Epigenetic Analysis of the Kallikrein Gene Family and Associated Pathways as a Novel Panel of Prostate Cancer Biomarkers

by

Ekaterina Olkhov-Mitsel

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy

Department of Laboratory Medicine and Pathobiology University of Toronto

© Copyright by Ekaterina Olkhov-Mitsel 2015

     

    ii

Epigenetic Analysis of the Kallikrein Gene Family and Associated Pathways as a Novel Panel of Prostate Cancer Biomarkers

Ekaterina Olkhov-Mitsel

Doctor of Philosophy

Department of Laboratory Medicine and Pathobiology

University of Toronto

2015

Abstract

Aberrant epigenetic landscape is a hallmark of prostate cancer (PCa), one of the most

common cancers among men worldwide. Currently, improved understanding of the molecular

mechanisms driving PCa pathogenesis as well as biomarkers for  improved diagnosis, prognosis

and treatment of the disease are needed. Kallikrein related peptidases (KLKs) have emerged as

important cancer biomarkers. However, the mechanisms leading to their deregulation in PCa are

poorly understood.

This thesis characterized the clinical impact of epigenetic mechanisms that regulate KLKs

and associated genes contributing to PCa pathogenesis. Utilizing genome-wide DNA

methylation profiling and quantitative MethyLight technology, novel tumor-specific

hypermethylation of KLK6 and KLK10 was identified in PCa. KLK10 DNA methylation was

associated with pathological stage and longer time to biochemical recurrence in two independent

cohorts of PCa patients. KLK6 hypermethylation was observed in normal-appearing prostate

tissue adjacent to PCa, which could be indicative of the epigenetic field effect. Through the use

of qPCR microRNA arrays a signature of 42 upregulated microRNAs predicted to target KLKs

was identified in association with Gleason score (GS)8 vs. GS6, but no signature was found in

correlation with biochemical recurrence. Among these, the association of increased miR-137

with increasing GS was established in an independent cohort. Notably, expression of miR-137,

     

    iii

predicted to target KLK10, was inversely correlated to KLK10 hypermethylation. The emerging

independent clinical significance of clinicopathologic entities such as intraductal carcinoma

(IDC) and carcinoma with cribriform architecture in PCa prognosis was further explored by

investigating DNA hypermethylation alterations of selected candidate genes including KLK10 in

association with these clinicopathologic entities. Although KLK10 methylation did not correlate

with either IDC or cribriform architecture, the previously discovered methylation of APC,

RASSF1A and TBX15 was significantly linked to these features in PCa. Further, the presence of

large cribriform architecture or HOXD3 hypermethylation was a strong independent predictor of

biochemical recurrence. Thus, this thesis presents a panel of methylated genes that demonstrated

promising prognostic performance and investigated their role in the etiology of PCa.

Accordingly, a novel multiplex MethyLight assay was developed for simultaneous analysis of

three methylated genes with potential applications in research and clinical settings.

     

    iv

Acknowledgments

Данная диссертация посвящается моим любимым и самым дорогим родителям!

This PhD project has been an amazing journey and I will never cease to be grateful to my supervisor, Dr. Bharati Bapat, for this opportunity. I would like to thank you for the limitless training, patience, support, guidance and advice you have given me in all matters of life. It has been a privilege to be mentored by you and instrumental to my growth as a PhD student.

I also gratefully acknowledge my thesis advisory members, Dr. Eleftherios Diamandis and Dr. Theo van der Kwast for their time, patience, intellectual contributions and good humor over the years. Your continuous support, proposed ideas, valuable discussions and constructive suggestions have been absolutely invaluable.

It has been my privilege to work with many collaborators over the years. Mainly, Drs. Zlotta, Fleshner, Siadat and Trudel. I have enjoyed the opportunity to watch and learn from your knowledge and expertise. Thank you for always being very generous with your time, knowledge and samples as well as always providing insightful discussions about the research.

I would like to thank all the teachers, guidance counsellors, professors, and all other acquaintances that have advised me over the years that I will not be able to achieve a high-school diploma, let alone an undergraduate degree. Unknowingly, you have been a great source of motivation for me to work harder and do better.

To the past and present members of the Bapat laboratory, thank you all. Jamie and Liyang, thank you for teaching me experimental techniques as well as providing me with a lot of feedback, tips, and tricks along the way. Sheron and George, thank you for sharing your knowledge with me as well as being so helpful and friendly in my first year in the lab. Ken, you have taught me so much about science, techniques, the lab and life. I could always come and talk to you about anything and everything. For this, I'll always be grateful. Andrea, thank you for talking endlessly with me about my experiments, proofreading my writing, and listening to the ideas that I suggested for your work. But most of all thank you for being such an amazing friend. Fang, I’m thankful for our thought-provoking discussions and newfound friendship. Vaiju, Bas, Linh, Thomas, Julia, Nicole, Carmelle and Shivani, I value the time in the lab we had together and will always carry fond memories. I look forward to keeping in touch with all of you in the future.

This thesis would not have been possible (or half as much fun) without the support of my great friends who have spent this period of my life with me and made Toronto my home. I especially thank my lab partner from first year undergrad chemistry class, Lusia, who has become like a twin sister to me over the years of our post-secondary education.

I would like to acknowledge my parents, Natalia and Vladimir, for raising me and going through two immigrations to provide a better life for my sister and I. Thank you for bearing with me during the challenges, joining me in the triumphs, never damping my sense of curiosity and always having confidence in me. I dedicate this thesis to you both. I also thank my grandparents who have played an important role in the development of my identity and choice to pursue a career in scientific research. I am extremely fortunate to have grandparents who have shown me unconditional love and support in my life. To my wonderful sister, Vika, thank you for your constant love and your willingness to help me in any- and every- thing I do.

Lastly, I would like to thank my husband and my best friend, Leonid. I have warned him on our second date in high school that my ambition in life is to win the Nobel Prize, which will require a remarkable amount of patience and sacrifices. Not only did he continue dating me after this, but he asked my hand in marriage. I feel your love and encouragement every single day. There are not enough words or expressions of gratitude to thank you.

     

    v

Table of Contents

Abstract ......................................................................................................................................... ii

Acknowledgements ..................................................................................................................... iv

Table of Contents ......................................................................................................................... v

List of Abbreviations .................................................................................................................. ix

List of Tables .............................................................................................................................. xiii

List of Figures ............................................................................................................................ xv

Chapter 1: Introduction ............................................................................................................... 1

1.1 The Prostate Gland .................................................................................................................. 1

1.2 Prostate Cancer ........................................................................................................................ 3

1.2.1 Epidemiology of Prostate Cancer ....................................................................................... 3

1.2.2 Prostate Cancer Risk Factors ............................................................................................. 3

1.2.3 Prostate Cancer Prevention ................................................................................................. 6

1.2.4 Familial and Hereditary Prostate Cancer ........................................................................... 9

1.2.5 Screening and Diagnosis of Prostate Cancer ................................................................... 13

1.2.6 Pathological Classification of Prostate Cancer ................................................................ 18

1.2.7 Prostate Cancer Treatment ................................................................................................ 22

1.3 Molecular Pathways to Prostate Cancer ................................................................................ 26

1.4 Epigenetics of Prostate Cancer .............................................................................................. 31

1.4.1 DNA methylation and Its Oxidation Derivatives ............................................................. 33

1.4.1.1 DNA Methylation Aberrations in Prostate Cancer ....................................................... 34

1.4.2 Histone Modifications ....................................................................................................... 39

1.4.3 Non-Coding RNA ............................................................................................................. 40

1.4.3.1 MicroRNAs ............................................................................................................... 40

1.4.3.2 Deregulation of MicroRNAs in Prostate Cancer ...................................................... 45

1.5 Prostate Cancer Biomarkers .................................................................................................. 47

1.6 Kallikrein and Kallikrein-Related Peptidases ....................................................................... 47

1.6.1 Clinical Relevance of Kallikrein Related Peptidases ........................................................ 50

1.6.2 Epigenetics of the Kallikrein Related Peptidases Gene Locus ......................................... 51

1.7 Rationale, Hypothesis and Objectives ................................................................................... 53

Chapter 2: Analysis of Kallikrein-Related Peptidases DNA Methylation in Prostate Cancer

Cell Lines, Tissue and Serum Samples .................................................................................... 54

     

    vi

2.1 Summary ............................................................................................................................... 55

2.2 Introduction .......................................................................................................................... 56

2.3 Materials and Methods ........................................................................................................ 57

2.3.1 Patients and Pathology ...................................................................................................... 57

2.3.2 MassARRAY EpiTYPER Analyses ................................................................................. 59

2.3.3 5-aza-2-deoxycytidine Treatment and RNA Extraction ................................................... 59

2.3.4 Reverse Transcription and RT-qPCR .............................................................................. 59

2.3.5 DNA Extraction, Sodium Bisulfate Modification and MethyLight ................................. 60

2.3.6 Tissue Microarray Construction ....................................................................................... 61

2.3.7 ERG immunohistochemistry ............................................................................................ 61

2.3.8 Statistical Analyses ........................................................................................................... 62

2.4 Results ................................................................................................................................... 66

2.4.1 Identification of Candidate KLKs for DNA Methylation Analyses .................................. 66

2.4.2 MassARRAY EpiTYPER Analyses ................................................................................. 67

2.4.3 DNA Methylation of KLK15 in Primary PCa and Matched Normal Tissue ................... 69

2.4.4 KLK15 DNA Methylation in Serum of PCa Patients and Healthy Controls ................... 71

2.4.5 KLKs 6 and 10 DNA Methylation in Primary PCa and Normal Tissue .......................... 71

2.4.6 KLK10 DNA Methylation in Serum of PCa Patients and Healthy Controls ................... 75

2.4.7 KLKs 6 and 10 DNA Methylation and Clinicopathological Features .............................. 75

2.4.8 ERG Immunohistochemistry ........................................................................................... 77

2.4.9 Correlation of ERG Expression and KLKs 6 and 10 DNA Methylation .......................... 77

2.4.10 KLKs 6 and 10 DNA Methylation and Biochemical Recurrence .................................. 77

2.4.11 Effect of 5-aza-2'-deoxycitidine on KLK Expression in PCa Cell Lines ...................... 81

2.5 Discussion .............................................................................................................................. 83

Chapter 3: Discovery of Alterations in DNA Methylation of KLKs and Associated Pathways

in Prostatic Carcinoma Glands with Cribriform Architecture or Intraductal Carcinoma 90

3.1 Summary ............................................................................................................................... 91

3.2 Introduction .......................................................................................................................... 92

3.3 Materials and Methods ........................................................................................................ 94

3.3.1 Histologic Pattern Annotation .......................................................................................... 94

3.3.2 DNA Methylation Analyses ............................................................................................. 96

3.3.3 Tissue Microarray Construction and ERG Immunohistochemistry ................................. 96

     

    vii

3.3.4 Statistical Analyses ........................................................................................................... 96

3.4 Results ................................................................................................................................... 99

3.4.1 IDC and Cribriform Architecture: Clinical Correlations in GS7 Cases .......................... 99

3.4.2 Aberrant DNA Methylation in GS7 IDC and Cribriform Architecture ......................... 102

3.4.3 DNA Hypermethylation in Areas with GP4 IDC and Cribriform Architecture ............ 105

3.4.4 Prognostic Impact of IDC, Cribriform and DNA Methylation Markers Within GS7 .... 109

3.5 Discussion ............................................................................................................................ 113

Chapter 4: Novel Multiplex MethyLight Protocol for Detection of DNA Methylation in

Patient Tissues and Bodily Fluids ........................................................................................... 118

4.1 Summary ............................................................................................................................. 119

4.2 Introduction ........................................................................................................................ 119

4.3 Materials and Methods ...................................................................................................... 121

4.3.1 Patient Samples and Cell Lines ....................................................................................... 121

4.3.2 DNA Extraction and Sodium Bisulfite Modification ..................................................... 121

4.3.3 MethyLight Assays ......................................................................................................... 122

4.4 Results ................................................................................................................................. 124

4.4.1 Parameters Important for Multiplex MethyLight Assay Design .................................... 124

4.4.2 Analytical Sensitivity and Specificity of the Multiplex MethyLight Assay ................... 125

4.4.3 Accuracy and Reproducibility of the Multiplex MethyLight Assay .............................. 129

4.4.4 Application of the Multiplex MethyLight Assay to Patient Samples ............................. 132

4.5 Discussion ............................................................................................................................ 134

Chapter 5: Assessment of Candidate MicroRNAs Targeting KLKs and their Association

with Prostate Cancer Progression .......................................................................................... 138

5.1 Summary ............................................................................................................................. 139

5.2 Introduction ........................................................................................................................ 140

5.3 Materials and Methods ...................................................................................................... 142

5.3.1 RNA Extraction ............................................................................................................. 142

5.3.2 RT-qPCR miRNA Arrays .............................................................................................. 142

5.3.3 RT-qPCR ........................................................................................................................ 143

5.3.4 MicroRNA Target Gene Prediction ............................................................................... 143

5.3.5 Statistical Analyses ......................................................................................................... 144

5.4 Results ................................................................................................................................. 146

     

    viii

5.4.1 MicroRNA Expression Profiling ................................................................................... 146

5.4.2 MicroRNA-KLK Axis of Interaction .............................................................................. 151

5.4.3 Expression of miR-21 and miR-137 in Primary PCa and Normal Tissue ...................... 151

5.4.4 Expression of miR-21, miR-137 and ERG .................................................................... 153

5.4.5 Correlation Between DNA Methylation, MicroRNA expression and ERG ................... 153

5.5 Discussion ............................................................................................................................ 155

Chapter 6: Discussion and Future Directions ........................................................................ 161

6.1 Epigenetic Marks in Association with PCa Initiation and Progression ............................ 161

6.2 Roles for Epigenetics in the Regulation of KLKs in PCa ................................................. 163

Appendix ................................................................................................................................... 166

Summary of publications ......................................................................................................... 167

References ................................................................................................................................. 168

     

    ix

List of Abbreviations

5-hmC 5-hydromethylcytosine 5ARI 5a-reductase inhibitor ACTB Beta actin ADRB2 Adrenoceptor beta 2, surface ADT Androgen deprivation therapy AFFIRM Safety and efficacy clinical trial of enzalutamide (MDV3100) AGTC Analytical Genetics Technology Centre ALSYMPCA Alpharadin in symptomatic prostate cancer clinical trial ALU Stretch of DNA originally characterized by arthrobacter luteus (Alu)

endonuclease APC Adenomatous polyposis coli AR Androgen receptor AUC Area under the curve BCL2 B-cell CLL/lymphoma 2 BMI Body mass index BPH Benign prostatic hyperplasia BRCA Breast cancer, early onset BSP bisulfite assisted genomic sequencing PCR c-FLIP Cellular ADD-like IL-1β-converting enzyme-inhibitory protein CAGE Cellular (FADD-like IL-1β-converting enzyme inhibitory protein CANT1 Calcium activated nucleotidase 1 CAPB Prostate cancer brain cancer susceptibility locus CCND Cyclin D cDNA Complementary DNA cfDNA Cancer cell-derived DNA CGI CpG Island CHEK2 Checkpoint kinase 2 CI Confidence interval CIMP CpG island methylator phenotype COU-AA-302 Cougar 302 clinical trial CP Cystoprostatectomy CpG Cytosine-guanine dinucleotide CRPC Castration-resistant prostate cancer CTC Circulating tumor cells CV Coefficient of variation CYP Cytochrome P450 CZ Central zone DAB2IP DAB2 interacting protein DGCR8 DiGeorge critical region 8 DHT Dihydrotestosterone DNA Deoxyribonucleic Acid DNMT DNA Methyltransferases

dNTP Deoxynucleotide

     

    x

DRE Digital rectal examination E2F2 E2F transcription factor 2 EGF Epidermal growth factor ERG V-ets erythroblastosis virus E26 homolog eRNA Enhancer RNA ERSPC European Randomized Study of Screening for Prostate Cancer ETS E26 transformation-specific family ETV Ets variant EZH2 enhancer of zeste homolog 2 FDA Food and Drug Administration FFPE Formalin-fixed paraffin embedded FGF Fibroblast growth factor fPSA Free PSA FRET Fluorescence resonance energy transfer GP Gleason pattern GS Gleason score GSTP1 Glutathione S-transferase P1 H&E Hematoxylin and eosin H2A.Z H2A histone family, member Z H3K(x)me1/2/3 Histone 3, lysine mono/di/trimethylation HDAC Histone deacetylases hK Human kallikreins HM High methylation HOXB13 Homeobox protein Hox-B13 HOXD Homeobox D gene HPC1 Hereditary Prostate Cancer 1 locus HPC2/ELAC2 Hereditary prostate cancer 2 HPC20 Hereditary prostate cancer 20 HPCX Hereditary Prostate Cancer, X-Linked HPSE Heparanase HR Hazard ratio HRP Horseradish peroxidase IDC Intraductal carcinoma IGF Insulin like growth factor IL-6 Interleukin 6 IMPACT Identification of men with a genetic predisposition to prostate cancer

trial KLF5 Kruppel-Like Factor 5 KLK Kallikrein and Kallikrein related peptidase LC Large cribriform LHRH Luteinising hormone-releasing hormone LINE-1 Long interspersed elements LM Low methylation LNA Locked nucleic acids

     

    xi

lncRNA Long non-coding RNA LRT Likelihood ratio test MAB Maximum androgen blockade MALDI-TOF MS Matrix-assisted laser desorption/ionization mass spectrometry MAPK Mitogen-activated protein kinase MBD Methyl-CpG-binding domain Mdm-2 Mouse double minute 2 homolog MgCl2 Magnesium Chloride MGMT O-6-methylguanine-DNA methyltransferase miRNA MicroRNA MLH1 MutL homolog 1 MLPA Multiplex ligation-dependent probe amplification MNAzymes Multi-component nucleic acid enzymes MRIGB Magnetic resonance imaging guided biopsy mRNA messenger Ribonucleic Acid MS-LAMP Loop mediated isothermal amplification MSP Methylation-specific PCR MSR1 Macrophage scavenger receptor 1, mTOR Mammalian target of rapamycin MYC v-myc avian myelocytomatosis viral oncogene homolog NBS1 Nibrin NCBI National Center for Biotechnology Information NF-κB Nuclear factor-κB NKX3.1 NK3 homeobox 1 PBC Peripheral blood cell PCa Prostate cancer PCA3 Prostate cancer antigen 3 PCaP Predisposing for prostate cancer PCPT Prostate cancer prevention trial PCR Polymerase chain reaction PHI Prostate health Index PI3K/Akt Phosphoinositide-3-kinase/protein kinase B PIN Prostatic intraepithelial neoplasia PIP3 Phosphatidylinositol (3,4,5)-trisphosphate piRNA Piwi-interacting RNA PIVOT Prostate Cancer Intervention versus Observation Trial (PIVOT) PLAU Plasminogen activator, urokinase PLCO Prostate, Lung, Colorectal and Ovary PMR Percent of methylated reference PP Padlock probes PSA Prostate specific antigen pT Pathological stage PTEN Phosphatase and tensin homolog PZ Peripheral zone

     

    xii

RALP Laparoscopic Robot-assisted Radical prostatectomy RARP Robot-assisted Radical prostatectomy RARβ2 Retinoic acid receptor, beta 2 RASSF Ras association domain family Rb Retinoblastoma RECK Reversion-inducing-cysteine-rich protein with kazal motifs REDUCE Reduction by Dutasteride of Prostate Cancer Events REMARK REporting recommendations for tumor MARKer prognostic studies RISC RNA-induced silencing complex RNASEL Ribonuclease L ROC Receiver-operator curve RP Radical prostatectomy RT Radiation therapy

RT-qPCR Reverse transcription quantitative real-time PCR SC Small cribriform SDS Sequence detection system SELECT Selenium and Vitamin E Cancer Prevention Trial SEPT9 Septin 9 SHOX2 Short Stature Homeobox 2 siRNA Small interfering RNA SLC45A3 Solute carrier family 45, member 3 SMAD4 Mothers against decapentaplegic homolog 4 snoRNA Small nucleolar RNA SNP Single-nucleotide polymorphisms snRNA Small nuclear RNA TBX15 T-Box transcription factor 15 TGFB2 Transforming growth factor, beta 2 tiRNA Transcription initiation RNA TMA Tissue microarray TMPRSS2 Transmembrane protease, serine 2 TNM Classification of malignant tumors, nodes, and metastasis tPSA Total PSA TRUSGB Transrectal ultrasound guided biopsy TZ Transition zone UHN University health network UTR Untranslated region VIM Vimentin vs. Versus Wif-1 Wnt inhibitory factor 1 Wnt Wingless-type MMTV integration site family  

 

 

     

    xiii

List of Tables Chapter 1

Table 1.1. Prostate cancer risk factors ........................................................................................... 5

Table 1.2. Dietary factors and supplements as prostate cancer chemopreventives ....................... 8

Table 1.3. Most commonly associated SNPs with risk of prostate cancer incidence ................. 12

Table 1.4. Pathological stages of prostate cancer tumors in accordance with the American Joint Committee on Cancer .................................................................................................................. 21 Table 1.5. Treatment options for metastatic castration-resistant prostate cancer (CRPC) ......... 25

Table 1.6. Frequently hypermethylated genes in prostate cancer ................................................ 38

Table 1.7. Types of non-coding RNAs ........................................................................................ 43

Table 1.8. Most commonly deregulated miRNAs in prostate cancer .......................................... 46

Chapter 2

Table 2.1. Clinical characteristics of two cohorts of prostate cancer patients ............................. 64 Table 2.2. Primer and probe sequences used in RT-qPCR and MethyLight assay for KLK6, KLK10, KLK15, β-actin and ALU genes ...................................................................................... 65 Table 2.3. KLK15 Median PMR, proportion of high methylation (HM) cases and P-values stratified according to Gleason Score and Pathological stage, Cohort I (2007-2011) ................. 70 Table 2.4. KLK6 and KLK10 median PMR values, proportion of high methylation (HM) cases and P-values for PCa specimens and normal tissues from the same prostate in cohort I (2007-2011) ............................................................................................................................................. 73 Table 2.5. Association between KLK6 and KLK10 methylation levels, age and prostatic zone in normal and cancerous tissues for patients in cohort I (2007-2011) ............................................ 73 Table 2.6. KLK6 and KLK10 Median PMR, proportion of high methylation (HM) cases and P-values stratified according to Gleason Score and Pathological stage, Cohorts I and II ............... 76 Table 2.7. Two independent multivariate Cox regression analyses of biochemical recurrence with Gleason Score, pathological stage, surgical margins, (A) KLK6 and (B) KLK10 methylation status, cohort II (1998-2001) ........................................................................................................ 80

     

    xiv

Chapter 3 Table 3.1. Primer and probe sequences used in MethyLight assay for APC, CYP26A1, HOXD3, HOXD8, KLK6, KLK10, RASSF1A, TBX15, TGFβ2 and ALU genes .......................................... 98 Table 3.2. Proportion of GS7 cases with IDC/LC and P-values stratified according to conventional pathology parameters, age, prostate weight and ERG expression status .............. 100 Table 3.3. Multivariate Cox regression analysis of biochemical recurrence with pathological stage, surgical margin status, preoperative PSA and IDC/LC in GS7 radical prostatectomy specimens ................................................................................................................................... 101 Table 3.4. Median PMR and Mann-Whitney P-values stratified by ERG expression status and further sub-stratified by IDC/LC status in GS7 radical prostatectomy specimens ................... 103 Table 3.5. Median PMR values stratified by IDC and cribriform status ................................... 107 Table 3.6. IDC and cribriform stratified by conventional pathology parameters, age, ERG status, PSA, and prostate weight ................................................................................................ 108 Table 3.7. Five independent multivariate Cox regression analyses of biochemical recurrence with pathological stage, surgical margin status, preoperative PSA, (A) LC, (B) cribriform, (C) HOXD3 methylation, (D) LC vs. HOXD3 methylation (E) LC and/or HOXD3 methylation .... 112 Chapter 4

Table 4.1. Multiplex MethyLight data (PMR) for APC, HOXD3 and TGFB2 in FFPE prostate cancer tissues, post-DRE urine samples of PCa patients as well as fresh frozen tissues and urine samples from PCa-free controls ................................................................................................. 133 Chapter 5

Table 5.1. Clinical characteristics of prostate cancer cases used for miRNA expression profiling .................................................................................................................................................... 145 Table 5.2. Clinicopathological parameters and ERG expression status in cohort I stratified according to cluster A and B on the basis of miRNA expression profiling ............................... 148 Table 5.3. Differentially expressed microRNAs comparing GS8 versus GS6 and recurrent versus non-recurrent cases ......................................................................................................... 149 Table 5.4. Proportion of cases highly methylated for KLK10, highly expressing miR-137 and ERG positive in cohort I ............................................................................................................ 154

     

    xv

List of Figures Chapter 1

Figure 1.1. (a) Prostate zonal anatomy and (b) simplified diagram of prostate cancer progression ..................................................................................................................................... 2  Figure 1.2. Predicted and experimentally verified substrates of prostate specific antigen ......... 17 Figure 1.3. (a) Gleason grading system (b) Intraductal carcinoma (IDC) of the prostate and (c) large cribriform (LC) prostate cancer ........................................................................................... 20 Figure 1.4. Schematic overview of androgen receptor reactivation during prostate carcinogenesis ............................................................................................................................. 30 Figure 1.5. Schematic overview of the main mechanisms of epigenetic regulation ................... 32 Figure 1.6. Schematic overview of microRNA biosynthesis ...................................................... 44 Figure 1.7. Schematic overview of the Kallikreins and Kallikrein-related peptidases (KLK) genomic locus ............................................................................................................................... 49 Chapter 2

Figure 2.1. EpiTYPER analysis of (a) KLK10 for samples 1 and 2, (b) genomic sequence of the region, (c) KLK15 for samples 2 and 3 and (d) genomic sequence of the region ........................ 68 Figure 2.2. Receiver Operator Curve analysis of (a) KLK6 (b) KLK10 (c) KLK15 percent of methylated reference (PMR) values in cohort I (2007-2011) ..................................... 74 Figure 2.3. Kaplan–Meier Curves of biochemical progression-free probability for (a) Gleason score, (b) stage, (c) surgical margin status, (d) KLK6 methylation status, and (e) KLK10 methylation status in cohort II (1998-2001) ................................................................................ 79 Figure 2.4. (1) KLK10 and (2) KLK6 (a) methylation and (b) expression levels following treatment with 5-aza-2’-deoxycytidine in PC-3 and 22RV1 cells .............................................. 82 Chapter 3

Figure 3.1. Schematic representation of study design ................................................................. 95

Figure 3.2. Kaplan-Meier curves of biochemical recurrence probability for (a) preoperative PSA, (b) pathological stage, (c) positive surgical margin status and (d) presence of IDC/LC .. 101 Figure 3.3. Median percent of methylated reference (PMR) values for APC, CYP26A1, HOXD3, HOXD8, RASSF1A, TBX15, TGFβ2, KLK10 and KLK6 genes in the presence and absence of intraductal carcinoma (IDC)/large cribriform (LC) .................................................................. 104

     

    xvi

Figure 3.4. Kaplan-Meier curves of biochemical recurrence probability for presence of (a) cribriform, (b) large cribriform (LC), (c) small cribriform (SC) and (d) intraductal carcinoma (IDC) ......................................................................................................................................... 110 Figure 3.5. Kaplan-Meier curves of biochemical recurrence probability for (a) HOXD3 methylation and (b) presence of large cribriform (LC) .............................................................. 111 Chapter 4

Figure 4.1. Analysis of the analytical sensitivity of the multiplex MethyLight technique for (A) APC, (B) HOXD3, (C) TGFB2 ................................................................................................... 127 Figure 4.2. Analysis of the analytical specificity of APC, HOXD3 and TGFB2 multiplex MethyLight assay ...................................................................................................................... 128 Figure 4.3. Representative analysis of the accuracy of multiplex MethyLight assay ............... 130 Figure 4.4. Analysis of the reproducibility of the multiplex Methylight assay ......................... 131 Chapter 5

Figure 5.1. Unsupervised hierarchical clustering of 32 PCa miRNA expression profiles ........ 147 Figure 5.2. Proportion of miR-137 and miR-21 high expression cases stratified by (a) Gleason score, (b) pathological stage and (c) ERG expression status ..................................................... 154

     

    1

Chapter 1 Introduction

1.1 The Prostate Gland

The prostate is a walnut-sized gland enclosed within a fibromuscular capsule and located in

the pelvis, surrounded by the rectum posteriorly and the bladder superiorly. It is part of the male

reproductive system that secretes fluid, which provides semen with nutritional support and

proteolytic enzymes necessary for fertilization. The gland is composed of 70% glandular

epithelium and 30% fibromuscular stroma, which provides structural support [1]. The prostatic

epithelium is made up of four major cell types: basal, transit amplifying, neuroendocrine and

secretory epithelial cells [2-4]. The stromal component of the prostate, surrounding the gland,

consists mainly of connective tissue and smooth muscle tissue.

The glandular region of the male prostate may be divided into three anatomical zones as

illustrated in Figure 1.1a: peripheral (PZ), central (CZ) and transition zone (TZ) [5]. PZ in young

men constitutes about 70% of the glandular tissue of the prostate and is most prone to tumor

development [6]. CZ occupies 25% of the prostatic glandular tissue and surrounds the

ejaculatory ducts. This region is rarely associated with carcinoma. TZ forms the remaining 5%

of the prostatic glandular tissue surrounding the prostatic urethra proximal to the ejaculatory

ducts and is most susceptible to the development of benign prostatic hyperplasia (BPH) [5, 7].

BPH is among the most common diagnosis in men over the age of 50 and is characterized by

benign increase in the size of the prostate (>30 cm3) caused by hyperplasia of prostatic epithelial

and stromal cells [8]. A small proportion of tumors (<30%) originate in the TZ. Compared to PZ

cancers, they are suggested to be less aggressive and have lower biochemical recurrence rates [6,

9].

 

     

    2

 

Figure 1.1. (a) Prostate zonal anatomy adapted from quizlet.com and (b) simplified diagram of prostate cancer progression.  

 

 

 

 

     

    3

1.2 Prostate Cancer

1.2.1 Epidemiology of Prostate Cancer

With 23,600 estimated new cases in 2014, prostate cancer (PCa) is the leading diagnosis of

cancer in Canadian men and is second only to lung cancer in men worldwide (1.1 million men

diagnosed in 2012) [10, 11]. In terms of mortality, it ranks third in Canada and fifth

internationally as a cause of cancer-related deaths in men. Each year, PCa kills approximately

4,000 men in Canada and 307,000 men worldwide. Given the large discordance between PCa

incidence and death, it is of fundamental importance to understand the clinical heterogeneity of

the disease for screening, prevention and therapeutic strategies.

1.2.2 Prostate Cancer Risk Factors

The only well-established etiological factors associated with PCa development and

progression are non-modifiable and include advancing age, family history and African, African-

American or Afro-Caribbean ancestry [10-14]. There are both hereditary and environmental

components to the underlying biological mechanisms linking these risk factors to PCa onset,

although they are not well understood. Numerous additional PCa risk factors are emerging as

summarized in Table 1.1.

Epidemiological, histopathological and molecular studies have shown that chronic

inflammation and prostatic atrophy, caused by infectious agents, chronic non-infectious

inflammatory diseases and/or other environmental factors, are associated with increased PCa

risk [15-17]. These lesions have been further hypothesized to serve as precursors to PCa as they

have some of the hallmark allelic variants and gene expression changes found in the disease.

Examples include loss of chromosome 8p22, p27 downregulation, c-MYC overexpression and

GSTP1 gene silencing by DNA methylation [15, 18-21]. On the contrary, inflammation on a

negative biopsy was also significantly associated with lower risk of developing PCa [22].

     

    4

Obesity and diet have also been extensively investigated in relation to PCa incidence [23, 24].

The most commonly proposed mechanisms for these associations include oxidative stress, DNA

damage, elevated androgen concentrations, adipokine signaling and the insulin-like growth

factor 1 (IGF-1) pathway [25, 26]. Studies on pre-diagnostic concentrations of circulating sex

hormones and adipocytokines in association with risk of PCa are limited and inconsistent [27-

30]. Therefore, the IGF-1 axis has been the most widely implicated mechanism linking obesity

and diet to PCa tumorigenesis. This is thought to be related to the mitogenic and anti-apoptotic

functions of this pathway [23]. GSTP1, an enzyme involved in stress response and

detoxification, is frequently silenced in PCa by DNA methylation, rendering cells sensitive to

heterocyclic amine carcinogens, thus providing additional explanation for the contribution of

diet to PCa development [31]. However, other studies provide no support for the association of

obesity with PCa incidence [32, 33]. Well-designed clinical trials that will reflect disease

heterogeneity and personalized PCa risk are warranted to validate the numerous proposed risk

factors for PCa.

     

    5

Table 1.1. Prostate cancer risk factors

Risk factor Proposed mechanism of action Level of evidence Reference

Inflammation Production of inflammatory cytokines, chemokines and growth factors Moderate [17]

Obesity

High body mass index and lack of physical activity are associated with endocrine changes including aberrant circulating sex hormone concentrations, adipokine signaling, and insulin-like growth factor 1 (IGF-1) levels

Inconsistent

[34] Red and processed meat

Circulating heme iron, heterocyclic amines, hydrocarbons and nitrosamines (formed through grilling or barbecuing) cause lipid peroxidation and DNA damage

Limited

[24] Dietary fat Interactions with a range of carcinogenic pathways,

most notably causing oxidative stress, DNA damage, and elevated synthesis of circulating androgens

Inconsistent [24]

Dairy Increased levels of circulating testosterone, IGF-1 and calcium (shown to suppress the production of 1,25-dihydroxyvitamin D3, a proposed chemopreventive agent in PCa)

Limited

[24] Low sun exposure

Lower production of 1,25-dihydroxyvitamin D3, a proposed chemopreventive agent in PCa Limited [35]

Cigarette smoking

Aberrant levels of circulating steroid hormones and/or exposure to carcinogens such as cadmium Limited [36]

Alcohol Exposure to carcinogens such as ethanol and/or acetaldehyde Inconsistent [37]

Male Pattern Baldness Elevated levels of dihydrotestosterone Inconsistent [38]

     

    6

1.2.3 Prostate Cancer Prevention

Importantly, PCa onset due to the abovementioned modifiable risk factors is potentially

preventable. Hence, dietary and lifestyle factors have been investigated as potential candidate

chemopreventive agents for PCa initiation and progression. Several dietary antioxidants,

including lycopene, vitamin E and selenium have been investigated with respect to PCa

incidence, suggesting oxidative processes play an important role in PCa neoplastic

transformation [39-41].

Lycopene and vitamin E are antioxidants with a wide range of anticancer properties as

summarized in Table 1.2. Further, selenium is a trace mineral that functions as a cofactor for

various enzymes with anti-tumorigenic functions in PCa cell cycle arrest, apoptosis, and immune

function [42-44]. PCa chemopreventive effects of lycopene, vitamin E and selenium have been

shown in preclinical in vitro and in vivo PCa models and epidemiological studies, but results

from randomized clinical trials have been conflicting [39, 44-51]. Most notably, the Selenium

and Vitamin E Cancer Prevention Trial (SELECT) enrolled 35,533 men to assess the efficacy of

each supplement individually, and in combination, on the incidence of PCa [51]. This large

randomized study found no reduction in PCa incidence. In fact, an elevated risk of PCa was

observed with increased vitamin E consumption (although this did not reach statistical

significance). Vitamin D is another supplement with pro-apoptotic, antiproliferative and anti-

metastatic functions in vitro and in vivo PCa mouse models [52]. However, reports regarding

Vitamin D in PCa chemoprevention are conflicting and additional studies are required to

elucidate its role [53, 54].

Endogenous hormones, especially androgens, are required for normal development and

maintenance of the prostate [55, 56]. Androgens and the androgen receptor (AR) are also

significant mediators of PCa development and progression. Therefore, the potential of androgen-

     

    7

disrupting chemicals in reducing PCa risk has been assessed in numerous studies. Two of the

largest of such randomized, placebo-controlled trials are the Prostate Cancer Prevention Trial

(PCPT) and the Reduction by Dutasteride of Prostate Cancer Events (REDUCE) trial [57, 58].

Both trials focused on 5α-reductase inhibitors (5ARIs), which inhibit the conversion of

testosterone to the more potent dihydrotestosterone (DHT), the primary androgen in the prostate.

The trials showed an overall reduction of about 25% in the relative risk of biopsy-detectable

low-grade PCa. However, there were significantly more high-grade tumors detected in the drug-

treated groups. The cause of this observation is unclear, although it is most commonly thought to

be due to either a detection bias or alternatively, an induction of higher grade PCa by 5ARIs

[59]. In terms of mortality, the most recent PCPT report demonstrated no differences in survival

with up to 18 years of follow-up in both arms of the study [60]. Accordingly, PCa

chemoprevention has been unsuccessful in clinical trials to date, possibly due to limited

understanding of PCa tumor heterogeneity and lack of biomarkers predictive of personalized

PCa risk and/or sensitivity to preventive agents.

     

    8

Table 1.2. Dietary factors and supplements as prostate cancer chemopreventives.

Chemopreventive Proposed mechanism of action Level of evidence

Cruciferous vegetables

Active components (e.g.isothiocyanates) show lowered androgen receptor signaling, promotion of apoptosis, induction of cell cycle arrest, inhibition of tumor invasion and angiogenesis

Limited

Green Tea Inhibition of angiogenesis, promotion of apoptosis Limited Lycopene Inhibition of DNA strand breaks, cell proliferation, reduction of

IGF, androgen receptor signaling, promotion of apoptosis Moderate

Omega-3 Fatty acid

Inhibition of cell proliferation and angiogenesis, promotion of apoptosis Inconsistent

Selenium Antioxidant, induction of cell cycle arrest and promotion of apoptosis Inconsistent

Vitamin A Induction of apoptosis Vitamin D Antiproliferative properties and promotion of apoptosis Limited Vitamin E Antioxidant, induction of cell cycle arrest and promotion of

apoptosis Limited

Modified from Venkateswaran V. and Klotz LH. Nature reviews Urology 2010, 7(8):442-453 [24].

 

     

    9

1.2.4 Familial and Hereditary Prostate Cancer

Most PCa cases are sporadic and only about 10% of cases display familial aggregation of the

disease, suggesting the existence of hereditary PCa [61]. Numerous epidemiological studies

have determined that the lifetime risk for the development of PCa is increased by 2- to 3-fold in

men with a first-degree relative with PCa [62-64]. Risk is further increased in men with family

history of early-onset PCa and multiple relatives with the disease. Segregation studies have

identified familial clustering patterns of PCa that are consistent with the presence of genetic

mutations that confer a Mendelian pattern of disease inheritance [13]. However, highly penetrant

susceptibility genes causing PCa have not yet been identified. Instead, multiple rare genetic

predisposition loci with a small to moderate effect have been shown to contribute to hereditary

PCa, as summarized below.

HPC1/RNASEL mutations

The Hereditary Prostate Cancer 1 (HPC1) locus on chromosome 1q24-25 was the first PCa

susceptibility locus identified in 1996 through genome wide scans of repeat markers in 91

affected families [65]. Large follow-up studies in additional cohorts have since both confirmed

and challenged the linkage of HPC1 with hereditary PCa [66-69]. Subsequent studies have

mapped the HPC1 locus to the to the RNASEL tumor suppressor gene as the causative factor in

the development of PCa [70, 71]. Others have found inactivating mutations of RNASEL to confer

only a modest PCa risk, suggesting HPC1 might play a role only in a subset of familial PCa

cases [72].

BRCA gene mutations

Mutations in BRCA1 and BRCA2 genes have been shown to confer high risk of PCa [73, 74].

PCa risks by age 65 have been estimated at 1.8-4.5-fold higher in BRCA1 germline mutation

carriers and 2.5-8.6-fold for BRCA2 mutation carriers [75-79]. Additionally, BRCA2 mutation

     

    10

carriers have been shown to present early-onset and aggressive disease characterized by higher

rates of lymph node involvement, distant metastasis, and a higher mortality rate compared to

non-carriers [80-83]. A recent report from the initial screening round of the Identification of

Men with a genetic predisposition to Prostate Cancer: Targeted screening in BRCA1 and 2

mutation carriers and controls (IMPACT) study has also demonstrated higher incidence of

intermediate- and high-risk PCa in BRCA2 mutation carriers [84].

HOXB13 mutations

Genome wide association studies have identified recurrent germline variations in the 17q21-

22 region as a locus associated with increased risk of hereditary PCa. This region was then

mapped to the HOXB13 gene and showed to harbor changes of glycine to glutamic acid (G84E)

[85]. The carrier frequency for this variant has been estimated at 0.1-4.9% and has been shown

in multiple cohorts to confer around 3-10-fold increased risk for the development of PCa,

particularly early-onset PCa [86, 87]. The association of HOXB13 G84E variant with PCa risk

has been further validated in the International Consortium for Prostate Cancer Genetics (ICPCG)

[88]. The 17q21-22 region also harbors the SPOP gene that has been suggested to incur a

germline missense mutation of asparagine to isoleucine (N296I) that is associated with

hereditary PCa [89].

Other significant loci

The predisposing for prostate cancer (PCaP) locus on chromosome 1q42.2–43 was the

second PCa susceptibility gene identified in southwestern European populations, but has not

been confirmed by other groups [90]. Similarly, carriers of mutations in mismatch repair genes,

HPCX, HPC20, HPC2/ELAC2, MSR1, NBS1, and CHEK2 have also been suggested to have

increased risk of PCa but were not confirmed by subsequent studies [91-96]. Therefore, their

role in hereditary PCa still needs to be confirmed. Additionally, CAPB, a gene located on

     

    11

chromosome 1p36, was shown to confer susceptibility to young onset of PCa in families with

relatives suffering from brain tumors [97].

Single Nucleotide Polymorphisms

Over the past decade numerous genome wide association studies have discovered 77

germline single nucleotide polymorphism (SNPs) to be commonly associated with a modest

(odds ratio 1.1-1.3) increase in risk of PCa incidence [98-104]. The top most cited SNPs are

listed in Table 1.3. Over 1,000 additional SNPs have been suggested to contribute to PCa risk

and need to be further investigated. Additionally, SNP variants have been reported to be

associated with aggressive PCa [105]. Yet, collectively SNPs are estimated to explain less than

30% of the familial risk for the disease. The region with the highest number of independently

associated variants with PCa risk is 8q24, near the MYC oncogene. Some of these SNPs are

functional and exert long-range tissue-specific regulation of MYC expression [106-108]. These

SNPs have also been shown to have race-dependent differences in the degree and frequency of

PCa incidence [109]. Further, risk prediction models have suggested that certain SNPs appear to

act multiplicatively with each other to confer a cumulative and significant 2.1-4.7-fold increase

in relative risk of PCa, which could direct PCa screening programs [100]. One notable example

is the prostate genetic score-33, calculated from analysis of 33 PCa risk-associated SNPs, which

has been shown to be a significant independent predictor of PCa risk, with an odds ratio of up to

2.1 when combined with PSA [110]. However, further validation is needed.

     

    12

Table 1.3. Most commonly associated SNPs with risk of prostate cancer incidence

Genomic position SNP Gene Reference 2p15 rs6545977 EHBP1 [100, 111] 4q24 rs7679673 TET2 [100, 111] 5p15.33 rs12653946 LPCAT1 [100, 112, 113] 5p15.33 rs2242652 TERT [114] 6p21.1 rs1983891 FOXP4 [100, 112, 113] 6q22.2 rs339331 RFX6 [100, 112, 113] 8p21.2 rs1512268 NKX3.1 [100, 111-113] 8q24.21 Numerous (example: rs6983267) MYC [100, 106-108] 9q31.2 rs817826 KLF4 [100, 104] 10q11 rs10993994 MSMB [106, 115] 11q13.2 rs10896449 CCND1 [100, 116] 17q21.2 rs7501939 and rs4430796 HNF1B [106, 111-114] 17q25.1 rs1859962 BC039327 [106, 111, 114, 117] 19q13.3 rs2735839 KLK15-2 [106] Xq12 rs5919432 AR [100, 114]

 

 

 

 

 

 

     

    13

1.2.5 Screening and Diagnosis of Prostate Cancer

The current standard of diagnosis for PCa is the histopathological examination of a transrectal

ultrasound guided biopsy (TRUSGB) [118]. Recently, magnetic resonance imaging guided

biopsy (MRIGB) has also been suggested as an additional diagnostic tool to detect clinically

significant PCa in patients with previous negative TRUSGB [119].

Screening for PCa is generally conducted via digital rectal exam (DRE) and Prostate-Specific

Antigen (PSA) testing, although pre-biopsy imaging using MRI and/or ultrasound is also

emerging. Over the past two decades, PSA screening has become a widely used non-invasive

clinical test for PCa; its widespread application has dramatically changed the disease landscape,

enhancing PCa detection at an early stage [120]. PSA, or KLK3, is an androgen-regulated serine

protease and a member of the Kallikrein-related peptidases (KLK) gene family, also known as

human kallikreins (hKs) [121-123]. It is expressed primarily in the luminal epithelium of the

prostate, secreted into the prostatic ducts and subsequently to the seminal plasma where it

cleaves semenogelin I and II for semen liquefaction [124, 125]. Other substrates for PSA have

also been described with respect to PCa development and progression (Figure 1.2) [126].

PSA is consistently expressed in normal prostate epithelium and is downregulated in PCa

tissue [127, 128]. PSA secretion to the circulation, however, is increased in PCa patients. The

exact mechanism of elevated PSA secretion in PCa has not yet been elucidated but it is thought

to occur through the disruption of the basal-cell layer [129]. Notably, with the development of

PSA as a PCa biomarker came the concept of biochemical recurrence, generally defined as a

serum PSA >0.2 ng/mL following treatment [130]. Biochemical recurrence precedes clinical

recurrence by many years, and thus is often used as a surrogate endpoint to evaluate treatment

efficacy. …

…Besides PCa, infection, irritation, prostatitis, urinary tract infection, BPH and medication,

     

    14

among other environmental factors, may elevate serum PSA levels [131, 132]. This limits the

specificity of the serum PSA test. An additional limitation of the PSA test is the lack of a

sufficiently low level of the biomarker at which PCa can be ruled out. Serum PSA levels are a

continuous parameter: the higher the value, the higher the risk of PCa. Thus, the appropriate

PSA threshold to indicate the need for prostate biopsy is under debate. PSA threshold of 4

ng/mL has long been regarded as an indicator for prostate biopsy. However, this threshold has a

positive detection rate of only 20-50%, resulting in a large number of negative biopsies [110,

133]. Further, PSA screening identifies a large number (7-56%) of indolent tumors [134-137].

Yet, since the natural history of screen-detected PCa remains poorly understood, many indolent

tumors receive treatment due to uncertainty. Therefore, overdiagnosis and overtreatment are

considered the most significant potential harms of PSA screening. PSA threshold of 4 ng/mL

has also been shown to miss 10-30% of PCa tumors, which may contribute to the continuous

mortality from the disease [133, 138, 139]. Lowering the PSA threshold has been suggested to

increase the sensitivity of the test [140]. However, studies have shown this also reduces the

specificity of the test, resulting in increased number of negative biopsies and indolent PCa

diagnoses. Thus, the effectiveness of PSA screening has become a controversial topic.

Two large randomized trials have recently assessed the effectiveness of PSA screening: the

Prostate, Lung, Colorectal and Ovary (PLCO) trial and the European Randomized Study of

Screening for Prostate Cancer (ERSPC) [141, 142]. The PLCO study, a US trial of 76,685 men

with 13 years of follow-up, found no mortality benefit with PSA screening. While the ERSPC

study of 182,160 men with 11-year follow-up, reported 20% reduction in mortality in the PSA-

screened group. Further, the results of the ERSPC trial indicated that to prevent one death from

PCa, 1055 men had to be screened and 37 had to be treated [142, 143]. A recent report from the

Gotenborg arm of the ERSPC study, which followed 20,000 men for 14 years, demonstrated that

     

    15

in their cohort only 293 men had to be screened and 12 had to be treated to avoid one PCa death

[144]. The seemingly conflicting results of these studies have been extensively reviewed and the

most commonly cited explanations include contamination with PSA testing in the control group,

the PLCO recruitment bias and insufficient follow-up period [145]. Based on the results of

PLCO and ERSPC studies, widespread mass screening for PCa using the PSA biomarker is not

recommended. In addition, the Canadian Task Force on Preventive Health Care recommends

against PSA testing to screen for PCa [146].

Additional improved screening methods have evolved in the past decade, with the

identification and validation of several new biomarkers, including measurements of different

molecular forms of PSA and PSA kinetics. Total PSA (tPSA) refers to the sum of free or

unbound PSA (fPSA) and PSA bound to other proteins. fPSA tends to be upregulated in BPH

compared to PCa, thus, the measurement of %fPSA and the ratio of fPSA-to-tPSA have shown

to improve the diagnostic specificity of PSA [147]. Additionally, several molecular forms of

ProPSA, non-catalytic preforms of the PSA enzyme, have been shown to have better accuracy at

detecting PCa and distinguishing indolent from aggressive PCa [148]. However, the use of these

variations of the PSA test have not yet achieved the sensitivity and specificity required to be

applicable in everyday clinical practice.

Recently, an FDA approved commercial test named Beckman Coulter Prostate Health Index

(PHI), which combines tPSA, fPSA and the [-2] form of proPSA into a single score, has been

reported [149, 150]. The test has demonstrated a great potential for clinical utility as it

outperformed tPSA and fPSA in numerous studies. Another multivariate model that has shown

high accuracy to predict risk of high-grade PCa on biopsy is called OPKO 4Kscore™ PCa Test

[151]. This four-kallikrein panel, which incorporates tPSA, fPSA, %fPSA and hK2, is in

commercialization phases and may soon be available in the clinic.

     

    16

High PSA density, referring to PSA concentration adjusted for prostate volume, has also been

suggested as an accurate predictor of PCa diagnosis, aggressiveness, recurrence and mortality

[152, 153]. Similarly, PSA kinetics, particularly PSA doubling-time, has also shown to enhance

the predictive value of PSA. Studies have demonstrated that shorter PSA doubling-time was

associated with increased risk of aggressive PCa and PCa-specific mortality [154, 155]. Yet, the

utility of PSA density and PSA doubling time is limited as their added benefit to the existing

PSA test is still under debate.

Prostate cancer antigen 3 (PCA3 also known as DD3) is another FDA approved urinary

biomarker that is commercially available as a PCa diagnostic test [156]. This non-coding,

prostate-specific RNA is overexpressed 10-100-fold in approximately 95% of PCa patients

compared to healthy or BPH patients. A PCA3 score of >35 in urine has been shown to have an

average sensitivity of 66% and specificity of 76%, which was an improvement on serum PSA

(65% sensitivity and 47% specificity) in the same study [157]. This biomarker has also shown

further utility in the decision-making process with regard to a repeat biopsy in men with a

negative first biopsy but a persistent suspicion of PCa.

Statistical models, known as nomograms, have also been suggested to increase the accuracy

of individualized PCa risk assessment tools. Nomograms such as the ones developed by Nam,

PCPT and Chun incorporate established PCa risk factors with PSA and one or more of the

additional abovementioned biomarkers to calculate the probability of PCa upon biopsy [158-

160]. Nomograms have shown improved diagnostic accuracy, over PSA screening alone, to

predict positive biopsies in a number of studies and some are available on-line for patients and

physicians. However, they have not received widespread acceptance in clinical decision-making.

It has been postulated that adding biomarkers specifically associated with PCa biological

behaviors might further improve the accuracy of existing nomograms.

     

    17

Figure 1.2. Predicted and experimentally verified substrates of prostate specific antigen (PSA). *Semenogelin I and II are considered to be the physiological substrates of PSA.

     

    18

1.2.6 Pathological Classification of Prostate Cancer

The most widely accepted pathological classification of PCa is the Gleason score (GS),

which is based on architectural features of the gland. The GS grading system, originally

developed in 1966 by DF Gleason and later modified in 2005 at the International Society of

Urological Pathology Consensus Conference, is currently one of the best prognosticators for

PCa [161-163]. In the GS grading system, all tumors fall into a 5-grade Gleason pattern (GP)

representing a continuum of progressively complex morphologies as illustrated in Figure 1.3a.

Each tumor is assigned a GS based on the sum of the two most prevalent GPs representing the

wide variation in morphology within a single PCa tumor. Assignment of GS 2-5 is rather

obsolete. Patients with GS ≤6 prostate tumors are considered low grade and have a good

prognosis, GS7 are intermediate grade and have variable outcomes, while GS ≥8 are high grade,

and generally have a poor outcome. Importantly, the presence of any GP 4 or 5, even in limited

amount, is typically considered clinically significant as it predicts more rapid progression and

suggests aggressive treatments are needed. However, the GS does not provide information on

altered molecular pathways and therapeutic targets for anticancer therapy selection.

