Cross-species Modeling of Replication Repair Deficient Cancers

Cross-species Modeling of Replication Repair Deficient Cancers

by

Melissa Anna Galati

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

Institute of Medical Science University of Toronto

© Copyright by Melissa Anna Galati 2021

ii

Cross-species Modeling of Replication Repair Deficient Cancers

Melissa Anna Galati

Doctor of Philosophy

Institute of Medical Science

University of Toronto

2021

Abstract

Replication repair deficiency (RRD) is a leading cause of cancer hypermutation. RRD cancers are

found across tissue types, often resistant to chemo-radiation therapy, and therefore universally

lethal. Among children, germline deficiencies in either replication repair mechanism—DNA

polymerase proofreading or the mismatch repair (MMR) system—result in early-onset cancers

which were considered incurable. Despite recent success of immune checkpoint inhibition (ICI) in

treating RRD cancers, most tumors remain refractory or unresponsive. Novel treatments and robust

models to study RRD cancers are lacking.

Using a multi-omics approach, we found that RRD hypermutant cancers are enriched for

RAS/MAPK mutations, exhibited increased RAS/MAPK pathway activity, and responded to MEK

inhibition. Treatment of patients with RRD gliomas revealed durable responses to MEK inhibition,

suggesting these tumors are addicted to oncogenic pathways resulting in favorable response to

targeted therapies.

To better study hypermutant cancers in vivo, we established mouse models harboring cancer-

associated POLE mutations P286R and S459F, which caused rapid yet distinct time to cancer

initiation, enabled stratification of POLE mutations into 3 groups based on clinical phenotype and

mutagenicity, and mirrored human cancer ultrahypermutation and mutational signatures. These

animals primarily developed lymphomas which could not be prevented, and were likely

exacerbated by ICI, as observed in humans.

We then used the Pole models to study combined RRD (MMRD+Pole mutations), which

commonly occur in childhood hypermutant cancers. We developed novel brain and

gastrointestinal-specific RRD mouse models. Both models resulted in highly penetrant, rapid onset

iii

cancers that genomically mimicked human RRD cancers. We observed that increased RRD likely

contributes to mutational loads not sustainable for normal cell function as manifested by

maldevelopment and malfunction of colon and brain from MMRD+Pole mutant homozygous

mice. Moreover, serial passaging of RRD brain tumors revealed differences in mutation

accumulation between intracranially and subcutaneously transplanted tumors in

immunocompromised and immunocompetent animals, likely due to differences in immune

surveillance within and outside the brain. Collectively, our observations provide insights into the

carcinogenesis of POLE-driven tumors, information for genetic counselling, surveillance, and

immunotherapy for patients. Finally, this work provides insight on immune editing, and platforms

to further study RRD cancers in vivo.

iv

Acknowledgments

Pursuing my doctoral thesis has undeniably been my most difficult, yet immensely rewarding experience.

I could not have arrived at the finish line without support from these and so many other people. For everyone

who has cheered me on and helped me stand back up during hardship, I am eternally grateful.

To my supervisor, Dr. Uri Tabori. Uri, the best decision I made in graduate school was joining your lab. I

could not have asked for a more supportive and understanding supervisor. When I started six years ago,

you told me that pursuing a PhD forces you to “grow up” – and wow, what a journey it has been. Thank

you for challenging me to stand on my own two feet, make difficult decisions, think critically, and stay

humble. Your true enthusiasm for science, and dedication to your patients continue to inspire me. I hope I

can one day pay it forward.

To my committee members, Dr. Cynthia Hawkins and Dr. Alberto Martin. Thank you both for sharing your

wealth of knowledge and expertise with me over the last six years. You brought so many terrific insights

to the table each committee meeting and made this stressful process so much more enjoyable.

Thank you to everyone, past and present, in the Tabori lab and SickKids. To say “it takes a village” is an

understatement. I feel so blessed to have worked with so many intelligent, hardworking, and exceedingly

kind people. In particular, I would like to acknowledge Cindy Zhang, Taylor Bridge, Sumedha Sudhaman,

Dana Elshaer, Miki Gams, and Lucie Stengs. Your friendships, advice, and technical help have been second

to none, and I do not know how I would have done this without you ladies.

To my friends who are really my family, thank you for riding the ups and downs of graduate school with

me, for being my sounding board when I was overwhelmed, and for always listening to me talk science.

Alex, Eryn, Jackie, Jungmin, Kat, Lori, Maddy, Megan, Rohit and so many others—you didn’t always

understand, and I’m sure you were bored 99% of the time, but your support was never unnoticed.

Alanna, I cannot begin to thank you enough for your love and support. Your drive and enthusiasm for life

inspire me daily. Thank you for being the first to read this thesis and for your steadfast belief in me.

Finally, to my big, and often crazy family. The personalities that have shaped me have been wonderful,

quirky, and incredibly strong. To my grandparents, Alfio, Anna, Felicia, and Giuseppe—thank you for

braving new land, raising me, and teaching me about hard work and dedication. To my sister, Laura – your

fearlessness, candor, and brilliance never cease to amaze me. You are my better half. To my father, Mario

– you were my original cheerleader. Thank you for your love and for being the calm amidst the storm.

Finally, to my mother, Maria. You are my rock. No words can possibly do justice to articulate what you

have given me, and I will spend the rest of my life thanking you.

v

Statement of Contributions

This thesis was prepared entirely by Melissa A. Galati. All aspects of this body of work including

the planning, execution, analysis, and writing of all original research and publications was

performed in whole or in part the author. The following contributions by the author and by other

individuals are formally and inclusively acknowledged:

Chapter 2

Chapter 2 was modified and reproduced from:

Campbell BB*, Galati MA*, Stone SC, Riemenschneider AN, Edwards M, Sudhaman S,

Siddaway R, Komosa M, Nunes NM, Nobre L, Morrissy AS, Zatzman M, Zapotocky M,

Joksimovic L, Kalimuthu SN, Samuel D, Mason G, Bouffet E, Morgenstern DA, Aronson M,

Durno C, Malkin D, Maris JM, Taylor MD, Shlien A, Pugh TJ, Ohashi PS, Hawkins CE, Tabori

U. “Mutations in the RAS/MAPK pathway drive replication repair deficient hypermutated tumors

and confer sensitivity to MEK inhibition.” Cancer Discov. 2021 Jun;11(6):1454-1467. DOI:

10.1158/2159-8290.CD-20-1050.

* These authors contributed equally.

Specific Contributions: All authors contributed to interpretation of data, and manuscript revision.

Study conceptualization, experimental design, and initial manuscript writing were conducted by

Dr. Uri Tabori, Dr. Brittany B. Campbell, and Ms. Melissa A. Galati. Melissa A. Galati performed

data analysis for DNA and RNA sequencing, in vitro experiments, contributed to in vivo

experiments, coordinated completion of study, and completed all revisions. Dr. Brittany B.

Campbell acquired and analysed Foundation Medicine targeted sequencing data, coordinated

CMMRD patient sample submission for DNA and RNA sequencing, and analysed Nanostring 3D

data. Simone C. Stone performed flow cytometry experiments of patient PBMCs. Dr. Robert

Siddaway performed signature analysis and clustering of RNA sequencing data including

PROGENy, RAS 18-gene signature panel, and GSEA. Ms. Alexandra N. Riemenschneider

performed western blotting. Melissa Edwards coordinated patient consenting and patient data

acquisition. Mr. Martin Komosa and Dr. Nuno M. Nunes assisted with technical experitise for in

vitro cell line experiments. Ms. Tatiana Lipman performed in vivo PDX experiments. Drs.

vi

Sumedha Sudhaman, Sorana Morrissy, and Mr. Matthew Zatzman provided informatics advice

and expertise. Drs. David Samuel, Gary Mason, Eric Bouffet, Daniel A. Morgenstern, Carol

Durno, David Malkin, John M. Maris, Uri Tabori, and Ms. Melyssa Aronson conducted patient

management and care. Drs. Adam Shlien, Trevor J. Pugh, Pamela S. Ohashi, Cynthia E. Hawkins,

and Uri Tabori provided supervision, and acquired funding and resources.

Chapter 3

Chapter 3 was modified and reproduced from:

Galati MA*, Hodel KP*, Gams MS, Sudhaman S, Bridge T, Zahurancik WJ, Ungerleider NA,

Park VS, Ercan AB, Joksimovic L, Siddiqui I, Siddaway R, Edwards M, de Borja R, Elshaer D,

Chung J, Forster VJ, Nunes NM, Aronson M, Wang X, Ramdas J, Seeley A, Sarosiek T, Dunn

GP, Byrd JN, Mordechai O, Durno C, Martin A, Shlien A, Bouffet E, Suo Z, Jackson JG, Hawkins

CE, Guidos CJ, Pursell ZF and Tabori U. “Cancers from novel Pole mutant mouse models provide

insights into polymerase-mediated hypermutagenesis and immune checkpoint blockade. Cancer

Res. 2020 Dec 15(80)(24) 5606-5618; DOI: 10.1158/0008-5472.CAN-20-0624.

* These authors contributed equally.

Specific Contributions: All authors contributed to interpretation of data, and manuscript revision.

Study conceptualization, experimental design, and initial manuscript writing were conducted by

Dr. Uri Tabori, Ms. Melissa A. Galati, Dr. Zachary F. Pursell, and Dr. Karl P. Hodel. Dr. Karl P.

Ms. Melissa A. Galati developed the PoleS459F mouse model, conducted all experiments related to

PoleS459F mice, analysed DNA sequencing data, assisted with interpretation of CyTOF

experiments, coordinated completion of study, and completed all revisions. Dr. Karl P. Hodel

developed and conducted experiments related to PoleP286R mutant mice which were housed at

Tulane University. Ms. Miki S. Gams performed CyTOF staining and data analysis related ot

CyTOF experiments. Mr. Walter Zahurancik performed enzymology experiment. Dr. Robert

Siddaway provided exome sequencing data from non-RRD mutant mice. Ms. Taylor Bridge, Ms.

Dana Elshear, and Ms. Vivian Park provided technical assistance with mouse husbandry. Ms. Ayse

B. Ercan assisted with immunohistochemistry. Drs. Sumedha Sudhaman, Nathan A. Ungerleider,

vii

Mr. Richard de Borja, and Mr. Lazar Joksimovic provided technical assistance with analysis of

sequencing data. Ms. Melyssa Aronson, Dr. Xia Wang, Dr. Jagadeesh Ramdas, Ms. Andrea

Seeley, Dr. Thomasz Sarosiek, Dr. Gavin P. Dunn, Dr. Jonathan N. Byrd, Dr. Oz Mordechai, and

Dr. Carol Durno provided data on cases of human POLE germline mutations. Drs. Alberto Martin,

Adam Shlien, Eric Bouffet, Zucai Suo, James G. Jackson, Cynthia E. Hawkins, Cynthia J. Guidos,

Zachary F. Pursell and Uri Tabori provided supervision, and acquired funding and resources.

Study conceptualization and experimental design Chapters 4 and 5 was conducted by Dr. Uri

Tabori and Ms. Melissa A. Galati. Ms. Melissa A. Galati conducted all in vivo experiments, data

interpretation, and bioinformatics analyses.

Chapter 4

Ms Taylor Bridge, Ms. Dana Elshaer, and Ms. Lucie Stengs provided technical assistance with

mouse colony maintenance and collection of in vivo data. Drs. Sumedha Sudhaman, Vanessa

Bianchi, and Anirban Das collected, analyzed, and provided human tumor data from the

International Replication Repair Deficiency Consortium.

Chapter 5

Ms. Lucie Stengs provided technical assistance with in vivo experiments and data collection. Mr.

Owen Crump conducted immunofluorescence experiments.

viii

Table of Contents

Acknowledgments.......................................................................................................................... iv

Statement of Contributions ..............................................................................................................v

Table of Contents ......................................................................................................................... viii

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

List of Figures .............................................................................................................................. xiv

List of Abbreviations .................................................................................................................. xvii

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

Introduction .................................................................................................................................1

1.1 DNA Replication & Replication Related Repair .................................................................1

1.1.1 Eukaryotic DNA Replication & Replicative Polymerases ......................................1

1.1.2 DNA Polymerase Fidelity & Exonuclease Proofreading .........................................2

1.1.3 DNA Mismatch Repair ............................................................................................3

1.2 The Mutator Phenotype Hypothesis & Cancer ....................................................................5

1.3 DNA Replication Repair & Cancer Predisposition .............................................................6

1.3.1 Lynch Syndrome/HNPCC .......................................................................................7

1.3.2 Constitutional/Biallelic Mismatch Repair Deficiency Syndrome............................9

1.3.3 Polymerase Proofreading Associated Polyposis ....................................................13

1.4 Treatment of Replication Repair Deficient Cancers ..........................................................15

1.4.1 Chemotherapies & Radiation .................................................................................15

1.4.2 Targeted & Synthetic Lethal Approaches ..............................................................18

1.4.3 Immune Checkpoint Inhibition ..............................................................................21

1.5 Replication Repair Deficient Cancer Modeling .................................................................25

1.5.1 Established cell lines, patient derived cell lines, & xenografts ..............................25

1.5.2 Mismatch Repair Deficient Mouse Models ...........................................................26

1.5.3 Polymerase Proofreading Deficient Mouse Models ..............................................28

ix

1.6 Thesis Rationale, Hypothesis, & Aims ..............................................................................30

1.6.1 Hypothesis..............................................................................................................30

1.6.2 Specific Aims .........................................................................................................30

Chapter 2 Mutations in the RAS/MAPK pathway drive replication repair deficient

hypermutated tumors and confer sensitivity to MEK inhibition ...............................................31

Mutations in the RAS/MAPK pathway drive replication repair deficient hypermutated

tumors and confer sensitivity to MEK inhibition ......................................................................32

2.1 Abstract ..............................................................................................................................32

2.2 Introduction ........................................................................................................................32

2.3 Materials & Methods .........................................................................................................34

2.3.1 Patient and sample collection for exome sequencing and PBMC collection.........34

2.3.2 Targeted panel sequencing .....................................................................................35

2.3.3 Whole exome sequencing ......................................................................................35

2.3.4 Subclone analysis ...................................................................................................35

2.3.5 Variant Impact Prediction Software.......................................................................36

2.3.6 RNA-Sequencing ...................................................................................................36

2.3.7 Nanostring nCounter gene expression system .......................................................36

2.3.8 In vivo serial xenograft experiment ........................................................................36

2.3.9 In vivo treatment of patient-derived xenografts .....................................................37

2.3.10 Western blotting .....................................................................................................37

2.3.11 Immunohistochemistry for RAS/MAPK pathway upregulation ............................37

2.3.12 Cell Lines ...............................................................................................................38

2.3.13 Flow Cytometry .....................................................................................................38

2.3.14 Statistical Analysis .................................................................................................38

2.3.15 Data Availability ....................................................................................................38

2.4 Results ................................................................................................................................39

x

2.4.1 Mutations in the RAS/MAPK pathway are common in hypermutated

childhood cancers...................................................................................................39

2.4.2 Replication repair deficient cancers activate the RAS/MAPK pathway ...............46

2.4.3 RRD hypermutant cancers respond to MEK inhibition .........................................57

2.4.4 Encouraging responses in patients with RRD gliomas to MEK inhibition ...........61

2.5 Discussion ..........................................................................................................................64

Chapter 3 Cancers from novel Pole mutant mouse models provide insights into polymerase-

mediated hypermutagenesis and immune checkpoint blockade ...............................................67

Cancers from novel Pole mutant mouse models provide insights into polymerase-mediated

hypermutagenesis and immune checkpoint blockade ...............................................................68

3.1 Abstract ..............................................................................................................................68

3.2 Introduction ........................................................................................................................68

3.3 Materials & Methods .........................................................................................................71

3.3.1 Experimental model & subject details ...................................................................71

3.3.2 Method Details .......................................................................................................73

3.3.3 Mass Cytometry Methods ......................................................................................76

3.4 Results ................................................................................................................................78

3.4.1 Pole mutant mice provide insight into genotype-phenotype in humans ................78

3.4.2 The genetic profiles of Pole mutant tumors resemble human cancers ..................85

3.4.3 Pole mutant mice develop two distinct types of T cell lymphoma ........................93

3.4.4 Immune checkpoint blockade (ICB) increases proliferation in Pole mutant T

cell lymphomas ....................................................................................................100

3.5 Discussion ........................................................................................................................103

Chapter 4 Generation of combined replication repair deficient (RRD) mouse models for the

study of RRD cancers..............................................................................................................107

Generation of combined replication repair deficient (RRD) mouse models for the study of

RRD cancers............................................................................................................................108

4.1 Introduction ......................................................................................................................108

4.2 Materials & Methods .......................................................................................................110

xi

4.2.1 Study Approval ....................................................................................................110

4.2.2 Mouse Models & Husbandry ...............................................................................110

4.2.3 Tissue Processing & Immunohistochemistry.......................................................110

4.2.4 Whole Exome Sequencing ...................................................................................111

4.2.5 Organoid Establishment .......................................................................................111

4.2.6 Statistical Analysis ...............................................................................................112

4.3 Results ..............................................................................................................................113

4.3.1 Intestinal specific ablation of MMR and Pole proofreading leads to tract-wide

dysplasia and greatly reduced survival in mice ...................................................113

4.3.2 Central nervous system wide ablation of MMR and Pole proofreading results

in rapid malignant brain tumorigenesis in the posterior fossa .............................116

4.3.3 The genomic landscape of mouse cancers with complete RRD mirrors the

human syndrome ..................................................................................................119

4.4 Discussion ........................................................................................................................121

Chapter 5 Immune surveillance affects the mutational landscape of RRD brain tumors ............123

Immune surveillance affects the mutational landscape of RRD brain tumors ........................124

5.1 Introduction ......................................................................................................................124

5.2 Materials & Methods .......................................................................................................127

5.2.1 Mouse Models & Husbandry ...............................................................................127

5.2.2 Orthotopic and subcutaneous serial tumor transplantation ..................................127

5.2.3 Tissue Processing & Immunofluorescence ..........................................................127

5.2.4 Whole Exome Sequencing ...................................................................................128

5.2.5 Statistical Analysis ...............................................................................................128

5.3 Results ..............................................................................................................................128

5.3.1 Anatomical location of transplanted RRD brain tumors influences growth

kinetics and overall survival in vivo.....................................................................128

5.3.2 Serial transplantation of RRD MBTs shed light on the impact of immune

surveillance on the genomic landscape of tumors ...............................................131

xii

5.4 Discussion ........................................................................................................................136

Chapter 6 General Discussion ......................................................................................................138

General Discussion..................................................................................................................138

6.1 Summary & Significance of Findings..............................................................................138

6.2 Future Directions .............................................................................................................144

6.2.1 In vitro small molecule screens & in vivo combinatorial therapies .....................144

6.2.2 Ablation of combined RRD in oligodendrocyte precursors for a more

clinically relevant model of brain tumorigenesis .................................................146

6.2.3 Emerging secondary driver events in CMMRD brain tumors .............................147

References ....................................................................................................................................149

Appendices ...................................................................................................................................179

Copyright Acknowledgements.....................................................................................................187

xiii

List of Tables

Table 1-1 Summary of MMR Deficient Mouse Models ............................................................... 27

Table 3-1 Spontaneous tumor incidence across PoleS459F/S459F, PoleS459F/+, and

PoleP286R/+ moribund mice ........................................................................................................ 85

xiv

List of Figures

Figure 1.1-1 Eukaryotic Replication and Mechanisms of Fidelity ................................................. 5

Figure 1.3-1 Tumor spectrum and Incidence for CMMRD patients within the International

Replication Repair Deficiency Consortium (IRRDC). ................................................................. 11

Figure 1.3-2 Frequency and location of germline and somatic Pol δ and Pol ε proofreading

exonuclease domain mutations in cancers .................................................................................... 15

Figure 1.5-1 Survival findings for two previously reported polymerase proofreading mutant

mouse models................................................................................................................................ 29

Figure 2.4-1 Prevalence of RAS/MAPK genetic events across 1215 pediatric cancers ............... 41

Figure 2.4-2 Major subtypes in1215 pediatric cancers (Ages 0-18) ............................................. 43

Figure 2.4-3 Genes with highest frequency of nonsynonymous point mutations across 1215

tumors ........................................................................................................................................... 44

Figure 2.4-4 Confirmatory tissue specific preference for MAPK pathway mutations in cBioPortal

Query............................................................................................................................................. 45

Figure 2.4-5 Prevalence of RAS/MAPK genetic events across 46 RRD cancers ........................ 49

Figure 2.4-6 Prevalence of RAS/MAPK mutations in polyclonal and temporal sampling of RRD

tumors ........................................................................................................................................... 50

Figure 2.4-7 Unsupervised transcriptomic sample clustering ....................................................... 51

Figure 2.4-8 Assessment of MAPK pathway activation in hypermutant RRD gliomas using

PROGENy..................................................................................................................................... 53

Figure 2.4-9 Transcriptome clustering based on housekeeping gene expression ......................... 53

Figure 2.4-10 Transcriptomic assessment of RAS pathway activation in hypermutant RRD ...... 54

Figure 2.4-11 Transcriptional assessment of RAS/MAPK pathway activation in hypermutant

RRD gliomas using NanoString3D ............................................................................................... 55

Figure 2.4-12 Proteomic assessment of RAS/MAPK pathway activation in hypermutant RRD

gliomas .......................................................................................................................................... 57

Figure 2.4-13 Sensitivity of established MMRD cell lines to MEK inhibition ........................... 58

Figure 2.4-14 Genomic characterization and subsequent xenografting and treatment of an

ultrahypermutant childhood colorectal cancer .............................................................................. 59

Figure 2.4-15 Genomic characterization, xenografting, and treatment of ultrahypermutant

childhood GBM ............................................................................................................................ 60

xv

Figure 2.4-16 Clinical response of Patient 1 with RRD high-grade glioma to Selumetinib ........ 62

Figure 2.4-17 Clinical response of Patient 2 with an RRD high-grade glioma to Trametinib and

flow cytometric analysis of anti-PD-1/MEKi synergism .............................................................. 64

Figure 3.4-1 PoleP286R mouse model design, validation, and tumor findings ........................... 82

Figure 3.4-2 PoleS459F mouse model design, validation, and tumor findings ............................ 84

Figure 3.4-3 Pole mutations confer variable tumorigenic capabilities in vivo and support a

genotype-phenotype correlation .................................................................................................... 84

Figure 3.4-4 The genomic landscape of Pole mutant mouse tumors resembles that of POLE-

driven human cancers ................................................................................................................... 88

Figure 3.4-5 Pole mutant mice tumors exhibit mutational signatures found in POLE driven

human cancers ............................................................................................................................... 88

Figure 3.4-6 The genomic landscape of Pole mutant mouse tumors gives insight into mutagenesis

mechanisms ................................................................................................................................... 89

Figure 3.4-7 (related to Figure 3.4-6) Exome data from tumor fractions from a single mouse were

compared for 3 additional mice .................................................................................................... 92

Figure 3.4-8 Determination of two types of T cell lymphoma in Pole mutant mice and their cell

of origins ....................................................................................................................................... 97

Figure 3.4-9 Identification of two distinct types of T cell lymphoma in Pole mutant mice ......... 97

Figure 3.4-10 Characterization of T cell malignancies in PoleS459F/S459F and PoleS459F/+

mice ............................................................................................................................................... 98

Figure 3.4-11 Extrinsic effect of ‘Group B’ malignant T cells on B cell population ................. 100

Figure 3.4-12 Impact of prophylactic immune checkpoint blockade on Pole driven

lymphomagenesis ........................................................................................................................ 102

Figure 4.3-1 GI specific ablation of RRD leads to decreased survival, high grade dysplasia, and

reduced GI stem-cell capabilities in vitro ................................................................................... 115

Figure 4.3-2 CNS-wide RRD leads to robust brain tumorigenesis in vivo and paradoxical

genotype-phenotype correlations ................................................................................................ 119

Figure 4.3-3 Cross species comparison of genomic tumor features reveals correlation between

mouse models and human cancers .............................................................................................. 120

Figure 5.1-1 Biomarkers predicting response to anti-PD-1 in RRD human cancers .................. 126

Figure 5.3-1 Anatomical location of transplanted RRD brain tumors influences growth kinetics

and overall survival in vivo......................................................................................................... 130

xvi

Figure 5.3-2 Schematic of serial transplantation experiment ..................................................... 134

Figure 5.3-3 Conservation of mutations between primary and transplanted RRD brain tumor . 134

Figure 5.3-4 Differences in mutation accumulation and changes to mutational landscape between

mechanisms of serial transplantation of RRD brain tumors ....................................................... 136

Figure 6.2-1 Established RRD brain tumor cell lines genomically resemble human and mouse

tumors and can be used for in vivo testing of therapies .............................................................. 145

Figure 6.2-2 OC+/Msh2L/L/PoleSF/+ mutant mice display delayed brain tumor onset and greater

incidence of glioma-like tumors ................................................................................................. 147

xvii

List of Abbreviations

5-FU – 5-Fluoro-uracil

6-TG – 6-thioguanine

APC – Antigen Presenting Cell

APC - Adenomatous Polyposis Coli

ATP – Adenosine Triphosphate

BMMRD – Biallelic Mismatch Repair Deficiency

CMMRD – Constitutional Mismatch Repair Deficiency

BWA – Burrows-Wheeler Alignment

CALMs – Café-au-lait macules

CD28 – Cluster of Differentiation 28

cGAS – Cyclic GMP-AMP synthase

CRC – Colorectal Cancer

CTLA-4 – Cytotoxic T-lymphocyte-associated protein 4

DNA – Deoxyribonucleic acid

DSB – Double stranded break

EPCAM – Epithelial cell adhesion molecule

ExoI – Exonuclease I

FAP – Familial Adenomatous Polyposis

FFPE – Formalin-fixed, paraffin embedded

xviii

GBM – Glioblastoma Multiforme

GEMM – Genetically engineered mouse model

HGD – High grade dysplasia

HGG – High grade glioma

HNPCC – Hereditary nonpolyposis colorectal cancer

ICI – Immune Checkpoint Inhibition

IDH1 – Isocitrate dehydrogenase 1

IHC - Immunohistochemisty

InSiGHT – International Society for Gastrointestinal and Hereditary Tumors

KiCS – SickKids Cancer Sequencing program

KRAS – Kirsten Rat Sarcoma Viral Oncogene Homolog

LGD – Low grade dysplasia

LS – Lynch Syndrome

MAPK – Mitogen activated protein kinase

MBT – Malignant brain tumor

MEK1 - Mitogen activated protein kinase kinase 1

MGMT – O6-Methylguanine-DNA Methyltransferase

MHC – Major Histocompatibility Complex

MLH1 – MutL Homolog 1

MMR – Mismatch repair

xix

MRI – Magnetic Resonance Imaging

MSH2 – MutS protein homolog 2

MSH3 – MutS homolog 3

MSH6 – MutS homolog 6

MSI – Microsatellite Instability

PCNA – Proliferating cell nuclear antigen

PCR – Polymerase chain reaction

PD-L1 – Programmed death-ligand 1

PD-1 – Programmed cell death protein 1

PDX – Patient-derived xenograft

PMS2 – PMS1 homolog 2

PNET – Primitive neuroectodermal tumor

POLa - Polymerase alpha

POLA1 – DNA polymerase alpha 1, catalytic subunit

POLA2 - DNA polymerase alpha 1

POLD1 – DNA polymerase delta catalytic subunit

POLE – DNA polymerase epsilon

PPAP – Polymerase Proofreading associated polyposis

PPD – Polymerase proofreading deficiency

PRIM1 – Primase DNA subunit 1

xx

RPA – Replication protein A

SBC – Small Bowel Cancer

SNV – Single nucleotide variant

STING – Stimulator of interferon genes

TCGA – The Cancer Genome Atlas

TGN - Thioguanine

TMB – Tumor mutation burden

TMZ - Temozolomide

TP53 – Tumor protein 53

UCSC – University of California Santa Cruz

WRN – Werner syndrome helicase

1

Chapter 1 Introduction

Introduction

1.1 DNA Replication & Replication Related Repair

1.1.1 Eukaryotic DNA Replication & Replicative Polymerases

The discovery of DNA as the substance of heredity and, subsequently, its structure in 1953,

necessitated a mechanism for faithfully replicating all six billion bases to progeny daughter cells

(Watson et al. 1953)—a mechanism revealed to be a tightly regulated and complex process

involving many different proteins, broadly termed the “replisome” (Burgers et al. 2017). While

components of the replisome such as the helicase, single-strand binding protein, clamp, and clamp

loader are essential for a functional eukaryotic replication fork, the proteins responsible for

synthesizing daughter DNA from a single stranded template are the DNA polymerases, with the

first DNA polymerase (Escherichia coli; Pol I) discovered in bacteria in 1958 (Lehman et al.

1958).

DNA polymerases are divided into A, B, C, D, X, Y, RT, (reverse transcriptase) and AEP (archaeo-

eukaryotic primase superfamily) families based on sequence conservation. These enzymes may

also be grouped based on their function as either replicative or non-replicative polymerases,

meaning they are either required solely during cell division to replicate the genome or they are

required throughout the cell’s life cycle, respectively. In humans, three DNA polymerases are

responsible for replicating nuclear DNA: Pol α (primase), Pol δ, and Pol ε—all three of which fall

into the B-family of DNA polymerases (Lange et al. 2011). Pol α is a heterotetramer composed of

two primase subunits—which initiate replication by synthesizing short (7-12 ribonucleotide) RNA

primers—and two polymerase subunits—which begin DNA synthesis, extending from RNA

primers (Pellegrini 2012). From the origin of replication, the replication fork moves bi-

directionally with all three DNA polymerases limited by the requirement of a 3’OH to conduct

synthesis meaning they may only proceed 5’ to 3’, resulting in the existence of Okazaki fragments

on the “lagging strand” (Okazaki et al. 1968). In eukaryotes, extensive studies in yeast have led to

a division of labor model whereby Pol ε is responsible for “leading” strand DNA synthesis (Pursell

2

et al. 2007), while Pol δ synthesizes the lagging strand (Morrison et al. 1990). In humans, however,

this division of labour at the replication fork is less clear with numerous studies supporting a role

for Pol δ in leading strand synthesis under certain conditions including cell stress (Pursell et al.

2007, Burgers 2009, Garbacz et al. 2018).

While Pol α is capable of both RNA priming and DNA extension, it does not participate in the

bulk of replication given its low processivity and lack of intrinsic proofreading capabilities

(Pellegrini 2012). Like Pol α, Pols ε and δ are multi-subunit enzymes. The Pol ε holoenzyme

consists of 4 subunits, p261 p59, p17, and p12, named for their predicted molecular weights in kD

and coded by the genes POLE, POLE2, POLE3, and POLE4, respectively. The p261 (POLE)

subunit (also known as the protein’s catalytic subunit) contains the highly conserved polymerase

and 3’ to 5’ exonuclease domains at its N-terminus, while the C-terminus is responsible for

interaction with other subunits in the holoenzyme (Henninger et al. 2014). Likewise, the Pol δ

heterotetramer consists of p125 (POLD1), p66 (POLD3), p50 (POLD4), and p12 (POLD4), with

p125 most like the N-terminal half of Pol ε p261 in that it also contains exonuclease and

polymerase domains (Pavlov et al. 2020). While all aspects of the holoenzyme are essential for

polymerase function, this review will focus on the catalytic subunits since they contain elements

most relevant to the work summarized in this thesis. A schematic of the eukaryotic replication fork

is provided in Figure 1.1-1 a.

1.1.2 DNA Polymerase Fidelity & Exonuclease Proofreading

In eukaryotes, under normal conditions, DNA replication is extremely accurate with an error rate

of 1 x 10-10 mutations per base pair per cell division (<1 mutation each cell cycle) (Drake et al.

