 mL of healthy blood and the protocol was
followed. The method was repeated in duplicates and the number of cells captured
and harvested was enumerated and plotted as shown in figure 4.2.
Figure: 4.2
Figure 4.2: Performance of Double Separation method. Stained LnCap cells were spiked into healthy
blood and cells captured in-cassette were enumerated before harvesting and subsequently separating for the
second time. Cells were enumerated in-cassette, and in second harvest. Harvest is expressed as percentage of
cells in second harvest compared with cells spiked-in. Recovery is expressed as percentage of cells in second
harvest compared with first capture.
Table: 4.1
1
st
capture 2
nd
capture Harvests Recovery WBCs
Mean 50.50 31.00 21.00 42.00 68
Std. Deviation 4.95 4.24 4.24 12.73 7.07
Table 4.1: Performance of Double Separation method. The data from the figure 4.2 was tabulated and also
the range of background WBCs enumerated after the harvest was tabulated.
The first capture rate was found to be in the desired range (50.50±4.95%).
But when the cells were captured again, there was further loss of spiked cells.
The background cells obtained were quite low with a range from 68±7 cells.
0%
20%
40%
60%
80%
100%
Cells (%)
Experiment 1
Experiment 2
32
The Double Separation of Parsortix was an efficient protocol for the
enrichment of CTCs, but some of the major issues are time required for
completing the protocol, reagents used for the protocol and the low harvest rate of
the desired cells. This method was time consuming, where the whole protocol
required more than 6 hours, which causes variability in the gene expression
profiling of the enriched fraction. The reagents used in the protocol was different
from the reagents used in the other protocol and all the reagents had to be made
in-house and so there could be variability involved in the experiments when the
protocol is used for large clinical trials. The harvest rate was low too, where the
desired cells were captured at the rate of 21(±4.24)% and harvest rates are
detrimental towards the enrichment ratio of the sample.
4.4. Single Separation in Parsortix followed by Negative depletion of the
background WBC by Easy Sep Kit:
Since Double Separation using Parsortix involved low capture and
required more time, we decided to pre-enrich the desired cells using the standard
Single Separation using Parsortix and later enrich the sample using Easy Sep Kit
(StemCell Technologies). The single separation protocol in Parsortix followed by
negative depletion of the background WBC by Easy Sep Kit was tested by
performing the spiking experiment in triplicates and with 200 LnCap cells spiked
into 7.5 mL of healthy blood.
33
Figure: 4.3
Figure 4.3: Performance of single separation in Parsortix followed by Negative depletion of the
background WBC by Easy Sep Kit. Stained LnCap cells were spiked into healthy blood and cells captured
in-cassette enumerated after harvesting. Cells were also enumerated after the Ferro Fluid separation which is
represented as harvest rate after Ferro Fluid Separation. Recovery is expressed as percentage of cells after the
Ferro Fluid separation compared with first capture. Percentage Enrichment is the ratio of number of cells
enumerated after the Ferro Fluid separation and the number of background WBCs enumerated.
Table: 4.2
LnCap after
Single
Separation in
Parsortix
LnCap after
FerroFluid
Separation
Recovery of
LnCap from
the Harvest
Percentage
Enrichment
WBCs
Experiment_1 39% 8% 21% 10% 129
Experiment_2 27% 14% 51% 1% 2335
Experiment_3 26% 10% 37% 18% 120
Table 4.2: Performance of single separation in Parsortix followed by Negative depletion of the
background WBC by Easy Sep Kit. The data from the figure 4.3 was tabulated and also the range of
background WBCs enumerated after the harvest was tabulated.
The data was validated by enumerating the stained spiked cells at different
stages as mentioned in the Materials and Methods section and the data was plotted
0%
20%
40%
60%
80%
100%
Cells (%)
Experiment 1 Experiment 2 Experiment 3
34
on a graph and tabulated as shown in figure 4.3 and table 4.2 respectively. The
capture rate of the stained LnCap cells was lower than the capture rate obtained
using other protocols. The protocol required 4 hours for completion which was
better compared to the Double Separation protocol. But this method was not
efficient for every single experiment because for one of the experiment, the
background cell remained as 2335 cells (approx.) which is generally obtained
only with the Single Separation. The reason why the background cells were not
depleted during the negative depletion using Easy Sep Kit was because of the red
blood cell carry over into the harvest material. Hence, the elimination of the red
blood cell during the Single Separation was important for the protocol to be
efficient. But this was a factor that could not be controlled as the whole blood
sample is used for Single Separation and the efficiency of separating different
cells depends on the instrument.
4.5. Rosette Sep depletion followed by Parsortix:
The Single Separation followed by negative depletion of the background
WBC by Easy Sep Kit was a reliable method but the process of elimination of the
WBCs after the Single separation led to the loss of the desired LnCap cells too.
Hence, the process of elimination of background WBCs was performed prior to
the Single Separation using the Parsortix. Since the elimination has to be
performed on whole blood sample, the Rosette Sep depletion cocktail for CD45
35
depletion was used to eliminate the background WBCs and also using this
method, eliminated the red blood cell carry over to the next step.
The Rosette Sep depletion followed by Parsortix protocol was tested by
performing the spiking experiment in duplicates and with 200 LnCap cells spiked
into 7.5 mL of healthy blood.
Figure: 4.4.
Figure 4.4: Performance of Rosette Sep depletion followed by Parsortix. Stained LnCap cells were spiked
into healthy blood and processed for Rosette Sep depletion. After the single separation in the Parsortix, cells
captured in cassette were enumerated as Capture. Harvest is expressed as percentage of number of cells
spiked in compared with the number of cells harvested. Recovery is expressed as percentage of cells
harvested compared with the cells captured in-cassette.
Table: 4.3.
Capture Harvest Recovery WBCs
RI-BLS-278 34% 28% 83% 47
RI-BLS-305 42% 33% 79% 121
Table 4.3: Performance of Rosette Sep depletion followed by Parsortix. The data from the figure 4.3 was
tabulated and also the range of background WBCs enumerated after the harvest was tabulated.
0%
20%
40%
60%
80%
100%
CELLS (%)
RI-BLS-278
RI-BLS-305
36
The data was tested and the enumeration of the cells captured and
harvested were plotted on a graph and tabulated. The enrichment using this
protocol was efficient and reliable as the background cells that was carried over
after enrichment was relatively low when compared to other methods and the
enrichment ratio was also higher when compared to the other methods. The
additional advantage of using this protocol was the relatively rapid time for
completion, 3 hours from start to finish. Hence this protocol which had relatively
high harvest rate and a low background was followed for the following
experiment and processing the patient samples.
4.6. Range of Detection of Rosette Sep Depletion followed by Parsortix:
Even though the protocol was decided to be followed for the processing
the samples, its range of detection was unknown. Hence the range of detection of
the protocol was determined by performing experiments with different number of
cells spiked in into the healthy blood and also enriching in different systems.
The data obtained was plotted on a graph as shown in figure 4.5 and
tabulated as shown in table. The experiment was performed in different
instruments to also include the variability across the instruments.
37
Figure: 4.5.
Figure 4.5: Range of detection of Rosette Sep depletion followed by Parsortix by comparing the harvest
rate. Stained LnCap cells of definite numbers (200,100, 50 and 25) were spiked into healthy blood and
processed for Rosette Sep depletion. After the single separation in the Parsortix, cells captured in cassette
were enumerated as Capture. Harvest is expressed as percentage of number of cells spiked in compared with
the number of cells harvested.