An emerging concept in the evolution of the GS is sub-stratification of the heterogeneous

high grade GPs, consisting of a broad range of histologic architectures and clinical outcomes, to

better predict PCa prognosis. Subtypes of histologic architectures in high grade GPs include

small and large fused glands, ill-formed glands embedded in loose or desmoplastic stroma, as

well as glomeruloid, mucinous, small and large cribriform (SC and LC) growth patterns (Figure

1.3c). Further, intraductal carcinoma (IDC) of the prostate has also been associated with poorer

prognosis (Figure 1.3b)[164-167]. The clinical importance and prognostic significance of each

of these architectural patterns, independently and in combination, needs further validation to

determine the utility of their potential incorporation into standard pathology reporting on biopsy.

     

    19

In addition to the GS, surgically removed primary prostate tumors are assigned a

pathological T (pT) stage. pT staging is assigned upon pathologic examination of the resection

specimen and ranges from pT2 to pT4, based on tumor size and extent of invasiveness as

outlined in Table 1.4. Tumor pT staging has also been established as a significant prognosticator

for PCa [168].

     

    20

Figure 1.3. (a) Gleason grading system illustration adapted from Harnden P. et al. The Lancet Oncology 2007, 8(5):411-419 [169] (b) Intraductal carcinoma (IDC) of the prostate and (c) Large cribriform (LC) prostate cancer adapted from Trudel D. et al. European Journal of Cancer 2014, 50(9):1610-1616 [170].  

     

    21

Table 1.4. Pathological stages of prostate cancer tumors in accordance with the American Joint Committee on Cancer  

Pathologic (pT) staging pT2 Organ confined pT2a Unilateral, one-half of one side or less pT2b Unilateral, involving more than one-half of side but not both sides pT2c Bilateral disease pT3 Extraprostatic extension pT3a Extraprostatic extension or microscopic invasion of bladder neck pT3b Seminal vesicles invasion pT4 Invasion of rectum, levator muscles, and/or pelvic wall

Modified from Cheng L. et al. Histopathology 2012, 60(1):87-117.

 

 

     

    22

1.2.7 Prostate Cancer Treatment

PCa is a spectrum of diseases with various stage-specific treatment options [171, 172].

Clinically indolent PCa, thought to pose minimal risk to patient’s life, is commonly monitored

by active surveillance to delay or avoid unnecessary definitive treatment. Criteria for placing

patients on active surveillance vary between institutions, but generally include GS ≤6, clinical

stage T1/T2, PSA ≤10 ng/ml and <50% biopsy cores positive for PCa [173, 174]. Patients on

active surveillance are rigorously monitored and curative treatment is recommended upon

detection of higher-risk features. Active surveillance studies have demonstrated that patients

with low risk disease in the program had comparable rates of biochemical recurrence to those

who received radical treatment [175]. Thus, this approach appears to be safe in the first 10 to 15

years following diagnosis. Additional extended follow-up is underway to obtain definitive

results on this approach.

Localized PCa is generally treated by radical prostatectomy (RP), radiation therapy (RT)

and/or androgen deprivation therapy (ADT). Additional therapeutic options for the disease are

emerging and include proton-beam and carbon ion–beam therapy, cryosurgical ablation of the

prostate and high-intensity focused ultrasound.

RP, including robot-assisted RP (RARP) and laparoscopic RARP (RALP), is a surgical

treatment available for localized PCa. RP has shown a PCa specific survival benefit in patients

younger than 65 with intermediate and high risk of disease progression. Therefore, most patients

with localized PCa undergo RP. However, there is a growing concern that PSA screen-detected

PCa patients are overtreated and unnecessarily suffer a significant amount of morbidity

associated with RP including bleeding, infection, urinary incontinence and erectile dysfunction.

Coordinately, the recently published Prostate Cancer Intervention versus (vs.) Observation Trial

(PIVOT) showed that patients with PSA-screen detected localized PCa who underwent RP did

     

    23

not have a significant reduction in all-cause or PCa-specific mortality compared with those who

opted for active surveillance over a 12-year follow-up [176, 177]. On the other hand,

biochemical and local or systematic recurrence after RP is observed in 15-30% of men that

require more aggressive treatment. Taken together, this highlights the need to better customize

treatment for PCa patients, ideally based on molecular therapeutic targets.

RT is an alternative to surgical intervention. It can be achieved through “seeded” radiation

(brachytherapy) or external beam therapy. Brachytherapy, the planting of a radioactive isotope

into the prostate, is typically only recommended for patients with low risk of PCa progression.

Long-term data of brachytherapy in these patients is promising benefits from treatment [178].

For PCa patients with intermediate risk of biochemical recurrence after RP, external beam RT is

usually offered. Meanwhile, high-risk patients are recommended to undergo ADT before and

during RT, as it results in increased overall survival [179, 180].

ADT, first identified by the Nobel laureate Dr. Huggins in 1941, is the standard first-line

therapy for locally advanced, recurrent or metastatic PCa [181]. It takes advantage of the fact

that PCa is a hormonally sensitive cancer and works by blocking circulating androgens essential

for PCa cell growth. ADT can be achieved through bilateral orchiectomy or pharmacotherapy.

Luteinising hormone-releasing hormone (LHRH) is produced by the pituitary gland and

regulates androgen production in the testes, adrenal glands and peripheral tissues such as the

prostate [182]. Therefore, LHRH agonists and antagonists have become the standard of care in

ADT [171]. Long-term administration of LHRH agonists causes downregulation of the pituitary

receptors for LHRH, which leads to testosterone suppression. On the other hand, LHRH

antagonists block LHRH receptor signaling causing a rapid decrease in testosterone levels.

Antiandrogens, both steroidal and non-steroidal, are another ADT treatment option in locally

advanced PCa [183]. However, the clinical benefits of this treatment alone remain marginal, thus

     

    24

it is not the recommended standard of care. Non-steroidal antiandrogens are generally given in

conjunction with LHRH agonists to achieve maximum androgen blockade (MAB), which has

been shown to provide survival advantage but has significant negative side effects such as hot

flashes, sexual and cognitive dysfunction [184]. Alternatively, intermittent ADT with alternating

phases of androgen suppression and treatment cessation, to allow androgen recovery between

phases, has shown similar survival benefits to MAB but with better tolerability and quality of

life [185].

Although ADT treatment is initially effective at slowing PCa progression, virtually all cases

eventually recur in a form that is unresponsive to such treatments. Such castration-resistant PCa

(CRPC) cases are currently incurable, but numerous treatments are recommended for disease

management (Table 1.5). These treatments generally reduce pain and skeletal-related events as

well as modestly prolong survival.

Currently, there is no specific sequence of therapy selection for PCa patients and physicians

generally adhere to PSA, staging or histopathological criteria for treatment selection. Therefore,

there is a need for biologically meaningful molecular classification of PCa subgroups that will

define specific stages of disease progression, direct personalized therapy selection and underpin

response or resistance to therapy.

     

    25

Table 1.5. Treatment options for metastatic castration-resistant prostate cancer (CRPC)

Treatment option Mechanism of action Reference Abiraterone acetate

* CYP17A inhibitor that blocks androgen synthesis * Demonstrates overall survival advantage * COU-302 trial [186-188]

Denosumab * Monoclonal antibody against the receptor activator of nuclear factor κ-B ligand * Demonstrates delay in skeletal-related events and prolong bone metastases–free survival [189, 190]

Docetaxel and/or Cabazitaxel

* Chemotherapy, may be used in combination with prednisone * Demonstrates overall survival advantage [191-193]

Enzalutamide (formerly MDV3100)

* Androgen receptor antagonist * Demonstrates overall survival advantage * AFFIRM trial [194-196]

Radium-223 * Radiopharmaceutical, acts as a calcium mimic and targets new bone growth in and around bone metastases to kill cancer cells * Demonstrates overall survival advantage * ALSYMPCA phase III clinical trial [197]

Sipuleucel-T *Immunotherapy - peripheral blood mononuclear antigen-presenting cells, collected from the patient and exposed to a fusion protein (prostatic acid phosphatase and granulocyte-macrophage colony-stimulating factor) to create loaded antigen presenting cells that then get reinfused to the patient. * Demonstrates overall survival advantage * Phase III IMPACT trial [198, 199]

Zoledronic acid * Demonstrates significantly delay and/or reduction in the incidence of both pain and the number of skeletal-related events [200]

 

 

 

     

    26

1.3 Molecular Pathways to Prostate Cancer

Somatic alterations let prostatic cells acquire hallmark capabilities shared by all cancers as

outlined by Hanahan and Weinberg in their landmark article “The Hallmarks of Cancer” [201,

202]. These hallmarks include self-sustained proliferative signaling, angiogenesis, genomic

instability, inflammation, deregulation of energy metabolism, evasion of growth suppressors,

immune destruction and apoptosis, enabled replicative immortality, invasion and metastasis.

However, despite the extensive research on the molecular pathways that lead to the development

and progression of PCa, the mechanisms underlying PCa tumorigenesis are not fully understood.

This challenge has been further complicated by the heterogeneity and multifocality of PCa.

AR is a nuclear receptor that, upon activation by androgens, mediates transcription of target

genes that are pivotal for normal prostate growth and development [56]. Further, the AR is a key

pathway in PCa initiation, tumor cell proliferation, inhibition of apoptosis and response to ADT

therapy [203, 204]. Accordingly, AR knockout mice and humans with inactivating mutations of

AR do not develop prostates or PCa [205].

AR signaling is overactive in PCa, thus ADT is an effective treatment that leads to PCa tumor

regression. However, tumor cells exposed to ADT undergo alterations that reactivate the AR

pathway despite the low levels of circulating androgenic ligands. Numerous mechanisms have

been implicated in the reactivation of AR signaling during PCa progression as illustrated in

Figure 1.4, including (1) intracrine androgen production and overexpression of enzymes

involved in the synthesis of steroids and androgens (2) overexpression of AR co-regulators (3)

AR amplification and overexpression (4) gain of function mutations in AR (5) promiscuous AR

ligand response (6) androgen independent AR activation (7) constitutive activation of certain AR

splice variants [206].

     

    27

Overactivation of the AR pathway leads to overexpression of many downstream target genes

such as PSA, c-FLIP and Mdm-2 [123, 207, 208]. Therefore, PSA screening is used as a readout

of AR activity for assessment of PCa progression and therapeutic outcome in the clinic. NKX3.1

is an example of androgen-regulated tumor suppressor gene, frequently lost in PCa and even

more so in CRPC [209-211]. The loss of function or deletion of this homeobox gene appears to

be an early event in PIN lesions and contributes to oxidative damage associated with PCa

initiation.

Additionally, chromosomal translocations that produce fusion transcripts between androgen-

regulated genes such as TMPRSS2, KLK2, CANT1, SLC45A3, or AX747630 and oncogenes such

as the ETS family transcription factors are prevalent in PCa [212]. Of these, fusions between

TMPRSS2 and ETS gene family (e.g. ERG, ETV1 and ETV4) are most common. Notably, the

TMPRSS2-ERG fusion which is present in approximately 50% of tumors often arises from

deletion of an interstitial fragment of chromosome 21, resulting in the AR responsive promoter

of TMPRSS2 driving the expression of the ERG oncogene [213-215]. This fusion is considered

an early event in PCa development, occurring early in PIN lesions, and has been suggested as a

diagnostic biomarker in PCa [216]. Further, the association of TMPRSS2-ERG fusions with PCa

progression and biochemical recurrence is controversial with some showing an association

whereas many other reports failing to do so [217, 218]. ERG overexpression has been shown to

stimulate cell migration and invasion, while its knockdown decreases the invasive properties of

PCa cells [219, 220]. Additionally, mouse models of TMPRSS2-ERG overexpression develop

PIN lesions, but require concomitant activation of additional pathways to develop PCa [221,

222]. Taken together this suggests chromosomal rearrangements as critical initiating events in

PCa.

     

    28

The phosphoinositide-3-kinase/protein kinase B (PI3K/AKT) pathway is another

overactivated pathway, found in approximately 40% of PCa and 70% of metastatic CRPC [223-

228]. This occurs most often though deletion mutations in the Phosphatase and tensin homolog

(PTEN) tumor suppressor gene, a genetic change that has been detected in PIN lesions but is

often a late event in PCa. Loss of PTEN leads to buildup of Phosphatidylinositol (3,4,5)-

trisphosphate (PIP3) and subsequent activation of AKT and mammalian target of rapamycin

(mTOR) signaling along with inhibition of p27, thus promoting cell cycle progression and

survival [229, 230]. Studies have shown that in some cases activated PI3K-AKT-mTOR

pathway is sufficient to compensate for AR signaling blockade by ADT [231, 232]. Therefore,

combination therapy with ADT and PI3K pathway inhibitors are being currently explored in

clinical trials.

In addition to the AKT/mTOR signaling, mitogen-activated protein kinase (MAPK) signaling

is also frequently activated in PCa and CRPC [233-236]. This has been shown to promote PCa

progression and CRPC in cell lines and mouse models. Similarly, other tyrosine kinase signaling

pathways, particularly Her2/Neu and SRC, have been implicated in aggressive PCa, CRPC, and

metastasis [237].

Numerous developmental pathways that govern prostate embryonic development have been

implicated in PCa initiation, progression to CRPC and epithelial-mesenchymal transition.

Examples of deregulated developmental pathways during prostate tumorigenesis include sonic

hedgehog, WNT and notch signaling as well as gene products of homeobox, forkhead and bone

morphogenetic proteins [238-241]. Growth factor signaling through TGFβ, EGF, IGF, FGF and

IL-6 as well as nuclear factor-κB (NF-κB) transcription factor members also play an important

role in the development and progression of PCa [241, 242]. Accordingly, clinical trials for

     

    29

therapeutic approaches that block each of the pathways mentioned in this section, independently

and in combination with ADT are ongoing.

It is becoming clear that besides the molecular pathways leading to PCa within epithelial

cells, changes in the tumor microenvironment are another major mechanism in PCa progression.

However, the specific mechanisms remain to be elucidated. During carcinogenesis, dynamic

interactions between epithelium, cells normally restricted to the stroma (smooth muscle cells,

fibroblasts, bone marrow-derived mesenchymal stem cells) and various inflammatory cells have

been shown to occur and promote the neoplastic phenotype. Most notable examples include

phenomena such as the inhibition of PCa growth when co-inoculated with normal prostatic

fibroblasts and the stimulation of PCa growth with cancer-associated or spontaneously-

immortalised fibroblasts [243-245].

     

    30

Figure 1.4. Schematic overview of androgen receptor reactivation during prostate carcinogenesis.

     

    31

1.4 Epigenetics of Prostate Cancer

Epigenetics is the study of reversible, heritable mechanisms that regulate gene expression

without altering the DNA sequence [246, 247]. Epigenetic inheritance refers to both mitotic and

meiotic transmission of gene expression states [248]. Although the significance of somatic

epigenetic inheritance (or epigenetic memory) has been well established in cell lineage, less is

known about transgenerational inheritance of epigenetic marks.

The main mechanisms of epigenetic regulation include DNA methylation and its oxidation

derivatives, histone modifications and non-coding RNAs (Figure 1.5). The interplay of these

modifications creates an epigenetic landscape that regulates normal cell biology. Alterations in

the epigenetic landscape are a hallmark of PCa with a potential for clinical translation [249]. The

reversibility of epigenetic mechanisms has gained increasing interest as therapeutic targets. At

present, epigenetic therapy has been established as a successful treatment approach for

hematological malignancies and research is underway for epigenetic therapy in solid tumors

[250-252]. Although epigenetic therapy is beyond the scope of this thesis, its relevance to the

topic of epigenetic biomarker research is important to note.

     

    32

Figure 1.5. Schematic overview of the main mechanisms of epigenetic regulation. Adapted from Ho LT. Genome-wide Distribution and Regulation of DNA Methylation and Hydroxymethylation in Prostate Cancer. Unpublished master's thesis 2014, University of Toronto, Toronto, Canada.

     

    33

1.4.1 DNA methylation and Its Oxidation Derivatives

DNA methylation is one of the most well studied epigenetic mechanisms in mammals. It

refers to the addition of a methyl group to the 5th carbon of a cytosine (5mC) that usually

precedes a guanine (CpG). Non-CpG methylation, including methylation of cytosines at CpN

and CpNpG (where N is A, T or C) also occurs, but is rare [253, 254]. Frequently, but not

exclusively, CpG dinucleotides occur in CG-rich DNA stretches known as CpG islands (CGIs)

[255]. CGIs are generally defined as DNA regions of 200 base pairs in length with a CG content

>50% and an observed/expected CpG ratio >0.6 [256]. The majority of CGIs are located at the 5'

region of genes and occupy approximately 60% of human gene promoters. While most CpG

sites in the genome are methylated, only about 10% of CGIs are methylated in normal cells. CGI

methylation is associated with long term transcriptional silencing such as X chromosome

inactivation, imprinting and silencing of certain germ cell-specific or tissue-specific genes [257,

258]. Thus, mutations in genes that regulate epigenetic changes across the genome and

alterations in genetic imprinting are associated with diverse forms of human disease [259].

DNA methylation-mediated gene silencing may be achieved by blocking DNA binding

proteins or by recruiting methyl-CpG-binding domain (MBD) proteins, which recruit histone

deacetylases (HDACs) and promote heterochromatin formation [260-262]. Alternatively, CGI

methylation within gene bodies is a feature of transcribed genes [263]. It is postulated to regulate

alternative promoters, splicing, activating elements and transcriptional elongation. Meanwhile,

intergenic methylation is usually associated with insulator or repeat regions, such as

centromeres, crucial for chromosomal stability [264-266]. The methylation of these regions is

also likely to suppress the expression of transposable elements and thus has an additional role in

genome stability.

     

    34

Beyond CGIs, methylation has also been shown to occur at regions of lower CpG density that

lie in close proximity, upstream and/or downstream, but not within CGIs (within 2000 base pairs

‘CpG shores’, or 4000 base pairs ‘CpG shelves’) [267, 268]. Despite having lower CpG content,

CGI shore methylation is more tightly associated with the tissue of origin and is more frequently

altered between normal and tumor cells.

DNA methylation patterns in mammalian cells are established and heritably maintained by

the cooperative activity of the de novo DNA methyltransferases, DNMT3A and DNMT3B, and

the maintenance methyltransferase, DNMT1 [269, 270]. Conversely, DNA demethylation

mechanisms are not fully understood and several models have been proposed [271-273]. The

passive demethylation model refers to loss of 5mC during replication. Alternatively, the active

demethylation model suggests a multi-step process that involves removal of methylation by ten-

eleven translocation family enzymes, which form intermediate 5-mC oxidation derivatives such

as 5-hydromethylcytosine (5-hmC). 5hmC is then further oxidized to 5-formylcytosine, then 5-

carboxylcytosine, and subsequently cytosine, independently of DNA replication [274].

1.4.1.1 DNA Methylation Aberrations in Prostate Cancer

Aberrant DNA methylation was the first epigenetic alteration identified in cancer initiation

and progression. In PCa, global DNA hypomethylation has been suggested to precede neoplasia

as it has been traced to PIN lesions [275, 276]. Alternatively, most studies report a progressive

global loss of methylation solely during PCa progression [277]. In particular, global

hypomethylation of the retrotransposable element LINE-1 has been most prominently associated

with PCa progression and metastases [278, 279]. Reductions in DNA methylation of imprinted

genes has also been demonstrated in PCa, most commonly in the IGF2 gene, resulting in

increased expression of this mitogenic peptide [280]. Locally, hypomethylation may also lead to

reactivation of genes repressed in normal prostate such as CAGE, CYP1B1, HPSE and PLAU, to

     

    35

name a few [281-284]. Besides activation of proto-oncogenes, global DNA hypomethylation has

been linked to chromosomal and genetic aberrations such as DNA recombinations and copy

number alterations during PCa progression [285, 286].

Along with global DNA hypomethylation, PCa cells are characterized by site-specific CGI

hypermethylation and increased expression of DNMTs [287-289]. Widespread CGI

hypermethylation has been shown to be an early event in prostate tumorigenesis and affect

hundreds of sites in the genome, both near genes and in regions without known genes.

Therefore, a CGI methylator phenotype (CIMP), analogous to the CIMP established in colon

cancer, has been proposed for PCa [290]. However, no CIMP-like specific gene panel has been

validated in PCa.

A pubmed search using the query “prostate and hypermethylation” returned 684 articles as of

September 20, 2014, most of which reported more than one locus that undergoes aberrant

hypermethylation in PCa. The most frequently reported hypermethylated genes are summarized

in Table 1.6. Notably, the prevalence of these hypermethylated sites varies between genes and

between studies, in part due to intra- and inter- tumor heterogeneity and multifocality.

Hypermethylated genes in PCa are implicated in all hallmarks of cancer. Accordingly, CGI

hypermethylation of various genes has been associated with early PCa development as well as

clinicopathological features of aggressiveness and poor prognosis. Hence, various investigative

teams have focused on characterizing the utility of DNA methylation signatures associated with

pathogenesis as biomarkers for PCa diagnosis, prognosis, disease monitoring and/or prediction

of response to therapy. Importantly, it has been found that assessment of a panel of such

biomarkers dramatically improves the sensitivity and specificity compared to any single marker

[290-292]. However, numerous hypermethylation events in PCa have been only investigated for

their potential as biomarkers. Consequently, functional analyses that will distinguish the

     

    36

hypermethylation events that have a causal functional role in the disease from epiphenomena are

needed.

Hypermethylation and subsequent silencing of GSTP1 is one of the most intensively studied

epigenetic marks in PCa, observed in 70-100% of all PCa tissues [293, 294]. Importantly,

GSTP1 hypermethylation can also be measured in body fluids, mainly serum and urine, of PCa

patients [295, 296]. Although this hypermethylation mark has been most commonly suggested as

a diagnostic marker, some studies have also reported it has additional utility as a prognostic

marker [297]. The GSTP1 enzyme plays an important role in detoxification of electrophilic

compounds such as carcinogens, thus its inactivation may render cells vulnerable to oxidative

DNA damage and promote carcinogenesis [298]. Further, GSTP1 hypermethylation has been

detected in 50–70% of PIN lesions and tumor adjacent non-malignant prostate tissue, indicating

it is part of the epigenetic field effect [299, 300]. The field effect (also known as field defect,

field cancerization effect and field carcinogenesis) refers to molecular alterations, for example

aberrant DNA methylation, in histologically benign epithelium contiguous to cancerous tissue,

which predisposes it to the development of neoplasms.

MDxHealth, Inc, has recently taken advantage of the concept of field effect in PCa to develop

the FDA approved ConfirmMDx test, which evaluates the methylation of GSTP1 as well as APC

and RASSF1 in histopathologically negative biopsy samples to identify occult PCa [301, 302].

To the best of my knowledge, this is the only DNA methylation based biomarker panel for PCa

that has transitioned to the clinic, despite the ever-growing number of reported hypermethylated

genes in the disease. This is likely related to various factors including lack of formal validation

of promising candidate biomarkers, shortcomings in DNA methylation detection technologies

and/or lack of suitable industrialization. Yet, DNA methylation-based tests for PCa are

promising in investigational research laboratory studies and will likely transition to clinical use

     

    37

in the future. This notion is further supported by the success demonstrated by commercially-

available DNA methylation-based tests for other cancers in the clinic. Examples include tests for

the hypermethylation of VIM, SEPT9, SHOX2, MGMT, and MLH1 [303].

Recent technologic advances have provided the ability to detect DNA methylation in intact

circulating tumor cells (CTCs) and/or cancer cell-derived DNA (cfDNA) from blood samples of

patients with solid tumors. This has opened the door for novel research fields in PCa diagnosis,

prognosis, disease monitoring, and prediction of response to therapy. cfDNA is present at

increased concentrations in PCa patients and its methylation patterns have been shown to

resemble those of the primary tumor DNA [304, 305]. Meanwhile, CTCs are becoming a hot

topic in the PCa field since their detection has been found to associate with metastatic spread of

PCa and their enumeration has been FDA approved for follow-up of patients with metastasis

[306]. The methylation pattern of CTCs in PCa patients with metastatic disease is not yet well

characterized.

     

    38

Table 1.6. Frequently hypermethylated genes in prostate cancer

Gene Name Gene Function Frequency APC WNT signaling 27% to 100% AR Androgen receptor signaling 0% to 28% ARF Tumor suppressor in the p53 signaling pathway 0% to 35% CD44 Cell–cell interactions, cell adhesion and migration 19% to 78% CDH1 Cell–cell adhesion 0% to 80% CDKN2A Cell cycle control 0% to 77% EDNRB G protein-coupled receptor, endothelin signaling 14% to 100% GSTP1 Detoxification 33% to 100% MDR1 Calcium signaling 55% to 100% MGMT DNA repair 0% to 75% PITX2 Transcription factor, involved in developmental

Pathways 30% to 100%

PTGS2 Prostaglandin biosynthesis 35% to 100% RARB Retinoic acid receptor 30% to 95% RASSF1A Cell cycle, DNA repair 49% to 100% TIMP3 Metallopeptidase inhibitor 0% to 97% Modified from Ahmed H. Biomarkers in cancer 2010, 2:17-33 [307].

     

    39

1.4.2 Histone Modifications

Histone proteins, which comprise the nucleosome core, cooperate with DNA methylation to

regulate chromatin structure and gene transcription in mammals. Histones, a unit comprising of

DNA wound around a histone octamer composed of four core subunits (H2A, H2B, H3, H4),

contain N-terminal tails that can undergo a variety of post-translational modifications including

methylation, acetylation, ubiquitylation, sumoylation and phosphorylation [308]. The location

and type of histone modifications (known as the histone code) determines whether they will

activate or repress transcription and promote chromatin remodeling. Examples of well-

characterized activating histone marks include acetylation of lysine (K) residues, trimethylation

of histone 3 at H3K4me3 or monomethylation of H4 at H4K20me1 as well as H3K27me1.

Alternatively, the most commonly reported repressive marks are trimethylation of H3K9me3

and H3K27me3, to name a few.

PCa is characterized by deregulation of the histone code and the enzymes responsible,

primarily histone acetyltransferases, deacetylases, methyltransferases and demethylases. This

results in altered positioning of nucleosomes and leads to aberrant gene expression. Perhaps one

of the most studied histone modification in PCa is the repressive H3K27me3 mark catalyzed by

the enhancer of zeste homolog 2 (EZH2) component of the multi-protein polycomb repressive

complex 2 [309-312]. In PCa, EZH2 overexpression is suggested to result from up-regulation of

ERG and deletions of microRNA-101 [313-315]. This results in H3K27me3 elevation and

transcriptional repression of numerous targets in association with poor PCa prognosis and

metastatic CRPC. Examples include the tumor suppressor genes NKX3.1, ADRB2 and DAB2IP

[315-317]. The oncogenic function of EZH2 has also been suggested to extend beyond the

H3K27me3 mark. For example, EZH2 has been shown to recruit DNMTs and act as AR co-

     

    40

activator to promote PCa progression [313, 318]. Therefore, EZH2 inhibitors have been

suggested as a therapeutic target for aggressive PCa.

In addition to histone modifications, histone variants can replace canonical histones in the

nucleosome to produce structural and gene expression alterations. Most notably, during PCa

progression increased expression and acetylation of the histone variant H2A.Z is associated with

upregulation of AR-regulated oncogenes [319].

1.4.3 Non-Coding RNA

Non-coding RNA (ncRNA) provide an additional level of epigenetic control of gene

expression [320]. There are a growing number of ncRNA categories described in the literature as

summarized in Table 1.7. These include transcription initiation RNA (tiRNA), microRNA

(miRNA), piwi-interacting RNA (piRNA), small interfering RNA (siRNA), enhancer RNA

(eRNA) and long non-coding RNA (lncRNA). Altered expression of all these ncRNAs has been

observed in cancer thus they may provide further insight into disease etiology and offer novel

therapeutic targets.

1.4.3.1 MicroRNAs

The most widely studied class of ncRNAs are miRNAs, which are small, single stranded

ncRNAs that post-transcriptionally repress the expression of over 60% of human mRNAs in a

sequence specific manner [321-323]. According to the most recent release of miRBase, the

primary database of all miRNA sequences and annotations, there are 2588 mature miRNAs

identified in humans to date, although this number is steadily increasing [324].

Expression patterns of miRNAs have been shown to have tissue specificity and involvement

in the regulation of an extensive array of cellular processes including development,

differentiation, proliferation and apoptosis [321, 322, 325]. The majority of miRNAs can

regulate the expression levels of numerous mRNAs simultaneously and cooperatively.

     

    41

Consequently, aberrant miRNA expression may deregulate key cellular processes and promote

carcinogenesis. Therefore, miRNA transcription and biogenesis is tightly regulated in normal

cells.

MiRNA genes are transcribed by RNA polymerase II or III, which form primary miRNA

transcripts (pri-miRNA)[326]. Alternatively, some miRNAs are located in intronic sequences of

host genes with which they are co-transcribed and excised by splicing events [327, 328]. The

multistage biogenesis of mature miRNA then follows as illustrated in Figure 1.6. Briefly, pri-

miRNAs, which are 33 base-pair double-stranded hairpin structures, are cleaved by the Drosha,

in complex with the dsRNA-binding protein DiGeorge critical region 8 (DGCR8) to release a 70

nt stem-loop precursor miRNA (pre-miRNA). The pre-miRNA is exported from the nucleus by

Exportin-5 in complex with Ran-GTP. In the cytoplasm, a second endoribonucleolytic reaction,

catalyzed by Dicer, generates a 19-24-nucleotide double stranded miRNA. Subsequently, one

strand of this miRNA duplex that is complementary to the target mRNA is incorporated into the

RNA-induced silencing complex (RISC), while the remaining strand is degraded. In mammals,

mature miRNAs typically bind mRNAs in their 3’ untranslated region (UTR) through base

pairing of miRNA nucleotides 2 to 8, also known as the seed sequence [329]. Subsequently,

miRNA-mediated gene silencing is achieved through multiple different pathways of translational

inhibition and/or mRNA degradation. Although miRNAs have been mostly shown to regulate

functional mRNAs, several studies have added a role for miRNAs in the regulation of

pseudogenes. For example, a recently discovered targeting of PTEN pseudogene by miR-20a,

miR-19b, miR-21, miR-26a, and miR-214 has been shown to rescue the expression levels of

PTEN by competing for the same set of miRNAs [330]. This suggests a novel active tumor

suppressive and/or oncogenic role for pseudogenes in PCa progression. Another interesting

concept in miRNA research are miRNA sponge transcripts, which are endogenous RNA

     

    42

transcripts with multiple miRNA target sites that compete for miRNA binding thus playing a

crucial role in gene regulation [331-333]. However, the biology and function of miRNA sponges

in PCa still remains to be elucidated.

Interestingly, miRNAs have been shown to regulate key enzymes involved in DNA

methylation and histone modifications such as DNMT3A, DNMT3B, and EZH2 [315, 334-336].

Meanwhile, the expression of miRNA genes is regulated by these same epigenetic mechanisms.

This highlights the integrated nature of the different components of the epigenetic landscape.

     

    43

Table 1.7. Types of non-coding RNAs Non-coding RNA

Size Function

tiRNAs 17-18 bp Regulation of transcription miRNAs 19-24 bp Post-transcriptional regulation of gene expression piRNAs 26-31 bp Transposon repression, DNA methylation siRNA 20-25 bp RNA interference pathway,post-transcriptional gene silencing

snoRNAs 60-300 bp Transcribed from intronic regions, guide chemical modifications of other RNA molecules

eRNA 50-2000 pb

Transcribed from enhancer regions, there is still no consensus on the functional significance, involved in recruitment of regulatory proteins

lncRNA >200 bp Involved in regulation of gene transcription, post-transcriptional regulation and epigenetic regulation

Modified from Esteller M. Nature reviews Genetics 2011, 12(12):861-874 [320].

     

    44

Figure 1.6. Schematic overview of microRNA biosynthesis. Modified from Winter J. et al. Nature cell biology 2009, 11(3):228-234.

     

    45

1.4.3.2 Deregulation of microRNAs in Prostate Cancer

Widespread dysregulation of miRNA expression is observed in PIN, primary PCa, as well as

different stages of disease progression and metastasis [337-346]. Further, several miRNAs have

also been found to influence tumor microenvironment, and response to therapy. Accordingly,

signatures of tissue and circulating miRNAs have been proposed as PCa biomarkers and

therapeutic targets. However, a conclusive PCa specific miRNA expression profile has not yet

been established. The list of individually analyzed deregulated miRNAs that act as tumor-

suppressors or oncogenes (also known as oncomirs) in PCa is long. Therefore, only the most

commonly reported dysregulated miRNAs in PCa are summarized in Table 1.8. Perhaps the

most prominent example of a miRNA described as highly deregulated in PCa tissues, serum and

plasma is miR-375 [347-350]. This miRNA has been shown to be upregulated in PCa and in

association with stage as well as progression to CRPC and metastatic CRPC. This miRNA has a

few known targets in PCa through which it is hypothesized to promote tumorigenesis. The first

is Sec23, downregulation of which impairs cellular immunogenicity of PCa [351]. Secondly,

miR-375 downregulates PHLPP2, which has been shown to strongly promote PCa cell growth

[347]. However, the expression of numerous additional miRNAs in PCa has shown conflicting

results in different studies. Technological limitations, lack of standardized methods of sample

collection and varied study designs have been suggested as potential explanations.

The majority of miRNA targets in PCa are currently identified using bioinformatic tools such

as TargetScan, miRanda and miRBase [352, 353]. However, these prediction algorithms are

loosely based on 3’UTR base complementarity, thermodynamic stability, target-site accessibility

and evolutionary conservation of miRNA binding sites, resulting in a large number of false

positives. Therefore, further experimental validation of all miRNA targets in PCa is crucial for

the elucidation of their true expression status and functional involvement in the disease.

     

    46

Table 1.8. Most commonly deregulated miRNAs in prostate cancer miRNA Expression Published targets let-7b ↓ HMGA1 miR-100 ↑ PLK1, FGFR3, IGF1R and MTOR miR-106b ↑ REST, caspase-7 and p21-activated cell cycle

arrest miR-125b ↓ p14(ARF), NCOR2 and p53 pathway miR-145 ↓ GOLM1, hexokinase-2, ERG and FSCN1 miR-16 ↓ CDK1, CDK2, BCL2, CCND1 and WNT3A mir-182 ↑ NDRG1, SNAI2 , GNA13 and FOXO1 mir-191 ↑ CDK6, TIMP3, MDM4-C and EGR1 miR-205 ↓ c-SRC, MAPK and AR pathway and BCL2 miR-20a ↑ CX43, E2F2 and E2F3 mir-21 ↑ PTEN, PDCD4, TPM1 and MARCKS miR-221 ↓ p27(Kip1), ARHI, PTEN, CDKN1B and MMP1 miR-222 ↓ p27(Kip1), ARHI, PTEN, CDKN1B and MMP1 miR-23b ↓ PTEN, Src and AKT miR-375 ↑ Sec23A miR-96 ↑ FOXO1 and hZIP1

Modified from Cannistraci A. et al. BioMed research international 2014, 2014:Article ID 146170 [337].

 

     

    47

1.5 Prostate Cancer Biomarkers

Emerging diagnostic and prognostic PCa biomarkers have been discussed throughout the

introduction chapter of this thesis, including prostate genetic score-33, improvements to the PSA

test, serum hK2 test, PHI, OPKO 4Kscore™, PCA3, IDC, histologic architectures such as

cribriform, ConfirmMDx and deregulated miRNAs such as miR-375. Several additional panels

of gene markers have been recently developed to improve the risk prediction of PCa

progression. Most notable examples include Oncotype DX (Genomics Health, Inc), Decipher

(DenomeDX) and Prolaris (Myriad Genetics, Inc) [156-158]. Moreover, a urinary marker that

detects the TMPRSS2-ERG fusion is currently under development, and may have utility in

distinguishing aggressive from low-risk PCa [354]. Numerous progression biomarkers in needle

biopsies have also been reported to distinguish between aggressive and indolent PCa. These

include immunohistochemistry and fluorescent in-situ hybridization assays for PTEN, MYC,

p53, SMAD4 and cyclin D1 to name a few [355-360]. However, the area under the curve (AUC)

achieved by these new markers outlines that there is still need for improvement.

1.6 Kallikrein and Kallikrein-Related Peptidases

Human tissue Kallikrein and Kallikrein-related peptidases (hKs) are a family of 15

homologous, secreted trypsin- or chymotrypsin- like serine proteases with a wide spectrum of

functions [122, 361, 362]. They are encoded by the KLK genes clustered contiguously on the

long arm of chromosome 19 at 19q13.3-13.4 (Figure 1.7). Thus, KLKs represent the largest

uninterrupted cluster of proteases in the human genome. KLK genes contain 5 coding exons and

most also contain one to three additional noncoding exons in the 5’UTR [362, 363]. The 15 KLK

genes give rise to >80 splice variants, yet their function and clinical significance is not well

understood [364, 365]. The hK proteins are translated as pre-proenzymes with conserved

catalytic triad of histidine, aspartate, and serine [366]. The pre-peptide targets hKs for secretion

     

    48

while the pro-portion keeps them as inactive zymogens until autocatalytic activation or

transactivation by other mature hKs or endoproteases. hKs are expressed by secretory epithelial

cells in a variety of tissues and accumulate in pericellular spaces and bodily fluids such as

serum, saliva and seminal plasma, to name a few [367-371]. hKs have been implicated in a wide

spectrum of normal physiological proteolytic cascades such as kinin formation, blood pressure

control, skin desquamation, semen liquefaction, innate immunity, tissue remodelling and

prohormone processing [372, 373]. Several specific well-studied examples include hK1 cleavage

of kininogen to release bradykinin, hK2 cleavage of pro-PSA, and hK4 activation of urokinase-

type plasminogen activator receptor [374-376]. However, the specific substrates and functions of

many hKs in various tissues still remain to be elucidated.

The mechanisms that regulate the tissue-specific expression profiles of hKs are also not well

understood, although various mechanisms have been proposed. The promoters of KLKs 1-4 and

10 contain TATA box variants and so are thought to be regulated by TATA binding proteins and

associated transcription factors [377]. Further, hormones have been implicated in the regulation

of KLKs since hormone response elements including those for androgens, glucocorticoids,

estrogen, vitamin D, and retinoic acid have been identified in most KLK promoters [373]. In

addition, shared regulatory elements such as enhancers have been hypothesized to regulate the

expression of KLKs due to their genomic co-localization and tissue specific co-expression. Gene

copy alterations, SNPs and epigenetic mechanisms, mainly DNA methylation, have also been

shown to have potential roles in regulating KLKs [378-382]. Post-transcriptionally, miRNAs

have been shown to target the regulation of hK protein synthesis [383-387]. Post-translationally,

activated hK proteins can be controlled by endogenous inhibitors and proteolytic cleavage [388-

391]. There are many unanswered questions regarding the individual regulatory mechanisms of

KLKs and their interplay that need to be experimentally investigated.

     

    49

Figure 1.7. Schematic overview of the Kallikreins and Kallikrein-related peptidases (KLK) genomic locus. CpG Islands (CGIs) and histone marks accumulated during prostate carcinogenesis are also illustrated in the diagram. Histone illustration was adopted from Ho LT. Genome-wide Distribution and Regulation of DNA Methylation and Hydroxymethylation in Prostate Cancer. Unpublished master's thesis 2014, University of Toronto, Toronto, Canada.

     

    50

1.6.1 Clinical Relevance of Kallikrein Related Peptidase

The clinical utility of KLKs is best exemplified by PSA screening. Increasing evidence

demonstrates great potential for other KLKs as biomarkers in urological and gynecological

malignancies, in addition to cancers of the lung, gastrointestinal tract, breast, head and neck,

kidney, brain, and skin [392-394]. Aberrant expression of several individual members of the

KLK family has been most commonly noted in hormone-dependent cancers and has been shown

to associate with cancer progression, clinical outcome and efficacy of anticancer drugs.

hKs have been shown to contribute to cancer development by stimulating cell proliferation,

enhancing vascular permeability and malignant cell dissemination via the mitogenic activities of

kinins [388]. This protease family also has been implicated in tumor angiogenesis, invasiveness

and metastasis due to dysregulation of ECM components and activation of signal transduction

cascades. However, the explicit functions and substrates of hKs in cancer development and

progression remain unclear and need to be further elucidated to better interpret their role as

cancer biomarkers.

In normal prostate tissue, all KLKs are expressed at the mRNA and, except for KLK8, at the

protein levels [394]. In PCa tissue, elevated levels of KLK2, 4 and 13-15 mRNA and/or protein

have been reported, while hKs 3, 5-7, 10 and 11 expression is decreased. Of these, increase in

KLKs 2, 14 and 15 is associated with poor prognosis, while KLK4 mRNA upregulation is a

marker of a favorable prognosis [151, 395-397]. Moreover, lower expression of the prostate-type

KLK11 mRNA is associated with tumor aggressiveness and poor prognosis [398]. Importantly,

since hKs are secreted, they hold a great potential for direct expression analysis in bodily fluids

that will reflect the state of the PCa cells at various stages of disease progression. Further

validation of KLK gene and protein expression data in independent patient tissues and bodily

     

    51

fluids using globally standardized operating procedures is required to better elucidate their

clinical utility.

1.6.2 Epigenetics of the Kallikrein Related Peptidases Gene Locus

Given the varied expression patterns of KLKs in different tissues and their widespread

deregulation in multiple cancers, understanding the regulatory mechanisms of this family of

proteases may have important implications for our understanding of normal cell functioning and

tumorigenesis.

Biocomputational and in silico analyses have identified that the KLK gene locus contains 16

CGIs in eight KLK genes (Figure 1.7) [399]. These include CGIs upstream or in the exon-intron

sequences of KLK15, KLK2-4, KLK5, KLK10-11, KLK13 and a CGI downstream of KLK1.

KLK5 and KLK10 have the most CGIs (each associated with four CGIs), followed by KLK15

(two CGIs) and the rest of the abovementioned KLK genes with one CGI each. Among these

CGIs, the largest one is found in the third exon of KLK10 while the smallest is located in intron

5 of KLK11. The presence of CGIs in certain KLK genes suggests that DNA methylation plays a

role in transcriptional regulation of these genes. Accordingly, a number of studies have

investigated the relationship between KLK gene methylation and expression using 5-aza-2’-

deoxycytidine, a cytosine analog that inhibits DNMTs. These studies have demonstrated that

treating cell lines derived from hepatocellular carcinoma, prostate, breast, ovarian, cervix,

gastric and lung cancer, with a demethylating drug can induce the expression of KLKs 1, 5, 6

and 10-13, thus establishing that DNA methylation plays a role in regulating their expression

[400-409]. Interestingly, KLK6 has been included in these studies although it does not contain a

CGI. However, it is CpG rich and may represent a CGI shore region.

The hypermethylation of CGI in the third exon of KLK10 has been most extensively

investigated thus far in cell lines and patient specimens from various cancers including acute

     

    52

lymphoblastic leukemia, non-small cell lung cancer, hepatocellular carcinoma, breast, ovarian,

prostate, lung, and gastric cancers [404-409]. Most notably, KLK10 exon 3 hypermethylation has

been shown to play an important role in the downregulation of hK10 expression and serve as an

independent prognostic biomarker in breast cancer, adult acute lymphoblastic leukemia, and

lung cancer. In PCa, genome wide methylation analysis of cell lines has identified a long-range

epigenetic activation through CpG demethylation in KLK15-KLK4, while KLK5-KLK12 were

found transcriptionally repressed by CpG hypermethylation. Additionally, KLK10 5’UTR

hypermethylation has been noted in PCa tissues using several microarray platforms [410, 411].

The regulation of KLKs by hormones in a subset of tissues suggests that open chromatin

conformation is necessary for the transcriptional activation of KLKs. However, very little is

known about the role of histone modifications in the regulation of KLK gene expression. To

date, the activation of KLKs 2 and 3 by androgens in PCa and the associated cascade of distinct

covalent histone modifications have been best investigated. These genes have been shown to

accumulate diacetylation in H3K9 and H3K14, phosphorylation of H3S10 and di- and tri-

methylation of H3K4 [412, 413]. Further, it was found that androgen-independent cell lines had

a significantly increased H3K9 and H3K14 methylation at the KLK3 locus compared to

androgen-dependent cells contributing to increases receptiveness to gene transcription. A

subsequent study of the KLK15-KLK4 gene locus in PCa cell lines have also found gain in the

active H3K9ac mark and loss of the H3K27me3 repressive mark in this region (Figure 1.7)

[414]. Other PCa associated histone modifications in the region were more discrete, with losses

of H3K9me2 specifically at KLK2 promoter, and gains of H3K4me3 at the KLK4 promoter.

Conversely, gains in the repressive H3K27me3 and H3K9me2 marks were observed in KLK5-

KLK12 region (Figure 1.7). Other studies investigated the role of histone deacetylation in the

silencing of KLK6 in breast cancer. They found that constitutive and inducible expression of

     

    53

KLK6 positively correlated with histone H4 acetylation located in its upstream sequences [400,

401].

Recent studies have shown that hundreds of miRNAs are predicted to target KLK mRNAs

using in silico prediction algorithms [383]. Majority of miRNAs were predicted to target more

than one KLK. Further, KLKs 2, 4, 5 and 10 were predicted to have multiple miRNA-targeting

sites and were predicted to be targeted by numerous miRNAs. Of these, KLK10 was predicted to

be targeted by the greatest number of miRNAs. However, miRNA-KLK target validation studies

are limited. Only let-7f, miR-224, miR-516a, miR-330, miR-331-3p and miR-143 were

experimentally verified to target KLKs 1, 4, 6 and 10 mRNA in breast, kidney, ovarian and

prostate cancer cell lines [384-387].

1.7 Hypothesis and Objectives

I hypothesize that epigenetic mechanisms such as DNA methylation and miRNAs co-operate

to play a role in regulating KLKs and associated pathways and can potentially serve as

diagnostic and/or prognostic PCa biomarkers. Further, using these biomarkers in conjunction

with clinical and histologic features such as intraductal carcinoma and cribriform will ultimately

help to improve the clinical management of the disease.

To address this hypothesis, the following objectives were formulated:

1. Analyze DNA methylation of KLKs in prostate cancer cell lines, tissue and serum samples.

2. Investigate associations between DNA methylation of KLKs and associated pathways, clinical

and morphological features of PCa.

3. Develop a multiplex protocol for simultaneous analysis of multiple promising DNA

methylation based PCa biomarkers.

4. Assess expression of miRNAs predicted to target KLKs as a signature of aggressive PCa.

     

    54

Chapter 2 Analysis of Kallikrein-Related Peptidases DNA Methylation in Prostate Cancer

Cell Lines, Tissue and Serum Samples

Ekaterina Olkhov-Mitsel,1,2 Theodorus Van der Kwast,2,3 Ken Kron,1,2 Hilmi Ozcelik,1 Laurent Briollais,1 Christine Massey,4 Franz Recker,5 Maciej Kwiatkowski,5 Neil E. Fleshner,6

Eleftherios P. Diamandis,2,7,8 Alexandre R. Zlotta9,10 and Bharati Bapat1,2,3

1Samuel Lunenfeld Research Institute; Mount Sinai Hospital; Toronto, ON Canada; 2Department of Laboratory Medicine and Pathobiology; University of Toronto; Toronto, ON Canada;

3Department of Pathology; University Health Network; University of Toronto; Toronto, ON Canada; 4Department of Biostatistics; Princess Margaret Hospital; Toronto, ON Canada;

5Department of Urology; Kantonsspital Aarau; Aarau, Switzerland; 6Department of Surgical Oncology; Division of Urology; University Health Network; University of Toronto; Toronto, ON Canada; 7Department of Clinical Biochemistry; University Health Network; Toronto, ON Canada; 8Department of Pathology and Laboratory Medicine; Mount Sinai Hospital; Toronto,

ON Canada; 9Division of Urology; Princess Margaret Hospital; Toronto, ON Canada; 10Division of Urology; Mount Sinai Hospital; Toronto, ON Canada.

The majority of the data presented in this chapter has been previously published in a manuscript entitled “Quantitative DNA methylation analysis of genes coding for kallikrein-related peptidases 6 and 10 as biomarkers for prostate cancer” Olkhov-Mitsel, E., Van der Kwast, T., Kron, K., Ozcelik, H., Briollais, L., Massey, C., Recker, F., Kwiatkowski, M., Fleshner, N.E., Diamandis, E.P., Zlotta A.R. and Bapat B. Epigenetics. 2012. 7(9); 1037-1045.

The work has been primarily contributed by Olkhov-Mitsel E. Pathologist van der Kwast T. assisted in specimen retrieval and histopathological confirmation. Former Ph.D student Kron K. extracted DNA from a subset of patients in cohort II and contributed to cohort II TMA construction as well as CGI methylation array data, used to select candidate KLKs for investigation in this chapter. Ozcelik H., Recker F., Kwiatkowski M., Fleshner N.E., Diamandis E.P., and Zlotta A.R. contributed to the development of the study and manuscript preparation. Briollais L. and Massey C. assisted with statistical analysis. Bapat B. supervised the project (including project development, experimental design, analysis and interpretation of data) and critically reviewed the manuscript and thesis.

     

    55

Chapter 2 Analysis of Kallikrein-Related Peptidases DNA Methylation in Prostate Cancer Cell Lines,

Tissue and Serum Samples  

2.1 Summary  

KLKs have emerged as an important family of cancer biomarkers, as exemplified by PSA

screening. Yet, few studies have examined the epigenetic regulation of KLKs and its

implications to PCa. In particular, DNA methylation, which plays an important role in prostate

carcinogenesis is being recognized as a promising diagnostic and prognostic biomarker for the

disease. Herein KLKs 6, 10 and 15 DNA methylation levels were quantified in two independent

cohorts of PCa patients operated by RP between 2007-2011 (cohort I, n=150) and 1998-2001

(cohort II, n=124). In Cohort I, DNA methylation levels of all three KLKs were significantly

higher in PCa tissue vs. normal (P-value<0.001). KLK6 methylation was significantly associated

with pathological stage and ERG oncogene expression status (P-value=0.029,0.009,

respectively) only in cohort I, while KLK10 methylation was significantly associated with

pathological stage (P-value=0.007, 0.031, respectively) and ERG (P-value<0.001) in both

cohorts. In cohort II, low KLK10 methylation was associated with biochemical recurrence in

univariate and multivariate analyses (P-value=0.046, 0.028). KLK15 methylation did not

correlate with clinicopathologic variables. Further, KLK10 and 15 methylation, analyzed in a

subset of serum samples from cohort I, did not correlate with the methylation levels in the

corresponding tissues. Subsequently, to assess the biological effect of DNA methylation on

KLK6 and KLK10 expression, PC-3 and 22RV1 PCa cells were treated with a demethylating

drug, 5-aza-2′deoxycytidine, resulting in increased expression of both KLKs, establishing that

DNA methylation plays a role in regulating gene expression. The results suggest that KLK6 and

KLK10 methylation distinguishes organ confined from locally invasive PCa and may have

prognostic value.

 

     

    56

2.2 Introduction

The development of novel PCa biomarkers remains an important and exciting challenge.

This is due to their great potential to be used in conjunction with existing makers for the

improvement of disease diagnosis and prognosis, while establishing a distinction between

aggressive and indolent forms of PCa. In particular, using the universally and readily available

formalin-fixed paraffin-embedded (FFPE) specimens.

Disease-specific aberrant DNA methylation events are a well-recognized hallmark of PCa

and are being recognized for their potential as promising diagnostic and prognostic biomarkers

for the disease. DNA methylation biomarkers offer several significant advantages over

expression-based markers. For instance, they are readily amplifiable and easily detectable using

polymerase chain reaction (PCR)-based approaches, even if alterations are present only in a

limited number of cells [415]. DNA methylation is a highly stable mark that can be readily

detected in a great variety of samples collected in a minimally invasive manner, such as saliva,

plasma, serum, urine, semen, and stool [416, 417]. Furthermore, disease-specific DNA

hypermethylation is a positively detectable signal. Recently, dysregulated DNA methylation

signatures have also been associated with TMPRSS2-ERG gene fusions, which have been

suggested as a predominant molecular subtype of PCa [213, 418, 419]. However, conflicting

results about the prognostic value of this gene fusion have been reported [218].

Accumulating evidence suggests that DNA methylation may regulate KLKs, an important

family of cancer biomarkers, in PCa [420]. Given the dysregulation of KLKs in PCa,

understanding the regulatory mechanisms of this family of proteases may provide important

information on PCa pathogenesis. Evidence to support this hypothesis includes the identification

of multiple CGIs in the 5'UTR and coding regions of numerous KLK genes using

biocomputational analysis [399]. Interestingly, no CpG islands are located in the major regions

     

    57

regulating KLK3 gene expression including the proximal promoter and the distal enhancers.

Secondly, majority of KLK genes lack classical promoter sequences [383]. Lastly, DNA

methylation has been previously suggested to regulate KLK expression in cell lines from

numerous different cancers such as breast, ovarian, lung and leukemia [399, 408, 409]. More

recently, KLK10 gene promoter was also found to be differentially methylated in PCa tissue vs.

benign adjacent prostate tissue in a few microarray studies [410, 411].