1998). Pols ε and δ possess an intrinsic base selectivity, which serves as the first barrier to

mutagenesis, incorporating an incorrect base pair, on average, every 104-105 nucleotides

(McCulloch et al. 2008). If such mutations were allowed to persist, over 100 000 point mutations

or single nucleotide variants (SNVs) would arise from each cell division. The repair processes

uniquely associated with DNA replication, preventing mutation accumulation, include polymerase

proofreading and the post-replicative DNA Mismatch Repair (MMR) system (see Section 1.1.3;

Figure 1.1-1 b).

As mentioned, both Pols ε and δ possess a functional 3’ to 5’ exonuclease proofreading domain in

their N-terminus, which is contained in a domain physically separated from the polymerase domain

3

(Beese et al. 1991). The exonuclease active site is composed of amino acids in conserved motifs

designated Exo I, II, and III, including carboxylate residues in motifs I and III that co-ordinate

divalent metal ions, essential for catalysis (Beese et al. 1991, Shamoo et al. 1999). Early

biochemistry studies in yeast (Saccharomyces cerevisiae) examined the error rates associated with

a mutant holoenzyme lacking exonuclease activity due to alanine replacements for the catalytic

carboxylates and found that this intrinsic proofreading activity increases replication fidelity by 10-

to 100-fold in Pol ε (Morrison et al. 1991, Shcherbakova et al. 1996, Shcherbakova et al. 2003)

and Pol δ (Fortune et al. 2005, Fortune et al. 2006). Additional work utilizing purified human Pol

ε and lacZ forward mutation assays supported early findings in yeast (Shinbrot et al. 2014). It was

these studies that caused scientists to postulate the potential link between DNA polymerase

proofreading and cancer (see below; Section 1.3). The exonuclease domain is highly conserved

across species, and indeed mutations causing changes to amino acids other than the catalytic

carboxylate residues also result in impaired proofreading ability, albeit to different extents

(Shinbrot et al. 2014, Barbari et al. 2018, Xing et al. 2019).

1.1.3 DNA Mismatch Repair

Post-replicative MMR acts as a final defence against incorrectly incorporated base pairs during S

phase. Paul Modrich is credited with mapping, at a molecular level, the MMR pathway, earning

him the Nobel Prize in Chemistry for his seminal work initially published in 1991. The MMR

system increases fidelity by approximately three orders of magnitude (Preston et al. 2010). In

addition to single base pair mismatches, the MMR system is also equipped to safeguard against

small insertions and deletions (indels; see further elaboration below). Modrich’s studies focused

on the MMR system in E. coli, and significant differences exist between the prokaryotic and

eukaryotic pathways; however the fundamental elements of these systems are the same. Generally,

the process of MMR includes four steps: 1) Mismatch recognition; 2) Signalling and

communication with downstream factors, 3) Mispaired base excision; and 4) DNA resynthesis

(Modrich 2006).

In E. coli, MMR is initiated by the protein MutS. As a homodimer, MutS scans DNA molecules

and recognizes mismatches and indels generated by DNA polymerase III, the major replicase for

both leading and lagging strands in this organism. Following mismatch detection, MutS undergoes

conformational changes which both induce asymmetry in the DNA at the site of the mismatch, as

4

well as changes in the protein’s ATPase domain and ATP binding site (Modrich 1991, Lamers et

al. 2000, Obmolova et al. 2000). In a mechanism not fully understood, these changes allow MutS

to associate with MutL, allowing communication with downstream effectors and assembly of the

MMR machinery (Grilley et al. 1989). To identify which strand must be corrected, MutH, an

endonuclease that specifically detects hemimethylated DNA is then recruited. MutH recognizes

methylated dGATC sites on the template (parent) strand and creates an incision on the transiently

unmethylated nascent (daughter) strand (Lu et al. 1983, Längle-Rouault et al. 1987). This creates

an entry point for DNA helicase II to unwind the DNA, which is coated and protected by single

strand binding (SSB) protein. Depending on where MutH nicks the DNA relative to the mismatch,

an appropriate exonuclease (four total with either 5’-to-3’ or 3’-to-5’ exonucleolytic activity) can

then excise the DNA surrounding and including the lesion. Finally, DNA is resynthesized by DNA

polymerase III and ligated by DNA ligase (Lahue et al. 1989).

In eukaryotes, the MMR machinery is considerably more complex. The eukaryotic counterparts to

prokaryote MutS and MutL are termed MutS homologue (Msh) and MutL homologue (Mlh).

Unlike in prokaryotes, the eukaryotic system utilizes several distinct heterodimers. There are two

mismatch recognition dimers MutSα and MutSβ. MutSα, consisting of the proteins Msh2 and

Msh6, recognizes single base pair mismatches and 1-2 base pair indels, while MutSβ, consisting

of Msh2 and Msh3, recognizes larger indels of 3-16 base pairs. While several functionally distinct

MutL homologue dimers exist in eukaryotes, the sole complex required for MMR is MutLα, which

consists of Mlh1 and postmeiotic segregation increased 2 (Pms2; Pms1 in yeast) and is recruited

to both MutSα and MutSβ once their respective targets have been identified. As in prokaryotes,

MutLα initiates downstream excision of the DNA surrounding and including the lesion (Kadyrov

et al. 2006), however, unlike in prokaryotes, hemimethylation of the DNA is not used as the means

with which to discriminate between DNA strands and a MutH homologue does not exist in

eukaryotes. Instead, it is postulated that strand breaks between lagging strand Okazaki fragments,

PCNA orientation, and a free 3’ hydroxyl on the leading strand act as signals for strand

discrimination (Kunkel et al. 2015). These predicted mechanisms of strand discrimination align

with studies in yeast which show that MMR more efficiently corrects lagging strand errors than

leading strand errors (Lujan et al. 2014). In the absence of MutH, MutLα was found to contain the

endonuclease function necessary for DNA cleavage (specifically this motif is found in Pms2 and

yeast Pms1) (Kadyrov et al. 2006). Repair events continue with the sole eukaryotic exonuclease,

5

Exo1, unwinding of the DNA by DNA helicase II, protection of template strand by RPA, DNA

resynthesis by Pol δ, and finally, ligation by ligase I (Longley et al. 1997, Nick McElhinny et al.

2008). Although both MutSα and MutSβ are involved in MMR, the mutator phenotype observed

from defects in MutSα are more severe than MutSβ and therefore have greater implications in

cancer initiation (see Section 1.3 below).

Figure 1.1-1 Eukaryotic Replication and Mechanisms of Fidelity

(A) Schematic of the eukaryotic replication fork highlighting the three mechanisms contributing to replication fidelity:

intrinsic DNA polymerase selectivity, DNA polymerase proofreading, and the mismatch repair system. Modified from

Preston et al. Semin. Cancer Biol. 2010. (B) Contribution of each mechanism indicated in (A) to overall replication

fidelity of 10-10 mutations/cell division. Modified from McCulloch & Kunkel, Cell Res. 2008.

1.2 The Mutator Phenotype Hypothesis & Cancer

It is, indeed, incredible now to think that only 50 years ago the link between mutational events and

cancer was merely speculative. The concept of the mutator phenotype hypothesis was put forth by

Lawrence A. Loeb and colleagues, who postulated that alterations in genes leading to increased

rates of spontaneous mutations underlie initiation and development of cancer (Loeb et al. 1974).

6

The earliest version of this hypothesis focused on errors in DNA replication (and therefore

mutations in genes that would ensure the fidelity of replicative DNA polymerases) as the basis for

malignant change. Such mutated genes are referred to as “mutators”. This early hypothesis was

predicated on several observations, such as, the incredible precision of normal DNA replication

machinery, experiments examining in vitro mutation rates of mutant forms of DNA polymerases

that abolished their proofreading activities (Kunkel et al. 1981), as well as observations of cancer

itself, including the accumulation of chromosomal aberrations throughout tumor development and

the ability of tumors to become resistant to chemotherapies. With growing evidence that mutations

could accumulate because of normal cellular processes (endogenous factors) and various

environmental insults (exogenous factors) (Lindahl et al. 1999), the mutator phenotype hypothesis

was expanded to include mutations in DNA repair genes more broadly.

A parallel, alternative theory of clonal evolution of tumor cell populations, put forth by Peter C.

Nowell, stated that genomic instability, governed by the generation and selection of mutations

(which could arise without increased mutation rates), contributes to a stepwise progression of

tumorigenesis (Nowell 1976). Although presented as separate hypotheses, the ideas of enhanced

mutagenesis and clonal selection are not mutually exclusive. Indeed, it was recognized that

alterations in some genes, including those coding for MMR machinery, both increase spontaneous

mutations and are selected for due to their role in the DNA damage response leading to cellular

resistance to certain chemotherapies (Fishel 2001). Further, in their revised “Hallmarks of Cancer”

Hanahan and Weinberg amended their original hallmarks to include genomic instability as an

enabling factor in tumor development and discussed how such instability leads to mutations

distributed randomly across the genome with a select few of these mutations conferring selective

advantage on subclones, enabling their outgrowth and dominance over less favorable clones

(Hanahan et al. 2011). This mutability hallmark is made possible through the breakdown of one or

more components of DNA-maintenance machinery—referred to as the genome’s “caretakers”.

1.3 DNA Replication Repair & Cancer Predisposition

The acceptance of genomic instability as the basis of cancer initiation and progression, coupled

with great advances in DNA sequencing technologies have led to many breakthroughs in the field

7

of cancer genetics. This includes the increased ability to determine the genetic aetiology of cancer,

with the hope of better diagnosis, stratification, and treatment of patients. In particular, these

technologies have aided in the detection and study of hypermutant cancers—those tumors with >

10 mutations per megabase (The Cancer Genome Atlas et al. 2012, Campbell et al. 2017).

Hypermutant cancers have many possible aetiologies, one of which is replication repair deficiency

(RRD)—deficiencies in either DNA polymerase proofreading or MMR. Such deficiencies can

arise de novo or occur as germline events. Given that inherited RRD is a cornerstone of this thesis,

this section will outline the three known cancer predisposition syndromes resulting from ablation

of either MMR or polymerase proofreading: 1) Lynch Syndrome (LS)—otherwise known as

hereditary nonpolyposis colorectal cancer (HNPCC); 2) Constitutional Mismatch Repair

Deficiency Syndrome (CMMRD); and 3) Polymerase Proofreading Associated Polyposis (PPAP).

1.3.1 Lynch Syndrome/HNPCC

The recorded history of LS, one of the most prevalent cancer predisposition syndromes, dates back

to 1913 with the characterization and pedigree mapping of ‘Family G’, one of the first

comprehensive recordings of familial cancers consisting mainly of colorectal, stomach, and

genitourinary cancers, with many cases displaying synchronous and/or metachronous tumors

(Warthin 1913). The disease was revisited by Lynch et al. in 1966, upon the characterization of

additional families, ‘Family N’ and ‘Family M’, both of whom displayed the same autosomal

dominant pattern of inheritance and cluster of cancer types as those characterized by Warthin

(Lynch et al. 1966). Prior to the identification of the genetic basis for LS throughout the 1990’s,

early work among LS families focused on carefully defining the clinical presentation of the

syndrome. It was observed in patients with LS that polyps arise at the same rate as the normal

population, however such polyps evolve into malignant tumors at a faster rate. This accelerated

adenoma-to-carcinoma transition is now a regarded as a hallmark of LS, a feature which

distinguishes it from familial adenomatous polyposis (FAP; characterized by multiple colonic

adenomatous polyps), hence the alternative name for LS, HNPCC (Boland et al. 1984, Tempero

et al. 1984). While early onset (< 50 years of age) colorectal cancer (CRC) and cancers of the

endometrium are the most common cancer types associated with LS, LS-associated cancers have

since expanded to include stomach, small bowel, ovarian, pancreas, ureter or renal pelvis,

hepatobiliary tract, and brain tumors (Watson et al. 2008), as well as sebaceous gland adenomas

and keratoacanthomas in a variant of LS, Muir–Torre syndrome (Lynch et al. 1981).

8

In parallel to the clinical characterization of LS, a series of basic science discoveries were made

which made possible the uncovering of the genetic basis for LS. Microsatellite instability (MSI),

the presence of frameshifts in tandem sequence repeats scattered throughout the genome, was a

phenomenon first discovered in studies on bacteriophages (Okada et al. 1972, Streisinger et al.

1985), and then subsequently seen to occur with higher frequency when these repeat tracts were

introduced into E. coli strains that were defective in MMR (Levinson et al. 1987). The same

phenomenon of increased frequency of MSI was also observed in MMRD yeast (Strand et al.

1993). Finally, Aaltonen and colleagues were the first to detect MSI in human cancers, in

particular, those cancers associated with LS (Aaltonen et al. 1993), and subsequent work

demonstrated that LS cells could not repair defects in repeat sequences nor single base mismatches,

thus providing a link between defective MMR and LS (Parsons et al. 1993). Based on these

findings, MSI screening in LS-associated cancers was adopted as a tool for the detection of this

disease and MMRD-driven cancers. Moreover, these findings provided early insight into the

hypermutability of MMRD cancers.

Once the links between MMRD, MSI, and LS were established, successive studies began mapping

the genetic loci responsible for LS through a combination of linkage analyses and positional

cloning. The MSH2 gene was the first to be mapped in 1993, with deleterious mutations

segregating with cancers in LS families (Fishel et al. 1993, Leach et al. 1993), followed by MLH1

(Bronner et al. 1994, Papadopoulos et al. 1994) and PMS2 (Nicolaides et al. 1994) in 1994, and

finally MSH6 in 1997 (Miyaki et al. 1997). Alterations in MLH1 are the most common causes of

LS, followed closely by alterations in MSH2, fewer cases caused by MSH6 mutations, and a

minority of cases caused by PMS2 mutations (Plazzer et al. 2013). This allelic distribution has

been explained by overlapping and redundant MMR activities of the various MSH and MLH

heterodimers (Fishel 2001), and caveated by the observation that some individuals with LS will

never develop cancers and those are likely individuals with PMS2 alterations (see Section 1.3.2

below on how this affects the diagnosis of CMMRD). Mutations in MMR lead to the inactivation

of the gene and thus can be point mutations, indels, duplications, or larger genomic

rearrangements, making detection via molecular genetics mechanisms sometimes difficult. A

curated database of cancer causing mutations in LS has been established by the International

Society for Gastrointestinal and Hereditary Tumors (InSiGHT) in an effort to aid in the diagnosis

of this and other hereditary CRC associated diseases (Plazzer et al. 2013). In addition to genetic

9

alterations in these four MMR genes, epigenetic alterations including hypermethylation of the

MLH1 promoter (associated with both hereditary and non-hereditary CRC) (Hitchins et al. 2005,

Hitchins et al. 2011) and hypermethylation of MSH2 caused by deletions in the adjacent gene,

epithelial cell adhesion molecule (EPCAM) (Ligtenberg et al. 2009), also contribute to the

aetiology of LS. Finally, consistent with the two-hit model of carcinogenesis, Hemminki et al

demonstrated that loss of the remaining wildtype allele of the mutated MMR gene consistently

occurred in LS (Hemminki et al. 1994). This observation that tumors from LS patients had

somatically acquired a second MMR mutation, thus leading to complete MMRD, was in

concordance the mutator hypothesis (discussed in Section 1.2 above). With the advent of next-

generation sequencing (NGS) technologies, it became possible to more comprehensively profile

the genetic landscape of human cancers and, due to their underlying MMRD and inability to

efficiently correct errors associated with DNA replication, LS tumors were found to harbour

hundreds to thousands of point mutations and small indels, far surpassing mutational burdens of

MMR-proficient cancers (Cancer Genome Atlas Network 2012).

The resulting loss of MMR protein expression has led to the routine use of immunohistochemistry

(IHC) to determine the MMR status of potential LS-associated tumors and gain insight into the

causative mutant MMR gene. Generally, loss of MSH6 or PSM2 by IHC points to alterations in

their respective genes, while loss of both MSH2 and MSH6 implies underlying MSH2 mutations

since the MSH6 protein is unstable in the absence of MSH2. The same is true for dual MLH1 and

PMS2 loss, which points to genetic alterations in MLH1 since this would destabilize PMS2.

Together, IHC and MSI screening provide robust, low-cost methods of detecting MMRD tumors

and suspected LS. Much of the foundational work on LS has informed the study of its childhood

predisposition counterpart, CMMRD.

1.3.2 Constitutional/Biallelic Mismatch Repair Deficiency Syndrome

CMMRD (sometimes referred to as BMMRD and previously known as Turcot’s Syndrome), like

LS, is typically caused by underlying genetic deficiencies in one of the four MMR genes. However,

while LS results from the inheritance of a single defective copy of an MMR gene, CMMRD is

caused by the biallelic inheritance of defective MMR genes. Some instances of CMMRD may also

be caused by compound heterozygous mutations in two of the four MMR genes (also referred to

10

as “digenic Lynch”). Like LS, the underlying cause of CMMRD was unknown prior to the genomic

era where MMRD, MSI and cancer initiation were linked. Indeed, initial descriptions of what was

then called Turcot’s Syndrome were a series of case reports summarizing findings of synchronous

or metachronous intestinal polyposis and central nervous system (CNS) cancers in children (Turcot

et al. 1959, Itoh et al. 1985). Although initial findings from Turcot were never validated at the

molecular level, it is likely these cases refer to the earliest reports of CMMRD given the clinical

presentations described. Moreover, early descriptions of CMMRD were likely conflated with

childhood onset FAP making it difficult to tease apart the genetic basis for the disease. Turcot’s

patients demonstrated DNA instability in both tumor and normal tissues, unlike LS patients which

only demonstrated DNA instability in their tumors, suggesting the underlying genetic basis for

Turcot’s was homozygous mutations in MMR genes (Miyaki et al. 1997). It was not until 1999

that the cause of this childhood onset predisposition syndrome was revealed to be defective MMR

with two studies identifying germline biallelic MLH1 deficiency as the cause of synchronous

childhood onset neurofibromatosis and hematological malignancies (Ricciardone et al. 1999,

Wang et al. 1999). As further clinical observations were made and the tumor spectrum of CMMRD

expanded beyond CRC and CNS tumors, the syndrome was renamed to better reflect the

underlying molecular basis for disease, hence CMMRD/BMMRD. ‘CMMRD’ will be used

throughout this thesis for consistency.

As was the case with LS, the spectrum of CMMRD-associated tumors and clinical presentation of

CMMRD has expanded and evolved over the last 20 years. The three most common types of cancer

associated with CMMRD include 1) hematological malignancies, 2) malignant brain tumors

(MBTs), and 3) CRCs or other LS-associated cancers occurring at a median age of 6, 9, and 17

years respectively, although some cancers may develop as early as infancy (Wimmer et al. 2014).

Moreover, due to germline ablation of MMR, a multitude of other cancer types have additionally

been observed with less frequency in CMMRD patients including skin cancers, osteosarcomas,

and recently even breast cancer (Bush et al. 2019) (summary of tumor frequencies in CMMRD

can be found in Figure 1.3-1). Although CMMRD is significantly less common and well-described

than its adult-onset counterpart, reported data demonstrates the disease is universally lethal with

individuals rarely surviving into adulthood (Lavoine et al. 2015). The cause of death in CMMRD

patients is usually their MBTs, most commonly high-grade gliomas (HGGs), as these tumors

typically have poor prognosis, compared to extracranial tumors. The acquisition of synchronous

11

and metachronous cancers is a hallmark of CMMRD (Vasen et al. 2014) and may act as a clue in

cases where CMMRD has yet to be diagnosed. In addition to the cancer-related clinical

presentations of disease, CMMRD patients can also present with non-neoplastic features, most

commonly café-au-lait macules (CALMs) and other signs reminiscent of neurofibromatosis-1

(NF1) as well as occasionally skin hypopigmentation or defects in immunoglobulin class switching

(Wimmer et al. 2014). Consanguinity is reported in nearly half of CMMRD families, and

additional homozygous mutations have been detected in families not reporting consanguinity,

suggesting a common founder mutation, particularly in isolated populations (Durno et al. 2015,

Lavoine et al. 2015, Amayiri et al. 2016).

Figure 1.3-1 Tumor spectrum and Incidence for CMMRD patients within the International Replication Repair

Deficiency Consortium (IRRDC).

Data collected by the author and Dr. Uri Tabori on behalf of the IRRDC. Durno, Ercan, Bianchi et al. J. Clin. Oncol.

(2021; in press).

Tumor Spectrum of CMMRD Patients in IRRDC

Brain Tumors* 44%

High Grade Gliomas 82%

Medulloblastomas 12%

CNS Embryonal 6%

GI Cancers 27%

Colorectal Cancers 71%

Small Bowel Cancers 21%

Stomach Cancers 6%

Liver Cancer 2%

Hematological

Malignancies

19%

Lymphoma 70%

Non-Hodgkin T-cell 48%

Non-Hodgkin B-cell 19%

Non-Hodgkin Unspecified 3%

Leukemia 30%

ALL 19%

AML 11%

Total Patients = 105

Total Tumors = 193

Other Malignant Tumors 10%

Retinoblastoma 5.3%

Breast Cancer 10.5%

Skin Cancer 15.8%

Neuroblastoma 5.3%

Wilms Tumor 10.5%

Kidney Cancer (other) 10.5%

Bladder Cancer 10.5%

Ovarian Cancer 5.3%

Endometrial Cancer 5.3%

Prostate Cancer 5.3%

Sarcoma 15.8%

12

Despite ongoing international collaboration through the European consortium “Care for CMMRD”

(C4CMMRD) and the International Replication Repair Deficiency Consortium (IRRDC), both

established to aid in the study of CMMRD, many challenges exist in the diagnosis and management

of CMMRD patients. Firstly, most cases of CMMRD (> 60%) are due to biallelic mutations in

PMS2, a notoriously difficult gene to sequence due to the presence of multiple pseudogenes,

followed by mutations in MSH6 (~20-30%), and then MLH1 and MSH2 (combined ~10-20% of

cases) (Vasen et al. 2014, Lavoine et al. 2015). The difficulty of PMS2 gene sequencing increases

the uncertainty of detected variants leading to higher numbers of variants of unknown significance

(VUSs). Moreover, the allelic frequencies observed in CMMRD are an inversion of those seen in

LS (see Section 1.3.1 above). Thus, because those with a germline heterozygous PMS2 mutation

are less likely to present with cancer than individuals with germline heterozygous MSH2 or MLH1

mutations (particularly those classical LS cancers including CRC and endometrial cancer), many

CMMRD families will not have a family history of cancer pointing to suspected LS (Bakry et al.

2014). Moreover, MSI detection, via the gold standard, low-cost Promega panel, which is often

used in the diagnosis of LS-associated cancers, is not effective in the diagnosis of CMMRD as

many tumors will appear MS stable (MSS). CMMRD tumors are, in fact, MSI-high, although NGS

tools are needed to more accurately assess the many MS loci throughout the genome beyond the

five loci included in the Promega panel (Chung et al. 2020). IHC, however, remains an accurate

method of detecting MMRD in both tumor and normal tissue for those patients presenting with

classical CMMRD (Bakry et al. 2014). Finally, treatment of CMMRD patients is difficult as

MMRD cancers are resistant to common first line chemotherapies (discussed in Section 1.4

below).

As with other cancers caused by germline or sporadic MMRD, CMMRD cancers are generally

hypermutant (>10 mut/Mb), with some subsets of tumors exhibiting ultrahypermutation (>100

mut/Mb) due to combined deficiencies in MMR and somatic polymerase proofreading loss (Shlien

et al. 2015, Campbell et al. 2017). Among pediatric cancers, CMMRD tumors are strikingly unique

in this hypermutated phenotype, as most childhood malignancies are characterized by incredibly

low mutational burdens. Moreover, because of this endogenous repair deficiency and

hypermutation, CMMRD tumors also exhibit specific mutational signatures, which are identifiable

patterns of SNVs that leave “imprints” on the genome and suggest a specific disease aetiology

(Alexandrov et al. 2013, Alexandrov et al. 2020). CMMRD cancers are enriched for specific

13

signatures including signatures 6, 15, 21, 26, and recently 44. While these genomic features of

high tumor mutational burden (TMB) and specific mutational and MS-indel signatures can aid in

diagnosis of CMMRD, and may prompt further investigation if detected, they are alone, not

sufficient. Further, this hypermutant phenotype may mask actionable tumor-promoting mutations,

making the study and development of potential treatments for CMMRD patients difficult.

1.3.3 Polymerase Proofreading Associated Polyposis

PPAP is the most recent of the three germline RRD-related syndromes to be uncovered. Prior to

the first reports of PPAP, several decades of research in non-human organisms had demonstrated

the mutator phenotype conferred by DNA polymerase proofreading loss (Shcherbakova et al.

2003, Fortune et al. 2005, Murphy et al. 2006) and suggested a role for DNA polymerase

proofreading deficiency (PPD) in carcinogenesis (Goldsby et al. 2001, Goldsby et al. 2002,

Albertson et al. 2009). Moreover, base pair substitutions in both the POLE and POLD1

exonuclease domains had previously been detected in human CRCs and CRC cell lines (da Costa

et al. 1995, Flohr et al. 1999, Yoshida et al. 2011, Cancer Genome Atlas Network 2012). The

Cancer Genome Atlas Network had also reported that POLE missense mutations were enriched

among hypermutated CRCs along with the alterations and epigenetic silencing of the MMR genes.

However, it was not until 2013 that the first cases of germline polymerase mutations as the

underlying genetic cause of a dominantly inherited predisposition syndrome were reported (Palles

et al. 2013). Two exonuclease domain mutations were identified in this publication: POLE L424V

and POLD1 S478N. The authors also noted that in addition to being hypermutant, all tumors

examined were MSS unlike adenomas and CRCs observed in LS, which are MSI-high.

The first cases of reported PPAP described a mild phenotype of adult-onset polyposis where

individuals developed tens of adenomas that did not always progress to invasive carcinomas (hence

the name PPAP). The tumor spectrum of PPAP, however, like LS, also includes high grade CRCs,

duodenal adenomas, and carcinomas, as well as extraintestinal lesions including genitourinary

cancers, high grade gliomas (HGG), and other benign lesions (Elsayed et al. 2015, Rosner et al.

2015, Spier et al. 2015). Although less well characterized than LS, the lifetime risk of CRC in

individuals with PPAP has been estimated to be 32-40% for POLE mutations carriers and 52-63%

for POLD1 carriers by 70 years of age (Buchanan et al. 2018). However, these predictions perhaps

underestimate the penetrance of this syndrome, as the authors likely included VUSs that are not

14

necessarily supported by yeast functional studies, good co-segregation among families, or

evidence from somatic events at the same residue. This is particularly an issue for POLE, which

is a large gene, and therefore more likely to incur VUSs. More recent reports have predicted the

lifetime penetrance of CRCs in individuals with familial POLE mutations is 11.1%, 48.5%, and

74% at 30, 50, and 70 years, respectively (Hamzaoui et al. 2020). The authors also noted that while

CRCs indeed possessed the highest incidence rate among PPAP individuals (present in 74% of

carriers), the incidence of HGGs (present in 15% of carriers) was significant enough to warrant

further discussion on inclusion in surveillance protocols. Notably, HGGs tend to occur in children

and young adults prior to CRCs which tend to occur later in life, although HGGs were observed

up to age 66 in affected individuals.

As more PPAP cases are reported, both the tumor spectrum and spectrum of cancer-causing

mutations have expanded (Rohlin et al. 2014, Valle et al. 2014, Aoude et al. 2015, Bellido et al.

2016). Nearly all catalogued mutations have been missense mutations found in the exonuclease

domains (or adjacent to the exonuclease domains) of either DNA polymerase (Figure 1.3-2). This

is expected since such mutations would need to impair only the protein’s exonuclease function

without disrupting essential polymerase function. In addition to the repertoire of germline

mutations detected, a larger, but overlapping group of polymerase exonuclease domain mutations

have been detected as underlying somatic events across human cancers (Church et al. 2013,

Kandoth et al. 2013, Rayner et al. 2016, Campbell et al. 2017). Both germline and somatic PPD

driven tumors are hypermutant, with mutational burdens higher than MMRD cancers (typically

>100 mut/Mb). Moreover, as with MMRD tumors, POLE driven cancers display specific

mutational signatures, most notably signature 10 (Alexandrov et al. 2013, Alexandrov et al. 2020).

Cancers driven by combined POLE/MMR mutations display signature 14, and cancers driven by

combined POLD1/MMR mutations display signature 20 (Haradhvala et al. 2018).

Despite recent large scale genomics studies that have shed light on several key characteristics of

PPD cancers, questions remain regarding the genotype/phenotype relationship of the different

polymerase mutations, particularly among POLE mutations which are more common across

human cancers. For example, we recently demonstrated that different amino acid substitutions at

the same protein residue in the DNA polymerases result in varying impacts on exonuclease

function and therefore TMB (Campbell et al. 2017). While studies in yeast and in vitro experiments

15

suggest that different mutations have variable impact on exonuclease function (Shinbrot et al.

2014, Barbari et al. 2018, Xing et al. 2019), more research is needed to study the impact of different

polymerase mutations on cancer predisposition in vivo.

Figure 1.3-2 Frequency and location of germline and somatic Pol δ and Pol ε proofreading exonuclease domain

mutations in cancers

A schematic representation of the exonuclease domains of Pol δ (part a) and Pol ε (part b) showing conserved exo

motifs (I–V), exo I active site residues, and the position and frequency of germline and somatic mutations. Exo sites

are highlighted in purple. Note that Pol ε residues L424V and P436R/S are the site of both germline and somatic

mutations and are colored according to which alteration is more frequent. Reused with permission from Nat. Rev.

Cancer (Rayner et al. 2016).

1.4 Treatment of Replication Repair Deficient Cancers

Given the diverse spectrum of cancer types across RRD driven cancers (discussed in Section 1.3

above) combined with the increased chance of synchronous tumors, treatment management for

such patients can be difficult. Moreover, underlying MMRD and PPD may preclude the use of

common therapeutic agents. Since much of this thesis focuses on novel treatments for RRD

cancers, it is necessary to explore the efficacy of current standards of care and emerging areas of

therapeutic exploration, particularly as they relate to RRD brain tumors which are commonly the

cause of death in these patients. Treatments for CRCs and hematological malignancies are also

discussed.

1.4.1 Chemotherapies & Radiation

The current standard of care for HGGs involves total surgical resection followed by radiation and

concomitant administration of the DNA-alkylating agent temozolomide (TMZ) (Stupp et al. 2005).

While the role of MMR in TMZ mediated cytotoxicity has been well studied (discussed further

below in this section), the data surrounding the efficacy of ionizing radiation (IR) in MMRD

16

cancers is controversial, with several studies demonstrating MMRD confers resistance to IR

(Brown et al. 2003) and others suggesting no effect at all (Cejka et al. 2004). Initial studies

examining the role of MMR in the DNA damage response post-IR focused on the ability of MMR

to recognize radiation induced damage such as 8-oxoguanine (8-oxoG). In these studies, which

utilized mouse embryonic fibroblasts (MEFs) deficient and proficient in one of the four MMR

proteins, authors found that MMRD was associated with radioresistance and accumulation of 8-

oxoG (Fritzell et al. 1997, DeWeese et al. 1998). However, further studies suggested that MMR

proteins are more likely to be involved in the repair of double stranded breaks (DSBs) induced by

IR since DSBs are the most deleterious source of damage caused by IR. Moreover, homologous

recombination (HR), one of the repair pathways essential for repair of DSBs, involves MMR

proteins (Yan et al. 2001). The proposed mechanism for the sensitivity of MMR-proficient cells

to IR is that MMR proteins are involved in DSB sensing and promote a G2/M cell cycle arrest and

ultimately cell death by apoptosis. More recent data demonstrates MMR-dependent sensitivity

may depend on radiation dose rate. MMR proficiency appears to be associated with

hypersensitivity to prolonged low dose rate IR (Martin et al. 2009), while MMRD appears to be

associated with radiosensitivity following acute high dose-rate IR (Franchitto et al. 2003). The role

of IR in the treatment of MMRD cancers remains a topic of interest and emerging questions

additionally seek to clarify whether any synergy exists between IR and the delivery of immune

checkpoint inhibitors (ICIs) (further explored in section 1.4.3 below).