Figure: 4.6.
Figure 4.6: Range of detection of Rosette Sep depletion followed by Parsortix by comparing the
background WBCs. After the cells were harvested, the background WBCs were enumerated and the graph
was plotted based on the data obtained.
0%
20%
40%
60%
80%
100%
HARVEST RATE AFTER THE ROSETTE SEP
DEPLETION FOLLOWED BY PARSORTIX
RI-BLS-278
RI-BLS-279
RI-BLS-305
IJKLMNO%= [(SMTTNUVIJKLMNO)/(SMTTNYZU[M\ UV)]×100
0
100
200
300
400
500
TOTAL NUMBER OF BACKGROUND WBC
BACKGROUND WBC AFTER THE ROSETTE SEP
DEPLETION FOLLOWED BY PARSORTIX
RI-BLS-278
RI-BLS-279
RI-BLS-305
38
The background WBCs were enumerated.
Table: 4.4.
Capure(%) Harvest(%) Recovery(%) WBCs
Mean 35.59 25.57 71.95 114.25
Standard Deviation 5.58 5.92 13.84 42.14
Table 4.4: Range of detection of Rosette Sep depletion followed by Parsortix. The data from the figure 4.3
was tabulated and also the range of background WBCs enumerated after the harvest was tabulated.
From the data obtained, it shows that this protocol retained its range of
capture and harvest rates intact, without much deviation for 200, 100 50 and 25
cells spiked in. The data about the background cells also did not have a larger
deviation and so the range of detection for the background using this protocol is
expected to be approximately around 114 ±42 cells.
This experiments not only focused on determining the range of detection
of the protocol, but it also showed that the protocol did not have much variation in
the data even though the desired spiked in cells were different. Also the efficiency
of this protocol was shown as it resulted in harvesting cells even though there
were only 25 LnCap cells spiked into the healthy blood.
4.7. Panel of genes chosen for gene expression profiling:
The qRT-PCR was performed was performed for a specific panel and the
panel was designed based on the study done by Miyamoto et. al., and the genes
are specific for CTCs and not WBCs. The are prostate specific lineage markers
39
that are only expressed by the prostate specific CTCs. Hence the panel was used
in this study and the LnCap cell line was chosen based on the following data:
Table: 4.5.
Cell line PSA
(KLK3)
KLK2 PSMA
(FOLH1)
AMACR AR KRT7 KRT8 EpCAM
(EPCAM,
CDH1)
22RV1 ^
NA
LnCap ^
PC3 ^ ^ ^ (low) ^
Du145 ^ ^ ^ ^ ^
Table 4.5: Comparing gene expression data for the specific panel designed by Miyamoto across the
most commonly used prostate cancer cell lines. The expression of each gene for the most frequently used
cell lines were found from other studies and tabulated.
4.7. Pre-Amplification of cDNA:
The enriched product of the spiked cells in healthy blood and patient
sample from the Parsortix was determined to be be analyzed for gene expression
profiling. But before using the product directly, the process of analysis of
extraction of RNA, synthesizing cDNA, pre-amplifying the cDNA and the
quantification of the pre-amplified product had to be optimized and checked.
Hence LnCap and 22RV1 cell lines were directly used to optimize.
RNA was extracted from the LnCap and 22RV1 cell lines, using the
RNaeasy MicroKit and cDNA was synthesized using the Quanta Biosciences Kit
as described in the Materials and Methods section. The RNA was quantified
before adding it into the cDNA synthesis mixture and 100ng was used as a
standard input. For pre-amplification of the cDNA, the cDNA was diluted to
40
2ng/5µL. The process of pre-amplification for 14 cycles was followed as
described in the Materials and Methods section for both the cell lines.
In order to check the relative amount of transcripts in the pre-amplified
cDNA, the LnCap pre-amplified product was compared with the LnCap cDNA of
100 ng that was not amplified and this comparison is shown in figure 4.7.
Figure: 4.7.
Figure 4.7: Comparing the the relative amount of transcripts in the pre-amplified cDNA with a
standard template in LnCap cells. RNA was extracted from LnCap cell line and cDNA was synthesize.
2ng/5µL of cDNA was used for pre-amplification of 14 cycles for 9 targeted genes. Amplification was
checked using a standard 100ng input along with the pre-amplified product using a qPCR. All qPCR was
performed in triplicates and are represented as means. Error bars represent the standard deviation.
The comparison from the graph shows that the relative expression levels
of the genes was not altered by the pre-amplificationstep for LnCap and the
product was obtained from all the 10 genes including the house-keeping gene. But
the amplification of KLK2 was found to be not the same as other genes. Another
cell line was chosen to see if the amplification of all the genes are the relative and
18.37
20.93 21.12
17.67
21.74
24.9
0
23.73
21.19
21.78
24.6
23.1
20.93
24.92
25.55
0
25.33
23.71
0
5
10
15
20
25
30
Ct values
PreAmp Ct
Standard template Ct
B-Actin AR PSA PSMA AMACR KLK2 KRT7 KRT8 EPCAM
41
if KLK2 has the same amplification rate. We used 22RV1 another cell line, that
expresses all the genes in the panel.
Figure: 4.8.
Figure 4.8: Comparing the the relative amount of transcripts in the pre-amplified cDNA with a
standard template in 22RV1 cells. RNA was extracted from 22RV1 cell line and cDNA was synthesize.
2ng/5µL of cDNA was used for pre-amplification of 14 cycles for 9 targeted genes. Amplification was
checked using a standard 100ng input along with the pre-amplified product using a qPCR. All qPCR was
performed in triplicates and are represented as means. Error bars represent the standard deviation.
The amplification of KLK2 was found to be the same for 22RV1 cell line
and hence the same primers were used for patient sample Pre-amplification as the
amplification of the genes remained same across different cell lines as seen in
figure 4.8.
Even though the amplification of the genes is found with 14 cycles in cell
lines, the patient samples would have relatively less number of CTCs than the
spiked cells and to obtain signal from such an enriched fraction, might be an
15.73
18.86
25.06
20.36
22.11
25.52
23.89
21
19.54 19.55
22.35
28.54
24.46
26.3
26.59
28.67
24.5
23.74
0
5
10
15
20
25
30
35
Ct values
PreAmp
Standard template
B-Actin AR PSA PSMA AMACR KLK2 KRT7 KRT8 EPCAM
42
issue. Hence we decided to check if there is amplification with 18 cycles using
LnCap cells. To validate the cycles used for amplification, both 14 cycles and 18
cycles were compared and the following data were obtained.
Figure: 4.9.
Figure 4.9: Comparing the the relative amount of transcripts in the pre-amplified cDNA after 18 and 14
cycles of pre-amplification in LnCap cells. RNA was extracted from LnCap cell line and cDNA was
synthesize. 2ng/5µL of cDNA was used for pre-amplification of 14 and 18 cycles for 9 targeted genes. The
amplification of both cycles were compared using qPCR. All qPCR was performed in triplicates and are
represented as means. Error bars represent the standard deviation.
The graph shows that relative gene expression of 14 and 18 cycles of pre-
amplification are equal for all genes.