However, very few studies to date have systematically investigated KLK DNA methylation in

primary prostate tissue and their potential contribution to PCa diagnosis or prognosis. By

genome-wide differential methylation hybridization CGI microarray profiling of GS8 vs. GS6

prostate tumors, the tumor specific hypermethylation of several genes including HOXD3 and

TGFβ2 has been previously established in the lab [291, 421, 422]. Using the same profiling

platform differential DNA methylation of KLKs 10 and 15 was identified in prostate tumors.

KLK6 was also chosen as a candidate gene for analysis since its promoter is CpG rich and its

hypermethylation was found to be associated with gene downregulation in breast cancer [400,

401].

In this chapter, the association between KLK 6, 10 and 15 DNA methylation status,

clinicopathological parameters and ERG expression was investigated for the first time in two

large independent cohorts of PCa patients operated by RP. Further, potential correlation between

their methylation levels in tumor tissue and serum DNA was explored.

2.3 Materials and methods  

2.3.1 Patients and Pathology

Two cohorts of PCa and normal prostatic tissues were used for all the studies described in

chapters 2 and 5 of this thesis. The samples in cohort I consisted of 150 patients diagnosed with

PCa who underwent RP between 2007 and 2011 at the University Health Network (UHN) in

     

    58

Toronto. Inclusion criteria for patients were based on the availability of serum samples in the

GU BioBank at UHN and self-reported European ancestry. Following the accrual of the first

available 115 cases, inclusion criteria was adjusted to identify GS≥8 cases due to the low

number of cases accrued in this category at that point of the study. Cohort II is made up of 124

patients diagnosed with PCa who underwent RP between 1998 and 2001 at UHN, with available

follow-up data. In Cohort II, biochemical recurrence was defined as a rise in PSA levels above

0.05 ng/mL following RP. Patients who received neo-adjuvant or hormone therapy before RP

were excluded from the study. The clinicopathological characteristics of both cohorts are listed

in Table 2.1. Additionally, serum samples were obtained from 15 patients in cohort I and 5 men

who had a negative prostate biopsy at UHN. Patient consent was obtained for specimen accrual

following RP or biopsy into the UHN tissue bank and UHN Genitourinary BioBank. All samples

and clinicopathological follow-up information were obtained according to the protocols

approved by the Research Ethics Board at Mount Sinai Hospital and UHN, Toronto. The

complete set of hematoxylin and eosin (H&E)-stained slides from each prostatectomy was

collected and reviewed by an expert pathologist (TVDK) to confirm/assign a GS (WHO/ISUP

2005 criteria), stage (TNM), and surgical margin status.

For each patient in cohort I, a subset of slides was selected based on the presence of

carcinoma with specific GS. Further, a subset of slides was selected for each of these patients

based on the presence of normal prostatic tissue containing at least 50% glandular content. In

cohort II, tumor areas of each Gleason pattern representing the overall GS were marked on the

H&E-stained slides. All of the marked tumors were corresponding to an area of at least 80%

neoplastic cellularity.

     

    59

2.3.2 MassARRAY EpiTYPER Analyses

Quantitative analysis of CpG dinucleotide methylation was performed using MassARRAY®

EpiTYPER analysis (Sequenom). EpiTYPER is a Time of flight Matrix-assisted laser

desorption/ionization mass spectrometry (MALDI-TOF MS) based method that provides a

quantitative view of CpG dinucleotide methylation status to single or multiple dinucleotide

resolution [423]. DNA from 2 fresh frozen samples was bisulfite modified, tagged with a T7

promoter, and transcribed into RNA. RNA was cleaved with ribonuclease A and the products

were resolved by the MS instrument. Analysis was performed in triplicate by the Analytical

Genetics Technology Centre (AGTC), Princess Margaret Hospital, Toronto. The regions

analyzed by EpiTYPER for KLK10 and KLK15 were chromosome 19: 51,522,948 to 51,523,410

and chromosome 19: 51,330,238-51,330,558 (NCBI37), respectively.

2.3.3 5-aza-2-deoxycytidine Treatment and RNA Extraction

Human PCa cell lines PC-3 and 22RV1 were cultured as monolayers in F-12 and RPMI 1640

media (Life Technologies), respectively, supplemented with 10% fetal bovine serum, and grown

in humidified atmosphere with 5% CO2 at 37ºC. PC-3 and 22RV1 cells were plated in 10cm

dishes, cultured for 24 hours, and treated with 4 µM 5-aza-2-deoxycytidine for 2 days [421].

After treatment, cells were washed with PBS, and fresh medium was added. Cells were then

further incubated for another 2 days, then harvested. Subsequently, genomic DNA and total

RNA were extracted simultaneously using AllPrep DNA/RNA mini kit (Qiagen), as

recommended by the supplier.

2.3.4 Reverse Transcription and RT-qPCR

For RT-qPCR analysis of KLK gene expression, total RNA was reverse transcribed using the

Bio-Rad iScript cDNA synthesis kit (Bio-Rad) following the manufacturer’s protocol. Following

synthesis, 50 ng of cDNA was used in 25 µL volume reactions with the following final reagent

     

    60

concentrations: 1x PCR buffer, 2mM MgCl2, 0.2µM of both forward and reverse primer, 0.25

units Taq polymerase, and 0.4mM each dNTP. Primer sequences are shown in Table 2.2. RT-

qPCR was performed on Applied Biosystems 7500 qPCR instrument using the following

program: 10 minutes initial incubation at 95ºC, followed by 40 cycles of 95ºC denaturation (15

seconds) and 60ºC simultaneous annealing and extension (1 minute).

2.3.5 DNA Extraction, Sodium Bisulfate Modification and MethyLight

FFPE tissue blocks matching the selected H&E slides were sectioned at a thickness of 10µm

and slides were stored in slide boxes at room temperature. FFPE tissue blocks fixation, storage

and processing prior to sectioning is unknown. These tissue slides were then superimposed on

H&E slides and each area of cancer or normal tissue was outlined. The circled areas of tissue

slides were then scraped with a scalpel and placed into 1.5 ml tubes. DNA was extracted from

tissues using QIAamp DNA Mini Kit (Qiagen) with a modified protocol as described previously

[422]. For tumors containing multiple foci, two patterns were combined in equal amounts to

represent the overall GS. DNA was isolated from serum samples using the QIAamp DNA Mini

Kit (Qiagen), according to the manufacturer’s protocol. The concentration and quality of DNA

samples was assessed using a NanoDrop 1000 spectrophotometer (Thermo Scientific). DNA was

stored at 4°C. Next, 100-400 ng of extracted DNA was converted using the EZ DNA

Methylation Gold Kit (Zymo Research) according to the manufacturer’s protocol and eluted to a

final concentration of 20 ng/µl. Bisulfite modified DNA was stored at -20°C.

DNA methylation analysis was performed using semi-quantitative MethyLight assay, a

Taqman-based technique that assesses percent DNA methylation at a defined gene locus, as

described earlier [424]. In brief, 20 ng of bisulfite-converted genomic DNA was amplified using

locus specific PCR primers flanking an oligonucleotide probe with a 5’ fluorescent reporter dye

and a 3’ quencher dye. Primer sequences are shown in Table 2.2. In each PCR reaction, 200 uM

     

    61

dNTPs, 0.3 uM forward and reverse primers, 0.1 uM probe, 3.5 mM MgCl2, 0.01% Tween-20,

0.05% gelatin and 0.5 units of Taq polymerase were combined in a 30 ul reaction volume. A

percent methylation ratio (PMR) score was calculated for each KLK gene locus according to the

formula of Eads et al. [424] by dividing the KLK gene:ALU ratio of a sample by the KLK

gene:ALU ratio of commercially available fully methylated DNA (Millipore) and multiplying by

100. The ALU PCR products (generated from a consensus CpG-devoid region of the ALU

repetitive element) were used as controls to normalize for input DNA [425]. Samples were

analyzed in duplicates in 96-well plates on an ABI 7500 RT-qPCR thermocycler (Life

technologies). The experiments were blinded to all variables, except GS.

2.3.6 Tissue Microarray Construction

Four 0.6mm cores were taken from each of the 150 cases in cohort I to represent the primary

tumor (two cores) as well as benign glandular tissue adjacent to the tumor (two cores). This

yielded a total of 600 cores within 2 tissue microarray (TMA) blocks. Serial sections of 5µm

from each microarray were used for H&E verification of tumor and benign prostatic tissue.

Additional 4µm unstained sections from each microarray were used for ERG

immunohistochemistry.

2.3.7 ERG Immunohistochemistry

Immunostaining of the cohort I TMA for ERG was conducted at the UHN Pathology

Research Program Laboratory as follows: deparaffinized 4µm TMA sections were dehydrated,

blocked in 0.6% hydrogen peroxide in methanol for 20 minutes and processed for antigen

retrieval in EDTA (pH 9.0) for 30 minutes in a microwave, followed by 30 minutes of cooling in

EDTA buffer. Sections were then blocked in 1% horse serum followed by an overnight

incubation with the ERG-mAb mouse monoclonal antibody (Biocare Medical; clone 9Fy),

diluted 1:300, at room temperature. Immunostaining was subsequently developed by the

     

    62

Polymer-HRP Immunohistochemistry Kit (Biogenex) according to manufacturers instructions.

Next, sections were counterstained in hematoxylin for 1 minute, dehydrated, cleared, and

mounted. Cohort II TMA construction and staining was performed as previously described[292].

Immunostained slides were reviewed (by TVDK) and individual TMA cores were scored. ERG

nuclear staining was scored as positive if >10% of cells in the core were stained whereas

staining in ≤10% of cells was considered negative. Cases were considered positive for ERG

expression if any of the arrayed cores from that case displayed positive ERG

immunohistochemistry.

2.3.8 Statistical Analyses

PMR scores for each sample analyzed were obtained from averaging duplicate runs. KLK

DNA methylation was separated into those with high methylation (HM) and those with low

methylation (LM) using PMR threshold values determined by Receiver-operating characteristic

(ROC) analysis as described earlier [426, 427]. Briefly, data for all PCa and normal specimens

was split based on PMR values above the threshold value determined by ROC analysis vs. those

below. Those PMRs above the ROC threshold value were classified as HM, whereas those

below this value were considered LM. Association between KLK PMR values and

clinicopathological features or ERG status was analyzed using the Kruskal-Wallis test. Further,

pairwise comparisons were performed using the Mann–Whitney U-test regardless of the results

of the Kruskal-Wallis test due to the exploratory nature of the study. DNA methylation in PCa

and matched normal specimen pairs as well as tissue-serum specimen pairs was assessed using

the Wilcoxon signed-rank test and Pearson correlation. Pearson χ2 tests were used to analyze

proportional differences between HM cases and ERG status in each cohort, each gene and in

each clinicopathological category. The Fisher exact method was used to replace the Pearson tests

when spreadsheet cell counts were <5. Univariate biochemical recurrence-free probabilities were

     

    63

estimated and plotted using Kaplan–Meier method and compared using the log-rank test.

Multivariate Cox proportional hazards regression analysis was used to analyze individual

contributions of each variable to biochemical recurrence-free probability. The likelihood ratio

test (LRT) was used with the backwards model selection to analyze the data. The LRT tests for

significance of variables by comparing the full model to a reduced model, with the reduced

model missing the variables being tested for significance. The criteria for staying in the reduced

model were set to P-value≤0.1. For all described methods, two-sided P-values of ≤0.05 were

considered significant. No adjustments to P-values were made for multiple comparisons[428].

ROC curves and all statistics were performed using SPSS version 21 software (SPSS, IBM

Software, Chicago, IL).

     

    64

Table 2.1. Clinical characteristics of two Cohorts of prostate cancer patients Clinical characteristic

Cohort I 2007-2011

Cohort II 1998-2001

Gleason Score§ No. of patients (%) No. of patients (%) 4 0

1 (0.8)

5 2 (1.3)

7 (5.6) 6 36 (24)

48 (38.7)

7 75 (50)

53 (42.7) 8 18 (12)

11 (8.9)

9 19 (12.7)

2 (1.6) 10 0

2 (1.6)

Pathological stage pT2 88 (58.7)

75 (60.5)

pT3a 42 (28)

30 (24.2) pT3b 18 (12)

14 (11.3)

pT4 2 (1.3)

5 (4.0) Age

Average 61.2

61.2 Median 62.0

62.0

Range 38-75

41.5-75.9 Surgical Margins

Positive 29 (19.3)

30 (24.2) Negative 121 (80.7)

94 (75.8)

Weight of Prostate Average 50a

N/A

Range 15-191

N/A Biochemical Recurrence Recurrences N/A

48 (38.7)

PSA Average 9.3b

9.1c

Range 0.21-165.43*

0.1-45.8

Total 150 124 § 2005 ISUP modified Gleason score assigned by TVDK

a Data was available for 146 patients b Data was available for 142 patients c Data was available for 116 patients

*One patient had a PSA of 165.43, the remaining 141 patients PSA levels range from 0.21-56.16

     

    65

Table 2.2. Primer and probe sequences used in RT-qPCR and MethyLight assay for KLK6, KLK10, KLK15, β-actin and ALU genes Gene Sequence Real-Time RT-qPCR β-actin Forward: 5’- ACAATGAGCTGCGTGTGGCT -3’

Reverse: 5’- TCTCCTTAATGTCACGCACGA -3’ KLK6 Forward: 5’- CCTGCAGCAGGAGCGGCC -3’

Reverse: 5’- TGTGAGGACCCACAGTGGATGGATA -3’ KLK10 Forward: 5’- TACAACAAGGGCCTGACCTGCT -3’

Reverse: 5’- GTCACTCTGGCAAGGGTCCTG -3’ MethyLight ALU Forward: 5’- GGTTAGGTATAGTGGTTTATATTTGTAATTTTAGTA -3’

Reverse: 5’- ATTAACTAAACTAATCTTAAACTCCTAACCTCA -3’ Probe: 5’FAM- CCTACCTTAACCTCCC -3’BHQ1

KLK6 Forward: 5’- AGGAAGTTATTGATGTAATCGTTTTTTCG -3’ Reverse: 5’- AAAACAATCGAACTTTATCCGCC -3’ Probe: 5’FAM - ACTCCGACCTCAACCTCTCTTCCGACGAACA -3’BHQ1

KLK10 Forward: 5’- GAGGGGGAAATTTCGGGCGC-3’ Reverse: 5’- CCCTCGCGACATCTTCCCG -3’ Probe: 5’FAM - ACCCGAATAAAACGCTCTCCGCGCCCCA -3’BHQ1

KLK15 Forward: 5’- CGTTACGAAGCGCGTAGTTATCG -3’ Reverse: 5’- CACACAAACCTCCCCCGAA -3’ Probe: 5’FAM - CAACGCGTGGGTAGCACCGCGGGGCGCACC -3’BHQ1

 

 

 

 

 

 

 

 

     

    66

2.4 Results  

2.4.1 Identification of Candidate KLKs for DNA Methylation Analyses

Candidate KLKs for epigenetic analysis were identified using Agilent Human CGI

microarrays performed by other members of the lab (V. Pethe and K. Kron) [421]. For this

study, 39 PCa cases and 5 benign tissues were analyzed for genome wide methylation levels

across 27,000 CGIs. KLKs 1, 15, 4, 9, 10, 12 and 13 were represented on the microarrays by 59

probes. KLK3, one of the most highly expressed genes in the prostate, does not contain CGIs in

the promoter, enhancers and gene body. Therefore, this gene was not represented on the array.

Further, the CGI between KLK3 and KLK2 (Figure 1.7) was not represented on the array. CpG

sites located within KLK4 gene body, another KLK of interest in PCa, were represented by 6

probes on the array. These probes were hypomethylated in tumor vs. benign adjacent prostate

tissues (average fold change = 0.47 and P-value = 0.008). However, CGI within the KLK10

5’UTR was most significantly hypermethylated in PCa compared to benign tissue (average fold

change =1.62 and P-value = 0.002), while KLK15 exon 3 CGI showed the greatest degree of

hypermethylation in Gleason pattern 4 vs. Gleason pattern 3 tumors (average fold change =1.57

and P-value = 0.039). Thus, both regions were chosen for further validation. In addition, KLK6

was selected as a candidate gene for analysis as the gene promoter is CpG rich and its

hypermethylation was found to be associated with gene downregulation in breast cancer [400].

KLK6 may be similarly downregulated in PCa potentially through the same epigenetic

mechanisms. Thus, KLK6 DNA methylation in PCa was further assessed, despite the fact it was

not represented on the Agilent Human CGI microarrays.

     

    67

2.4.2 MassARRAY EpiTYPER Analyses

To validate the methylation signal from the CGI microarrays, quantitative DNA methylation

of CpG dinucleotides within KLK10 5'UTR and KLK15 exon 3 CGIs was measured by

MassARRAY EpiTYPER. For this analysis, four fresh frozen PCa tissues were chosen for

validation of (i) methylation signal enrichment of ≥3 and (ii) methylation signal less than 1.1-

fold on the microarrays. On EpiTYPER analysis the average methylation levels across 34 CpG

sites within KLK10 5'UTR in sample 1 were 82% compared to 30% in sample 2 (Figure 2.1A).

For KLK15, sample 3 had an average methylation of 87% across 31 CpG residues compared to

75% in sample 4 (Figure 2.1B). The most robust differences in methylation were observed

across the last 6 CpG dinucleotides, with sample 3 averaging 84% methylation and sample 4

having 45% methylation. These results confirmed the DNA methylation profiles of KLKs 10 and

15 from the CGI microarrays. Additionally, these results were used to identify the most distinct,

differentially methylated CpGs within the selected regions to design DNA methylation-specific

primers and probes for subsequent MethyLight analysis (Figure 2.1C&D). For KLK6, selection

of differentially methylated CpGs for MethyLight analysis was achieved using a previously

published bisulfite-sequencing map of the promoter region [400].

     

    68

Figure 2.1. EpiTYPER analysis of (a) KLK10 for samples 1 and 2, (b) genomic sequence of the region, (c) KLK15 for samples 2 and 3 and (d) genomic sequence of the region. Colored bars represent the average methylation over three replicates with standard error bars displayed. Representative cases with high methylation (¢) and low methylation (¢) as detected by Agilent array screen are shown. The primer and probe sequence used for MethyLight analysis are highlighted in red.

     

    69

2.4.3 DNA Methylation of KLK15 in Primary PCa and Matched Normal Tissue

KLK15 methylation levels were assessed in cohort I consisting of 150 PCa specimens and

matched 147 normal tissues from the same prostate. Table 2.1 displays the clinicopathological

characteristics of the cohort. MethyLight analyses identified that the median PMR values for

PCa tissues were significantly higher than in matched normal tissues; 22.7 vs. 20.4 (Wilcoxon P-

value <0.001, Table 2.3). Next, ROC curve analysis was performed to determine the ability of

KLK15 DNA methylation to accurately distinguish cancer from normal tissues (Figure 2.2A).

The AUC was 0.592 (95% CI: 0.527–0.657). The ROC curve was used to establish the PMR

threshold value (PMR 36.5%) that allowed optimal separation between benign and malignant

tissue with maximum combined sensitivity (31%) and specificity (87%). Based on this

threshold, KLK15 DNA methylation was classified into a HM group, which was equal or greater

than the PMR threshold values 36.5%, and an LM group, which accounted for the rest of the

samples. χ2 analysis showed a greater proportion of HM cases in PCa specimens compared with

normal; 31% vs. 13% (P-value<0.001). KLK15 methylation in normal tissue was positively

associated with methylation levels in the matched cancerous tissue from the same patient

(Pearson correlation P-value <0.001), and did not correlate with patient’s age or prostate weight

(Pearson correlation P-value >0.05).

KLK15 methylation levels and frequencies in PCa tissues did not increase with increase in GS

(Table 2.3). Similarly, there were no significant differences in KLK15 methylation between

organ-confined pT2 cases and locally advanced pT3 and pT4 cases.

     

    70

Table 2.3. KLK15 Median PMR, proportion of high methylation (HM) cases and P-values stratified according to Gleason Score and Pathological stage, Cohort I (2007-2011)

Cohort I (2007-2011) Median PMR n (%) HM Gleason Score

≤6 27.6 11 (30%) 7 27.9 28 (37%) ≥8 17.0 6 (16%)

Pathological Stage pT2 22.0 22 (25%) pT3a 27.9 15 (36%)

pT3b+pT4 21.1 8 (40%) pT3a+pT3b 27.9 23 (38%)

Comparison Kruskal-Wallis P-value χ2 P-value ≤6 vs. 7 vs. ≥8 0.009 0.071 pT2 vs. pT3a vs. pT3b+pT4 0.505 0.265

     

    71

2.4.4 KLK15 DNA Methylation in Serum of PCa Patients and Healthy Controls

To examine whether KLK15 hypermethylation can be detected in serum, I analyzed the sera

of 15 PCa patients in cohort I and five healthy patients, confirmed by a negative biopsy. In

healthy individuals, the median KLK15 DNA methylation PMR value was higher than in PCa

patients; 29.3% vs. 21.3% (Mann Whitney U P-value = 1). Next, I compared KLK15 DNA

methylation for the 15 PCa cases, for which both tissue and serum were available. The median

methylation levels were significantly higher in serum (21.3%) than tumor tissues (14.1%) from

the same patient (Wilcoxon P-value = 0.011). Further, there was no significant correlation

between methylation levels in tumor and serum from the same patient (Pearson correlation

coefficient of 0.409 and P-value = 0.146).

The preliminary data presented thus far on KLK15 methylation revealed it is not a significant

diagnostic or prognostic marker in PCa. As a result, it was not analyzed further.

2.4.5 KLKs 6 and 10 DNA Methylation in Primary PCa and Normal Tissue

Next, I quantified the methylation levels of KLK6 and KLK10 in cohorts I and II (Table 2.1).

In cohort I, median PMR values for both KLKs, given in Table 2.4, were significantly higher in

PCa compared to normal tissue (Wilcoxon Signed-Rank Test P-value <0.001). Additionally, I

performed ROC curve analysis to determine the ability of DNA methylation of each KLK to

accurately distinguish cancer from normal tissues (Figure 2.2). The AUC for KLK6 and KLK10

were 0.622 (95% CI: 0.558–0.686), and 0.816 (95% CI: 0.767–0.865), respectively.

Consequently, I established PMR threshold values using the ROC curves that allowed optimal

separation between benign and malignant tissue with maximum combined sensitivity and

specificity. The PMR threshold values were 11.54% for KLK6 and 4.01% for KLK10 as the

maximized combined sensitivity and specificity were 72% and 52%, 70% and 82%,

respectively. I further classified KLK DNA methylation into a HM group, which was equal or

     

    72

greater than the PMR threshold values determined by the ROC analysis, and LM group, which

accounted for the rest of the samples. χ2 analysis showed a greater proportion of HM cases in

PCa specimens compared with normal (Table 2.4); KLK6 - 72% vs. 48% (P-value<0.001),

KLK10 - 71% vs. 18% (P-value <0.001).

DNA methylation levels in each KLK gene were positively correlated with DNA methylation

in the other gene, in both cohorts (Pearson correlation coefficient = 0.339, P-value<0.001,

Pearson correlation coefficient = 0.182, P-value=0.047, respectively). Additionally, patients in

cohort I with one KLK HM were significantly more likely to have the other KLK also highly

methylated (χ2 P-value<0.001). This association was not observed in cohort II (χ2 P-value =

0.336).

KLK6 DNA methylation observed in normal tissue significantly correlated with age and with

DNA methylation in the matched cancerous tissue from the same prostate (Pearson correlation

coefficient = 0.216, P-value=0.009, Pearson correlation coefficient = 0.652, P-value<0.001,

respectively, Table 2.5). However, such association was not observed for KLK10 DNA

methylation in normal tissue. KLK10, but not KLK6 DNA methylation levels significantly

differed between normal tissue acquired from the transition zone, median PMR 0, compared to

the peripheral zone, median PMR 1.40 (P-value<0.001).

     

    73

Table 2.4. KLK6 and KLK10 median PMR values, proportion of high methylation (HM) cases and P-values for PCa specimens and normal tissues from the same prostate in cohort I (2007-2011)

Median PMR Wilcoxon signed-rank test P-value n (%) HM χ2 P-value

KLK6

Cancer 16.27 <0.001 106 (72%) <0.001 Normal 11.47 70 (48%)

KLK10

Cancer 10.08 <0.001 106 (71%) <0.001 Normal 0.79 26 (18%)

Table 2.5. Association between KLK6 and KLK10 methylation levels, age and prostatic zone in normal and cancerous tissues from the same prostate in cohort I (2007-2011)

KLK6 Normal KLK10 Normal

Pearson correlation coefficient P-value

χ2 P-value

Pearson correlation coefficient P-value

χ2 P-value

Age 0.216 0.009 0.278a 0.063 0.448 0.453a

Cancer KLK6 PMR

0.652 <0.001 <0.001 0.006 0.940 0.152 Cancer KLK10 PMR

0.137 0.099 0.481 0.136 0.102 0.189

Mann-Whitney U Test P-value

χ2 P-value

Mann-Whitney U Test P-value

χ2 P-value

PZ vs. TZ 0.168 0.739 <0.001 <0.001 a Median age was used as a cutpoint to convert the continuous age variable to a binary variable for χ2 analysis.

     

    74

Figure 2.2. Receiver Operator Curve analysis of (a) KLK6 (b) KLK10 (c) KLK15 percent of methylated reference (PMR) values in cohort I (2007-2011). Abbreviations: AUC - area under the curve.

 

 

 

 

 

 

 

     

    75

2.4.6 KLK10 DNA Methylation in Serum of PCa Patients and Healthy Controls

I proceeded to analyze KLK10 methylation in sera of 15 PCa patients from cohort I and five

healthy controls, confirmed by a negative biopsy. In healthy controls, median KLK10 PMR

values were higher than in PCa; 19.4% vs. 8.1% (Mann-Whitney P-value = 0.168). Next, I

compared KLK10 DNA methylation for the 15 PCa cases, for which both tissue and serum were

available. This revealed no correlation of methylation level between tumor and serum from the

same patient, with a Pearson correlation coefficient -0.240 (P-value=0.408). Further, KLK10

PMR values were higher in serum (8.1%) compared to tumor tissues (5.6%) from the same

patient (Wilcoxon P-value P-value = 0.683).

2.4.7 KLKs 6 and 10 DNA Methylation and Clinicopathological Features

Next, KLK6 and KLK10 DNA methylation status in association with GS was examined. PCa

specimens were separated into 3 groups based on GS; low (GS≤6), intermediate (GS=7), and

high grade (GS≥8). Overall, no significant association between the 3 GS groups and KLK DNA

methylation was observed (Table 2.6).

Similarly, the relationship between quantitative KLK DNA methylation and pathological

stage was analyzed. PCa specimens were separated into 3 groups; organ confined pT2 cases,

pT3a cases that extended into the periprostatic tissue and pT3b+pT4 cases that infiltrated into

the seminal vesicles or bladder musculature. In cohort I, KLK6 PMR values and prevalence of

HM cases were significantly different between the three pathological stage groups (P-value =

0.029, 0.012, respectively). However, this association between KLK6 DNA methylation and

pathological stage was not observed in cohort II. KLK10 DNA methylation PMR values were

significantly associated with pathological stage in both cohorts (P-value =0.007, 0.031). Further,

in cohort I, the proportion of KLK10 HM cases significantly associated with pathological stage

(P-value=0.031) and a similar trend was also observed in cohort II (P-value=0.057).

     

    76

Table 2.6. KLK6 and KLK10 Median PMR, proportion of high methylation (HM) cases and P-values stratified according to Gleason Score and Pathological stage, Cohort I and II

Cohort I (2007-2011) Cohort II (1998-2001)

Median PMR

n (%) HM

Median PMR

n (%) HM

Median PMR

n (%) HM

Median PMR

n (%) HM

Gleason Score KLK6 KLK10 KLK6 KLK10

≤6 15.2 26 (72%) 7.44 26 (68%) 28.55 49 (96%) 14.97 41 (73%) 7 18.36 55 (73%) 14.24 57 (76%) 22.74 43 (96%) 19.06 43 (81%) ≥8 13.23 25 (68%) 7.06 23 (62%) 28.32 11 (85%) 7.39 9 (60%)

Pathological Stage KLK6 KLK10 KLK6 KLK10

pT2 14.37 54 (63%) 7.54 55 (63%) 26.49 62 (94%) 12.94 51 (68%) pT3a 18.91 35 (83%) 13.1 34 (81%) 28.55 24 (96%) 22.12 27 (90%)

pT3b+pT4 20.71 17 (85%) 23.11 17 (83%) 22.44 17 (94%) 20.81 15 (79%) pT3a+pT3b 19.67 50 (83%) 14.36 49 (82%) 23.88 36 (95%) 21.01 39 (80%)

Comparison P-valuea χ2

P-value P-valuea χ2

P-value P-valuea χ2

P-value P-valuea χ2

P-value ≤6 vs. 7 vs. ≥8 0.219 0.771 0.064 0.299 0.225 0.35 0.349 0.228 pT2 vs. pT3a vs.pT3b+pT4 0.0286 0.012 0.007 0.031 0.953 0.804 0.031 0.057

a Kruskal–Wallis one-way analysis of variance P-value

     

    77

2.4.8 ERG Immunohistochemistry

The ERG antibody used in this study has been previously shown to accurately represent the

presence of TMPRSS2-ERG gene fusions in PCa [292, 429]. Consistent with previous reports,

none of the benign prostate tissue analyzed in this study expressed the ERG protein [430, 431].

In neoplastic cells, the observed ERG nuclear staining patterns were similar to

immunohistochemical results previously reported for similar antibodies [292, 431]. Of the 150

cases that were represented on the TMA, 40 cases did not yield information due to insufficient

representation of tissue of interest. Positive ERG protein expression was observed in 37 of 110

(33.6%) of cases, consistent with other studies [432, 433]. ERG protein expression status did not

correlate with clinicopathological parameters including GS, pathological stage, and surgical

margins (Figure 1, χ2 P-values = 0.806 and 0.257, respectively). Further, age, preoperative PSA,

and prostate weight did not differ between ERG positive and ERG negative tumors (Mann-

Whitney U P-value > 0.05).

2.4.9 Correlation of ERG Expression and KLKs 6 and 10 DNA Methylation

KLKs 6 and 10 DNA methylation was significantly greater in ERG positive vs. ERG negative

PCa in cohort I (Mann-Whitney U P-values < 0.001 and 0.009, respectively). In Cohort II,

among the 124 cases analyzed, only KLK10 DNA methylation was significantly positively

associated with ERG expression (Mann-Whitney U P-values < 0.001 and 0.726, respectively).

2.4.10 DNA Methylation and Biochemical Recurrence

Univariate log-rank analysis of GS, stage, and surgical margin status showed that each

variable is a significant predictor of biochemical recurrence, indicating that the series of patients

in cohort II is representative of other populations studied [434, 435]. The two groups that

demonstrated the most significant separation for each variable in terms of biochemical

recurrence in univariate analysis are illustrated in Figure 2.3, a-c. Univariate Kaplan–Meier/log-

     

    78

rank analysis showed no association between ERG expression and biochemical recurrence (P-

value = 0.145). The relationship between biochemical recurrence and DNA methylation status of

KLK6 and KLK10 genes in cohort II was then examined. Patients with low levels of KLK10

DNA methylation showed a trend to shorter time to biochemical recurrence than patients with

high levels of KLK10 DNA methylation (log-rank P-value=0.046, Figure 2e). The overall

proportion of biochemical recurrences in the LM group was 55%, whereas in the HM group it

was only 33%. A similar trend for KLK6 DNA methylation status was observed, but was not

significant (P-value = 0.107, Figure 2d). Furthermore, multivariate Cox regression analysis of

the data was performed (Table 2.7A). In this analysis, PCa specimens were separated into two

groups that demonstrated the most significant separation in terms of biochemical recurrence in

univariate analysis. This was performed to limit the number of covariates considered in the

multivariate analysis and to maintain the power of the survival analysis. Significant predictors of

biochemical recurrence included GS, pathological stage, and surgical margin status. Low KLK10

DNA methylation status was a borderline significant predictor of biochemical recurrence

(HR=2.108, P-value=0.028, Table 2.7B) and KLK6 DNA methylation status has also shown a

trend (HR=2.155, P-value=0.074, Table 2.7A).

     

    79

Figure 2.3. Kaplan–Meier Curves of biochemical progression-free probability for (a) Gleason score, (b) stage, (c) surgical margin status, (d) KLK6 methylation status, and (e) KLK10 methylation status in cohort II (1998-2001).   Abbreviations: HM-high methylation; LM-low methylation. PCa specimens were separated into two groups that demonstrated the most significant separation in terms of biochemical recurrence in univariate analysis for each variable.

     

    80

Table 2.7. Two Independent Multivariate Cox regression analyses of biochemical recurrence with Gleason Score, pathological stage, surgical margins, (A) KLK6 and (B) KLK10 methylation status, cohort II (1998-2001)

Cohort II (1998-2001) (A) Hazard ratio 95% CI LRT P-value Gleason Score 5.08 2.37-10.88 < 0.001 Pathological stage 2.41 1.33-4.36 0.004 Surgical margin status 3.07 1.63-5.76 0.001 Low KLK6 methylation 2.15 0.99-4.70 0.074 (B)

Gleason Score 4.19 1.94-9.02 0.001 Pathological stage 3.00 1.61-5.57 0.001 Surgical margin status 2.61 1.40-4.87 0.004 Low KLK10 methylation 2.11 1.10-4.02 0.028

     

    81

2.4.11 Effect of 5-aza-2'-deoxycitidine on KLK Expression in PCa Cell Lines

To assess the biological effect of DNA methylation on KLK6 and KLK10 gene expression in

the PCa cell lines, PC-3 and 22RV1, I used the demethylating agent 5-aza-2'-deoxycitidine, a

cytidine analog that sequesters DNMTs after its incorporation into genomic DNA [436]. PC-3

and 22RV1 cells were chosen because they displayed significant DNA methylation of KLK6 and

KLK10 as determined by MethyLight (Figure 2.4). Following treatment, 9.5-fold and 6-fold

increase in KLK10 expression was observed in PC-3 and 22RV1 cells, respectively, and a 4-fold

increase in KLK6 expression was observed in PC-3 cells (Figure 2.4). KLK6 expression was not

detectable in 22RV1 cells. Hence, this cell line was excluded from this analysis. To correlate the

increase in KLK6 and KLK10 expression with 5-aza-2'-deoxycitidine-induced change in DNA

methylation, MethyLight analysis was performed. 5-aza-2'-deoxycitidine treatment led to 57%

and 49% decrease in KLK10 DNA methylation in PC-3 and 22RV1 cells, respectively, and 29%

decrease in KLK6 DNA methylation in PC-3 cells.

     

    82

Figure 2.4. (1) KLK10 and (2) KLK6 (a) methylation and (b) expression levels following treatment with 5-aza-2’-deoxycytidine in PC-3 and 22RV1 cells. Cells were treated with 4 µM 5-aza-2’-deoxycytidine for 2 days and KLK6 and KLK10 quantitative methylation and expression analysis was determined using MethyLight and RT-qPCR, respectively.

 

65.9 69.6

28.6

35.8

0

20

40

60

80

100

22RV1 PC3

Aver

age

PMR

Val

ue

11

6.0

9.5

02468

101214

22RV1 PC3

Rel

ativ

e Exp

ress

ion

1.00

3.95

0

1

2

3

4

5

PC 3R

elat

ive E

xpre

ssio

n

49.86

35.52

0

10

20

30

40

50

60

PC3

Aver

age

PMR

Val

ueTreatment

Control

(1) (2)(a) (a)

(b) (d)

     

    83

2.5 Discussion

This chapter describes a significant, tumor specific DNA hypermethylation of KLKs 6, 10 and

15 in PCa, as well as a significant association between KLK10 DNA methylation, pathological

stage, and biochemical recurrence. I chose to investigate KLK6 promoter DNA methylation,

which has been previously shown to be differently methylated in breast cancer, as well as KLKs

10 and 15, which were identified as differently methylated by genome-wide profiling using CGI

microarrays [421]. EpiTYPER analysis was then used to verify KLKs 10 and 15 DNA

methylation within individual CpG sites and for the purpose of designing primers and probes for

MethyLight analysis. For these purposes, analysis of 2 samples was sufficient.

Although earlier studies have focused on KLK10 exon 3 DNA methylation, the 5’UTR region

may have more biological significance because numerous studies have shown evidence of

abnormal methylation of CGIs within gene promoters and 5’UTR may be associated with

transcriptional gene silencing, thus potentially contributing to cancer pathogenesis [399, 404,

407-409, 437, 438].

DNA methylation levels and prevalence of KLKs 6, 10 and 15 were significantly higher in

PCa tissue compared to normal prostate tissue in cohort I. The expression of hK6 and hK10 has

been previously shown to be downregulated in PCa compared to non-malignant prostatic tissue

[439]. These results combined with our 5-aza-2'-deoxycitidine treatment data suggest that DNA

methylation plays a role in regulating the expression of these genes in PCa. This is consistent

with previous reports regarding the epigenetic regulation of KLK6 and KLK10 in breast cancer

[399-401, 440]. Alternatively, KLK15 mRNA is upregulated in PCa [397, 441-444]. The

methylation of CGI within exon 3 of KLK15 investigated in this study is thus consistent with the

notion that CGI methylation within gene bodies is a feature of transcribed genes. However,

among these DNA methylation markers, only KLK10 methylation shows good diagnostic

     

    84

characteristics and can most efficiently distinguish malignant from benign prostate tissue. This is

consistent with two previous studies that have identified promoter KLK10 DNA methylation as a

PCa associated epigenetic event using a microarray platform [410, 411]. A potential limitation

of the study is that the benign tissue used was from the vicinity of the tumor. These normal-

appearing tissues may show field effect or other changes secondary to the nearby cancer

consequently skewing the results. Future work is necessary to examine DNA methylation in

KLK10 5’UTR in normal tissue from a large independent cohort of non-PCa patients to better

define its diagnostic potential in PCa. Additionally, since the epithelium to stroma ratio varies

between normal and cancerous tissue, KLK10 methylation should be further independently

investigated in tumor vs. normal epithelia as well as tumor-associated stroma vs. normal stroma.

Our pilot analysis of KLK10 and KLK15 methylation in serum revealed it did not correlate to the

methylation levels in the corresponding tissues. Further, median KLK10 and KLK15 methylation

was elevated in the serum compared to the tissue of the same patients. Surprisingly, median

KLK10 and KLK15 methylation was also elevated in healthy controls compared to PCa. Taken

together, this suggests that the circulating DNA methylation patterns of KLKs 10 and 15 in

serum are not PCa specific and likely originate from other tissues. Given the suggested role of

DNA methylation in the regulation of KLK gene expression, it is possible that they are

methylated in tissues that do not express these KLKs such as kidneys [445].

The chapter reports that KLK10 DNA methylation in normal tissue from prostates harboring

tumors was associated with the zonal anatomy of the prostate, with significantly lower DNA

methylation status in the transition zone, independently of age. It was previously shown that

different anatomic regions of the prostate have diverse biological characteristics, including

differences in gene expression for such genes as ASPA, and differences in age-related DNA

methylation for genes such as TIMP3 and S100A2 [446, 447]. It is possible that the epigenetic

     

    85

mechanisms regulating KLK10 expression are different between the prostate zones as is KLK10

expression, also contributing to the biological differences between the zones. Further

quantitative DNA methylation and expression analyses in different prostate anatomical regions

of normal tissue from non-PCa patients are required to investigate this possibility. On the other

hand, KLK6 DNA methylation in normal tissue from prostates harboring tumors seems to be

age-related. These results are not surprising since it has been previously shown that DNA

hypermethylation of numerous genes, such as GSTP1, increases with age in the normal prostate

tissue [448]. Additionally, KLK6 DNA methylation in the normal tissue (from PCa patients) was

associated with higher DNA methylation in the matched cancerous tissue from the same

prostate. This may be possibly explained by the field effect, which refers to molecular

abnormalities in histologically benign epithelium contiguous to cancerous tissue [449]. This has

been previously shown for other genes such as GSTP1 and RARβ2 [450, 451]. However, it is

necessary to analyze KLK6 DNA methylation in normal prostatic tissue from healthy individuals

to support this possibility. Additionally, since the epithelium to stroma ratio varies between

normal and cancerous tissue, KLK6 methylation should be further independently investigated in

tumor vs. normal epithelia as well as tumor-associated stroma vs. normal stroma.

The results show that DNA methylation of KLK6 and KLK10 in PCa is not correlated with

GS, but is associated with pathological stage. Significant increase in KLK10 DNA methylation

was observed for pathological stages pT3+pT4 vs. pT2 in both cohorts. This observation may

have important implications for PCa detection and management as high KLK10 DNA

methylation levels detected in prostate tissue may mark clinically relevant disease and could be

incorporated in predictive models for preoperative prostate cancer staging. Additionally, these

results imply that the tumor suppressive effect of this protease is linked to tumor invasion rather

than gland differentiation. However, a previous publication found no significant association

     

    86

between KLK mRNA or protein expression and pathological stage in cancerous prostate tissues

[439, 452]. Therefore, future work is necessary to elucidate the biological significance of KLK10

DNA methylation to protein expression and other epigenetic mechanisms that could potentially

regulate these proteins expression. Similarly, a significant increase in KLK6 DNA methylation

from pathological stage pT2 to pT3+pT4 in cohort I was demonstrated. However, this significant

finding generated in cohort I was not confirmed in cohort II. This can be possibly explained by

differences in patient cohort characteristics or a spurious correlation in cohort I. This

discrepancy needs to be further investigated in a larger independent cohort.

In PCa, the discovery of frequent TMPRSS2-ERG gene fusion and its biological implications

has led to numerous investigations into its diagnostic and prognostic clinical applications [213].

To date, studies addressing the relationship between TMPRSS2-ERG gene fusions and PCa

aggressiveness or clinical outcome have provided conflicting results [218]. This study seems to

lend support to the lack of prognostic role for TMPRSS2-ERG status. The conflicting evidence

on the prognostic significance of TMPRSS2-ERG fusion is likely due to inter- and intra-

individual heterogeneity in PCa biology, differences in examined cohorts, ascertainment biases

in patient recruitment and/or methodological differences. More detailed molecular sub-

classifications of ERG positive PCa cases may better explain why some act as indolent while

others act as aggressive disease in different cohorts. In fact, several studies investigating the

biological role of TMPRSS2-ERG fusion and the subsequent ERG overexpression in PCa

development in transgenic murine prostates have found that on its own, it is insufficient to

induce the development of invasive carcinoma, indicating that other genetic and epigenetic

factors are necessary for PCa initiation and progression [453]. Although the epigenetic factors

have not been as well investigated, TMPRSS2-ERG has been associated with distinct

methylation, histone modifications, and miRNA expression profiles [418, 419]. Accordingly,

     

    87

significant correlations between ERG-positive tumors and methylation of KLK6 in cohort I and

KLK10 in both cohorts was observed. The functional mechanisms that underlie the relationship

between KLK methylation and ERG expression is unknown and needs further investigation. One

possibility is that ERG drives the overexpression of EZH2 and DNMTs in PCa, subsequently

leading to increased methylation at specific gene loci [314, 454]. Alternatively, the observed

association may be merely the result of accumulation of genomic and epigenomic alterations

known to be associated with aggressive PCa.

On multivariate Cox regression analyses, low KLK10 DNA methylation in cohort II was

associated with shorter time to biochemical recurrence, but not more significant than GS,

pathological stage or surgical margin status. There may be multiple reasons for this occurrence.

For example, the KLK10 DNA methylation might be one of a panel of methylation markers,

which together predict PCa disease course with a high degree of accuracy. This phenomenon has

been demonstrated in numerous studies that have shown the added value of a multi-gene DNA

methylation panels for diagnosis and prognosis of PCa in comparison to single gene DNA

methylation [455, 456]. Similarly, the methylation of KLK10 might be used in a panel of

methylation markers, which together will accurately predict PCa disease course and/or outcome.

In addition, low KLK10 DNA methylation was also found to be associated with tumors of

lower pathological stage. This paradoxical observation is intriguing and may have several

possible explanations. First, KLK10 DNA methylation may have two phases in PCa: an initial

increase in DNA methylation from organ confined pT2 to invasive pT3+pT4 stages, followed by

a decrease in DNA methylation after tumor progression. Second, perhaps most cells in a prostate

tumor have an increase in KLK10 DNA methylation during the progression from organ-confined

to locally-invasive PCa. However, a subset of organ-confined tumors that have a unique

epigenetic identity, including low DNA methylation of KLK10, are more prone to be involved in

     

    88

biochemical recurrence of PCa. Since previous studies have demonstrated no statistically

significant difference in the biochemical recurrence rate between patients with hK10 protein

positive and negative tumors, it is possible that KLK10 DNA methylation is just an

epiphenomenon associated with these processes [439, 452]. Yet, KLK15 DNA methylation was

not associated with clinicopathological variables associated with aggressive PCa. This suggests

that not all aberrant DNA methylation marks in the KLK locus (19q13.3-13.4) are

indiscriminately associated with PCa aggressiveness and suggests a specific role for KLK10 as a

prognostic biomarker for the disease.

The chapter reports the use of MethyLight technology for KLK methylation analysis. This

approach has several advantages. It is a high-throughput, sensitive, specific and quantitative

assay that requires very small amounts of DNA, thus making it suitable to be used in clinical

laboratories. However, this approach also has certain limitations, including reliance on bisulfite-

treated DNA as a template for PCR, and inability to recognize heterogeneously methylated

molecules and their significance to PCa. Another potential limitation of the study is that there is

a higher percentage of GS>8 cases in cohort I compared to cohort II. Having more GS>8 cases

in cohort I vs. cohort II may introduce a bias, skew the survival analysis of our study and skew

association analysis between KLK DNA methylation and GS. However, no follow-up data is

available for cohort I and no significant association between KLK DNA methylation and GS was

found in the study.

In conclusion, KLK6, 10 and 15 DNA methylation was characterized for the first time in two

independent cohorts of PCa specimens and matched normal prostatic tissues. Significant, tumor-

specific DNA hypermethylation of all three KLKs was observed, implying a diagnostic potential.

The results suggest that serum may not be an appropriate bodily fluid for further analysis of the

diagnostic potential of KLK10 methylation. The potential of KLK6 DNA methylation as a

     

    89

marker of field effect in normal peripheral zone prostate tissue was demonstrated. Further, the

significant association of increased KLK10 methylation with advanced pathological stage and

ERG status was validated. Low DNA methylation of KLK10 was found to potentially further

predict patient biochemical recurrence risk. Future work is necessary to further elucidate both

the clinical utility and functional relevance of KLK10 DNA methylation as a progression

biomarker.

     

    90

Chapter 3 Discovery of Alterations in DNA Methylation of KLKs and Associated

Pathways in Prostatic Carcinoma Glands with Cribriform Architecture or Intraductal Carcinoma

Ekaterina Olkhov-Mitsel,1,2 Farshid Siadat,3 Dominique Trudel,3 Ken Kron,1,2 Liyang Liu, 1,2 Theodorus van der Kwast 2,3 and Bharati Bapat 1,2,3

1Samuel Lunenfeld Research Institute; Mount Sinai Hospital; Toronto, ON Canada; 2Department of Laboratory Medicine and Pathobiology; University of Toronto; Toronto, ON Canada;

3Department of Pathology; University Health Network; University of Toronto; Toronto, ON Canada.

The work presented in this chapter has been primarily contributed by Olkhov-Mitsel E., Van der Kwast T., a pathologist, assisted in specimen retrieval and histopathological confirmation. The pathology residents Siadat F. and Trudel D. have assisted with histopathological review of cribriform architecture and intraductal carcinoma. Trudel D. also read the ERG immunohistochemistry data used in this chapter. Kron K., a former Ph.D. student, and Liu L., a former MSc. student, performed MethyLight analysis for APC, CYP26A1, HOXD3, HOXD8, RASSF1A, TBX15 and TGFβ2. Kron K. has also constructed the TMA used for ERG immunohistochemistry in this chapter.   Bapat B. supervised the project (including project development, experimental design, analysis and interpretation of data) and critically reviewed the manuscript and thesis.

A manuscript based on the data presented in this chapter is currently in preparation.

     

    91

Chapter 3 Discovery of Alterations in DNA Methylation of KLKs and Associated Pathways in

Prostatic Carcinoma Glands with Cribriform Architecture or Intraductal Carcinoma

3.1 Summary

Intraductal carcinoma (IDC) and cribriform architecture are emerging as clinically significant

indicators of aggressive PCa. However, limited studies have investigated molecular alterations

in association with these clinicopathologic entities. Herein, DNA-methylation of KLKs and the

previously discovered methylation markers APC, CYP26A1, HOXD3, HOXD8, RASSF1A,

TBX15 and TGFβ2 were investigated in association with IDC and/or cribriform architecture in

GP4 PCa of 91 GS7 RP specimens. DNA methylation of KLK6 and KLK10 were not

significantly associated with the presence of IDC or cribriform architecture. Median methylation

levels of TBX15 were significantly elevated in the presence of any amount of IDC (P-

value=0.020). Stratification of cribriform architecture revealed that LC was associated with a

significant increase in median methylation levels for APC and RASSF1A, meanwhile SC was

linked to elevated APC and TBX15 methylation (P-values = 0.008, 0.012, 0.033, 0.023,

respectively). Stratification of cases by ERG expression showed that APC, RASSF1A and TBX15

methylation was significantly associated with IDC and/or cribriform only in ERG positive cases.

LC and HOXD3 methylation were both significant independent predictors of biochemical

recurrence in univariate and multivariate analyses. Importantly, patients with either high levels

of HOXD3 methylation or LC in GP4 showed a more significant association with biochemical

recurrence than pathological stage or either marker on its own in multivariate analyses. The

findings suggest that among the various morphological GP4 features, those with cribriform

architecture and/or IDC have a distinct DNA methylation profile with significant prognostic

value. The mere presence of cribriform architecture or IDC seems to be associated with the

observed increase in APC, RASSF1A and TBX15 methylation.

     

    92

3.2 Introduction

The GS classification system is among the best predictors of clinical outcome after RP, and is

universally used to guide PCa treatment. Within this system, GP4 prostatic adenocarcinomas

represent the most heterogeneous group of neoplasms with a wide range of clinical outcomes

[162]. In the context of GS7, it is considered intermediate risk for biochemical recurrence [457].

Therefore, further stratification of risk in this patient population would be of particular value for

clinical decision-making.

GP4 is assigned to adenocarcinomas when they contain any one of the following architectural

patterns: small or large fused growth pattern, poorly formed, glomeruloid and cribriform

architectures [162]. Moreover, this pattern is associated with IDC, a distinct histopathologic

entity characterized by malignant cells spanning the lumen of prostatic ducts and acini [164-

166]. Although specific morphologies of PCa, such as IDC and cribriform, have been recognized

for decades, their independent clinical significance is only now emerging. Studies have shown

that IDC on biopsy is indicative of invasive PCa with adverse clinical course, warranting a

definite treatment [166, 167]. Further, IDC detected in RP specimens is strongly associated with

large tumor volume, increased risk of extraprostatic extension and seminal vesicle invasion,

positive surgical margins and a worse clinical outcome [165, 458, 459]. Similarly, cribriform

architecture, characterized by confluent epithelial proliferation with multiple lumina, without

intervening stroma, has been demonstrated as a significant independent prognosticator for PCa

[460]. Cribriform has been associated with lymph node and distant metastasis as well as

biochemical recurrence and disease-specific death, outperforming any of the other GP4

architectural features as a prognosticator [461-463].

Therefore, it is of utmost importance to distinguish IDC and cribriform growth patterns from

other histological entities of PCa specimens. However, difficulties in the distinction of these

     

    93

clinicopathologic entities and other lesions based on morphologic criteria alone have been

reported [464]. In this regard, molecular features such as DNA hypermethylation may be of

value. Discovery of DNA hypermethylation events that specifically associate with IDC and

cribriform architectural patterns may expand our understanding of their biological basis and

have potential clinical implications in their distinction on biopsies and RPs. Furthermore, DNA

methylation markers may potentially serve as molecular indicators of IDC and/or cribriform

architectural patterns on biopsies and/or in biofluids. Biopsies may miss these morphologies,

especially when they occupy a very small percentage of the prostate while alterations in DNA

methylation may be detected as a result of the field effect. Importantly, DNA is stable and easy

to isolate from biofluids, while methylated DNA is readily amplifiable and easily detectable

using PCR-based approaches [415, 416]. This may allow for DNA methylation markers to be

used as surrogate indicators of IDC and/or cribriform in biofluids, avoiding the need for

histopathological examination of biopsies.