TMZ and other SN1-type chemotherapeutic agents such as procarbazine, N-methyl-N’-nitro-N-

nitrosoguanidine (MNNG), and N-methyl-N-nitrosourea (MNU) trigger their cytotoxic effects

through the methylation of guanine to produce O6-methylguanine (MeG), which triggers cell cycle

arrest and/or cell death. The mechanism through which these agents exert their effects has been

shown to be dependent on intact MMR. Early studies in bacteria demonstrated that the cytotoxic

effects of MNNG on E. coli could be ameliorated with the introduction of either a mutL or mutS

mutation (Karran et al. 1982). Further work in human and mouse cell lines defective in either

MutSα or MutLα demonstrated these cells were highly resistant to MeG-inducing agents (Levati et

al. 1998, Humbert et al. 1999, Takagi et al. 2003), and tolerant to MeG even at the expense of point

mutations induced (Branch et al. 1993). Evidence suggests that although the MMR system does

not detect the MeG lesion directly, it does recognize the MeG-thymine mismatch that occurs after

erroneous incorporation of a thymine rather than a cytosine opposite the MeG by DNA polymerases

17

during the next replication cycle. From here, two possible theories suggest either that MMR acts

as a DNA damage sensor to signal apoptosis directly, or that MMR launches into “futile cycling”.

In this latter theory, post-replicative MMR removes stretches of single stranded DNA (ssDNA)

containing thymines opposite MeG and creating a gap in DNA, which is filled by DNA polymerases

that reinsert thymine opposite MeG, stimulating another round of MMR. This “futile cycling”

process of MMR results in the accumulation of ssDNA and generation of double stranded breaks

after several rounds of replication, leading to cell cycle arrest and apoptosis (Mojas et al. 2007,

Quiros et al. 2010). For example, small decreases in the expression of MSH2, for example, have

been shown to lead to striking resistance to patient derived glioblastoma cell lines in vitro

(McFaline-Figueroa et al. 2015), and resistance to MeG-inducing agents has even been proposed as

a screening tool to identify CMMRD individuals (Bodo et al. 2015). Moreover, many recurrent

HGGs that demonstrate resistance to temozolomide, are shown to have acquired MMRD,

suggesting tumors select for MMRD cells (Cahill et al. 2007, Yip et al. 2009, Felsberg et al. 2011).

Several other chemotherapies have also been shown to create DNA adducts which result in

mismatches recognized by the MMR system, thus either directly signalling apoptosis and cell

death or engaging in futile repair cycles as described for methylating agents. Several commonly

used anti-metabolites—compounds that mimic DNA nucleotides but interfere with normal cellular

function—have been shown to be less effective in MMRD cancers likely due to lack of

downstream mismatch recognition. 5-Fluorourocil (5-FU), a pyrimidine analogue used in CRC

treatment does not improve survival in MSI-high CRCs but does improve survival in MSS CRCs

(Ribic et al. 2003, Carethers et al. 2004). These findings are supported by in vitro studies

demonstrating that MMR recognizes mismatches generated by 5-FU lesions (Meyers et al. 2005).

Likewise, thiopurines such as 6-thioguanine (6-TG) and 6-mercaptopurine (6-MP), have

commonly been used in the treatment of childhood leukemia as well as other diseases such as

inflammatory bowel disease (IBS) and as immunosuppressants post-transplant. Active MMR has

been shown to be required for the cytotoxic action of these drugs (Swann et al. 1996), and human

leukemia cells lacking MSH2 demonstrate resistance to thiopurines (Diouf et al. 2011).

Immunosuppression post-transplant using the thiopurine azathioprine has also been correlated with

increased risk of MMRD acute myeloid leukemia (AML), suggesting a selective pressure for

MMRD azathioprine-resistance cells (Offman et al. 2004). MMRD cancers are also resistant to

several platinum salts, such as cisplatin, which is used in the treatment of medulloblastoma

18

(another common brain tumor present in CMMRD individuals) (Aebi et al. 1996, Drummond et

al. 1996).

Thus, the use of these agents in cancers driven by MMRD is likely to be unsuccessful, which may

prove clinically challenging particularly when patients have not yet been diagnosed as CMMRD

or present with synchronous cancers (Amayiri et al. 2016) and novel treatment options are needed.

1.4.2 Targeted & Synthetic Lethal Approaches

Due to the hypermutant status of RRD cancers, it can be difficult to determine which mutations

are true drivers and which are passengers and will not be targetable. Moreover, the mutator

phenotype conferred by underlying RRD likely means that tumors will easily develop resistance

to this type of approach by accumulating additional mutations.

For example, although several targeted monoclonal antibodies have significantly improved

outcomes for acute lymphoblastic leukemia, this may not be applicable for RRD cancers. These

therapies include the bispecific T-cell engaging antibody (BiTE®) blinatumomab, which redirects

T-cells to lyse CD19 positive B cells (Bargou et al. 2008, Klinger et al. 2012), as well as the

conjugated anti-CD22 antibody, inotuzumab ozogamicin, which, once internalized within CD22

expressing cells, releases the DNA-damaging agent calicheamicin, causing DSBs and eventually

cell death (DiJoseph et al. 2004). Both are used in the treatment of refractory B-cell acute

lymphoblastic leukaemia (B-ALL). Although not fully explored in MMRD driven B-ALL, a recent

case report suggests these targeted approaches may not be useful in the context of CMMRD

(Oshrine et al. 2019). In this report, the authors described one CMMRD patient’s progression and

death following combination blinatumomab and inotuzumab ozogamicin for the treatment of his

refractory B-ALL. Analysis of the tumor after administration of combined targeted therapy

demonstrated expansion of a CD19— population, possibly due to hypermutation and selection of

this malignant clone, which was shown to have evolved many new mutations (Oshrine et al. 2019).

Notably, loss of CD19 has been shown to occur only in a minority of blinatumomab non-

responders (Jabbour et al. 2018). Although this report provides just one case, it suggests that the

CMMRD mutator phenotype may preclude the use of such targeted therapies for MMRD

hematological malignancies.

19

Nevertheless, uncovering synthetic lethal targets and oncogene addictions in these cancers may

result in more favorable clinical outcomes. Several new studies have provided insight into the

efficacy of synthetic lethal and targetable approaches for MMRD/MSI-high cancers.

In examining data from large-scale cancer cell line screens and functional genomics assays, several

recent studies have found that MSI-high cancer cells (due to MMRD) are selectively vulnerable to

inhibition of Werner syndrome helicase (WRN), a RecQ family DNA helicase with diverse roles

in DNA repair, replication, transcription, and telomere maintenance (Chu et al. 2009, Behan et al.

2019, Chan et al. 2019, Kategaya et al. 2019, Lieb et al. 2019). All four studies reporting the

synergism between dual ablation of MMR and WRN, validated these findings in vitro using MSI-

high and MSS cell lines, demonstrating selective WRN dependency only in MSI-high cell lines

specifically on the helicase activity of WRN and not its exonuclease activity. Two studies also

tested this dependency in vivo using xenografted MSI high colon cancer cell lines and showed that

WRN inhibition was sufficient to slow or suppress tumor growth (Behan et al. 2019, Chan et al.

2019). Interestingly, while reconstitution of WRN rescued proliferation and viability phenotypes

in MSI high cells, re-expression of MLH1 in deficient cell lines did not rescue WRN knockdown

(Chan et al. 2019, Kategaya et al. 2019, Lieb et al. 2019). Moreover, knockout of MMR in MSS

cell lines was not sufficient to induce WRN dependency (Behan et al. 2019), suggesting the

efficacy of WRN inhibition depends on the presence of specific genomic lesions (MSI). Three of

these studies demonstrated WRN inhibition in the context of MSI induces DSBs, cell cycle arrest,

and apoptosis although the exact mechanism through which WRN protects from these DSB events

was unclear (Chan et al. 2019, Kategaya et al. 2019, Lieb et al. 2019). However, one recent study

demonstrated that expanded TA-dinucleotide repeats in MSI-high cells form cruciform-like

structures which stall replication forks, activate the ATR checkpoint kinase, and require unwinding

by the WRN helicase, thus providing a mechanism for dependence of MMRD cells on WRN

activity (van Wietmarschen et al. 2020). Although further work is required to examine the presence

of these specific repeat expansions across MMRD cancers, these promising findings provide a

biomarker for the selection of patients for WRN inhibition.

Another recent study identified proteome instability as a therapeutic vulnerability in MMRD

cancers (McGrail et al. 2020). The authors utilized isogenic MMR proficient and deficient cell

lines along with data from MSI-high TCGA cancers to develop an MMRD transcriptomic

20

signature and identify potential therapeutics for MSI-high cancers. Using this method, the Nedd8-

Activating Enzyme E1 (NAE) inhibitor MLN4924 was identified. The authors demonstrated that

MMRD-induced destabilizing mutations lead to proteome instability and an abundance of

misfolded protein aggregates, requiring a Nedd8-mediated degradation pathway for clearance of

these misfolded proteins. Blockade of this pathway using MLN4924 led to accumulation of protein

aggregates and immunogenic cell death. The authors also demonstrated combinatorial anti-PD-1

and NAE inhibition in vivo was more efficacious than either monotherapy in the treatment of an

Msh2-null mouse cancer cell line, suggesting blockade of neddylation potentiates immune

mediated rejection. The mechanism through which proteome instability increases the

immunogenicity of MMRD tumors is still unclear.

While our catalogues of cancer-causing mutations and genes have become increasingly more

detailed (Kandoth et al. 2013, Lawrence et al. 2014), it can be difficult to identify which genomic

alterations are true driver mutations, particularly in the context of hypermutant cancers. One study

that sought to quantify and identify driver mutations across 29 cancer types, demonstrated that as

TMB increased, so did the number of driver mutations, although the increase was sublinear

(Martincorena et al. 2017). The same authors additionally demonstrated that missense mutations

in certain genes were under positive selection regardless of amino acid alteration. Of the top six

genes, four of them were KRAS, HRAS, NRAS, and BRAF. Another study that similarly sought to

identify new mutational hotspots across 322 cancer types (32 organ sites) demonstrated that

signalling members of the RAS/MAPK downstream of the typical receptor tyrosine kinases, such

as MAP2K1, were found to be more frequently mutated than expected by chance (Chang et al.

2018). Although the RAS/MAPK pathway has not yet been explored extensively in the context of

RRD cancers, activation of the MAPK pathway commonly drives pediatric gliomas and other

cancers (Zhang et al. 2013, Northcott et al. 2015, Ryall et al. 2016, Guerreiro Stucklin et al. 2019).

Further, the large size of the NF1 gene, which codes for a tumor suppressor and negative regulator

of the MAPK pathway, makes it particularly susceptible to mutation in the context of mutator

phenotypes conferred by RRD. Indeed, the notable presence of CALMs in CMMRD individuals,

often resulting in suspicion of NF-1 rather than CMMRD, suggests somatic NF1 events may occur

in CMMRD. Moreover, the availability of inhibitors targeting members of the RAS/MAPK

pathway, such as MEK (selumetinib and trametinib) and BRAF (dabrafenib and vemurafenib),

makes this an attractive pathway to investigate in the context of RRD.

21

1.4.3 Immune Checkpoint Inhibition

Finally, ICIs, arguably one of the greatest advances in adult oncology in recent years, harness the

power of an individual’s immune system to provoke an anti-tumor response. At the centre of ICI

efficacy are the mechanisms balancing T lymphocyte activation and tolerance. For a T-cell to

launch an immune response, its T-cell receptor (TCR) must engage its specific antigen via an

antigen presenting cell (APC) and CD80/CD86 co-stimulation must occur via CD28. In addition

to this two-step model of activation, co-inhibitory signalling pathways can negatively regulate and

prevent T-cell activation, which is critical to immune homeostasis (Chen et al. 2013).

Due to their genomic alterations, tumors can display what are referred to as neoantigens that no

longer resemble “self” and therefore provoke T-cell attack. Much of the seminal work supporting

a role for T-cells in tumor immune surveillance came from murine model experiments

demonstrating the role of T-cells or their effector molecules in the control of tumor growth

(Shankaran et al. 2001). To persist, many cancers evolve immune evasion strategies that can

include co-opting co-inhibitory immune checkpoints. Two checkpoints important in the context of

cancer immunotherapy are the cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) and

programmed cell death protein-1 (PD-1) checkpoints. CTLA-4, which is expressed on newly

activated T-cells, prevents T-cell mediated attack by engaging the same ligands as CD28

(CD80/CD86), but with higher affinity, thereby outcompeting co-stimulatory signals and

transmitting inhibitory signals to suppress T-cell activation (Walunas et al. 1994, Krummel et al.

1995). PD-1, conversely, has been shown to be upregulated on T-cells following prolonged antigen

exposure, promoting T-cell exhaustion (Barber et al. 2006, Wei et al. 2017) via engagement of

PD-1 ligands (PD-L1/PD-L2), which are commonly upregulated on tumor cells (Dong et al. 2002).

Thus, ICI seeks to sensitize or re-sensitize T-cells to their antigenic targets on tumor cells by

targeting CTLA-4 and/or PD-1 using humanized monoclonal antibodies, and thereby “removing

the breaks” on T-cell mediated tumor immune surveillance.

The first commercially available ICI that received FDA approval for use in cancer therapy was

ipilimumab, a monoclonal antibody targeting CTLA-4, following its success in metastatic

melanoma patients (Hodi et al. 2010). Indeed, long term assessments of ipilimumab have

demonstrated 22% survival for all patients at 3-years with long term follow-up of up to 10 years,

compared to 12% in patients treated with standard chemotherapy (Schadendorf et al. 2015).

22

The first commercially available FDA-approved anti-PD-1 ICIs were nivolumab and

pembrolizumab, following their success in the treatment of unresectable or metastatic melanoma,

where they outperformed ipilimumab (Ribas et al. 2015, Robert et al. 2015, Weber et al. 2017),

with fewer high-grade adverse events (Ribas et al. 2016). In addition to fewer severe side effects,

PD-1 blockade was also shown to be more efficacious in a wider spectrum of cancers than CTLA-

4 blockade. In addition to melanoma, PD-1 blockade has proven effective and approved for use in

non-small cell lung carcinoma (NSCLC) (Brahmer et al. 2015, Herbst et al. 2016), small cell lung

carcinoma (Antonia et al. 2016), urothelial bladder cancer (Sharma et al. 2017), Hodgkin’s

lymphoma (Ansell et al. 2015), advanced head and neck squamous cell cancer (HNSCC) (Ferris

et al. 2016), Merkel cell carcinoma (Nghiem et al. 2019), renal cell carcinoma (RCC) (Motzer et

al. 2015), hepatocellular carcinoma (El-Khoueiry et al. 2017), primary mediastinal large B-cell

lymphoma (PMBCL) (Zinzani et al. 2017), gastroesophageal junction (GEJ) adenocarcinoma

(Fuchs et al. 2018), cervical cancer (Chung et al. 2019), and MSI-high/MMRD CRCs (discussed

further in this section below) (Overman et al. 2017). Combinatorial PD-1 and CTLA-4 blockade

regimes have also proven more efficacious than either monotherapy in certain cancers, albeit with

increased instances of immune adverse events (Larkin et al. 2015, Postow et al. 2015).

Finally, disruption of the PD-1 checkpoint pathway has also proved feasible using inhibitors

targeting PD-1’s ligand, PD-L1. Atezolizumab, durvalumab, and avelumab are anti-PD-L1

monoclonal antibodies that have shown efficacy and been approved for use in a similar range of

cancers to PD-1 inhibitors (Kaufman et al. 2016, Massard et al. 2016, Reck et al. 2019).

While durable, long-term responses have been life-changing for a minority of individuals, most

patients will continue to experience disease progression. Given that a significant number of

patients do not benefit from ICIs, much of the research surrounding immunotherapy has aimed to

identify predictive biomarkers. Given the scope of this thesis, discussion on some of the most

salient biomarkers including several genomic and microenvironment specific markers is

necessary.

One of the first biomarkers reported by several studies shortly after the approval of ipilimumab

was high mutational loads (Snyder et al. 2014, Rizvi et al. 2015, Van Allen et al. 2015). All three

studies found that high TMB (100 nonsynonymous mutations per exome or median 308

nonsynonymous mutations) was significantly associated with clinical benefit from ICI. Moreover,

23

using patient human leukocyte antigen (HLA)-type and neoantigen prediction tools, the authors

demonstrated that increased mutational burdens correlated with predicted neoantigens and clinical

benefit. This supported the “neoantigen roulette” theory, the idea that higher numbers of

nonsynonymous mutations would lead to a greater chance of “winning” neoantigens that would be

presented by a tumor cell’s major histocompatibility complex (MHC) and incite a T-cell response

(Gubin et al. 2015). These findings explained the success of immunotherapy in cancers commonly

associated with hypermutation such as melanoma, NSCLC, and MSI-high CRCs (Alexandrov et

al. 2013). Indeed, aggregate data demonstrate significant pan-cancer association between TMB

and ICI response by tumor type (Yarchoan et al. 2017), with exceptions seen in some tumor types

including RCC and human papilloma virus (HPV)-positive HNSCC where no association between

pre-treatment TMB and clinical benefit was seen. Moreover, neither TMB nor total neoantigen

load alone can clearly discriminate responders from non-responders. To refine TMB prediction,

one study demonstrated an inverse relationship between intratumoral heterogeneity and overall

survival, showing that clonal neoantigens together with increased TMB was more significantly

associated with clinical benefit (McGranahan et al. 2016). Another study demonstrated that a large

change in TMB (∆TMB) during anti-PD-1 therapy was strongly associated with response and

survival in patients with melanoma (Riaz et al. 2017), although this biomarker metric may be

clinically difficult to implement.

One of the most notable predictive biomarkers in ICI has been tumor MMRD status. MSI-

high/MMRD tumors are highly sensitive to ICIs regardless of tissue of origin (Le et al. 2015, Le

et al. 2017). Given the mutator phenotype conferred by MMRD, these findings are, perhaps, not

surprising. MMRD tumors develop many neoantigens due to their hypermutant phenotype

(Germano et al. 2017). Indeed, we have demonstrated durable benefit from ICIs can be achieved

in CMMRD patients with recurrent HGGs (Bouffet et al. 2016). These findings are promising

given the challenges associated with care standards in the context of CMMRD. The success of

ICIs in MMRD cancers has led to the approval of PD-1 inhibitors for the treatment of

MMRD/MSI-high tumors, the first-ever tissue-agnostic approval based on biomarker analyses

(Prasad et al. 2018). Notably, MMRD tumors appear to respond to ICIs despite being highly

heterogenous (Alexandrov et al. 2013), suggesting that the shear volume of mutations ensures

subclones will possess immunogenic mutations. Another recently uncovered mechanism through

which MMRD tumors may instigate an immune response is via the cyclic GMP-AMP synthase

24

(cGAS)-stimulator of interferon genes (STING) pathway. Two studies published in tandem

demonstrated that MLH1 deficiency leads to hyperactivity of Exo1, causing excessive genomic

instability and the escape of nuclear DNA into the cytoplasm (Guan et al. 2021, Lu et al. 2021).

This increase in cytosolic DNA triggers cGAS-STING signaling, increased interferon secretion,

activation of antigen-presenting dendritic cells, which may then prime T-cells to recognize and kill

cancer cells. Importantly, the authors showed that inactivation of STING or EXO1 in MLH1

deficient cells could impair the priming of T cells in vitro and infiltration of T-cells into tumors in

vivo. While these findings provide an additional cancer-cell intrinsic mechanism for promoting

immune recognition, it is not clear whether these findings are unique in the context of MLH1

deficiency or whether they would hold true in cancers driven by ablation of other MMR genes.

In addition to these genomic correlates, aspects of the immune microenvironment have also

demonstrated prognostic potential. PD-L1 expression on tumor cells is expected for therapies

targeting the PD-1 axis to have an effect. Although PD-L1 expression has been shown to positively

correlate with response to PD-1 blockade (Topalian et al. 2012, Garon et al. 2015), other studies

show no correlation (Motzer et al. 2015). These conflicting data may be due to discrepancies in

criteria for assessing PD-L1 expression, use of different detection assays, and even the cell-type

bearing PD-L1 expression (immune vs tumor cells). In addition to PD-L1 expression, the density

of tumor-infiltrating lymphocytes (TILs) within a tumor is another potential predictor of ICI

response (Fridman et al. 2012). Notably, MSI-high CRCs demonstrate markedly increased levels

of TILs compared to MSS CRCs and “immunoscores”, associated with high numbers of TILs,

were better predictors of patient survival than MSI (Mlecnik et al. 2016). Further work is needed

to assess TIL infiltration into other MMRD and POLE driven hypermutant cancers, particularly

those considered to be “immunologically cold”, such as gliomas. Moreover, while “immune-

inflamed” tumors seem to respond better to ICIs, emerging data also suggests that specific T-cell

phenotypic markers (Sade-Feldman et al. 2018, Thommen et al. 2018) and TIL clonality (Riaz et

al. 2017) are strongly associated with response to ICI. However, studies examining the phenotype

and clonality of TILs have not arrived at a consensus regarding whether such markers are important

prior, during, or after therapy, suggesting more work is needed.

It is clear that no one biomarker can currently predict response to ICI. Moreover, data surrounding

the efficacy of ICIs in PPD tumors is lacking. One retrospective study demonstrated that in the

25

context of CRCs, PPD tumors showed favorable prognosis and high TIL levels (Domingo et al.

2016), but as discussed above, the spectrum of PPD and MMRD tumors is broad. In the context

of hypermutant cancers, more work is needed to explore the influence of hypermutation on

immune surveillance and vice versa.

1.5 Replication Repair Deficient Cancer Modeling

Preclinical models of human cancer have been instrumental in the translatability of oncology-

related research from “bench to bedside”. Since the establishment and use of preclinical models of

RRD is the focus of this thesis, a review of existing models is appropriate. Broadly, with respect

to RRD, models relevant to this thesis include established human and mouse cancer cell lines,

patient-derived cancer cell lines and xenografts, and genetically engineered mouse models

(GEMMs).

1.5.1 Established cell lines, patient derived cell lines, & xenografts

Together, established and patient-derived cell line models are, by far, the most common models

used in basic and translational oncology (Gengenbacher et al. 2017). These encompass mouse and

human-derived established cell lines as well as patient-derived lines. These cell lines provide

opportunity for both in vitro and in vivo (subcutaneous and orthotopic) experimentation. Although

not autochthonous, cell line models are low-cost, efficient tools for oncology research, particularly

for elucidating relevant cancer genes and high-throughput screening of drug candidates.

In the context of MMRD, many established human cell lines are CRC-derived such as the HCT116,

HCT15, and LoVo cell lines which have been used extensively in the study of MMR (Aebi et al.

1996, Flohr et al. 1999, Cejka et al. 2004, Meyers et al. 2005). Alternatively, the genetic

manipulation of MMR proficient mouse and human cancer cell lines to generate MMRD isogenic

lines is also common (Russo et al. 2019, McGrail et al. 2020); in the context of mouse-derived

lines, these models allow for the study of MMR in syngeneic, immunocompetent mouse strains,

which is important to examine tumor-immune interactions (Germano et al. 2017, Mandal et al.

2019, Guan et al. 2021, Lu et al. 2021). Despite the ease of use with which these established cell

lines afford, they are often critiqued as oversimplified models of cancer due to differences in tumor

microenvironment and lack of genetic heterogeneity that results from long term in vitro culture.

26

Patient derived xenografts (PDX), which are typically transplanted directly into

immunocompromised mice, may overcome the limitations of established cell lines since they

should more accurately recapitulate the attributes of the original tumor. While the use of MMRD

PDXs has been limited (Nguyen et al. 2014), several studies have utilized PDX models to study

MMR as a mechanism of therapeutic resistance (Nagel et al. 2017, Touat et al. 2020). Although

PDXs have been shown to reasonably predict patient response to therapy (Hidalgo et al. 2014),

their high cost and variability in growth kinetics make their use limited. Moreover, outside of the

use of humanized mouse models, the immunocompromised animals used for PDXs do not allow

for the study of immune based therapies.

1.5.2 Mismatch Repair Deficient Mouse Models

GEMMs, unlike established cell line and patient-derived models discussed above, reproduce the

underlying genetic events of specific cancers, allowing for de novo tumor formation in an

autochthonous in vivo setting. Due to a lack of GEMMs of RRD tumors, it is difficult to analyse

tumor progression, identify driver mutations, and test potential therapeutics for these hypermutant

cancers.

Since the discovery of the human MMR system and its role in cancer development, mouse models

of MMRD have been established and studied. Msh2 (de Wind et al. 1995, Reitmair et al. 1995),

Msh6 (Edelmann et al. 1997), Mlh1 (Edelmann et al. 1999), and Pms2 (Prolla et al. 1998)

constitutional germline knockout mice were generated primarily to study LS-associated cancers.

The severity of tumor onset and survival in these animals correlates with our understanding of the

MMR system and the contributions of each MMR component to LS. Msh2 and Mlh1 knockout

mice develop tumors and succumb to them faster than Msh6 and Pms2 knockout mice, supporting

the critical role of these genes in tumorigenesis. Although mutation spectra differ slightly between

MMRD mice, most homozygous animals die from lymphoma and thymomas, with few

gastrointestinal (GI) cancers. Although the hematological malignancy phenotype was largely

considered a weakness of these models in the study of LS, such cancers in CMMRD are common,

thus more accurately modeling the constitutional syndrome rather than the heterozygous

syndrome. Indeed, the penetrance of cancers in the heterozygous phenotype was minimal across

knockout animals. A summary of survival and tumor findings from these animals is provided in

Table 1.1.

27

To circumvent the germline phenotype of hematological malignancies, Msh2LoxP (Kucherlapati et

al. 2010) and Mlh1LoxP (Reiss et al. 2010) conditional knockout mice were designed. The

phenotypes in these animals were dependent on Cre-driver strains used in the study but have been

used to model intestinal cancers in isolation, thus providing a more accurate model of LS. Findings

from studies using MMR condition knockout mice are also summarized in Table 1.1.

Table 1-1 Summary of MMR Deficient Mouse Models

Genotype Cre-

Transgene

Target Tissue Phenotype Average

Age

(months)

Other notable

features

Ref

Msh2-/- n/a All tissues T-cell lymphomas

B-cell lymphomas

Gastrointestinal tumors

6 MSI (de Wind

et al.

1995,

Reitmair

et al.

1995)

Msh6-/- n/a All tissues T-cell lymphomas

B-cell lymphomas

Gastrointestinal tumors

10 Low MSI (Edelman

n et al.

1997)

Mlh1-/- n/a All tissues T-cell lymphomas

B-cell lymphomas

Gastrointestinal tumors

7 MSI, males and

females infertile

(Prolla et

al. 1998,

Edelmann

et al.

1999)

Pms2-/- n/a All tissues T-cell lymphomas 10 MSI, males

infertile

(Prolla et

al. 1998)

Msh2LoxP Ella-Cre All tissues T-cell lymphomas

Small intestine tumors

6 (Kucherla

pati et al.

2010)

Msh2LoxP Villin-Cre Small and Large

intestine

Small intestine tumors 12 MSI (Kucherla

pati et al.

2010)

Msh2LoxP

APCLoxP

Adenoviral-

Cre

Adenovirus

expressing Cre

transgene

Large intestine tumors (in

combination with

APCLoxP allele)

U/N U/N (Kucherla

pati et al.

2013)

Msh2LoxP Lgr5-

CreERT2

Lgr5 expressing

stem cells

Small intestine tumors 19

3*

*Temozolomide

treated

(Wojciech

owicz et

al. 2014)

Msh2LoxP D9-Cre Medium-spiny

GABA-ergic

projection

neurons (MSNs)

Reduced HTT-CAG

expansions and modified

huntingtin phenotypes

n/a* *No reduced

viability

observed

(Kovalenk

o et al.

2012)

Mlh1LoxP Ella-Cre All tissues T-cell lymphomas

Gastrointestinal tumors

7 MSI, infertile (Reiss et

al. 2010)

Mlh1LoxP Lck-Cre Lck expressing

early stage

thymocytes

T-cell lymphomas 9 MSI (Reiss et

al. 2010)

28

1.5.3 Polymerase Proofreading Deficient Mouse Models

Fewer PPD than MMRD GEMMs have been generated, likely due to the lack of human data

surrounding PPD as an underlying mechanism for cancer initiation until the last decade. As a

result, the first models of Pol ε and Pol δ exonuclease deficient mice were generated through

mutation of the active sites discovered from experiments in yeast (Polee/e mice modeling mutations

at D275 and E277 in POLE; Pold1e/e mice modeling mutations D316 and E318 in POLD1)

(Goldsby et al. 2001, Goldsby et al. 2002, Albertson et al. 2009). Pol ε mutant mice lived longer

(~16 months) and developed a wider spectrum of tumors than Pol δ mutant mice (~7 months,

similar to MMRD models). Interestingly, unlike in humans where PPD is consistently detected as

heterozygous events, heterozygous PPD mice did not develop cancers at a rate different than

wildtype littermates. Survival and tumor spectrum findings from studies on such mice are shown

in Figure 1.5-1.

During the preparation of this thesis, an additional Pol ε exonuclease deficient cancer model was

reported (Figure 1.5-1) (Li et al. 2018). This inducible model utilized a lox-stop-lox (LSL) system

to conditionally knock-in one of the most prevalent POLE mutations detected in human cancers:

P286R. The authors used this model to generate germline mutant Pole+/P286R mice. The

heterozygous P286R mice demonstrated much more rapid time to tumor development and death

than even the homozygous Polee/e animals, suggesting substantial variability in the mutagenesis of

different polymerase mutations. Homozygous mutant mice were rarely observed (n=2 vs. expected

41), essentially exhibiting embryonic lethality. Heterozygous mice developed multiple tumor types

including T-cell lymphomas, small cell lung cancer, adenocarcinomas of the colon and

endometrium, and angiosarcomas. Importantly, whole genome sequencing (WGS) of tumors from

PoleP286R mice exhibited expected hypermutation (> 10 mut/Mb in all cancers), although not

completely recapitulating the human POLE signature 10. The reasons for this are unclear,

however, the fact that this model is a conditional knock-in makes it an attractive tool for developing

tissue-specific models of PPD cancers.

Finally, mouse models of pediatric combined RRD cancers have yet to be established. While

Albertson et al had combined PPD and Mlh1-/- mice, embryonic lethality was observed due to the

germline status of both alleles (Albertson et al. 2009). A recent combinatorial RRD mouse model

using the conditional knock-in LSL-PoleP286R model and conditional knock-out Msh2LoxP model

29

described above, was generated to study RRD endometrial cancers (Li et al. 2020). This study

demonstrated a synergy between PPD and MMRD, resulting in faster tumor development and

median time to death than PPD alone. Moreover, the authors used the model to demonstrate the

efficacy of ICIs for the treatment of RRD endometrial cancers.

Figure 1.5-1 Survival findings for two previously reported polymerase proofreading mutant mouse models

(A) Survival of Polee/e and Pole+/e mutant mice on a C57BL/6J background are indicated in red (light red indicates

same genotype but mixed C57BL/6J:129/Sv genetic background). Polee = exo allele = dual amino acid substitutions

at Pole exonuclease active sites: D272A and E274A. Reused with permission from PNAS (Albertson et al. 2009). (B)

Survival of PoleP286R/+, 2 PoleP286R/LSL, and wildtype siblings. The LSL-PoleP286R mice are ‘lox-stop-lox’ inducible and

bred to exhibit germline P286R mutations. P286R is one of the most prevalent POLE exonuclease alterations found

in human cancers. Reused with permission from the American Society for Clinical Investigation (Li et al. 2018). Note

the striking survival differences between the two models.