4.8. Rosette Sep Depletion followed by Parsortix protocol by qRT-PCR:
To further ensure the entire molecular analysis, the test was performed on
spiked in samples where 100, 50 and 25 cells were spiked-in, along with a control
13.35
16.44
18.22
12.33
20.51
17.49
0
18.73
15.7
18.5
21.65
23.12
18.09
25.75
21.98
0
24.52
21.03
0
5
10
15
20
25
30
Ct values
18cycles
14cycles
B-Actin AR PSA PSMA AMACR KLK2 KRT7 KRT8 EPCAM
43
where no cells were spiked in. The Rosette Depletion followed by Parsortix
protocol was performed for all the spiked in samples. The harvest was carefully
processed for RNA extraction. cDNA was synthesized and pre-amplification was
also performed which was followed by qPCR.
Figure: 4.10.
Figure 4.10: Detection of gene expression using Rosette Sep depletion followed by Parsortix protocol by
qRT-PCR in spiked-in samples. Stained LnCap cells were spiked into healthy blood and processed using
Rosette Sep depletion followed by Parsortix protocol. After the enrichment, the harvest was collected and
RNA was extracted followed by cDNA synthesis. Pre-amplification of cDNA was performed for the 9
targeted genes and was quantified using qPCR. All qPCR was performed in triplicates and are represented as
means. Error bars represent the standard deviation.
Prostate-specific gene transcripts were detected from the enriched cancer
cells for 7 of 8 genes in the panel, whereas no signal was detectable from control
blood samples without cancer cells (green) for all genes except KRT8.. One gene,
KRT7, was not detected in any of the samples, because KRT7 is negative in
LnCap cells that was used for spiking in. There is no significant difference in Ct
values between the different spike in experiments. A trend was expected where
44
the Ct values of 100 cells spiked in was much less than the 25 cells spiked in
sample. But since these experiments involve such low number of cells, the
technical variability is higher.
The data was also validated by performing the spiking experiments at
different time points since patient samples sometimes are received only after 24
hours from remote sites. The experiment was performed initially by spiking in
200 LnCap cells stained with CFDA-SE and the protocol was followed at time 0,
3, 6 and 24 hours after spiking in.
Figure: 4.11.
Figure 4.11: Detection of gene expression using Rosette Sep depletion followed by Parsortix protocol by
qRT-PCR in spiked-in samples at time points 0, 3, 6 and 24 hours. Stained LnCap cells were spiked into
healthy blood and processed using Rosette Sep depletion followed by Parsortix protocol at time points 0, 3, 6
and 24 hours. After the enrichment, the harvest was collected and RNA was extracted followed by cDNA
synthesis. Pre-amplification of cDNA was performed for the 9 targeted genes and was quantified using
qPCR. All qPCR was performed in triplicates and are represented as means. Error bars represent the standard
deviation.
21.22
30.92
27.84
27.36
29.44
35.03
0
30.36
29.96
21
28.17
25.57
25.93
30.53
31.59
0
25.99
27.4
20.74
28.27
25.82
25.84
29.23
31.22
0
27.1
29.36
17.95
29.98
28.59 28.66
0
31.98
0
27.28
0
0
5
10
15
20
25
30
35
40
B-Actin AR PSA PSMA AMACR KLK2 KRT7 KRT8 EPCAM
Ct values
Time Point = 0
Time Point = 3hours
Time Point = 6hours
Time Point = 24hours
45
The Ct value was detected all the time points and there was a trend shown
where the Ct values of time point 0 was higher than other time points, which was
not expected. This could be due to several reasons where time point 0 is processed
immediately and as the cells are being tortured constantly, the transcripts might
have degraded or it could be a technical variability. The experiment was repeated
and since there was no significant difference between the time points 3 and 6
hours, the time point 6 hours was not repeated.
Figure: 4.12.
Figure 4.12: Detection of gene expression using Rosette Sep depletion followed by Parsortix protocol by
qRT-PCR in spiked-in samples at time points 0, 3 and 24 hours. Stained LnCap cells were spiked into
healthy blood and processed using Rosette Sep depletion followed by Parsortix protocol at time points 0, 3
and 24 hours. After the enrichment, the harvest was collected and RNA was extracted followed by cDNA
synthesis. Pre-amplification of cDNA was performed for the 9 targeted genes and was quantified using
qPCR. All qPCR was performed in triplicates and are represented as means. Error bars represent the standard
deviation.
The Ct values of the time point 0 has been showing a similar trend. But the
Ct values of the sample after 24 hours is higher for certain genes. This data
indicates that the genes can be detected, albeit at a lower rate due to cell
19.8
27.62
25.14
24.58
29.49
32.83
0
27.35
27.82
18.89
25.23
23.09
22.25
27.08
28.79
0
26.31
25.64
17.72
28.76
25.76
25.46
30.23
31.28
0
27.68
30.72
0
5
10
15
20
25
30
35
B-Actin AR PSA PSMA AMACR KLK2 KRT7 KRT8 EPCAM
Raw Ct values
Time Point = 0
Time Point = 3 hours
Time Point = 24 hours
46
degradation, variability in the efficiency of the instrument used or technical
variability.
4.9. Gene Expression profiling of Patient samples:
The assay was optimized and checked for enrichment and gene expression
profiling. Hence patient samples were collected for determining signatures from
the enriched material. Advanced Metastatic Prostate Cancer Patients were the
ideal candidates that were needed for the study since they are found to have more
CTCs. The samples from the patients were collected with an informed consent
under an IRB-approved protocol and the details regarding the disease state and the
treatment is shown in the table 4.6. Most of these patients were in the advanced
stage of cancer and castration resistant. They had very levels of PSA in their
blood and were under treatment. Four patient samples were obtained. The patient
samples PS_1, PS_2 and PS_3 were enriched through the Rosette Sep Depletion
followed by Parsortix protocol at time point 0 and 24 hours and in duplicates and
the blood samples were collected in 4 EDTA tubes. The enriched sample was then
processed for gene expression profiling. As a control, the WBC pellet from the
same patient was collected at time point 0 and 24 hours and profiled for gene
expression.
47
Table: 4.6.
Patient
ID
Date
Received Disease Age Notes
PS_1 4/25/17
Castration
resistant
metastatic
prostate
cancer
79
Early castrate resistance, progressed after abiraterone
and enzalutamide, currently on Docetaxel. High PSA
decreasing.
PS_2 4/26/17
Stage 4
prostate
cancer
62
Adjuvant radiation, Lupron started but showed signs of
progression, started Enzalutamide and Everolimus but
progressed, started Docetaxel but might have stopped
mid-treatment
PS_3 5/3/17
Metastatic
prostate
cancer
81
Began radiation + Lupron, then started Abiraterone and
Prednisone but progressed, received Rad223 which left
stable disease, found new metastasis and started
Docetaxel -- seemingly continuing treatment. High
PSA at last follow-up.
PS_4 5/4/17
Metastatic
prostate
cancer
60
Began ADT + abiraterone but progressed recently.
Hence abiraterone was stopped and replaced with
enzalutamide. PSA has been rising since last follow-up.
Table 4.6: A brief report of the patient samples collected. The date of collection of the patient samples
along with their disease states were tabulated. The treatment according to the stage of disease given to the
patient is also explained in the notes.
For patient sample PS_1, the sample was enriched using the Rosette Sep
depletion followed by Parsortix protocol and the gene expression profiling was
performed as discussed earlier in the spiking samples.
Figure: 4.13.