In this chapter, IDC and cribriform architectural features on RP were characterized and

compared for their associations with DNA methylation of KLK6 and KLK10 genes, given their

prognostic potential discovered in Chapter 2. Further, the previously discovered methylation

markers APC, CYP26A1, HOXD3, HOXD8, RASSF1A, TBX15 and TGFβ2 were assessed in

association with IDC and cribriform as well as clinicopathologic features of PCa. Importantly,

the abovementioned genes are involved in numerous pathways that associate them with KLKs

and implicate them in carcinogenesis, including developmental and proliferative signaling, DNA

damage response, as well as evasion of growth suppressors and apoptosis [465-473].

     

    94

3.3 Materials and methods  

3.3.1 Histologic Pattern Annotation

A total of 98 GS7 patients who underwent RP from 1998 to 2001 at UHN, Toronto were

identified from a previously published cohort [291, 292, 422]. Patient consent was obtained for

specimen accrual following RP into the UHN tissue bank. All samples and clinicopathological

follow-up information were obtained according to the protocols approved by the Research Ethics

Board at Mount Sinai Hospital and UHN, Toronto. The complete set of H&E-stained slides from

each prostatectomy was collected and reviewed by an expert pathologist (TVDK) to confirm GS

(WHO/ISUP criteria), stage (TNM), and surgical margin status. The complete set of H&E slides

were then reevaluated by DT and the presence or absence of IDC and/or LC was recorded as

described previously by Trudel et al.[170]. This data was used for exploratory analysis of

potential associations between DNA methylation patterns and IDC and/or LC within the same

prostate for the first time (Figure 3.1). Next, to identify whether the alterations in DNA

methylation in association with IDC and/or LC occur within the same tumor area, the marked

areas on the H&E stained slides used for DNA methylation analysis were then reevaluated by FS

(Figure 3.1). The selected H&E slides were evaluated by FS for the presence or absence of (1)

IDC, defined as a lumen-spanning solid or cribriform neoplastic proliferation distending

antecedent prostate glands or ducts; (2) cribriform, confluent epithelial proliferation with

multiple lumina, without intervening stroma; (3) SC, an area of cribriform growth pattern below

the size of an average prostatic benign gland; (4) LC, characterized by cribriform growth pattern

exceeding the size of an average benign gland. For each specimen, the percentage of tumor

(estimated on H&E) occupied by the four histologic patterns was also recorded.

     

    95

 

Figure 3.1. Schematic representation of study design

 

 

     

    96

3.3.2 DNA Methylation Analyses

DNA was extracted from 10 µm sections of FFPE blocks matching the selected H&E slides

using a QIAamp DNA Mini Kit (Qiagen) as described in section 2.3.5 of this thesis (page 60).

The concentration and quality of DNA samples was assessed using NanoDrop 1000

spectrophotometer (Thermo Scientific). Next, 100-400 ng of extracted DNA from GP3 and GP4

for each GS7 case was bisulfite modified separately using EZ DNA Methylation Gold Kit

(Zymo Research) according to the manufacturer’s protocol. The quantitative MethyLight assay

was performed by KK and LL using 20 ng of converted DNA for GP3 and GP4 separately as

described in section 2.3.5. The published primer sequences are shown in Table 3.1 [291, 292,

422]. A PMR score for each gene was calculated for GP3 and GP4 separately per each case as

described in section 2.3.5. The PMR values for each individual pattern within a GS7 case were

then averaged to calculate the PMR value for that case.

3.3.3 Tissue microarray Construction and ERG Immunohistochemistry

A range from 3 to 13 of 0.6 mm cores were taken from each of the 98 GS7 cases to represent

each of the primary and secondary GP present within the case. This was performed as part of a

larger TMA described previously by Kron et al. [292]. ERG immunohistochemistry was

performed as described in section 2.3.7 of this thesis (page 61).

3.3.4 Statistical Analyses

The Pearson χ2-square test was used to analyze proportional differences in the presence of

IDC and/or cribriform between clinicopathological categories and ERG groups. The Fisher exact

method was used to replace the Pearson test when spreadsheet cell counts were <5. Association

between PMR values and IDC and/or architectural pattern was analyzed using the Mann-

Whitney U-test (for two-group comparisons) and Kruskal-Wallis test (for four-group

comparisons). Mann-Whitney U-test was also applied to correlate between PMR values and

     

    97

ERG status. For univariate and multivariate biochemical recurrence analyses, the continuous

PMR variable was separated into those with HM and those with LM based on a third quartile

threshold as previously described [422]. Univariate biochemical recurrence-free probabilities

were assessed using the Kaplan-Meier curve and log-rank tests. Multivariate Cox proportional

hazards regression analysis and LRT were used to analyze individual contributions of each

variable to biochemical recurrence-free probabilities. Each factor was coded as a binary variable

for Cox regression. Stage was coded as organ confined (pT2) or locally advanced (pT3/pT4),

surgical margin status as negative or positive, PSA as above or below median, and methylation

as LM or HM. For all described methods, two-sided P-values of ≤0.05 were considered

significant. All statistical analyses were conducted using SPSS version 21 software (SPSS, IBM

Software).

     

    98

Table 3.1. Primer and probe sequences used in MethyLight assay for APC, CYP26A1, HOXD3, HOXD8, KLK6, KLK10, RASSF1A, TBX15, TGFβ2 and ALU genes

Gene Sequence ALU Forward: 5’- GGTTAGGTATAGTGGTTTATATTTGTAATTTTAGTA -3’

Reverse: 5’- ATTAACTAAACTAATCTTAAACTCCTAACCTCA -3’ Probe: 5’FAM- CCTACCTTAACCTCCC -3’BHQ1

APC ENSG00000134982* Chr5: 112737781-112737856§

Forward: 5’- GAACCAAAACGCTCCCCAT -3’ Reverse: 5’- TTATATGTCGGTTACGTGCGTTTATAT -3’ Probe: 5’FAM - CCCGTCGAAAACCCGCCGATTA -3’BHQ1

CYP26A1 ENSG00000095596 Chr10: 93069103-93069230

Forward: 5’- TTGTAGAGATTCGACGTACGCGG -3’ Reverse: 5’- AAAACCTTCCGTCAAACATCCTCTACG -3’ Probe: 5’FAM - ACGCCCACGACGTACCCGCTTCCTTAC-3’BHQ1

HOXD3 ENSG00000128652 Chr2: 176163186-  176163274

Forward: 5’- TTAAAGGTTTATGGTTGCGC-3’ Reverse: 5’- TTACGAACACTAAACTACACCCG-3’ Probe: 5’FAM - ACAAAACGTTCCCGACGCTTCTAAAA -3’BHQ1

HOXD8 ENSG00000175879 Chr2: 176128934-  176129040

Forward: 5’- TAGTCGGTTTTGGTTCGTTGC -3’ Reverse: 5’- CGTTCTAAAACGAAAAAAAAAACTCGCG -3’ Probe: 5’FAM - TCCTCGAACAAAACGCGACTCCCGAATCTC -3’BHQ1

KLK6 ENSG00000167755 Chr19: 50969651-50969764

Forward: 5’- AGGAAGTTATTGATGTAATCGTTTTTTCG -3’ Reverse: 5’- AAAACAATCGAACTTTATCCGCC -3’ Probe: 5’FAM - ACTCCGACCTCAACCTCTCTTCCGACGAACA -3’BHQ1

KLK10 ENSG00000129451 Chr19:  51019335-51019434

Forward: 5’- GAGGGGGAAATTTCGGGCGC-3’ Reverse: 5’- CCCTCGCGACATCTTCCCG -3’ Probe: 5’FAM - ACCCGAATAAAACGCTCTCCGCGCCCCA -3’BHQ1

RASSF1A ENSG00000068028 Chr3:  50340720-50340784¶

Forward: 5’- ATTGAGTTGCGGGAGTTGGT -3’ Reverse: 5’- ACACGCTCCAACCGAATACG-3’ Probe:5’FAM - CCCTTCCCAACGCGCCCA -3’BHQ1

TBX15 ENSG00000092607 Chr1: 118984438-118984535

Forward: 5’- GCGGTTTTGTAAGTATATTGTTGCG -3’ Reverse: 5’- ACTCCGAATAAAACAAAAACTAAAATCCG -3’ Probe: 5’FAM - CAAATAACGCCGCCGAACGCCT -3’BHQ1¶

TGFβ2 ENSG00000092969 Chr1: 218346933-  218347007

Forward: 5’- TTTTAGGAGAAGGCGAGTCG -3’ Reverse: 5’- CTCCTTAACGTAATACTCTTCGTCG -3’ Probe: 5’FAM - TCTCGCGCTCGCAAACGACC -3’BHQ1

*Ensembl gene accession number §Human genome assembly GRCh38/hg38 ¶The RASSF1A TaqMan hydrolysis probe sequence contains the SNP rs4688725

 

     

    99

3.4 Results  

3.4.1 IDC and Cribriform Architecture: Clinical Correlations in GS7 Cases

A review of the 98 GS7 RP specimens revealed that 91 cases had available H&E slides. Of

these, IDC and/or LC growth pattern was present in 48 men (52.7%). GS7 cases composed

predominantly of GP4 had a higher prevalence of IDC and/or LC architectural features

compared to cases composed predominantly of GP3 (71.4% [20/28] vs. 44.4% [28/63], χ2 P-

value < 0.001, Table 3.2). Further, cases positive for IDC/LC patterns were associated with ERG

expression and higher pathological stage (χ2 P-values = 0.017 and 0.022, respectively, Table

3.2).

A total of 31 of 91 GS7 cases (42.9%) in this cohort had biochemical recurrence. Univariate

log-rank analysis with Kaplan-Meier curves revealed that statistically significant predictive

factors for biochemical recurrence were preoperative PSA above median, advanced pathological

stage and positive surgical margin status (Figure 3.2, a-c), while age, prostate weight and ERG

expression were not significant factors. Importantly, cases positive for IDC/LC were also

associated with a significantly shorter time to biochemical recurrence (Figure 3.2d). However, in

multivariate Cox proportional hazards regression analysis, adjusting for the effects of significant

clinicopathological predictive factors, IDC/LC did not show an independent significant

prognostic effect on biochemical recurrence (Table 3.3).

     

    100

Table 3.2. Proportion of GS7 cases with IDC/LC and P-values stratified according to conventional pathology parameters, age, prostate weight and ERG expression status

IDC/LC

Not Present (0%) Present (>0%)

Gleason Score 7, predominant GP 3 56% 44%

7, predominant GP 4 29% 71% Pearson Chi-Square P-value 0.017 Pathological Stage

pT2 68% 32% pT3 28% 72% Pearson Chi-Square P-value 0.0002 Surgical Margins

Negative 49% 51% Positive 42% 58% Pearson Chi-Square P-value 0.523 Age

Below Median 49% 51% Above Median 46% 54% Pearson Chi-Square P-value 0.774 Preoperative PSA

Below Median 58% 42% Above Median 42% 58% Pearson Chi-Square P-value 0.135 Prostate Weight

Negative 61% 39% Positive 42% 58% Pearson Chi-Square P-value 0.082 ERG status of case

Negative 60% 40% Positive 35% 65% Pearson Chi-Square P-value 0.022

     

    101

Figure 3.2. Kaplan-Meier curves of biochemical recurrence probability within GS7 for (a) preoperative PSA, (b) pathological stage, (c) surgical margin status and (d) presence of IDC/LC

Table 3.3. Multivariate Cox regression analysis of biochemical recurrence with pathological stage, surgical margin status, preoperative PSA and IDC/LC in GS7 radical prostatectomy specimens

Hazard ratio 95% CI LRT P-value

Pathological Stage 4.283 1.81-9.81 0.001 Surgical margin Status 2.314 1.12-4.80 0.024 Preoperative PSA 1.647 0.81-3.34 0.167 IDC/LC

1.274 0.57-2.83 0.551

     

    102

3.4.2 Aberrant DNA Methylation in GS7 IDC and Cribriform Architecture

First, exploratory analysis for correlations between IDC/LC presence anywhere within the

PCa GS7 cases in cohort II and DNA methylation within the outlined areas selected as described

in section 3.3.1 was performed (Figure 3.1). KLK6 and KLK10 mean methylation levels were

elevated in cases positive for IDC/LC but did not reach statistical significance (Figure 3.2).

Mann-Whitney analysis of seven additional methylation marks previously assessed in the lab,

demonstrated significantly greater APC, CYP26A1, RASSF1A and TBX15 methylation in cases

positive for IDC/LC (Mann-Whitney P-values = 0.013, 0.015, 0.007, 0.005, respectively, Figure

3.3). Meanwhile, HOXD3, HOXD8, and TGFβ2 methylation did not significantly differ with the

presence of these clinicopathologic entities.

ERG expression in this cohort significantly correlated with the prevalence of IDC/LC (Table

3.2) and methylation of APC, HOXD3, HOXD8 and TBX15 genes (Mann Whitney P-values =

0.035, 0.002, 0.007, <0.00002, respectively, Table 3.4). Accordingly, stratification of cases by

ERG expression revealed that APC, CYP26A1, RASSF1A and TBX15 methylation was

significantly associated with ID/LC only in ERG positive cases (Table 3.4).

     

    103

Table 3.4. Median PMR and Mann-Whitney P-values stratified by ERG expression status and further sub-stratified by IDC/LC status in GS7 radical prostatectomy specimens

ERG status of case

Negative Positive P-value

APC PMR 35.3 46.25 0.035 IDC/LC - IDC/LC + P-value IDC/LC - IDC/LC + P-value 25.75 43.4 0.788 30.65 51.4 0.005 CYP26A1 PMR 16.3 20.28 0.145 IDC/LC - IDC/LC + P-value IDC/LC - IDC/LC + P-value 16.24 19.92 0.808 13.36 22.56 0.006 HOXD3 PMR 20.0 32.2 0.002 IDC/LC - IDC/LC + P-value IDC/LC - IDC/LC + P-value 19 21 0.72 37 32 0.876 HOXD8 PMR 23.5 32.85 0.007 IDC/LC - IDC/LC + P-value IDC/LC - IDC/LC + P-value 24.825 23.35 0.71 30.8 32.85 0.633 KLK6 PMR 27.2 26.9 0.981 IDC/LC - IDC/LC + P-value IDC/LC - IDC/LC + P-value 28.1 30.55 0.491 30.45 23.9 0.164 KLK10 PMR 14.7 15 0.841 IDC/LC - IDC/LC + P-value IDC/LC - IDC/LC + P-value 18.25 9.65 0.162 17 13 0.947 RASSF1A PMR 74.7 85.95 0.114 IDC/LC - IDC/LC + P-value IDC/LC - IDC/LC + P-value 71.4 86.55 0.115 75.93 103.25 0.050 TBX15 PMR* 8.1 21.1 <0.001 IDC/LC - IDC/LC + P-value IDC/LC - IDC/LC + P-value 3.8 8.95 0.691 10.4 29.29 0.005 TGFb2 PMR 1.3 0.2 0.067 IDC/LC - IDC/LC + P-value IDC/LC - IDC/LC + P-value 1.2 1.125 0.907 0.1 0.63 0.188

Abbreviations: PMR = percent of methylated reference. IDC = intraductal carcinoma. LC= large cribriform.

* TBX15 PMR values showed the most significant differences between IDC/LC positive and negative cases in terms of fold change and p-value.

     

    104

Figure 3.3. Median percent of methylated reference (PMR) values for APC, CYP26A1, HOXD3, HOXD8, RASSF1A, TBX15, TGFβ2, KLK10 and KLK6 genes in the presence and absence of intraductal carcinoma (IDC)/large cribriform (LC) within GS7. Circles represent outliers (between 1.5 and 3 interquartile ranges from 3rd quartile) and asterisks represent extreme cases (greater than 3 interquartile ranges from 3rd quartile). Mann-Whitney U P-values ≤0.05 are labeled.

     

    105

3.4.3 DNA Hypermethylation in Areas with GP4 IDC and Cribriform Architecture

In an effort to explore further the significance of DNA methylation events occurring in IDC

and cribriform morphological patterns, the GP4 area of the GS7 cases were re-reviewed for the

presence of IDC and cribriform architecture (including SC and LC) only within the areas

previously analyzed for DNA methylation of APC, CYP26A1, HOXD3, HOXD8, RASSF1A,

TBX15 and TGFβ2 (Figure 3.1). KLK methylation was not analyzed in a pattern specific manner

in these cases and thus was excluded from analysis.

The presence of cribriform growth pattern was identified in GP4 of 61 of 91 GS7 cases

(67.0%), while IDC was present in 21 of 91 GP4 cases (23.0%). Breakdown of percentage

involvement of each pattern in the tissue region selected for DNA methylation analysis is listed

in Table 3.5. Notably, the proportion of cases positive for IDC and/or cribriform (including SC

and LC) in the specific area used for methylation analysis did not differ between GS7 cases

composed predominantly of GP4 vs. predominant GP3, but did associate with pathological stage

and preoperative PSA (Table 3.6).

Thereafter, we sought to investigate gene specific DNA methylation differences between GP4

positive for IDC or cribriform compared to those that do not within the same area. GP4 with

IDC had significantly elevated median TBX15 methylation PMR values within the same area

(Mann-Whitney U P-value=0.020, Table 3.5). Similarly, GP4 with a cribriform component had a

significant increase in median PMR for APC, RASSF1A and TBX15 (Mann-Whitney U P-values

= 0.045, 0.007, 0.013, respectively, Table 3.5). Further, stratifying the cribriform architectural

pattern into SC and LC, Mann-Whitney analysis demonstrated that TBX15 median PMR was

significantly elevated in GP4 with SC component, while RASSF1A median PMR was

significantly elevated with the presence of LC, and median APC PMR was significantly higher

     

    106

in GP4 with either pattern (Mann-Whitney U P-values = 0.023, 0.012, 0.019, 0.008,

respectively, Table 3.5).

The IDC, SC, and LC architectural patterns were then divided into four groups based on

percentage involvement of each lesion in the GP4 area used for DNA methylation analysis as

indicated in Table 3.5. APC, RASSF1A and TBX15 median methylation was elevated from 0% to

>0-5% involvement of each morphology. However, no clear pattern of incremental

hypermethylation in accordance with stepwise increase in percentage of IDC, SC or LC pattern

involvement was observed.

     

    107

Table 3.5. Median PMR values stratified by IDC and cribriform status within the same GP4 area

APC HOXD3 HOXD8 CYP26A1

RASSF1A

TBX15 TGFb2

IDC N Median

PMR Median

PMR Median

PMR Median

PMR Median

PMR Median

PMR Median

PMR 0% 70 41.0 25.3 25.6 16.3 87.7 12.2 0.0

>0% 21 48.9 27.5 30.7 18.9 101.8 30.1 0.0 P-valuea

0.106 0.765 0.146 0.515 0.341 0.020 0.512

IDC Category N 0% 70 41.0 25.3 25.6 16.3 87.7 12.2 0.0

>0 to 5% 6 62.1 31.0 34.2 21.3 75.2 19.7 0.0 >5 to 25% 9 46.2 25.4 30.7 33.3 101.8 36.6 1.1

>25% 6 55.0 28.1 28.0 11.1 126.8 30.1 0.0 P-valueb

0.338 0.794 0.516 0.242 0.432 0.093 0.702

Cribriform N 0% 30 31.7 25.8 25.7 16.3 69.5 10.0 0.0

>0% 61 47.3 27.1 27.7 18.5 99.2 21.6 0.0 P-valuea

0.045 0.888 0.278 0.426 0.007 0.013 0.287

SC§ N 0% 35 33.0 25.6 25.9 17.1 81.5 10.0 0.0

>0% 56 49.3 28.0 27.7 18.3 92.4 23.3 0.0 P-valuea

0.033 0.879 0.636 0.495 0.181 0.023 0.852

SC Category N 0% 35 33.0 25.6 25.9 17.1 81.5 10.0 0.0

>0 to 5% 19 38.3 31.7 23.4 13.4 101.6 19.2 0.0 >5 to 25% 24 67.9 29.4 29.2 18.1 86.3 27.4 0.0

>25% 13 46.6 25.3 28.2 26.0 95.4 23.3 0.0 P-valueb

0.019 0.794 0.591 0.644 0.418 0.083 0.782

LC* N 0% 76 39.4 27.5 25.6 17.9 84.7 15.4 0.0

>0% 15 66.1 25.3 46.2 23.3 111.5 30.1 1.1 P-valuea

0.008 0.743 0.060 0.209 0.012 0.100 0.221

LC Category 0% 76 39.4 27.5 25.6 17.9 84.7 15.4 0.0

>0 to 5% 5 61.1 21.6 46.2 46.4 109.4 30.1 0.0 >5 to 25% 4 60.0 23.2 45.2 13.0 113.7 27.7 5.1

>25% 6 74.6 25.3 41.3 23.3 138.2 30.2 5.1 P-valueb

0.068 0.867 0.271 0.114 0.082 0.436 0.319

a Mann-Whitney U P-value bKruskal-Wallis P-value §SC, Small Cribriform - an area of cribriform growth pattern below the size of an average prostatic benign gland *LC, Large Cribriform - an area of cribriform growth pattern exceeding the size of an average benign gland.

     

    108

Table 3.6. IDC and cribriform stratified by conventional pathology parameters, age, ERG status, PSA, and prostate weight

IDC Cribriform Small Cribriform§ Large Cribriform* Gleason Score Negative Positive Negative Positive Negative Positive Negative Positive

7a 83% 17% 35% 65% 40% 60% 84% 16% 7b 64% 36% 29% 71% 36% 64% 82% 18% P-value 0.056 0.552 0.720 0.814 Pathological Stage

pT2 89% 11% 47% 53% 47% 53% 91% 9% pT3 64% 36% 20% 80% 31% 69% 78% 22% P-value 0.006 0.007 0.130 0.081 Surgical Margins

Negative 77% 23% 31% 69% 40% 60% 80% 20% Positive 77% 23% 38% 62% 35% 65% 92% 8% P-value 1.000 0.481 0.633 0.153 Age Below Median

74% 26% 33% 67% 43% 57% 89% 11% Above Median

80% 20% 33% 67% 33% 67% 78% 22% P-value 0.491 0.941 0.320 0.144 Preoperative PSA Negative 77% 23% 45% 55% 52% 48% 77% 23% Positive 78% 22% 22% 78% 27% 73% 78% 22% P-value 0.932 0.022 0.017 0.057 Prostate Weight

Negative 79% 21% 38% 62% 40% 60% 79% 21% Positive 80% 20% 32% 68% 36% 64% 80% 20% P-value 0.912 0.542 0.695 0.044

§Small Cribriform - an area of cribriform growth pattern below the size of an average prostatic benign gland *Large Cribriform - an area of cribriform growth pattern exceeding the size of an average benign gland

 

     

    109

3.4.4 Prognostic Impact of IDC, Cribriform and DNA Methylation Markers Within GS7

The presence of cribriform, and more significantly LC, in GP4 areas (within GS7) used for

DNA methylation analysis were predictors of biochemical recurrence in univariate analysis

(Figure 3.4, a-b, log-rank P-value = 0.032, 0.022, respectively). A similar trend was observed for

the presence of IDC and SC, but it did not reach significance, likely due to limited number of

samples (Figure 3.4, c-d, log-rank P-value = 0.061, 0.115, respectively). Furthermore,

multivariate Cox regression analysis of our data was performed. Significant predictors of

biochemical recurrence included pathological stage, surgical margin status and presence of LC

(Table 3.7A). The presence of cribriform has also shown a trend (Table 3.7B). Next, the

relationship between biochemical recurrence and each of the methylation markers analyzed in

GP4 of the GS7 RP cases was examined. Of the seven methylated genes, only HM of HOXD3

and HOXD8 in GP4 showed significant association with shorter time to biochemical recurrence

in univariate analysis (Figure 3.5, log-rank P-values = 0.015, 0.017, respectively). However,

only HODX3 methylation remained significant in multivariate analysis (Table 3.7C).

In a multivariate model including both LC and HOXD3 methylation status within the same

GP4 area, both were significant independent predictors of biochemical recurrence (Table 3.7D).

Importantly, combining patients with either high levels of HOXD3 methylation or LC in GP4

into one category showed a more significant association with biochemical recurrence than

pathological stage or either marker on its own (Table 3.7E). However, the additive predictive

effect of having both LC and HOXD3 HM could not be assessed because only three GP4 tumors

contained both marks in this cohort.

     

    110

Figure 3.4. Kaplan-Meier curves of biochemical recurrence probability for presence of (a) cribriform, (b) large cribriform (LC), (c) small cribriform (SC) and (d) intraductal carcinoma (IDC) in GP4 within GS7.

     

    111

Figure 3.5. Kaplan-Meier curves of biochemical recurrence probability for (a) HOXD3 methylation and (b) presence of large cribriform (LC) in GP4 within GS7. Abbreviations: HM-high methylation; LM-low methylation.

     

    112

Table 3.7. Five independent Multivariate Cox regression analyses of biochemical recurrence with pathological stage, surgical margin status, preoperative PSA, (A) LC, (B) cribriform, (C) HOXD3 methylation, (D) LC vs. HOXD3 methylation (E) LC and/or HOXD3 methylation in GP4 within GS7.

(A)

Hazard ratio 95% CI LRT P-value

Pathological Stage 3.889 1.75-8.67 0.001 Surgical margin Status 3.691 1.68-8.13 0.001 Preoperative PSA 1.796 0.82-3.93 0.143 LC

2.562 1.03-6.37 0.043

(B)

Hazard ratio 95% CI LRT P-value Pathological Stage 3.999 1.80-8.88 0.001 Surgical margin Status 3.384 1.61-7.12 0.001 Preoperative PSA 1.929 0.92-4.03 0.080 Cribriform

2.439 0.98-6.09 0.056

(C)

Hazard ratio 95% CI LRT P-value Pathological Stage 4.929 2.30-10.57 0.000042 Surgical margin Status 3.498 1.70-7.19 0.001 Preoperative PSA 2.468 1.21-5.02 0.013 HOXD3 High Methylation 3.158 1.40-7.12 0.006 (D)

Hazard ratio 95% CI LRT P-value

Pathological Stage 4.535 1.964-10.471 0.000399 Surgical margin Status 4.246 1.925-9.366 0.000341 Preoperative PSA 1.932 0.869-4.297 0.106 LC 4.750 1.92-11.72 0.001 HOXD3 High Methylation 3.130 1.29-7.59 0.011 (E)

Hazard ratio 95% CI LRT P-value

Pathological Stage 3.673 1.65-8.19 0.001 Surgical margin Status 5.280 2.22-12.55 0.0002 Preoperative PSA 1.747 0.81-3.79 0.158 LC/ HOXD3 High Methylation 5.413 2.20-13.34 0.0002

     

    113

3.5 Discussion

The identification of clinicopathologic entities such IDC and cribriform architecture in PCa is

emerging as a remarkable prognosticator in PCa. Therefore, it is of paramount importance to

ascertain the biological basis of these lesions. In this chapter, DNA methylation has been

investigated for the first time in IDC and cribriform architecture in GP4 carcinoma glands in a

series of 91 RP specimens with GS7 PCa. We have selected to investigate the methylation of

APC, CYP26A1, HOXD3, HOXD8, KLK6, KLK10, RASSF1A, TBX15 and TGFβ2 genes, which

have also previously shown promise as prognostic biomarkers for PCa with potential for clinical

utility.

Median methylation levels of APC, RASSF1A and TBX15 were significantly linked to IDC

and/or cribriform architecture within the same GP4 area of GS7 RP cases. CYP26A1, HOXD3,

HOXD8, KLK6 and KLK10 methylation was not associated with GP4 cribriform architecture or

IDC. APC was significantly hypermethylated in association with the presence of cribriform,

including LC and SC, suggesting it may serve as a molecular indicator of the presence of

cribriform morphology in GP4. Further, RASSF1A and TBX15 hypermethylation may assist in

the molecular distinction of LC from SC. Meanwhile TBX15 hypermethylation in the absence of

any of the other hypermethylated genes may potentially serve as a molecular indicator of IDC.

This provides the first evidence for a link between IDC and/or cribriform architecture and DNA

methylation, which warrants further investigation of additional methylated genes in IDC and

various architectural patterns of GP4 PCa.

Notably, the mere presence of cribriform architecture or IDC seemed to associate with

increase in APC, RASSF1A, and TBX15 methylation within the same area (Table 3.5). This

observation may have great consequences for the identification of IDC and/or cribriform

architectural patterns on biopsy, which may miss these morphologies, especially when they

occupy a very small percentage of the prostate. Yet, the DNA hypermethylation signature

     

    114

associated with IDC and/or cribriform may be detected in nearby tissue, in line with the concept

of field effect [449]. However, the association between the abovementioned methylation

markers, IDC and cribriform architecture needs to be further validated on additional larger

cohorts of RP and biopsy specimens. Importantly, larger cohorts may better define the ability of

APC, RASSF1A and TBX15 to distinguish GP4 with and without IDC/cribriform architecture

using ROC curve analysis, which could not be achieved in this study due the limited number of

specimens in the cohort. Future work is also necessary to explore the exciting potential for the

identification of IDC and/or cribriform specific DNA methylation signature in easily accessible

biofluids as it may serve as a surrogate for epigenetic events occurring in these clinicopathologic

entities in the prostate, thus avoiding the need for a biopsy. Given the increased potential of IDC

cells to travel into the prostatic urethra via the prostatic ductal system, we hypothesize that

detection of TBX15 hypermethylation in association with IDC may be possible in post-DRE

urine and should be further investigated.

ERG immunohistochemistry reflective of TMPRSS2-ERG gene fusion was more prevalent in

the presence of IDC/LC in this cohort, consistent with other published studies [474-476]. This

validates the association of ERG with IDC/LC, which may have significant implications for

improving the accuracy of PCa prognosis. The prognostic value of ERG immunohistochemistry

and TMPRSS2-ERG gene fusion is currently controversial in PCa [218]. The discrepancies in the

reported prognostic value of ERG could be explained by stratification of PCa cases by IDC/LC

status.

The hypermethylation of CYP26A1, HOXD3 and TBX15, among other hypermethylated

genes, has also been associated with ERG expression and TMPRSS2-ERG gene fusion [292,

419]. Therefore, the association of APC, CYP26A1, RASSF1A and TBX15 methylation with

IDC/LC stratified by ERG expression status was investigated. IDC/LC was associated with

     

    115

elevated median methylation levels of these genes exclusively in ERG positive GP4. This

demonstrates that the association between IDC/LC and DNA methylation may in part be

dependent on TMPRSS2-ERG status and suggests a functional relationship that requires further

investigation.

Median CYP26A1 methylation, averaged for all GS7 patterns per case, was significantly

elevated in association with the presence of IDC/LC. On the contrary, median CYP26A1

methylation in GP4 of these cases was not associated with the presence of IDC or cribriform

within the same area. The observed association between CYP26A1 hypermethylation in GS7

cases and IDC/LC may be driven by the methylation values in GP3 or alternatively, it may have

been a spurious correlation that did not withstand a more stringent analysis.

Notably, not all of the nine methylation biomarkers investigated were associated with IDC

and/or cribriform architecture, suggesting it may not be a generalized phenomenon. However,

the biological roles of APC, RASSF1A and TBX15 methylation in the formation of these

clinicopathologic entities are not known and need to be elucidated. One hypothesis is that

elevated methylation of APC, RASSF1A and TBX15 may further downregulate the expression of

these genes thus inducing proliferative signaling pathways such as MYC, ERK1/2 and p21 [465,

466, 468, 472]. In the context of these signaling pathways, if the observed APC

hypermethylation in association with cribriform architecture results in gene silencing, it may

promote upregulation of MYC leading to potential stimulation of cribriform proliferation.

Similarly, DNA methylation-mediated silencing of RASSF1A will promote upregulation of

MYC and phosphorylation of ERK1/2 thus activating the downstream signaling cascade, which

may promote LC proliferation. The role of TBX15 in cancer has not yet been characterized. It is

known to be involved in embryonic development processes that include mesoderm formation,

patterning and organogenesis. During development, TBX15 has been shown to interact and

     

    116

functionally cooperate with PAX3 [465]. If such functional cooperation between PAX3 and

TBX15 occurs in cancer, it may affect the expression of many downstream targets such as

CDKN1A (p21) [468]. Thus, DNA methylation-mediated downregulation of TBX15 may

promote cell proliferation, which may explain its association with IDC and cribriform

architecture. However, an alternative to this would be that APC, RASSF1A and TBX15

methylation might be just an epiphenomenon in clinicopathologic entities of PCa. Future work is

necessary to address this issue.

In this cohort of RPs, we demonstrated that IDC and cribriform architecture has a prevalence

of 26% and 67%, respectively, in GP4 of GS7 PCa, similar to that observed in previous

publications [165, 461]. Our results further demonstrated that the presence of cribriform

architecture in GP4 of GS7 carcinoma glands is associated with advanced pathological stage and

shorter time to biochemical recurrence in univariate analysis, consistent with previous studies

[461, 477]. Nevertheless, cribriform architecture was not an independent predictor of disease

recurrence on multivariate analysis, likely due to the limited number of specimens available in

the given cohort. Importantly, LC in GP4 demonstrated a more significant prognostic effect on

biochemical recurrence in univariate and multivariate analysis, in accordance with the

hypothesis that among the various architectural patterns constituting GP4 PCa, the LC variant

conveys the most unfavourable prognosis [170]. Further, the presence of even a small

component of LC (<5%) in GP4 was sufficient to convey this unfavourable prognosis,

suggesting that any amount of LC is clinically relevant. This finding could also imply that the

presence of LC as tertiary or even minor component in GS6 carcinoma on RP should be

investigated as an indicator of aggressive PCa. However, contradictory to other studies, IDC was

not significantly associated with biochemical recurrence in this cohort, most likely due to limited

sample size.

     

    117

On multivariate Cox regression analyses, the presence of LC and HOXD3 HM in GP4, each,

was found to independently associated with shorter time to biochemical recurrence, but not more

significant than pathological stage or surgical margin status. Additionally, in patients with GP4

tumors with LC or HOXD3 HM, the presence of either marker was the best predictor of

biochemical recurrence. One of the limitations of this study is that only three GP4 carcinomas

had both LC and HOXD3 hypermethylation. Thus, the possibility that these markers

complement each other or have an additive effect could not be evaluated. The intriguing results

of this study warrant future analysis of larger cohorts to elucidate the clinical utility and

biological relevance of HOXD3 methylation and LC architecture in GP4 PCa.

In summary, in this chapter, the association of nine DNA methylation marks, ERG expression

status, IDC and cribriform architecture was assessed for the first time in a series of 91 GP4

carcinoma glands in RPs with GS7 PCa. The prevalence and prognostic significance of LC,

similar to previously reported series of GS7 RPs was confirmed. Further, significant increase in

APC, RASSF1A and TBX15 methylation in association with IDC and/or cribriform was

demonstrated, which may be dependent on ERG expression status. In biochemical recurrence

probability analyses, HOXD3 methylation or LC architecture in GP4 PCa was found to have a

potentially higher risk of biochemical recurrence. Future work is necessary to further validate

the clinical utility and functional relevance of DNA methylation events in IDC and cribriform

architecture as well as should it be included routinely in the RP and pathology reports.

     

    118

Chapter 4 Novel Multiplex MethyLight Protocol for Detection of DNA Methylation in

Patient Tissues and Bodily Fluids

Ekaterina Olkhov-Mitsel1,2, Darko Zdravic1,2, Ken Kron3, Theodorus van der Kwast2,4, Neil Fleshner5 and Bharati Bapat1,2,4

1Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada; 2Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON,

Canada; 3Ontario Cancer Institute, Princess Margaret Cancer Center-University Health Network, Toronto, ON, Canada, 4Department of Pathology and Laboratory Medicine, University

Health Network, Toronto, Ontario, Canada; 5Department of Urology, University Health Network, Toronto, ON, Canada.

The data presented in this chapter has been previously published in a manuscript entitled “Novel Multiplex MethyLight Protocol for Detection of DNA Methylation in Patient Tissues and Bodily Fluids” Olkhov-Mitsel E., Zdravic D., Kron K., Van der Kwast T., Fleshner N.E., and Bapat B. Sci Rep. 2014, 4; 4432. The work has been primarily contributed by Olkhov-Mitsel E. Zdravic D. and Kron, K., former Ph.D. students, contributed to experimental design, analysis and interpretation of data. Van der Kwast T., a pathologist, assisted in specimen retrival and histopathological confirmation. Fleshner N.E. identified and recruited suitable patients who provided urine samples for the study. Bapat B. supervised the project (including project development, experimental design, analysis and interpretation of data) and critically reviewed the manuscript and thesis.

     

    119

Chapter 4 Novel Multiplex MethyLight Protocol for Detection of DNA Methylation in Patient Tissues

and Bodily Fluids

4.1 Summary  

Aberrant DNA methylation is a hallmark of cancer and is an important potential biomarker.

Particularly, as shown previously in this thesis, combined analysis of a panel of hypermethylated

genes shows the most promising clinical performance. Herein, a multiplex MethyLight assay is

developed, optimized and standardized to simultaneously detect hypermethylation of APC,

HOXD3 and TGFβ2 in DNA extracted from PCa cell lines, archival tissue specimens, and urine

samples. I established that the assay is capable of discriminating between fully methylated and

unmethylated alleles with 100% analytical specificity and demonstrated the assay to be as highly

accurate and reproducible as the singleplex approach. For proof of principle, I analyzed the

methylation status of these genes in tissue and urine samples of PCa patients as well as PCa-free

controls. These data show that the multiplex MethyLight assay offers a significant advantage

when working with limited quantities of DNA and has potential applications in research and

clinical settings.

4.2 Introduction

Assessment of tumor-specific DNA methylation changes has a broad range of applications

from providing insights into disease pathogenesis to biomarker discovery. Accordingly,

differential methylation patterns of selected candidate genes have been shown to serve as

promising biomarkers for early diagnosis, prognosis, disease monitoring, prediction of response

to therapy, and assessment of risk of recurrence [478-483]. Importantly, assessment of a panel of

such biomarkers dramatically improves sensitivity and specificity as compared to any single

marker [484]. Therefore, analysis of multiple DNA methylation-based biomarkers is becoming

increasingly important in translational research, as demonstrated in chapter 3 of this thesis. DNA

methylation events have several advantages with respect to their use as cancer biomarkers since

     

    120

they can be tissue- and tumor-type specific [485]. DNA is stable and easy to isolate from various

biological sources and methylation analysis can be performed on a diverse array of specimens

ranging from fresh, frozen, or formalin fixed tissues to different types of biofluids including

plasma, serum, urine or saliva samples [416]. However, despite their promise, DNA methylation

measurements have not come into widespread use in the clinic [486]. There are a number of

potential reasons for this. One significant technological challenge is the sensitive detection of

specific DNA methylation patterns occurring at low abundance and/or availability of limited

quantities of clinical samples. Methylation specific PCR-based techniques are among the easiest

and quickest methods for DNA methylation analysis that have the potential to be used in

everyday laboratory practice for screening of a large number of samples [487]. One such

technique, MethyLight, is based on sensitive and quantitative, methylation-specific, fluorescence

based qPCR assay [424, 488]. MethyLight assays quantify DNA methylation at a particular

locus by using DNA oligonucleotides that anneal differentially to bisulfite-converted DNA

according to their methylation status in the original genomic DNA. MethyLight is broadly

applicable for analysis of large series of samples. Although highly sensitive for the detection of

methylation signal, the conventional MethyLight assay can only analyze one gene at a time,

which can be potentially limiting when attempting to analyze multiple markers using particularly

small quantities of DNA obtained from sources such as needle biopsy, saliva, serum, fine needle

aspirates or urine.

In this chapter, the development of a multiplex MethyLight assay that allows for the co-

amplification of multiple genes in one reaction mixture is described. A panel of three well-

studied methylation marks from chapter 3 was selected, namely, APC, HOXD3 and TGFβ2 [291,

421, 422]. The limits of the technique were first established by measuring its analytical

sensitivity, specificity, accuracy and reproducibility of detection of methylated vs. unmethylated

     

    121

DNA. Subsequently, the methylation status of these genes was analyzed in PCa cell lines,

archival FFPE PCa tissue, fresh frozen benign prostatic tissue, and urine samples from PCa

patients as well as PCa-free controls, in order to test the feasibility and applicability of this

technique.

4.3 Materials and methods

4.3.1 Patient Samples and Cell Lines

RP FFPE tissue samples reported in this chapter were a subgroup of 10 samples in cohort I,

described in chapter 2, section 2.3.1 (page 57). FFPE tissue blocks fixation, storage and

processing prior to sectioning is unknown. The 10 samples were selected based on high KLK10

methylation status and were expected to harbor methylation of other genes, thus serving as good

test samples for our proof-of-principle experiment. Another series of five cystoprostatectomy

(CP) fresh frozen prostatic tissue samples was collected from UHN, reviewed by a pathologist

(TVDK) and confirmed as benign. Additionally, post-DRE urine specimens of 10 unrelated

patients diagnosed with organ-confined, intermediate risk PCa at UHN and examined by the

attending uro-oncologist (NF) as well as five voided urine samples from PCa-free controls were

obtained from the UHN Genito-Urinary Prostate Clinic. Patient consent was obtained for accrual

of urine and surgically excised tissue following RP or CP at the UHN tissue bank, and according

to the approved protocols of the REB at UHN and Mount Sinai Hospital in Toronto. The human

prostate cancer cell lines LNCaP (ATCC# CRL-1740) and PC-3 (ATCC# 59500) were obtained

from Drs. R. Bristow and E. Diamandis at the University of Toronto.

4.3.2 DNA Extraction and Sodium Bisulfite Modification

Cell pellets were isolated from 35–70 mL of whole urine by centrifugation (2000 g, 10

minutes) and DNA was extracted using Qiagen QIAampDNA Microkit (Qiagen) according to

the manufacturer’s protocol. Meanwhile, the QIAamp DNA Mini Kit was used to extract DNA

     

    122

from cell lines and FFPE tissue (following macrodissection) as described in section 2.3.5 (page

60). The concentration and quality of DNA samples was assessed using a NanoDrop 1000

spectrophotometer (Thermo Scientific). A260/280 ratio ranged from 1.8 to 2.2 and A260/230 ranged

from 1.8 to 2.1 for the DNA samples used in this chapter. DNA extracted from cell lines, FFPE

tissue and urine samples (100–400 ng genomic DNA for each) was bisulfite modified using the

Zymogen EZ DNA Methylation Gold Kit (Zymo Research) as per manufacturer’s protocol and

was eluted into a final concentration ranging from 10 ng/uL - 20 ng/uL. In each PCR reaction 1

ul of bisulfite modified DNA was used. For experiments involving serial dilutions, the

concentration of CpGenomeTM universal methylated DNA (Chemicon, Millipore) and peripheral

blood cell (PBC) DNA was measured by NanoDrop 1000 (Thermo Scientific) prior to bisulfite

modification and diluted to a final concentration of 5 ng/uL. Following bisulfite conversion, the

DNA concentration was again verified by NanoDrop. Next, based on these measurements, 70 ng

of bisulfite modified CpGenomeTM universal methylated DNA and 140 ng bisulfite modified

PBC DNA were mixed at a concentration of 5 ng/uL and serially diluted in 3-fold increments

from 153 (1.67 ng/uL) up to 1519683 (0.0003 ng/uL). In each PCR reaction 2 ul of each dilution

were run in triplicate.

4.3.3 MethyLight Assays

Singleplex MethyLight assays for each gene were performed by incubating the bisulfite

converted DNA as described in chapter 2, section 2.3.5 (page 60). For the multiplex MethyLight

assay, bisulfite converted DNA was incubated with 400 uM dNTPs, 10.5 mMMgCl2, 0.01%

Tween-20, 0.05% gelatin, 1.0 units of Taq polymerase and varied concentrations of each primer

and TaqMan hydrolysis probe set as follows: 8.0 uM APC, HOXD3 and TGFβ2 forward and

reverse primers each, 1.0 uM ALU forward and reverse primers each, 2.66 uM APC and TGFβ2,

2.0 uM HOXD3 and 0.1 uM ALU probe each, in a 30 ul reaction volume on a 96-well plate

     

    123

sealed with MicroAmpTM Optical adhesive film (Life Technologies). The ratio between the

values of APC, HOXD3 or TGFβ2 and ALU was used as a measure of PMR according to the

formula of Eads et al. [424] as follows:

[(gene/ALU)sample/(gene/ALU) CpGenomeTM universal methylated DNA]x100%

APC, HOXD3, TGFβ2 and ALU primers and TaqMan hydrolysis probe sequences used for

multiplex MethyLight are listed in Table 3.1. Oligonucleotides for unmethylated reactions were

identical, except CG was replaced with TG and CA in the forward and reverse sequences,

respectively. All oligonucleotide sequences were ordered from Intergrated DNA Technologies

(IDT) and analyzed in silico for potential non-specific interactions using the IDT heterodimer

oligoanalyzer algorithm: http://www.idtdna.com/analyzer/applications/oligoanalyzer. All PCR

reactions were performed on an Applied Biosystems 7500 qPCR system (Life Technologies)

with an initial denaturation at 95°C for 10 minutes, followed by 50 cycles of 95°C for 15

seconds and 60°C for 1 minute. Data acquisition and analysis was performed using the SDS

2.0.6. In each run, a no-template control reaction was included that was always reliably negative

(undetermined Ct value, also known as quantification cycle or Cq value). CpGenomeTM

universal methylated DNA (Chemicon, Millipore) was also integrated in each run as a positive

control as well as standard curve material. Standard curves were prepared using seven three-fold

dilutions starting from a concentration of 40 ng/uL. APC, HOXD3 and TGFβ2 standard curve R2

was ≥0.95 while ALU standard curve R2 was ≥0.99. Multiplex ALU reactions were considered

positive when sufficient input DNA was present, which was defined as ALU Ct value ≥19.

Above this cutoff, not detecting methylated DNA of either assay could be either due to a lack of

methylation or due to insufficient quality/quantity of template DNA. The selection of this Ct

cutoff was based on technical parameters determined during multiplex MethyLight assay

optimization and standardization. Briefly, ALU Ct value of 19 corresponded to the lowest DNA

     

    124

concentration on the standard curve that reliably detected APC, HOXD3 and TGFβ2 methylation

in the same reaction. Intra-assay and inter-assay coefficient of variation (CV) and differences

between singleplex and multiplex MethyLight PMR values were determined using paired t-test

on Microsoft® Office Excel for Mac 2011.

4.4 Results

4.4.1 Parameters Important for Multiplex MethyLight Assay Design

Important parameters for the successful development of multiplex MethyLight are primer

design, fluorescent dye selection and optimization of reaction conditions. For the initial

development of a multiplex MethyLight assay, MethyLight primers and TaqMan hydrolysis

probe sets used for DNA methylation analysis of APC, HOXD3 and TGFβ2 genes were chosen

as described in previous studies [291, 421, 422]. For methylation-independent MethyLight

control reaction, ALU sequences were selected to measure the amount of input DNA. ALU

repetitive elements represent a significant portion of the human genome hence they are less

susceptible to normalization errors caused by cancer associated aneuploidy and copy number

changes [425]. In silico analysis of these primer and probe sequences revealed minimal

unfavorable interactions, defined as minimal propensity to form dimers based on sequence

homology and delta Gibbs free energy, allowing them to be multiplexed. The ABI 7500 qPCR

System, reported in this thesis, has five channels for target detection; one of these (the ROX dye,

emission maximum at 605 nm) is required in our master mix to serve as a passive reference.

Thus, four fluorescent dyes were selected to label four different gene probes which have

sufficient spectral separation to avoid overlap between signals of different targets: HEX (554

nm) for APC, CY5 (650 nm) for HOXD3, TAMRA (583 nm) for TGFβ2, and FAM (520 nm) for

ALU. Next, the multiplex MethyLight reaction conditions were optimized by running a set of

qPCR reactions with a range of concentrations of dNTPs (200–600 uM), MgCl2 (3.5–10.5 mM),

     

    125

Taq polymerase enzyme (0.5–1.0 units), primers (0.5–8.0 uM) and probes (0.1–4.0 uM). Thus,

the optimal multiplex MethyLight reaction conditions were established at 400 uM dNTPs, 10.5

mM MgCl2, 1.0 units of Taq polymerase enzyme, primer concentrations of 8.0 uM for APC,

HOXD3 and TGFβ2, and 1.0 uM for ALU, probe concentrations of 2.66 uM for APC and

TGFβ2, 2.0 uM for HOXD3 and 0.1 uM for ALU (data not shown).

4.4.2 Analytical Sensitivity and Specificity of the Multiplex MethyLight Assay

Serial dilution experiments were performed to determine the analytical sensitivity of

detection of APC, HOXD3 and TGFβ2 methylation in the multiplex MethyLight assay. Seventy

nanograms of bisulfite modified CpGenomeTM universal methylated DNA was serially diluted in

3-fold increments from 153 up to 1519683. The APC, HOXD3 and TGFβ2 methylated reactions

were then used to track the decreasing amount of methylated DNA in the serial dilution series

(Figure 4.1). The APC, HOXD3 and TGFβ2 reactions show decreasing detection of methylated

alleles in each subsequent dilution, as indicated by the increasing cycle number at which each

reaction signal crosses the detection threshold. In a multiplex MethyLight setting, the analytical

sensitivity limit of APC, HOXD3 and TGFβ2 methylation was determined to be the 1527

dilution (estimated to contain 0.37 ng of methylated DNA). The ALU control reaction analytical

sensitivity of detection was established at the 156561 dilution (0.0016 ng). Similarly, bisulfite

modified CpGenomeTM was serially diluted in human peripheral blood cell (PBC) DNA, which

is expected to harbor minimal methylation of the selected genes. The PBC DNA and all of the

dilution reactions had overlapping ALU amplification curves, indicating they contained equal

amounts of DNA (10 ng). PBC DNA displayed minimal methylation of TGFβ2 (<1 PMR). In

this setting, the analytical sensitivity limit of APC, HOXD3 and TGFβ2 methylation was also the

1527 dilution (0.37 ng of methylated DNA and 9.63 ng of PBC DNA, respectively). Next, to

assess the analytical specificity of detection of the APC, HOXD3 and TGFβ2 multiplex

     

    126

MethyLight assay, separate reactions using either the methylated or unmethylated primer-probe

sets of each gene were performed on both PBC DNA and CpGenomeTM universal methylated

DNA (expected to show robust methylation, Figure 4.2). Consistent with its unmethylated status,

PBC DNA yielded an amplification signal for multiplex MethyLight reactions with the

unmethylated primers, while there was no detectable signal for the methylated reaction. In

contrast, CpGenomeTM universal methylated DNA yielded a robust amplification signal in the

methylated, but not in the unmethylated reactions for all three genes. Both the PBC and

CpGenomeTM universal methylated DNA were positive for their ALU reactions, indicating that

the minimal required amount of input DNA was present in the multiplex assay. Therefore,

multiplex MethyLight technology can efficiently discriminate between fully methylated vs.

unmethylated alleles with 100% specificity.

     

    127

Figure 4.1. Analysis of the analytical sensitivity of the multiplex MethyLight technique for (A) APC, (B) HOXD3, (C) TGFB2. CpGenome™ universal methylated DNA was serially diluted in 3-fold increments from 1:3 to 1:19683 and subsequent multiplex analysis was performed. The resulting relative fluorescence (ΔRn) was plotted as a function of cycle number. Increasing dilutions are indicated by different colors and shapes, as shown on the right. (D) ALU control reaction was included to determine the detection of input DNA in each dilution. The black horizontal line indicates the threshold used for calculating the amount of template DNA. A positive signal was noted for a specific dilution when the corresponding amplification curve had passed the threshold line.

     

    128

Figure 4.2. Analysis of the analytical specificity of APC, HOXD3 and TGFB2 multiplex MethyLight assay. Separate reactions using either the methylated or unmethylated primers were performed on both human peripheral blood cell (PBC) DNA and CpGenome™ universal methylated DNA. The resulting relative reaction fluorescence (ΔRn) was plotted as a function of cycle number. The (A) methylated or (B) unmethylated reactions were 100% specific because the methylated primers did not cross-react with PBC DNA (0% methylated) and the methylated primers did not cross-react with CpGenome™ universal methylated DNA (100% methylated). Both CpGenome™ universal methylated DNA and PBC were positive for their ALU reactions, indicating that there was sufficient input DNA in each sample.

     

    129

4.4.3 Accuracy and Reproducibility of the Multiplex MethyLight Assay

To test the accuracy of the APC, HOXD3 and TGFβ2 reactions in the multiplex MethyLight

setting, a series of conventional singleplex MethyLight assays was first performed to quantify

methylation levels of each gene separately in the human PCa derived cell lines LNCaP and PC-

3. Multiplex MethyLight assays were then performed to analyze the methylation levels of APC,

HOXD3 and TGFβ2 in the same cell lines. As shown in Figure 4.3, there were no significant

differences in PMR values between singleplex and multiplex MethyLight assays (mean fold

change=1.3, paired t-test P-value > 0.05) for each of the three genes in either of the two cell

lines. Additionally, to test the reproducibility of the multiplex MethyLight assay, 2–6 additional

independent runs of the assay were performed in the same PCa cell lines. The mean PMR values

as well as inter-assay and intra-assay PMR CV are shown in Figure 4.4. The inter-assay and

intra-assay PMR CV values, ranging from 0.044 to 0.138 and 0.014 to 0.110 respectively, are

indicative of only a modest discordance in PMR values between independent multiplex

MethyLight runs, suggesting that the technique generates highly reproducible results. In PC-3

cells only, a CV value of 1.569 is observed for TGFβ2 due to high variance in the methylation of

this gene in this cell line. Although TGFβ2 methylation in PC-3 cell line shows high variance

and PMR CV value, it is not statistically significant between singleplex and multiplex assays by

Student’s t-test.