A B

30

1.6 Thesis Rationale, Hypothesis, & Aims

Over the last decade, advances in DNA sequencing technologies, coupled with the promise of

immunotherapies such as ICIs have provided, for the first time, opportunities to treat the myriad

of RRD cancers (discussed in Sections 1.3 and 1.4). While advances in our ability to diagnose

patients and cancers driven by RRD have increased greatly, its clear that using only established

tumor biopsies limits our ability to study events contributing to initiation, progression, and

response to therapy in these cancers.

In addition to established human cancer cell lines and PDX models that provide opportunities to

assess therapeutic efficacy and track tumor development and evolution, immune competent

transgenic models are necessary for the study of immune based therapies, which have

demonstrated great promise in RRD cancers.

1.6.1 Hypothesis

I hypothesize that combining clinical and human multi-omics data with PDX and immune

competent transgenic animal models will advance our knowledge of the biology of RRD cancers

as well as the identification and preclinical testing of therapies.

1.6.2 Specific Aims

1. Utilize human RRD informatics data and patient-derived tumors to study the dependency

of RRD tumors on RAS/MAPK pathway activation.

2. Build and characterize novel mouse models of DNA Polymerase Ɛ exonuclease deficiency.

3. Model and study complete, tissue-specific, replication repair deficiency by combining Pole

mutant mice with a conditional knockout MMR mouse model.

4. Study the relationship between immune surveillance and tumor mutational burdens in a

combined replication repair deficient mouse model.

31

Chapter 2

Mutations in the RAS/MAPK pathway drive replication repair deficient hypermutated tumors and confer sensitivity to MEK

inhibition

Chapter 2 was modified and reproduced with permission from:

Campbell BB*, Galati MA*, Stone SC, Riemenschneider AN, Edwards M, Sudhaman S,

Siddaway R, Komosa M, Nunes NM, Nobre L, Morrissy AS, Zatzman M, Zapotocky M,

Joksimovic L, Kalimuthu SN, Samuel D, Mason G, Bouffet E, Morgenstern DA, Aronson M,

Durno C, Malkin D, Maris JM, Taylor MD, Shlien A, Pugh TJ, Ohashi PS, Hawkins CE, Tabori

U. “Mutations in the RAS/MAPK pathway drive replication repair deficient hypermutated tumors

and confer sensitivity to MEK inhibition.” Cancer Discov. 2021 Jun;11(6):1454-1467. DOI:

10.1158/2159-8290.CD-20-1050.

* These authors contributed equally. See ‘Statement of Contributions’ for detailed author

contributions.

32

Mutations in the RAS/MAPK pathway drive replication repair deficient hypermutated tumors and confer sensitivity to MEK inhibition

2.1 Abstract

The RAS/MAPK pathway is an emerging targeted pathway across a spectrum of both adult and

pediatric cancers. Typically, this is associated with a single, well-characterized point mutation in

an oncogene. Hypermutant tumors which harbor many somatic mutations may obscure the

interpretation of such targetable genomic events. We find that replication repair deficient (RRD)

cancers which are universally hypermutant and affect children born with RRD cancer

predisposition, are enriched for RAS/MAPK mutations (p=10-8). These mutations are not random,

exist in subclones, and increase in allelic frequency over time. The RAS/MAPK pathway is

activated both transcriptionally and at the protein level in patient derived RRD tumors and these

tumors responded to MEK inhibition in vitro and in vivo. Treatment of patients with RAS/MAPK

hypermutant gliomas reveal durable responses to MEK inhibition. Our observations suggest that

hypermutant tumors may be addicted to oncogenic pathways resulting in favorable response to

targeted therapies.

2.2 Introduction

Tumor hypermutation is defined by an excess of somatic mutations (>10 mut/Mb). Based on this

definition, nearly one fifth of human cancers are hypermutated, and nearly all tissue types have a

subset of tumors which are hypermutated (Pritchard et al. 2014, Gargiulo et al. 2016, Campbell et

al. 2017). Multiple reports have concluded that hypermutant tumors are unique in their evolution,

resistant to conventional chemotherapies, and confer poor patient survival (Hunter et al. 2006,

Touat et al. 2020).

Hypermutation arises as the result of various underlying processes (Alexandrov et al. 2013). These

include environmental exposures (UV radiation (Pfeifer et al. 2005, Cancer Genome Atlas 2015),

smoking (Pleasance et al. 2010), aristolochic acid exposure (Poon et al. 2013)), chemotherapy

(alkylators) (van Thuijl et al. 2015) and replication repair deficiency (RRD), among others.

33

Mutations arising from replication repair-related errors have been causally implicated in two-thirds

of human cancer (Loeb et al. 1974, Tomasetti et al. 2017). RRD and associated hypermutation can

result from germline or somatic mutations in either DNA polymerases (POLE and POLD1) (Palles

et al. 2013) or in the mismatch repair genes (MSH2, MSH6, MLH1 and PMS2) (Bonadona et al.

2011). Furthermore, germline biallelic loss of any of the mismatch repair genes results in one of

the most aggressive human cancer syndromes termed Constitutional Mismatch Repair Deficiency

Syndrome (CMMRD). These patients develop cancers in numerous organs during childhood and

harbor the highest mutation burden (>300 mutations/MB) of any human cancer type (Shlien et al.

2015). Due to the tumor burden and their inherent resistance to conventional therapies, these

patients tend to have extremely poor outcomes.

Targeting hypermutated tumors with immune checkpoint inhibitors results in clinical

benefit (Errico 2015, Le et al. 2015, Rizvi et al. 2015, Bouffet et al. 2016, Johanns et al. 2016,

Santin et al. 2016). Recent data has demonstrated that in some cancer types, high mutational burden

is a strong predictor of response to immune checkpoint inhibition (Vandeven et al. 2018). The PD-

1 inhibitor, Pembrolizumab is now approved for therapy in all RRD cancers reinforcing the

agnostic, mechanism-based indication for a cancer therapeutic (Overman et al. 2017). However,

immune checkpoint inhibition alone has not been universally effective (Overman et al. 2018, Le

et al. 2020, Marabelle et al. 2020). Defining other vulnerabilities of hypermutant cancers is a major

aim of the cancer research community.

Single nucleotide alterations in proto-oncogenes that regulate cell growth and proliferation

pathways are a commonly observed tumor-promoting event (Montagut et al. 2009). One such

pathway is the well-characterized RAS/MAPK pathway, where gain-of-function mutations in RAS

and/or downstream effectors results in constitutive dysregulation of signalling to upregulate

transcription factors involved in diverse cellular processes conferring survival advantage. While

canonical amino acid changes, such as a valine to glutamic acid at position 600 in BRAF (V600E),

or a glycine to aspartic acid substitution at position 12 in KRAS (G12D), are commonly observed

in multiple cancers (Sebolt-Leopold et al. 2004), the spectrum of potentially actionable

RAS/MAPK activating mutations in hypermutated cancers and whether these mutations are

actively selected in the context of RRD-derived hypermutation is largely unknown.

34

Importantly, recent reports have noted that nonsynonymous mutations in RAS/MAPK pathway

genes (i.e. HRAS, NRAS, BRAF) exhibit strong selection, regardless of their amino acid position

(Martincorena et al. 2017). In addition, mutations in RAS/MAPK signalling members downstream

of the typical receptor tyrosine kinases, such as MAPK1, MAP2K1, MAP2K2, MAP2K4, MAP3K13

have been found to be significantly more frequently mutated than expected by chance by genomic

analyses of large tumor datasets of diverse types (Chang et al. 2018). Thus, the diversity of

RAS/MAPK pathway driver mutations, and the genes in which they arise, may be larger than

previously anticipated.

Since RAS/MAPK pathway genes are commonly mutated in several childhood cancers, including

cancers types commonly associated with CMMRD, we reasoned that hypermutated RRD tumors

will harbor functionally impactful mutations in key genes of the RAS/MAPK signalling pathway

and therefore would be amenable to RAS/MAPK pathway inhibition.

Here we show that RAS/MAPK mutations are enriched in RRD hypermutated tumors and result

in activation of the pathway. Treatment of patients with RRD cancers reveal encouraging

responses in heavily pretreated children with otherwise poor prognoses.

2.3 Materials & Methods

2.3.1 Patient and sample collection for exome sequencing and PBMC collection

A cohort of germline replication-repair deficient patients with known clinical history was collected

as described previously (Shlien et al. 2015). In brief, patients were registered as a part of the

International Replication Repair Deficiency Consortium, which includes multiple centers

worldwide. The study was conducted in accordance with the Canadian Tri-Council Policy

Statement II (TCPS II). Following Institutional Research Ethics Board approval, all data were

centralized in the Division of Haematology/Oncology at The Hospital for Sick Children

(SickKids). Written, informed consent was obtained from patients’ parents or guardians, or from

the patients, where applicable. Family history, demographic and clinical data were obtained from

the responsible physician and/or genetic counselor at the corresponding centers.

35

2.3.2 Targeted panel sequencing

FoundationOne panel sequencing was performed on 1215 pediatric tumors and variants were

shared via the pediatric portal https://pediatric-data.foundationmedicine.com/.

Targeted sequencing was performed as previously described (Campbell et al. 2017). In brief,

exonic hybridization capture of >400 genes implicated in cancer was applied to a minimum of 50

ng of DNA extracted from formalin-fixed paraffin-embedded clinical cancer specimens.

Pathologic diagnosis of each case was confirmed by review of hematoxylin and eosin (H&E)

stained slides and samples were excluded if found to contain <20% tumor cells. Libraries were

sequenced to high uniform median coverage (>500x) and assessed for base substitutions, copy

number alterations, and gene fusions/rearrangements.

2.3.3 Whole exome sequencing

High-throughput sequencing, read mapping, and identification of mutations was performed at The

Center for Applied Genomics at the Hospital for Sick Children. Tumor and matched blood derived

DNA were run using Agilent’s exome enrichment kit (Sure Select V4/V5; with >50% of baits

above 25x coverage), on an Illumina HiSeq2500. Base calls and intensities from the Illumina

HiSeq 2500 were processed into FASTQ files using CASAVA and/or HAS. The paired-end

FASTQ files were aligned to UCSC’s hg19 GRCh37 with BWA. Aligned reads were realigned for

known insertion/deletion events using SRMA and/or GATK. Base quality scores were recalibrated

using the Genome Analysis Toolkit26 (v1.1-28). Somatic substitutions were identified using

MuTect (v1.1.4). Mutations were then filtered against common single-nucleotide polymorphisms

(SNPs) found in dbSNP (v132), the 1000 Genomes Project (Feb 2012), a 69-sample Complete

Genomics data set, and the Exome Sequencing Project (v6500) and the ExAc database.

2.3.4 Subclone analysis

Tumor subclones were determined using the R package SciClone (Miller et al. 2014). Specific

RAS/MAPK variants specific to each clone were identified following distribution of the variants

to their respective subclones.

36

2.3.5 Variant Impact Prediction Software

Variant Impact prediction was performed using PolyPhen-2 (Adzhubei et al. 2013)

http://genetics.bwh.harvard.edu/pph2/.

2.3.6 RNA-Sequencing

21 hypermutant pediatric GBMs collected by the RRD consortium underwent stranded, poly-A

capture RNA sequencing following library preparation using the NEBNext® Poly(A) mRNA

Magnetic Isolation kit. Raw RNA-seq reads were mapped to the human genome (build hg38) using

the STAR aligner (v2.4.2a) in two-pass mode and gene annotations from GENCODE v25.

Duplicates were marked and removed with Picard Tools prior to quantification of gene counts.

Gene counts were calculated using the python script htseq-count (v0.6.1). Raw counts were

normalized with the R package DEseq and pathway activation determined with the R package

PROGENy. Differential expression was analysed with edgeR and pre-ranked gene set enrichment

analysis performed on the Hallmarks gene sets after ranking genes according to rankScore

= sign(log2FC) x -log10(edgeR-adjusted p-value).

2.3.7 Nanostring nCounter gene expression system

Total RNA was isolated from 5- to 10-mm scrolls of Formalin Fixed Paraffin Embedded preserved

tissue with the RNeasy FFPE extraction kit (QIAGEN, Valencia, CA). RNA quantity and quality

were assessed using the NanoDrop 2000 (Thermo Scientific, Waltham, MA). Samples displaying

nanodrop values of 2.0 to 2.1 were utilized.

Five hundred nanograms of total RNA input was used on the NanoString nCounter system

(nanoString Technologies, Seattle, WA).

2.3.8 In vivo serial xenograft experiment

All mouse studies were approved and performed in accordance with the policies and regulations

of the Institutional Animal Care Committee of the Hospital for Sick Children, in Toronto. Athymic

NOD scid gamma mice (The Jackson Laboratory, Bar Harbor, ME) underwent subcutaneous

implantation of tumor of a 1x1 cm piece of a patient derived RRD hypermutant colorectal tumor

and propagated for 90 days. Half of the tumor was then harvested for DNA extraction and exome

37

sequencing, while the second half was implanted into a new mouse. The procedure was repeated

3-4 times.

2.3.9 In vivo treatment of patient-derived xenografts

Primary human colorectal and brain pathological and post-surgical core biopsies were obtained

following Institutional Research Ethics Board approval and written informed consent obtained

through the IRRDC. 1x106 isolated glioblastoma multiforme (GBM) cells were implanted

subcutaneously into three groups of ten mice each (vehicle and Selumetinib and/or Trametinib).

Engrafted mice were administered Trametinib (1 mg/kg)(Lin et al. 2015, Yao et al. 2017, Fedele

et al. 2018) or Selumetinib (100 mg/kg)(Bartholomeusz et al. 2012, Troiani et al. 2012, Lin et al.

2019) via oral gavage daily until endpoint.

2.3.10 Western blotting

Tissue samples were flash frozen with liquid nitrogen and then mechanically pulverized to fine

powder using a mortar and pestle. Samples were lysed using hot 2x SDS buffer with sodium

orthovanidate and phosphoinhibitor cocktails II (Millipore Corp Cat #524625-1SET) and IV

(Calbiochem Cat #524628-1SET). Protein concentrations were quantified using ThermoFisher’s

Pierce BCA protein assay kit (catalog #23227). Equal loading (20ug protein) was ensured for each

blot. Fresh bromophenol blue dye and dithiothreitol (DTT; final concentration 100mM) were

added to lysate aliquots and boiled immediately before resolution. All samples were resolved using

SDS-PAGE, 12% bis-acrylamide gels, and transferred using BioRad’s semi-dry transfer system to

a PVDF membrane. α-tubulin was used as a loading control. Membranes were probed with primary

antibodies specific to protein of interest: MEK1/2, phospho-MEK1/2, ERK1/2, phospho-ERK1/2,

caspase-3, cleaved caspase-3, PARP; α-tubulin, T5168, Sigma. Lysis buffer composition: 2mL

100% glycerol + 2mL (10% SDS) + 5.4mL H2O + 400uL 0.5M EDTA + 200uL 1.0M TRIS +

50uL protease and phosphoinhibitors.

2.3.11 Immunohistochemistry for RAS/MAPK pathway upregulation

Nuclear pERK staining of 11 RRD Glioma and 2 CRC was performed using Phospho-p44/42

MAPK (Erk1/2) (Thr202/Tyr204) Antibody (Cell Signalling Cat #9101). All slides were subject

to pathological review. Images were reviewed and captured using the open source software for

digital pathology analysis: QuPath.

38

2.3.12 Cell Lines

Colorectal cancer cell lines HCT116 and HCT15 were purchased from ATCC® and certified

Mycoplasma free (August 2013). Passage 3 from both HCT116 and HCT15 were used for in vitro

and sequencing experiments. LoVo was obtained from C. Pearson, The Hospital for Sick Children,

Toronto, Canada (April 2016) and passage 4 was used for in vitro and sequencing experiments.

Both LoVo and HCT116 were confirmed Mycoplasma free using the PCR Mycoplasma Test Kit;

Catalogue #409010 (mdbiosciences, March 2017). The patient derived high grade glioma cell line

(Fig. 5) was established in December 2016, passaged 21 times before used in in vitro experiments,

and certified Mycoplasma free using the Mycoplasma PCR Detection Kit Catalogue #G238

(abm®; August 2020). Treatments were performed with MEK inhibitors Trametinib (SelleckChem

GSK1120212) or Selumetinib (SelleckChem AZD6244).

2.3.13 Flow Cytometry

Viable frozen PBMCs were incubated with Fc block (BD Biosciences) prior to staining for surface

markers (anti-CD3 - clone UCHT1, anti-CD4 – clone RPA-T4, anti-CD8 – clone RPA-T8, anti-

CD39 – clone A1, anti-Ki67 – clone 20Raj1) and viability dye (eBioscience). Cells were fixed and

permeabilized for intercellular staining with the Foxp3 transcription factor staining buffer set

(BD). Flow cytometry voltages were set using Rainbow beads (Spherotech) with the same setting

between experiments. Samples were acquired on a BD LSR Fortessa flow cytometer and data were

analyzed using the FlowJo software.

2.3.14 Statistical Analysis

Statistical analyses were performed using the GraphPad Prismv8 software. The log-rank (Mantel-

Cox) test was used to analyze the survival difference for in vivo mouse experiments. P values <0.05

were determined to be significant. Asterisks denote P values as follows: *, P < 0.05; **, P < 0.01;

***, P < 0.001, **** P<0.0001.

2.3.15 Data Availability

Existing whole exome sequencing data have previously been deposited in the European Genome

Phenome Archive under study accession numbers EGAS00001000579 and EGAS00001001112.

New whole exome sequencing and RNA sequencing data generated from this study have been

39

deposited in the European Genome Phenome Archive under study accession number

EGAS00001005008. Public datasets used include cBioPortal (https://www.cbioportal.org/).

Further information and/or requests for data will be fulfilled by the corresponding author, Uri

Tabori ([email protected]).

2.4 Results

2.4.1 Mutations in the RAS/MAPK pathway are common in hypermutated childhood cancers

First, to determine prevalence of RAS/MAPK mutations in pediatric cancers, we analysed all

single nucleotide mutations resulting in nonsynonymous amino acid changes in a large cohort of

1215 pediatric cancer patients (ages 0-18; 55% male, 45% female) using targeted sequencing of

>400 well defined cancer genes, sequenced and made publicly available by Foundation Medicine

(Frampton et al. 2013). Major tumor subtypes within this cohort consisted of brain tumors,

haematological malignancies, extracranial embryonal tumors, sarcomas, and carcinomas (Figure

2.4-1 a; 2.4-2). Further information on these tumors and their specific genetic alterations, can be

found in Figure 2.4-2 and Supplemental Table S1 – see Appendices. Nonsynonymous mutations

in NRAS, NF-1 and KRAS were the second, third and fourth most common alterations observed

following TP53 across cancers of all types (Figure 2.4-1 a). The negative RAS/MAPK pathway

regulator NF-1 showed predominantly truncating point mutations, while point mutations in

oncogenes NRAS, KRAS, BRAF and PTNP11 oncogenes were almost exclusively missense

(Figure 2.4-3). Of the 15 most common genes harboring SNVs in these childhood cancers, 5 were

RAS/MAPK pathway members, affecting 18% (218/1215) of all childhood tumors. These

mutations were observed across most childhood cancers suggesting that the potential for

RAS/MAPK pathway inhibition in pediatric cancer as a therapeutic option may be under-explored.

We then considered hypermutated tumors. These cancers are commonly devoid of copy number

alterations and are characterised by a preponderance of single nucleotide variants (Shlien et al.

2015). We observed different RAS/MAPK pathway mutation enrichment among tumor types

suggesting that these mutations are not simply a by-product of hypermutation. First, the prevalence

of RAS/MAPK mutations were most commonly observed in gastrointestinal tract cancers and

40

gliomas followed by hematopoietic malignancies (Fisher’s Exact test, p < 0.0001). These cancers

are associated with the cancer predisposition syndrome CMMRD (Durno et al. 2015) and therefore

driven by RRD. Second, RAS/MAPK mutations were enriched in tumor types known to harbor

these alterations regardless of their tumor mutational burden. For example, 60% of hypermutant

neuroblastoma harbored mutations in RAS/MAPK genes, consistent with recent reports (Eleveld et

al. 2015) while hypermutated Wilms tumors and rhabdomyosarcomas rarely harbored mutations

in the pathway (Figure 2.4-1 b). Third, specific genes in the pathway dominated each respective

tumor type. KRAS mutations were dominant in gastrointestinal tumors, while NF1 mutations were

more dominant in brain tumors (Figure 2.4-1 c). This tissue-of-origin selectivity was further

supported by comparing CNS vs. colorectal cancers for NF1 mutations versus KRAS dominance

in cBioPortal (Figure 2.4-4). These data support the notion that alterations in the RAS/MAPK

pathway in hypermutated cancers are beyond what would be expected by chance and may be

confer a selective advantage to hypermutated cancer cells.

41

Figure 2.4-1 Prevalence of RAS/MAPK genetic events across 1215 pediatric cancers

(A) Pie graph indicates the tissue of origin of 1215 pediatric cancers. 1803 Single nucleotide variants were detected

across the entire cohort, RAS/MAPK activating events are indicated in yellow. (B) Prevalence of RAS/MAPK

mutations in hypermutant pediatric tumors, stratified by tumor type. (C) Pie graphs demonstrating the distribution of

the 11 most commonly mutated RAS/MAPK genes across different tissues of origin.

42

43

Figure 2.4-2 Major subtypes in1215 pediatric cancers (Ages 0-18)

Colored bar charts display age distribution per subtype and grey bar charts display frequency of RAS/MAPK pathway

mutations.

44

Figure 2.4-3 Genes with highest frequency of nonsynonymous point mutations across 1215 tumors

Bar plot displaying frequency of protein coding mutations resulting in a single amino acid change or truncation in

1215 pediatric tumors. Genes with less than 3 occurrences in the cohort were excluded. * = RAS/MAPK pathway

actors.

45

Figure 2.4-4 Confirmatory tissue specific preference for MAPK pathway mutations in cBioPortal Query

Tissue preferences for specific MAPK mutations were observed for 1825 colorectal cancers (A) and 1764 gliomas

(B). KRAS dominance is observed in GI vs. NF1 dominance in CNS tumors, mimicking pediatric data.

46

2.4.2 Replication repair deficient cancers activate the RAS/MAPK pathway

To further explore the role of RAS/MAPK mutations in RRD hypermutant gliomas and

gastrointestinal cancers, we performed high coverage exome sequencing (~98x) of 46 tumors from

the International RRD consortium (IRRDC), an international group of clinicians and medical

scientists studying and treating children and young adults with RRD cancers. Matched controls

were obtained and sequenced for all but two of 46 tumors (D291_2, and X12). All 46 cancers

harbored several point mutations in diverse members of the RAS/MAPK pathway (Figure 2.4-5;

Supplemental Table S2 – see Appendices). These included at least one loss of function NF1

mutation in 95% (39/41) of gliomas and often additional missense mutations in intermediary

RAS/MAPK pathway members (MAPK1, MAP2K1, MAP2K2, MAP3K1, MAP3K2).

Clonal heterogeneity is a major contributor to acquired resistance to current targeted therapies. We

anticipated that this may also be a concern in RRD cancers due to their ongoing mutation

accumulation and highly polyclonal nature. To determine whether RAS/MAPK mutations exist in

different clones within RRD cancers, we performed high resolution subclonal analysis of

hypermutated cancers that were sequenced to a mean depth of 750x. We confirmed that every

clone in polyclonal tumors contained a nonsynonymous mutation in a RAS/MAPK pathway gene

(Figure 2.4-6 a). For additional support, we employed the variant impact prediction software

POLYPHEN-2 to confirm that the mutations that arose were predicted damaging in the case of

negative regulators, while positive regulator genes preferred benign/activating mutations. As

observed in general hypermutant childhood cancers, RRD gliomas demonstrated multiple

truncating mutations in NF1, while colorectal tumors preferred activating mutations in tyrosine

protein kinase receptors (Figure 2.4-6 a).

To determine whether RAS/MAPK mutations persist during clonal evolution of RRD cancers, we

established serial xenografts of a hypermutant (20 mutations/MB) colorectal cancer surgically

resected from a patient with a homozygous germline MLH1 mutation. The primary tumor

contained one activating mutation in MAP3K1, which persisted as a consistently expanding variant

allele fraction (VAF) throughout the serial engrafting (Figure 2.4-6 b). Two additional

RAS/MAPK pathway activating mutations in KRAS and MAP3K2 arose during serial

transplantations and increased from low to a high variant allele fraction, suggesting enrichment for

these mutations (Figure 2.4-6 b). This enrichment in RAS/MAPK mutations in progressive clones

47

was independent of the original variants in the tumor, which remained at a stable, clonal VAF

between 0.4 and 0.45 across serial xenografts further suggesting a functional role to these

mutations. We conclude that within the RRD cancers, RAS/MAPK mutations are non-randomly

enriched and exist in multiple clones during cancer progression.

Next, we explored whether these mutations result in activation of the RAS/MAPK pathway that

can be observed at the transcriptional and translational level. First, we sequenced the full

transcriptome of 21 RRD hypermutant glioblastomas that had undergone exome sequencing and

harbored confirmed RAS/MAPK pathway alterations (Figure 2.4-5; Supplemental Table S2 – see

Appendices) and compared them to five normal adult brain and four normal fetal brain samples.

After confirming that global transcriptional patterns between tumor and normal were distinct

(Figure 2.4-7), we used several pathway signature models to assess RAS/MAPK pathway

activation. We assessed MAPK pathway activity using the signature-based model PROGENy

(Schubert et al. 2018), which derives signatures using 100 responsive genes that are most

consistently deregulated as a result of MAPK activation. RRD tumors had significantly higher

MAPK signature scores than normal samples (Figure 2.4-8 a) and distinctly clustered from

normals based on the transcriptional output of the PROGENy MAPK pathway genes (Figure 2.4-

8 b). This was not true for housekeeping genes (Figure 2.4-9), suggesting that such clustering does

not occur randomly. Using the same RNA sequencing data, we could additionally distinctly cluster

samples using a gene expression RAS pathway signature comprised of 18 genes (Dry et al. 2010)

(Figure 2.4-10 a). Finally, we performed Gene Set Enrichment Analysis (GSEA) of the hallmark

gene set for genes upregulated by KRAS activation (Liberzon et al. 2015) and found that RRD

tumors are significantly enriched for KRAS upregulated genes compared to normal adult and fetal

brains (Figure 2.4-10 b; normalized enrichment score (NES) = 3.13).

To further directly measure pathway activation, we employed a multiplexed RNA and protein

detection strategy in which multiple RAS/MAPK pathway related mRNA and downstream

proteins and their phosphorylation are probed and counted directly without previous amplification

(NanoString nCounter gene expression system). RRD hypermutant gliomas had significantly

higher expression of genes in the pathway when compared to normal brain (p=0.0002) (Figure

2.4-11 a). This correlated well with other childhood gliomas with known driver mutations in NF1,

BRAF-V600E and BRAF-KIAA1549 fusion which are known to harbor RAS/MAPK pathway

48

activation and respond well to MEK inhibitors (Figure 2.4-11 a). Further examination of the

pathway activation using NanoString revealed that RRD hypermutant cancers preferentially

expressed high levels of downstream target transcription factors (FOS, JUN, MYC) and less of

membrane receptors, further supporting the role of mutations in mid-pathway tyrosine kinases

which result in activation of the effector genes and a possible loop repression of receptor

expression (Figure 2.4-11 b).

Activation of RAS/MAPK kinases was also observed at the protein level when comparing

phospho-ERK to total ERK and phospho-c-RAF in RRD hypermutant gliomas and normal brain

using NanoString. RRD gliomas harbored comparable levels of these phosphorylated targets to

gliomas driven by BRAF and NF1 alterations (Figure 2.4-12 a). As a further confirmation of the

assay, protein expression of Ki-67 and p53 was significantly higher in RRD high grade gliomas as

compared to low grade gliomas harboring the RAS/MAPK alterations (Figure 2.4-12 a), indicating

that the increase in phosphorylation observed was not random. Lastly, immunohistochemical

staining for phospho-ERK was performed on 11 hypermutant high grade glioma and 2 colorectal

cancers in children with germline mutations in replication repair genes (MMR, POLE) and 100%

of tumors displayed strong positive phospho-ERK staining (Figure 2.4-12 b; Supplementary

Figure S1 - see Appendices). These data further suggest that RRD hypermutant cancers are

driven, in part, by non-random mutations resulting in growth-factor receptor independent

activation of the RAS/MAPK pathway.

49

Figure 2.4-5 Prevalence of RAS/MAPK genetic events across 46 RRD cancers

Exome sequencing of 46 GBM and CRC from patients with germline mutations in MMR and POLE. Top panel

indicates tumor mutation burden in mutations/mb, middle panels indicate tumor type and number of nonsynonymous,

protein-coding events in 11 commonly mutated RAS/MAPK genes. GBM = glioblastoma multiforme; CRC =

colorectal cancer

50

Figure 2.4-6 Prevalence of RAS/MAPK mutations in polyclonal and temporal sampling of RRD tumors

(A) Clonal analysis of three heavily mutated tumors and the respective RAS/MAPK mutations in each clone. Polyphen

scores for each mutation distinguish damaging mutations in negative regulators, in contrast to benign or activating

mutations in gain-of-function genes (B) Schematic of serial xenografting experiment performed on one hypermutant

patient-derived colorectal cancer xenograft (left). Changes in variant allele fraction of RAS/MAPK genes mutated in

the primary and subsequent xenografts. All mutations increase in tumor fraction, suggesting positive selection, while

other variants in the primary remain stable (right).

51

Figure 2.4-7 Unsupervised transcriptomic sample clustering

Multidimensional scaling plot demonstrating clustering of CMMRD tumors, normal fetal brain, and normal adult brain

based on RNA sequencing count data. CMMRD tumors cluster distinctly from normal fetal and adult brain. Normal

fetal brain and adult brain also show distinct expression patterns.

52

53

Figure 2.4-8 Assessment of MAPK pathway activation in hypermutant RRD gliomas using PROGENy

(A) Violin plot of RNA sequencing-derived MAPK pathway PROGENy signature scores of CMMRD GBMs (n=21)

compared to normal fetal (n=4) and adult (n=5) brains (p < 0.0001; Welch t-test). Lines indicate median and quartile

values. (B) Unsupervised clustering of CMMRD tumors (n=22), normal fetal brains (n=4), and normal adult brains

(n=5) based on expression of PROGENy 100 MAPK pathway transcriptional output signature genes.

Figure 2.4-9 Transcriptome clustering based on housekeeping gene expression

Control clustering for all tumors based on expression of a set of randomly selected housekeeping genes demonstrates

that clustering distinctly does not occur non-specifically.

54

Figure 2.4-10 Transcriptomic assessment of RAS pathway activation in hypermutant RRD

(A) Unsupervised clustering of all samples in (Figure 2.4-8) based on expression of an 18-gene RAS transcriptional

output signature. (B) GSEA enrichment plot for genes upregulated by KRAS activation in CMMRD GBM vs. normal

brain samples in (A). Normalized enrichment score (NES) is reported (FDR = 0).

55

Figure 2.4-11 Transcriptional assessment of RAS/MAPK pathway activation in hypermutant RRD gliomas

using NanoString3D

(A) NanoString counts of single mRNA molecules in 20 RAS/MAPK pathway related probes, including EGFR, BRAF,

KRAS, MAP3K1, FOS and JUN, in CMMRD GBM compared to tissue-matched BRAF-V600E mutant low-grade

gliomas. (B) NanoString counts of single mRNA molecules in 20 RAS/MAPK pathway related probes, stratified by

protein-type. Transcription factor targets of the RAS/MAPK pathway were the most upregulated. For boxplots,

median and quartile values are depicted.