Figure 4.13: Detection of gene expression using Rosette Sep depletion followed by Parsortix protocol by
qRT-PCR in patient sample PS_1 at time points 0 and 24 hours. Patient sample PS_1 was processed
using Rosette Sep depletion followed by Parsortix protocol at time points 0 and 24 hours in duplicates. After
the enrichment, the harvest was collected and RNA was extracted followed by cDNA synthesis. Pre-
22.82
0 0 0 0 0 0
26.95
0
21.17
0 0 0 0 0
34.74
28.99
0
19.07
31.14
0 0 0 0 0
25.39
0
19.27
0 0 0 0 0
34.93
25.57
0
0
5
10
15
20
25
30
35
40
B-Actin AR PSA PSMA AMACR KLK2 KRT7 KRT8 EPCAM
Ct values
Patient Sample PS_1
0hrs-EnrichedFraction_1 0hrs-EnrichedFraction_2
24hrs-EnrichedFraction_1 24hrs-EnrichedFraction_2
48
amplification of cDNA was performed for the 9 targeted genes and was quantified using qPCR. All qPCR
was performed in triplicates and are represented as means. Error bars represent the standard deviation.
There were no prostate specific genes that were detected from PS_1,
except for AR in one of the duplicate after 24 hours of processing. This could be a
non-specific signal as certain signals are altered and the signal was after 24 hours
which was not seen at the time point 0. And also the expression was a higher Ct
value of 31.14. There were kertain genes detected from the enriched fraction. In
order to identify if the genes were expressed from the enriched fraction, we also
did a control experiment by profiling the gene expression from WBC pellet from
a healthy blood sample.
Figure: 4.14.
Figure 4.14: Expression of Miyamoto’s panel of genes and PBMC specific genes in PBMC/WBC. RNA
was extracted from WBC pellet from healthy blood sample and cDNA was synthesized. The expression of
the genes in was quantified using qPCR. All qPCR was performed in triplicates and are represented as means.
Error bars represent the standard deviation.
Certain prostae specific genes like AMACR was detected in the WBC
pellet along with the keratin specific genes like KRT7 and KRT8. As a positive
22
0 0 0
31.77
0
33.43
26.35
0
21.97
25.02
31.49
0
10
20
30
40
B-Actin AR PSA PSMA AMACR KLK2 KRT7 KRT8 EpCAM CD45 CD16 CD19
Ct values
Expression of Miyamoto’s panel of genes and PBMC specific
genes in PBMC
49
control, WBC specific genes like CD45, CD16 and CD19 genes were also
checked for expression and they were detected.
Both patient sample PS_2 and PS_3 were enriched using the Rosete Sep
Depletion followed by Parsortix and profiled for gene expression as patient
sample PS_1.
Figure: 4.15.
Figure 4.15: Detection of gene expression using Rosette Sep depletion followed by Parsortix protocol by
qRT-PCR in patient sample PS_2 at time points 0 and 24 hours. Patient sample PS_2 was processed
using Rosette Sep depletion followed by Parsortix protocol at time points 0 and 24 hours in duplicates. After
the enrichment, the harvest was collected and RNA was extracted followed by cDNA synthesis. Pre-
amplification of cDNA was performed for the 9 targeted genes and was quantified using qPCR. All qPCR
was performed in triplicates and are represented as means. Error bars represent the standard deviation.
In patient sample PS_2, prostate specific genes AR, PSA, AMACR and
KLK2 were detected along with keratin specific genes KRT7 and KRT8. The
detection of prostate specific genes indiactes the presence of CTCs as these genes
are not expressed by the WBC pellet. There was no expression of EpCAM in this
sample which shows that there is variability in the expression of EpCAM by
CTCs and all CTCs do not necessarily express EpCAM.
17.99
32.02
0
34.89
31.4
33.44
0
27.91
0
18.88
31.83
0
33.89
32.59
0
31.86
30.69
0
18.64
0 0
33.62
30.54
33.82
0
28.65
0
21.03
0 0 0 0
33.35
33.78
28.95
0
0
5
10
15
20
25
30
35
40
B-Actin AR PSA PSMA AMACR KLK2 KRT7 KRT8 EPCAM
Ct values
Patient Sample PS_2
0hrs-EnrichedFraction_1 0hrs-EnrichedFraction_2
24hrs-EnrichedFraction_1 24hrs-EnrichedFraction_2
50
Figure: 4.16.
Figure 4.16: Detection of gene expression using Rosette Sep depletion followed by Parsortix protocol by
qRT-PCR in patient sample PS_3 at time points 0 and 24 hours. Patient sample PS_3 was processed
using Rosette Sep depletion followed by Parsortix protocol at time points 0 and 24 hours in duplicates. After
the enrichment, the harvest was collected and RNA was extracted followed by cDNA synthesis. Pre-
amplification of cDNA was performed for the 9 targeted genes and was quantified using qPCR. All qPCR
was performed in triplicates and are represented as means. Error bars represent the standard deviation.
In patient sample PS_3, the only prostate specific gene that was detected
was KLK2, but the other genes were not detected. Keratin gene KRT8 was also
detected but this gene is also expressed by the WBC pellet. Hence, it was
concluded that no CTCs were detected from this sample.
For patient sample PS_4, the sample was enriched using the same protocol
but the sample was processed only at time point 0, in duplicates and profiled for
gene expression. The sample was processed in parallel with the Cell Search and
the number of CTCs in the sample was enumerated.
23.93
0 0 0 0
32.2
0
32.45
0
24.41
0 0 0 0
32.41
0 0 0
23.08
0 0 0 0
33.35
0
26.77
0
22.7
0 0 0 0 0 0
28.61
0
0
5
10
15
20
25
30
35
40
B-Actin AR PSA PSMA AMACR KLK2 KRT7 KRT8 EPCAM
Raw Ct values
Patient sample PS_3
0hrs-EnrichedFraction_1 0hrs-EnrichedFraction_2
24hrs-EnrichedFraction_1 24hrs-EnrichedFraction_2
51
Figure: 4.17.
Figure 4.17: Detection of gene expression using Rosette Sep depletion followed by Parsortix protocol by
qRT-PCR in patient sample PS_4. Patient sample PS_4 was processed using Rosette Sep depletion
followed by Parsortix protocol at time points 0 in duplicates. After the enrichment, the harvest was collected
and RNA was extracted followed by cDNA synthesis. Pre-amplification of cDNA was performed for the 9
targeted genes and was quantified using qPCR. All qPCR was performed in triplicates and are represented as
means. Error bars represent the standard deviation.
The prostate specific gene that was detected from the sample was KLK2
and even this particular gene had a very high Ct value. Keratin genes KRT7 and
KRT8 was detected but these genes are also expressed by the WBC pellet. Hence
we concluded that there is no CTCs that was detected from this sample which was
validated by the Cell Search data which also did not enumerate any CTCs from
the sample. Cell Search has been used by many other groups to validate their
method or assay for the detection of CTCs since it’s the only FDA approved
technology for the enumeration of CTCs.