     

    130

Figure 4.3. Representative analysis of the accuracy of multiplex MethyLight assay through the comparison of PMR values for APC, HOXD3 and TGFB2 genes in DNA from human prostate cancer cell lines LNCaP and PC-3 to singleplex MethyLight PMR values. The PMR for each gene was calculated from an average of 3-7 independent assays with duplicate reactions per gene in each assay. Paired t-test analysis was performed to assess differences in PMR values between singleplex and multiplex MethyLight assays for each of the three genes in either cell line (paired t-test P-value>0.05).

     

    131

Figure 4.4. Analysis of the reproducibility of the multiplex Methylight assay reactions for APC, HOXD3, and TGFB2. Three independent multiplex MethyLight assays were performed in duplicate reactions for the prostate cancer cell lines LNCaP and PC-3. The intra-assay coefficient of variation (CV) was calculated as the average for the duplicate reactions in each independent multiplex assay. The inter-assay CV was calculated between the 3-7 independent multiplex assays. Although TGFB2 methylation in the PC-3 cell line shows high CV value, it is not statistically significant between singleplex and multiples assays by student’s t-test.

     

    132

4.4.4 Application of the Multiplex MethyLight Assay to Patient Samples

The main potential application of the multiplex MethyLight technology is aimed at rapid

DNA methylation analysis of multiple loci in clinical samples with limited DNA quantities. A

number of studies have shown that APC, HOXD3 and TGFβ2 are aberrantly methylated in PCa

and each is associated with disease progression and biochemical recurrence [421, 422]. Further,

in previous studies, our lab has shown that when combined, the methylation status of these genes

outperformed any single marker for the prediction of disease progression and biochemical

recurrence [291]. Therefore, as a proof of principle experiment, the multiplex MethyLight

method was tested by measuring the methylation levels of APC, HOXD3 and TGFβ2 in 10 FFPE

PCa tissues, five fresh frozen benign prostatic tissues, 10 urine samples obtained from unrelated

PCa patients, and five urine samples from PCa-free controls. As seen in Table 4.1, PMR values

above zero were detected for at least one gene in all but one of the tested tissues and all urine

samples from PCa patients. Two of the urine samples from PCa-free controls had PMR values

above zero for the APC gene (PMR 50.2 and 0.3 for samples 4 and 5, respectively). All 30 DNA

samples gave rise to a positive signal in their ALU reactions (Ct<19), confirming that the

minimal required amount of input DNA was present in each reaction to perform the multiplex

assay.

     

    133

Table 4.1. Multiplex MethyLight data (PMR) for APC, HOXD3 and TGFB2 in FFPE prostate cancer tissues, post-DRE urine samples of PCa patients as well as fresh frozen tissues and urine samples from PCa-free controls

Average PMR

Prostate Cancer Tissues APC HOXD3 TGFB2 1

39.7 4.1 10.0

2

0.0 0.0 1.1 3

63.4 29.1 22.1

4

44.0 15.9 10.8 5

35.0 7.8 13.1

6

11.9 9.8 3.3 7

34.3 13.3 13.9

8

42.3 1.8 14.0 9

13.2 10.7 12.7

10

16.7 16.9 3.4 Benign Prostatic Tissues 1 0.0 0.0 0.0 2 0.2 7.9 2.5 3 0.6 17.0 1.2 4 5.2 12.2 2.4 5 0.9 12.7 1.3 Prostate Cancer Patients Urine 1

4.8 1.6 0.0

2

0.0 12.9 0.0 3

1.6 7.6 0.8

4

0.0 3.9 0.0 5

0.6 4.5 0.0

6

0.8 0.2 0.0 7

0.0 1.3 0.0

8

0.0 0.8 0.6 9

0.0 0.4 0.0

10 0.1 0.3 0.1 Prostate Cancer-free Controls Urine 1 0.0 0.0 0.0 2 0.0 0.0 0.0 3 0.0 0.0 0.0 4 0.3 0.0 0.0 5 0.2 0.0 0.0

     

    134

4.5 Discussion

The chapter reports the development of a multiplex MethyLight assay for the simultaneous

analysis of APC, HOXD3, TGFΒ2 methylation markers and the reference marker ALU in

agreement with the minimum information for publication of quantitative real-time PCR

experiments (MIQE) guidelines [489]. This method has demonstrated high analytical specificity,

accuracy and reproducibility in the quantitative detection of methylated vs. unmethylated alleles.

Further, the applicability of this approach to rapidly detect DNA methylation in patients’ fresh

frozen and FFPE tissues as well as urine samples was demonstrated. Incorporating a multiplex

protocol for MethyLight assay makes the technique better suited for use with precious clinical

samples, is more economical and enhances throughput, thereby increasing the overall efficiency

of the methylation detection experiments. Multiplex MethyLight requires smaller amounts of

template DNA compared to singleplex MethyLight because it can be used to analyze multiple

loci concomitantly in one assay. This is especially beneficial for analyzing clinical samples with

limited DNA quantities such as biopsy material. The Multiplex protocol is also more

economical, as it allows for smaller amounts of reagents (i.e. oligonucleotides, MgCl2, enzyme)

to be used in each multiplex assay when compared to the singleplex approach. A standard

multiplex run requires only one standard curve for quantification of four genes. This decreases

the amount of CpGenomeTM universal methylated DNA used and allows for the simultaneous

analysis of three DNA methylation markers in up to 41 samples in duplicate in a 96-well plate

format in each assay, thus saving time and increasing efficiency. Some limitations of the assay

should be noted. The multiplex MethyLight assay reported in this chapter was performed on

ABI 7500 qPCR system, which supports multiplexing of up to five fluorescent dyes. A greater

number of methylation marks could be potentially incorporated into the multiplex MethyLight

assay when used with other platforms such as the QuantStudioTM 12 K Flex qPCR System

(Applied Biosystems), which supports multiplexing of up to six fluorogenic probes. Another

     

    135

factor that may limit the number of markers that can be multiplexed in a MethyLight assay is

primer and TaqMan hydrolysis probe design. MethyLight technology is based on the detection

of methylation-dependent sequence differences in DNA following bisulfite modification, which

leads to a significant reduction in DNA complexity. This makes the design of highly specific

primers and probes challenging and requires special attention to be paid to parameters such as

homology of primers and probes with their target sequences, oligonucleotides length, CG

content, and concentration. Lastly, the analytical sensitivity of the multiplex MethyLight assay

developed in this study is limited, yet sufficient for the detection of APC, HOXD3 and TGFβ2

methylation in FFPE tissue and urine samples. Further, we have shown that in a multiplex

setting the methylation of APC, HOXD3 and TGFβ2 could be detected in the presence of large

excess of unmethylated DNA. This further suggests that this approach can be used to detect

aberrant methylation patterns in heterogeneous clinical samples with substantial unmethylated

DNA contamination. It is important to note that the presence of sufficient DNA amounts was

evaluated by the ALU reactions. However, this only indicates that the minimal required amount

of input DNA was present in the multiplex assay, not that sufficient cancer DNA is present in

the sample. The analytical sensitivity of detection for this assay could be potentially further

improved by incorporating Locked Nucleic Acids (LNA) in primers and/or probes [490, 491].

LNA probes and primers are nucleic acid analogues that contain a bicyclic furanose ring in the

ribose sugar which is chemically locked in an RNA-mimicking conformation. This chemistry

allows for increased analytical specificity and improved detection limit for qPCR assays.

Different multiplex approaches have been developed to evaluate DNA methylation in various

sources of patient DNA samples. Some of these include multiplex ligation-dependent probe

amplification (MLPA), bisulfite assisted genomic sequencing PCR (BSP), loop mediated

isothermal amplification (MS-LAMP), nested MSP, fluorescence resonance energy transfer

     

    136

(FRET), padlock probes (PP), and multi-component nucleic acid enzymes (MNAzymes)[492-

501]. However, all these multiplex DNA methylation analysis methods have several

disadvantages compared to MethyLight technology. First, approaches such as BSP, PP and

nested MSP are based on gel electrophoresis, which is generally less sensitive than qPCR. These

approaches also require relatively greater quantities of template DNA. Second, probe

development for approaches such as MLPA and MS-LAMP is much more complicated, costly

and time consuming. Third, most of the above-mentioned approaches are multi-step procedures

that cannot be performed in a single tube increasing the risk of contamination. Lastly, novel

approaches such as FRET and MNAzymes are based on chemistries that are not as well

established and need to be further validated. Other approaches based on methylation specific

fluorescent probe detection by qPCR have been previously reported [502-505]. However, the

control reactions utilized in these approaches were ACTB and not ALU based, which is a less

stable and/or less reproducible measure of bisulfite-converted input DNA, especially when

analyzing tumor samples with local amplifications or deletions of single genes. Multiplex digital

MethyLight has also been utilized for DNA methylation analysis [506]. Digital MethyLight is

shown to have greater analytical sensitivity and accuracy than the singleplex MethyLight assay

in detecting individual methylated DNA molecules in clinical samples [425]. Therefore, Digital

MethyLight technology can potentially be applied to the APC, HOXD3, TGFβ2 multiplex

MethyLight assay in the future. This is the first study to develop a multiplex MethyLight assay

for the analysis of the three-gene panel consisting of APC, HOXD3 and TGFβ2 genes, which

have previously shown promise as prognostic biomarkers for PCa. Our multiplex MethyLight

assay was able to successfully determine methylation levels of these three genes in fresh frozen

and FFPE tissue as well as urine samples obtained from PCa patients and PCa-free individuals.

This proof of principal cohort included 20 PCa samples and 10 healthy controls, thus is only

     

    137

adequate to demonstrate the feasibility of the multiplex assay. To ensure high specificity of this

multiplex assay in future studies, an individual PMR threshold value will need to be set for each

individual gene that will allow for clear distinction between PCa and normal samples, as has

been done for other multiplex methylation assays [503]. Taking an approach similar to that used

in this chapter, multiplex MethyLight assays can be further developed for other promising DNA

methylation biomarker panels. One such example is the APC, MGMT, RASSF2A and Wif-1

biomarker panel, which has shown high diagnostic sensitivity and specificity in colon cancer

[483, 507]. Another example is the APC, RASSF1A and TBX15 methylation panel, which

demonstrated association with IDC and cribriform architecture of aggressive PCa in chapter 3 of

this thesis.

Future studies based on this multiplex MethyLight assay will focus on analyzing the clinical

utility of APC, HOXD3 and TGFβ2 as methylation biomarkers, as well as expanding the number

of candidate markers and/or different combinations of methylation gene markers that can be

multiplexed in the MethyLight assay, and increasing the number of samples that can be analyzed

employing a single standard curve by using 384- or 1536- well plates. Additionally, different

types of clinical samples such as needle biopsies, oral lavage, fine needle aspirates and serum

samples will be tested.

     

    138

Chapter 5 Assessment of Candidate MicroRNAs Targeting KLKs and their Association

with Prostate Cancer Progression Ekaterina Olkhov-Mitsel1,2, Theodorus Van der Kwast2,3, Neil E. Fleshner4, Eleftherios P.

Diamandis1,2,5,6, Alexandre R. Zlotta7, and Bharati Bapat1,2,3

1 Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, Ontario, Canada; 2

Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada; 3 Department of Pathology, University Health Network, University of Toronto,

Toronto, Ontario, Canada; 4 Department of Surgical Oncology, Division of Urology, University Health Network, University of Toronto, Toronto, Ontario, Canada; 5 Department of Pathology and Laboratory Medicine, Mount Sinai Hospital, Toronto, Ontario, Canada; 6 Department of Clinical Biochemistry, University Health Network, Toronto, Ontario, Canada; 7 Division of Urology, Princess Margaret Hospital and Mount Sinai Hospital, Toronto, Ontario, Canada.

The work presented in this chapter has been primarily contributed by Olkhov-Mitsel E. Van der Kwast T., a pathologist, assisted in specimen retrieval, histopathological confirmation and reading of ERG immunohistochemistry. Fleshner N. and Zlotta A. assisted with sample retrieval. Bapat B. supervised the project (including project development, experimental design, analysis and interpretation of data) and critically reviewed the manuscript and thesis. A manuscript based on the data presented in this chapter is currently in preparation.

     

    139

Chapter 5 Assessment of Candidate MicroRNAs Targeting KLKs and their Association with Prostate

Cancer Progression  

5.1 Summary  

PCa is characterized by aberrant expression of miRNAs and a recurrent TMPRSS2-ERG

fusion. However, the utility of these molecular alterations as PCa prognosticators is not yet well

established. Herein, the expression of 84 miRNAs was analyzed in 36 PCa FFPE RPs.

Upregulation of 56 of these miRNAs correlated with GS but none were differentially expressed

in association with biochemical recurrence. Among these, the expression of miR-137 and miR-

21 was analyzed for the first time in association with ERG expression status, and

clinicopathological features of PCa in cohort I. High miR-137 expression was associated with

GS and positive ERG expression (χ2 P-value =0.026 and 0.017, respectively), whereas no

significant association was observed for miR-21 expression (χ2 P-value=0.66 and 0.59,

respectively). No significant associations were observed between miRNA expression and

pathological stage, age, preoperative PSA or prostate weight. ERG protein expression also did

not correlate with clinicopathological parameters in this cohort. Correlation analysis controlling

for ERG revealed that DNA methylation of KLK10 was inversely correlated to the expression of

miR-137, a miRNA predicted to target it. This is the first study to correlate DNA methylation

and miRNA expression involved in the regulation of the KLK10 gene. The results suggest

tumors with different GS appear to have unique miRNA expression patterns. Further, the

correlation of miR-137 expression with GS suggests future studies could evaluate its

independent prognostic value in PCa. Lack of differential expression of miR-21 and miR-137

between PCa and adjacent normal tissue warrants further investigation into miRNA expression

as part of the epigenetic field effect. The association between miR-137 and ERG expression

status provides a new potential link between miRNAs and TMPRSS2-ERG gene fusion.

     

    140

5.2 Introduction  

miRNAs are a class of small, single stranded, non-coding RNAs that post-transcriptionally

regulate the expression of over 60% of mRNAs in a sequence specific manner [508, 509]. They

have been shown to regulate multiple protein-coding genes involved in divergent cellular

functions including development, differentiation, proliferation, and apoptosis [510-513].

Emerging evidence has implicated various miRNAs in the development and progression of

multiple human cancers, including PCa [338, 514, 515]. Further, it has been demonstrated that

miRNAs are remarkably stable in FFPE tissue, urine and plasma, making them attractive

potential biomarkers in cancer diagnosis and prognosis [516-518]. Besides potential roles as

biomarkers, miRNAs have been suggested as candidate therapeutic tools [519]. Accordingly,

strategies are being developed to restore the expression of miRNAs to normal physiological

levels in cancer cells. Importantly, this therapeutic approach offers the possibility of targeting

one miRNA molecule to modulate multiple pathways deregulated in tumors.

Recently, numerous PCa miRNA profiling studies have been published, each suggesting a

distinct signature of miRNAs for PCa diagnosis and prognosis [337]. However, the results from

the various studies are inconsistent and there is no clear miRNA expression profile for PCa.

Some groups suggest a widespread downregulation of miRNAs in PCa, while others suggest a

general overexpression of miRNAs in the disease. Most promising miRNAs that have been

specifically investigated and suggested as prognostic biomarkers in PCa are miR-205, miR-221

and miR-145, to name a few [339, 340, 343, 344, 520-522]. However, the AUC achieved by

these miRNA-based biomarkers outlines that there is still need for improvement.

During prostate carcinogenesis, in addition to miRNA deregulation, many other epigenetic

alterations, genetic mutations and chromosomal aberrations occur and accumulate. Among

these, deletion of an interstitial fragment of chromosome 21, resulting in the AR responsive

     

    141

promoter and 5′ end of TMPRSS2 driving the expression of the ERG oncogene is the most

common gene fusion in PCa, present in 40-80% of all tumors [213]. ERG overexpression has

been shown to regulate highly important pathways in PCa including the AR pathway, Wnt/TCF

signal transduction, polycomb group proteins, epithelial-to-mesenchymal transition (EMT) and

cell motility [313, 523-527]. Additionally, tumors positive for the oncogenic fusion protein have

been shown to have dysregulated epigenetic signatures, including altered miRNA expression

[418, 419]. However, conflicting results about the prognostic value of this gene fusion have been

reported [218].

Given the controversial results of large-scale miRNA profiles in PCa, a more focused

analysis of miRNA expression was performed in this chapter using RT-qPCR miRNA arrays to

shed light on miRNAs deregulated in association with aggressive PCa. These arrays allow for

analysis of 84 miRNAs that have been previously implicated in tumorigenesis. Additionally, the

majority (73%) of these miRNAs are predicted to target KLKs which may have potential

biological implications in PCa as KLKs are a promising biomarker family frequently

deregulated in the disease. Taken together with data from chapter 2, this also provides potential

novel avenues for the investigation of different epigenetic regulatory mechanisms that alter the

transcriptional potential of the same KLK gene. Taken together, this approach provides a more

focused analysis on miRNAs whose deregulation might have a pathogenic significance in PCa.

Among the panel of novel and/or differentially expressed miRNAs identified, the prognostic

impact of miR-21 was further characterized in a large cohort of PCa patients. Further, the

expression of miR-137 was investigated for the first time in PCa as a novel candidate biomarker.

The expression of these miRNAs was also correlated to ERG expression status, as it denotes an

etiological subgroup of PCa tumors.

     

    142

5.3 Materials and methods  

5.3.1 RNA extraction

Human prostate FFPE tissue blocks from patients in cohort I and II (described in chapter 2,

section 2.3.1, page 57) were selected to represent GS6 vs. GS8 and early (<3 years) vs. late

(>7.8 years) biochemical recurrence status (Table 5.1). The tissue blocks were then matched to

selected H&E slides reviewed by TDVK as described in section 2.3.1 and sectioned at a

thickness of 10 µm. The tissue slides were then superimposed on the H&E slides and each area

of cancer and/or matched normal tissue was outlined. The circled areas of tissue were then

scraped with a scalpel and tissue was placed into 1.5 ml tubes. Total RNA, including miRNA,

was extracted from tissues using miReasy FFPE Kit (Qiagen) according to the manufacturer’s

instructions. Total RNA was eluted in 20ul of RNase-free water and quantified using NanoDrop

ND-1000 spectrophotometer (Thermo Scientific).

5.3.2 RT-qPCR miRNA Arrays

Expression profiles of miRNAs were quantified in 36 cases from cohort I and II of different

GS and biochemical recurrence status (Table 5.1) using a miRNA RT-qPCR array platform

(Human Cancer Pathway Finder miScript miRNA qPCR array, MIHS-102Z, Qiagen), as per the

manufacturer's protocol. Briefly, miRNAs were converted into cDNA using miScript II Reverse

Transcription Kit (Qiagen). RT-qPCR was performed using miScript SYBR Green PCR Kit

(Qiagen) and the Applied Biosystems 7500 real-time thermocycler in 96-well PCR microplates

(Life Technologies). The Human Cancer Pathway Finder miScript miRNA PCR array contained

lyophilized miScript primer assays for 84 miRNAs plus six normalization controls (SNORD61,

SNORD68, SNORD72, SNORD95, SNORD96A and RNU6-2) as well as controls to assess

RNA recovery, reverse transcription performance, and qPCR performance. Data were quantified

     

    143

using SDS version 2.0.6 (ABI, Life Technologies) and analyzed using the miScript miRNA PCR

Array Data Analysis tool (http://pcrdataanalysis.sabiosciences.com/ mirna/arrayanalysis.php).

5.3.3 RT-qPCR

A total of 10ng of RNA per sample was converted to cDNA using Taqman® miRNA reverse

transcription kit (Applied Biosystems by Life Technologies). RT-qPCR was then performed in

duplicate to measure the levels of mature hsa-miR-137, hsa-miR-21 and U6 snRNA using

individual TaqMan® microRNA Assays (Applied Biosystems, Life Technologies) according to

the manufacturer’s instructions. RT-qPCR was performed using Taqman® universal mastermix

II no UNG on Applied Biosystems 7500 real-time thermocycler in 96-well PCR microplates

(Life Technologies). Data was analyzed on SDS version 2.0.6. Relative miRNA expression

levels were normalized against endogenous control U6 snRNA using the formula 2-ΔCt, where

ΔCt was equal to the difference between the threshold cycle of the target miRNA and that of U6

snRNA in each replicate of each sample. U6 snRNA was selected for normalization purposes

owing to its broad use as a reference molecule for miRNA expression studies in PCa and its

reported biological stability [528, 529].

5.3.4 MicroRNA Target Gene Prediction

Putative miRNAs predicted to target the 3’UTR of KLKs were identified utilizing the

algorithms in the microRNA Data Integration Portal (mirDIP; http://ophid.utoronto.ca/mirDIP/);

a publicly available data portal integrating in silico miRNA target predictions from seven

databases (RNA22, picTar, microRNA.org, miRBase, PITA, DIANA-microT and TargetScan).

Only miRNA-KLK interactions predicted by two or more databases were considered for

analysis.

     

    144

5.3.5 Statistical Analyses

Differential miRNA expression was determined using the Bayesian adjusted t-statistic from

the miScript miRNA PCR Array Data Analysis tool. Differentially expressed miRNAs were

identified using fold-change threshold of 2 and P-values of ≤0.05. Interquartile range analysis

identified 4 samples as outliers. Therefore, they were removed from analysis. Hierarchical

clustering of miRNA 2-ΔCt expression values was performed using GenePattern software (Broad

Institute, Cambridge, CA).

For miR-21 and miR-137 expression analysis, each amplification reaction was performed in

duplicate and the mean 2-ΔCt was used for statistical analysis. The continuous 2-ΔCt variable was

separated into high expression and low expression based on a third quartile threshold. Pearson χ2

tests were used to analyze proportional differences between ERG and miRNA expression and

between each category of GS, pathological stage and surgical margin status. Mann–Whitney U-

test was used to assess differences in age, preoperative PSA, and prostate weight between ERG

positive and negative cases as well as between miRNA high expression and low expression

cases. Partial correlation and χ2 tests were used to assess the association between miRNA

expression and KLK10 DNA methylation. For all described methods, two-sided P-values of

≤0.05 were considered significant. All tests were conducted with SPSS version 21 software

(SPSS, IBM Software).

     

    145

Table 5.1. Clinical characteristics of prostate cancer cases used for miRNA expression profiling

Clinical characteristic Gleason Score No. of patients (%)

6

24 (67) 8

12 (33)

Pathological stage pT2

22 (61)

pT3

13 (36) pT4

1 (3)

Age Average

61.4

Median

62.9 Range

47-72

Surgical Margins Positive

7 (19)

Negative

29 (81) Prostate Weight

Average

53.6a Range

15-112

Biochemical Recurrence Recurrences

18 (50)

PSA Average

9.1b

Range

3.8-35

Total 36 a. Available for 34 of 36 cases b. Available for 35 of 36 samples

 

 

     

    146

5.4 Results  

5.4.1 MicroRNA Expression Profiling

RT-qPCR array analysis of 84 human miRNAs was performed in 36 samples of

macrodissected PCa tissues selected to reflect subgroups of different GS and biochemical

recurrence status (Table 5.1). Unsupervised hierarchical clustering was performed on 32 of 36

cases to cluster the samples based on their miRNA expression patterns. Interquartile range

analysis identified 4 samples as outliers, therefore they were removed from analysis. PCa

samples were distinctly separated into two groups, cluster A (n=19) and cluster B (n=13, Figure

5.1). Cluster A consisted of a significantly higher percentage (53%) of GS8 cases compared to

Cluster B (8%) (χ2 P-value = 0.009). There were no other significant differences between the

two clusters in terms of biochemical recurrence status, ERG expression, preoperative PSA, age,

or prostate weight (Table 5.2).

Next, supervised analysis of miRNA expression patterns was performed for GS8 vs. GS6

PCa. Using a 2-fold enrichment signal difference and a P-value ≤ 0.05 between the two grades

as a threshold, a total of 56 miRNAs were significantly up-regulated in GS8 vs. GS6 (Table 5.3).

Among the significantly up-regulated miRNAs are miR-19a (7.9-fold), miR-96 (6.5-fold), miR-

32 (6.4-fold), and miR-21 (5.3-fold), which have been previously identified to be upregulated in

association with PCa progression [340, 344, 530-535]. Meanwhile, 4 (7%) of the miRNAs that

were significantly differentially expressed between GS8 and GS6 were novel candidates.

However, no significant differentially expressed miRNAs were identified between GS6

recurrent vs. GS6 non-recurrent cases (Table 5.3). Similarly, comparison of all recurrent vs. non-

recurrent cases, independent of GS, also did not yield any significant differentially expressed

miRNAs.

     

    147

Figure 5.1. Unsupervised hierarchical clustering of 32 PCa miRNA expression profiles. Summary of clinicopathological and ERG expression details stratified according to cluster A and cluster B PCa are labeled on top.

     

    148

Table 5.2. Clinicopathological parameters and ERG expression status in cohort I stratified according to cluster A and B on the basis of miRNA expression profiling

Cluster Gleason Score A B 6 47% 92% 8 53% 8% χ2 P-value 0.009 Pathological Stage

pT2 47% 69% pT3+pT4 53% 31% χ2 P-value 0.220 Recurrence No Recurrence 47% 46% Recurrence 53% 54% χ2 P-value 0.946 Surgical Margins

Negative 84% 69% Positive 16% 31% χ2 P-value 0.314 Age

Below Median 63% 46% Above Median 37% 54% χ2 P-value 0.341 ERG status

Negative 67% 62% Positive 33% 38% χ2 P-value 0.778 Preoperative PSA Negative 42% 67% Positive 58% 33% χ2 P-value 0.183 Prostate Weight

Negative 67% 38% Positive 33% 62% χ2 P-value 0.119

     

    149

Table 5.3. Differentially expressed microRNAs comparing GS8 versus GS6 and recurrent versus non-recurrent cases

GS 8 vs. 6 GS6 recurrent vs. non-

recurrent cases All recurrent vs. non-

recurrent cases miRNA Fold Change P-value Fold Change P-value Fold Change P-value miR-140-5p 8.2 0.000000 1.7 0.603767 1.3 0.940 miR-19a-3p 7.9 0.000054 1.4 0.297079 1.2 0.403 miR-142-5p 7.6 0.000004 1.4 0.232038 1.3 0.561 miR-301a-3p 7.0 0.000001 1.1 0.564516 1.1 0.599 miR-96-5p 6.5 0.000146 1.0 0.859521 1.1 0.363 miR-144-3p 6.5 0.129733 2.4 0.208683 2.1 0.095 miR-32-5p 6.4 0.011114 0.8 0.976441 1.1 0.337 miR-335-5p 6.4 0.000099 1.0 0.622322 1.1 0.387 miR-18a-5p 6.2 0.000023 1.4 0.763273 1.3 0.517 miR-135b-5p 5.9 0.000218 1.5 0.302947 1.1 0.856 miR-148b-3p 5.9 0.000012 1.2 0.309683 1.0 0.867 miR-15b-5p 5.5 0.000004 1.1 0.964809 1.0 0.889 miR-9-5p 5.4 0.100962 1.3 0.440944 1.4 0.369 miR-21-5p 5.3 0.000111 1.1 0.826598 1.2 0.435 miR-148a-3p 5.0 0.000073 1.0 0.908981 0.9 0.906 miR-29a-3p 4.9 0.000010 1.2 0.565349 1.1 0.623 miR-27a-3p 4.7 0.000001 1.1 0.843525 1.2 0.498 miR-218-5p 4.6 0.001579 1.2 0.756812 1.2 0.528 miR-183-5p 4.4 0.000582 0.9 0.484076 1.0 0.999 miR-30c-5p 4.3 0.000006 1.1 0.977257 1.0 0.853 miR-29b-3p 4.3 0.000420 1.1 0.918305 1.1 0.471 miR-146b-5p 4.1 0.000016 1.3 0.644966 1.1 0.630 miR-15a-5p 4.1 0.000343 1.1 0.942526 1.1 0.547 miR-143-3p 4.1 0.000099 1.1 0.829648 1.1 0.735 miR-126-3p 4.1 0.000011 1.1 0.738764 1.1 0.925 let-7f-5p 3.9 0.000524 2.5 0.435775 1.9 0.644 miR-27b-3p 3.9 0.000002 1.2 0.567563 1.2 0.499 miR-34c-5p 3.8 0.000019 1.3 0.290113 1.2 0.766 miR-16-5p 3.7 0.000033 1.2 0.826082 1.1 0.889 miR-7-5p 3.6 0.000209 0.9 0.443910 0.9 0.793 miR-10b-5p 3.6 0.000439 1.1 0.733043 1.1 0.915 miR-20b-5p 3.6 0.001817 1.1 0.596109 1.1 0.464 miR-20a-5p 3.5 0.000316 1.1 0.809793 1.1 0.535 miR-98-5p 3.3 0.000179 1.5 0.538970 1.4 0.685 miR-181c-5p 3.2 0.000003 0.9 0.997964 1.0 0.924 miR-17-5p 3.1 0.000517 1.1 0.961188 1.1 0.619 let-7a-5p 3.1 0.000399 2.5 0.585605 1.8 0.847 miR-100-5p 3.1 0.000642 1.3 0.445892 1.1 0.829 miR-10a-5p 3.0 0.000277 1.3 0.506810 1.1 0.675 miR-203a-3p 3.0 0.000699 0.9 0.623086 0.9 0.933 miR-146a-5p 2.9 0.005473 1.0 0.45192 1.0 0.300 miR-191-5p 2.8 0.000204 1.0 0.638465 0.9 0.860 let-7g-5p 2.8 0.000391 1.0 0.685972 1.0 0.971 let-7e-5p 2.7 0.000976 2.1 0.307166 1.5 0.874 miR-128-3p 2.7 0.000214 1.0 0.813452 1.0 0.997 miR-130a-3p 2.7 0.000019 1.0 0.886900 1.0 0.824 let-7i-5p 2.7 0.000192 1.1 0.793234 1.1 0.989 miR-23b-3p 2.6 0.000078 1.0 0.731240 1.0 0.896 miR-25-3p 2.6 0.001605 1.0 0.789911 1.0 0.669 miR-127-5p 2.5 0.000965 1.0 0.556145 1.0 0.888 miR-125b-5p 2.5 0.003437 1.1 0.740848 1.0 0.815 miR-132-3p 2.5 0.000559 1.1 0.925652 1.0 0.775

     

    150

Table 5.3. continued

GS 8 vs. 6 GS6 recurrent vs. non-

recurrent cases All recurrent vs. non-

recurrent cases miRNA Fold Change P-value Fold Change P-value Fold Change P-value miR-181b-5p 2.4 0.000112 1.1 0.716415 1.1 0.872 miR-138-5p 2.3 0.011003 1.9 0.242481 1.5 0.720 let-7d-5p 2.3 0.000369 1.1 0.593440 1.1 0.970 miR-34a-5p 2.1 0.001728 1.0 0.892657 1.1 0.671 miR-150-5p 2.1 0.070805 1.3 0.798448 1.2 0.546 miR-1-3p 2.1 0.051332 1.2 0.523319 1.3 0.566 miR-214-3p 2.1 0.003294 1.1 0.503359 0.9 0.797 let-7c-5p 1.9 0.006180 1.1 0.560055 1.0 0.882 miR-125a-5p 1.9 0.004894 1.0 0.728400

0 1.0 0.919

miR-134-5p 1.9 0.026081 1.2 0.388384 1.0 0.815 let-7b-5p 1.8 0.005107 1.3 0.683336 1.2 0.927 miR-210-3p 1.7 0.010824 1.0 0.510338 1.1 0.280 miR-181a-5p 1.6 0.011280 1.0 0.739256 1.1 0.848 miR-200c-3p 1.6 0.040513 0.7 0.217573 0.8 0.291 miR-193b-3p 1.6 0.020879 1.0 0.697669 1.0 0.805 miR-149-5p 1.5 0.031575 1.0 0.949319 1.0 0.840 miR-181d-5p 1.5 0.078435 1.1 0.740821 1.1 0.728 miR-92a-3p 1.5 0.044409 1.0 0.621269 1.0 0.748 miR-122-5p 1.3 0.194743 0.7 0.682770 0.9 0.406 miR-133b 1.3 0.515455 1.0 0.700180 1.1 0.591 miR-193a-5p 1.2 0.112166 1.0 0.776667 0.9 0.744 miR-155-5p 1.2 0.257197 1.3 0.390177 1.2 0.908 miR-222-3p 1.2 0.762204 1.1 0.554805 1.0 0.890 miR-378a-3p 1.1 0.469184 1.1 0.651664 1.0 0.731 miR-373-3p 1.1 0.801045 1.1 0.558651 1.0 0.848 miR-124-3p 1.0 0.997885 1.2 0.420276 1.1 0.573 miR-372-3p 0.9 0.986890 1.0 0.458503 1.0 0.789 miR-206 0.9 0.403740 2.1 0.181675 1.3 0.205 miR-215-5p 0.7 0.378594 0.9 0.561744 0.9 0.562 miR-184 0.5 0.387075 0.7 0.637296 0.8 0.663

     

    151

5.4.2 MicroRNA-KLK Axis of Interaction

In silico analysis was performed to identify a panel of miRNAs, part of the miScript RT-

qPCR array, implicated in KLK regulation and, potentially, PCa pathogenesis using mirDIP:

microRNA Data Integration Portal, which integrates data from seven databases. The mirDIP

program predicted 521 unique miRNA-KLK interactions by ≥2 databases (20% of mature

miRNAs identified in humans to date). Among these, 42 were significantly up-regulated in GS8

vs. GS6 cases in the miScript RT-qPCR array analysis previously described. KLK10 was

predicted to be the most targeted KLK mRNA, with 231 (44%) miRNAs predicted to target it.

Contrarily, KLK1 was predicted to be the least targeted KLK, with only 2 predicted targeting

miRNAs. The majority (67%) of miRNAs were predicted to target more than one KLK. The

miRNAs predicted to target the greatest number of KLKs are mir-296-3p, mir-612, and miR-

637, with eight predicted interactions each. Many of the predicted interactions might have

potential significant implications for PCa pathogenesis. For example, miR-205 and miR-221,

two of the most consistently studied downregulated miRNAs in PCa, are predicted to target

KLK2, which is known to be upregulated in PCa. Importantly, miRNAs predicted to target

KLKs have additional predicted experimentally verified targets and functions in PCa. A few

notable examples are miR-34c which has been shown to be a direct target of p53 to induce cell

cycle arrest and/or apoptosis [536, 537], miR-16 that targets BCL2, CCND1 and WNT3A to

inhibit proliferation and invasiveness [538], and let-7a that targets E2F2 and CCND2 to induce

cell cycle arrest [539].

5.4.3 Expression of miR-21 and miR-137 in Primary PCa and Normal Tissue

The identification of previously reported upregulated miRNAs in GS8 vs. GS6 cases

provided confidence in our approach. Therefore, from the RT-qPCR array data, miR-21 was

selected for further analysis on the basis of its overexpression in GS8 vs. GS6 cases, previously

     

    152

established biological function in PCa and predicted interaction with KLK10 3’UTR. Notably,

functional studies have found that miR-21 induces downregulation of numerous additional

tumor-suppressor genes such as PTEN, PDCD4, and RECK [540]. Additionally, a novel miRNA

in PCa was selected for further analysis, based on its predicted interaction with KLK10 3’UTR

and hypomethylation on genome-wide methylation arrays of PCa tissues previously performed

in our lab [421]. This miRNA was previously shown to be regulated by DNA methylation, thus

it was expected to be upregulated in PCa [541]. The predicted interaction between these

miRNAs and KLK10 mRNA, which is downregulated in PCa, provides further rationale for the

upregulation of these miRNAs in the disease and an additional potential avenue for the

epigenetic regulation of KLKs. Notably, in silico analysis using the mirDIP: microRNA Data

Integration Portal has revealed numerous predicted tumor suppressor gene targets for this

miRNA including PTEN, KLF5, and SMAD4 [542].

Expression levels of miR-137 and miR-21 were established in PCa tumors and matched

benign tissue from the same prostate. Chi-squared analysis revealed no significant differential

expression of these miRNAs between matched cancer and benign tissue (P-value = 0.505 and

0.91, respectively). A significantly greater proportion of cases expressed high miR-137, but not

miR-21, in histologically normal tissue acquired from PZ compared to TZ (χ2 P-value=0.018).

Next, the association between the proportion of highly expressing miR-137 or miR-21 PCa

specimens and GS was examined (Figure 5.2a). PCa specimens were separated into 3 groups

based on GS; low (GS≤6), intermediate (GS=7), and high grade (GS≥8). High miR-137

expression was associated with higher GS (χ2 P-value = 0.026), whereas a similar significant

association was not observed for miR-21 expression (χ2 P-value = 0.66). Similarly, the

relationship between high miRNA expression and pathological stage was analyzed (Figure

5.4b). PCa specimens were separated into 3 groups; organ confined pT2 cases, pT3a cases that

     

    153

extended into the periprostatic tissue and pT3b+pT4 cases that infiltrated into the seminal

vesicles or bladder musculature. Overall, no significant association between the three groups and

miRNA expression status was observed. Age, preoperative PSA, prostate weight and surgical

margin status were also not associated with the expression of either miRNA. Further, no

correlation between the expression of miR-137 and miR-21 was observed (χ2 P-value = 0.103).

5.4.4 Expression of miR-21, miR-137 and ERG

High expression levels of miR-137, but not miR-21, were significantly associated with ERG-

positive compared to ERG-negative specimens (Figure 5.2c, χ2 P-values = 0.017 and 0.595,

respectively). This correlation suggests miR-137 expression status may have different

significance in ERG-positive vs. ERG-negative cases. Therefore, miR-137 expression data was

further explored through stratification by ERG expression status and tested for association with

GS and/or pathological stage. However, ERG-stratified analysis revealed no additional

significant associations.

5.4.5 Correlation Between DNA Methylation, MicroRNA expression and ERG

Association between DNA methylation of KLK10 and the expression of the two miRNAs

predicted to target it (miR-21 and miR-137) was assessed in Cohort I consisting of 150 PCa

specimens. In this cohort, each of these epigenetic events was independently correlated to ERG

expression. Therefore, a partial correlation analysis that controlled for ERG expression revealed

a significant inverse linear correlation between KLK10 methylation and expression of miR-137,

but not miR-21 (Partial Correlation coefficients = -0.237 and 0.105, P-value = 0.023 and 0.323,

respectively). This correlation was further verified using a χ2 test (Table 5.5). Finally, KLK10

methylation, miR-137 and ERG expression was integrated to evaluate their combined

relationship to clinicopathologic parameters (Table 5.4). The prevalence of all three markers

increased with pathological stage but not GS.

     

    154

Figure 5.2. Proportion of miR-137 and miR-21 high expression cases stratified by (a) Gleason score, (b) pathological stage and (c) ERG expression status

Table 5.4. Proportion of cases highly methylated for KLK10, highly expressing miR-137 and ERG positive in cohort I

KLK10 methylation, miR-137 and ERG expression

None Any 1 Any 2 All 3

Gleason Score GS=<7 69% 78% 82% 55%

GS>=8 31% 22% 18% 45% χ2 P-value 0.28 Pathological Stage

pT2 69% 62% 52% 18% pT3 31% 38% 48% 82% χ2 P-value 0.04 Surgical Margins

Negative 81% 78% 71% 82% Positive 19% 22% 29% 18% χ2 P-value 0.839 Age

Below Median 44% 48% 68% 45% Above Median 56% 52% 32% 55% χ2 P-value 0.288 Preoperative PSA

Below Median 69% 48% 35% 45% Above Median 31% 52% 65% 55% χ2 P-value 0.199 Prostate Weight

Below Median 31% 49% 68% 45% Above Median 69% 51% 32% 55% χ2 P-value 0.117

     

    155

5.5 Discussion

This chapter presents 56 differentially expressed miRNAs in a group of 32 prostate tumors

representing different GS (GS8 vs. GS6) that may be potentially involved in PCa progression.

Previously published miRNA expression profiles have demonstrated a widespread

downregulation of miRNA expression levels correlating with PCa progression. Our study does

not contradict this as we used a more targeted platform that includes numerous upregulated

miRNA sets. Up-regulation of miRNAs also occurs in PCa tumors, and it is consistent with the

known oncogenic activity of many miRNAs [340]. This study utilized RT-qPCR miRNA arrays,

similar to those reported by others, which were recently identified as having one of the highest

analytical sensitivity and reproducibility metrics for detection of miRNA expression among 12

commercially available platforms [543-545]. Our observations were supported by several other

published studies on PCa miRNA expression that reported upregulation of many of the same

miRNAs such as miR-96, miR-20a, and miR-32 to name a few [344, 530, 532]. However, the

expression of numerous miRNAs was difficult to compare to previously published studies due to

lack of consistent “-3p” and “-5p” designation in publications, utilization of different types of

samples and control reference RNAs.

Previous reports regarding a miRNA signature in association with PCa recurrence are

conflicting. A study by Ozen et al. found no single miRNA was useful as a predictor of

recurrence [341]. Tong et al. reported 16 miRNAs that could distinguish early recurrence from

late recurrence using hierarchical clustering, but expression of no single miRNA was changed

over 2-fold with a P-value<0.05 between the two groups in the study [343]. Later, Karatas et al.,

Schaefer et al. and Schubert et al. reported 93, 15 and 7 miRNAs, respectively, differentially

expressed in association with biochemical recurrence after RP [344, 546, 547]. This further

highlights the inconsistencies across prior studies regarding alterations in miRNA expression in

association with biochemical recurrence. The findings of this chapter seem to be in agreement

     

    156

with the former studies. The lack of significantly differentially expressed miRNAs in

association with recurrence is intriguing. A potential explanation could be substantial variability

in the expression of miRNAs and ncRNAs used for reference that is obscuring the identification

of robust differential expression. Additionally, the association between miRNA expression and

disease outcome may be weak and/or cannot be accurately detected with the analytical methods

currently available.

The chapter presents the first evidence for association between miR-137 and ERG expression

status and clinicopathological parameters in a large cohort of RPs. Elevated miR-137 expression

correlated with GS, suggesting it could be investigated in future studies as a biomarker of PCa

progression. This is consistent with previous publications on the upregulation of miR-137 in

tongue squamous cell carcinoma and bladder cancer but in contrast to reports on the

downregulation of miR-137 in oral, ovarian, colon, gastric and lung cancers [548-554].

However, this is not surprising since numerous miRNAs have shown tissue- and tumor- specific

expression signatures [338]. Alternatively, the association between miR-137 expression and GS

may have just a spurious correlation. Although there are no experimentally validated gene

targets for mir-137 in PCa, in silico analysis using the mirDIP: microRNA Data Integration

Portal has revealed numerous predicted tumor suppressor gene targets for this miRNA including

PTEN, KLF5, and SMAD4 [542].  

Further evidence that increased mir-137 expression may be associated with PCa

aggressiveness stems from our analysis of its expression in the adjacent normal tissues. First, the

expression of miR-137 in tumor adjacent normal tissues was not associated with GS. Secondly,

expression of mir-137 was significantly lower in TZ tissue, which has been suggested to give

rise to less aggressive tumors that have a lower biochemical recurrence rate compared to cancers

that develop in the PZ [6, 555]. Currently, little is known about the differences in miRNA

     

    157

expression between tumor adjacent prostate zones or their potential role in prostate biology and

carcinogenesis. Our results suggest further miRNA expression analyses in different prostate

anatomical regions of normal tissue from PCa patients and healthy individuals are required.

One of the scientific rationales for selecting miR-137 for RT-qPCR analysis was its predicted

targeting of KLK10, which was found to be hypermethylated in PCa in chapter 2 of this thesis.

This is the first study to report an association between a decrease in 5’UTR DNA methylation of

KLK10 and increase in expression of a miRNA that is predicted to target it at the 3’UTR, with

ERG expression status as a covariate. This is consistent with the concept of functional

complementation between DNA methylation and miRNA regulation in the human genome

reported by Su et al. which suggests that genes under strong promoter DNA methylation control

tend to avoid miRNA regulation, and vice versa [556]. Yet, KLK10 has one of the largest

number of both CGIs and miRNAs predicted to target it in the KLK family. This is intriguing but

difficult to interpret as there are numerous unanswered questions regarding the methylation

status of the various CGIs in KLK10, their functional relevance, the functional role of numerous

predicted KLK10-miRNA interactions in post-transcriptional regulation, the dominance of each

regulatory mechanism, the functional role of hK10 in PCa, and lastly, the role of ERG in this

context? There is a need for future studies to answer all of the abovementioned questions to

elucidate if this correlation has a biological role or if it is just a spurious correlation. Further

systematic genome-wide examination of the coordinated effects of DNA methylation and

miRNAs on regulating expression of their common target genes in PCa is also required.

The second miRNA investigated in this study, miR-21 has been extensively studied in

various cancers and was shown to be overexpressed in tumors of the breast, colon, and cervix

[540]. Moreover, upregulation of miR-21 has been suggested as a prognostic marker in

numerous cancers such as cervical, colon and breast cancers [540]. Functional studies have

     

    158

found that miR-21 induces downregulation of numerous tumor-suppressor genes such as PTEN,

PDCD4, and RECK [540]. In PCa, miR-21 was initially identified to be upregulated in PCa in

two large-scale studies conducted by Volinia et al. [338] and Szczyrba et al. [345].

Subsequently, conflicting reports have come out on the expression status of miR-21 in PCa and

its association with clinicopathological features [528, 540, 557, 558]. However, our study

performed the largest RT-qPCR based analysis of miR-21 expression in localized PCa tissue and

is the first to examine its association with TMPRSS2-ERG. In a large and diverse cohort of PCa

samples with or without ERG expression and with different clinicopathological features, miR-21

was neither associated with ERG expression status nor aggressive PCa clinical state defined by

GS, pathological stage or preoperative PSA. The clinical significance of miR-21 expression in

PCa prognosis may vary across studies due to differences in cohort recruitment, size and

clinicopathological characteristics as well as platforms. This suggests that mir-21 upregulation is

not a common event to all PCa tissue cohorts and/or it cannot be accurately quantified by all

platforms.

Further, miR-21, similar to miR-137, was not differently expressed in PCa compared to

matched benign tissues from the same prostate. This is an intriguing observation since overall

differential expression of miRNAs between PCa and tumor adjacent normal was previously

reported. Yet, expression of miR-137 has not been systematically investigated in cancer in the

literature, thus it is difficult to interpret these results. Publications investigating miR-137

expression in colon cancer suggest its CGI hypermethylation and subsequent downregulation is

a tumor-specific event when compared to adjacent normal tissue [541]. However, our results

suggest this is not the case in PCa. The expression of miR-21 was previously shown to be

elevated in tumor compared to tumor-adjacent normal in gastric cancer [559]. Yet, a previous

publication investigating the expression of miR-21 in FFPE PCa tissue and adjacent normal

     

    159

found no significant differences in expression, similar to our observations [528]. We hypothesize

this observation may be explained by the field effect [449]. Numerous molecular alterations

have been reported to constitute the field effect including aberrant DNA methylation, mutations

and copy number alterations. The clinical significance of the field effect in PCa has been

recently demonstrated by the commercial ConfirmMDx test, an epigenetic assay for aberrant

DNA methylation on negative PCa biopsy as an indication of high risk for occult PCa [302].

Yet, to the best of my knowledge, only one study has previously explored alterations in miRNA

expression between normal tissues from PCa-negative vs. PCa-positive biopsies [560]. It is

worth exploring further miRNA expression in normal prostatic tissue to support the possibility

that it is part of the field effect. With the development of more sophisticated and sensitive

platforms for miRNA analysis one can envision that miRNA expression will be incorporated

into PCa biopsy analysis to improve risk assessment for occult PCa.

TMPRSS2-ERG has been associated with distinct DNA methylation, histone modifications,

and miRNA expression profiles. For example, oncogenic TMPRSS2-ERG fusion has been

correlated with downregulation of mir-200c and mir-221 as well as upregulation of miR-26a

[418, 561, 562]. There are many possible functional mechanisms that link miR-137 and ERG

expression. MiR-137 may be part of signaling cascades that synergize with ERG such as the AR

pathway. Alternately, ERG may regulate epigenetic pathways that alter DNA methylation,

known to regulate miR-137 in cancer [551]. For example, Borno et al. have shown that

TMPRSS2-ERG negative tumors display higher expression levels of EZH2 compared to tumors

containing the translocation [418]. This may promote interactions between EZH2 (within the

context of the PRC2 and 3) with DNMTs resulting in DNA hypermethylation of miR-137 and

subsequent decrease in miR-137 expression in ERG negative tumors [454]. However, little is

     

    160

known about miR-137 regulation in PCa and it needs to be further investigated to elucidate the

link between this miRNA and ERG.

In conclusion, in this chapter, a signature of 56 differentially expressed miRNAs was found

in association with GS8 vs. GS6 PCa. Further, miR-137 and miR-21 expression was

characterized for the first time in a large cohort of 150 matched normal prostatic and adjacent

PCa specimens in correlation with ERG expression status. Significant miR-137 upregulation

with more advanced GS and ERG expression was demonstrated. Lastly, expression of miR-137,

predicted to target KLK10 3’UTR, was inversely correlated to KLK10 5’UTR methylation

(controlling for ERG expression status) which warrants future additional research into

simultaneous coordinated regulation of gene expression by multiple epigenetic mechanisms.

     

    161

Chapter 6: Discussion and future directions

6.1 Epigenetic Marks in Association with PCa Initiation and Progression

The aim of PCa biomarker research is to identify a panel of markers that can be easily and

inexpensively tested in biofluid samples preferably obtained using non-invasive strategies. The

goal is to not only analyze these markers in asymptomatic individuals to accurately and quickly

diagnose PCa, but also to guide treatment decision making, predict disease aggressiveness and

outcome among PCa patients at different stages of disease progression. The work of this thesis

contributes to the body of knowledge that will help accomplish this goal in the future.

This thesis is a preclinical exploratory biomarker study, as defined by Pepe et al. in 2001

[563], conducted on retrospectively collected RP specimens. The studies reported in this thesis

suggest several potentially promising biomarkers, including: (1) KLK10 DNA methylation,

which may have diagnostic and prognostic value, (2) Methylation of APC, RASSF1A, TBX15,

HOXD3 and presence of LC as a prognostic biomarker panel and (3) miR-137 expression that

may potentially have prognostic value. Further, a multiplex MethyLight protocol has been

developed as a nascent technology with greater potential for translation into a clinical setting

than individual gene analysis.

The next phase of research for these biomarkers is further exploration and validation in

additional large international cohorts. In addition to the biomarkers presented in this thesis, there

are thousands of reports on potentially promising biomarkers for PCa, yet few make it to the

clinic. Therefore, we have taken care to adhere to the REMARK guidelines (appendix) in

reporting our tissue tumor biomarkers, which should ease and facilitate the assessment of their

value as diagnostic and/or prognostic markers [564].

To test the potential of KLK10 methylation as a diagnostic biomarker, it needs to be tested on

multiple large international cohorts of prospectively collected specimens that are obtained non-

     

    162

invasively. If to be tested as potential prognostic marker, it should be also tested in independent

validation cohorts with follow-up data available. Our pilot analysis of KLK10 methylation in

serum revealed it may not be a suitable biofluid for these studies. The next possibility is to

investigate KLK10 methylation in post-DRE urine. Preliminary experiments by other members

of the laboratory suggest the detection of KLK10 methylation in post-DRE urine specimens is

feasible. Based on these findings, KLK10 methylation is currently being investigated for the first

time, as a potential prognostic methylation biomarker to discriminate between aggressive vs.

indolent PCa, through the Movember Global Action Plan funded “urinary biomarker initiative”

(http://ca.movember.com/programs/global-collaboration).

Further studies need to assess assay reproducibility, perform an ROC curve and compare the

AUC to PSA, as well as estimate the true- and false- positive rates between healthy controls,

men with abnormalities of the prostate (e.g. BPH) and PCa patients. The effect of factors such as

age and lifestyle on the detection of KLK10 methylation in DRE-urine will also need to be

determined. If detection of KLK10 DNA methylation in post-DRE urine samples proves to have

sufficient sensitivity and specificity, it may help reduce the rate of PCa-negative biopsies.