56

57

Figure 2.4-12 Proteomic assessment of RAS/MAPK pathway activation in hypermutant RRD gliomas

(A) NanoString counts of single protein molecules in involved in RAS/MAPK pathway activation, including phospho-

ERK and phospho-C-RAF. Ki-67 protein levels are shown to highlight the distinguishing growth characteristics of

LGG versus GBM. (B) Positive immunohistochemistry (IHC) staining for phospho-ERK (pERK) on two

representative CMMRD brain tumors harboring several RAS/MAPK pathway alterations (bottom left; MMR190 and

MMR134), compared to a CMMRD normal postmortem brain sample and non-RRD pediatric medulloblastoma

demonstrating negative staining (top), and a human renal cell carcinoma sample demonstrating nuclear and

cytoplasmic staining of phospho-ERK in tubular epithelial structures surrounded by negatively staining lymphocytes

(lower right). Scale bars indicate 50μm.

2.4.3 RRD hypermutant cancers respond to MEK inhibition

Several RAS/MAPK pathway inhibitors have been developed to block tumor growth driven by

this pathway. Those that target the MEK intracellular kinase are already indicated for BRAF-

V600E melanoma and are currently in various phases of trials for other tumor types including

glioma and colon cancer (Shannon et al. 2017, Yaeger et al. 2017). To test the preclinical benefit

of RAS/MAPK inhibitors in RRD hypermutant tumors, we modelled glioma and colon cancers as

these are the tumors most commonly observed in children with RRD. Established colorectal cancer

cell lines were first confirmed to be hypermutant and driven by MMR deficiency via signature

analysis (COSMIC signatures 6, 14, and 15; Figure 2.4-13). Dose response analysis revealed that

both established and patient-derived cells were sensitive to MEK inhibition (Figure 2.4-13). We

then tested the ability of MEK inhibitors to impede cancer growth in vivo using an RRD patient-

derived colorectal cancer xenograft with germline MLH1 mutations (Figure 2.4-14). First, deep

sequencing was performed to confirm that the tumor harbored clonal and subclonal RAS/MAPK

mutations (Figure 2.4-14 a). Mice injected with patient-derived colorectal cancer cells and treated

daily with Trametinib demonstrated significantly increased survival benefit (Figure 2.4-14 b).

Moreover, daily treatment with either MEK inhibitor (Selumetinib and Trametinib) independently

resulted in reduced tumor growth, and significant RAS/MAPK pathway-specific growth arrest as

exhibited by decreased phosphorylation of ERK and MEK and increased apoptosis (Figure 2.4-

14 b,c).

Similarly, use of MEK inhibitors for the treatment of a patient-derived pediatric high-grade glioma

with confirmed hypermutation (299 mut/Mb), multiple RAS/MAPK pathway mutations, and a clear

RRD signature (Figure 2.4-15 a-d) resulted in significant response to both Trametinib and

Selumetinib in vitro (Figure 2.4-15 e). Moreover, xenografts established from these primary

cultures demonstrated improved survival when treated with Trametinib (p=0.0001) and reduced

tumor growth in response to either inhibitor in vivo (p=0.0039, Figure 2.4-15 f, g).

58

Figure 2.4-13 Sensitivity of established MMRD cell lines to MEK inhibition

Three established hypermutant colorectal cancer MMR-deficient cell lines (LoVo, HCT116, and HCT15) harbor

several RAS/MAPK pathway promoting mutations and COSMIC signatures related to MMR and/or polymerase

exonuclease deficiency (left). Mutational signatures reflect MMR deficiency as a source of hypermutation. Cell lines

were treated with Trametinib concentrations ranging from 1nM - 100 nM and viable cells were measured using a

hemocytometer at 24, 48, and 72 hours post treatment (right). Cell number percentages were normalized to DMSO.

59

Three technical replicates were analysed and are shown as mean and SD. Statistical significance was assessed by

ANOVA (*p < 0.05; **p < 0.01; ***p < 0.001; and ****p < 0.0001).

Figure 2.4-14 Genomic characterization and subsequent xenografting and treatment of an ultrahypermutant

childhood colorectal cancer

(A) Schematic of tumor sequencing analysis performed on a patient derived hypermutant colorectal cancer. Targeted

panel sequencing revealed two major clones. A secondary KRAS G13D known RAS/MAPK driver pathway mutation

was acquired in the secondary clone. Mutation signatures within the tumor reveal a characteristic pattern of replication

repair deficiency. (B) Tumor growth experiments in response to MEK inhibition therapy (Trametinib) for flank-

implanted colorectal cancer xenograft. Tumor volume was measured biweekly in Trametinib treated vs. vehicle group.

Western blot analysis of tissues and the conclusion of the experiment reveal a decrease in phospho-MEK and phospho-

ERK. A second repeated experiment to demonstrate survival differences is shown. (C) Tumor growth experiments in

response to MEK inhibition therapy using Selumetinib. Mice were sacrificed after 24 days to display differences in

tumor sizes.

60

Figure 2.4-15 Genomic characterization, xenografting, and treatment of ultrahypermutant childhood GBM

(A) Schematic of experiments performed on a CMMRD pediatric Glioblastoma (B) Subclonal analysis reveal multiple

mutations per clone. (C) Identity of RAS/MAPK pathway-related alterations in this tumor. (D) Mutational signature

analysis reveals the source of mutations as replication repair deficiency. (E) Primary culturing of cells derived from

this tumor, without prior culturing, and treatment with either Trametinib or Selumetinib. (F) Survival experiment

following flank implantation of this ultrahypermutant GBM and subsequent treatment in vivo with Trametinib. (G) A

61

second set of flank engraftment experiments were conducted to assess tumor growth following flank implantation and

treatment with Trametinib or Selumetinib. Arrow indicates time of treatment commencement.

2.4.4 Encouraging responses in patients with RRD gliomas to MEK inhibition

Two patients from the IRRDC who had RAS/MAPK active gliomas were treated with MEK

inhibitors. Patient 1, a CMMRD patient harboring homozygous PMS2 germline mutations was

diagnosed at age 3 years with an inoperable diffuse glioma (Figure 2.4-16). The patient underwent

three distinct chemotherapy protocols (combination vincristine plus carboplatin, vinblastine, and

temozolomide respectively) and failed to achieve a measurable response by RECIST criteria

(Figure 2.4-16 – left scan). The patient was then treated with Selumetinib as a fourth line of

treatment which resulted in the first measurable response (80% reduction in tumor volume) as

observed by MRI imaging and three years of stable disease (Figure 2.4-16 – right scan).

Patient 2, a CMMRD patient harboring homozygous germline PMS2 mutations, was diagnosed at

age 14 years with glioblastoma multiforme (GBM) (WHO Grade IV). The patient was initially

treated with gross total surgical resection and radiation and, upon tumor recurrence, was treated

with the anti-PD1 inhibitor Nivolumab. In parallel, molecular analysis of the tumor confirmed

ultrahypermutation and multiple RAS/MAPK pathway mutations (Figure 2.4-17 a). Following

multifocal progression on Nivolumab alone, addition of Trametinib resulted in remarkable

resolution of these lesions with maximum response in 5 months (Figure 2.4-17 b). This response

persisted for a total of 9 months as a third line of therapy for this cancer. To further elucidate the

immune response to therapy we collected serial blood samples and isolated peripheral blood

mononuclear cells (PBMCs) throughout the duration of Nivolumab administration for patient 2.

Using flow cytometric analysis, we observed an increase in proliferating CD3+CD8+ T cells one

month after addition of Trametinib to anti-PD1 therapy (Figure 2.4-17 c, d). These proliferating

T cells also expressed tumor-reactive T cell marker CD39 (Figure 2.4-17 e) (Canale et al. 2018,

Duhen et al. 2018). This increase in proliferation was followed by an increase in the total

CD3+CD8+ T cell compartment (Figure 2.4-17 c) corresponding with tumor regression as

observed by MRI imaging (Figure 2.4-17 b). Similar increases in proliferating and total

CD3+CD8+ T cells one month post anti-PD1 initiation were observed in other CMMRD patients

with a GBM that responded to anti-PD1 alone, while no changes to the PBMC T cell compartment

62

were observed in CMMRD patients with a GBM that showed no clinical response to anti-PD-1

treatment (Figure 2.4-17 c, d). Collectively, these data further suggest that RRD hypermutant

cancers which upregulate the RAS/MAPK pathway are susceptible to MEK inhibition and the

addition of MEK inhibitors to immune checkpoint blockade may promote an immune-mediated

anti-tumor response.

Figure 2.4-16 Clinical response of Patient 1 with RRD high-grade glioma to Selumetinib

Previous failed treatment courses and sustained responses to Selumetinib

63

64

Figure 2.4-17 Clinical response of Patient 2 with an RRD high-grade glioma to Trametinib and flow cytometric

analysis of anti-PD-1/MEKi synergism

(A) Mutational burden and RAS/MAPK pathway mutations detected by exome sequencing of GMB tumor from

patient MMR190. (B) Previous failed treatment courses and sustained responses to Trametinib. (C) Correlation

between immune activation and clinical response to MEK inhibition in patient 2 (MMR190). Percentage of CD3+CD8+

PBMCs expressing Ki67 (left axis) and fold change in total CD3+CD8+ PBMCs (right axis) over time in patients with

GBM tumors that showed response to anti-PD1 monotherapy, no response to anti-PD1 monotherapy, and response to

combination anti-PD1 and the MEK inhibitor Trametinib (MMR190 – patient 2). Arrow indicates start of Trametinib

administration. (D) Ki67 expression on CD3+CD8+ PBMCs one month following anti-PD1 treatment initiation, and

one month following combination anti-PD1/MEK inhibition (MMR190) or three months following anti-PD1 initiation

(anti-PD1 responder and non responder). (E) Expression of CD39 on Ki67+CD3+CD8+ PBMCs using flow

cytommetry post anti-PD1 treatment alone (left) and combination anti-PD1 + Trametinib treatment (right) for patient

MMR190. * = stopgain mutation; fs = frameshift mutation; ICI = anti-PD1; MEKi = Trametinib.

2.5 Discussion

Data presented in this study demonstrate that within the polyclonal nature of hypermutant

replication repair deficient cancers, multiple mutations in the RAS/MAPK pathway are enriched

and can co-exist, which together render these tumors susceptible to MEK inhibition. Under these

circumstances, our current model of subgrouping tumor types by major driver mutation and

targeting the respective pathway will preclude many potential candidates for this type of targeted

therapy. The one-mutation-one-drug paradigm, as evidenced by targeting BRAF-V600E in

melanoma and glioma may be limited in scope. In the context of hypermutation, a spectrum of

mutations may activate the RAS/MAPK pathway which can collectively be targeted

therapeutically.

While Foundation Medicine data provides a sufficiently large cohort to examine the prevalence of

RAS/MAPK alterations across pediatric cancers, this data is derived from sequencing of tumor only

without comparative normals and some mutations reported in NF1 and PTPN11 may result from

germline predisposition, not an enrichment for somatic events across pediatric cancers. However,

pediatric tumors driven by germline mutations in NF1 and PTPN11 do not result in hypermutated

cancers but rather have extremely low mutational burdens (D'Angelo et al. 2019). Thus, genetic

events in RAS/MAPK genes found in the subset of Foundation Medicine tumors deemed to be

hypermutant are likely to be somatic and not germline events.

As we learn more about the emergence, clonal evolution, and progression of RRD driven

hypermutant cancers (Zehir et al. 2017) a different approach is warranted to manage these patients.

Novel therapies that rely on a specific target, such as CD19 CAR T-cell and tyrosine kinase

65

inhibitor therapy may fail as RRD tumors are polyclonal and continuously accumulate mutations,

ultimately resulting in loss of the antigen or acquiring resistance to the target respectively (Oshrine

et al. 2019).

Dependence of these hypermutant cancers on RAS/MAPK alterations may be explained

conceptually by “oncogene addiction”, defined as cancers that contain multiple genetic and

chromosomal abnormalities but are dependent on or “addicted” to one or a few mutations for both

maintenance of the malignant phenotype and cell survival (Shannon et al. 2017). Moreover,

previous studies have demonstrated striking synthetic lethal interactions between MEK inhibitors

and inhibitors of other DNA repair pathways including homologous recombination (Sun et al.

2017, Maertens et al. 2019). Our findings provide further support that activation of the

RAS/MAPK pathway may synergize with functional DNA repair elements, which may be required

to protect tumor cells undergoing excessive DNA damage. Germline RRD cancers from this study

provide a unique opportunity to study synthetic lethal interactions between RRD and MEK

inhibition.

Mutations in the MMR and polymerase genes result in a hypermutation phenotype, and the

oncogenic mutations that drive these cancers may initially be obscured. A multi-dimensional

analysis, which integrates evidence for pathway activation in the form of multi-clonal point

mutations (DNA level), gene expression (RNA level) and phospho-protein probing (protein level)

provides a strong rationale for therapy targeting a specific pathway even if canonical mutations

are not present.

One oft-cited caveat to employing targeted therapies is the risk of a tumor not responding

completely due to inherent resistance of some clones in the tumor that do not carry the pathway-

specific mutation. Our data suggests that RRD cancers invariably harbor mutations in the

RAS/MAPK pathway, with unique mutations being observed across subclones (Figure 2.4-6 a)

and are found to enrich over time (Figure 2.4-6 b). While our observations suggest that activation

of the RAS/MAPK oncogenic pathway is a necessary process which governs RRD tumors,

additional studies may be needed to further assess tumor dependence on this pathway and whether

such addiction is sufficient to overcome the genomic instability of these cancers. Moreover,

mechanisms of resistance in RRD tumors that respond and subsequently relapse after treatment

66

with MEK inhibitors are currently unknown. Therefore, combination with other pathway inhibitors

or immune based approaches should be considered.

The second patient in this study was treated with the anti-PD1 inhibitor Nivolumab, currently

indicated for MMR deficient and hypermutant cancers (Errico 2015, Overman et al. 2017,

Hellmann et al. 2018). Combinatorial immune checkpoint inhibitors have been suggested to be the

next step in the evolution of immune-based therapies against cancer (Mahoney et al. 2015, Sharma

et al. 2015, Kyi et al. 2016). The success of such combinations, however, may be related to the

underlying mechanisms driving tumorigenesis and hypermutation (Hellmann et al. 2019). Thus,

recent data suggesting a lack of response in non-hypermutant colorectal cancers to combined

immune checkpoint inhibition with MEK inhibitors may not be relevant to RRD hypermutant

cancers (Eng et al. 2019). Our data and the encouraging responses in CMMRD children with RRD

hypermutant cancers suggest that a high mutation burden due to replication repair deficiency may

be required to achieve durable clinical benefit, however clinical trials will be necessary to better

assess the efficacy of MEK inhibitors in these patients.

In summary, this study suggests that a nuanced analytical approach to the pattern of pathway-

specific mutations that arise in hypermutant cancers can reveal a targetable therapeutic

opportunity. Given that hypermutation is observed in up to 20% of cancers at presentation and

possibly higher at recurrence, these approaches should be explored in future clinical trials which

are agnostic to tissue of origin and focus on common mechanisms such as hypermutation.

While the preclinical work conducted in this study is highly promising—particularly when

presented with the encouraging CMMRD patient responses to MEK inhibition (Figure 2.4-16;

Figure 2.4-17)—the model systems used here are entirely patient-derived and are limited with

respect to several properties: 1) Established and patient-derived cell lines lack the multitude of cell

types and structural organization of primary tumor microenvironments; and 2) PDX models, while

replicating the three dimensional tumor architecture, rely on immunocompromised animals, which

are not useful for testing immune based therapies. Transgenic animal models, which mimic the

human syndrome and produce spontaneous cancers, are necessary to further our understanding of

tumor biology and for preclinical testing. Subsequent results chapters contained in this thesis detail

findings from the development of several transgenic mouse models of RRD and the use of such

models for preclinical work.

67

Chapter 3 Cancers from novel Pole mutant mouse models provide insights

into polymerase-mediated hypermutagenesis and immune checkpoint blockade

Chapter 3 was modified and reproduced from:

Galati MA*, Hodel KP*, Gams MS, Sudhaman S, Bridge T, Zahurancik WJ, Ungerleider NA,

Park VS, Ercan AB, Joksimovic L, Siddiqui I, Siddaway R, Edwards M, de Borja R, Elshaer D,

Chung J, Forster VJ, Nunes NM, Aronson M, Wang X, Ramdas J, Seeley A, Sarosiek T, Dunn

GP, Byrd JN, Mordechai O, Durno C, Martin A, Shlien A, Bouffet E, Suo Z, Jackson JG, Hawkins

CE, Guidos CJ, Pursell ZF and Tabori U. “Cancers from novel Pole mutant mouse models provide

insights into polymerase-mediated hypermutagenesis and immune checkpoint blockade. Cancer

Res. 2020 Dec 15(80)(24) 5606-5618; DOI: 10.1158/0008-5472.CAN-20-0624.

* These authors contributed equally. See ‘Statement of Contributions’ for detailed author

contributions.

68

Cancers from novel Pole mutant mouse models provide insights into polymerase-mediated hypermutagenesis and immune checkpoint blockade

3.1 Abstract

POLE mutations are a major cause of hypermutant cancers, yet questions remain regarding

mechanisms of tumorigenesis, genotype-phenotype correlation, and therapeutic considerations. In

this study, we establish mouse models harboring cancer-associated POLE mutations P286R and

S459F, which cause rapid albeit distinct time to cancer initiation in vivo, independent of their

exonuclease activity. Mouse and human correlates enabled novel stratification of POLE mutations

into 3 groups based on clinical phenotype and mutagenicity. Cancers driven by these mutations

displayed striking resemblance to the human ultra-hypermutation and specific signatures.

Furthermore, Pole-driven cancers exhibited a continuous and stochastic mutagenesis mechanism,

resulting in inter- and intratumoral heterogeneity. Checkpoint blockade did not prevent Pole

lymphomas, but rather likely promoted lymphomagenesis as observed in humans. These

observations provide insights into the carcinogenesis of POLE-driven tumors and valuable

information for genetic counselling, surveillance, and immunotherapy for patients.

3.2 Introduction

Eukaryotic DNA replication is a highly accurate process with an error rate of 10-10 mutations per

base per cell division (Drake et al. 1998). This is enabled by the replication repair machinery which

consists of two components. DNA polymerases δ (Pol δ) and ε (Pol ε) are the only nuclear DNA

polymerases capable of highly accurate, processive DNA synthesis and current models place Pol

δ and Pol ε at the lagging and leading strands, respectively (Bebenek et al. 2004, Pursell et al.

2007, Lange et al. 2011, Miyabe et al. 2011, Hogg et al. 2014). These polymerases replicate the

vast majority of the ~3 x 109 basepairs present in the human genome, which must be copied with

high fidelity to preserve the transmission of sequence integrity to cell progeny. Two enzymatic

properties intrinsic to Pol δ and ε that help to accomplish this task include nucleotide selectivity

and 3’ to 5’ exonuclease proofreading.

69

Should a polymerase incorrectly insert a nucleotide and fail to excise the mispaired base, the

mismatch repair (MMR) system can correct these polymerase errors in a post-replicative manner

(Ganai et al. 2016). Complete inactivation of MMR occurs somatically in several tumor types

(Haraldsdottir et al. 2014) and as germline events in constitutional mismatch repair deficiency

(CMMRD) (Bakry et al. 2014) and Lynch Syndrome (LS) (Lynch et al. 2003). Thus, MMR

inactivation is a well-established pathway to mutagenesis and cancer susceptibility. Mutations in

POLE are present in multiple cancers with high tumor mutation burdens (TMB) (The Cancer

Genome Atlas et al. 2012, Church et al. 2013, Levine et al. 2013, Palles et al. 2013, Campbell et

al. 2017). We established the International Replication Repair Deficiency Consortium (IRRDC) to

study these replication repair deficient syndromes and tumors in humans. These studies uncovered

several key questions regarding POLE tumorigenesis that existing human data cannot sufficiently

resolve.

Firstly, only a limited number of POLE mutations cause ultra-hypermutation (>100 mutations per

megabase [Mut/Mb]) and cancer, since such mutations must maintain the protein’s polymerase

ability while disabling proofreading function(Campbell et al. 2017). The nature of genotype-

phenotype correlation of POLE mutations is therefore poorly understood. Initial reports of cases

with germline POLE mutations that disrupt exonuclease activity described a mild phenotype,

mainly consisting of adult-onset polyposis (Briggs et al. 2013, Church et al. 2013, Palles et al.

2013). The phenotypic disparity between this benign polyposis syndrome and more aggressive

cancers attributable to either germline (Johanns et al. 2016) or somatic (Shlien et al. 2015) POLE

mutations suggests variable mutagenic driver capabilities between these mutations. Understanding

the nature of this genotype-phenotype relationship is required to better counsel patients and initiate

appropriate surveillance protocols.

Secondly, the process of mutation accumulation and tumorigenesis in POLE-driven cancers is

poorly described. Human POLE-driven cancers exhibit a unique mutational signature dominated

by three distinct base pair substitutions (BPS) defined by the context of their 5’ and 3’ flanking

bases: C>A-TCT, C>T-TCG (COSMIC Signature 10) (Alexandrov et al. 2013) and T>G-TTT

(COSMIC Signature 28) (Alexandrov et al. 2018, Petljak et al. 2019). Various models of POLE

proofreading deficiency, including experiments in yeast, human cells in culture, and with purified

Pol ε protein in vitro, have independently provided evidence that mutant Pol ε is responsible for

70

portions of the above mutational signature but the signature has not been entirely reconstructed in

cancers in vivo (Shinbrot et al. 2014, Barbari et al. 2018, Hodel et al. 2018). Moreover, analysis

of such signatures in vivo can be used to identify additional secondary mutagenic processes that

act during tumor progression.

Finally, immune checkpoint inhibitors (ICI) have shown dramatic responses in hypermutant

cancers including melanoma, lung, and MMR deficient tumors (Snyder et al. 2014, Le et al. 2015,

Rizvi et al. 2015, Van Allen et al. 2015). However, it is unclear whether POLE mutant tumors will

respond to such therapies. Furthermore, preclinical testing for these drugs is hampered by lack of

immunocompetent mouse models with hypermutant cancers.

Therefore, animal models which can robustly mimic the human POLE syndrome and cancer

development are urgently needed to answer the questions above and to model response of

polymerase mutant hypermutant cancers to immunotherapy.

Two previously reported models revealed striking differences in phenotype and failed to answer

the above questions. Albertson and colleagues generated a mouse model of Pol ε exonuclease

deficiency by engineering double-alanine substitutions at catalytic residues, D272 and E274 (exo-

) in the Pole Exo I motif (Albertson et al. 2009). Strikingly, Pole+/exo- mice do not display any

tumor-susceptibility or abnormal survival despite global, germline expression of the exo- allele,

which retains negligible exonuclease activity at steady-state (Korona et al. 2011, Shinbrot et al.

2014). In contrast, recently reported, LSL-PoleP286R mice that conditionally express the most

recurrent POLE variants in human cancer, rapidly develop a diverse array of ultra-hypermutated

tumors, indicating that the tumorigenic potential of these Pole mutations is variable (Li et al. 2018).

These studies did not discuss the mutagenic processes throughout cancer development or their

therapeutic implications. Moreover, tumors from LSL-PoleP286R mice did not recapitulate

expected Pole-related signatures in their entirety.

To address the tumorigenic and therapeutic potential for POLE cancer variants in vivo, we

developed CRISPR/Cas9-mediated knock-in mice harboring germline P286R and S459F

mutations. The P286R and S459F alleles have been identified in multiple independent human

tumors, with the former representing the most common mutations observed in human POLE driven

cancers (Rayner et al. 2016, Barbari et al. 2017) and the latter is commonly observed as a second

71

somatic occurrence in MMR deficient cancers associated with extremely aggressive cancers with

the highest TMBs(Shlien et al. 2015). Our models exhibit exceptionally strong, yet different,

reduction in survival independent of the degree of exonuclease activity providing insight on human

and mouse genotype-phenotype. Detailed analysis of these tumors revealed a complete

recapitulation of Pole-related mutational signatures and a continuous, stochastic accumulation of

mutations resulting in a high degree of inter and intratumoral heterogeneity affecting tumor

propagation and treatment.

3.3 Materials & Methods

CONTACT FOR REAGENT AND RESOURCE SHARING

Further information and requests for resources and reagents should be directed and will be fulfilled

by co-lead contact Uri Tabori ([email protected]).

3.3.1 Experimental model & subject details

PoleS459F Mouse Model

PoleS459F mice were generated in house at The Centre for Phenogenomics (TCP) (Toronto, ON,

Canada) by direct delivery of Cas9 reagents to C57BL/6J (The Jackson Laboratory, Stock 000664)

mouse zygotes at TCP as performed previously (Gertsenstein et al. 2018). Briefly, a single guide

RNA (sgRNA) with the desired spacer sequence (5’-AGCATCTGACACTGAGTAAG-3’) was

synthesized by in vitro transcription from a PCR-derived template. A microinjection mixture of 20

ng/µL Cas9 mRNA (ThermoFisher, A29378), 10 ng/µL sgRNA, and 10 ng/µL single-stranded

oligodeoxynucleotide (ssODN) template (5’-

ATGAATTTCCTTGGACCCTTGGTTTTTAATGGTCTTGCTCTCTGATGTTCTCCTCAGA

CTCTGGCtACTTACTtcGTGTCAGATGCTGTGGCTACTTACTACCTGTACATGAAATAC

GTCCACCCCTTCATATTCGCCCTGTGCA-3’; mutated nucleotides indicated in lowercase)

was microinjected into C57BL/6J zygotes. Injected zygotes were incubated in KSOMAA media

(Zenith Biotech, ZEKS-50) at 37ºC with 6% CO2 until same-day transfer into CD-1 (Charles River

Labs, Strain 022) surrogate host mothers. PCR primers (5’-

72

AAACTTGGCTATGACCCTGTAGAG-3’ and 5’-GGGATATCACTTCTGAAGACACCAG-

3’) flanking the sgRNA target site and outside of the repair template homology arms, were used to

amplify the region of interest from founder progeny. PCR amplicons were subjected to Sanger

sequencing. Founders with the desired nucleotide changes were selected for breeding with

C57BL/6J mice to produce N1 progeny that were confirmed by sequence analysis of PCR

amplicons using the same primers. PoleS459F mice were housed at TCP and all procedures involving

animals were performed in compliance with the Animals for Research Act of Ontario and the

Guidelines of the Canadian Council on Animal Care. TCP Animal Care Committee reviewed and

approved all procedures conducted on animals at TCP.

PoleP286R Mouse Model

PoleP286R mice were generated at The Jackson Laboratory (Bar Harbor, ME, USA) using a

CRISPR/Cas9-mediated knock-in strategy. Three overlapping sgRNAs with the desired spacer

sequences (5’-TCAGCATCAGGGAATTTGAG-3’, 5’-TCTGATCGGTCTCAGCATCA-3’, 5’-

ATCTGATCGGTCTCAGCATC-3’) were synthesized by in vitro transcription. A microinjection

mixture of wildtype Cas9, sgRNA, and ssODN (5’-

GTCCTTTTAGGACCCTGTGGTTTTGGCATTTGACATCGAGACGACCAAACTGCCTCTC

AAATTCCgTGATGCcGAGACCGATCAGATCATGATGATCTCCTATATGATTGATGGCC

AGGTGAACAGAATC-3’; mutated nucleotides indicated in lowercase) was microinjected into

C57BL/6J zygotes. Injected zygotes were then transferred to pseudopregnant female mice. PCR

primers (5’-TCCAAGATGAAGATGTTGTCCC-3’ and 5’-CTAATCCACCCACAAGCCTC-3’)

residing outside the of the repair template homology arms and capturing the gRNA target site were

used to amplify the region of interest from founder progeny. Sanger-sequencing of PCR amplicons

were performed to identify PoleP286R/+ mice, which were selected for breeding with C57BL/6J

mice to produced N1 progeny. Genotyping of N1 mice were performed as above. All animal care

and procedures were approved by the Tulane School of Medicine Institutional Animal Care and

Use Committee (Protocol #4445). All PoleP286R/+ and wildtype control mice were housed at the

Tulane University SOM Vivarium (New Orleans, LA, USA).

Mlh1-/- Mouse Model

73

Mlh1-/- mice (Edelmann et al. 1999) were provided by A. Martin (University of Toronto) and

housed as indicated above for PoleS459F mice. Mlh1 genotyping was performed as indicated

previously(Edelmann et al. 1999) using the following primers: primer A,

TGTCAATAGGCTGCCCTAGG; primer B, TGGAAGGATTGGAGCTACGG; and primer C,

TTTTCAGTGCAGCCTATGCTC.

3.3.2 Method Details

Determination of Pol ε Expression

Expression of the PoleS459F mutant allele was assessed in ear notches derived from adult

PoleS459F/+, PoleS459F/S459F, and Pole+/+ mice. Total RNA was purified using the RNeasy Mini Kit

(Qiagen). cDNA was synthesized using SuperScript® IV Reverse Transcriptase (Thermo Fisher

Scientific) on 1μg of RNA. PCR was subsequently performed on cDNA with intron-spanning

primers (forward: 5’-ACTGCCTCAGGTGGGTGAAG-3’ and reverse: 5’-

CTTCCGCAGCACCTCACTAGG-3’). Products were visualized by agarose gel electrophoresis

using GelRed (Biotium). Gel purification was performed followed by Sanger sequencing to detect

mutant and WT alleles.

Pathology

Complete necropsies were performed, and the following tissues were taken for histology: brain,

liver, spleen, heart, lungs, thymus, small and large intestines, and any abnormal tissue. Tissues

samples were fixed in 10% phosphate buffered formalin, embedded in paraffin, sectioned, stained

with hematoxylin and eosin, and examined by light microscopy. Pathologists (I. Siddiqui and C.

Hawkins) reviewed all slides. Tumors were confirmed histologically.

Immunohistochemistry

Immunostaining was performed on unstained 5 µm slides from formalin-fixed, paraffin embedded

(FFPE) blocks. Sections were deparaffinized, rehydrated, and subjected to heat-mediated epitope

retrieval using 0.01M Citrate Buffer pH 6.0. Sections were washed 2x 5 min in PBS-T and, blocked

in 2.5% Normal Goat Serum and incubated with either anti-CD3 (Dako, A0452; 1:200 dilution)

or anti-B220 (Pharmingen, 553084; 1:2000) primary antibodies for 1 hour at room temperature.

74

Endogenous peroxidases were blocked using BLOXALL Blocking Solution (SP-6000) at room

temperature for 15 minutes. Sections were subsequently washed and treated with ImmPRESS

(Perixodase) Polymer IgG reagent (either ImmPRESS HRP Anti-Rabbit IgG (Peroxidase) Polymer

Detection Kit, Vector Laboratories MP7451 or ImmPRESS HRP Anti-Rat IgG (Peroxidase)

Polymer Detection Kit, Vector Laboratories MP-7404) for 30 minutes at room temperature.

Sections were washed and then treated with DAB (Vector Laboratories SK-4100) for visualization

followed by double distilled water for to stop the reaction. Sections were counterstained with

Hematoxylin, dehydrated, mounted with a coverslip using VectaMount Permanent Mounting

Media (Vector Laboratories, H-5000).

Literature Search for case reports of germline POLE mutations

MeSH search terms: ((((((("hereditary colorectal cancer") OR "colorectal cancer") OR

"endometrial cancer") OR polymerase proofreading associated polyposis) OR "genetic risk

factors") AND POLE) AND mutations) AND germline).