22.95
0 0 0 0
33.51 33.26
34.04
0
23.17
0 0 0 0
33.46
34.08
0 0
0
5
10
15
20
25
30
35
40
B-Actin AR PSA PSMA AMACR KLK2 KRT7 KRT8 EPCAM
Raw Ct values
Patient Sample PS_4
0hrs-EnrichedFraction_1 0hrs-EnrichedFraction_2
52
Chapter 5.0 Discussion
There has been a constant need for molecular characterization of CTCs as
the data from such analysis would give a better understanding of the tumor and
also emphasizes on the prognosis of the patient. The advantage that the molecular
profiling of CTCs can be used instead of tumor tissue profiling and the data can
be used to determine the resistance or progression of the disease are some of the
important reasons why the identification, recovery and characterization of CTCs
has become more accurate and cost-effective. The information obtained from the
CTCs about gene rearrangements, translocations, or any other mutation that
would lead to the difference in expression of the receptor and localization or
splice variation can be directly used as a tool for directing targeted therapy. Gene
expression is important as it provides all these information, but the current
methods either do not enrich CTCs that is reliable for gene expression profiling as
there is high background WBCs, or they isolate single cell for gene expression
profiling. Hence there was a need to find a suitable approach that can be
informative about the cancer cell fraction and also be rapid and efficient and cost
effective for multiple samples. And so we decided to set out to develop using the
new technology like Parsortix, a size and deformability based microfluidic system
that is available to us
The experimental approach was followed and based on the data obtained,
the Rosette Sep depletion followed by Parsortix protocol was chosen. This
protocol has been robust and less time consuming. It provides high enrichment of
the sample which is much needed for the gene expression profiling of the
53
samples. Since this method is faster and robust when compared to other protocol
available, it also provides good quality RNA which is required to obtain reliable
data from the patient samples.
The efficiency of Rosette Sep depletion followed by Parsortix was
validated by performing spiking in different number of cells (100, 50, 25 and 0).
The gene expression profiling was done from all the enriched fractions and it
clearly shows that the expression is only from the LnCap cells as the control did
not have expression. But there was no drop in Ct values as the spiked-in cells
gradually decreases. This is because of the variability involved in capturing the
cells and the technical variability when the molecular technique was performed.
The second experiment N=2 in determining qPCR signaling from the harvest of
the sample enriched by Rosette Sep Depletion followed by Parsortix (figure 4.10)
did not have AR signaling but it could be proven by repeating the experiment
several times that this was one of the many experiment where the transcript of AR
was lost during RNA extraction. This could be due to the technical variability or
the variability when using different systems.
The experiment was also performed at different time point experiment to
prove that the cells are captured and gene expression profiling could be performed
even after 24 hours for certain genes. But the data states that there was no
significant difference between time point 0 and 24 hours. This could be due to
more background and lower recovery rate which led to the loss of most of the
cells captured at time point 0. It could also be due to the assumption that cells
which were already dead were captured at time point 0 and gene expression
54
analysis was performed, whereas those cells were no longer present in the 24 hr
time point.
After validating the protocol, patient samples were drawn and the samples
were processed using the Rosette Sep depletion followed by Parsortix protocol.
The patient samples PS_1, PS_2 and PS_3 were obtained in 4 EDTA tubes and
the samples were processed in two different time points 0 and 24 hours in
duplicates. The sample PS_2 was the only sample that had expression of Prostate
specific genes, which helps in understanding that there is CTCs in that particular
sample. This enrichment/PCR protocol may be less reliable to gauge “quantitative
changes” in expression and may be more useful simply as a qualitative measure of
whether cancer cells are there. Hence there could be technical variability involved
in the assay. But after 14 cycles, AR signaling was determined along with PSMA.
Interestingly, Miyamoto et al., 2012 showed that the presence of "AR-mixed"
CTCs and increasing "AR-on" cells despite treatment with abiraterone acetate
were associated with an adverse treatment outcome and in these cases the PSMA
is found to be OFF which could be related to the data obtained and needs to be
confirmed with further experiments (Miyamoto et al., 2012). In the same sample
EpCAM was not expressed which could be related to the data obtained by Yu et
al., where they explain that there were a subset of CTCs that did not express
EpCAM (Yu et al., 2013).
The patient samples PS_1 and PS_3 did not have the signatures from the
prostate specific genes. Hence it indicates that there were no CTCs enriched in
that sample. The patient sample PS_4 was processed at time point 0 hours and in
55
duplicates. The sample was also processed for Cell Search and there was no CTCs
found in the sample. The gene expression data also proved that there was no
CTCs in the sample.
This assay was initially designed for detecting signatures from the
enriched fraction of patient sample and thereby characterizing the CTCs which
could be further used as a information for targeted therapy. While the assay was
optimized using cell lines, the gene expression was detected as the expression of
genes by the cell line spiked is known. But, when the assay was used to process
patient samples, the enriched fraction obtained is so different from the LnCap
cells. Cell lines are not a best representation of CTCs as the tumor in patients are
so different and there is heterogeneity within and in between patients. But there
were certain genes like KLK2, KRT7 and KRT8 which was detected in both the
patient sample and WBC pellet. Hence, we decided to replace these genes with
HOXB13, FOXA1 and GRHL2, which were used in the gene expression profiling
from PAXgene tubes (Danila et al., 2014). This is preliminary but serves as a
proof of concept that small numbers of cancer cells can be recovered with low
background allowing for detection of cancer specific transcripts. This small panel
can thus serve as a “litmus” test of whether prostate CTCs are present.
Further processing in this assay development would involve processing
patient samples in the optimized gene panel that has been optimized for only
CTCs specifically. The samples that would be collected in the future would be
processed with side-by-side CellSearch to see if presence of CTCs by CellSearch
correlates with panel and the results obtained in the assay. But the caveat of
56
processing samples and comparing it with CellSearch is that it is a different
enrichment method from the one used in this assay and thus may target different
cells (or more or fewer cells). The CellSearch is based on cell surface marker
enrichment system which is again a biased way of detecting CTCs as they are
found to have varying expression of cell surface markers and EpCAM.
This assay developed could be used to detect the presence of CTCs in the
sample as antibody staining is not reliable. The data obtained from this assay can
be used to perform an Open Array with a customized panel and determine
signatures from other targets and thereby obtain essential data. This data could be
also used to run a discovery, where new signatures can be found. The Rosette Sep
depletion followed by Parsortix is also used to obtain TRAP (Telomeric Repeat
Amplification Protocol) assay to detect telomerase activity and the protocol is yet
to be optimized for TRAP assay.
57
Chapter 6.0 References
1. Alix-Panabieres, C., & Pantel, K. (2014). Challenges in circulating tumour
cell research. Nat Rev Cancer, 14(9), 623–631.
https://doi.org/10.1038/nrc3686
2. Ashworth T. A case of cancer in which cells similar to those in the tumors
were seen in the blood after death. Aust. Med. J. 14, 146–147 (1869).
3. Baccelli, I., Schneeweiss, A., Riethdorf, S., Stenzinger, A., Schillert, A.,
Vogel, V., … Trumpp, A. (2013). Identification of a population of blood
circulating tumor cells from breast cancer patients that initiates metastasis
in a xenograft assay. Nature Biotechnology, 31(6), 539–544.
https://doi.org/10.1038/nbt.2576
4. Balic, M., Dandachi, N., Hofmann, G., Samonigg, H., Loibner, H.,
Obwaller, A., … Bauernhofer, T. (2005). Comparison of two methods for
enumerating circulating tumor cells in carcinoma patients. Cytometry Part
B - Clinical Cytometry, 68(1), 25–30. https://doi.org/10.1002/cyto.b.20065
5. Bedard, P. L., Hansen, A. R., Ratain, M. J., & Siu, L. L. (2013). Tumour
heterogeneity in the clinic. Nature, 501(7467), 355–364.
https://doi.org/10.1038/nature12627
6. Burrell, R. A., McGranahan, N., Bartek, J., & Swanton, C. (2013). The
causes and consequences of genetic heterogeneity in cancer evolution.