The potential prognostic value of KLK10, APC, RASSF1A, TBX15 and HOXD3 methylation,

miR-137 expression and presence of LC as a PCa biosignature also needs to be validated in

many large prospective outcome-based cohorts of biopsy and RP biospecimens. These studies

need to test the reproducibility of the results in this thesis and investigate the effects of a large

list of factors on the prognostic value of this biosignature, as outlined in numerous publications

[565-567]. These include, but are not limited to cohort selection bias, biospecimen selection,

collection, handling and storage, methodological artifacts and appropriateness of statistical

analysis. Next, the ability of this biosignature to predict outcome needs to be compared to the

current histopathology-based gold standard for PCa classification. Importantly future studies

     

    163

should explore the exciting potential of this biosignature consisting of DNA methylation,

miRNA and histopathology based biomarkers to predict PCa disease aggressiveness on biopsy

and/or in biofluids, as discussed previously.

6.2 Roles for Epigenetics in the Regulation of KLKs in PCa

The selection of biomarkers for this thesis was primarily focused around the KLK gene

family, as there is a large body of evidence that this protease family is deregulated in PCa at the

mRNA and protein levels. Therefore, the hypothesis of this thesis was that the mechanisms that

regulate the KLK gene locus are also likely deregulated. Although the regulation of KLK

expression is still not well understood, the presence of multiple CGIs and miRNA target

sequences together with previous exploratory publications provide clear evidence of

involvement of epigenetic mechanisms in their regulation. The role of DNA methylation and

miRNAs in KLK biology is an important and interesting topic that needs to be analyzed in future

studies.

In addition to the specific hypermethylation events in CGIs within KLKs 6, 10 and 15,

investigated in this thesis, hypermethylation of numerous CGIs within KLKs 1, 9, 12 and 13 in

conjunction with hypomethylation of KLK4 was observed in the CGI microarrays published by

Kron et al. [421]. Further, a regulatory role for DNA methylation has been previously implicated

in KLK5 and KLK7 [399, 568]. Taken together, this suggests a broader regulatory role for DNA

methylation in this large contiguous gene cluster. Future work is warranted to elucidate the KLK

locus-wide CGI methylation events and their contribution to gene expression. Importantly, the

role of DNA methylation in enhancer and potentially common regulatory regions of KLKs is

needed. Additional investigation into DNA methylation as a regulator of the observed

coordinated ’cassette-type’ regulatory elements of numerous KLK genes located in close

proximity on chromosome 19 in PCa is warranted.

     

    164

There are a number of potential mechanisms behind the widespread DNA methylation events

in the KLK gene locus. First, as previously discussed, DNMTs and EZH2 (which directly targets

DNMTs) are overexpressed in PCa and may contribute to the hypermethylation of certain genes

in the KLK locus, among others [287, 454]. Second, DNA methylation of several KLK genes

may occur during normal prostate physiology and undergo spreading throughout the cluster

during carcinogenesis, as has been shown previously for other genes [569, 570].

While DNA methylation has been shown to influence KLK6 and KLK10 gene expression in

this thesis, it is not the only mechanism that controls the expression of these genes. There is a

need to integrate data on chromosomal changes, SNPs, mutations, DNA methylation, histone

modifications, miRNAs and hormonal regulation to truly elucidate the function, significance and

dominance of each regulatory mechanism in the regulation of this gene family in PCa. The

cooperativity and integrative contributions of these mechanisms as part of the genetic and

epigenetic landscape also needs to be investigated. Importantly, when performing this integrated

analysis in PCa, relevant SNPs that have been associated with the incidence of the disease need

to be incorporated (summary of publications). For example, in the regulation of KLK4 in PCa, it

seems that the H3Ac and H3K4me2/3 histone marks previously observed to be enriched in the

region are in line with the hypomethylation observed on the methylation microarrays published

by Kron et al. to confer an open chromatin region and allow transcription factors access to the

androgen response elements implicated in the regulation of this gene [414, 421, 571, 572].

Further, there are no predicted interactions between this KLK and miRNAs known to be

deregulated in PCa. Taken together, these regulatory mechanisms seem to cooperate to bring

about the observed upregulation of KLK4 in PCa.

The results of this thesis also demonstrate cooperativity between KLK10 CGI methylation

and miR-137 expression. This phenomenon is likely not specific to KLK10 mRNA and may

     

    165

have various explanations. Genome-wide, the expression of numerous miRNAs is regulated by

DNA methylation [573, 574]. Thus, cases displaying elevated methylation may have subsequent

downregulation of miRNAs and vice verse. Alternative mechanism could be that cases with

elevated micronome may have elevated expression of miRNAs that target the DNA methylation

machinery. Subsequently, these cases display lower DNA methylation.

Future work should also investigate the biology behind the association of KLK10

methylation, miR-137 expression and LC lesions with ERG expression and/or TMPRSS2-ERG

fusion. These associations could involve the ERG, EZH2 and DNMTs pathway, as discussed

earlier. Thus future work may involve elucidating the direct link of TMPRSS2-ERG fusions in

creating an altered epigenetic landscape.

Several questions pertaining to the role of KLKs in the biology of PCa also remain to be

elucidated. Both KLKs 6 and 10 have been shown to be downregulated in PCa and suggested to

function as tumor suppressor genes since their re-expression in breast cancer cell lines result in

slower proliferation rates and decrease in tumorigenicity following injection into nude mice

[401, 439, 440]. The proposed substrate for hK6 in breast cancer tumorigenesis is vimentin

while hK10 substrates remain unknown. Future work is needed to investigate whether the

downregulation of KLK6 and KLK10 genes may have a functional contribution in the

development and progression of PCa.

In conclusion, the work presented in this thesis creates a better understanding of alterations in

DNA methylation events in the KLK locus and miRNAs predicted to target KLKs post-

transcriptionally. Additionally, the studies summarized here support the hypothesis that DNA

hypermethylation mediates, in part, the observed downregulation of KLKs 6 and 10 in PCa.

Clinically, these epigenetic aberrations associate with PCa molecular subtypes defined by ERG

expression, aggressive disease histopathology and recurrence following RP.

     

    166

Appendix

Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK)

INTRODUCTION 1. State the marker examined, the study objectives, and any prespecified hypotheses. MATERIALS AND METHODS Patients 2. Describe the characteristics (e.g., disease stage or comorbidities) of the study patients, including their source and inclusion and exclusion criteria. 3. Describe treatments received and how chosen (e.g., randomized or rule-based). Specimen characteristics 4. Describe type of biological material used (including control samples) and methods of preservation and storage. Assay methods 5. Specify the assay method used and provide (or reference) a detailed protocol, including specific reagents or kits used, quality control procedures, reproducibility assessments, quantitation methods, and scoring and reporting protocols. Specify whether and how assays were performed blinded to the study endpoint. Study design 6. State the method of case selection, including whether prospective or retrospective and whether stratifi cation or matching (e.g., by stage of disease or age) was used. Specify the time period from which cases were taken, the end of the follow-up period, and the median follow-up time. 7. Precisely define all clinical endpoints examined. 8. List all candidate variables initially examined or considered for inclusion in models. 9. Give rationale for sample size; if the study was designed to detect a specifi ed effect size, give the target power and effect size. Statistical analysis methods 10. Specify all statistical methods, including details of any variable selection procedures and other model-building issues, how model assumptions were verified, and how missing data were handled. 11. Clarify how marker values were handled in the analyses; if relevant, describe methods used for cutpoint determination. RESULTS Data 12. Describe the fl ow of patients through the study, including the number of patients included in each stage of the analysis (a diagram may be helpful) and reasons for dropout. Specifically, both overall and for each subgroup extensively examined report the numbers of patients and the number of events. 13. Report distributions of basic demographic characteristics (at least age and sex), standard (disease-specific) prognostic variables, and tumor marker, including numbers of missing values. Analysis and presentation 14. Show the relation of the marker to standard prognostic variables. 15. Present univariate analyses showing the relation between the marker and outcome, with the estimated effect (e.g., hazard ratio and survival probability). Preferably provide similar analyses for all other variables being analyzed. For the effect of a tumor marker on a time-to-event outcome, a Kaplan – Meier plot is recommended. 16. For key multivariable analyses, report estimated effects (e.g., hazard ratio) with confi dence intervals for the marker and, at least for the fi nal model, all other variables in the model. 17. Among reported results, provide estimated effects with confi dence intervals from an analysis in which the marker and standard prognostic variables are included, regardless of their statistical signifi cance. 18. If done, report results of further investigations, such as checking assumptions, sensitivity analyses, and internal validation. DISCUSSION 19. Interpret the results in the context of the prespecified hypotheses and other relevant studies; include a discussion of limitations of the study. 20. Discuss implications for future research and clinical value.

     

    167

Summary of publications

Olkhov-Mitsel E, Zdravic D, Kron K, van der Kwast T, Fleshner N, Bapat B: Novel Multiplex MethyLight Protocol for Detection of DNA Methylation in Patient Tissues and Bodily Fluids. Scientific reports 2014, 4:4432. White-Al Habeeb NM, Ho LT, Olkhov-Mitsel E, Kron K, Pethe V, Lehman M, Jovanovic L, Fleshner N, van der Kwast T, Nelson CC et al: Integrated analysis of epigenomic and genomic changes by DNA methylation dependent mechanisms provides potential novel biomarkers for prostate cancer. Oncotarget 2014, 5(17):7858-7869. Olkhov-Mitsel E, Bapat B. Quantitative DNA methylation analysis of genes coding for kallikrein-related peptidases 6 and 10 as biomarkers for prostate cancer, "Beyond the Abstract". Invited commentary for UroToday.com. URL:www.urotoday.com/index.php?option=com_content&Itemid=830&catid=1134& id=56806&lang=en&view=article Olkhov-Mitsel E, Van der Kwast T, Kron K, Ozcelik H, Briollais L, Massaey C, Recker F, Kwiatkowski M, Fleshner N, Diamandis EP, Zlotta AR and Bapat B. Quantitative DNA methylation analysis of genes coding for kallikrein-related peptidases 6 and 10 as biomarkers for prostate cancer. Epigenetics 2012, 7(9):1037-1045. Olkhov-Mitsel E, Bapat B. Strategies for discovery and validation of methylated and hydroxymethylated DNA biomarkers. Cancer Med 2012, 1(2):237-260. Pasic MD, Olkhov E, Bapat B, Yousef GM. Epigenetic regulation of kallikrein-related peptidases: there is a whole new world out there. Biological chemistry 2012, 393(5):319-330.

Briollais L, Ozcelik H, Xu J, Kwiatkowski M, Savas S, Recker F, Kuk C, Fleshner N, Olkhov-Mitsel E Hanna S, Juvet T, Friedlander M, Li H, Chadwick K, Trachtenberg J, Toi A, van der Kwast T, Bapat B, Diamandis EP, Zlotta A. Fine-Mapping of the Kallikrein Locus Supports a Role for the Kallikrein 6 Region in Genetic Predisposition for Aggressive Prostate Cancer: Results from a Canadian Cohort and the Swiss Arm of the European Randomized Study for Prostate Cancer Screening. Manuscript submitted to Human Molecular Genetics. Manuscripts in Preperation Olkhov-Mitsel E, Siadat F, Trudel D, Kron K, Liu L, Van der Kwast T, and Bapat B. Epigenetic Alterations in Prostatic Carcinoma Glands with Cribriform Architecture or Intraductal Carcinoma. Manuscript in preperation. Olkhov-Mitsel E, Van der Kwast T, Fleshner N, Diamandis EP, Zlotta A, and Bapat B. Alterations in miR-137 and miR-21 Expression with ERG Expression and Prostate Cancer Aggressiveness. Manuscript in preperation. Olkhov-Mitsel E, Van der Kwast T, Hermanns T, Mortezavi A, Wild PJ, Zlotta A, Fleshner N, Bapat B. Assessment of MMR Protein Expression and Its Prognostic Significance in Prostate Cancer. Manuscript in preperation. Zhao F, Vesprini D, Olkhov-Mitsel E, Zdravic D, Zlotta AR, Venkateswaran V, Loblaw A, Van Der Kwast T, Fleshner NE, Klotz L, Bapat B. Urinary DNA methylation biomarkers: a non-invasive method for prostate cancer monitoring. Manuscript in preperation. Olkhov-Mitsel E, Savio A, Kron K, Pethe V, van Rhijn BW, van der Kwast T, Bapat B. Epigenome-Wide DNA Methylation Profiling Identifies Differential Methylation Biomarkers in High-Grade Bladder Cancer. Manuscript in preperation.

     

    168

References 1. McNeal JE. The anatomic heterogeneity of the prostate. Progress in clinical and biological

research 1980, 37:149-160. 2. Isaacs JT, Coffey DS. Etiology and disease process of benign prostatic hyperplasia. The Prostate

Supplement 1989, 2:33-50. 3. Wang Y, Hayward S, Cao M, Thayer K, Cunha G. Cell differentiation lineage in the prostate.

Differentiation; research in biological diversity 2001, 68(4-5):270-279. 4. Wang X, Kruithof-de Julio M, Economides KD, Walker D, Yu H, Halili MV, Hu YP, Price SM,

Abate-Shen C, Shen MM. A luminal epithelial stem cell that is a cell of origin for prostate cancer. Nature 2009, 461(7263):495-500.

5. McNeal JE. Normal histology of the prostate. The American journal of surgical pathology 1988, 12(8):619-633.

6. McNeal JE, Redwine EA, Freiha FS, Stamey TA. Zonal distribution of prostatic adenocarcinoma. Correlation with histologic pattern and direction of spread. The American journal of surgical pathology 1988, 12(12):897-906.

7. Thorpe A, Neal D. Benign prostatic hyperplasia. Lancet 2003, 361(9366):1359-1367. 8. Garraway WM, Collins GN, Lee RJ. High prevalence of benign prostatic hypertrophy in the

community. Lancet 1991, 338(8765):469-471. 9. King CR, Ferrari M, Brooks JD. Prognostic significance of prostate cancer originating from the

transition zone. Urologic oncology 2009, 27(6):592-597. 10. Ferlay J, Soerjomataram I, Ervik M, Dikshit R, Eser S, Mathers C, Rebelo M, Parkin DM,

Forman D, Bray, F. GLOBOCAN 2012 v1.0, Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 11 [Internet]. Lyon, France: International Agency for Research on Cancer; 2013. Available from: http://globocan.iarc.fr, accessed on 4/9/2014.

11. Canadian Cancer Statistics 2014. In. Produced by Canadian Cancer Society and Statistics Canada; 2014.

12. Lichtenstein P, Holm NV, Verkasalo PK, Iliadou A, Kaprio J, Koskenvuo M, Pukkala E, Skytthe A, Hemminki K. Environmental and heritable factors in the causation of cancer-analyses of cohorts of twins from Sweden, Denmark, and Finland. The New England journal of medicine 2000, 343(2):78-85.

13. Carter BS, Beaty TH, Steinberg GD, Childs B, Walsh PC. Mendelian inheritance of familial prostate cancer. Proceedings of the National Academy of Sciences of the United States of America 1992, 89(8):3367-3371.

14. Gaines AR, Turner EL, Moorman PG, Freedland SJ, Keto CJ, McPhail ME, Grant DJ, Vidal AC, Hoyo C. The association between race and prostate cancer risk on initial biopsy in an equal access, multiethnic cohort. Cancer causes & control : CCC 2014, 25(8):1029-1035.

15. De Marzo AM, Platz EA, Sutcliffe S, Xu J, Gronberg H, Drake CG, Nakai Y, Isaacs WB, Nelson WG. Inflammation in prostate carcinogenesis. Nature reviews Cancer 2007, 7(4):256-269.

16. Sfanos KS, De Marzo AM. Prostate cancer and inflammation: the evidence. Histopathology 2012, 60(1):199-215.

17. Vignozzi L, Maggi M. Prostate cancer: Intriguing data on inflammation and prostate cancer. Nature reviews Urology 2014, 11(7):369-370.

18. Gurel B, Iwata T, Koh CM, Jenkins RB, Lan F, Van Dang C, Hicks JL, Morgan J, Cornish TC, Sutcliffe S et al. Nuclear MYC protein overexpression is an early alteration in human prostate carcinogenesis. Modern pathology 2008, 21(9):1156-1167.

19. Nakayama M, Bennett CJ, Hicks JL, Epstein JI, Platz EA, Nelson WG, De Marzo AM. Hypermethylation of the human glutathione S-transferase-pi gene (GSTP1) CpG island is present in a subset of proliferative inflammatory atrophy lesions but not in normal or hyperplastic epithelium of the prostate: a detailed study using laser-capture microdissection. The American journal of pathology 2003, 163(3):923-933.

     

    169

20. van Leenders GJ, Gage WR, Hicks JL, van Balken B, Aalders TW, Schalken JA, De Marzo AM. Intermediate cells in human prostate epithelium are enriched in proliferative inflammatory atrophy. The American journal of pathology 2003, 162(5):1529-1537.

21. Yildiz-Sezer S, Verdorfer I, Schafer G, Rogatsch H, Bartsch G, Mikuz G. Assessment of aberrations on chromosome 8 in prostatic atrophy. BJU international 2006, 98(1):184-188.

22. Moreira DM, Nickel JC, Gerber L, Muller RL, Andriole GL, Castro-Santamaria R, Freedland SJ. Baseline prostate inflammation is associated with a reduced risk of prostate cancer in men undergoing repeat prostate biopsy: results from the REDUCE study. Cancer 2014, 120(2):190-196.

23. Allott EH, Masko EM, Freedland SJ. Obesity and prostate cancer: weighing the evidence. European urology 2013, 63(5):800-809.

24. Venkateswaran V, Klotz LH. Diet and prostate cancer: mechanisms of action and implications for chemoprevention. Nature reviews Urology 2010, 7(8):442-453.

25. Chan JM, Stampfer MJ, Giovannucci E, Gann PH, Ma J, Wilkinson P, Hennekens CH, Pollak M. Plasma insulin-like growth factor-I and prostate cancer risk: a prospective study. Science 1998, 279(5350):563-566.

26. Roddam AW, Allen NE, Appleby P, Key TJ, Ferrucci L, Carter HB, Metter EJ, Chen C, Weiss NS, Fitzpatrick A et al. Insulin-like growth factors, their binding proteins, and prostate cancer risk: analysis of individual patient data from 12 prospective studies. Annals of internal medicine 2008, 149(7):461-471.

27. Roddam AW, Allen NE, Appleby P, Key TJ. Endogenous sex hormones and prostate cancer: a collaborative analysis of 18 prospective studies. Journal of the National Cancer Institute 2008, 100(3):170-183.

28. Hsing AW, Chua S, Jr., Gao YT, Gentzschein E, Chang L, Deng J, Stanczyk FZ. Prostate cancer risk and serum levels of insulin and leptin: a population-based study. Journal of the National Cancer Institute 2001, 93(10):783-789.

29. Morgentaler A, Bruning CO, 3rd, DeWolf WC. Occult prostate cancer in men with low serum testosterone levels. Jama 1996, 276(23):1904-1906.

30. Stattin P, Soderberg S, Hallmans G, Bylund A, Kaaks R, Stenman UH, Bergh A, Olsson T. Leptin is associated with increased prostate cancer risk: a nested case-referent study. The Journal of clinical endocrinology and metabolism 2001, 86(3):1341-1345.

31. Nelson WG, De Marzo AM, Deweese TL, Lin X, Brooks JD, Putzi MJ, Nelson CP, Groopman JD, Kensler TW. Preneoplastic prostate lesions: an opportunity for prostate cancer prevention. Annals of the New York Academy of Sciences 2001, 952:135-144.

32. MacInnis RJ, English DR. Body size and composition and prostate cancer risk: systematic review and meta-regression analysis. Cancer causes & control : CCC 2006, 17(8):989-1003.

33. Kristal AR, Arnold KB, Neuhouser ML, Goodman P, Platz EA, Albanes D, Thompson IM. Diet, supplement use, and prostate cancer risk: results from the prostate cancer prevention trial. American journal of epidemiology 2010, 172(5):566-577.

34. Liu Y, Hu F, Li D, Wang F, Zhu L, Chen W, Ge J, An R, Zhao Y. Does physical activity reduce the risk of prostate cancer? A systematic review and meta-analysis. European urology 2011, 60(5):1029-1044.

35. Gilbert R, Metcalfe C, Oliver SE, Whiteman DC, Bain C, Ness A, Donovan J, Hamdy F, Neal DE, Lane JA et al. Life course sun exposure and risk of prostate cancer: population-based nested case-control study and meta-analysis. International journal of cancer 2009, 125(6):1414-1423.

36. Huncharek M, Haddock KS, Reid R, Kupelnick B. Smoking as a risk factor for prostate cancer: a meta-analysis of 24 prospective cohort studies. American journal of public health 2010, 100(4):693-701.

37. Rota M, Scotti L, Turati F, Tramacere I, Islami F, Bellocco R, Negri E, Corrao G, Boffetta P, La Vecchia C et al. Alcohol consumption and prostate cancer risk: a meta-analysis of the dose-risk relation. European Journal of Cancer Prevention 2012, 21(4):350-359.

38. Zhou CK, Pfeiffer RM, Cleary SD, Hoffman HJ, Levine PH, Chu LW, Hsing AW, Cook MB. Relationship Between Male Pattern Baldness and the Risk of Aggressive Prostate Cancer: An

     

    170

Analysis of the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial. Journal of clinical oncology 2015, 33(5):419-25.

39. The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group. The New England journal of medicine 1994, 330(15):1029-1035.

40. Giovannucci E, Rimm EB, Liu Y, Stampfer MJ, Willett WC. A prospective study of tomato products, lycopene, and prostate cancer risk. Journal of the National Cancer Institute 2002, 94(5):391-398.

41. Li H, Stampfer MJ, Giovannucci EL, Morris JS, Willett WC, Gaziano JM, Ma J. A prospective study of plasma selenium levels and prostate cancer risk. Journal of the National Cancer Institute 2004, 96(9):696-703.

42. Morris JD, Pramanik R, Zhang X, Carey AM, Ragavan N, Martin FL, Muir GH. Selenium- or quercetin-induced retardation of DNA synthesis in primary prostate cells occurs in the presence of a concomitant reduction in androgen-receptor activity. Cancer letters 2006, 239(1):111-122.

43. Ryan-Harshman M, Aldoori W. The relevance of selenium to immunity, cancer, and infectious/inflammatory diseases. Canadian journal of dietetic practice and research 2005, 66(2):98-102.

44. Venkateswaran V, Klotz LH, Fleshner NE. Selenium modulation of cell proliferation and cell cycle biomarkers in human prostate carcinoma cell lines. Cancer research 2002, 62(9):2540-2545.

45. Liu X, Allen JD, Arnold JT, Blackman MR. Lycopene inhibits IGF-I signal transduction and growth in normal prostate epithelial cells by decreasing DHT-modulated IGF-I production in co-cultured reactive stromal cells. Carcinogenesis 2008, 29(4):816-823.

46. Jiang Q, Wong J, Fyrst H, Saba JD, Ames BN. gamma-Tocopherol or combinations of vitamin E forms induce cell death in human prostate cancer cells by interrupting sphingolipid synthesis. Proceedings of the National Academy of Sciences of the United States of America 2004, 101(51):17825-17830.

47. Venkateswaran V, Klotz LH, Ramani M, Sugar LM, Jacob LE, Nam RK, Fleshner NE. A combination of micronutrients is beneficial in reducing the incidence of prostate cancer and increasing survival in the Lady transgenic model. Cancer Prev Res (Phila) 2009, 2(5):473-483.

48. Duffield-Lillico AJ, Dalkin BL, Reid ME, Turnbull BW, Slate EH, Jacobs ET, Marshall JR, Clark LC. Selenium supplementation, baseline plasma selenium status and incidence of prostate cancer: an analysis of the complete treatment period of the Nutritional Prevention of Cancer Trial. BJU international 2003, 91(7):608-612.

49. Ellinger S, Ellinger J, Stehle P. Tomatoes, tomato products and lycopene in the prevention and treatment of prostate cancer: do we have the evidence from intervention studies? Current opinion in clinical nutrition and metabolic care 2006, 9(6):722-727.

50. Kavanaugh CJ, Trumbo PR, Ellwood KC. The U.S. Food and Drug Administration's evidence-based review for qualified health claims: tomatoes, lycopene, and cancer. Journal of the National Cancer Institute 2007, 99(14):1074-1085.

51. Lippman SM, Klein EA, Goodman PJ, Lucia MS, Thompson IM, Ford LG, Parnes HL, Minasian LM, Gaziano JM, Hartline JA et al. Effect of selenium and vitamin E on risk of prostate cancer and other cancers: the Selenium and Vitamin E Cancer Prevention Trial (SELECT). Jama 2009, 301(1):39-51.

52. Deeb KK, Trump DL, Johnson CS. Vitamin D signalling pathways in cancer: potential for anticancer therapeutics. Nature reviews Cancer 2007, 7(9):684-700.

53. Schwartz GG. Vitamin D in blood and risk of prostate cancer: lessons from the Selenium and Vitamin E Cancer Prevention Trial and the Prostate Cancer Prevention Trial. Cancer epidemiology, biomarkers & prevention 2014, 23(8):1447-1449.

54. Gilbert R, Martin RM, Beynon R, Harris R, Savovic J, Zuccolo L, Bekkering GE, Fraser WD, Sterne JA, Metcalfe C. Associations of circulating and dietary vitamin D with prostate cancer risk: a systematic review and dose-response meta-analysis. Cancer causes & control : CCC 2011, 22(3):319-340.

     

    171

55. Heinlein CA, Chang C. Androgen receptor in prostate cancer. Endocrine reviews 2004, 25(2):276-308.

56. Roy AK, Lavrovsky Y, Song CS, Chen S, Jung MH, Velu NK, Bi BY, Chatterjee B. Regulation of androgen action. Vitamins and hormones 1999, 55:309-352.

57. Thompson IM, Goodman PJ, Tangen CM, Lucia MS, Miller GJ, Ford LG, Lieber MM, Cespedes RD, Atkins JN, Lippman SM et al. The influence of finasteride on the development of prostate cancer. The New England journal of medicine 2003, 349(3):215-224.

58. Andriole G, Bostwick D, Brawley O, Gomella L, Marberger M, Tindall D, Breed S, Somerville M, Rittmaster R. Chemoprevention of prostate cancer in men at high risk: rationale and design of the reduction by dutasteride of prostate cancer events (REDUCE) trial. The Journal of urology 2004, 172(4 Pt 1):1314-1317.

59. Theoret MR, Ning YM, Zhang JJ, Justice R, Keegan P, Pazdur R. The risks and benefits of 5alpha-reductase inhibitors for prostate-cancer prevention. The New England journal of medicine 2011, 365(2):97-99.

60. Thompson IM, Jr., Goodman PJ, Tangen CM, Parnes HL, Minasian LM, Godley PA, Lucia MS, Ford LG. Long-term survival of participants in the prostate cancer prevention trial. The New England journal of medicine 2013, 369(7):603-610.

61. Carter BS, Bova GS, Beaty TH, Steinberg GD, Childs B, Isaacs WB, Walsh PC. Hereditary prostate cancer: epidemiologic and clinical features. The Journal of urology 1993, 150(3):797-802.

62. Lesko SM, Rosenberg L, Shapiro S. Family history and prostate cancer risk. American journal of epidemiology 1996, 144(11):1041-1047.

63. Narod SA, Dupont A, Cusan L, Diamond P, Gomez JL, Suburu R, Labrie F. The impact of family history on early detection of prostate cancer. Nature medicine 1995, 1(2):99-101.

64. Steinberg GD, Carter BS, Beaty TH, Childs B, Walsh PC. Family history and the risk of prostate cancer. The Prostate 1990, 17(4):337-347.

65. Smith JR, Freije D, Carpten JD, Gronberg H, Xu J, Isaacs SD, Brownstein MJ, Bova GS, Guo H, Bujnovszky P et al. Major susceptibility locus for prostate cancer on chromosome 1 suggested by a genome-wide search. Science 1996, 274(5291):1371-1374.

66. Cooney KA, McCarthy JD, Lange E, Huang L, Miesfeldt S, Montie JE, Oesterling JE, Sandler HM, Lange K. Prostate cancer susceptibility locus on chromosome 1q: a confirmatory study. Journal of the National Cancer Institute 1997, 89(13):955-959.

67. Xu J, Zheng SL, Chang B, Smith JR, Carpten JD, Stine OC, Isaacs SD, Wiley KE, Henning L, Ewing C et al. Linkage of prostate cancer susceptibility loci to chromosome 1. Human genetics 2001, 108(4):335-345.

68. McIndoe RA, Stanford JL, Gibbs M, Jarvik GP, Brandzel S, Neal CL, Li S, Gammack JT, Gay AA, Goode EL et al. Linkage analysis of 49 high-risk families does not support a common familial prostate cancer-susceptibility gene at 1q24-25. American journal of human genetics 1997, 61(2):347-353.

69. Suarez BK, Lin J, Burmester JK, Broman KW, Weber JL, Banerjee TK, Goddard KA, Witte JS, Elston RC, Catalona WJ. A genome screen of multiplex sibships with prostate cancer. American journal of human genetics 2000, 66(3):933-944.

70. Rokman A, Ikonen T, Seppala EH, Nupponen N, Autio V, Mononen N, Bailey-Wilson J, Trent J, Carpten J, Matikainen MP et al. Germline alterations of the RNASEL gene, a candidate HPC1 gene at 1q25, in patients and families with prostate cancer. American journal of human genetics 2002, 70(5):1299-1304.

71. Carpten J, Nupponen N, Isaacs S, Sood R, Robbins C, Xu J, Faruque M, Moses T, Ewing C, Gillanders E et al. Germline mutations in the ribonuclease L gene in families showing linkage with HPC1. Nature genetics 2002, 30(2):181-184.

72. Casey G, Neville PJ, Plummer SJ, Xiang Y, Krumroy LM, Klein EA, Catalona WJ, Nupponen N, Carpten JD, Trent JM et al. RNASEL Arg462Gln variant is implicated in up to 13% of prostate cancer cases. Nature genetics 2002, 32(4):581-583.

     

    172

73. Tulinius H, Olafsdottir GH, Sigvaldason H, Tryggvadottir L, Bjarnadottir K. Neoplastic diseases in families of breast cancer patients. J Med Genet 1994, 31(8):618-621.

74. Sellers TA, Potter JD, Rich SS, Drinkard CR, Bostick RM, Kushi LH, Zheng W, Folsom AR. Familial clustering of breast and prostate cancers and risk of postmenopausal breast cancer. Journal of the National Cancer Institute 1994, 86(24):1860-1865.

75. Leongamornlert D, Mahmud N, Tymrakiewicz M, Saunders E, Dadaev T, Castro E, Goh C, Govindasami K, Guy M, O'Brien L et al. Germline BRCA1 mutations increase prostate cancer risk. British journal of cancer 2012, 106(10):1697-1701.

76. Kote-Jarai Z, Leongamornlert D, Saunders E, Tymrakiewicz M, Castro E, Mahmud N, Guy M, Edwards S, O'Brien L, Sawyer E et al. BRCA2 is a moderate penetrance gene contributing to young-onset prostate cancer: implications for genetic testing in prostate cancer patients. British journal of cancer 2011, 105(8):1230-1234.

77. Thompson D, Easton DF. Cancer Incidence in BRCA1 mutation carriers. Journal of the National Cancer Institute 2002, 94(18):1358-1365.

78. Cancer risks in BRCA2 mutation carriers. Journal of the National Cancer Institute 1999, 91(15):1310-1316.

79. Castro E, Eeles R. The role of BRCA1 and BRCA2 in prostate cancer. Asian journal of andrology 2012, 14(3):409-414.

80. Castro E, Goh C, Olmos D, Saunders E, Leongamornlert D, Tymrakiewicz M, Mahmud N, Dadaev T, Govindasami K, Guy M et al. Germline BRCA mutations are associated with higher risk of nodal involvement, distant metastasis, and poor survival outcomes in prostate cancer. Journal of clinical oncology 2013, 31(14):1748-1757.

81. Gallagher DJ, Gaudet MM, Pal P, Kirchhoff T, Balistreri L, Vora K, Bhatia J, Stadler Z, Fine SW, Reuter V et al. Germline BRCA mutations denote a clinicopathologic subset of prostate cancer. Clinical cancer research 2010, 16(7):2115-2121.

82. Mitra A, Fisher C, Foster CS, Jameson C, Barbachanno Y, Bartlett J, Bancroft E, Doherty R, Kote-Jarai Z, Peock S et al. Prostate cancer in male BRCA1 and BRCA2 mutation carriers has a more aggressive phenotype. British journal of cancer 2008, 98(2):502-507.

83. Tryggvadottir L, Vidarsdottir L, Thorgeirsson T, Jonasson JG, Olafsdottir EJ, Olafsdottir GH, Rafnar T, Thorlacius S, Jonsson E, Eyfjord JE et al. Prostate cancer progression and survival in BRCA2 mutation carriers. Journal of the National Cancer Institute 2007, 99(12):929-935.

84. Bancroft EK, Page EC, Castro E, Lilja H, Vickers A, Sjoberg D, Assel M, Foster CS, Mitchell G, Drew K et al. Targeted Prostate Cancer Screening in BRCA1 and BRCA2 Mutation Carriers: Results from the Initial Screening Round of the IMPACT Study. European urology 2014, 66(3):489-99.

85. Ewing CM, Ray AM, Lange EM, Zuhlke KA, Robbins CM, Tembe WD, Wiley KE, Isaacs SD, Johng D, Wang Y et al. Germline mutations in HOXB13 and prostate-cancer risk. The New England journal of medicine 2012, 366(2):141-149.

86. Huang H, Cai B. G84E mutation in HOXB13 is firmly associated with prostate cancer risk: a meta-analysis. Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine 2014, 35(2):1177-1182.

87. Smith SC, Palanisamy N, Zuhlke KA, Johnson AM, Siddiqui J, Chinnaiyan AM, Kunju LP, Cooney KA, Tomlins SA. HOXB13 G84E-related familial prostate cancers: a clinical, histologic, and molecular survey. The American journal of surgical pathology 2014, 38(5):615-626.

88. Xu J, Lange EM, Lu L, Zheng SL, Wang Z, Thibodeau SN, Cannon-Albright LA, Teerlink CC, Camp NJ, Johnson AM et al. HOXB13 is a susceptibility gene for prostate cancer: results from the International Consortium for Prostate Cancer Genetics (ICPCG). Human genetics 2013, 132(1):5-14.

89. Zuhlke KA, Johnson AM, Tomlins SA, Palanisamy N, Carpten JD, Lange EM, Isaacs WB, Cooney KA. Identification of a novel germline SPOP mutation in a family with hereditary prostate cancer. The Prostate 2014, 74(9):983-990.

     

    173

90. Berthon P, Valeri A, Cohen-Akenine A, Drelon E, Paiss T, Wohr G, Latil A, Millasseau P, Mellah I, Cohen N et al. Predisposing gene for early-onset prostate cancer, localized on chromosome 1q42.2-43. American journal of human genetics 1998, 62(6):1416-1424.

91. Xu J, Meyers D, Freije D, Isaacs S, Wiley K, Nusskern D, Ewing C, Wilkens E, Bujnovszky P, Bova GS et al. Evidence for a prostate cancer susceptibility locus on the X chromosome. Nature genetics 1998, 20(2):175-179.

92. Berry R, Schroeder JJ, French AJ, McDonnell SK, Peterson BJ, Cunningham JM, Thibodeau SN, Schaid DJ. Evidence for a prostate cancer-susceptibility locus on chromosome 20. American journal of human genetics 2000, 67(1):82-91.

93. Rebbeck TR, Walker AH, Zeigler-Johnson C, Weisburg S, Martin AM, Nathanson KL, Wein AJ, Malkowicz SB. Association of HPC2/ELAC2 genotypes and prostate cancer. American journal of human genetics 2000, 67(4):1014-1019.

94. Dejager S, Mietus-Snyder M, Friera A, Pitas RE. Dominant negative mutations of the scavenger receptor. Native receptor inactivation by expression of truncated variants. The Journal of clinical investigation 1993, 92(2):894-902.

95. Demuth I, Frappart PO, Hildebrand G, Melchers A, Lobitz S, Stockl L, Varon R, Herceg Z, Sperling K, Wang ZQ et al. An inducible null mutant murine model of Nijmegen breakage syndrome proves the essential function of NBS1 in chromosomal stability and cell viability. Human molecular genetics 2004, 13(20):2385-2397.

96. Dong X, Wang L, Taniguchi K, Wang X, Cunningham JM, McDonnell SK, Qian C, Marks AF, Slager SL, Peterson BJ et al. Mutations in CHEK2 associated with prostate cancer risk. American journal of human genetics 2003, 72(2):270-280.

97. Gibbs M, Stanford JL, McIndoe RA, Jarvik GP, Kolb S, Goode EL, Chakrabarti L, Schuster EF, Buckley VA, Miller EL et al. Evidence for a rare prostate cancer-susceptibility locus at chromosome 1p36. American journal of human genetics 1999, 64(3):776-787.

98. Akamatsu S, Takata R, Haiman CA, Takahashi A, Inoue T, Kubo M, Furihata M, Kamatani N, Inazawa J, Chen GK et al. Common variants at 11q12, 10q26 and 3p11.2 are associated with prostate cancer susceptibility in Japanese. Nature genetics 2012, 44(4):426-429.

99. Amin Al Olama A, Kote-Jarai Z, Schumacher FR, Wiklund F, Berndt SI, Benlloch S, Giles GG, Severi G, Neal DE, Hamdy FC et al. A meta-analysis of genome-wide association studies to identify prostate cancer susceptibility loci associated with aggressive and non-aggressive disease. Human molecular genetics 2013, 22(2):408-415.

100. Eeles RA, Olama AA, Benlloch S, Saunders EJ, Leongamornlert DA, Tymrakiewicz M, Ghoussaini M, Luccarini C, Dennis J, Jugurnauth-Little S et al. Identification of 23 new prostate cancer susceptibility loci using the iCOGS custom genotyping array. Nature genetics 2013, 45(4):385-391, 391e381-382.

101. Goh CL, Schumacher FR, Easton D, Muir K, Henderson B, Kote-Jarai Z, Eeles RA. Genetic variants associated with predisposition to prostate cancer and potential clinical implications. Journal of internal medicine 2012, 271(4):353-365.

102. Gudmundsson J, Sulem P, Gudbjartsson DF, Blondal T, Gylfason A, Agnarsson BA, Benediktsdottir KR, Magnusdottir DN, Orlygsdottir G, Jakobsdottir M et al. Genome-wide association and replication studies identify four variants associated with prostate cancer susceptibility. Nature genetics 2009, 41(10):1122-1126.

103. Gudmundsson J, Sulem P, Gudbjartsson DF, Masson G, Agnarsson BA, Benediktsdottir KR, Sigurdsson A, Magnusson OT, Gudjonsson SA, Magnusdottir DN et al. A study based on whole-genome sequencing yields a rare variant at 8q24 associated with prostate cancer. Nature genetics 2012, 44(12):1326-1329.

104. Xu J, Mo Z, Ye D, Wang M, Liu F, Jin G, Xu C, Wang X, Shao Q, Chen Z et al. Genome-wide association study in Chinese men identifies two new prostate cancer risk loci at 9q31.2 and 19q13.4. Nature genetics 2012, 44(11):1231-1235.

105. Xu J, Zheng SL, Isaacs SD, Wiley KE, Wiklund F, Sun J, Kader AK, Li G, Purcell LD, Kim ST et al. Inherited genetic variant predisposes to aggressive but not indolent prostate cancer.

     

    174

Proceedings of the National Academy of Sciences of the United States of America 2010, 107(5):2136-2140.

106. Eeles RA, Kote-Jarai Z, Giles GG, Olama AA, Guy M, Jugurnauth SK, Mulholland S, Leongamornlert DA, Edwards SM, Morrison J et al. Multiple newly identified loci associated with prostate cancer susceptibility. Nature genetics 2008, 40(3):316-321.

107. Al Olama AA, Kote-Jarai Z, Giles GG, Guy M, Morrison J, Severi G, Leongamornlert DA, Tymrakiewicz M, Jhavar S, Saunders E et al. Multiple loci on 8q24 associated with prostate cancer susceptibility. Nature genetics 2009, 41(10):1058-1060.

108. Yeager M, Chatterjee N, Ciampa J, Jacobs KB, Gonzalez-Bosquet J, Hayes RB, Kraft P, Wacholder S, Orr N, Berndt S et al. Identification of a new prostate cancer susceptibility locus on chromosome 8q24. Nature genetics 2009, 41(10):1055-1057.

109. Freedman ML, Haiman CA, Patterson N, McDonald GJ, Tandon A, Waliszewska A, Penney K, Steen RG, Ardlie K, John EM et al. Admixture mapping identifies 8q24 as a prostate cancer risk locus in African-American men. Proceedings of the National Academy of Sciences of the United States of America 2006, 103(38):14068-14073.

110. Kader AK, Liss MA, Trottier G, Kim ST, Sun J, Zheng SL, Chadwick K, Lockwood G, Xu J, Fleshner NE. Impact of Prostate-specific Antigen on a Baseline Prostate Cancer Risk Assessment Including Genetic Risk. Urology 2015, 85(1):165-171.

111. Eeles RA, Kote-Jarai Z, Al Olama AA, Giles GG, Guy M, Severi G, Muir K, Hopper JL, Henderson BE, Haiman CA et al. Identification of seven new prostate cancer susceptibility loci through a genome-wide association study. Nature genetics 2009, 41(10):1116-1121.

112. Takata R, Akamatsu S, Kubo M, Takahashi A, Hosono N, Kawaguchi T, Tsunoda T, Inazawa J, Kamatani N, Ogawa O et al. Genome-wide association study identifies five new susceptibility loci for prostate cancer in the Japanese population. Nature genetics 2010, 42(9):751-754.

113. Schumacher FR, Berndt SI, Siddiq A, Jacobs KB, Wang Z, Lindstrom S, Stevens VL, Chen C, Mondul AM, Travis RC et al. Genome-wide association study identifies new prostate cancer susceptibility loci. Human molecular genetics 2011, 20(19):3867-3875.

114. Kote-Jarai Z, Olama AA, Giles GG, Severi G, Schleutker J, Weischer M, Campa D, Riboli E, Key T, Gronberg H et al. Seven prostate cancer susceptibility loci identified by a multi-stage genome-wide association study. Nature genetics 2011, 43(8):785-791.

115. Kote-Jarai Z, Leongamornlert D, Tymrakiewicz M, Field H, Guy M, Al Olama AA, Morrison J, O'Brien L, Wilkinson R, Hall A et al. Mutation analysis of the MSMB gene in familial prostate cancer. British journal of cancer 2010, 102(2):414-418.

116. Thomas G, Jacobs KB, Yeager M, Kraft P, Wacholder S, Orr N, Yu K, Chatterjee N, Welch R, Hutchinson A et al. Multiple loci identified in a genome-wide association study of prostate cancer. Nature genetics 2008, 40(3):310-315.

117. Gudmundsson J, Sulem P, Steinthorsdottir V, Bergthorsson JT, Thorleifsson G, Manolescu A, Rafnar T, Gudbjartsson D, Agnarsson BA, Baker A et al. Two variants on chromosome 17 confer prostate cancer risk, and the one in TCF2 protects against type 2 diabetes. Nature genetics 2007, 39(8):977-983.

118. Heidenreich A, Bellmunt J, Bolla M, Joniau S, Mason M, Matveev V, Mottet N, Schmid HP, van der Kwast T, Wiegel T et al. EAU guidelines on prostate cancer. Part 1: screening, diagnosis, and treatment of clinically localised disease. European urology 2011, 59(1):61-71.

119. Pokorny MR, de Rooij M, Duncan E, Schroder FH, Parkinson R, Barentsz JO, Thompson LC. Prospective study of diagnostic accuracy comparing prostate cancer detection by transrectal ultrasound-guided biopsy versus magnetic resonance (MR) imaging with subsequent MR-guided biopsy in men without previous prostate biopsies. European urology 2014, 66(1):22-29.

120. Damber JE, Aus G. Prostate cancer. Lancet 2008, 371(9625):1710-1721. 121. Riegman PH, Vlietstra RJ, van der Korput JA, Romijn JC, Trapman J. Characterization of the

prostate-specific antigen gene: a novel human kallikrein-like gene. Biochemical and biophysical research communications 1989, 159(1):95-102.

     

    175

122. Diamandis EP, Yousef GM, Clements J, Ashworth LK, Yoshida S, Egelrud T, Nelson PS, Shiosaka S, Little S, Lilja H et al. New nomenclature for the human tissue kallikrein gene family. Clinical chemistry 2000, 46(11):1855-1858.

123. Kim J, Coetzee GA. Prostate specific antigen gene regulation by androgen receptor. Journal of cellular biochemistry 2004, 93(2):233-241.

124. Lilja H, Laurell CB. The predominant protein in human seminal coagulate. Scandinavian journal of clinical and laboratory investigation 1985, 45(7):635-641.

125. Lilja H, Oldbring J, Rannevik G, Laurell CB. Seminal vesicle-secreted proteins and their reactions during gelation and liquefaction of human semen. The Journal of clinical investigation 1987, 80(2):281-285.

126. Thorek DL, Evans MJ, Carlsson SV, Ulmert D, Lilja H. Prostate-specific kallikrein-related peptidases and their relation to prostate cancer biology and detection. Established relevance and emerging roles. Thrombosis and haemostasis 2013, 110(3):484-492.

127. Pretlow TG, Pretlow TP, Yang B, Kaetzel CS, Delmoro CM, Kamis SM, Bodner DR, Kursh E, Resnick MI, Bradley EL, Jr. Tissue concentrations of prostate-specific antigen in prostatic carcinoma and benign prostatic hyperplasia. International journal of cancer 1991, 49(5):645-649.

128. Magklara A, Scorilas A, Stephan C, Kristiansen GO, Hauptmann S, Jung K, Diamandis EP. Decreased concentrations of prostate-specific antigen and human glandular kallikrein 2 in malignant versus nonmalignant prostatic tissue. Urology 2000, 56(3):527-532.

129. Kulasingam V, Diamandis EP. Strategies for discovering novel cancer biomarkers through utilization of emerging technologies. Nature clinical practice Oncology 2008, 5(10):588-599.

130. Freedland SJ, Sutter ME, Dorey F, Aronson WJ. Defining the ideal cutpoint for determining PSA recurrence after radical prostatectomy. Prostate-specific antigen. Urology 2003, 61(2):365-369.

131. Nadler RB, Humphrey PA, Smith DS, Catalona WJ, Ratliff TL. Effect of inflammation and benign prostatic hyperplasia on elevated serum prostate specific antigen levels. The Journal of urology 1995, 154(2 Pt 1):407-413.

132. Ornish D, Weidner G, Fair WR, Marlin R, Pettengill EB, Raisin CJ, Dunn-Emke S, Crutchfield L, Jacobs FN, Barnard RJ et al. Intensive lifestyle changes may affect the progression of prostate cancer. The Journal of urology 2005, 174(3):1065-1069; discussion 1069-1070.

133. Thompson IM, Pauler DK, Goodman PJ, Tangen CM, Lucia MS, Parnes HL, Minasian LM, Ford LG, Lippman SM, Crawford ED et al. Prevalence of prostate cancer among men with a prostate-specific antigen level <=4.0 ng per milliliter. The New England journal of medicine 2004, 350(22):2239-2246.

134. Draisma G, Etzioni R, Tsodikov A, Mariotto A, Wever E, Gulati R, Feuer E, de Koning H. Lead time and overdiagnosis in prostate-specific antigen screening: importance of methods and context. Journal of the National Cancer Institute 2009, 101(6):374-383.

135. Draisma G, Boer R, Otto SJ, van der Cruijsen IW, Damhuis RA, Schroder FH, de Koning HJ. Lead times and overdetection due to prostate-specific antigen screening: estimates from the European Randomized Study of Screening for Prostate Cancer. Journal of the National Cancer Institute 2003, 95(12):868-878.

136. Zappa M, Ciatto S, Bonardi R, Mazzotta A. Overdiagnosis of prostate carcinoma by screening: an estimate based on the results of the Florence Screening Pilot Study. Annals of oncology : official journal of the European Society for Medical Oncology / ESMO 1998, 9(12):1297-1300.

137. Telesca D, Etzioni R, Gulati R. Estimating lead time and overdiagnosis associated with PSA screening from prostate cancer incidence trends. Biometrics 2008, 64(1):10-19.

138. Mistry K, Cable G. Meta-analysis of prostate-specific antigen and digital rectal examination as screening tests for prostate carcinoma. The Journal of the American Board of Family Practice 2003, 16(2):95-101.

139. Schroder FH, van der Cruijsen-Koeter I, de Koning HJ, Vis AN, Hoedemaeker RF, Kranse R. Prostate cancer detection at low prostate specific antigen. The Journal of urology 2000, 163(3):806-812.

     

    176

140. Krumholtz JS, Carvalhal GF, Ramos CG, Smith DS, Thorson P, Yan Y, Humphrey PA, Roehl KA, Catalona WJ. Prostate-specific antigen cutoff of 2.6 ng/mL for prostate cancer screening is associated with favorable pathologic tumor features. Urology 2002, 60(3):469-473.

141. Andriole GL, Crawford ED, Grubb RL, 3rd, Buys SS, Chia D, Church TR, Fouad MN, Gelmann EP, Kvale PA, Reding DJ et al. Mortality results from a randomized prostate-cancer screening trial. The New England journal of medicine 2009, 360(13):1310-1319.

142. Schroder FH, Hugosson J, Roobol MJ, Tammela TL, Ciatto S, Nelen V, Kwiatkowski M, Lujan M, Lilja H, Zappa M et al. Screening and prostate-cancer mortality in a randomized European study. The New England journal of medicine 2009, 360(13):1320-1328.

143. Wolters T, Roobol MJ, Steyerberg EW, van den Bergh RC, Bangma CH, Hugosson J, Ciatto S, Kwiatkowski M, Villers A, Lujan M et al. The effect of study arm on prostate cancer treatment in the large screening trial ERSPC. International journal of cancer 2010, 126(10):2387-2393.

144. Hugosson J, Carlsson S, Aus G, Bergdahl S, Khatami A, Lodding P, Pihl CG, Stranne J, Holmberg E, Lilja H. Mortality results from the Goteborg randomised population-based prostate-cancer screening trial. The Lancet Oncology 2010, 11(8):725-732.

145. Schroder FH, Roobol MJ. ERSPC and PLCO prostate cancer screening studies: what are the differences? European urology 2010, 58(1):46-52.

146. Bell N, Gorber SC, Shane A, Joffres M, Singh H, Dickinson J, Shaw E, Dunfield L, Tonelli M. Recommendations on screening for prostate cancer with the prostate-specific antigen test. Canadian Medical Association journal 2014, 186(16):1225-1234.

147. Catalona WJ, Smith DS, Ornstein DK. Prostate cancer detection in men with serum PSA concentrations of 2.6 to 4.0 ng/mL and benign prostate examination. Enhancement of specificity with free PSA measurements. Jama 1997, 277(18):1452-1455.

148. Makarov DV, Isharwal S, Sokoll LJ, Landis P, Marlow C, Epstein JI, Partin AW, Carter HB, Veltri RW. Pro-prostate-specific antigen measurements in serum and tissue are associated with treatment necessity among men enrolled in expectant management for prostate cancer. Clinical cancer research 2009, 15(23):7316-7321.

149. Loeb S. Prostate cancer: Prostate Health Index--improving screening in men with family history. Nature reviews Urology 2013, 10(9):497-498.

150. Catalona WJ, Partin AW, Sanda MG, Wei JT, Klee GG, Bangma CH, Slawin KM, Marks LS, Loeb S, Broyles DL et al. A multicenter study of [-2]pro-prostate specific antigen combined with prostate specific antigen and free prostate specific antigen for prostate cancer detection in the 2.0 to 10.0 ng/ml prostate specific antigen range. The Journal of urology 2011, 185(5):1650-1655.

151. Vickers AJ, Cronin AM, Aus G, Pihl CG, Becker C, Pettersson K, Scardino PT, Hugosson J, Lilja H. A panel of kallikrein markers can reduce unnecessary biopsy for prostate cancer: data from the European Randomized Study of Prostate Cancer Screening in Goteborg, Sweden. BMC medicine 2008, 6:19.

152. Catalona WJ, Beiser JA, Smith DS. Serum free prostate specific antigen and prostate specific antigen density measurements for predicting cancer in men with prior negative prostatic biopsies. The Journal of urology 1997, 158(6):2162-2167.

153. Catalona WJ, Southwick PC, Slawin KM, Partin AW, Brawer MK, Flanigan RC, Patel A, Richie JP, Walsh PC, Scardino PT et al. Comparison of percent free PSA, PSA density, and age-specific PSA cutoffs for prostate cancer detection and staging. Urology 2000, 56(2):255-260.

154. Carter HB, Pearson JD, Metter EJ, Brant LJ, Chan DW, Andres R, Fozard JL, Walsh PC. Longitudinal evaluation of prostate-specific antigen levels in men with and without prostate disease. Jama 1992, 267(16):2215-2220.