POLE exonuclease excision rate assay

Excision rate constants were measured as previously described (Zahurancik et al. 2014,

Zahurancik et al. 2015). Briefly, a pre-incubated solution of Pol e (100 nM) and 5′-32P-labeled

DNA substrate (20 nM) was rapidly mixed with Mg2+ (8 mM) in reaction buffer at 37°C. After

various incubation times, the reaction was quenched with the addition of EDTA. The excision rate

constants for POLE wild-type, D275A/E277A, P286H, P286R, F367S, L424V, L424I, and S459F

were measured using a rapid chemical quench-flow apparatus. Product concentration was plotted

versus time and fit to a single-exponential equation, [product] = Aexp(-kexot), to yield the excision

rate constant, kexo.

Prophylactic Immunotherapy

PoleS459F/S459F mice were stratified into four treatment groups and beginning at 6-8 weeks of age

were injected intraperitoneally (IP) with either: 1. Vehicle-only (PBS, pH = 7.0); 2. Anti-mouse

CTLA-4 200μg (BioXcell, Catalog No. 648317J2B); 3. Anti-mouse PD1 250μg (BioXcell,

Catalog No. 665418F1); or 4. Anti-mouse CTLA-4 200μg + anti-mouse PD1 250μg. Mice were

injected every 3-4 days until endpoint. At endpoint, mice were euthanized and whole-body

75

necropsied. All macroscopically visible abnormal tissue was collected and subsequently divided

to formalin-fix and paraffin-embed, snap freeze, and dissociate for viable freezing of tumor cells.

Statistical Analysis

Kaplan-Meier curves of mouse survival logs were generated via the GraphPad Prismv7 software.

Log-rank test was performed with the same software to calculate the p-value. Statistical

significance was set at p < 0.05.

Whole Exome Sequencing and Variant Calling

DNA from both tumor and normal tissues was isolated using the Qiagen DNeasy Kit (Catalog No.

69504) and subsequently submitted to The Center for Applied Genomics (Hospital for Sick

Children, Toronto, ON, Canada), for whole exome sequencing (WES), alignment to the reference

genome and variant calling. Agilent’s SureSelectXT Mouse All Exon kit was used for enrichment

and paired-end sequencing was done on Illumina HiSeq4000. For two PoleP286R/+ tumors (1098

and 1144), tumor and control tail gDNA was submitted to the Beijing Genomics Institute (BGI)

Americas at Children’s Hospital of Philadelphia (CHOP). Prior to sequencing, DNA was enriched

using Agilent SureSelect XT Mouse All Exon Kit. Paired-end WES was performed on an Illumina

HiSeq4000. FASTQ files from these two mice were processed in the same manner as all other

samples in the current study as follows. The software bcl2fastq2 v2.17 was used to generate raw

FASTQ files. Alignment to the mouse mm10 reference genome was done using BWA-MEM

0.7.12, followed by PicardTools 1.133 to mark duplicates and sort the BAM file, and GATK 3.4-

46 ‘IndelRealigner’ for local realignment of reads. Variant calling was done as described by Doran

et al., 2016 using SAMtools 1.3.1 (for mpileup), BCFtools 1.3.1 and VCFtools 0.1.12a(Danecek

et al. 2011). GATK MuTect2 was also used to compare tumor and matched normal samples to call

somatic SNVs and indels and annotation of variants was done using ANNOVAR (version Feb

2016).

Data Analysis

The TMB from WES data was plotted per chromosome using easyGgplot2 on R.

DeConstructSigs(Rosenthal et al. 2016) was used to determine COSMIC signatures (Alexandrov

et al., 2013) in the mutation spectrum within a tri-nucleotide context for each sample. All analyses

76

were done on R version 3.4.4 using the high-performance computing cluster at the Hospital for

Sick Children.

Data Availability

Data generated during our study have been deposited in the NCBI’s Sequence Read Archive (SRA)

database under the accession code PRJNA659565.

3.3.3 Mass Cytometry Methods

Cell Staining for Mass Cytometry

Spleen, lymph node or thymus from PoleS459F/+ or PoleS459F/S459F mice showing clinical signs of

malignancy and enlargement of one or more lymphoid tissues were mechanically digested to

release single cells and cryopreserved. Freshly harvested spleen from Pole+/+ control mice was

dissociated into single cell suspensions on the day of each immunophenotyping experiment to use

as a staining control. Cryopreserved cells were thawed and washed twice in pre-warmed complete

(c) RPMI (RPMI, 10% FBS, 25mM Hepes, 55 μM β-mercaptoethanol, 0.1 mM non-essential

amino acids, 1mM sodium pyruvate, 2mM L-Glutamine) containing 200 μg/ml DNAse. Cells were

pelleted, resuspended in cRPMI, and counted using Trypan Blue via a hemocytometer. Cell

concentration was adjusted to 2x106 cells/ml and placed at 37°C in a humidified 5% CO2

incubator for 30ʹ. After 30’, 500nM 127IdU (5-127Iodo-2’ deoxyuridine) was added for an additional

60ʹ to label newly synthesized DNA in proliferating cells. Cells were then washed prior to blocking

Fc Receptors as previously described(An et al. 2018), washed again and then stained for 30’ at RT

with anti-CD45 tagged with 89Y or 156Gd. After washing, pairs of bar-coded samples (stained

with different CD45 conjugates) were combined into a single 5 mL tube polypropylene to perform

multi-plexed staining with pre-determined optimal concentrations of metal-tagged antibodies

specific for cell surface markers, Cisplatin (BioVision Inc., USA) to stain dead cells and

transcription factor antibodies as previously described(An et al. 2018). After the final wash, cells

were re-suspended in PBS containing 0.3% saponin, 1.6% formaldehyde, and 100 μM

191/193Iridium to stain nuclear DNA for up to 48h at 4°C. Prior to analyzing stained cells on the

Helios, cells were washed and re-suspended in Maxpar Cell Acquisition Solution (Fluidigm,

Markham ON Canada) at 2-5x105/ml followed by addition of 5-element EQ normalization beads

77

(Fluidigm, Markham ON, Canada). Samples were acquired on the Helios according to Fluidigm’s

protocols. The Helios software (v6.7.1014) was used for pre-processing to generate and normalize

FCS 3.0 datafiles.

Metal-tagged Antibodies

The following antibodies were directly purchased from Fluidigm: CD45-89Y, CD150-167Er, and

Lag3-174Yb. All other antibodies were purified from hybridoma supernatants in-house or

purchased as purified carrier-free antibodies and metal-tagged using Fluidigm Maxpar Metal

Conjugation Kits according to the manufacturer’s instructions. Panel and antibody information can

be found in Supplemental Table S3 – see Appendices. This panel was used in two experiments

to stain single cell suspensions from enlarged lymphoid tissues from a total of 16 untreated mice,

4 anti-CTLA-4 treated mice, 5 anti-PD1 treated mice, and 6 combination therapy treated mice.

CyTOF Data Analysis

FCS 3.0 files were uploaded into Cytobank (Santa Clara, CA) and each parameter was scaled using

the Arcsinh transformation (scale argument of 5). Each 2-plex FCS file was manually de-

convolved into separate FCS files (using the ‘split files by population’ feature in Cytobank)

containing either CD45-89Y+ or CD45-156Gd+ cells, which were then further gated to remove EQ

beads, dead cells, debris, and aggregates. FCS files containing 12,500 CD45+ live single cells from

each sample were exported for clustering using the FlowSOM algorithm

(github.com/SofieVG/FlowSOM) (Van Gassen et al. 2015). Clustering was performed (k = 30,

Arcsinh scale argument of 5) on 2 WT spleen and 16 tumor samples using the markers indicated

in Supplemental Table S3 – see Appendices. The FlowFrame utilility within the Bioconductor

R package ‘flowCore’ was used to create new FCS files that include the FlowSOM cluster IDs,

which were then uploaded to Cytobank where t-SNE dimensionality reduction was performed

(iterations: 3000, perplexity:30, theta: 0.5) using the clustering markers. The FlowSOM cluster

IDs were also included to enhance visualization of the FlowSOM clusters in the t-SNE

embedding(Amir el et al. 2013).

Statistical Analysis for Mass Cytometry Data

78

Prismv8.2.0 was used to perform ordinary one-way ANOVA with post-hoc comparisons of means

of each column with every other column (Figure 3.4-9 b, c; 3.4-10 c) or the mean of untreated

control with the other columns (Figure 3.4-12 c). Correction for multiple testing was performed

using the two-stage linear step-up procedure of Benjamini, Krieger and Yekutieli (FDR=5%) to

calculate adjusted p values (ie., q values). In Figures 3.4-8, 3.4-9, 3.4-10, and 3.4-12, *: q<0.033;

**: q<0.002; ***: q<0.001.

3.4 Results

3.4.1 Pole mutant mice provide insight into genotype-phenotype in humans

To model the role of different POLE exonuclease domain mutants (EDM) on tumorigenesis in

vivo, we used a CRISPR/Cas9-mediated knock-in approach to generate mice harboring the

clinically relevant, P286R or S459F substitutions (Figure 3.4-1; 3.4-2). Expression of mutant and

WT alleles was confirmed by sequencing of cDNA amplified using intron-spanning primers

(Figure 3.4-1; 3.4-2). Litters from PoleS459F/+ × PoleS459F/+ breeders exhibited normal Mendelian

distribution of WT and mutant Pole alleles (p = 0.9869, Chi-Square independence test). However,

litters from PoleP286R/+ × PoleP286R/+ breeders never produced PoleP286R/P286R neonates (p = 0.0014,

Chi-Square independence test) and PoleP286R/P286R embryos were not observed after embryonic day

E13.5, suggesting embryonic lethality (Figure 3.4-1). In contrast, male and female PoleS459F/+,

PoleS459F/S459F, and PoleP286R/+ mice reached sexual maturity and were fertile.

Although all PoleS459F/+, PoleS459F/S459F, and PoleP286R/+ mice survived into adulthood, they rapidly

succumbed to aggressive cancers beginning as early as 1.2 (PoleS459F/S459F), 4.1 (PoleS459F/+) and

2.0 (PoleP286R/+) months. Tumor-free survival differed significantly between genotypes (Figure

3.4-3 a). As expected, given the additional mutant allele, PoleS459F/S459F mice had significantly

shorter median survival compared to PoleS459F/+ mice (p < 0.0001). Moreover, PoleP286R/P286R mice

were non-viable, confirming a more severe phenotype in homozygous mutants compared to

heterozygous P286R mutants in vivo consistent with previously characterized Pole mutant mice

(Albertson et al. 2009, Li et al. 2018). Importantly, PoleP286R/+ mice had shorter tumor-free survival

than PoleS459F/+ mice, and since PoleS459F/S459F were viable, the same could be presumed for

79

homozygous mutants. Furthermore, the influence of a single P286R or S459F allele on survival

observed here was remarkably more severe than the Pole exo- allele previously reported (Albertson

et al. 2009). Together, these findings reveal a genotype-phenotype correlation between Pole

mutant genotypes and tumor penetrance in mice.

Necropsy and pathology findings revealed that the majority of PoleS459F/+, PoleS459F/S459F, and

PoleP286R/+ mice had hepatosplenomegaly, large mediastinal (thymic) masses, and/or enlarged

lymph nodes, suggesting the presence of lymphomas with similar characteristics to those described

in replication repair deficient (RRD) models (Reitmair et al. 1995, Prolla et al. 1998, Goldsby et

al. 2002) (Table 3.1 and Figure 3.4-1; 3.4-2). Other tumor types observed in PoleS459F mice

included those observed in humans such as adenoma and adenocarcinoma of the gastrointestinal

(GI) tract, and other uncommon cancers such as sarcoma, and germ cell.

To determine the relevance of this correlation to human disease, particularly with respect to

instances of germline POLE mutations, we collected clinical data from the IRRDC (Shlien et al.

2015, Campbell et al. 2017) and published data on cases harboring germline or sporadic POLE

mutations (summarized in Supplementary Table S4 – see Appendices). We found that driver

status POLE mutations could be classified into three distinct categories corresponding to their

clinical presentation (Figure 3.4-3 b). Less severe mutations, including several outside the

exonuclease domain, were detected in the germline of individuals across families, had low

penetrance, predisposing to a mild phenotype of adult-onset polyposis of the GI tract and late-

onset colorectal cancer (CRC). A second set of mutations, including P436R, were also found in

the germline and predisposed to early-onset, more aggressive tumors such as brain malignancies

and childhood-onset CRC. Finally, a third group of mutations, including P286R and S459F, were

only observed as somatically acquired mutations in various cancers suggesting that these mutations

are more aggressive than mutations tolerated in the germline.

To explore the potential mechanisms of this phenomenon and its correlation to the human

syndrome we first determined whether survival differences between Pole mutant mice could be

explained by differing effects of mutations on exonuclease function. We performed in vitro

enzymatic reactions as previously conducted to assess exonuclease activity of several POLE

mutants (Zahurancik et al. 2014, Zahurancik et al. 2015). Interestingly, the S459F mutation ablated

80

exonuclease activity nearly 100-fold more than either the P286R or D275A/E277A mutations,

suggesting that exonuclease activity alone cannot explain the ability of a mutation to drive tumor

development in vivo (Figure 3.4-3 c). Recent data in Saccharomyces cerevisiae demonstrate that

the mutation rate of P286R is the highest observed, and both P286R and S459F are significantly

higher than D275A/E277A (Barbari et al. 2018). Moreover, the ultramutator phenotype conferred

by the P286R mutation may be due to hyperactive polymerase activity rather than decreased

exonuclease activity alone (Xing et al. 2019). These findings provide possible mechanisms that

explain differences in survival observed between Pole mutants and support a genotype-phenotype

correlation of POLE mutations in humans.

81

5’-TTTGGGACTGAGGTG-(178)-CCAAACTGCCTCTCAAATTCCCTGATGCTGAGACCGATCAGATCATGAT-(46)-AAACTTAGAATGCTTTTGTGGGG-

3’

3’-AAACCCTGACTCCAC-(178)-GGTTTGACGGAGAGTTTAAGGGACTACGACTCTGGCTAGTCTAGTACTA-(46)-TTTGAATCTTACGAAAACACCCC-

5’

Allele

WT

Mutant 5’-TTTGGGACTGAGGTG-(178)-CCAAACTGCCTCTCAAATTCCGTGATGCCGAGACCGATCAGATCATGAT-(46)-AAACTTAGAATGCTTTTGTGGGG-3’

3’-AAACCCTGACTCCAC-(178)-GGTTTGACGGAGAGTTTAAGGCACTACGGCTCTGGCTAGTCTAGTACTA-(46)-TTTGAATCTTACGAAAACACCCC-5’

DdeI

312 bp (full PCR product)

288 bp

76 bp212 bp

DdeIP286

R286

DdeI

Pole +/+ +/P286R P286R/P286R

DdeI + + - +

L

100

200

300

76

212

288312

82

Figure 3.4-1 PoleP286R mouse model design, validation, and tumor findings

(A) Schematic showing a portion of the mouse Pole gene used for CRISPR-Cas9 targeting for the P286R mutation.

Altered bases are underlined in black and amino acid changes are bolded. Numbers indicate amino acid position.

Green line indicates the sequence of the sgRNA and red line indicates the PAM site. (B) Validation of mutant Pole

allele expression in PoleP286R/+ mice via Sanger sequencing of RT-PCR products from Pole+/+ and two PoleP286R/+

mice. (C) Genotyping Pole+/P286R and PoleP286R/P286R mouse E13.5 embryos. (Top) The indicated 312bp PCR-amplified

genomic DNA sequence surrounding the Pro286 codon is shown. The Pole WT allele contains three DdeI restriction

sites. CRISPR editing knocked in the P286R codon change (red) and a silent T:A>C:G change (blue) that inactivates

on DdeI site. Diagnostic DdeI restriction fragment sizes are shown. (Bottom) Agarose gel showing DdeI digests of

Pole+/+, Pole+/P286R controls and PoleP286R/P286R embryo. Fragment sizes (bp) shown on left. (D-H) (Left) Necropsy and

(Right) histological findings from PoleP286R and PoleS459F mice. White bar indicates 50μm.

83

84

Figure 3.4-2 PoleS459F mouse model design, validation, and tumor findings

(A) Schematic showing a portion of the mouse Pole gene used for CRISPR-Cas9 targeting for the S459F mutation.

Altered bases are underlined in black and amino acid changes are bolded. Numbers indicate amino acid position.

Green line indicates the sequence of the sgRNA and red line indicates the PAM site. (B) Validation of mutant Pole

allele expression in PoleS459F/S459F and PoleS459F/+ mice via Sanger sequencing of RT-PCR products from PoleS459F/S459F,

PoleS459F/+, and Pole+/+ mice. (C-L) (Left) Necropsy and (Right) histological findings from PoleS459F mice. Black bar

indicates 50μm.

Figure 3.4-3 Pole mutations confer variable tumorigenic capabilities in vivo and support a genotype-phenotype

correlation

(A) Kaplan-Meier survival estimates. Mice were followed for long-term survival and observed daily until endpoint.

Upper: PoleS459F/S459F (n= 31); PoleS459F/+ (n=33); Pole+/+ (n=16); Lower: PoleP286R/+ (n = 27); Pole+/+ (n=17). One

month = 4.3 weeks. Significance are indicated using Log-rank test. (B) Landscape of germline and somatic POLE

driver mutations. Mutations designated as germline were collected from the IRRDC or were previously reported.

Mutations designated as “somatic only” are not found in germline cases but were validated drivers as determined by

our previous comprehensive characterization (Campbell et al. 2017). Residues upstream of the exonuclease domains

(1-268) and downstream of the polymerase domain (1100 – 2286) are omitted for clarity. (C) Excision rate constants

measured for 7 POLE exonuclease mutants. POLE mutants are indicated on the x-axis. Error fitting for each curve

was performed as described previously (Zahurancik et al. 2014).

85

Table 3-1 Spontaneous tumor incidence across PoleS459F/S459F, PoleS459F/+, and PoleP286R/+ moribund

mice

Moribund mice were euthanized and necropsied, and tumors were diagnosed by histology. *Incidences (%) among 31

PoleS459F/S459F, 33 PoleS459F/+, and 27 PoleP286R/+ mice. ¥For lymphomas mice were divided into two groups: 1) mice

with mediastinal masses (thymic lymphomas), which typically included infiltration and enlargement of the spleen and

other lymph nodes, and 2) mice with spleen and/or lymph node enlargement (without macroscopic thymic

enlargement).

Incidence (%)*

Tumor PoleS459F/S459F PoleS459F/+ PoleP286R/+

Lymphoma¥

Thymic 45 39 26

Splenic/Lymph Node 43 52 74

Adenocarcinoma

Small Intestine 6 9

Other Neoplasms

Testicular 3

Sarcoma 3

3.4.2 The genetic profiles of Pole mutant tumors resemble human cancers

POLE exonuclease deficiency in human malignancy is genetically characterized by genome-wide

ultra-hypermutation and mutational signatures 10 (Alexandrov et al. 2013) and 28 (Alexandrov et

al. 2018, Petljak et al. 2019). However, the kinetics of mutation accumulation over time is not

known. To study these processes, we performed whole exome sequencing (WES) on multiple

tumors and spatial locations from PoleP286R/+, PoleS459F/+, and PoleS459F/S459F mice. Sequenced

tumors mainly consisted of lymphomas (26/29 across all three Pole genotypes) since these were

the predominant tumors observed, but also included several non-lymphoma tumors (1 GI, 1

testicular cancer, and 1 sarcoma) (Figure 3.4-4 a). All Pole mutant tumors were ultra-hypermutant

with an average TMB of 170.5 Mut/Mb across all three genotypes (Figure 3.4-4 a), comparable

to findings from human POLE mutant cancers. To ensure these findings were unique to Pole

mutant mouse tumors, we sequenced tumors (5 lymphomas and 1 GI cancer) from MMRD mice

harboring homozygous deletion of Mlh1 (Mlh1-/-)(Edelmann et al. 1999), as well as a cohort of

86

tumors (5 lymphomas, 2 brain tumors, and 1 lung cancer) from replication repair proficient mice.

Mlh1-/- tumors were hypermutant and exhibited mutational burdens consistent with human LS and

CMMRD cancers (between 10 to 100 Mut/MB) but were significantly lower than all Pole mutant

tumors (p=0.00022). Both Pole mutant and Mlh1-/- mouse tumors harboured significantly higher

mutational loads than replication repair proficient mouse cancers (p=2x10-5 and 0.0037,

respectively; Figure 3.4-4 a). These mutations were evenly distributed throughout the exome

(Figure 3.4-4 b) suggesting a stochastic mutagenesis mechanism.

Mutational signatures are often indicative of specific mechanisms of mutagenesis. We plotted the

proportion of tumor SNVs within specific trinucleotide contexts and performed signatures analysis

using the R package deconstructSigs (Rosenthal et al. 2016). Signatures 10 and 28—characteristic

of POLE exonuclease deficient tumors (Alexandrov et al. 2013, Alexandrov et al. 2018, Petljak et

al. 2019)—were found in all Pole mutant mouse tumors and together constituted the largest

proportion of signatures present in each tumor (Figure 3.4-5).

To study how mutations accumulate in POLE mutant cancers, we examined tumors from multiple

sites. As germline RRD in humans can result in multiple synchronous cancers (Bakry et al. 2014),

we first determined whether all lesions from several sites (liver, spleen, lymph node, and thymus)

indeed originated from a single tumor. We examined sequencing data from 4 such mice (2

PoleP286R/+, 1 PoleS459F/+, and 1 PoleS459F/S459F). In all cases, there was a significant number of

common SNVs present in all tumor tissues (Figure 3.4-6 a; Figure 3.4-7). These data suggest that

a single hematopoietic malignancy arose in Pole mutant mice allowing us to study mutational

processes temporally and spatially.

We next studied the mutational landscape of multi-site tumors in more detail. Despite significant

overlap in mutations (>1000 SNVs shared), each tumor fraction also had a significant proportion

of mutations unique to that fraction. By plotting the variant allele fractions (VAFs) of mutations

unique and common to tumor fractions we were able to observe the emergence of both early and

late clones (Figure 3.4-6 b; Figure 3.4-7), supporting a stochastic mutagenesis mechanism that

acts continually.

We then used mutational analysis as described above to detect unique signatures in specific tumor

clones in sites absent in other fractions. For example, PoleS459F/S459F tumors from mouse 4802

87

revealed signatures specific to MMR deficiency and combined MMR and POLE exonuclease

deficiency only in the thymus (Figure 3.4-6 c). This was not observed in common, shared, or

private SNVs in the liver and spleen. The phenomenon of complete RRD (loss of both MMR and

polymerase proofreading) recapitulates our observations in human cancers which may start as

polymerase mutant, and subsequently develop MMRD (Campbell et al. 2017) and in cases of

CMMRD that develop secondary somatic polymerase exonuclease deficiency (Shlien et al. 2015).

Together, ultrahypermutation, the pattern of mutational accumulation, and signatures suggest that

the underlying mutagenesis mechanism for Pol ε exonuclease deficiency is consistent between

humans and mice and is a stochastic and continuous process.

A

0.027

0.0014

0.00045

0.0037

0.00094

0.0015

0.00033

Muta

tions/M

b B

P286R Mouse 1144 Thymus

MutM

B

Chromosome

MutM

B

1

10

100

1000S459F Mouse 4802 Thymus

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

1

10

100

1000

1

10

100

250

1000

Genotype

PoleS459F/S459F

PoleS459F/+

PoleP286R/+

Mlh1−/−

NonRRD

Brain

GI

Lung

Lymph

SC

Test

Tumor Type PoleS459F/S459F PoleS459F/+ PoleP286R/+ Mlh1−/− NonRRD

2

-

1

5

-

1

-

5

-

-

-

-

-

-

-

11

-

-

-

1

-

7

-

-

-

-

-

8

1

1

88

Figure 3.4-4 The genomic landscape of Pole mutant mouse tumors resembles that of POLE-driven human

cancers

(A) Left -Violin plot showing mutation frequency from tumor whole exome sequencing (WES) of mouse tumors. Pole

mutant mouse cancers (PoleS459F/S459F n = 10; PoleS459F/+ n = 8; PoleP286R/+, n = 11) were compared to MMRD mouse

tumors (n=6) and replication repair proficient mouse-derived tumors (n =8). The red dashed line indicates 100

mutations/Mb. Significance are indicated using Student’s T-test. Scatter plot symbols indicate tumor type for each

group and are summarized in: Right – table summarizing tumor types sequenced for each genotypic group. Brain =

brain tumor; GI = gastrointestinal adenocarcinoma; Lung = lung adenocarcinoma; Lymph = lymphoma; SC = sarcoma;

Test = testicular germ cell tumor. (B) Mutation frequencies, as calculated by the number of mutations per target region

covered, are plotted per chromosome, and reveal no evidence of localized hypermutation.

Figure 3.4-5 Pole mutant mice tumors exhibit mutational signatures found in POLE driven human cancers

Whole-exome sequencing (WES) derived mutational signatures in PoleS459F/S459F, PoleS459F/+, and PoleP286R/+ mouse

tumors. The 96 possible mutation types based on trinucleotide context are shown for each sample that underwent WES

in this study. Cosmic signatures as called by deconstructSigs are shown for each sample (right).

Pole S459F/+ mouse tumours cont’d

VN

Sl

ato

Tf

ot

ne

cre

P

C>A C>G C>T T>A T>C T>G

3031-Thymus

0%

5%

10%

15%

20%

ACA

ACG

CCA

CCG

GCA

GCG

TCA

TCG

ACA

ACG

CCA

CCG

GCA

GCG

TCA

TCG

ACA

ACG

CCA

CCG

GCA

GCG

TCA

TCG

ATA

ATG

CTA

CTG

GTA

GTG

TTA

TTG

ATA

ATG

CTA

CTG

GTA

GTG

TTA

TTG

ATA

ATG

CTA

CTG

GTA

GTG

TTA

TTG

Signature.1

Signature.10

Signature.12

Signature.15

Signature.17

Signature.21

Signature.28

unknown

VN

Sl

ato

Tf

ot

ne

cre

P

C>A C>G C>T T>A T>C T>G

3124-Thymus

0%

5%

10%

15%

20%

ACA

ACG

CCA

CCG

GCA

GCG

TCA

TCG

ACA

ACG

CCA

CCG

GCA

GCG

TCA

TCG

ACA

ACG

CCA

CCG

GCA

GCG

TCA

TCG

ATA

ATG

CTA

CTG

GTA

GTG

TTA

TTG

ATA

ATG

CTA

CTG

GTA

GTG

TTA

TTG

ATA

ATG

CTA

CTG

GTA

GTG

TTA

TTG

Signatur e.1

Signatur e.10

Signatur e.15

Signatur e.28

unkn own

VN

Sl

ato

Tf

ot

ne

cre

P

C>A C>G C>T T>A T>C T>G

3124-Ureter

0%

5%

10%

15%

20%

ACA

ACG

CCA

CCG

GCA

GCG

TCA

TCG

ACA

ACG

CCA

CCG

GCA

GCG

TCA

TCG

ACA

ACG

CCA

CCG

GCA

GCG

TCA

TCG

ATA

ATG

CTA

CTG

GTA

GTG

TTA

TTG

ATA

ATG

CTA

CTG

GTA

GTG

TTA

TTG

ATA

ATG

CTA

CTG

GTA

GTG

TTA

TTG

Signatur e.1

Signatur e.10

Signatur e.28

unknown

Pole S459F/S459F mouse tumours

VN

Sl

ato

Tf

ot

ne

cre

P

C>A C>G C>T T>A T>C T>G

46-Testes

0%

5%

10%

15%

20%

ACA

ACG

CCA

CCG

GCA

GCG

TCA

TCG

ACA

ACG

CCA

CCG

GCA

GCG

TCA

TCG

ACA

ACG

CCA

CCG

GCA

GCG

TCA

TCG

ATA

ATG

CTA

CTG

GTA

GTG

TTA

TTG

ATA

ATG

CTA

CTG

GTA

GTG

TTA

TTG

ATA

ATG

CTA

CTG

GTA

GTG

TTA

TTG

Signature.1

Signature.10

Signature.28

unknown

89

Figure 3.4-6 The genomic landscape of Pole mutant mouse tumors gives insight into mutagenesis mechanisms

(A) Exome data from tumor fractions from a single mouse were compared. Private SNVs were defined as SNVs that

were present only in one specific fraction. Shared SNVs were present in two fractions. Common SNVs were present

in all fractions. The number of private, shared, and common SNVs are all indicated. Corresponding tumor fractions

are indicated. (B) Density plots of the number of single nucleotide variants (SNVs) by variant allele fraction (VAF)

in each of the tumor fractions. SNVs unique to indicated tumor fractions and common to all tumor fractions were

subsetted and plotted for each fraction. Red arrows indicate the presence of common SNVs in early and late clones

for thymus vs. spleen and liver, respectively. (C) Constructed evolutionary tree for tumor 4802 based on three

sequenced fractions. Length of branch is proportional to the number of SNVs. Colors correspond to private, shared,

and common SNVs indicated in (C). Stacked bars indicate the proportion and presence of mutational signatures.

90

91

92

Figure 3.4-7 (related to Figure 3.4-6) Exome data from tumor fractions from a single mouse were compared

for 3 additional mice

93

3.4.3 Pole mutant mice develop two distinct types of T cell lymphoma

Having established the genetic processes affecting tumorigenesis in Pole mutant mice, we used

high dimensional immune phenotyping to better understand whether the unique genetic

mechanism affects the cellular origin and heterogeneity of lymphomas in mutant mice. We used

mass cytometry to simultaneously analyze expression of 30 markers of hematopoietic cell lineage

and activation/differentiation state (Supplementary Table S3 – see Appendices) among single

cells from enlarged thymus (T), spleen (S or SPL) or lymph nodes (L) of moribund PoleS459F/S459F

(n=12) and PoleS459F/+(n=4) mice. Unsupervised FlowSOM clustering followed by t-Stochastic

Neighbor Embedding (tSNE) revealed striking differences in the phenotype and abundance of T

and B cell subsets in lymphoma samples versus those in wild-type (WT) SPL (Figure 3.4-8 a-c).

Cell clusters expressing T lineage markers (TCRβ, CD4 and/or CD8β) were more abundant in the

lymphoma samples than in WT SPL, suggesting that the lymphomas were of T lineage origin.

However, in some samples there were multiple T lineage clusters with varying expression of

TCRβ, CD4, and/or CD8β, whereas in others there was a single predominant TCRβ+ CD4+ CD8β-

cluster.

The t-SNE plots also showed that two different T cell subsets were proliferating across samples,

based on incorporation of 127I-iododeoxyuridine (IdU), a thymidine analogue. There was little

overlap among the T cell clusters or proliferating cells on the t-SNE plots in these two groups,

suggesting two distinct T cell lymphoma subtypes, which we termed Group A and Group B. Group

A lymphomas likely came from thymus where T cells develop, and subsequently tumors spread to

SPL, lymph node and liver (see Table 3.1). Group B lymphomas likely came from SPL or lymph

node, which are secondary lymphoid tissues where naïve T and B cells undergo antigen-triggered

differentiation into immune effector cells. Group B samples contained more B and myeloid cells

than Group A samples (Figure 3.4-8 c), likely reflecting their different tissues of origin.

Immunohistochemistry (IHC) staining for CD3 in WT SPL, Group A and Group B tumors as well

as staining for B220 in WT SPL and Group B tumors supported these findings (Figure 3.4-8 d).

Because non-T cells were abundant in Group B lymphomas, we generated a heatmap to better

visualize group-specific expression of T cell markers among cells lacking B and myeloid cell

markers. These analyses revealed significant inter-and intratumoral heterogenity both between and

94

within Group A and Group B lymphomas. Group A samples expressed high amounts of CD8β

together with variable amounts of TCRβ and CD4, but low amounts of the T cell activation markers

CD44, PD1 and ICOS. Group B samples expressed high amounts of TCRb and CD4 (with one

exception) and little CD8β but expressed high amounts of T cell activation markers (Figure 3.4-8

b; 3.4-9 a). Manual analysis confirmed that Group A samples had significantly higher abundance

of TCRβ+ CD8β+ and TCRβ- CD8β+ cells than Group B samples (Figure 3.4-8 a,b; 3.4-9 a,b).