Nature, 501(7467), 338–345. https://doi.org/10.1038/nature12625
7. Chen, C., Mahalingam, D., Osmulski, P., Jadhav, R. R., Wang, M., Leach,
R. J., … Huang, T. H. (2012). Single-cell analysis of circulating tumor
58
cells identifies cumulative expression patterns of EMT-related genes in
metastatic prostate cancer. The Prostate, 73(8), 813–826.
https://doi.org/10.1002/pros.22625.Single-cell
8. Cohen, S. J., Punt, C. J. A., Iannotti, N., Saidman, B. H., Sabbath, K. D.,
Gabrail, N. Y., … Meropol, N. J. (2008). Relationship of circulating tumor
cells to tumor response, progression-free survival, and overall survival in
patients with metastatic colorectal cancer. Journal of Clinical Oncology,
26(19), 3213–3221. https://doi.org/10.1200/JCO.2007.15.8923
9. Collins, F. S., & Varmus, H. (2015). A New Initiative on Precision
Medicine. New England Journal of Medicine, 372(9), 793–795. JOUR.
https://doi.org/10.1056/NEJMp1500523
10. Crawford, E. D., Bennett, C. L., Andriole, G. L., Garnick, M. B., &
Petrylak, D. P. (2013). The utility of prostate-specific antigen in the
management of advanced prostate cancer. BJU International, 112(5), 548–
560. https://doi.org/10.1111/bju.12061
11. Cristofanilli, M., Hayes, D. F., Budd, G. T., Ellis, M. J., Stopeck, A.,
Reuben, J. M., … Terstappen, L. W. M. M. (2005). Circulating tumor
cells: A novel prognostic factor for newly diagnosed metastatic breast
cancer. Journal of Clinical Oncology, 23(7), 1420–1430.
https://doi.org/10.1200/JCO.2005.08.140
12. Danila, D. C., Anand, A., Schultz, N., Heller, G., Wan, M., Sung, C. C.,
… Scher, H. I. (2014). Analytic and clinical validation of a prostate
cancer-enhanced messenger RNA detection assay in whole blood as a
59
prognostic biomarker for survival. European Urology.
https://doi.org/10.1016/j.eururo.2013.07.006
13. Darshan, M. S., Loftus, M. S., Thadani-Mulero, M., Levy, B. P., Escuin,
D., Zhou, X. K., … Giannakakou, P. (2011). Taxane-induced blockade to
nuclear accumulation of the androgen receptor predicts clinical responses
in metastatic prostate cancer. Cancer Research, 71(18), 6019–6029.
https://doi.org/10.1158/0008-5472.CAN-11-1417
14. De Bono, J. S., Scher, H. I., Montgomery, R. B., Parker, C., Miller, M. C.,
Tissing, H., … Raghavan, D. (2008). Circulating tumor cells predict
survival benefit from treatment in metastatic castration-resistant prostate
cancer. Clinical Cancer Research, 14(19), 6302–6309.
https://doi.org/10.1158/1078-0432.CCR-08-0872
15. De Giorgi, V., Pinzani, P., Salvianti, F., Panelos, J., Paglierani, M.,
Janowska, A., … Massi, D. (2010). Application of a Filtration- and
Isolation-by-Size Technique for the Detection of Circulating Tumor Cells
in Cutaneous Melanoma. Journal of Investigative Dermatology, 130(10),
2440–2447. https://doi.org/10.1038/jid.2010.141
16. Emmanuel S. Antonarakis, M.D., Changxue Lu, Ph.D., Hao Wang, Ph.D.,
B. L., Sc.M., Mary Nakazawa, M.H.S., Jeffrey C. Roeser, B.S., Yan Chen,
Ph.D., T. A., Mohammad, Ph.D., Yidong Chen, Ph.D., Helen L. Fedor,
B.S., Tamara L. Lotan, M.D., Q., Zheng, M.D., Angelo M. De Marzo,
M.D., Ph.D., John T. Isaacs, Ph.D., W. B. I., Ph.D., Rosa Nadal, M.D.,
Channing J. Paller, M.D., Samuel R. Denmeade, M.D., M. A., &
60
Carducci, M.D., Mario A. Eisenberger, M.D., and Jun Luo, P. . (2015).
AR-V7 and resistance to enzalutamide and abiraterone in prostate cancer,
371(11), 1028–1038. https://doi.org/10.1056/NEJMoa1315815.AR-V7
17. Ferraldeschi, R., Nava Rodrigues, D., Riisnaes, R., Miranda, S.,
Figueiredo, I., Rescigno, P., … De Bono, J. (2015). PTEN protein loss and
clinical outcome from castration-resistant prostate cancer treated with
abiraterone acetate. European Urology, 67(4), 795–802.
https://doi.org/10.1016/j.eururo.2014.10.027
18. Fischer, J. C., Niederacher, D., Topp, S. A., Honisch, E., Schumacher, S.,
Schmitz, N., … Stoecklein, N. H. (2013). Diagnostic leukapheresis
enables reliable detection of circulating tumor cells of nonmetastatic
cancer patients. Proceedings of the National Academy of Sciences,
110(41), 16580–16585. https://doi.org/10.1073/pnas.1313594110
19. Gorges, T. M., Kuske, A., R??ck, K., Mauermann, O., M??ller, V., Peine,
S., … Pantel, K. (2016). Accession of tumor heterogeneity by multiplex
transcriptome profiling of single circulating tumor cells. Clinical
Chemistry, 62(11), 1504–1515.
https://doi.org/10.1373/clinchem.2016.260299
20. He, W., Kularatne, S. A., Kalli, K. R., Prendergast, F. G., Amato, R. J.,
Klee, G. G., … Low, P. S. (2008). Quantitation of circulating tumor cells
in blood samples from ovarian and prostate cancer patients using tumor-
specific fluorescent ligands. International Journal of Cancer, 123(8),
1968–1973. https://doi.org/10.1002/ijc.23717
61
21. Hofman, V. J., Ilie, M. I., Bonnetaud, C., Selva, E., Long, E., Molina, T.,
… Hofman, P. M. (2011). Cytopathologic detection of circulating tumor
cells using the isolation by size of epithelial tumor cell method: Promises
and pitfalls. American Journal of Clinical Pathology, 135(1), 146–156.
https://doi.org/10.1309/AJCP9X8OZBEIQVVI
22. Hu, B., & Goldkorn, A. (2014). Circulating tumor cells in prostate cancer:
a better predictor of survival than prostate specific antigen? Expert Review
of Anticancer Therapy, 14(11), 1257–1260.
https://doi.org/10.1586/14737140.2014.957190
23. Huang, J., Wang, K., Xu, J., Huang, J., & Zhang, T. (2013). Prognostic
significance of circulating tumor cells in non-small-cell lung cancer
patients: A meta-analysis. PLoS ONE, 8(11).