155. D'Amico AV, Chen MH, Roehl KA, Catalona WJ. Preoperative PSA velocity and the risk of death from prostate cancer after radical prostatectomy. The New England journal of medicine 2004, 351(2):125-135.

156. Bussemakers MJ, van Bokhoven A, Verhaegh GW, Smit FP, Karthaus HF, Schalken JA, Debruyne FM, Ru N, Isaacs WB. DD3: a new prostate-specific gene, highly overexpressed in prostate cancer. Cancer research 1999, 59(23):5975-5979.

     

    177

157. van Gils MP, Hessels D, van Hooij O, Jannink SA, Peelen WP, Hanssen SL, Witjes JA, Cornel EB, Karthaus HF, Smits GA et al. The time-resolved fluorescence-based PCA3 test on urinary sediments after digital rectal examination; a Dutch multicenter validation of the diagnostic performance. Clinical cancer research 2007, 13(3):939-943.

158. Thompson IM, Ankerst DP, Chi C, Goodman PJ, Tangen CM, Lucia MS, Feng Z, Parnes HL, Coltman CA, Jr. Assessing prostate cancer risk: results from the Prostate Cancer Prevention Trial. Journal of the National Cancer Institute 2006, 98(8):529-534.

159. Chun FK, de la Taille A, van Poppel H, Marberger M, Stenzl A, Mulders PF, Huland H, Abbou CC, Stillebroer AB, van Gils MP et al. Prostate cancer gene 3 (PCA3): development and internal validation of a novel biopsy nomogram. European urology 2009, 56(4):659-667.

160. Nam RK, Kattan MW, Chin JL, Trachtenberg J, Singal R, Rendon R, Klotz LH, Sugar L, Sherman C, Izawa J et al. Prospective multi-institutional study evaluating the performance of prostate cancer risk calculators. Journal of clinical oncology 2011, 29(22):2959-2964.

161. Gleason DF. Classification of prostatic carcinomas. Cancer Chemother Rep 1966, 50(3):125-128. 162. Epstein JI, Allsbrook WC, Jr., Amin MB, Egevad LL. The 2005 International Society of

Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma. The American journal of surgical pathology 2005, 29(9):1228-1242.

163. Albertsen PC, Hanley JA, Gleason DF, Barry MJ. Competing risk analysis of men aged 55 to 74 years at diagnosis managed conservatively for clinically localized prostate cancer. Jama 1998, 280(11):975-980.

164. Kovi J, Jackson MA, Heshmat MY. Ductal spread in prostatic carcinoma. Cancer 1985, 56(7):1566-1573.

165. McNeal JE, Yemoto CE. Spread of adenocarcinoma within prostatic ducts and acini. Morphologic and clinical correlations. The American journal of surgical pathology 1996, 20(7):802-814.

166. Guo CC, Epstein JI. Intraductal carcinoma of the prostate on needle biopsy: Histologic features and clinical significance. Modern pathology 2006, 19(12):1528-1535.

167. Robinson BD, Epstein JI. Intraductal carcinoma of the prostate without invasive carcinoma on needle biopsy: emphasis on radical prostatectomy findings. The Journal of urology 2010, 184(4):1328-1333.

168. Cheng L, Montironi R, Bostwick DG, Lopez-Beltran A, Berney DM. Staging of prostate cancer. Histopathology 2012, 60(1):87-117.

169. Harnden P, Shelley MD, Coles B, Staffurth J, Mason MD. Should the Gleason grading system for prostate cancer be modified to account for high-grade tertiary components? A systematic review and meta-analysis. The Lancet Oncology 2007, 8(5):411-419.

170. Trudel D, Downes MR, Sykes J, Kron KJ, Trachtenberg J, van der Kwast TH. Prognostic impact of intraductal carcinoma and large cribriform carcinoma architecture after prostatectomy in a contemporary cohort. European Journal of Cancer 2014, 50(9):1610-1616.

171. Heidenreich A, Bastian PJ, Bellmunt J, Bolla M, Joniau S, van der Kwast T, Mason M, Matveev V, Wiegel T, Zattoni F et al. EAU guidelines on prostate cancer. Part II: Treatment of advanced, relapsing, and castration-resistant prostate cancer. European urology 2014, 65(2):467-479.

172. Heidenreich A, Bastian PJ, Bellmunt J, Bolla M, Joniau S, van der Kwast T, Mason M, Matveev V, Wiegel T, Zattoni F et al. EAU guidelines on prostate cancer. part 1: screening, diagnosis, and local treatment with curative intent-update 2013. European urology 2014, 65(1):124-137.

173. D'Amico AV, Whittington R, Malkowicz SB, Schultz D, Schnall M, Tomaszewski JE, Wein A. A multivariate analysis of clinical and pathological factors that predict for prostate specific antigen failure after radical prostatectomy for prostate cancer. The Journal of urology 1995, 154(1):131-138.

174. Epstein JI. Prognostic significance of tumor volume in radical prostatectomy and needle biopsy specimens. The Journal of urology 2011, 186(3):790-797.

175. Klotz L, Emberton M. Management of low risk prostate cancer-active surveillance and focal therapy. Nature reviews Clinical oncology 2014, 11(6):324-334.

     

    178

176. Wilt TJ, Brawer MK, Jones KM, Barry MJ, Aronson WJ, Fox S, Gingrich JR, Wei JT, Gilhooly P, Grob BM et al. Radical prostatectomy versus observation for localized prostate cancer. The New England journal of medicine 2012, 367(3):203-213.

177. Bill-Axelson A, Holmberg L, Garmo H, Rider JR, Taari K, Busch C, Nordling S, Haggman M, Andersson SO, Spangberg A et al. Radical prostatectomy or watchful waiting in early prostate cancer. The New England journal of medicine 2014, 370(10):932-942.

178. Morris WJ, Keyes M, Spadinger I, Kwan W, Liu M, McKenzie M, Pai H, Pickles T, Tyldesley S. Population-based 10-year oncologic outcomes after low-dose-rate brachytherapy for low-risk and intermediate-risk prostate cancer. Cancer 2013, 119(8):1537-1546.

179. Widmark A, Klepp O, Solberg A, Damber JE, Angelsen A, Fransson P, Lund JA, Tasdemir I, Hoyer M, Wiklund F et al. Endocrine treatment, with or without radiotherapy, in locally advanced prostate cancer (SPCG-7/SFUO-3): an open randomised phase III trial. Lancet 2009, 373(9660):301-308.

180. Warde P, Mason M, Ding K, Kirkbride P, Brundage M, Cowan R, Gospodarowicz M, Sanders K, Kostashuk E, Swanson G et al. Combined androgen deprivation therapy and radiation therapy for locally advanced prostate cancer: a randomised, phase 3 trial. Lancet 2011, 378(9809):2104-2111.

181. Wong YN, Ferraldeschi R, Attard G, de Bono J. Evolution of androgen receptor targeted therapy for advanced prostate cancer. Nature reviews Clinical oncology 2014, 11(6):365-376.

182. Schally AV, Kastin AJ, Arimura A. Hypothalamic follicle-stimulating hormone (FSH) and luteinizing hormone (LH)-regulating hormone: structure, physiology, and clinical studies. Fertility and sterility 1971, 22(11):703-721.

183. Wirth MP, Hakenberg OW, Froehner M. Antiandrogens in the treatment of prostate cancer. European urology 2007, 51(2):306-313; discussion 314.

184. Schmitt B, Wilt TJ, Schellhammer PF, DeMasi V, Sartor O, Crawford ED, Bennett CL. Combined androgen blockade with nonsteroidal antiandrogens for advanced prostate cancer: a systematic review. Urology 2001, 57(4):727-732.

185. Silva FC, Silva FM, Goncalves F, Santos A, Kliment J, Whelan P, Oliver T, Antoniou N, Pastidis S, Marques Queimadelos A et al. Locally advanced and metastatic prostate cancer treated with intermittent androgen monotherapy or maximal androgen blockade: results from a randomised phase 3 study by the South European uroncological group. European urology 2014, 66(2):232-239.

186. Attard G, Reid AH, Yap TA, Raynaud F, Dowsett M, Settatree S, Barrett M, Parker C, Martins V, Folkerd E et al. Phase I clinical trial of a selective inhibitor of CYP17, abiraterone acetate, confirms that castration-resistant prostate cancer commonly remains hormone driven. Journal of clinical oncology 2008, 26(28):4563-4571.

187. Attard G, Reid AH, Olmos D, de Bono JS. Antitumor activity with CYP17 blockade indicates that castration-resistant prostate cancer frequently remains hormone driven. Cancer research 2009, 69(12):4937-4940.

188. Ryan CJ, Smith MR, de Bono JS, Molina A, Logothetis CJ, de Souza P, Fizazi K, Mainwaring P, Piulats JM, Ng S et al. Abiraterone in metastatic prostate cancer without previous chemotherapy. The New England journal of medicine 2013, 368(2):138-148.

189. Pageau SC. Denosumab. Monoclonal antibodies 2009, 1(3):210-215. 190. Fizazi K, Carducci M, Smith M, Damiao R, Brown J, Karsh L, Milecki P, Shore N, Rader M,

Wang H et al. Denosumab versus zoledronic acid for treatment of bone metastases in men with castration-resistant prostate cancer: a randomised, double-blind study. Lancet 2011, 377(9768):813-822.

191. Tannock IF, de Wit R, Berry WR, Horti J, Pluzanska A, Chi KN, Oudard S, Theodore C, James ND, Turesson I et al. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. The New England journal of medicine 2004, 351(15):1502-1512.

192. Bergstralh DT, Ting JP. Microtubule stabilizing agents: their molecular signaling consequences and the potential for enhancement by drug combination. Cancer treatment reviews 2006, 32(3):166-179.

     

    179

193. de Bono JS, Oudard S, Ozguroglu M, Hansen S, Machiels JP, Kocak I, Gravis G, Bodrogi I, Mackenzie MJ, Shen L et al. Prednisone plus cabazitaxel or mitoxantrone for metastatic castration-resistant prostate cancer progressing after docetaxel treatment: a randomised open-label trial. Lancet 2010, 376(9747):1147-1154.

194. Tran C, Ouk S, Clegg NJ, Chen Y, Watson PA, Arora V, Wongvipat J, Smith-Jones PM, Yoo D, Kwon A et al. Development of a second-generation antiandrogen for treatment of advanced prostate cancer. Science 2009, 324(5928):787-790.

195. Scher HI, Beer TM, Higano CS, Anand A, Taplin ME, Efstathiou E, Rathkopf D, Shelkey J, Yu EY, Alumkal J et al. Antitumour activity of MDV3100 in castration-resistant prostate cancer: a phase 1-2 study. Lancet 2010, 375(9724):1437-1446.

196. Scher HI, Fizazi K, Saad F, Taplin ME, Sternberg CN, Miller K, de Wit R, Mulders P, Chi KN, Shore ND et al. Increased survival with enzalutamide in prostate cancer after chemotherapy. The New England journal of medicine 2012, 367(13):1187-1197.

197. Parker C, Nilsson S, Heinrich D, Helle SI, O'Sullivan JM, Fossa SD, Chodacki A, Wiechno P, Logue J, Seke M et al. Alpha emitter radium-223 and survival in metastatic prostate cancer. The New England journal of medicine 2013, 369(3):213-223.

198. Kantoff PW, Higano CS, Shore ND, Berger ER, Small EJ, Penson DF, Redfern CH, Ferrari AC, Dreicer R, Sims RB et al. Sipuleucel-T immunotherapy for castration-resistant prostate cancer. The New England journal of medicine 2010, 363(5):411-422.

199. Schellhammer PF, Chodak G, Whitmore JB, Sims R, Frohlich MW, Kantoff PW. Lower baseline prostate-specific antigen is associated with a greater overall survival benefit from sipuleucel-T in the Immunotherapy for Prostate Adenocarcinoma Treatment (IMPACT) trial. Urology 2013, 81(6):1297-1302.

200. Saad F, Gleason DM, Murray R, Tchekmedyian S, Venner P, Lacombe L, Chin JL, Vinholes JJ, Goas JA, Zheng M. Long-term efficacy of zoledronic acid for the prevention of skeletal complications in patients with metastatic hormone-refractory prostate cancer. Journal of the National Cancer Institute 2004, 96(11):879-882.

201. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000, 100(1):57-70. 202. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 2011, 144(5):646-674. 203. Xu Y, Chen SY, Ross KN, Balk SP. Androgens induce prostate cancer cell proliferation through

mammalian target of rapamycin activation and post-transcriptional increases in cyclin D proteins. Cancer research 2006, 66(15):7783-7792.

204. Wang Q, Li W, Zhang Y, Yuan X, Xu K, Yu J, Chen Z, Beroukhim R, Wang H, Lupien M et al. Androgen receptor regulates a distinct transcription program in androgen-independent prostate cancer. Cell 2009, 138(2):245-256.

205. Tan ME, Li J, Xu HE, Melcher K, Yong EL. Androgen receptor: structure, role in prostate cancer and drug discovery. Acta pharmacologica Sinica 2015, 36(1):3-23.

206. Feldman BJ, Feldman D. The development of androgen-independent prostate cancer. Nature reviews Cancer 2001, 1(1):34-45.

207. Gao S, Lee P, Wang H, Gerald W, Adler M, Zhang L, Wang YF, Wang Z. The androgen receptor directly targets the cellular Fas/FasL-associated death domain protein-like inhibitory protein gene to promote the androgen-independent growth of prostate cancer cells. Mol Endocrinol 2005, 19(7):1792-1802.

208. Dehm SM, Tindall DJ. Molecular regulation of androgen action in prostate cancer. Journal of cellular biochemistry 2006, 99(2):333-344.

209. Bhatia-Gaur R, Donjacour AA, Sciavolino PJ, Kim M, Desai N, Young P, Norton CR, Gridley T, Cardiff RD, Cunha GR et al. Roles for Nkx3.1 in prostate development and cancer. Genes & development 1999, 13(8):966-977.

210. Bowen C, Bubendorf L, Voeller HJ, Slack R, Willi N, Sauter G, Gasser TC, Koivisto P, Lack EE, Kononen J et al. Loss of NKX3.1 expression in human prostate cancers correlates with tumor progression. Cancer research 2000, 60(21):6111-6115.

     

    180

211. Abdulkadir SA, Magee JA, Peters TJ, Kaleem Z, Naughton CK, Humphrey PA, Milbrandt J. Conditional loss of Nkx3.1 in adult mice induces prostatic intraepithelial neoplasia. Molecular and cellular biology 2002, 22(5):1495-1503.

212. Kumar-Sinha C, Tomlins SA, Chinnaiyan AM. Recurrent gene fusions in prostate cancer. Nature reviews Cancer 2008, 8(7):497-511.

213. Tomlins SA, Rhodes DR, Perner S, Dhanasekaran SM, Mehra R, Sun XW, Varambally S, Cao X, Tchinda J, Kuefer R et al. Recurrent fusion of TMPRSS2 and ETS transcription factor genes in prostate cancer. Science 2005, 310(5748):644-648.

214. Perner S, Demichelis F, Beroukhim R, Schmidt FH, Mosquera JM, Setlur S, Tchinda J, Tomlins SA, Hofer MD, Pienta KG et al. TMPRSS2:ERG fusion-associated deletions provide insight into the heterogeneity of prostate cancer. Cancer research 2006, 66(17):8337-8341.

215. Tu JJ, Rohan S, Kao J, Kitabayashi N, Mathew S, Chen YT. Gene fusions between TMPRSS2 and ETS family genes in prostate cancer: frequency and transcript variant analysis by RT-PCR and FISH on paraffin-embedded tissues. Modern pathology 2007, 20(9):921-928.

216. Furusato B, Gao CL, Ravindranath L, Chen Y, Cullen J, McLeod DG, Dobi A, Srivastava S, Petrovics G, Sesterhenn IA. Mapping of TMPRSS2-ERG fusions in the context of multi-focal prostate cancer. Modern pathology 2008, 21(2):67-75.

217. Clark JP, Cooper CS. ETS gene fusions in prostate cancer. Nature reviews Urology 2009, 6(8):429-439.

218. St John J, Powell K, Conley-Lacomb MK, Chinni SR. TMPRSS2-ERG Fusion Gene Expression in Prostate Tumor Cells and Its Clinical and Biological Significance in Prostate Cancer Progression. Journal of cancer science & therapy 2012, 4(4):94-101.

219. Cai C, Hsieh CL, Shemshedini L. c-Jun has multiple enhancing activities in the novel cross talk between the androgen receptor and Ets variant gene 1 in prostate cancer. Molecular cancer research 2007, 5(7):725-735.

220. Klezovitch O, Risk M, Coleman I, Lucas JM, Null M, True LD, Nelson PS, Vasioukhin V. A causal role for ERG in neoplastic transformation of prostate epithelium. Proceedings of the National Academy of Sciences of the United States of America 2008, 105(6):2105-2110.

221. Carver BS, Tran J, Gopalan A, Chen Z, Shaikh S, Carracedo A, Alimonti A, Nardella C, Varmeh S, Scardino PT et al. Aberrant ERG expression cooperates with loss of PTEN to promote cancer progression in the prostate. Nature genetics 2009, 41(5):619-624.

222. King JC, Xu J, Wongvipat J, Hieronymus H, Carver BS, Leung DH, Taylor BS, Sander C, Cardiff RD, Couto SS et al. Cooperativity of TMPRSS2-ERG with PI3-kinase pathway activation in prostate oncogenesis. Nature genetics 2009, 41(5):524-526.

223. Cairns P, Okami K, Halachmi S, Halachmi N, Esteller M, Herman JG, Jen J, Isaacs WB, Bova GS, Sidransky D. Frequent inactivation of PTEN/MMAC1 in primary prostate cancer. Cancer research 1997, 57(22):4997-5000.

224. Sun X, Huang J, Homma T, Kita D, Klocker H, Schafer G, Boyle P, Ohgaki H. Genetic alterations in the PI3K pathway in prostate cancer. Anticancer research 2009, 29(5):1739-1743.

225. Clegg NJ, Couto SS, Wongvipat J, Hieronymus H, Carver BS, Taylor BS, Ellwood-Yen K, Gerald WL, Sander C, Sawyers CL. MYC cooperates with AKT in prostate tumorigenesis and alters sensitivity to mTOR inhibitors. PloS one 2011, 6(3):e17449.

226. Barbieri CE, Baca SC, Lawrence MS, Demichelis F, Blattner M, Theurillat JP, White TA, Stojanov P, Van Allen E, Stransky N et al. Exome sequencing identifies recurrent SPOP, FOXA1 and MED12 mutations in prostate cancer. Nature genetics 2012, 44(6):685-689.

227. Grasso CS, Wu YM, Robinson DR, Cao X, Dhanasekaran SM, Khan AP, Quist MJ, Jing X, Lonigro RJ, Brenner JC et al. The mutational landscape of lethal castration-resistant prostate cancer. Nature 2012, 487(7406):239-243.

228. Taylor BS, Schultz N, Hieronymus H, Gopalan A, Xiao Y, Carver BS, Arora VK, Kaushik P, Cerami E, Reva B et al. Integrative genomic profiling of human prostate cancer. Cancer cell 2010, 18(1):11-22.

     

    181

229. Kremer CL, Klein RR, Mendelson J, Browne W, Samadzedeh LK, Vanpatten K, Highstrom L, Pestano GA, Nagle RB. Expression of mTOR signaling pathway markers in prostate cancer progression. The Prostate 2006, 66(11):1203-1212.

230. Di Cristofano A, De Acetis M, Koff A, Cordon-Cardo C, Pandolfi PP. Pten and p27KIP1 cooperate in prostate cancer tumor suppression in the mouse. Nature genetics 2001, 27(2):222-224.

231. Carver BS, Chapinski C, Wongvipat J, Hieronymus H, Chen Y, Chandarlapaty S, Arora VK, Le C, Koutcher J, Scher H et al. Reciprocal feedback regulation of PI3K and androgen receptor signaling in PTEN-deficient prostate cancer. Cancer cell 2011, 19(5):575-586.

232. Mulholland DJ, Tran LM, Li Y, Cai H, Morim A, Wang S, Plaisier S, Garraway IP, Huang J, Graeber TG et al. Cell autonomous role of PTEN in regulating castration-resistant prostate cancer growth. Cancer cell 2011, 19(6):792-804.

233. Abreu-Martin MT, Chari A, Palladino AA, Craft NA, Sawyers CL. Mitogen-activated protein kinase kinase kinase 1 activates androgen receptor-dependent transcription and apoptosis in prostate cancer. Molecular and cellular biology 1999, 19(7):5143-5154.

234. Gioeli D, Mandell JW, Petroni GR, Frierson HF, Jr., Weber MJ. Activation of mitogen-activated protein kinase associated with prostate cancer progression. Cancer research 1999, 59(2):279-284.

235. Uzgare AR, Isaacs JT. Enhanced redundancy in Akt and mitogen-activated protein kinase-induced survival of malignant versus normal prostate epithelial cells. Cancer research 2004, 64(17):6190-6199.

236. Gao H, Ouyang X, Banach-Petrosky WA, Gerald WL, Shen MM, Abate-Shen C. Combinatorial activities of Akt and B-Raf/Erk signaling in a mouse model of androgen-independent prostate cancer. Proceedings of the National Academy of Sciences of the United States of America 2006, 103(39):14477-14482.

237. Drake JM, Graham NA, Stoyanova T, Sedghi A, Goldstein AS, Cai H, Smith DA, Zhang H, Komisopoulou E, Huang J et al. Oncogene-specific activation of tyrosine kinase networks during prostate cancer progression. Proceedings of the National Academy of Sciences of the United States of America 2012, 109(5):1643-1648.

238. Shaw A, Bushman W. Hedgehog signaling in the prostate. The Journal of urology 2007, 177(3):832-838.

239. Bisson I, Prowse DM. WNT signaling regulates self-renewal and differentiation of prostate cancer cells with stem cell characteristics. Cell research 2009, 19(6):683-697.

240. Carvalho FL, Simons BW, Eberhart CG, Berman DM. Notch signaling in prostate cancer: a moving target. The Prostate 2014, 74(9):933-945.

241. Meeks JJ, Schaeffer EM. Genetic regulation of prostate development. Journal of andrology 2011, 32(3):210-217.

242. Park JI, Lee MG, Cho K, Park BJ, Chae KS, Byun DS, Ryu BK, Park YK, Chi SG. Transforming growth factor-beta1 activates interleukin-6 expression in prostate cancer cells through the synergistic collaboration of the Smad2, p38-NF-kappaB, JNK, and Ras signaling pathways. Oncogene 2003, 22(28):4314-4332.

243. Stephenson RA, Dinney CP, Gohji K, Ordonez NG, Killion JJ, Fidler IJ. Metastatic model for human prostate cancer using orthotopic implantation in nude mice. Journal of the National Cancer Institute 1992, 84(12):951-957.

244. Yan G, Fukabori Y, McBride G, Nikolaropolous S, McKeehan WL. Exon switching and activation of stromal and embryonic fibroblast growth factor (FGF)-FGF receptor genes in prostate epithelial cells accompany stromal independence and malignancy. Molecular and cellular biology 1993, 13(8):4513-4522.

245. Olumi AF, Grossfeld GD, Hayward SW, Carroll PR, Tlsty TD, Cunha GR. Carcinoma-associated fibroblasts direct tumor progression of initiated human prostatic epithelium. Cancer research 1999, 59(19):5002-5011.

246. Waddington CH. The epigenotype. Endeavour 1942, 1:18-20. 247. Jablonka E, Lamb MJ. The changing concept of epigenetics. Annals of the New York Academy

of Sciences 2002, 981:82-96.

     

    182

248. Richards EJ. Inherited epigenetic variation--revisiting soft inheritance. Nature reviews Genetics 2006, 7(5):395-401.

249. Jeronimo C, Bastian PJ, Bjartell A, Carbone GM, Catto JW, Clark SJ, Henrique R, Nelson WG, Shariat SF. Epigenetics in prostate cancer: biologic and clinical relevance. European urology 2011, 60(4):753-766.

250. Braiteh F, Soriano AO, Garcia-Manero G, Hong D, Johnson MM, Silva Lde P, Yang H, Alexander S, Wolff J, Kurzrock R. Phase I study of epigenetic modulation with 5-azacytidine and valproic acid in patients with advanced cancers. Clinical cancer research 2008, 14(19):6296-6301.

251. Rasheed W, Bishton M, Johnstone RW, Prince HM. Histone deacetylase inhibitors in lymphoma and solid malignancies. Expert review of anticancer therapy 2008, 8(3):413-432.

252. Jones PA. At the tipping point for epigenetic therapies in cancer. The Journal of clinical investigation 2014, 124(1):14-16.

253. Woodcock DM, Crowther PJ, Diver WP. The majority of methylated deoxycytidines in human DNA are not in the CpG dinucleotide. Biochemical and biophysical research communications 1987, 145(2):888-894.

254. Clark SJ, Harrison J, Frommer M. CpNpG methylation in mammalian cells. Nature genetics 1995, 10(1):20-27.

255. Bird A, Taggart M, Frommer M, Miller OJ, Macleod D. A fraction of the mouse genome that is derived from islands of nonmethylated, CpG-rich DNA. Cell 1985, 40(1):91-99.

256. Gardiner-Garden M, Frommer M. CpG islands in vertebrate genomes. Journal of molecular biology 1987, 196(2):261-282.

257. Holliday R, Pugh JE. DNA modification mechanisms and gene activity during development. Science 1975, 187(4173):226-232.

258. Riggs AD. X inactivation, differentiation, and DNA methylation. Cytogenetics and cell genetics 1975, 14(1):9-25.

259. Peters J. The role of genomic imprinting in biology and disease: an expanding view. Nature reviews Genetics 2014, 15(8):517-530.

260. Boyes J, Bird A. DNA methylation inhibits transcription indirectly via a methyl-CpG binding protein. Cell 1991, 64(6):1123-1134.

261. Clark SJ, Harrison J, Molloy PL. Sp1 binding is inhibited by (m)Cp(m)CpG methylation. Gene 1997, 195(1):67-71.

262. Nan X, Ng HH, Johnson CA, Laherty CD, Turner BM, Eisenman RN, Bird A. Transcriptional repression by the methyl-CpG-binding protein MeCP2 involves a histone deacetylase complex. Nature 1998, 393(6683):386-389.

263. Yang X, Han H, De Carvalho DD, Lay FD, Jones PA, Liang G. Gene body methylation can alter gene expression and is a therapeutic target in cancer. Cancer cell 2014, 26(4):577-590.

264. Moarefi AH, Chedin F. ICF syndrome mutations cause a broad spectrum of biochemical defects in DNMT3B-mediated de novo DNA methylation. Journal of molecular biology 2011, 409(5):758-772.

265. Yoder JA, Walsh CP, Bestor TH. Cytosine methylation and the ecology of intragenomic parasites. Trends in genetics 1997, 13(8):335-340.

266. Bell AC, Felsenfeld G. Methylation of a CTCF-dependent boundary controls imprinted expression of the Igf2 gene. Nature 2000, 405(6785):482-485.

267. Irizarry RA, Ladd-Acosta C, Wen B, Wu Z, Montano C, Onyango P, Cui H, Gabo K, Rongione M, Webster M et al. The human colon cancer methylome shows similar hypo- and hypermethylation at conserved tissue-specific CpG island shores. Nat Genet 2009, 41(2):178-186.

268. Doi A, Park IH, Wen B, Murakami P, Aryee MJ, Irizarry R, Herb B, Ladd-Acosta C, Rho J, Loewer S et al. Differential methylation of tissue- and cancer-specific CpG island shores distinguishes human induced pluripotent stem cells, embryonic stem cells and fibroblasts. Nat Genet 2009, 41(12):1350-1353.

     

    183

269. Okano M, Bell DW, Haber DA, Li E. DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell 1999, 99(3):247-257.

270. Kim GD, Ni J, Kelesoglu N, Roberts RJ, Pradhan S. Co-operation and communication between the human maintenance and de novo DNA (cytosine-5) methyltransferases. The EMBO journal 2002, 21(15):4183-4195.

271. Valinluck V, Sowers LC. Endogenous cytosine damage products alter the site selectivity of human DNA maintenance methyltransferase DNMT1. Cancer research 2007, 67(3):946-950.

272. Tahiliani M, Koh KP, Shen Y, Pastor WA, Bandukwala H, Brudno Y, Agarwal S, Iyer LM, Liu DR, Aravind L et al. Conversion of 5-methylcytosine to 5-hydroxymethylcytosine in mammalian DNA by MLL partner TET1. Science 2009, 324(5929):930-935.

273. Cortellino S, Xu J, Sannai M, Moore R, Caretti E, Cigliano A, Le Coz M, Devarajan K, Wessels A, Soprano D et al. Thymine DNA glycosylase is essential for active DNA demethylation by linked deamination-base excision repair. Cell 2011, 146(1):67-79.

274. Ito S, Shen L, Dai Q, Wu SC, Collins LB, Swenberg JA, He C, Zhang Y. Tet proteins can convert 5-methylcytosine to 5-formylcytosine and 5-carboxylcytosine. Science 2011, 333(6047):1300-1303.

275. Yang B, Sun H, Lin W, Hou W, Li H, Zhang L, Li F, Gu Y, Song Y, Li Q et al. Evaluation of global DNA hypomethylation in human prostate cancer and prostatic intraepithelial neoplasm tissues by immunohistochemistry. Urologic oncology 2013, 31(5):628-634.

276. Cho NY, Kim JH, Moon KC, Kang GH. Genomic hypomethylation and CpG island hypermethylation in prostatic intraepithelial neoplasm. Virchows Arch 2009, 454(1):17-23.

277. Yegnasubramanian S, Haffner MC, Zhang Y, Gurel B, Cornish TC, Wu Z, Irizarry RA, Morgan J, Hicks J, DeWeese TL et al. DNA hypomethylation arises later in prostate cancer progression than CpG island hypermethylation and contributes to metastatic tumor heterogeneity. Cancer research 2008, 68(21):8954-8967.

278. Santourlidis S, Florl A, Ackermann R, Wirtz HC, Schulz WA. High frequency of alterations in DNA methylation in adenocarcinoma of the prostate. The Prostate 1999, 39(3):166-174.

279. Florl AR, Steinhoff C, Muller M, Seifert HH, Hader C, Engers R, Ackermann R, Schulz WA. Coordinate hypermethylation at specific genes in prostate carcinoma precedes LINE-1 hypomethylation. British journal of cancer 2004, 91(5):985-994.

280. Jarrard DF, Bussemakers MJ, Bova GS, Isaacs WB. Regional loss of imprinting of the insulin-like growth factor II gene occurs in human prostate tissues. Clinical cancer research 1995, 1(12):1471-1478.

281. Cho B, Lee H, Jeong S, Bang YJ, Lee HJ, Hwang KS, Kim HY, Lee YS, Kang GH, Jeoung DI. Promoter hypomethylation of a novel cancer/testis antigen gene CAGE is correlated with its aberrant expression and is seen in premalignant stage of gastric carcinoma. Biochemical and biophysical research communications 2003, 307(1):52-63.

282. Tokizane T, Shiina H, Igawa M, Enokida H, Urakami S, Kawakami T, Ogishima T, Okino ST, Li LC, Tanaka Y et al. Cytochrome P450 1B1 is overexpressed and regulated by hypomethylation in prostate cancer. Clinical cancer research 2005, 11(16):5793-5801.

283. Ogishima T, Shiina H, Breault JE, Tabatabai L, Bassett WW, Enokida H, Li LC, Kawakami T, Urakami S, Ribeiro-Filho LA et al. Increased heparanase expression is caused by promoter hypomethylation and up-regulation of transcriptional factor early growth response-1 in human prostate cancer. Clinical cancer research 2005, 11(3):1028-1036.

284. Pakneshan P, Xing RH, Rabbani SA. Methylation status of uPA promoter as a molecular mechanism regulating prostate cancer invasion and growth in vitro and in vivo. Federation of American Societies for Experimental Biology journal 2003, 17(9):1081-1088.

285. Schulz WA, Elo JP, Florl AR, Pennanen S, Santourlidis S, Engers R, Buchardt M, Seifert HH, Visakorpi T. Genomewide DNA hypomethylation is associated with alterations on chromosome 8 in prostate carcinoma. Genes, chromosomes & cancer 2002, 35(1):58-65.

286. Gaudet F, Hodgson JG, Eden A, Jackson-Grusby L, Dausman J, Gray JW, Leonhardt H, Jaenisch R. Induction of tumors in mice by genomic hypomethylation. Science 2003, 300(5618):489-492.

     

    184

287. Patra SK, Patra A, Zhao H, Dahiya R. DNA methyltransferase and demethylase in human prostate cancer. Molecular carcinogenesis 2002, 33(3):163-171.

288. Gillio-Tos A, Fiano V, Zugna D, Vizzini L, Pearce N, Delsedime L, Merletti F, Richiardi L. DNA methyltransferase 3b (DNMT3b), tumor tissue DNA methylation, Gleason score, and prostate cancer mortality: investigating causal relationships. Cancer causes & control : CCC 2012, 23(9):1549-1555.

289. Gravina GL, Ranieri G, Muzi P, Marampon F, Mancini A, Di Pasquale B, Di Clemente L, Dolo V, D'Alessandro AM, Festuccia C. Increased levels of DNA methyltransferases are associated with the tumorigenic capacity of prostate cancer cells. Oncology reports 2013, 29(3):1189-1195.

290. Maruyama R, Toyooka S, Toyooka KO, Virmani AK, Zochbauer-Muller S, Farinas AJ, Minna JD, McConnell J, Frenkel EP, Gazdar AF. Aberrant promoter methylation profile of prostate cancers and its relationship to clinicopathological features. Clinical cancer research 2002, 8(2):514-519.

291. Liu L, Kron KJ, Pethe VV, Demetrashvili N, Nesbitt ME, Trachtenberg J, Ozcelik H, Fleshner NE, Briollais L, van der Kwast TH et al. Association of tissue promoter methylation levels of APC, TGFbeta2, HOXD3 and RASSF1A with prostate cancer progression. International journal of cancer 2011, 129(10):2454-2462.

292. Kron K, Liu L, Trudel D, Pethe V, Trachtenberg J, Fleshner N, Bapat B, van der Kwast T. Correlation of ERG expression and DNA methylation biomarkers with adverse clinicopathologic features of prostate cancer. Clinical cancer research 2012, 18(10):2896-2904.

293. Lee WH, Morton RA, Epstein JI, Brooks JD, Campbell PA, Bova GS, Hsieh WS, Isaacs WB, Nelson WG. Cytidine methylation of regulatory sequences near the pi-class glutathione S-transferase gene accompanies human prostatic carcinogenesis. Proceedings of the National Academy of Sciences of the United States of America 1994, 91(24):11733-11737.

294. Hopkins TG, Burns PA, Routledge MN. DNA methylation of GSTP1 as biomarker in diagnosis of prostate cancer. Urology 2007, 69(1):11-16.

295. Cairns P, Esteller M, Herman JG, Schoenberg M, Jeronimo C, Sanchez-Cespedes M, Chow NH, Grasso M, Wu L, Westra WB et al. Molecular detection of prostate cancer in urine by GSTP1 hypermethylation. Clinical cancer research 2001, 7(9):2727-2730.

296. Bastian PJ, Palapattu GS, Lin X, Yegnasubramanian S, Mangold LA, Trock B, Eisenberger MA, Partin AW, Nelson WG. Preoperative serum DNA GSTP1 CpG island hypermethylation and the risk of early prostate-specific antigen recurrence following radical prostatectomy. Clinical cancer research 2005, 11(11):4037-4043.

297. Mahon KL, Qu W, Devaney J, Paul C, Castillo L, Wykes RJ, Chatfield MD, Boyer MJ, Stockler MR, Marx G et al. Methylated Glutathione S-transferase 1 (mGSTP1) is a potential plasma free DNA epigenetic marker of prognosis and response to chemotherapy in castrate-resistant prostate cancer. British journal of cancer 2014, 111(9):1802-1809.

298. Di Pietro G, Magno LA, Rios-Santos F. Glutathione S-transferases: an overview in cancer research. Expert opinion on drug metabolism & toxicology 2010, 6(2):153-170.

299. Brooks JD, Weinstein M, Lin X, Sun Y, Pin SS, Bova GS, Epstein JI, Isaacs WB, Nelson WG. CG island methylation changes near the GSTP1 gene in prostatic intraepithelial neoplasia. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology 1998, 7(6):531-536.

300. Ushijima T. Epigenetic field for cancerization. Journal of biochemistry and molecular biology 2007, 40(2):142-150.

301. Stewart GD, Van Neste L, Delvenne P, Delree P, Delga A, McNeill SA, O'Donnell M, Clark J, Van Criekinge W, Bigley J et al. Clinical utility of an epigenetic assay to detect occult prostate cancer in histopathologically negative biopsies: results of the MATLOC study. The Journal of urology 2013, 189(3):1110-1116.

302. Wojno KJ, Costa FJ, Cornell RJ, Small JD, Pasin E, Van Criekinge W, Bigley JW, Van Neste L. Reduced Rate of Repeated Prostate Biopsies Observed in ConfirmMDx Clinical Utility Field Study. American health & drug benefits 2014, 7(3):129-134.

     

    185

303. Mikeska T, Bock C, Do H, Dobrovic A. DNA methylation biomarkers in cancer: progress towards clinical implementation. Expert review of molecular diagnostics 2012, 12(5):473-487.

304. Wu TL, Zhang D, Chia JH, Tsao K, Sun CF, Wu JT. Cell-free DNA: measurement in various carcinomas and establishment of normal reference range. Clinica chimica acta; international journal of clinical chemistry 2002, 321(1-2):77-87.

305. Ellinger J, Haan K, Heukamp LC, Kahl P, Buttner R, Muller SC, von Ruecker A, Bastian PJ. CpG island hypermethylation in cell-free serum DNA identifies patients with localized prostate cancer. The Prostate 2008, 68(1):42-49.

306. Scher HI, Jia X, de Bono JS, Fleisher M, Pienta KJ, Raghavan D, Heller G. Circulating tumour cells as prognostic markers in progressive, castration-resistant prostate cancer: a reanalysis of IMMC38 trial data. The Lancet Oncology 2009, 10(3):233-239.

307. Ahmed H. Promoter methylation in prostate cancer and its application for the early detection of prostate cancer using serum and urine samples. Biomarkers in cancer 2010, 2:17-33.

308. Kouzarides T. Chromatin modifications and their function. Cell 2007, 128(4):693-705. 309. Varambally S, Dhanasekaran SM, Zhou M, Barrette TR, Kumar-Sinha C, Sanda MG, Ghosh D,

Pienta KJ, Sewalt RG, Otte AP et al. The polycomb group protein EZH2 is involved in progression of prostate cancer. Nature 2002, 419(6907):624-629.

310. Yu J, Yu J, Rhodes DR, Tomlins SA, Cao X, Chen G, Mehra R, Wang X, Ghosh D, Shah RB et al. A polycomb repression signature in metastatic prostate cancer predicts cancer outcome. Cancer research 2007, 67(22):10657-10663.

311. Simon JA, Lange CA. Roles of the EZH2 histone methyltransferase in cancer epigenetics. Mutation research 2008, 647(1-2):21-29.

312. Ke XS, Qu Y, Rostad K, Li WC, Lin B, Halvorsen OJ, Haukaas SA, Jonassen I, Petersen K, Goldfinger N et al. Genome-wide profiling of histone h3 lysine 4 and lysine 27 trimethylation reveals an epigenetic signature in prostate carcinogenesis. PloS one 2009, 4(3):e4687.

313. Yu J, Yu J, Mani RS, Cao Q, Brenner CJ, Cao X, Wang X, Wu L, Li J, Hu M et al. An integrated network of androgen receptor, polycomb, and TMPRSS2-ERG gene fusions in prostate cancer progression. Cancer cell 2010, 17(5):443-454.

314. Kunderfranco P, Mello-Grand M, Cangemi R, Pellini S, Mensah A, Albertini V, Malek A, Chiorino G, Catapano CV, Carbone GM. ETS transcription factors control transcription of EZH2 and epigenetic silencing of the tumor suppressor gene Nkx3.1 in prostate cancer. PloS one 2010, 5(5):e10547.

315. Varambally S, Cao Q, Mani RS, Shankar S, Wang X, Ateeq B, Laxman B, Cao X, Jing X, Ramnarayanan K et al. Genomic loss of microRNA-101 leads to overexpression of histone methyltransferase EZH2 in cancer. Science 2008, 322(5908):1695-1699.

316. Chen H, Tu SW, Hsieh JT. Down-regulation of human DAB2IP gene expression mediated by polycomb Ezh2 complex and histone deacetylase in prostate cancer. The Journal of biological chemistry 2005, 280(23):22437-22444.

317. Xie D, Gore C, Liu J, Pong RC, Mason R, Hao G, Long M, Kabbani W, Yu L, Zhang H et al. Role of DAB2IP in modulating epithelial-to-mesenchymal transition and prostate cancer metastasis. Proceedings of the National Academy of Sciences of the United States of America 2010, 107(6):2485-2490.

318. Hoffmann MJ, Engers R, Florl AR, Otte AP, Muller M, Schulz WA. Expression changes in EZH2, but not in BMI-1, SIRT1, DNMT1 or DNMT3B are associated with DNA methylation changes in prostate cancer. Cancer biology & therapy 2007, 6(9):1403-1412.

319. Draker R, Sarcinella E, Cheung P. USP10 deubiquitylates the histone variant H2A.Z and both are required for androgen receptor-mediated gene activation. Nucleic acids research 2011, 39(9):3529-3542.

320. Esteller M. Non-coding RNAs in human disease. Nature reviews Genetics 2011, 12(12):861-874. 321. He L, Hannon GJ. MicroRNAs: small RNAs with a big role in gene regulation. Nature reviews

Genetics 2004, 5(7):522-531. 322. Mendell JT. MicroRNAs: critical regulators of development, cellular physiology and

malignancy. Cell Cycle 2005, 4(9):1179-1184.

     

    186

323. Friedman RC, Farh KK, Burge CB, Bartel DP. Most mammalian mRNAs are conserved targets of microRNAs. Genome research 2009, 19(1):92-105.

324. Kozomara A, Griffiths-Jones S. miRBase: integrating microRNA annotation and deep-sequencing data. Nucleic acids research 2011, 39:D152-157.

325. Chen CZ, Li L, Lodish HF, Bartel DP. MicroRNAs modulate hematopoietic lineage differentiation. Science 2004, 303(5654):83-86.

326. Krol J, Loedige I, Filipowicz W. The widespread regulation of microRNA biogenesis, function and decay. Nature reviews Genetics 2010, 11(9):597-610.

327. Li SC, Tang P, Lin WC. Intronic microRNA: discovery and biological implications. DNA and cell biology 2007, 26(4):195-207.

328. Rodriguez A, Griffiths-Jones S, Ashurst JL, Bradley A. Identification of mammalian microRNA host genes and transcription units. Genome research 2004, 14(10A):1902-1910.

329. Huntzinger E, Izaurralde E. Gene silencing by microRNAs: contributions of translational repression and mRNA decay. Nature reviews Genetics 2011, 12(2):99-110.

330. Ioffe YJ, Chiappinelli KB, Mutch DG, Zighelboim I, Goodfellow PJ. Phosphatase and tensin homolog (PTEN) pseudogene expression in endometrial cancer: a conserved regulatory mechanism important in tumorigenesis? Gynecologic oncology 2012, 124(2):340-346.

331. Franco-Zorrilla JM, Valli A, Todesco M, Mateos I, Puga MI, Rubio-Somoza I, Leyva A, Weigel D, Garcia JA, Paz-Ares J. Target mimicry provides a new mechanism for regulation of microRNA activity. Nature genetics 2007, 39(8):1033-1037.

332. Poliseno L, Salmena L, Zhang J, Carver B, Haveman WJ, Pandolfi PP. A coding-independent function of gene and pseudogene mRNAs regulates tumour biology. Nature 2010, 465(7301):1033-1038.

333. Cesana M, Cacchiarelli D, Legnini I, Santini T, Sthandier O, Chinappi M, Tramontano A, Bozzoni I. A long noncoding RNA controls muscle differentiation by functioning as a competing endogenous RNA. Cell 2011, 147(2):358-369.

334. Duursma AM, Kedde M, Schrier M, le Sage C, Agami R. miR-148 targets human DNMT3b protein coding region. RNA 2008, 14(5):872-877.

335. Noonan EJ, Place RF, Pookot D, Basak S, Whitson JM, Hirata H, Giardina C, Dahiya R. miR-449a targets HDAC-1 and induces growth arrest in prostate cancer. Oncogene 2009, 28(14):1714-1724.

336. Braconi C, Huang N, Patel T. MicroRNA-dependent regulation of DNA methyltransferase-1 and tumor suppressor gene expression by interleukin-6 in human malignant cholangiocytes. Hepatology 2010, 51(3):881-890.

337. Cannistraci A, Di Pace AL, De Maria R, Bonci D. MicroRNA as New Tools for Prostate Cancer Risk Assessment and Therapeutic Intervention: Results from Clinical Data Set and Patients' Samples. BioMed research international 2014, 2014: Article ID 146170.

338. Volinia S, Calin GA, Liu CG, Ambs S, Cimmino A, Petrocca F, Visone R, Iorio M, Roldo C, Ferracin M et al. A microRNA expression signature of human solid tumors defines cancer gene targets. Proceedings of the National Academy of Sciences of the United States of America 2006, 103(7):2257-2261.

339. Porkka KP, Pfeiffer MJ, Waltering KK, Vessella RL, Tammela TL, Visakorpi T. MicroRNA expression profiling in prostate cancer. Cancer research 2007, 67(13):6130-6135.

340. Ambs S, Prueitt RL, Yi M, Hudson RS, Howe TM, Petrocca F, Wallace TA, Liu CG, Volinia S, Calin GA et al. Genomic profiling of microRNA and messenger RNA reveals deregulated microRNA expression in prostate cancer. Cancer research 2008, 68(15):6162-6170.

341. Ozen M, Creighton CJ, Ozdemir M, Ittmann M. Widespread deregulation of microRNA expression in human prostate cancer. Oncogene 2008, 27(12):1788-1793.

342. Prueitt RL, Yi M, Hudson RS, Wallace TA, Howe TM, Yfantis HG, Lee DH, Stephens RM, Liu CG, Calin GA et al. Expression of microRNAs and protein-coding genes associated with perineural invasion in prostate cancer. The Prostate 2008, 68(11):1152-1164.

     

    187

343. Tong AW, Fulgham P, Jay C, Chen P, Khalil I, Liu S, Senzer N, Eklund AC, Han J, Nemunaitis J. MicroRNA profile analysis of human prostate cancers. Cancer gene therapy 2009, 16(3):206-216.

344. Schaefer A, Jung M, Mollenkopf HJ, Wagner I, Stephan C, Jentzmik F, Miller K, Lein M, Kristiansen G, Jung K. Diagnostic and prognostic implications of microRNA profiling in prostate carcinoma. International journal of cancer 2010, 126(5):1166-1176.

345. Szczyrba J, Loprich E, Wach S, Jung V, Unteregger G, Barth S, Grobholz R, Wieland W, Stohr R, Hartmann A et al. The microRNA profile of prostate carcinoma obtained by deep sequencing. Molecular cancer research 2010, 8(4):529-538.

346. Carlsson J, Davidsson S, Helenius G, Karlsson M, Lubovac Z, Andren O, Olsson B, Klinga-Levan K. A miRNA expression signature that separates between normal and malignant prostate tissues. Cancer cell international 2011, 11(1):14.

347. Hart M, Nolte E, Wach S, Szczyrba J, Taubert H, Rau TT, Hartmann A, Grasser FA, Wullich B. Comparative microRNA profiling of prostate carcinomas with increasing tumor stage by deep sequencing. Molecular cancer research 2014, 12(2):250-263.

348. Watahiki A, Macfarlane RJ, Gleave ME, Crea F, Wang Y, Helgason CD, Chi KN. Plasma miRNAs as Biomarkers to Identify Patients with Castration-Resistant Metastatic Prostate Cancer. International journal of molecular sciences 2013, 14(4):7757-7770.

349. Nguyen HC, Xie W, Yang M, Hsieh CL, Drouin S, Lee GS, Kantoff PW. Expression differences of circulating microRNAs in metastatic castration resistant prostate cancer and low-risk, localized prostate cancer. The Prostate 2013, 73(4):346-354.

350. Bryant RJ, Pawlowski T, Catto JW, Marsden G, Vessella RL, Rhees B, Kuslich C, Visakorpi T, Hamdy FC. Changes in circulating microRNA levels associated with prostate cancer. British journal of cancer 2012, 106(4):768-774.

351. Szczyrba J, Nolte E, Wach S, Kremmer E, Stohr R, Hartmann A, Wieland W, Wullich B, Grasser FA. Downregulation of Sec23A protein by miRNA-375 in prostate carcinoma. Molecular cancer research 2011, 9(6):791-800.

352. Peterson SM, Thompson JA, Ufkin ML, Sathyanarayana P, Liaw L, Congdon CB. Common features of microRNA target prediction tools. Frontiers in genetics 2014, 5: Article 23.

353. Huang Y, Zou Q, Song H, Song F, Wang L, Zhang G, Shen X. A study of miRNAs targets prediction and experimental validation. Protein & cell 2010, 1(11):979-986.

354. Leyten GH, Hessels D, Jannink SA, Smit FP, de Jong H, Cornel EB, de Reijke TM, Vergunst H, Kil P, Knipscheer BC et al. Prospective multicentre evaluation of PCA3 and TMPRSS2-ERG gene fusions as diagnostic and prognostic urinary biomarkers for prostate cancer. European urology 2014, 65(3):534-542.

355. Berney DM, Gopalan A, Kudahetti S, Fisher G, Ambroisine L, Foster CS, Reuter V, Eastham J, Moller H, Kattan MW et al. Ki-67 and outcome in clinically localised prostate cancer: analysis of conservatively treated prostate cancer patients from the Trans-Atlantic Prostate Group study. British journal of cancer 2009, 100(6):888-893.

356. Lotan TL, Gurel B, Sutcliffe S, Esopi D, Liu W, Xu J, Hicks JL, Park BH, Humphreys E, Partin AW et al. PTEN protein loss by immunostaining: analytic validation and prognostic indicator for a high risk surgical cohort of prostate cancer patients. Clinical cancer research 2011, 17(20):6563-6573.

357. Ribeiro FR, Henrique R, Martins AT, Jeronimo C, Teixeira MR. Relative copy number gain of MYC in diagnostic needle biopsies is an independent prognostic factor for prostate cancer patients. European urology 2007, 52(1):116-125.

358. Kudahetti S, Fisher G, Ambroisine L, Foster C, Reuter V, Eastham J, Moller H, Kattan MW, Cooper CS, Scardino P et al. p53 immunochemistry is an independent prognostic marker for outcome in conservatively treated prostate cancer. BJU international 2009, 104(1):20-24.

359. Ding Z, Wu CJ, Chu GC, Xiao Y, Ho D, Zhang J, Perry SR, Labrot ES, Wu X, Lis R et al. SMAD4-dependent barrier constrains prostate cancer growth and metastatic progression. Nature 2011, 470(7333):269-273.

     

    188

360. Markert EK, Mizuno H, Vazquez A, Levine AJ. Molecular classification of prostate cancer using curated expression signatures. Proceedings of the National Academy of Sciences of the United States of America 2011, 108(52):21276-21281.

361. Diamandis EP, Yousef GM, Luo LY, Magklara A, Obiezu CV. The new human kallikrein gene family: implications in carcinogenesis. Trends in endocrinology and metabolism 2000, 11(2):54-60.

362. Yousef GM, Diamandis EP. The expanded human kallikrein gene family: locus characterization and molecular cloning of a new member, KLK-L3 (KLK9). Genomics 2000, 65(2):184-194.

363. Harvey TJ, Hooper JD, Myers SA, Stephenson SA, Ashworth LK, Clements JA. Tissue-specific expression patterns and fine mapping of the human kallikrein (KLK) locus on proximal 19q13.4. The Journal of biological chemistry 2000, 275(48):37397-37406.

364. Kurlender L, Borgono C, Michael IP, Obiezu C, Elliott MB, Yousef GM, Diamandis EP. A survey of alternative transcripts of human tissue kallikrein genes. Biochimica et biophysica acta 2005, 1755(1):1-14.

365. Tan OL, Whitbread AK, Clements JA, Dong Y. Kallikrein-related peptidase (KLK) family mRNA variants and protein isoforms in hormone-related cancers: do they have a function? Biological chemistry 2006, 387(6):697-705.

366. Sotiropoulou G, Pampalakis G, Diamandis EP. Functional roles of human kallikrein-related peptidases. The Journal of biological chemistry 2009, 284(48):32989-32994.