CD4 was also variably expressed by CD8β+ cells in some Group A samples (Figure 3.4-8 b; 3.4-

9 a), a feature of normal immature CD4-/+ CD8+ thymocytes that proliferate extensively before

differentiating into mature CD4+ and CD8+ T cells (Guidos et al. 1989, Boudil et al. 2015). Both

CD8β+ subsets were proliferating in Group A samples, as indicated by their incorporation of high

amounts of IdU, a thymidine analogue (Figure 3.4-9 c). Thus, Group A lymphomas originate

from T cell precursors, similar to human T cell lymphoblastic leukemias (also known as thymic

lymphomas) that commonly arise during in mice and humans with mutations in the Ataxia

telangiectasia mutated (Atm) gene or others that regulate DNA repair and DNA damage

checkpoints (Figure 3.4-10 a) (Matei et al. 2006).

By contrast, Group B lymphomas had significantly more mature TCRβ+ cells expressing CD4 but

not CD8β than Group A lymphomas and WT SPL (Figure 3.4-8 a,b; Figure 3.4-9 a,b). TCRβ+

cells in Group B also included significantly more CD4+ cells that co-expressed CD44, ICOS or

PD1 than Group A samples or WT SPL, suggesting that they were activated (Figure 3.4-10 b,c).

TCRβ+ CD4+ CD8β- cells that expressed PD1 and ICOS were proliferating in Group B but not

Group A or WT SPL (Figure 3.4-9 c; Figure 3.4-10 d). Finally, Group B samples exhibited both

inter- and intratumoral heterogenity with respect to activation markers (Figure 3.4-11 a). In WT

mice, these markers are expressed by antigen-activatated CD4+ T-follicular helper (TFH) cells that

proliferate and differentiate in lymph nodes and spleen during immune responses to foreign

antigens. These data suggest that Group B lymphomas arise in peripheral lymphoid tissues from

activated TFH-like CD4 T cells.

In mice and humans, normal TFH cells promote proliferation and immunoglobulin (Ig) class

switching of activated B cells in germinal centers (Crotty 2019). Group B lymphomas had more

class-switched IgD- IgMhi B cells expressing variable amounts of the GC B cell markers CD150,

CD95 and Bcl6 than WT spleen (Figure 3.4-11 b,c), suggesting that the TFH-like lymphoma cells

95

acted cell non-autonomously to promote B cell proliferation and Ig class-switching in Group B

tumors, as is seen in cases of human TFH-like lymphoma (Vallois et al. 2016).

In Group B cases where multiple tumor-infiltrated tissues were examined (n=3), both spleen and

thymus were infiltrated with TFH-like cells and exhibited similar intratumoral heterogeneity

(Figure 3.4-11 d) suggesting the same tumor migrated to different tissues. Given that Group B

tumors likely originated from peripheral CD4 T cells, it is possible the lymphoma cells recirculated

to the thymus as is known to occur for effector/memory T cells in normal mice.

Collectively, these data suggest that lymphomas in Pole mutant mice have distinct cells types and

tissues of origin: T cell precursors in the thymus (Group A) or TFH-like T cells in peripheral

lymphoid tissues (Group B).

96

97

Figure 3.4-8 Determination of two types of T cell lymphoma in Pole mutant mice and their cell of origins

(A) Representative t-SNE plots of FlowSOM clusters present in wild-type (WT) SPL (top) compared to representative

Group A (middle; 3031) and Group B (bottom; 3158) lymphomas. Maps were colored in the Z dimension by

FlowSOM cluster number (left) or the indicated lineage markers. (B) Representative 10% probability contour plots of

TCRβ versus B220 or CD8b on live single cells (left 2 columns) or CD4 versus CD8β expression by and TCRβ+ (3rd

column) and TCRβ- CD11b- B220- cells (last column) WT SPL (top row) Group A (middle row; 3031), and Group

B (bottom row; 3158) samples. Quadrant gates (blue) show percentage of cells in each population. (C) Scatter plots

show the frequency of B cells (CD19+ CD22+) and myeloid cells (CD11b+) among total live cells from each group.

(D) Representative immunohistochemistry (IHC) staining with anti-CD3 (top) and anti-B220 (bottom) on WT SPL

(left), Group A mediastinal mass (middle), and Group B enlarged SPL and mediastinal mass (right). scale bar = 50μm;

inlet scale bar = 500μm.

Figure 3.4-9 Identification of two distinct types of T cell lymphoma in Pole mutant mice

(A) Heatmap of marker expression (columns) by CD11b- B220- cells in each sample (rows). Marker intensity was

normalized to the transformed ratio of medians by the column’s minimum. Samples were manually grouped as control

WT SPL (n=2), Group A lymphomas with high CD8b and variable TCRb and CD4 expression (n=9), and Group B

lymphomas with high TCRb and CD4 and little CD8b expression (n=7) subsets. Sample ID to the right indicates Pole

genotype (S459F/S459F = SF/SF; S459F/+ = SF/+) and tissue: thymus (T), spleen (S) and lymph node (L). (B) Scatter

plots show the % of each subset (y axis), identified by manual gating, of the indicated subsets among live single cells:

TCRβ+ CD8β- (left), TCRβ+ CD8β+ (middle) and TCRβ- CD8β+ (right) in WT SPL (n=5, circles) versus Group A

(n=9, squares) and Group B (n=7, triangles) lymphomas. Means are identified with a dashed horizontal line on each

bar and whiskers show the SD. These graphs include all the lymphoma samples shown on the heatmap in part B. (C)

Scatter plots show the percentage of IdU+ CD8b+ and IdU+ CD8b- cells within the TCRb+ subset (left and middle)

compared to the % IdU+ CD8b+ cells in the TCRb- B220- CD11b- subset (right) defined as shown in Figure 3.4-8 b.

Data are shown for the same samples shown in part B. Differences between groups were evaluated by ordinary one-

way ANOVA, using a false discovery framework (FDR) of 5% to yield q values. Significant differences are noted by:

***, q<0.001; **, q<0.002, *, q<0.033.

98

Figure 3.4-10 Characterization of T cell malignancies in PoleS459F/S459F and PoleS459F/+ mice

(A) Contour plots show TCRβ versus the indicated markers on total live cells on 2 thymic lymphomas from Atm-/-

mice stained with the same mass cytometry panel as the Pole lymphomas. (B) Contour plots show CD4 versus the

indicated markers on TCRβ+ cells from the samples shown in Figure 3.4-8 b. (C) Scatter plots configured as described

in (C) show the relative abundance of CD4+ cells expressing CD44, PD1 or ICOS cells among TCRβ+ cells from the

groups shown in Figure 3.4-10 b. Differences between groups were evaluated by ordinary one-way ANOVA, using

a false discovery framework (FDR) of 5% to yield q values. Significant differences are noted by: ***, q<0.001; **,

q<0.002, *, q<0.033. (D) Contour plots show IdU versus the indicated markers on TCRβ+ cells from the samples

shown in Figure 3.4-8 b.

99

100

Figure 3.4-11 Extrinsic effect of ‘Group B’ malignant T cells on B cell population

(A) Representative 10% probability contour plots of CD4 (y-axis) vs the indicated T cell activation markers (x-axis)

expressed by TCRβ+ cells from WT spleen (top row) versus three Group B samples (18, 3124, and 3158). Quadrant

gates (blue) show percentage of cells in each population. (B) Heatmap of Germinal Center markers (columns) among

B cells present in the listed samples (rows). Marker intensity is shown as the raw median value for each sample.

Samples are manually grouped as control WT (n= 4) and Group B (n= 7) samples. Sample ID to the right indicates

genotype and tissue code, wherein PP=Peyer’s patch, S=spleen, and L=lymph node. (C) Contour plots of IgD vs

CD95, CD150, BCL6, and IdU expression on CD22+ (B cells) between WT Peyer’s patch and indicated Group B

samples. Quadrant gates (blue) show percentage of IgD- GC-like B cells amongst total B cell population. (D)

Probability contour plots (10%) of CD4 vs PD-1 (left), or ICOS (right) expression by TCRβ+ cells from thymus (top)

versus spleen (bottom) isolated from three different mice with Group B lymphomas (Top, 18; middle, 3124; Bottom,

4219). Quadrant gates show percentages of CD4-expressing PD-1+ and ICOS+ T cells in blue.

3.4.4 Immune checkpoint blockade (ICB) increases proliferation in Pole mutant T cell lymphomas

Hypermutation is linked with positive responses to ICB in some solid tumors (Rooney et al. 2015,

McGranahan et al. 2016) including RRD cancers (Le et al. 2015, Bouffet et al. 2016). Furthermore,

using immunotherapy as a preventive approach is currently discussed in the context of germline

RRD carriers. We sought to determine whether prophylactic treatment with anti-PD1 and/or anti-

CTLA-4 would delay lymphoma incidence and increase survival in Pole mutant mice. We focused

this trial (and CyTOF analysis) on PoleS459F/S459F mice because they develop lymphoma with the

shortest and least variable latency (Figure 3.4-3). We treated cohorts of 6-week old mutant mice

twice weekly with either anti-PD1, anti-CTLA-4 or both antibodies and monitored them for tumor

initiation and survival. Surprisingly, despite the extreme hypermutation, treatment with either or

both checkpoint inhibitors did not significantly alter survival (Figure 3.4-12 a).

To search for a plausible explanation, we examined tumor infiltrated SPL cells from each treatment

group using the same mass cytometry panel we used to characterize lymphoma in untreated mice.

Heatmap visualization of T cell marker expression by cells lacking B and myeloid cell markers

showed that the most abundant cells in 15/16 treated mice were TCRb+ CD8b – cells (Figure 3.4-

12 b) that resembled Group B lymphomas in the untreated cohort. Although more TCRb+ CD8b-

lymphoma cells lacked CD4 in the treated (5/15) cohort versus untreated (1/7) Group B

lymphomas, they all expressed high levels CD44, PD1 and ICOS (Figure 3.4-12 b). Thus, most

lymphomas in the treated cohorts were Group B tumors. We therefore asked if any of the

101

treatments altered the abundance or proliferation of activated T cells using untreated Group B

samples as the comparison group. We observed no significant differences in the proportion of

TCRb+ cells or of CD4+ PD1+ and CD4+ ICOS+ cells within the TCRb+ subset in Group B tumors

from treated vs untreated mice (Figure 3.4-12 c top). These data suggest that the treatments had

no impact on the abundance or phenotype of T lymphoma cells. However, the proportion of

proliferating IdU+ T cells was significantly higher in mice treated with both anti-CTLA-4 and anti-

PD1 (Figure 3.4-12 c bottom). This finding suggests that the combination ICB therapy failed

because PD1 and CTLA-4 co-operatively restrain the proliferation of TFH-like Group B

lymphoma cells that express high levels of PD1 and ICOS. These findings are consistent with

previous reports on that PD1 has a tumor suppressor function in lymphoma models (Wartewig et

al. 2017) as well as with human data from our consortium showing that human T cell lymphomas

fail to respond to ICB.

102

Figure 3.4-12 Impact of prophylactic immune checkpoint blockade on Pole driven lymphomagenesis

(A) Kaplan-Meier survival estimates from control, anti-PD1, anti-CTLA-4, and combination treated PoleS459F/S459F

mice. Treatments were delivered intraperitoneally (IP) twice weekly beginning at 6 weeks of age until mice were

endpoint. Tumor-free, and overall survival findings are identical. One month = 4.3 weeks. (p = 0.82; Log-rank test).

(B) Heatmap of marker expression (columns) by CD11b- B220- cells in each sample (rows) from WT SPL or each

treatment group at end-point configured as described for Figure 3.4-9 a. Data for the untreated mice (n=7) was

replotted from Figure 3.4-9. (C) Top: Scatter plots (configured as described in Figure 3.4-9 show the percentage of

TCRβ+ cells among total live cells (left) and the %CD4+ PD1+ (middle) or %CD4+ ICOS+ cells among TCRβ+ cells in

each treatment groups. There was a single Group A tumor in the Combo treatment group in which most proliferating

cells were TCRb- CDb+ PD1- (not shown), but we chose not to exclude this sample to keep the analysis unbiased.

Bottom: Scatter plots show relative the %IdU+ cells in the TCRβ+ subset from each treatment group. Anti-CTLA-4

(n=4), anti-PD1 (n-5), Combo (n=7). Significant differences in abundances were tested and displayed as in Figure

3.4-9 except that each treatment group was compared only to the untreated group.

103

3.5 Discussion

In this report, we utilize novel mouse models to comprehensively study the penetrance and

tumorigenic processes of POLE mutants, enabling characterization of genotype/phenotype

differences observed among humans with germline POLE mutations. We find that mouse tumors

driven by Pole mutations model human POLE mutant tumors genetically and have unique

mechanisms of tumor progression, intra-tumoral heterogeneity, and response to ICB.

Although initially thought to be driven by MMRD alone, mutations in DNA polymerases are a

major cause of RRD and hypermutation in cancer (Briggs et al. 2013, Campbell et al. 2017).

Determining whether a POLE mutation can drive tumorigenesis is difficult since mutations must

retain polymerase function of the protein while causing dysfunction of its proofreading ability.

This large gene has been shown to harbor many mutations both in the germline and somatically in

hypermutant tumors. Thus, defining true drivers is very difficult and is key for genetic counselling

and initiation of surveillance protocols. While initial reports on POLE driven mutagenesis in

human cancers described a mild phenotype of adult-onset polyposis, data from our international

consortium and case reports reveal that some mutations result in a more aggressive genetic

syndrome (Figure 3.4-3 b). Importantly, some of the most common somatic mutations have not

been reported to date in the germline. The aggressive cancer phenotype exhibited in our mouse

models provides further clarity to this genotype-phenotype observation as mutations, such as

P286R and S459F, may be incompatible with human development.

In humans, POLE mutations are most commonly drivers in GI and endometrial cancers (Briggs et

al. 2013, Church et al. 2013). This is thus a limitation of studying germline PoleP286R and PoleS459F

mice which develop mainly lymphomas. Nevertheless, these mutant mice are robust models to

provide insight on POLE tumorigenesis as tumors recapitulate both the mutational burden and

processes observed in POLE driven human tumors. Tumors from both species are driven mainly

by SNVs, are evenly dispersed throughout the genome, and exhibit a similar TMB accumulation

ceiling. Moreover, both models exhibited specific mutations and mutational signatures consistent

with POLE mutagenesis—C>A-TCT, C>T-TCG (COSMIC Signature 10) (Alexandrov et al.

2013); T>G-TTT (COSMIC Signature 28) (Alexandrov et al. 2018, Petljak et al. 2019). These

signatures were not clearly observed in previous Pole-mutant models, notably in genome

104

sequencing from LSL-PoleP286R mouse embryonic fibroblasts and tumors (Li et al. 2018). This may

reflect differences in sequencing facility pipelines or gene dosing since LSL-PoleP286R mice are

hemizygous for Pole. Differences in gene dosing and genetic backgrounds may also explain

differences in tumor spectra observed.

Using several algorithms, we provide further insight on the pattern and process of Pol ε driven

tumor progression. Firstly, our data suggest obligatory continuous mutation accumulation and lack

of clonal stability. The constructed clonal evolutionary tree (Figure 3.4-6 c) reveals a continuous

steady accumulation of mutations. This is very different than the classic clonal evolution pattern

where an aggressive clone is dominant and stable, such as in melanoma (Birkeland et al. 2018).

Finally, it is interesting to note that both polymerase mutant and MMRD tumors can gain

secondary mutations to become combined RRD in humans and mice (Figure 3.4-6 c) (Shlien et

al. 2015).

Human RRD hematopoietic malignancies are predominantly of T cell origin (Wimmer et al. 2014),

but T cell malignancies arise from many different stages of T cell differentiation. Differences in

cell of origin underly important differences in prognosis and outcome for many types of cancer

(Visvader 2011). In mice and humans, T cell lymphoblastic lymphomas arise in the thymus from

TCRβ-/+ CD4-/+ CD8β+ precursors that undergo cell cycle arrest and recombine TCRA prior to

completing maturation (Guidos et al. 1989, Boudil et al. 2015). Group A lymphomas in PoleS459F

mutant mice were arrested during this developmental transition, which is known to be highly prone

to oncogenic transformation from studies of mice and humans with genetic defects in other DNA

damage/repair checkpoints (Danska et al. 1994). Indeed, both LSL-PoleP286R mutant mice (Li et al.

2018) as well as Pold1 proofreading deficient mice (Goldsby et al. 2002) also appeared to develop

such lymphomas.

Outcomes for T cell lymphoblastic lymphomas are better than for mature T cell lymphomas, a

heterogenous group of hematopoietic malignancies that develop when mutations accumulate

during antigen-driven T cell proliferation in peripheral lymphoid tissues. This group includes

angioimmunoblastic T cell lymphoma (AITL), a nodal T cell lymphoma that originates from TFH

cells that harbor mutations in genes that regulate TCR signaling and epigenetic processes, occurs

in older adults and has non-autonomous effects on B cells (Vallois et al. 2016). Group B

105

lymphomas may provide a mouse model for this a rare poor prognosis type of T cell lymphoma.

Collectively, our finding that PoleS459F mutant mice develop two distinct types of T cell lymphoma

suggests that there are multiple stages during Pole-driven lymphomagenesis in which a T cell can

gain advantage by hypermutation.

Furthermore, our detailed mass cytometry analyses reveal that, in contrast to most lymphoid

malignancies, Pole driven lymphomas exhibit substantial phenotypic intratumoral heterogeneity

(Figure 3.4-8; Figure 3.4-12 a,d). This phenotypic heterogeneity may be explained by the

genomic intra-tumoral heterogeneity but also suggest that a more comprehensive analysis of

human lymphomas which are driven by RRD is required to better tackle these aggressive cancers.

Treatment of PoleS459F/S459F mice prophylactically with ICB, revealed splenomegaly in all mice

with clinical signs of lymphoma. Since our earlier analysis indicated that lymphoma infiltration to

various organs in a single mouse was always from the same cells phenotypically (Figure 3.4-12

d) and genomically (Figure 3.4-6 a; Figure 3.4-7), we therefore focused our analysis of tumor-

infiltrated spleen cells from all treatment groups. Upon analysis by CyTOF, we observed that 15/16

treated tumors were Group B with 1/16 classifying as Group A (Figure 3.4-12 b). This enrichment

is important since PD1 was not highly expressed by mouse Group A lymphomas and this can

explain the survival benefit for Group B tumors and their lack of response to (and potentially

exacerbation by) ICB. In hypermutant solid tumors ICB is used to counteract the PD1 immune

inhibitory checkpoint restraints, unleashing intra-tumoral CD8 T cells to kill tumor cells that

express mutationally generated “neoantigens” (Sharma et al. 2015, McGranahan et al. 2016).

However, prophylactic treatment of Pole mutant mice with anti-PD1 alone or in combination with

anti-CTLA-4 did not improve survival or decrease the abundance of TFH-like CD4 cells.

While this lack of therapeutic ICB effect could reflect “immuno-editing” or other upregulation of

other inhibitory checkpoints (Rooney et al. 2015, Sharma et al. 2017, Zappasodi et al. 2018), it

more likely reflects the unique biology of the Group B lymphomas, which had few CD8 T cells

and consisted predominantly of TFH-like CD4 cells expressing high levels of PD1, ICOS and

CD44. PD1 (Shi et al. 2018, Crotty 2019) and CTLA-4 (Sage et al. 2014) can both restrain ICOS-

induced co-stimulation of normal TFH cells. Thus, combined interference with the PD1 and

CTLA-4 inhibitory checkpoints on malignant Group B T cell lymphomas may allow ICOS-

106

mediated co-stimulation to promote their growth. In accordance with this idea, proliferation of the

TFH-like lymphoma was increased by combination ICB therapy (Figure 3.4-12 d). This finding

accords with a recent study showing that anti-CTLA-4 expands ICOS+ TH1-like CD4 T cells in

murine tumor models (Wei et al. 2017). Interestingly, mono- and bi-allelic deletions of PDCD1,

the gene encoding PD1, are frequent in some human T cell lymphoma types (Wartewig et al. 2017),

suggesting a tumor suppressor function in these cases. While these data may not be generalizable

to human solid tumors driven by POLE mutations (indeed ICB has been shown to effectively treat

hypermutant solid tumors (Le et al. 2015, Bouffet et al. 2016), this tumor suppressor function of

PD1 may explain data from our international consortium suggest that anti-PD1 therapy will not

prevent or delay occurrences of T cell malignancies in RRD mutant patients. Taken together, our

findings suggest that the use of ICB therapies warrants urgent further investigation for PD1+ T cell

lymphomas or RRD hematological malignancies.

In summary, our mouse models provide a platform to study hypermutant lymphomas. They

provide valuable information about specific mutations and their potential phenotype in humans.

Establishing the kinetics and clonal evolution of RRD cancers, which cannot be done by studying

yeast or human tumors, may be used as an Achilles’ heel for these cancers. Finally, this highly

penetrant model of RRD tumors may be the most reliable approach for future robust preclinical

testing of immune-based drugs on spontaneous tumors rather than established cell lines on

immunocompromised mice. Specifically, these mice offer an opportunity to study combined

models of RRD as they can be combined with existing transgenic MMRD mouse models

(Kucherlapati et al. 2010). This avenue of study is now possible, especially given the propensity

of our heterozygous Pole mice to acquire tumors, which more closely resembles the clinical

landscape of polymerase germline and somatic mutations in cancer.

107

Chapter 4 Generation of combined replication repair deficient (RRD) mouse

models for the study of RRD cancers

108

Generation of combined replication repair deficient (RRD) mouse models for the study of RRD cancers

4.1 Introduction

Two-thirds of mutations across human cancers result from errors in DNA replication (Tomasetti

et al. 2017). The two mechanisms of repair responsible for correcting mistakes made by DNA

polymerases during replication include the intrinsic exonuclease proofreading capabilities of Pols

ε and δ and post-replicative MMR. Ablation of either of these mechanisms through germline

inheritance or somatically acquired mutations leads to replication repair deficiency (RRD),

resulting in hypermutation, microsatellite instability (MSI), and increased likelihood of cancer

initiation (Cortez 2019).

Germline inheritance of either polymerase proofreading deficiency (PPD) or a monoallelic

pathogenic mutation in one of the four MMR genes (MSH2, MSH6, MLH1, and PMS2) typically

leads to adult onset of gastrointestinal and genitourinary cancers (Lynch et al. 2003, Palles et al.

2013). However, inheritance of biallelic pathogenic mutations in one of the MMR genes result in

constitutional mismatch repair deficiency syndrome (CMMRD), a childhood onset, extremely

aggressive cancer predisposition syndrome. CMMRD is typically characterized by CNS,

gastrointestinal, and hematological cancers, although patients can develop many other types of

cancers, as each cell in their body lacks MMR (Tabori et al. 2017, Bush et al. 2019). Further, brain

tumors from patients with CMMRD were found to have somatically acquired secondary mutations

in the proofreading domains of genes coding for the Pol ε and δ catalytic subunits (POLE and

POLD1) (Shlien et al. 2015). Indeed, it is now clear that inactivation of both replication repair

mechanisms (resulting in complete RRD) can occur across many cancer types including

endometrial, prostate, and gastrointestinal cancers. Moreover, while the temporal order of genetic

events in CMMRD tumors with complete RRD clearly proceeds from primary MMRD to

secondary PPD, the reverse order of primary PPD with secondary MMRD is also common

(Campbell et al. 2017, León-Castillo et al. 2020). Combined RRD tumors display

“ultrahypermutation” with mutational loads between 100-1000 mutations/Mb. Due to this

hypermutant phenotype, these tumors respond well to immune checkpoint inhibition (ICI) (Bouffet

et al. 2016). However, not all tumors respond to ICI treatment and a more comprehensive approach

109

including the development of in vivo models to study combined RRD cancers are needed to explore

tumor biology and determine why therapeutic response is variable.

While evidence suggests a synergy exists between MMRD and PPD leading to some of the highest

mutational loads observed in human cancers (and one of the strongest mutator phenotypes), others

have shown that these high mutation rates decrease the fitness of an organism (Herr et al. 2014).

Indeed, mice with combined germline MMRD and PPD demonstrated embryonic lethality,

suggesting an incompatibility with mammalian life (Albertson et al. 2009).

Thus, while most studies in mice exploring RRD have focused on the ablation of either MMRD or

PPD, there is a need to model and study combinatorial RRD, particularly for those solid tumors

arising in cases of CMMRD, including gastrointestinal and brain cancers.

We hypothesized that combined RRD would result in rapid time to tumorigenesis in vivo for

cancers commonly found in children with RRD (brain and gastrointestinal) and could be used in

the study of such cancers. Here we characterize the generation of three novel models of combined

RRD cancers using the germline PoleS459F mouse model described and characterized above in

Chapter 3 in combination with an existing Msh2LoxP mouse model (used also by Li and colleagues

in the modeling of combined RRD endometrial carcinoma), as well as several different Cre-

recombinase driver mouse strains. One brain tumor model and one gastrointestinal cancer model

were developed and evaluated as potential platforms to study the development of complete RRD

tumors and use for preclinical therapeutic testing.

110

4.2 Materials & Methods

4.2.1 Study Approval

All mouse experiments from this study were reviewed and approved by The Centre for

Phenogenomics (TCP) Animal Care Committee (Toronto, ON, Canada). Experiments were

performed in compliance with the Animals for Research Act of Ontario and the Guidelines of the

Canadian Council on Animal Care.

4.2.2 Mouse Models & Husbandry

All mice from this study were housed at TCP in a pathogen-free environment and fed ad libitum

on standard chow. PoleS459F mice were generated in house and genotyped as described above in

Chapter 3.3.1. Both Villin-Cre (El Marjou et al. 2004) and Msh2LoxP (Kucherlapati et al. 2010)

mice were obtained from W. Edelmann (Albert Einstein College of Medicine, New York, NY,

USA). Nestin-Cre mice (Tronche et al. 1999) (JAX stock #003771) were obtained from M. Taylor

(The Hospital for Sick Children, Toronto, ON, Canada). Olig2-Cre mice (Zawadzka et al. 2010)

(JAX stock #025567) were obtained from C. Hawkins (The Hospital for Sick Children, Toronto,

ON, Canada). Msh2LoxP mice were genotyped as per Kucherlapati et al using PCR primers: (5’-

TACTGATGCGGGTTGAAGG-3’ and 5’-AACCAGAGCCTCAACTAGC-3’). Knockout of

exon 12 was detected through genotyping specific tissues using the PCR primers: (5’-

TACTGATGCGGGTTGAAGG-3’, 5’-TGTGCTGGCTCACTTAGACG-3’ and 5’-

GGCAAACTCCTCAAATCACG-3’). All Cre mouse strains were genotyped using universal Cre

PCR primers (Mouse Clinical Instiute): control band (5’-

ACTGGGATCTTCGAACTCTTTGGAC-3’ and 5’-GATGTTGGGGCACTGCTCATTCACC-

3’), Cre band (5’-CCATCTGCCACCAGCCAG-3’ and 5’-TCGCCATCTTCCAGCAGG-3’).

Both male and female experimental animals were studied from each model.

4.2.3 Tissue Processing & Immunohistochemistry

Complete necropsies were performed as described in Chapter 3.3. Tissue samples were obtained

and either snap frozen or fixed in 10% phosphate buffered formalin, embedded in paraffin,

sectioned, stained with hematoxylin and eosin, and examined by light microscopy. Tumors were

111

confirmed histologically either by I. Siddiqui (small and large bowel tumors) or C. Hawkins (brain

neoplasms).

IHC was performed on unstained 5 μm slides from formalin-fixed paraffin embedded blocks.

Sections were deparaffinized, rehydrated, and subjected to heat-mediated epitope retrieval using

0.01M Citrate Buffer pH 6.0. Sections were washed 3 x 5 min in PBS-T and endogenous

peroxidases were blocked using BloXall (Vector Labs, SP- 6000-100) for 10 min at RT. Sections

were then washed 2 x 2 min with PBS-T, blocked with 2.5% Normal Horse Serum for 20 min at

RT, and incubated with either anti-GFAP (DAKO, GA524), anti-synaptophysin (Abcam,

ab32127), or anti-Olig2 (Millipore Sigma, AB9610) for 1.5 hr at RT. Sections were then washed

and incubated with ImmPRESS Universal Antibody (anti-mouse IgG/anti-rabbit IgG, peroxidase)

Polymer Reagent (Vector Labs, MP-7800) for 30 min at RT. Sections were subsequently washed

and then treated with DAB (Vector Laboratories SK-4100) for visualization followed by double

distilled water for to stop the reaction. Sections were counterstained with Hematoxylin,

dehydrated, mounted with a coverslip using VectaMount Permanent Mounting Media (Vector

Laboratories, H-5000).

4.2.4 Whole Exome Sequencing

WES, variant calling, and signatures analyses were conducted as outlined in Chapters 2 (2.3.2)

and 3 (3.3.2) for human and mouse cancers, respectively. Sequencing data will be made available

upon publication.

4.2.5 Organoid Establishment

Organoids were established from 12-week-old VC+/Msh2L/+, VC+/Msh2LoxP/+, and

VC+/Msh2L/LPoleSF/+ mice. Mice were euthanized and small intestines from each mouse were

harvested, flushed, placed into a dish with cold PBSO (Ca-Mg free phosphate buffered saline), and

washed several times. Intestines were cut open down the length of the organ and a coverslip was

used to scrap off villi, leaving crypts attached. Villi were washed off and intestines were then cut

into small 2-4mm pieces and transferred into a 50 mL falcon tube. Intestinal pieces were washed

with 10 ml PBSO, supernatant was removed, and this was repeated until the supernatant was clear.

Intestinal pieces were then incubated in 25 mL 2 mM EDTA in cold PBSO for 30 min at 4°C.

After incubation intestines were allowed to settle to the bottom of the tube and supernatant was

112

removed and discarded. 10 mL of PBSO with 10% FBS was added, and the intestines were pipetted

3-5 times to release the crypts. The supernatant containing the crypts was removed and passed

through a 70 μM strainer to remove debris and isolate crypts. This was repeated 2-3 times to obtain

multiple crypt fractions which were inspected using an inverted microscope to determine which

fractions contained the most crypts and least debris. Select fractions were combined and

centrifuged at 600 rpm at 4°C for 5 min, washed and resuspended in PBSO with 10% FBS,

centrifuged as previously, and resuspended in DMEM/F-12. Subsequent steps followed

STEMCELL technologies IntestiCultTM organoid protocol for mouse (STEMCELL Technologies,

Catalog #06005). Crypt numbers were estimated by counting the number of crypts in a 10 μL

aliquot from each sample using an inverted microscope and crypts were subsequently aliquoted

and adjusted such that the number of crypts was equal for each sample. Samples were centrifuged

as above and resuspended in 150 μL of prepared IntestiCultTM Organoid Growth Medium at room

temperature. An equal amount of thawed Matrigel® (Corning, Cat #356231) was added to each

crypt sample and mixed thoroughly. Each sample crypt preparation was pipetted into 4 wells of a

pre-warmed 24-well plate at 50 μL crypt mixture/well and carefully placed in a 37°C incubator for

10 min until the Matrigel® domes solidified. Finally, 750 μL IntestiCultTM Organoid Growth

Medium was carefully added to dome-containing well. Organoids were incubated at 37°C and 5%

CO2 and observed over 72hr.