https://doi.org/10.1371/journal.pone.0078070
24. Kelloff, G. J., & Sigman, C. C. (2012). Cancer biomarkers: selecting the
right drug for the right patient. Nature Reviews Drug Discovery, 11(3),
201–214. https://doi.org/10.1038/nrd3651
25. Lara, O., Tong, X., Zborowski, M., Chalmers, J. J., Lara, O., Tong, X., …
Chalmers, J. J. (2004). Enrichment of rare cancer cells through depletion
of normal cells using density and ow-through, immunomagnetic cell
separation. Experimental Hematology, 32(10), 891–904.
https://doi.org/10.1016/j.exphem.2004.07.007
26. Lecharpentier, A., Vielh, P., Perez-Moreno, P., Planchard, D., Soria, J. C.,
& Farace, F. (2011). Detection of circulating tumour cells with a hybrid
62
(epithelial/mesenchymal) phenotype in patients with metastatic non-small
cell lung cancer. British Journal of Cancer, 105(9), 1338–1341.
https://doi.org/10.1038/bjc.2011.405
27. Liu, X., Ledet, E., Li, D., Dotiwala, A., Steinberger, A., Feibus, A., …
Zhang, H. (2016). A Whole Blood Assay for AR-V7 and ARv567es in
Patients with Prostate Cancer. Journal of Urology, 196(6), 1758–1763.
https://doi.org/10.1016/j.juro.2016.06.095
28. Liu, S. V, Dempsey, P. W., Strauss, W., Xu, Y., Xu, T., Pinski, J. K., …
Goldkorn, A. (2013). Targeted next-generation sequencing (NGS) of
circulating tumor cells (CTCs) in hormone-sensitive prostate cancer
(HSPC). Journal of Clinical Oncology, 31(15_suppl), 11040. JOUR.
https://doi.org/10.1200/jco.2013.31.15_suppl.11040
29. Lopez, N. E., Weiss, A. C., Robles, J., Fanta, P., & Ramamoorthy, S. L.
(2016). A systematic review of clinically available gene expression
profiling assays for stage II colorectal cancer: initial steps toward genetic
staging. American Journal of Surgery, 212(4), 700–714.
https://doi.org/10.1016/j.amjsurg.2016.06.019
30. Marusyk, A., Almendro, V., & Polyak, K. (2012). Intra-tumour
heterogeneity: a looking glass for cancer? Nature Reviews Cancer, 12(5),
323–334. https://doi.org/10.1038/nrc3261
31. Merlo, L. M. F., Pepper, J. W., Reid, B. J., & Maley, C. C. (2006). Cancer
as an evolutionary and ecological process. Nature Reviews Cancer, 6(12),
924–935. https://doi.org/10.1038/nrc2013
63
32. Miller, M. C., Doyle, G. V, & Terstappen, L. W. M. M. (2010).
Significance of Circulating Tumor Cells Detected by the CellSearch
System in Patients with Metastatic Breast Colorectal and Prostate Cancer.
Journal of Oncology, 2010, 617421. https://doi.org/10.1155/2010/617421
33. Miyamoto, D. T., Lee, R. J., Stott, S. L., Ting, D. T., Wittner, B. S.,
Ulman, M., … Haber, D. A. (2012). Androgen receptor signaling in
circulating tumor cells as a marker of hormonally responsive prostate
cancer. Cancer Discovery, 2(11), 995–1003. https://doi.org/10.1158/2159-
8290.CD-12-0222
34. Miyamoto, D. T., Zheng, Y., Wittner, B. S., Lee, R. J., Zhu, H., Broderick,
K. T., … Haber, D. A. (2015). RNA-Seq of single prostate CTCs
implicates noncanonical Wnt signaling in antiandrogen resistance.
35. Müller, V., Stahmann, N., Riethdorf, S., Rau, T., Zabel, T., Goetz, A., …
Pantel, K. (2005). Circulating Tumor Cells in Breast Cancer : Correlation
to Bone Marrow Micrometastases , Heterogeneous Response to Systemic
Therapy and Low Proliferative Activity. Clin Cancer Res, 11(5), 3678–
3685. https://doi.org/10.1158/1078-0432.CCR-04-2469
36. Ozkumur, E., Shah, A. M., Ciciliano, J. C., Emmink, B. L., David, T.,
Brachtel, E., … Haber, D. A. (2013). Intertial focusing for tumor antigen-
dependent and -indpendent sorting of rare circulating tumor cells. Sci
Transl Med., 5(179), 1–20.
https://doi.org/10.1126/scitranslmed.3005616.Inertial
37. Rosenberg, R., Gertler, R., Friederichs, J., Fuehrer, K., Dahm, M., Phelps,
64
R., … Siewert, J. R. (2002). Comparison of two density gradient
centrifugation systems for the enrichment of disseminated tumor cells in
blood. Cytometry, 49(4), 150–158. https://doi.org/10.1002/cyto.10161
38. Roychowdhury, S., & Chinnaiyan, A. M. (2013). Advancing Precision
Medicine for Prostate Cancer Through Genomics. Journal of Clinical
Oncology, 31(15), 1866–1873. JOUR.
https://doi.org/10.1200/JCO.2012.45.3662
39. Scher, H. I., Lu, D., & Schreiber, N. A. (2016). Association of ar-v7 on
circulating tumor cells as a treatment-specific biomarker with outcomes
and survival in castration-resistant prostate cancer. JAMA Oncology,
2(11), 1441–1449. JOUR. Retrieved from
http://dx.doi.org/10.1001/jamaoncol.2016.1828
40. Siegel, R. L., Miller, K. D., & Jemal, A. (2016). Cancer statistics. CA
Cancer J Clin, 66(1), 7–30. https://doi.org/10.3322/caac.21332.
41. Singh, D., Febbo, P. G., Ross, K., Jackson, D. G., Manola, J., Ladd, C., …
Sellers, W. R. (2002). Gene expression correlates of clinical prostate
cancer behavior. Cancer Cell, 1(2), 203–209.
https://doi.org/10.1016/S1535-6108(02)00030-2
42. Siravegna, G., Marsoni, S., Siena, S., & Bardelli, A. (2017). Integrating
liquid biopsies into the management of cancer. Nature Reviews Clinical
Oncology. https://doi.org/10.1038/nrclinonc.2017.14
43. Smerage, J. B., & Hayes, D. F. (2006). The measurement and therapeutic
implications of circulating tumour cells in breast cancer. British Journal of
65
Cancer, 94(1), 8–12. https://doi.org/10.1038/sj.bjc.6602871
44. Steinestel, J., Luedeke, M., Arndt, A., Schnoeller, T. J., Lennerz, J. K.,
Wurm, C., … Schrader, A. J. (2015). Detecting predictive androgen
receptor modifications in circulating prostate cancer cells. Oncotarget,
(Table 1), 1–11. https://doi.org/10.18632/oncotarget.3925
45. Stott, S., Lee, R., & Nagrath, S. (2010). Isolation and characterization of
circulating tumor cells from patients with localized and metastatic prostate
cancer. Science Translational Medicine, 2(25), 25ra23.
https://doi.org/10.1126/scitranslmed.3000403.Isolation
46. Subramanian, J., & Simon, R. (2010). Gene expression-based prognostic
signatures in lung cancer: Ready for clinical use? Journal of the National
Cancer Institute, 102(7), 464–474. https://doi.org/10.1093/jnci/djq025
47. Ting, D. T., Wittner, B. S., Ligorio, M., Vincent Jordan, N., Shah, A. M.,
Miyamoto, D. T., … Haber, D. A. (2014). Single-cell RNA sequencing
identifies extracellular matrix gene expression by pancreatic circulating
tumor cells. Cell Reports, 8(6), 1905–1918.