367. Michael IP, Pampalakis G, Mikolajczyk SD, Malm J, Sotiropoulou G, Diamandis EP. Human tissue kallikrein 5 is a member of a proteolytic cascade pathway involved in seminal clot liquefaction and potentially in prostate cancer progression. The Journal of biological chemistry 2006, 281(18):12743-12750.

368. Kishibe M, Bando Y, Terayama R, Namikawa K, Takahashi H, Hashimoto Y, Ishida-Yamamoto A, Jiang YP, Mitrovic B, Perez D et al. Kallikrein 8 is involved in skin desquamation in cooperation with other kallikreins. The Journal of biological chemistry 2007, 282(8):5834-5841.

369. Emami N, Diamandis EP. Human kallikrein-related peptidase 14 (KLK14) is a new activator component of the KLK proteolytic cascade. Possible function in seminal plasma and skin. The Journal of biological chemistry 2008, 283(6):3031-3041.

370. Eissa A, Amodeo V, Smith CR, Diamandis EP. Kallikrein-related peptidase-8 (KLK8) is an active serine protease in human epidermis and sweat and is involved in a skin barrier proteolytic cascade. The Journal of biological chemistry 2011, 286(1):687-706.

371. Saraswati S, Block AS, Davidson MK, Rank RG, Mahadevan M, Diekman AB. Galectin-3 is a substrate for prostate specific antigen (PSA) in human seminal plasma. The Prostate 2011, 71(2):197-208.

372. Emami N, Diamandis EP. New insights into the functional mechanisms and clinical applications of the kallikrein-related peptidase family. Molecular oncology 2007, 1(3):269-287.

373. Lawrence MG, Lai J, Clements JA. Kallikreins on steroids: structure, function, and hormonal regulation of prostate-specific antigen and the extended kallikrein locus. Endocrine reviews 2010, 31(4):407-446.

374. Mills IH. Kallikrein, kininogen and kinins in control of blood pressure. Nephron 1979, 23(2-3):61-71.

375. Lovgren J, Rajakoski K, Karp M, Lundwall a, Lilja H. Activation of the zymogen form of prostate-specific antigen by human glandular kallikrein 2. Biochemical and biophysical research communications 1997, 238(2):549-555.

376. Takayama TK, McMullen BA, Nelson PS, Matsumura M, Fujikawa K. Characterization of hK4 (prostase), a prostate-specific serine protease: activation of the precursor of prostate specific antigen (pro-PSA) and single-chain urokinase-type plasminogen activator and degradation of prostatic acid phosphatase. Biochemistry 2001, 40(50):15341-15348.

377. Yousef GM, Diamandis EP. The new human tissue kallikrein gene family: structure, function, and association to disease. Endocrine reviews 2001, 22(2):184-204.

     

    189

378. Bayani J, Marrano P, Graham C, Zheng Y, Li L, Katsaros D, Lassus H, Butzow R, Squire JA, Diamandis EP. Genomic instability and copy-number heterogeneity of chromosome 19q, including the kallikrein locus, in ovarian carcinomas. Molecular oncology 2011, 5(1):48-60.

379. Girgis AH, Bui A, White NM, Yousef GM. Integrated genomic characterization of the kallikrein gene locus in cancer. Anticancer research 2012, 32(3):957-963.

380. Lindahl P, Sall T, Bjartell A, Johansson AM, Lilja H, Hallden C. Copy number variants in the kallikrein gene cluster. PloS one 2013, 8(7):e69097.

381. Goard CA, Bromberg IL, Elliott MB, Diamandis EP. A consolidated catalogue and graphical annotation of dbSNP polymorphisms in the human tissue kallikrein (KLK) locus. Molecular oncology 2007, 1(3):303-312.

382. Sukumar N, Scott E, Dimitromanolakis A, Misiak A, Prassas I, Diamandis EP, Konvalinka A. Mining for single nucleotide variants (SNVs) at the kallikrein locus with predicted functional consequences. Biological chemistry 2014, 395(9):1037-1050.

383. Chow TF, Crow M, Earle T, El-Said H, Diamandis EP, Yousef GM. Kallikreins as microRNA targets: an in silico and experimental-based analysis. Biological chemistry 2008, 389(6):731-738.

384. Yousef GM. microRNAs: a new frontier in kallikrein research. Biological chemistry 2008, 389(6):689-694.

385. White NM, Bui A, Mejia-Guerrero S, Chao J, Soosaipillai A, Youssef Y, Mankaruos M, Honey RJ, Stewart R, Pace KT et al. Dysregulation of kallikrein-related peptidases in renal cell carcinoma: potential targets of miRNAs. Biological chemistry 2010, 391(4):411-423.

386. White NM, Chow TF, Mejia-Guerrero S, Diamandis M, Rofael Y, Faragalla H, Mankaruous M, Gabril M, Girgis A, Yousef GM. Three dysregulated miRNAs control kallikrein 10 expression and cell proliferation in ovarian cancer. British journal of cancer 2010, 102(8):1244-1253.

387. White NM, Youssef YM, Fendler A, Stephan C, Jung K, Yousef GM. The miRNA-kallikrein axis of interaction: a new dimension in the pathogenesis of prostate cancer. Biological chemistry 2012, 393(5):379-389.

388. Borgono CA, Diamandis EP. The emerging roles of human tissue kallikreins in cancer. Nature reviews Cancer 2004, 4(11):876-890.

389. Schechter NM, Choi EJ, Wang ZM, Hanakawa Y, Stanley JR, Kang Y, Clayman GL, Jayakumar A. Inhibition of human kallikreins 5 and 7 by the serine protease inhibitor lympho-epithelial Kazal-type inhibitor (LEKTI). Biological chemistry 2005, 386(11):1173-1184.

390. Yousef GM, Obiezu CV, Luo LY, Magklara A, Borgono CA, Kishi T, Memari N, Michael l P, Sidiropoulos M, Kurlender L et al. Human tissue kallikreins: from gene structure to function and clinical applications. Advances in clinical chemistry 2005, 39:11-79.

391. Luo LY, Jiang W. Inhibition profiles of human tissue kallikreins by serine protease inhibitors. Biological chemistry 2006, 387(6):813-816.

392. Avgeris M, Mavridis K, Scorilas A. Kallikrein-related peptidase genes as promising biomarkers for prognosis and monitoring of human malignancies. Biological chemistry 2010, 391(5):505-511.

393. Kontos CK, Scorilas A. Kallikrein-related peptidases (KLKs): a gene family of novel cancer biomarkers. Clinical chemistry and laboratory medicine 2012, 50(11):1877-1891.

394. Schmitt M, Magdolen V, Yang F, Kiechle M, Bayani J, Yousef GM, Scorilas A, Diamandis EP, Dorn J. Emerging clinical importance of the cancer biomarkers kallikrein-related peptidases (KLK) in female and male reproductive organ malignancies. Radiology and oncology 2013, 47(4):319-329.

395. Obiezu CV, Soosaipillai A, Jung K, Stephan C, Scorilas A, Howarth DH, Diamandis EP. Detection of human kallikrein 4 in healthy and cancerous prostatic tissues by immunofluorometry and immunohistochemistry. Clinical chemistry 2002, 48(8):1232-1240.

396. Yousef GM, Stephan C, Scorilas A, Ellatif MA, Jung K, Kristiansen G, Jung M, Polymeris ME, Diamandis EP. Differential expression of the human kallikrein gene 14 (KLK14) in normal and cancerous prostatic tissues. The Prostate 2003, 56(4):287-292.

     

    190

397. Stephan C, Yousef GM, Scorilas A, Jung K, Jung M, Kristiansen G, Hauptmann S, Bharaj BS, Nakamura T, Loening SA et al. Quantitative analysis of kallikrein 15 gene expression in prostate tissue. The Journal of urology 2003, 169(1):361-364.

398. Nakamura T, Stephan C, Scorilas A, Yousef GM, Jung K, Diamandis EP. Quantitative analysis of hippostasin/KLK11 gene expression in cancerous and noncancerous prostatic tissues. Urology 2003, 61(5):1042-1046.

399. Pampalakis G, Diamandis EP, Sotiropoulou G. The epigenetic basis for the aberrant expression of kallikreins in human cancers. Biological chemistry 2006, 387(6):795-799.

400. Pampalakis G, Sotiropoulou G. Multiple mechanisms underlie the aberrant expression of the human kallikrein 6 gene in breast cancer. Biological chemistry 2006, 387(6):773-782.

401. Pampalakis G, Prosnikli E, Agalioti T, Vlahou A, Zoumpourlis V, Sotiropoulou G. A tumor-protective role for human kallikrein-related peptidase 6 in breast cancer mediated by inhibition of epithelial-to-mesenchymal transition. Cancer research 2009, 69(9):3779-3787.

402. Chou RH, Lin SC, Wen HC, Wu CW, Chang WS. Epigenetic activation of human kallikrein 13 enhances malignancy of lung adenocarcinoma by promoting N-cadherin expression and laminin degradation. Biochemical and biophysical research communications 2011, 409(3):442-447.

403. Wong J, Sia YY, Misso NL, Aggarwal S, Ng A, Bhoola KD. Effects of the demethylating agent, 5-azacytidine, on expression of the kallikrein-kinin genes in carcinoma cells of the lung and pleura. Pathology research international 2011, 2011: Article ID 167046.

404. Li B, Goyal J, Dhar S, Dimri G, Evron E, Sukumar S, Wazer DE, Band V. CpG methylation as a basis for breast tumor-specific loss of NES1/kallikrein 10 expression. Cancer Research 2001, 61(21):8014-8021.

405. Huang W, Zhong J, Wu LY, Yu LF, Tian XL, Zhang YF, Li B. Downregulation and CpG island hypermethylation of NES1/hK10 gene in the pathogenesis of human gastric cancer. Cancer letters 2007, 251(1):78-85.

406. Kioulafa M, Kaklamanis L, Stathopoulos E, Mavroudis D, Georgoulias V, Lianidou ES. Kallikrein 10 (KLK10) methylation as a novel prognostic biomarker in early breast cancer. Annals of oncology 2009, 20(6):1020-1025.

407. Sidiropoulos M, Pampalakis G, Sotiropoulou G, Katsaros D, Diamandis EP. Downregulation of human kallikrein 10 (KLK10/NES1) by CpG island hypermethylation in breast, ovarian and prostate cancers. Tumour biology 2005, 26(6):324-336.

408. Roman-Gomez J, Jimenez-Velasco A, Agirre X, Castillejo JA, Barrios M, Andreu EJ, Prosper F, Heiniger A, Torres A. The normal epithelial cell-specific 1 (NES1) gene, a candidate tumor suppressor gene on chromosome 19q13.3-4, is downregulated by hypermethylation in acute lymphoblastic leukemia. Leukemia 2004, 18(2):362-365.

409. Zhang Y, Song H, Miao Y, Wang R, Chen L. Frequent transcriptional inactivation of Kallikrein 10 gene by CpG island hypermethylation in non-small cell lung cancer. Cancer science 2010, 101(4):934-940.

410. Ashour N, Angulo JC, Andres G, Alelu R, Gonzalez-Corpas A, Toledo MV, Rodriguez-Barbero JM, Lopez JI, Sanchez-Chapado M, Ropero S. A DNA hypermethylation profile reveals new potential biomarkers for prostate cancer diagnosis and prognosis. The Prostate 2014, 74(12):1171-1182.

411. Kobayashi Y, Absher DM, Gulzar ZG, Young SR, McKenney JK, Peehl DM, Brooks JD, Myers RM, Sherlock G. DNA methylation profiling reveals novel biomarkers and important roles for DNA methyltransferases in prostate cancer. Genome research 2011, 21(7):1017-1027.

412. Kang Z, Janne OA, Palvimo JJ. Coregulator recruitment and histone modifications in transcriptional regulation by the androgen receptor. Molecular endocrinology 2004, 18(11):2633-2648.

413. Jia L, Shen HC, Wantroba M, Khalid O, Liang G, Wang Q, Gentzschein E, Pinski JK, Stanczyk FZ, Jones PA et al. Locus-wide chromatin remodeling and enhanced androgen receptor-mediated transcription in recurrent prostate tumor cells. Molecular and cellular biology 2006, 26(19):7331-7341.

     

    191

414. Bert SA, Robinson MD, Strbenac D, Statham AL, Song JZ, Hulf T, Sutherland RL, Coolen MW, Stirzaker C, Clark SJ. Regional activation of the cancer genome by long-range epigenetic remodeling. Cancer cell 2013, 23(1):9-22.

415. Herman JG, Graff JR, Myohanen S, Nelkin BD, Baylin SB. Methylation-specific PCR: a novel PCR assay for methylation status of CpG islands. Proceedings of the National Academy of Sciences of the United States of America 1996, 93(18):9821-9826.

416. Paluszczak J, Baer-Dubowska W. Epigenetic diagnostics of cancer--the application of DNA methylation markers. Journal of applied genetics 2006, 47(4):365-375.

417. Heyn H, Esteller M. DNA methylation profiling in the clinic: applications and challenges. Nature reviews Genetics 2012, 13(10):679-692.

418. Borno ST, Fischer A, Kerick M, Falth M, Laible M, Brase JC, Kuner R, Dahl A, Grimm C, Sayanjali B et al. Genome-wide DNA methylation events in TMPRSS2-ERG fusion-negative prostate cancers implicate an EZH2-dependent mechanism with miR-26a hypermethylation. Cancer discovery 2012, 2(11):1024-1035.

419. Kron K, Trudel D, Pethe V, Briollais L, Fleshner N, van der Kwast T, Bapat B. Altered DNA methylation landscapes of polycomb-repressed loci are associated with prostate cancer progression and ERG oncogene expression in prostate cancer. Clinical cancer research 2013, 19(13):3450-3461.

420. Pasic MD, Olkhov E, Bapat B, Yousef GM. Epigenetic regulation of kallikrein-related peptidases: there is a whole new world out there. Biological chemistry 2012, 393(5):319-330.

421. Kron K, Pethe V, Briollais L, Sadikovic B, Ozcelik H, Sunderji A, Venkateswaran V, Pinthus J, Fleshner N, van der Kwast T et al. Discovery of novel hypermethylated genes in prostate cancer using genomic CpG island microarrays. PLoS One 2009, 4(3):e4830.

422. Kron KJ, Liu L, Pethe VV, Demetrashvili N, Nesbitt ME, Trachtenberg J, Ozcelik H, Fleshner NE, Briollais L, van der Kwast TH et al. DNA methylation of HOXD3 as a marker of prostate cancer progression. Lab Invest 2010, 90(7):1060-1067.

423. Ehrich M, Bocker S, van den Boom D. Multiplexed discovery of sequence polymorphisms using base-specific cleavage and MALDI-TOF MS. Nucleic acids research 2005, 33(4):e38.

424. Eads CA, Danenberg KD, Kawakami K, Saltz LB, Blake C, Shibata D, Danenberg PV, Laird PW. MethyLight: a high-throughput assay to measure DNA methylation. Nucleic acids research 2000, 28(8):E32.

425. Weisenberger DJ, Campan M, Long TI, Kim M, Woods C, Fiala E, Ehrlich M, Laird PW. Analysis of repetitive element DNA methylation by MethyLight. Nucleic acids research 2005, 33(21):6823-6836.

426. Rosenbaum E, Hoque MO, Cohen Y, Zahurak M, Eisenberger MA, Epstein JI, Partin AW, Sidransky D. Promoter hypermethylation as an independent prognostic factor for relapse in patients with prostate cancer following radical prostatectomy. Clinical cancer research 2005, 11(23):8321-8325.

427. Henrique R, Ribeiro FR, Fonseca D, Hoque MO, Carvalho AL, Costa VL, Pinto M, Oliveira J, Teixeira MR, Sidransky D et al. High promoter methylation levels of APC predict poor prognosis in sextant biopsies from prostate cancer patients. Clinical cancer research 2007, 13(20):6122-6129.

428. Rothman KJ. No adjustments are needed for multiple comparisons. Epidemiology 1990, 1(1):43-46.

429. Furusato B, Tan SH, Young D, Dobi A, Sun C, Mohamed AA, Thangapazham R, Chen Y, McMaster G, Sreenath T et al. ERG oncoprotein expression in prostate cancer: clonal progression of ERG-positive tumor cells and potential for ERG-based stratification. Prostate cancer and prostatic diseases 2010, 13(3):228-237.

430. Zhang S, Pavlovitz B, Tull J, Wang Y, Deng FM, Fuller C. Detection of TMPRSS2 gene deletions and translocations in carcinoma, intraepithelial neoplasia, and normal epithelium of the prostate by direct fluorescence in situ hybridization. Diagnostic molecular pathology 2010, 19(3):151-156.

     

    192

431. Park K, Tomlins SA, Mudaliar KM, Chiu YL, Esgueva R, Mehra R, Suleman K, Varambally S, Brenner JC, MacDonald T et al. Antibody-based detection of ERG rearrangement-positive prostate cancer. Neoplasia 2010, 12(7):590-598.

432. Saramaki OR, Harjula AE, Martikainen PM, Vessella RL, Tammela TL, Visakorpi T. TMPRSS2:ERG fusion identifies a subgroup of prostate cancers with a favorable prognosis. Clinical cancer research 2008, 14(11):3395-3400.

433. Falzarano SM, Navas M, Simmerman K, Klein EA, Rubin MA, Zhou M, Magi-Galluzzi C. ERG rearrangement is present in a subset of transition zone prostatic tumors. Modern pathology 2010, 23(11):1499-1506.

434. Epstein JI, Carmichael M, Walsh PC. Adenocarcinoma of the prostate invading the seminal vesicle: definition and relation of tumor volume, grade and margins of resection to prognosis. The Journal of urology 1993, 149(5):1040-1045.

435. Veltri RW, Miller MC, Partin AW, Coffey DS, Epstein JI. Ability to predict biochemical progression using Gleason score and a computer-generated quantitative nuclear grade derived from cancer cell nuclei. Urology 1996, 48(5):685-691.

436. Christman JK. 5-Azacytidine and 5-aza-2'-deoxycytidine as inhibitors of DNA methylation: mechanistic studies and their implications for cancer therapy. Oncogene 2002, 21(35):5483-5495.

437. Lu CY, Hsieh SY, Lu YJ, Wu CS, Chen LC, Lo SJ, Wu CT, Chou MY, Huang TH, Chang YS. Aberrant DNA methylation profile and frequent methylation of KLK10 and OXGR1 genes in hepatocellular carcinoma. Genes Chromosomes Cancer 2009, 48(12):1057-1068.

438. Jones PA. The DNA methylation paradox. Trends in genetics 1999, 15(1):34-37. 439. Petraki CD, Gregorakis AK, Papanastasiou PA, Karavana VN, Luo LY, Diamandis EP.

Immunohistochemical localization of human kallikreins 6, 10 and 13 in benign and malignant prostatic tissues. Prostate cancer and prostatic diseases 2003, 6(3):223-227.

440. Goyal J, Smith KM, Cowan JM, Wazer DE, Lee SW, Band V. The role for NES1 serine protease as a novel tumor suppressor. Cancer research 1998, 58(21):4782-4786.

441. Yousef GM, Scorilas A, Jung K, Ashworth LK, Diamandis EP. Molecular cloning of the human kallikrein 15 gene (KLK15). Up-regulation in prostate cancer. The Journal of biological chemistry 2001, 276(1):53-61.

442. Mavridis K, Avgeris M, Koutalellis G, Stravodimos K, Scorilas A. Expression analysis and study of the KLK15 mRNA splice variants in prostate cancer and benign prostatic hyperplasia. Cancer science 2010, 101(3):693-699.

443. Rabien A, Fritzsche FR, Jung M, Tolle A, Diamandis EP, Miller K, Jung K, Kristiansen G, Stephan C. KLK15 is a prognostic marker for progression-free survival in patients with radical prostatectomy. International journal of cancer 2010, 127(10):2386-2394.

444. Mavridis K, Stravodimos K, Scorilas A. Quantified KLK15 gene expression levels discriminate prostate cancer from benign tumors and constitute a novel independent predictor of disease progression. The Prostate 2013, 73(11):1191-1201.

445. Shaw JL, Diamandis EP. Distribution of 15 human kallikreins in tissues and biological fluids. Clinical chemistry 2007, 53(8):1423-1432.

446. van der Heul-Nieuwenhuijsen L, Hendriksen PJ, van der Kwast TH, Jenster G. Gene expression profiling of the human prostate zones. BJU international 2006, 98(4):886-897.

447. Jeronimo C, Henrique R, Hoque MO, Mambo E, Ribeiro FR, Varzim G, Oliveira J, Teixeira MR, Lopes C, Sidransky D. A quantitative promoter methylation profile of prostate cancer. Clinical Cancer Research 2004, 10(24):8472-8478.

448. Kwabi-Addo B, Chung W, Shen L, Ittmann M, Wheeler T, Jelinek J, Issa JP. Age-related DNA methylation changes in normal human prostate tissues. Clin Cancer Res 2007, 13(13):3796-3802.

449. Nonn L, Ananthanarayanan V, Gann PH. Evidence for field cancerization of the prostate. The Prostate 2009, 69(13):1470-1479.

450. Slaughter DP, Southwick HW, Smejkal W. Field cancerization in oral stratified squamous epithelium; clinical implications of multicentric origin. Cancer 1953, 6(5):963-968.

     

    193

451. Hanson JA, Gillespie JW, Grover A, Tangrea MA, Chuaqui RF, Emmert-Buck MR, Tangrea JA, Libutti SK, Linehan WM, Woodson KG. Gene promoter methylation in prostate tumor-associated stromal cells. Journal of the National Cancer Institute 2006, 98(4):255-261.

452. Yousef GM, White NM, Michael IP, Cho JC, Robb JD, Kurlender L, Khan S, Diamandis EP. Identification of new splice variants and differential expression of the human kallikrein 10 gene, a candidate cancer biomarker. Tumour Biology 2005, 26(5):227-235.

453. Zong Y, Xin L, Goldstein AS, Lawson DA, Teitell MA, Witte ON. ETS family transcription factors collaborate with alternative signaling pathways to induce carcinoma from adult murine prostate cells. Proceedings of the National Academy of Sciences of the United States of America 2009, 106(30):12465-12470.

454. Vire E, Brenner C, Deplus R, Blanchon L, Fraga M, Didelot C, Morey L, Van Eynde A, Bernard D, Vanderwinden JM et al. The Polycomb group protein EZH2 directly controls DNA methylation. Nature 2006, 439(7078):871-874.

455. Richiardi L, Fiano V, Vizzini L, De Marco L, Delsedime L, Akre O, Tos AG, Merletti F. Promoter methylation in APC, RUNX3, and GSTP1 and mortality in prostate cancer patients. Journal of clinical oncology 2009, 27(19):3161-3168.

456. Phe V, Cussenot O, Roupret M. Methylated genes as potential biomarkers in prostate cancer. BJU International 2010, 105(10):1364-1370.

457. Epstein JI, Partin AW, Sauvageot J, Walsh PC. Prediction of progression following radical prostatectomy. A multivariate analysis of 721 men with long-term follow-up. The American journal of surgical pathology 1996, 20(3):286-292.

458. Cohen RJ, Chan WC, Edgar SG, Robinson E, Dodd N, Hoscek S, Mundy IP. Prediction of pathological stage and clinical outcome in prostate cancer: an improved pre-operative model incorporating biopsy-determined intraductal carcinoma. British journal of urology 1998, 81(3):413-418.

459. Wilcox G, Soh S, Chakraborty S, Scardino PT, Wheeler TM. Patterns of high-grade prostatic intraepithelial neoplasia associated with clinically aggressive prostate cancer. Human pathology 1998, 29(10):1119-1123.

460. Totten RS, Heinemann MW, Hudson PB, Sproul EE, Stout AP. Microscopic differential diagnosis of latent carcinoma of prostate. AMA archives of pathology 1953, 55(2):131-141.

461. Dong F, Yang P, Wang C, Wu S, Xiao Y, McDougal WS, Young RH, Wu CL. Architectural heterogeneity and cribriform pattern predict adverse clinical outcome for Gleason grade 4 prostatic adenocarcinoma. The American journal of surgical pathology 2013, 37(12):1855-1861.

462. Kryvenko ON, Gupta NS, Virani N, Schultz D, Gomez J, Amin A, Lane Z, Epstein JI. Gleason score 7 adenocarcinoma of the prostate with lymph node metastases: analysis of 184 radical prostatectomy specimens. Archives of pathology & laboratory medicine 2013, 137(5):610-617.

463. Kweldam CF, Wildhagen MF, Steyerberg EW, Bangma CH, van der Kwast TH, van Leenders GJ. Cribriform growth is highly predictive for postoperative metastasis and disease-specific death in Gleason score 7 prostate cancer. Modern pathology 2014 advance online publication.

464. Shah RB, Magi-Galluzzi C, Han B, Zhou M. Atypical cribriform lesions of the prostate: relationship to prostatic carcinoma and implication for diagnosis in prostate biopsies. The American journal of surgical pathology 2010, 34(4):470-477.

465. Farin HF, Mansouri A, Petry M, Kispert A. T-box protein Tbx18 interacts with the paired box protein Pax3 in the development of the paraxial mesoderm. The Journal of biological chemistry 2008, 283(37):25372-25380.

466. He TC, Sparks AB, Rago C, Hermeking H, Zawel L, da Costa LT, Morin PJ, Vogelstein B, Kinzler KW. Identification of c-MYC as a target of the APC pathway. Science 1998, 281(5382):1509-1512.

467. Ikushima H, Miyazono K. TGFbeta signalling: a complex web in cancer progression. Nature reviews Cancer 2010, 10(6):415-424.

468. Mayanil CS, George D, Freilich L, Miljan EJ, Mania-Farnell B, McLone DG, Bremer EG. Microarray analysis detects novel Pax3 downstream target genes. The Journal of biological chemistry 2001, 276(52):49299-49309.

     

    194

469. Nathalie HV, Chris P, Serge G, Catherine C, Benjamin B, Claire B, Christelle P, Briollais L, Pascale R, Marie-Lise J et al. High kallikrein-related peptidase 6 in non-small cell lung cancer cells: an indicator of tumour proliferation and poor prognosis. Journal of cellular and molecular medicine 2009, 13(9B):4014-4022.

470. Rezsohazy R. Non-transcriptional interactions of Hox proteins: inventory, facts, and future directions. Developmental dynamics : an official publication of the American Association of Anatomists 2014, 243(1):117-131.

471. Tang XH, Gudas LJ. Retinoids, retinoic acid receptors, and cancer. Annual review of pathology 2011, 6:345-364.

472. Thaler S, Hahnel PS, Schad A, Dammann R, Schuler M. RASSF1A mediates p21Cip1/Waf1-dependent cell cycle arrest and senescence through modulation of the Raf-MEK-ERK pathway and inhibition of Akt. Cancer research 2009, 69(5):1748-1757.

473. Treeck O, Schuler S, Haring J, Skrzypczak M, Lattrich C, Ortmann O. icb-1 Gene counteracts growth of ovarian cancer cell lines. Cancer letters 2013, 335(2):441-446.

474. Fine SW, Gopalan A, Leversha MA, Al-Ahmadie HA, Tickoo SK, Zhou Q, Satagopan JM, Scardino PT, Gerald WL, Reuter VE. TMPRSS2-ERG gene fusion is associated with low Gleason scores and not with high-grade morphological features. Modern pathology 2010, 23(10):1325-1333.

475. Han B, Suleman K, Wang L, Siddiqui J, Sercia L, Magi-Galluzzi C, Palanisamy N, Chinnaiyan AM, Zhou M, Shah RB. ETS gene aberrations in atypical cribriform lesions of the prostate: Implications for the distinction between intraductal carcinoma of the prostate and cribriform high-grade prostatic intraepithelial neoplasia. The American journal of surgical pathology 2010, 34(4):478-485.

476. Lee K, Chae JY, Kwak C, Ku JH, Moon KC. TMPRSS2-ERG gene fusion and clinicopathologic characteristics of Korean prostate cancer patients. Urology 2010, 76(5):1268 e1267-1213.

477. Kir G, Sarbay BC, Gumus E, Topal CS. The association of the cribriform pattern with outcome for prostatic adenocarcinomas. Pathology, research and practice 2014, 210(10):640-644.

478. Brock MV, Hooker CM, Ota-Machida E, Han Y, Guo M, Ames S, Glockner S, Piantadosi S, Gabrielson E, Pridham G et al. DNA methylation markers and early recurrence in stage I lung cancer. The New England journal of medicine 2008, 358(11):1118-1128.

479. Esteller M, Corn PG, Urena JM, Gabrielson E, Baylin SB, Herman JG. Inactivation of glutathione S-transferase P1 gene by promoter hypermethylation in human neoplasia. Cancer research 1998, 58(20):4515-4518.

480. Esteller M, Garcia-Foncillas J, Andion E, Goodman SN, Hidalgo OF, Vanaclocha V, Baylin SB, Herman JG. Inactivation of the DNA-repair gene MGMT and the clinical response of gliomas to alkylating agents. The New England journal of medicine 2000, 343(19):1350-1354.

481. Milani L, Lundmark A, Kiialainen A, Nordlund J, Flaegstad T, Forestier E, Heyman M, Jonmundsson G, Kanerva J, Schmiegelow K et al. DNA methylation for subtype classification and prediction of treatment outcome in patients with childhood acute lymphoblastic leukemia. Blood 2010, 115(6):1214-1225.

482. Muller HM, Widschwendter A, Fiegl H, Ivarsson L, Goebel G, Perkmann E, Marth C, Widschwendter M. DNA methylation in serum of breast cancer patients: an independent prognostic marker. Cancer research 2003, 63(22):7641-7645.

483. Rawson JB, Manno M, Mrkonjic M, Daftary D, Dicks E, Buchanan DD, Younghusband HB, Parfrey PS, Young JP, Pollett A et al. Promoter methylation of Wnt antagonists DKK1 and SFRP1 is associated with opposing tumor subtypes in two large populations of colorectal cancer patients. Carcinogenesis 2011, 32(5):741-747.

484. Nikolaidis G, Raji OY, Markopoulou S, Gosney JR, Bryan J, Warburton C, Walshaw M, Sheard J, Field JK, Liloglou T. DNA methylation biomarkers offer improved diagnostic efficiency in lung cancer. Cancer research 2012, 72(22):5692-5701.

485. Esteller M, Corn PG, Baylin SB, Herman JG. A gene hypermethylation profile of human cancer. Cancer research 2001, 61(8):3225-3229.

     

    195

486. Rodriguez-Paredes M, Esteller M. Cancer epigenetics reaches mainstream oncology. Nature Medicine 2011, 17(3):330-339.

487. Herman JG, Baylin SB. Methylation-specific PCR. Current protocols in human genetics 2001, Chapter 10:Unit 10.6.

488. Trinh BN, Long TI, Laird PW. DNA methylation analysis by MethyLight technology. Methods 2001, 25(4):456-462.

489. Bustin SA, Benes V, Garson JA, Hellemans J, Huggett J, Kubista M, Mueller R, Nolan T, Pfaffl MW, Shipley GL et al. The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clinical chemistry 2009, 55(4):611-622.

490. Koshkin AA, Wengel J. Synthesis of Novel 2',3'-Linked Bicyclic Thymine Ribonucleosides. The Journal of Organic Chemistry 1998, 63(8):2778-2781.

491. Reynisson E, Josefsen MH, Krause M, Hoorfar J. Evaluation of probe chemistries and platforms to improve the detection limit of real-time PCR. Journal of Microbiological Methods 2006, 66(2):206-216.

492. Dahl C, Guldberg P. A ligation assay for multiplex analysis of CpG methylation using bisulfite-treated DNA. Nucleic acids research 2007, 35(21):e144.

493. Kandimalla R, Masius R, Beukers W, Bangma CH, Orntoft TF, Dyrskjot L, van Leeuwen N, Lingsma H, van Tilborg AA, Zwarthoff EC. A 3-Plex Methylation Assay Combined with the FGFR3 Mutation Assay Sensitively Detects Recurrent Bladder Cancer in Voided Urine. Clinical cancer research 2013.

494. Mokany E, Tan YL, Bone SM, Fuery CJ, Todd AV. MNAzyme qPCR with superior multiplexing capacity. Clinical chemistry 2013, 59(2):419-426.

495. Shi X, Tang C, Zhou D, Zhao H, Lu Z. Multiplex detection of CpG methylation using microarray combining with target-selection-padlock probe. Clinica chimica acta 2010, 411(17-18):1187-1194.

496. Wang J, Yu M, Li K, Xiao J, Zhou Y. Improved PCR-BSP assay for multiplex methylation pattern analysis in minimal amount of DNA. Mol Biotechnol 2009, 42(3):333-340.

497. Zerilli F, Bonanno C, Shehi E, Amicarelli G, Adlerstein D, Makrigiorgos GM. Methylation-specific loop-mediated isothermal amplification for detecting hypermethylated DNA in simplex and multiplex formats. Clinical chemistry 2010, 56(8):1287-1296.

498. Zhang Q, Hu G, Yang Q, Dong R, Xie X, Ma D, Shen K, Kong B. A multiplex methylation-specific PCR assay for the detection of early-stage ovarian cancer using cell-free serum DNA. Gynecologic oncology 2013, 130(1):132-139.

499. Zhang Z, Sun D, Hutajulu SH, Nawaz I, Nguyen Van D, Huang G, Haryana SM, Middeldorp JM, Ernberg I, Hu LF. Development of a non-invasive method, multiplex methylation specific PCR (MMSP), for early diagnosis of nasopharyngeal carcinoma. PloS one 2012, 7(11):e45908.

500. Zou H, Allawi H, Cao X, Domanico M, Harrington J, Taylor WR, Yab T, Ahlquist DA, Lidgard G. Quantification of methylated markers with a multiplex methylation-specific technology. Clinical chemistry 2012, 58(2):375-383.

501. Bormann F, Sers C, Seliger B, Handke D, Bergmann T, Seibt S, Lehrach H, Dahl A. Methylation-specific ligation detection reaction (msLDR): a new approach for multiplex evaluation of methylation patterns. Molecular genetics and genomics 2011, 286(3-4):279-291.

502. Fackler MJ, McVeigh M, Mehrotra J, Blum MA, Lange J, Lapides A, Garrett E, Argani P, Sukumar S. Quantitative multiplex methylation-specific PCR assay for the detection of promoter hypermethylation in multiple genes in breast cancer. Cancer research 2004, 64(13):4442-4452.

503. He Q, Chen HY, Bai EQ, Luo YX, Fu RJ, He YS, Jiang J, Wang HQ. Development of a multiplex MethyLight assay for the detection of multigene methylation in human colorectal cancer. Cancer genetics and cytogenetics 2010, 202(1):1-10.

504. Snellenberg S, De Strooper LM, Hesselink AT, Meijer CJ, Snijders PJ, Heideman DA, Steenbergen RD. Development of a multiplex methylation-specific PCR as candidate triage test for women with an HPV-positive cervical scrape. BMC cancer 2012, 12:551.

     

    196

505. Van Neste L, Bigley J, Toll A, Otto G, Clark J, Delree P, Van Criekinge W, Epstein JI. A tissue biopsy-based epigenetic multiplex PCR assay for prostate cancer detection. BMC urology 2012, 12:16.

506. Weisenberger DJ, Trinh BN, Campan M, Sharma S, Long TI, Ananthnarayan S, Liang G, Esteva FJ, Hortobagyi GN, McCormick F et al. DNA methylation analysis by digital bisulfite genomic sequencing and digital MethyLight. Nucleic acids research 2008, 36(14):4689-4698.

507. Lee BB, Lee EJ, Jung EH, Chun HK, Chang DK, Song SY, Park J, Kim DH. Aberrant methylation of APC, MGMT, RASSF2A, and Wif-1 genes in plasma as a biomarker for early detection of colorectal cancer. Clinical cancer research 2009, 15(19):6185-6191.

508. Ambros V. microRNAs: tiny regulators with great potential. Cell 2001, 107(7):823-826. 509. Bartel DP. MicroRNAs: target recognition and regulatory functions. Cell 2009, 136(2):215-233. 510. Harfe BD. MicroRNAs in vertebrate development. Current opinion in genetics & development

2005, 15(4):410-415. 511. Carleton M, Cleary MA, Linsley PS. MicroRNAs and cell cycle regulation. Cell Cycle 2007,

6(17):2127-2132. 512. Ivey KN, Srivastava D. MicroRNAs as regulators of differentiation and cell fate decisions. Cell

Stem Cell 2010, 7(1):36-41. 513. Lima RT, Busacca S, Almeida GM, Gaudino G, Fennell DA, Vasconcelos MH. MicroRNA

regulation of core apoptosis pathways in cancer. European journal of cancer 2011, 47(2):163-174.

514. Calin GA, Dumitru CD, Shimizu M, Bichi R, Zupo S, Noch E, Aldler H, Rattan S, Keating M, Rai K et al. Frequent deletions and down-regulation of micro- RNA genes miR15 and miR16 at 13q14 in chronic lymphocytic leukemia. Proceedings of the National Academy of Sciences of the United States of America 2002, 99(24):15524-15529.

515. Liu CG, Calin GA, Meloon B, Gamliel N, Sevignani C, Ferracin M, Dumitru CD, Shimizu M, Zupo S, Dono M et al. An oligonucleotide microchip for genome-wide microRNA profiling in human and mouse tissues. Proceedings of the National Academy of Sciences of the United States of America 2004, 101(26):9740-9744.

516. Xi Y, Nakajima G, Gavin E, Morris CG, Kudo K, Hayashi K, Ju J. Systematic analysis of microRNA expression of RNA extracted from fresh frozen and formalin-fixed paraffin-embedded samples. RNA 2007, 13(10):1668-1674.

517. Kolbert CP, Feddersen RM, Rakhshan F, Grill DE, Simon G, Middha S, Jang JS, Simon V, Schultz DA, Zschunke M et al. Multi-platform analysis of microRNA expression measurements in RNA from fresh frozen and FFPE tissues. PloS one 2013, 8(1):e52517.

518. Leite KR, Canavez JM, Reis ST, Tomiyama AH, Piantino CB, Sanudo A, Camara-Lopes LH, Srougi M. miRNA analysis of prostate cancer by quantitative real time PCR: comparison between formalin-fixed paraffin embedded and fresh-frozen tissue. Urologic oncology 2011, 29(5):533-537.

519. Kasinski AL, Slack FJ. Epigenetics and genetics. MicroRNAs en route to the clinic: progress in validating and targeting microRNAs for cancer therapy. Nature reviews Cancer 2011, 11(12):849-864.

520. Spahn M, Kneitz S, Scholz CJ, Stenger N, Rudiger T, Strobel P, Riedmiller H, Kneitz B. Expression of microRNA-221 is progressively reduced in aggressive prostate cancer and metastasis and predicts clinical recurrence. International journal of cancer 2010, 127(2):394-403.

521. Tucci P, Agostini M, Grespi F, Markert EK, Terrinoni A, Vousden KH, Muller PA, Dotsch V, Kehrloesser S, Sayan BS et al. Loss of p63 and its microRNA-205 target results in enhanced cell migration and metastasis in prostate cancer. Proceedings of the National Academy of Sciences of the United States of America 2012, 109(38):15312-15317.

522. Avgeris M, Stravodimos K, Fragoulis EG, Scorilas A. The loss of the tumour-suppressor miR-145 results in the shorter disease-free survival of prostate cancer patients. British journal of cancer 2013, 108(12):2573-2581.

     

    197

523. Becker-Santos DD, Guo Y, Ghaffari M, Vickers ED, Lehman M, Altamirano-Dimas M, Oloumi A, Furukawa J, Sharma M, Wang Y et al. Integrin-linked kinase as a target for ERG-mediated invasive properties in prostate cancer models. Carcinogenesis 2012, 33(12):2558-2567.

524. Cai C, Wang H, Xu Y, Chen S, Balk SP. Reactivation of androgen receptor-regulated TMPRSS2:ERG gene expression in castration-resistant prostate cancer. Cancer research 2009, 69(15):6027-6032.

525. Gupta S, Iljin K, Sara H, Mpindi JP, Mirtti T, Vainio P, Rantala J, Alanen K, Nees M, Kallioniemi O. FZD4 as a mediator of ERG oncogene-induced WNT signaling and epithelial-to-mesenchymal transition in human prostate cancer cells. Cancer research 2010, 70(17):6735-6745.

526. Iljin K, Wolf M, Edgren H, Gupta S, Kilpinen S, Skotheim RI, Peltola M, Smit F, Verhaegh G, Schalken J et al. TMPRSS2 fusions with oncogenic ETS factors in prostate cancer involve unbalanced genomic rearrangements and are associated with HDAC1 and epigenetic reprogramming. Cancer research 2006, 66(21):10242-10246.

527. Leshem O, Madar S, Kogan-Sakin I, Kamer I, Goldstein I, Brosh R, Cohen Y, Jacob-Hirsch J, Ehrlich M, Ben-Sasson S et al. TMPRSS2/ERG promotes epithelial to mesenchymal transition through the ZEB1/ZEB2 axis in a prostate cancer model. PloS one 2011, 6(7):e21650.

528. Folini M, Gandellini P, Longoni N, Profumo V, Callari M, Pennati M, Colecchia M, Supino R, Veneroni S, Salvioni R et al. miR-21: an oncomir on strike in prostate cancer. Molecular cancer 2010, 9:12.

529. Schaefer A, Jung M, Miller K, Lein M, Kristiansen G, Erbersdobler A, Jung K. Suitable reference genes for relative quantification of miRNA expression in prostate cancer. Experimental & molecular medicine 2010, 42(11):749-758.

530. Martens-Uzunova ES, Jalava SE, Dits NF, van Leenders GJ, Moller S, Trapman J, Bangma CH, Litman T, Visakorpi T, Jenster G. Diagnostic and prognostic signatures from the small non-coding RNA transcriptome in prostate cancer. Oncogene 2012, 31(8):978-991.

531. Wang SY, Shiboski S, Belair CD, Cooperberg MR, Simko JP, Stoppler H, Cowan J, Carroll PR, Blelloch R. miR-19, miR-345, miR-519c-5p serum levels predict adverse pathology in prostate cancer patients eligible for active surveillance. PloS one 2014, 9(6):e98597.

532. Larne O, Martens-Uzunova E, Hagman Z, Edsjo A, Lippolis G, den Berg MS, Bjartell A, Jenster G, Ceder Y. miQ--a novel microRNA based diagnostic and prognostic tool for prostate cancer. International journal of cancer 2013, 132(12):2867-2875.

533. Jalava SE, Urbanucci A, Latonen L, Waltering KK, Sahu B, Janne OA, Seppala J, Lahdesmaki H, Tammela TL, Visakorpi T. Androgen-regulated miR-32 targets BTG2 and is overexpressed in castration-resistant prostate cancer. Oncogene 2012, 31(41):4460-4471.

534. Shen J, Hruby GW, McKiernan JM, Gurvich I, Lipsky MJ, Benson MC, Santella RM. Dysregulation of circulating microRNAs and prediction of aggressive prostate cancer. The Prostate 2012, 72(13):1469-1477.

535. Coppola V, Musumeci M, Patrizii M, Cannistraci A, Addario A, Maugeri-Sacca M, Biffoni M, Francescangeli F, Cordenonsi M, Piccolo S et al. BTG2 loss and miR-21 upregulation contribute to prostate cell transformation by inducing luminal markers expression and epithelial-mesenchymal transition. Oncogene 2013, 32(14):1843-1853.

536. Corney DC, Flesken-Nikitin A, Godwin AK, Wang W, Nikitin AY. MicroRNA-34b and MicroRNA-34c are targets of p53 and cooperate in control of cell proliferation and adhesion-independent growth. Cancer research 2007, 67(18):8433-8438.

537. He L, He X, Lim LP, de Stanchina E, Xuan Z, Liang Y, Xue W, Zender L, Magnus J, Ridzon D et al. A microRNA component of the p53 tumour suppressor network. Nature 2007, 447(7148):1130-1134.

538. Bonci D, Coppola V, Musumeci M, Addario A, Giuffrida R, Memeo L, D'Urso L, Pagliuca A, Biffoni M, Labbaye C et al. The miR-15a-miR-16-1 cluster controls prostate cancer by targeting multiple oncogenic activities. Nature medicine 2008, 14(11):1271-1277.

539. Dong Q, Meng P, Wang T, Qin W, Qin W, Wang F, Yuan J, Chen Z, Yang A, Wang H. MicroRNA let-7a inhibits proliferation of human prostate cancer cells in vitro and in vivo by targeting E2F2 and CCND2. PloS one 2010, 5(4):e10147.

     

    198

540. Jazbutyte V, Thum T. MicroRNA-21: from cancer to cardiovascular disease. Current drug targets 2010, 11(8):926-935.

541. Balaguer F, Link A, Lozano JJ, Cuatrecasas M, Nagasaka T, Boland CR, Goel A. Epigenetic silencing of miR-137 is an early event in colorectal carcinogenesis. Cancer research 2010, 70(16):6609-6618.

542. Shirdel EA, Xie W, Mak TW, Jurisica I. NAViGaTing the micronome--using multiple microRNA prediction databases to identify signalling pathway-associated microRNAs. PloS one 2011, 6(2):e17429.

543. Ngalame NN, Tokar EJ, Person RJ, Xu Y, Waalkes MP. Aberrant microRNA expression likely controls RAS oncogene activation during malignant transformation of human prostate epithelial and stem cells by arsenic. Toxicological sciences 2014, 138(2):268-277.

544. Calvano Filho CM, Calvano-Mendes DC, Carvalho KC, Maciel GA, Ricci MD, Torres AP, Filassi JR, Baracat EC. Triple-negative and luminal A breast tumors: differential expression of miR-18a-5p, miR-17-5p, and miR-20a-5p. Tumour biology 2014, 35(8):7733-7741.

545. Mestdagh P, Hartmann N, Baeriswyl L, Andreasen D, Bernard N, Chen C, Cheo D, D'Andrade P, DeMayo M, Dennis L et al. Evaluation of quantitative miRNA expression platforms in the microRNA quality control (miRQC) study. Nature Methods 2014, 11(8):809-815.

546. Karatas OF, Guzel E, Suer I, Ekici ID, Caskurlu T, Creighton CJ, Ittmann M, Ozen M. miR-1 and miR-133b are differentially expressed in patients with recurrent prostate cancer. PloS one 2014, 9(6):e98675.

547. Schubert M, Spahn M, Kneitz S, Scholz CJ, Joniau S, Stroebel P, Riedmiller H, Kneitz B. Distinct microRNA expression profile in prostate cancer patients with early clinical failure and the impact of let-7 as prognostic marker in high-risk prostate cancer. PloS one 2013, 8(6):e65064.

548. Bi Y, Han Y, Bi H, Gao F, Wang X. miR-137 impairs the proliferative and migratory capacity of human non-small cell lung cancer cells by targeting paxillin. Human cell 2014, 27(3):95-102.

549. Chen Q, Chen X, Zhang M, Fan Q, Luo S, Cao X. miR-137 is frequently down-regulated in gastric cancer and is a negative regulator of Cdc42. Dig Dis Sci 2011, 56(7):2009-2016.

550. Guo J, Xia B, Meng F, Lou G. miR-137 suppresses cell growth in ovarian cancer by targeting AEG-1. Biochemical and biophysical research communications 2013, 441(2):357-363.

551. Langevin SM, Stone RA, Bunker CH, Grandis JR, Sobol RW, Taioli E. MicroRNA-137 promoter methylation in oral rinses from patients with squamous cell carcinoma of the head and neck is associated with gender and body mass index. Carcinogenesis 2010, 31(5):864-870.

552. Liu M, Lang N, Qiu M, Xu F, Li Q, Tang Q, Chen J, Chen X, Zhang S, Liu Z et al. miR-137 targets Cdc42 expression, induces cell cycle G1 arrest and inhibits invasion in colorectal cancer cells. International journal of cancer 2011, 128(6):1269-1279.

553. Wong TS, Liu XB, Wong BY, Ng RW, Yuen AP, Wei WI. Mature miR-184 as Potential Oncogenic microRNA of Squamous Cell Carcinoma of Tongue. Clinical cancer research 2008, 14(9):2588-2592.

554. Xiu Y, Liu Z, Xia S, Jin C, Yin H, Zhao W, Wu Q. MicroRNA-137 upregulation increases bladder cancer cell proliferation and invasion by targeting PAQR3. PloS one 2014, 9(10):e109734.

555. Noguchi M, Stamey TA, Neal JE, Yemoto CE. An analysis of 148 consecutive transition zone cancers: clinical and histological characteristics. The Journal of urology 2000, 163(6):1751-1755.

556. Su Z, Xia J, Zhao Z. Functional complementation between transcriptional methylation regulation and post-transcriptional microRNA regulation in the human genome. BMC Genomics 2011, 12 Suppl 5:S15.

557. Leite KR, Tomiyama A, Reis ST, Sousa-Canavez JM, Sanudo A, Camara-Lopes LH, Srougi M. MicroRNA expression profiles in the progression of prostate cancer--from high-grade prostate intraepithelial neoplasia to metastasis. Urologic oncology 2013, 31(6):796-801.

558. Amankwah EK, Anegbe E, Park H, Pow-Sang J, Hakam A, Park JY. miR-21, miR-221 and miR-222 expression and prostate cancer recurrence among obese and non-obese cases. Asian journal of andrology 2013, 15(2):226-230.

     

    199

559. Chan SH, Wu CW, Li AF, Chi CW, Lin WC. miR-21 microRNA expression in human gastric carcinomas and its clinical association. Anticancer research 2008, 28(2A):907-911.

560. Hellwinkel OJ, Sellier C, Sylvester YM, Brase JC, Isbarn H, Erbersdobler A, Steuber T, Sultmann H, Schlomm T, Wagner C. A Cancer-Indicative microRNA Pattern in Normal Prostate Tissue. International journal of molecular sciences 2013, 14(3):5239-5249.

561. Kim J, Wu L, Zhao JC, Jin HJ, Yu J. TMPRSS2-ERG gene fusions induce prostate tumorigenesis by modulating microRNA miR-200c. Oncogene 2014, 33(44):5183-92.

562. Gordanpour A, Stanimirovic A, Nam RK, Moreno CS, Sherman C, Sugar L, Seth A. miR-221 Is down-regulated in TMPRSS2:ERG fusion-positive prostate cancer. Anticancer research 2011, 31(2):403-410.

563. Pepe MS, Etzioni R, Feng Z, Potter JD, Thompson ML, Thornquist M, Winget M, Yasui Y. Phases of biomarker development for early detection of cancer. Journal of the National Cancer Institute 2001, 93(14):1054-1061.

564. McShane LM, Altman DG, Sauerbrei W, Taube SE, Gion M, Clark GM. Reporting recommendations for tumor marker prognostic studies (REMARK). Journal of the National Cancer Institute 2005, 97(16):1180-1184.

565. Dancey JE, Dobbin KK, Groshen S, Jessup JM, Hruszkewycz AH, Koehler M, Parchment R, Ratain MJ, Shankar LK, Stadler WM et al. Guidelines for the development and incorporation of biomarker studies in early clinical trials of novel agents. Clinical cancer research 2010, 16(6):1745-1755.

566. Diamandis EP. Cancer biomarkers: can we turn recent failures into success? Journal of the National Cancer Institute 2010, 102(19):1462-1467.

567. Buyse M, Sargent DJ, Grothey A, Matheson A, de Gramont A. Biomarkers and surrogate end points--the challenge of statistical validation. Nature reviews Clinical oncology 2010, 7(6):309-317.

568. Yang TY, Hsu LI, Chiu AW, Pu YS, Wang SH, Liao YT, Wu MM, Wang YH, Chang CH, Lee TC et al. Comparison of genome-wide DNA methylation in urothelial carcinomas of patients with and without arsenic exposure. Environmental research 2014, 128:57-63.

569. Wang X, Fan J, Liu D, Fu S, Ingvarsson S, Chen H. Spreading of Alu methylation to the promoter of the MLH1 gene in gastrointestinal cancer. PloS one 2011, 6(10):e25913.

570. Zhang Y, Shu J, Si J, Shen L, Estecio MR, Issa JP. Repetitive elements and enforced transcriptional repression co-operate to enhance DNA methylation spreading into a promoter CpG-island. Nucleic acids research 2012, 40(15):7257-7268.

571. Lai J, Myers SA, Lawrence MG, Odorico DM, Clements JA. Direct progesterone receptor and indirect androgen receptor interactions with the kallikrein-related peptidase 4 gene promoter in breast and prostate cancer. Molecular cancer research 2009, 7(1):129-141.

572. Nelson PS, Gan L, Ferguson C, Moss P, Gelinas R, Hood L, Wang K. Molecular cloning and characterization of prostase, an androgen-regulated serine protease with prostate-restricted expression. Proceedings of the National Academy of Sciences of the United States of America 1999, 96(6):3114-3119.

573. Saito Y, Jones PA. Epigenetic activation of tumor suppressor microRNAs in human cancer cells. Cell Cycle 2006, 5(19):2220-2222.

574. Lujambio A, Esteller M. CpG island hypermethylation of tumor suppressor microRNAs in human cancer. Cell Cycle 2007, 6(12):1455-1459.