https://doi.org/10.1016/j.celrep.2014.08.029
48. Tsuchiya, N., Matsui, S., Narita, S., Kamba, T., Mitsuzuka, K.,
Hatakeyama, S., … Habuchi, T. (2013). Distinct Cancer-Specific Survival
in Metastatic Prostate Cancer Patients Classified by a Panel of Single
Nucleotide Polymorphisms of Cancer-Associated Genes. Genes &
Cancer, 4(1–2), 54–60. JOUR.
https://doi.org/10.1177/1947601913481354
66
49. Vona, G., Estepa, L., Béroud, C., Damotte, D., Capron, F., Nalpas, B., …
Paterlini-Bréchot, P. (2004). Impact of Cytomorphological Detection of
Circulating Tumor Cells in Patients with Liver Cancer. Hepatology, 39(3),
792–797. https://doi.org/10.1002/hep.20091
50. Weitz, J. (1998). Dissemination of tumor cells in patients undergoing
surgery for colorectal cancer. Clinical Cancer Research, 4(February),
343–348.
51. West, H., & JO, J. (2015). Circulating tumor cells. JAMA Oncology, 1(3),
394. JOUR. Retrieved from
http://dx.doi.org/10.1001/jamaoncol.2015.0711
52. Xu, T., Lu, B., Tai, Y. C., & Goldkorn, A. (2010). A cancer detection
platform which measures telomerase activity from live circulating tumor
cells captured on a microfilter. Cancer Research, 70(16), 6420–6426.
https://doi.org/10.1158/0008-5472.CAN-10-0686
53. Yang, L., Lang, J. C., Balasubramanian, P., Jatana, K. R., Schuller, D.,
Agrawal, A., … Chalmers, J. J. (2009). Optimization of an enrichment
process for circulating tumor cells from the blood of head and neck cancer
patients through depletion of normal cells. Biotechnology and
Bioengineering, 102(2), 521–534. https://doi.org/10.1002/bit.22066
54. Yap, T. A., Gerlinger, M., Futreal, P. A., Pusztai, L., & Swanton, C.
(2012). Intratumor Heterogeneity: Seeing the Wood for the Trees. Science
Translational Medicine, 4(127), 127ps10-127ps10.
https://doi.org/10.1126/scitranslmed.3003854
67
55. Yu, M., Bardia, A., Wittner, B. S., Stott, S. L., Smas, M. E., Ting, D. T.,
… Toner, M. (2013). Circulating breast tumor cells exhibit dynamic
changes in epithelial and mesenchymal compositionand Mesenchymal
Composition, 339(February), 580–584.
56. Zheng, S., Lin, H. K., Lu, B., Williams, A., Datar, R., Cote, R. J., & Tai,
Y. C. (2011). 3D microfilter device for viable circulating tumor cell (CTC)
enrichment from blood. Biomedical Microdevices, 13(1), 203–213.
https://doi.org/10.1007/s10544-010-9485-3
Abstract (if available)
Abstract
Circulating tumor cells (CTCs) are tumor cells that are disseminated from the local site into the vasculature and thereby can be obtained by a simple blood draw from the patients as a biopsy of the tumor. The technology of using CTCs as a source to understand the current state of the tumor has been advancing and there are many studies that have demonstrated the importance of characterizing these cells. Gene expression profiling of CTCs has been a challenge since RNA degrades faster when compared to genomic DNA. Even though there are studies that have demonstrated single cell RNA-seq, the methods used are not robust and reproducible and were also expensive for the large cohort study. Hence we decided to use Parsortix (Angle. Inc, UK) which is a simple, cheap and robust technology to enrich CTCs. One of the challenges that remained even after using Parsortix was the number of background WBCs that were along with the enriched fraction. Therefore, we performed different methods to optimize the enrichment fraction obtained using this technology and determined an efficient method by using Rosette Sep Depletion followed by Parsortix. Therefore, we hypothesized that using this technology in our enrichment process would enhance the ratio of CTCs to WBC we obtain in the fraction and enable more reliable and high-throughput sample analysis for detection of prostate cancer specific transcripts. To test this hypothesis, we decided to use a specific panel of genes that are prostate specific. Before using the method to identify signatures from patient samples, the method optimized were tested using LnCap cells. Further, this assay was validated using patient sample who were diagnosed with prostate cancer.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Circulating tumor cells as biomarkers in lung cancer
PDF
Exploring the effects of CXCR4 inhibition on circulating tumor cell populations in metastatic prostate cancer
PDF
Deconvolution of circulating tumor cell heterogeneity and implications for aggressive variant prostate cancer
PDF
An image-based identification of aggressive circulating tumor cell subtypes
PDF
Tri-specific T cell engager immunotherapy targeting tumor initiating cells
PDF
Exploration of the roles of cancer stem cells and survivin in the pathogenesis and progression of prostate cancer
PDF
Decreased levels of expression of transmembrane protein 56 (TMEM56) in breast cancer tissues
PDF
Role of a novel transmembrane protein, MTTS1 in mitochondrial regulation and tumor suppression
PDF
Peripheral blood mononuclear cell capture and sequencing: optimization of a droplet based capture method and its applications
PDF
Evaluation of preservatives in blood collection tubes for cell-free RNA transcriptional profiles in human plasma
PDF
RNA methylation in cancer plasticity and drug resistance
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
Genetics and the environment: evaluating the role of noncoding RNA in autism spectrum disorder
PDF
The cancer stem-like phenotype: therapeutics, phenotypic plasticity and mechanistic studies
PDF
Modeling and simulation of circulating tumor cells in flow. Part I: Low-dimensional deformation models for circulating tumor cells in flow; Part II: Procoagulant circulating tumor cells in flow
PDF
Role of a novel transmembrane protein, TMEM56, in tumorigenic growth of human PC3 prostate cancer cell line
PDF
Gene expression and angiogenesis pathway across DNA methylation subtypes in colon adenocarcinoma
PDF
Development of immunotherapy for small cell lung cancer using novel modified antigens
PDF
The effect of tumor-mediated immune suppression on prostate cancer immunotherapy
PDF
Developing a bioluminescence tracking system targeting on tumor cells and T cells for cancer immunotherapy
Asset Metadata
Creator
Jojo, Nita Elizabeth
(author)
Core Title
Optimization of circulating tumor cells isolation for gene expression analysis
School
Keck School of Medicine
Degree
Master of Science
Degree Program
Molecular Microbiology and Immunology
Publication Date
07/11/2017
Defense Date
06/20/2017
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
circulating tumor cells,gene expression profiling,low input RNA,OAI-PMH Harvest,Parsortix
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Goldkorn, Amir (
committee chair
), Schönthal, Axel (
committee member
), Zandi, Ebrahim (
committee member
)
Creator Email
jojo@usc.edu,nitaej@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c40-396914
Unique identifier
UC11264316
Identifier
etd-JojoNitaEl-5497.pdf (filename),usctheses-c40-396914 (legacy record id)
Legacy Identifier
etd-JojoNitaEl-5497.pdf
Dmrecord
396914
Document Type
Thesis
Rights
Jojo, Nita Elizabeth
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
Tags
circulating tumor cells
gene expression profiling
low input RNA
Parsortix