4.2.6 Statistical Analysis

Statistical analyses were performed using the GraphPad Prismv9 software. The log-rank (Mantel-

Cox) test was used to analyze survival differences, unless otherwise indicated. P values <0.05 were

determined to be significant. Asterisks denote P values as follows: *, P < 0.05; **, P < 0.01; ***,

P < 0.001, **** P<0.0001.

113

4.3 Results

4.3.1 Intestinal specific ablation of MMR and Pole proofreading leads to tract-wide dysplasia and greatly reduced survival in mice

Mice harboring Villin-Cre, the conditional allele Msh2LoxP, and PoleS459F were interbred to generate

Villin-Cre+/Msh2LoxP/LoxP (abbreviated VC+Msh2L/L), Villin-Cre+/Msh2LoxP/LoxP/PoleS459F/+

(abbreviated VC+Msh2L/LPoleSF/+), and Villin-Cre+/Msh2LoxP/LoxP/PoleS459F/S459F (abbreviated mice

VC+Msh2L/LPoleSF/SF) mice. Villin-Cre is a transgenic mouse line which is expressed exclusively

in the intestinal epithelium by embryonic day (E) 12.5 (El Marjou et al. 2004). VC+Msh2L/L were

previously characterized and found to survive on average 12 months with all mice dying by 17

months. Moreover, these mice developed primarily highly invasive adenocarcinomas of the small

intestine (Kucherlapati et al. 2010). Like VC+Msh2L/L, both VC+Msh2L/LPoleSF/+ and

VC+Msh2L/LPoleSF/SF mice were viable and fertile, however, a significant difference in survival

was observed between all lines (Figure 4.3-1 a), with survival significantly decreasing with the

addition of each subsequent PoleS459F allele suggesting a synergy between MMRD and PPD.

At endpoint, VC+Msh2L/LPoleSF/+ mice displayed weight loss, pale toes, and extremely hunched

posture. Histological analysis of the gastrointestinal tracts revealed a mix of low and high-grade

dysplasia, luminal necrosis, and adenocarcinoma in situ sporadically across small and large bowels

(Figure 4.3-1 b). This was unlike previously characterized VC+Msh2L/L mice that tended to exhibit

a single focal disseminated mass at endpoint. VC+Msh2L/LPoleSF/SF mice demonstrated an even

more severe phenotype and in many instances prior to endpoint, diarrhea was observed in the cages

indicating difficulty conserving water. Necropsy findings demonstrated loose intestinal tracts that

had noticeable polyps throughout intestines. It is possible that tract-wide inactivation of both MMR

and PPD disrupts normal intestinal function, affecting animal welfare, and leading to endpoint

before invasive disease can develop.

To further study the effects of RRD on intestinal crypt function, we attempted to establish

gastrointestinal organoids from mouse colonic crypts. We harvested crypts from the small

intestines of 12-week-old VC+Msh2L/+ (Lynch), VC+Msh2L/L (CMMRD), and VC+Msh2L/LPoleSF/+

(RRD) mice, plated crypts in Matrigel domes, and observed their growth and ability to form

organoids over 72 hours (Figure 4.3-1 c). Interestingly, after 72 hrs, organoids from VC+Msh2L/+

114

mouse crypts appeared to begin to develop characteristic branched growth structures, while those

derived from VC+Msh2L/L mouse crypts exhibited more heterogeneity in their growth kinetics with

a mixture of branched structures as well as more cystic, spheroid-like structures. This was

consistent with findings from Keysselt and colleagues, who also observed heterogeneity in growth

kinetics among VC+Msh2L/L derived organoids indicative of early abnormalities likely resulting

from the presence of genetic instability (Keysselt et al. 2017). Strikingly, crypts derived from

VC+Msh2L/LPoleSF/+ GI tracts failed to establish intestinal organoids in culture (Figure 4.3-1 c),

with few crypts developing into early-stage luminal organoid structures. Collectively, histology,

survival, and in vitro findings suggest that pervasive, combined gastrointestinal-specific RRD

results in a level of instability perhaps not sustainable for normal intestinal epithelial function.

115

Figure 4.3-1 GI specific ablation of RRD leads to decreased survival, high grade dysplasia, and reduced GI

stem-cell capabilities in vitro

(A) Kaplan-Meier tumor-free survival estimates for VC+Msh2L/L (n=31), VC+Msh2L/LPoleSF/+ (n=48), and

VC+Msh2L/LPoleSF/SF mice (n=5). P value was determined by Log-rank test. (B) Representative HE stained slides from

swiss rolled VC+Msh2L/LPoleSF/+ large and small bowels demonstrating luminal necrosis, high grade dysplasia, and

adenocarcinoma in situ. Blue scale bars indicate 1000 μM and 200 μM for upper and lower panels, respectively. (C)

Representative light microscope images of 72hr timepoint GI organoid experiment using crypts harvested from 12-

Small Bowel Large Bowel

A

B

C

0 5 10 150

50

100

Age (months)S

urv

ival (%

)

Log Rankp < 0.0001

VC+/MSH2

L/L/Pole

SF/SF

VC+/MSH2

L/L/Pole

SF/+

VC+/MSH2

L/L

116

week-old VC+Msh2L/+, VC+Msh2L/L, and VC+Msh2L/LPoleSF/+ mice. Experiment performed twice with four wells

plated/genotype.

4.3.2 Central nervous system wide ablation of MMR and Pole proofreading results in rapid malignant brain tumorigenesis in the posterior fossa

Next, we wanted to explore the interaction of MMRD and PPD in the brain, given the prevalence

of combined RRD in brain tumors from children with either CMMRD or germline heterozygous

PPD. Mice harboring Nestin-Cre, the conditional allele Msh2LoxP, and PoleS459F were interbred to

generate Nestin-Cre+/Msh2LoxP/LoxP (abbreviated NC+Msh2L/L), Nestin-

Cre+/Msh2LoxP/LoxP/PoleS459F/+ (abbreviated NC+Msh2L/LPoleSF/+), and Nestin-

Cre+/Msh2LoxP/LoxP/PoleS459F/S459F (abbreviated mice NC+Msh2L/LPoleSF/SF) mice. Nestin-Cre is a

transgenic mouse line which is expressed throughout the CNS by E11.5 (Tronche et al. 1999).

While specific ablation of MMR in the murine intestinal tract and endometrium have previously

been studied, brain specific MMRD has not yet been examined in a murine model.

Both NC+Msh2L/L and NC+Msh2L/LPoleSF/+ mice were viable and fertile, however due to severe

early-onset neurological phenotypes observed in NC+Msh2L/LPoleSF/SF mice, these animals were

not tested for breeding capabilities. Indeed, striking differences in survival and phenotype were

observed between all genotypes (Figure 4.3-2 a), supporting the synergy observed in models with

intestinal specific ablation of replication repair. Unlike VC+Msh2L/L animals, NC+Msh2L/L mice

never developed cancer and demonstrated lifespans comparable to control animals lacking the

Nestin-Cre allele, suggesting that MMRD alone is insufficient to drive brain tumorigenesis in vivo.

However, the addition of a single PoleS459F allele resulted in strikingly reduced survival with

NC+Msh2L/LPoleSF/+ mice exhibiting an average lifespan of only 11.6 weeks and no mice surviving

beyond 15.1 weeks (Figure 4.3-2 a left). Necropsy and histology findings from euthanized

NC+Msh2L/LPoleSF/+ mice showed that all animals developed tumors in the posterior fossa with

100% penetrance (Figure 4.3-2 a right).

Although these cancers developed in the posterior fossa (Figure 4.3-2 b), IHC for specific markers

including Gfap, Olig2, and synaptophysin, revealed the tumors to be negative for both Gfap and

117

synaptophysin, but partially positive for Olig2 (Figure 4.3-2 b), suggesting these tumors may be

glioma-like rather than typical medulloblastomas. This is curiously similar to human RRD brain

tumors which resemble HGGs (Shlien et al. 2015) but may cluster distinctly based on methylation

analysis and can arise in the posterior fossa (Dodgshun et al. 2020).

As in the Villin-Cre driven model, NC+Msh2L/LPoleSF/SF mice exhibited neurological deficits

(ataxia, abnormal gait, and frequent seizures) even within the first few weeks of life. Indeed, by

the time mice were weaned at 3 weeks, NC+Msh2L/LPoleSF/SF mice weighed significantly less than

their NC+Msh2L/LPoleSF/+ littermates (Figure 4.3-2 c). Necropsy findings from age matched

animals also revealed that the brains of NC+Msh2L/LPoleSF/SF mice were smaller than those of their

NC+Msh2L/LPoleSF/+ or WT littermates. Interestingly, histological examination of brains from

NC+Msh2L/LPoleSF/SF mice that were euthanized due to severe seizures did not reveal any signs of

brain tumorigenesis, rather cerebella from these mice appeared hypoproliferative with lack of

normal observed architecture, suggesting improper migration of cells during development (Figure

4.3-2 d). However, if NC+Msh2L/LPoleSF/SF mice were managed such that they were handled less

frequently, and provided supportive high calorie chow, we could extend their lifespans such that

they could reach adulthood just as their NC+Msh2L/LPoleSF/+ littermates. In these cases (n = 17),

NC+Msh2L/LPoleSF/SF mice lived, on average, longer than NC+Msh2L/LPoleSF/+ mice with a mean

survival of 14.4 weeks with a maximum survival of 20.6 weeks (Figure 4.3-2 e). These animals

developed mainly brain tumors (n = 13) but also died of lymphomas (n = 2) or both tumor types

(n = 2) due to their germline homozygous polymerase mutations. Similar to data from intestinal

specific ablation of RRD, these data suggest that brain specific genetic instability conferred by

homozygous MMRD and PPD results in decreased cellular fitness and disruption of normal

neuronal function, resulting in paradoxical delayed tumor onset.

118

NC

+M

sh

2L/L

/Po

leS

F/S

FW

T

GFAP OLIG2 Synaptophysin

0 5 10 15 20 250

50

100

Age (months)

Su

rviv

al

(%)

Log Rank

p < 0.0001

NC+/MSH2LoxP/LoxP/PoleS459F/+

NC+/MSH2LoxP/LoxP

NC+/MSH2LoxP/LoxP/PoleS459F/S459F

NC+Msh2L/L/PoleSF/SF NC+Msh2L/L WT

Msh2Hom

/PoleHet

Msh2Hom

/PoleHom

0

5

10

15

Mass at 3 weeks old

Genotype

Ma

ss

(g

)

****

A

B

C

HE

D

E

0 2 4 60

50

100

Age (months)

Su

rviv

al

(%)

NC+/Msh2

L/L/Pole

SF/+

NC+/Msh2

L/L/Pole

SF/SF

(neuro endpoints)

NC+/Msh2

L/L/Pole

SF/SF

(tumor endpoints)

✱✱✱✱

✱✱

✱✱✱✱

119

Figure 4.3-2 CNS-wide RRD leads to robust brain tumorigenesis in vivo and paradoxical genotype-phenotype

correlations

(A) Kaplan-Meier survival estimates for NC+Msh2L/L (n=15), NC+Msh2L/LPoleSF/+ (n=21), and NC+Msh2L/LPoleSF/SF

mice (n=7). P value was determined by Log-rank test (left). Representative image of typical posterior fossa brain

tumor found in NC+Msh2L/LPoleSF/+ mice (right). (B) Representative histology from NC+Msh2L/LPoleSF/+ mouse brain

tumor. From left to right: HE and IHC stains for Gfap, Olig2, and Synaptophysin. Blue scale bars indicate 50 μM and

1000 μM for high magnification and inlet, respectively. (C) Average mass of NC+Msh2L/LPoleSF/+ (n=12) and

NC+Msh2L/LPoleSF/SF (n=6) mice at wean. P value calculated by unpaired t test with Welch’s correction. (D)

Representative brains dissected from NC+Msh2L/LPoleSF/SF, NC+Msh2L/L, and WT mice demonstrating the smaller size

of NC+Msh2L/LPoleSF/SF mouse brains (upper). HE images from age matched WT and NC+Msh2L/LPoleSF/SF mouse

brains demonstrating differences in cerebellar structure, suggesting abnormal migration of cells, resulting in mouse

ataxia and seizures. (E) Kaplan-Meier survival estimates for NC+Msh2L/LPoleSF/+ (n=21), and NC+Msh2L/LPoleSF/SF

mice sacrificed due to severe neurological symptoms (n=7) or NC+Msh2L/LPoleSF/SF mice minimally handled and

sacrificed due to brain tumor related endpoints (n=17). P values calculated using Wilcoxon test.

4.3.3 The genomic landscape of mouse cancers with complete RRD mirrors the human syndrome

Given the unique hypermutant status of RRD tumors and the role this type of genomic instability

plays in such tumors’ susceptibility or resistance to therapies, it is critical that the genomic

landscape of tumors from RRD mice resemble human combined RRD cancers. We performed

WES on brain tumors from NC+Msh2L/LPoleSF/+ and intestinal polyps from VC+Msh2L/LPoleSF/+

mice and assessed their mutational burdens in comparison with cancers from mice with either PPD

or MMRD alone (data used as reported in Chapter 3). Good concordance was observed between

mutational loads observed from mouse cancers with human MMRD, PPD, and RRD cancers

(Figure 4.3-3). However, while all combined RRD cancers displayed hallmark hypermutation,

mutational loads observed from polyps derived from VC+Msh2L/LPoleSF/+ mice (mean = 57

mut/Mb) were less than those observed in RRD brain tumors from NC+Msh2L/LPoleSF/+ mice

(mean = 357 mut/Mb) or Pole mutant driven lymphomas (mean = 170 mut/Mb). Likely, this is a

result of lack of advanced disease in these animals.

Finally, when mutational signatures were assessed and compared between mouse and human

tumors, again mouse cancers recapitulated mutational signatures in human cancers (Figure 4.3-

3). As discussed in Chapter 3, Pole mutant mouse lymphomas displayed signature 10, the hallmark

signature associated with Pol ε mutations in cancer. MMRD mutant cancers from Mlh1-/- mice

exhibited MMRD related signatures 6, 15, and 20. Finally, combined RRD tumors exhibited

120

signature 14, characteristic of human tumors with both MMRD and PPD. Taken together,

mutational burden and signatures data from RRD animals provide support for the use of such

models in the study and preclinical modeling of RRD cancers.

Figure 4.3-3 Cross species comparison of genomic tumor features reveals correlation between mouse models

and human cancers

(A) Tumor mutational burdens given in total SNVs/Mb for human: h-RRD (n=40); h-POLE (n=4); h-MMRD (n=30);

h-NonRRD (n=7) and mouse cancers: m-Br-RRD (n=7); m-GI-RRD (n=3); m-POLE (n=29); m-NonRRD (n=7). (B)

COSMIC mutational signatures called using the decontructsigs algorithm for both human and mouse cancers.

Representative signature contributions are shown for each species and replication repair status. RRD = complete

replication repair deficiency; MMRD = mismatch repair deficient; POLE = polymerase epsilon proofreading deficient;

Non-RRD = no replication repair deficiency; Br = brain tumor from NC+Msh2L/LPoleSF/+ mice; GI = gastrointestinal

adenoma from VC+Msh2L/LPoleSF/+ mice.

121

4.4 Discussion

This chapter summarized the development of several novel models of combined MMRD and

polymerase proofreading loss with the aim of creating new tools to study and treat RRD cancers.

To our knowledge, these are the first animal models of combined RRD brain and GI cancers

reported. During the preparation of this thesis, one recent study by Li and colleagues utilized LSL-

PoleP286R mice (Pole mutant conditional knock-in; discussed in Chapter 3) combined with

Msh2LoxP mice to study the in vivo characteristics of endometrial cancers driven by ablation of

either or both repair mechanisms (Li et al. 2020). Results from this study, as with ours,

demonstrated a synergy between PPD and MMRD in endometrial carcinogenesis, with

PoleP286R/+Msh2-/- mice exhibiting decreased survival compared to PPD alone or MMRD-only

animals which never developed endometrial cancer. These endometrial cancers displayed hallmark

ultrahypermutation, and tumors from both PPD and combined RRD mice displayed sensitivity to

treatment with combinatorial αPD1/αCTLA4. While this study provides additional preclinical

support for the use of ICIs in the treatment of endometrial cancers with PPD or combined RRD,

questions remain as to whether the same principles hold true for RRD childhood tumors. Such

questions may be addressed with our models, although some considerations for future studies

should be noted.

Firstly, while NC+Msh2L/LPoleSF/+ mice developed MBTs with striking penetrance and are

therefore useful in the study of disease progression, VC+Msh2L/LPoleSF/+ and VC+Msh2L/LPoleSF/SF

mice only exhibited diffuse disease with lack of focal mass with dissemination (Figure 4.3-1).

Although the time to tumor onset and death was relatively rapid in VC/Msh2/Pole mice, the lack

of advanced disease precludes the use of these animals in the study of high grade RRD intestinal

cancers. These intestinal models do, however, give some insight on human RRD cancers as the

prevalence of intestinal tumors with complete RRD in humans is relatively low, suggesting that

complete RRD may affect the fitness of intestinal epithelia. This was also consistent with findings

from NC+Msh2L/LPoleSF/SF mice, which exhibited severe neurological abnormalities, but delayed

tumor onset compared to their NC+Msh2L/LPoleSF/+ littermates (Figure 4.3-2). Collectively, the

impacts of excessive mutational burden on normal tissue function as observed in our models are

consistent with previous work in mice which demonstrated that combined germline RRD results

in embryonic lethality (Albertson et al. 2009) as well as work in diploid yeast showing decreased

122

cellular fitness (Herr et al. 2014). Moreover, human tumors with homozygous POLE exonuclease

mutations have never been discovered, suggesting the accumulation of mutations may be too rapid

and stochastic for both normal and tumor cells.

Future studies may seek to limit inactivation of both MMR and polymerase proofreading to

specific areas of the intestinal tract through delivery of Adenoviral-Cre (Kucherlapati et al. 2013),

mosaic inactivation through the use of Lgr5-EGFP-IRES-creERT2 mice (Barker et al. 2007), and the use

of inducible polymerase mutant models such as LSL-PoleP286R mice (Li et al. 2020). Collectively,

these alterations in model design will likely ameliorate observed tract-wide disease prevalence and

allow for more advanced disease to occur.

The NC+Msh2L/LPoleSF/+ tumors mimic the human tumors in several avenues. First, these are all

rapidly growing tumors. Indeed, using our surveillance protocol, we discovered that imaging must

occur at least every 6 months to avoid large brain tumors even when the previous scan is normal.

Second, these tumors manifest the complex definition of RRD brain tumors as gliomas or tumors

of embryonal origin. Our model results primarily in posterior fossa tumors. While tumors

histologically mimicked medulloblastomas or small cell glioblastomas, most of these brain tumors

are largely synaptophysin negative with strong Olig2 staining, suggesting a “glioma-like”

phenotype. Furthermore, we are currently developing similar models using different drivers such

as Olig2-Cre resulting in tumors in various locations and histology, which may more accurately

recapitulate the diversity of tumor types seen in CMMRD patients. Moreover, while polymerase

mutations are common drivers, they are not universally present across cancers. Studies focused on

identifying mouse and human RRD tumor cell of origins and specific additional driver events may

help in our understanding of the different phenotypes observed in mouse and human RRD brain

tumors (see section 6.3 for further discussion on future directions).

Finally, combined RRD MBTs from NC+Msh2L/LPoleSF/+ mice genomically mimic the human

disease, exhibiting strikingly high mutational burdens and relevant mutational signatures (Figure

4.3-3). This is the first known animal model to recapitulate childhood RRD MBTs and provides a

platform for further studies on such cancers which is explored further in the fourth and final study

of this thesis.

123

Chapter 5

Immune surveillance affects the mutational landscape of RRD brain tumors

124

Immune surveillance affects the mutational landscape of RRD brain tumors

5.1 Introduction

RRD cancers are universally hypermutant due to the continuous acquisition of somatic mutations

(Campbell et al. 2017). In adults, hypermutant cancers such as melanoma and lung cancers have

been shown to respond well to ICI therapy targeting PD-1 signalling (Hodi et al. 2010, Garon et

al. 2015). Although ICI is less efficacious in the treatment of childhood cancers, RRD childhood

cancers prove to be an exception to this rule, likely due to their hypermutation status (Bouffet et

al. 2016).

Various biomarkers including TMB, MSI, and PD-L1 expression have been studied as

mechanisms to predict response to ICI (Geoerger et al. 2020, Le et al. 2020, Marabelle et al. 2020).

Although ICIs have led to encouraging responses in patients with universally lethal cancers

resulting from germline RRD, not all cancers respond to ICI, and questions remain regarding

predictors of response to therapy.

Firstly, among children with germline RRD, non-CNS solid tumors have been shown to respond

better to ICI treatment than CNS tumors (Figure 5.1-1 a). While this may be due to the immune

privileged nature of the brain compared to extracranial sites, this hypothesis has not been formally

tested although more CD8+ T cells are observed within RRD tumors outside the CNS compared to

those within the CNS (Figure 5.1-1 b).

Secondly, RRD childhood cancers that respond best to therapy are those which exhibit

ultrahypermutation (100-1000 mut/Mb) compared to those which exhibit hypermutation (10 – 100

mut/Mb; Figure 5.1-1 c). Presumably, this is due to a higher likelihood of obtaining immunogenic

mutations which are those expressed as neoantigens (Figure 5.1-1 d) and targeted by CD8 T cells,

however this has not been explicitly tested in vivo. Further, RRD tumors that relapse do not appear

to continue to accumulate mutations despite underlying repair deficiencies that should continue to

cause mutation accumulation (Figure 5.1-1 e). It has been hypothesized that such a ceiling in

mutational burden is due to cell autonomous mechanisms whereby the number of mutations is too

great for cancer cell function (Shlien et al. 2015), however, whether these maximums are indeed

125

caused by cell autonomous or cell non autonomous mechanisms, such as immunoediting through

immune surveillance is unclear.

To address these questions related to the immune surveillance of RRD brain tumors, we utilized

the NC+Msh2L/LPoleSF/+ mouse model (discussed in Chapter 4), as this model displayed the most

penetrant phenotype with a rapid time to brain tumor related endpoint (~11 weeks). The final study

of this thesis explores the interaction of hypermutant RRD brain tumors with the immune system

and demonstrates the importance of anatomical tumor location on perceived mutational

maximums, survival, and immunoediting.

126

Figure 5.1-1 Biomarkers predicting response to anti-PD-1 in RRD human cancers

(A) Progression-free survival for non-CNS solid tumors (n = 11) vs. CNS tumors (n = 27). (B) Responder to anti-PD-

1 blockade had higher levels of CD8 T-cell infiltration as compared to non-responders. Green=gastrointestinal (GI);

blue=central nervous system (CNS) tumors. (C) Response to anti-PD1 and overall survival by non-synonymous

variants for all solid RRD tumors. (D) Response to anti-PD1 and overall survival by total predicted neoantigens for

RRD brain tumors. (E) Primary vs relapsed mutational loads in combined RRD CNS tumors (n=6). Data from

Morgenstern, Sudhaman, Das et al. (unpublished) (A-D) and the IRRDC (E).

127

5.2 Materials & Methods

5.2.1 Mouse Models & Husbandry

Study approvals were obtained as per section 4.2.1 above. All mice from this study were housed

at TCP in a pathogen-free environment and fed ad libitum on standard chow. NOD scid gamma

(Jax strain #5557) and C57BL/6J were used and obtained from TCP in house colony.

5.2.2 Orthotopic and subcutaneous serial tumor transplantation

Primary brain tumors from endpoint NC+Msh2L/LPoleSF/+ mice were harvested and mechanically

dissociated to achieve a single cell suspension. The cells were allocated for injection either

subcutaneously in NSG or C57BL/6J animals or intracranially in NSG or C57BL/6J animals. For

subcutaneously injected tumors 200,000 cells were resuspended in a 1:1 mixture of PBS and

Matrigel® (Corning, Cat #356231) and injected subcutaneously in the flanks of 6 to 8 weeks old

female mice and allowed to grow until tumors reach 1700mm3. Mice were the euthanized and

tumors were harvest using sterile scissors and forceps, minced as previously into single cell

suspensions, and reinjected as described previously for subcutaneous tumors. For intracranially

implanted tumors, 50,000 cells were implanted by stereotactic injection into the posterior fossas

of 6 to 8 weeks old female mice using the following stereotactic coordinates: 2mm posterior to

lambda, 2mm lateral and 2mm deep. Tumors were allowed to develop until neurological endpoint.

Mice were euthanized, and tumors were harvest, dissociated as above, and reinjected intracranially

until 3 serial passages were achieved.

5.2.3 Tissue Processing & Immunofluorescence

Complete necropsies were performed as described in Chapter 3.3. Tissue samples were obtained

and dissociated into single cells or fixed in 10% phosphate buffered formalin, embedded in

paraffin, sectioned, stained with hematoxylin and eosin, and examined by light microscopy.

Immunofluorescence was performed on unstained 5 μm slides from formalin-fixed paraffin

embedded blocks. Sections were deparaffinized, rehydrated, and subjected to heat-mediated

epitope retrieval using 0.01M Citrate Buffer pH 6.0. Slides were washed with ddH2O for 1 min

and then blocked with Blocking Solution (2% bovine serum albumin; 0.2% fish gelatin; 1:50

donkey serum in PBS) at 37°C for 30 mins. Slides were then incubated with mouse anti-CD3

128

(DAKO, Clone F7.2.38) and rabbit anti-CD8 (Abcam, Clone EPR20305) for 1 hr at room

temperature. Slides were washed 3x with Blocking Solution and incubated with conjugated

secondary antibodies donkey anti-rabbit IgG (Alexa Fluor 647; abcam, Cat# ab150075) and

donkey anti-mouse IgG (Cy5; Jackson ImmunoResearch Laboratories, Cat# 715-175-151). Slides

were washed 3x with PBS and then nuclei were stained with DAPI (1:1000 in PBS) at room

temperature for 5 mins. Slides were washed again and coverslipped using slowfade (SlowFade

Diamond Antifade, S36972; Fisher Scientific). Slides were scanned and images taken using

PanoramicViewer (v.1.15.4).

5.2.4 Whole Exome Sequencing

WES and variant calling analyses were conducted as outlined in section 3.3.2 for mouse cancers.

Sequencing data will be made available upon publication. Overlapping mutations were identified

using bcftools v.1.6. and plotted using R v.4.0.3 and the package UpSetR.

5.2.5 Statistical Analysis

Statistical analyses were performed using the GraphPad Prismv9 software. The log-rank (Mantel-

Cox) test was used to analyze survival differences, unless otherwise indicated. P values <0.05 were

determined to be significant. Asterisks denote P values as follows: *, P < 0.05; **, P < 0.01; ***,

P < 0.001, **** P<0.0001.

5.3 Results

5.3.1 Anatomical location of transplanted RRD brain tumors influences growth kinetics and overall survival in vivo

To study the impact of a competent immune system on the growth of RRD brain tumors, we first

transplanted several RRD MBTs derived from endpoint NC+Msh2L/LPoleSF/+ mice, subcutaneously

in flanks of both immunocompromised (NSG) and immunocompetent syngeneic (C57BL/6J)

animals. Tumors consistently demonstrated significantly slower growth in syngeneic, immune

competent animals compared to immunocompromised animals (Figure 5.3-1 a), and therefore

significantly longer survival (Figure 5.3-1 b-top), consistent with previous work that demonstrated

129

delayed growth of mouse MMRD cancer cell lines in vivo (Germano et al. 2017). Based on

observations from human RRD cancers suggesting greater immune surveillance in non-CNS

cancers compared to CNS cancers, we hypothesized that the differences observed upon

subcutaneous transplantation of RRD murine MBTs would not be observed when tumors were

transplanted orthotopically (intracranially). Indeed, no differences in survival existed between

intracranially injected immunocompromised and immunocompetent mice (Figure 5.3-1 b-

bottom), and these tumors displayed similar size and histology at endpoint (Figure 5.3-1 c),

suggesting less immune surveillance in the brain compared to extracranial sites where tumor

growth and survival differences were significant.

Since cytotoxic CD8+ T cells are the major immune cell type mediating immune surveillance of

tumors, we performed immunofluorescence to examine the extent of CD8+ T cell infiltration in the

brain compared to flank. In humans, greater T cell infiltration is observed in extracranial tumors

compared to CNS cancers. In transplanted mice, both flank and intracranially implanted tumors

appeared to have infiltration of CD8+ T cells (Figure 5.3-1 d). Ongoing experiments utilizing flow

cytometry aim to quantify both the number of infiltrating T cells as well as assess their

characteristics to determine whether they are activated or anergic. Taken together, these findings

suggest that immune infiltration and surveillance occur in the intracranial and subcutaneous

setting, but that these the extent of surveillance may differ between sites.

130

Figure 5.3-1 Anatomical location of transplanted RRD brain tumors influences growth kinetics and overall

survival in vivo

(A) Tumor growth kinetics of primary RRD tumors transplanted in NSG vs C57BL/6J mice (n=5 mice/group; p

<0.0001 at day 30). (B) Kaplan-Meier survival estimates for primary RRD tumors transplanted in NSG and C57BL/6J

mice (n=20 mice/group) for flank (top; p <0.0001) and intracranially (bottom; p=ns) transplanted tumors. (C)

Representative H&E stains from intracranially implanted RRD brain tumors in both NSG (top) and C57BL/6J

(bottom) animals. Black scale bar indicates 1000 μm; magnified inlet white scale bar indicates 20 μm. Tumor area is

indicated. (D) Immunofluorescence stains for CD3 and CD8 from representative flank and intracranially implanted

tumors in C57BL/6J mice. Orange = CD3; red=CD8; blue=DAPI. White scale bar indicates 50 μm.

131

5.3.2 Serial transplantation of RRD MBTs shed light on the impact of immune surveillance on the genomic landscape of tumors

Cell autonomous mechanisms have previously been thought to control the number of mutations

allowed to persist in tumor cells and define a “ceiling” (upper limit), compatible with cancer cell

survival. However, immune surveillance, and by extension immunoediting of tumor neoantigens

may additionally contribute to perceived mutational maximums. To determine the extent to which

immune surveillance influences a tumor’s genomic landscape within both intracranial and

extracranial contexts, we serially transplanted primary NC+Msh2L/LPoleSF/+ tumors over three

generations in immunocompromised and immunocompetent syngeneic animals both intracranially

and subcutaneously (Figure 5.3-2).

Animals transplanted within each generation of serial passaging were necropsied when they

reached endpoint, tumors were harvested, processed, and dissociated, or fixed for FFPE. DNA was

extracted from the primary tumor and n = 3 tumors from each passage, which underwent WES.

The number of non-synonymous SNVs and indels were quantified for each tumor as these

mutations are most likely to generate neoantigens and be targeted by the immune system. We first

examined the number of non-synonymous mutations and indels conserved between the primary

tumor and the first passage in each transplant environment. Interestingly, the fraction of mutations

conserved from the primary tumor was highest in immunocompromised animals regardless of

anatomical location transplanted and significantly lower in subcutaneously transplanted syngeneic

animals (Figure 5.3-3). Those tumors transplanted intracranially in syngeneic animals exhibited

significantly more conserved mutations than those transplanted subcutaneously and appeared to

conserve less mutations than those transplanted in immunocompromised animals, although this

was not statistically significant (Figure 5.3-3).

We then looked at the mutational loads of tumors over time. Strikingly, we observed that those

tumors serially transplanted subcutaneously in immunocompetent animals progressively decreased

in mutational loads over subsequent passages (Figure 5.3-4). This was not true of tumors serially

passaged intracranially in immunocompetent animals nor of tumors passaged in either anatomical

location in immunocompromised animals (Figure 5.3-4). Indeed, tumors passaged intracranially

in immunocompromised animals increased in mutational burden with each successive passage.

Interestingly, while tumors transplanted subcutaneously in immunocompromised animals did

132

steadily increase in non-synonymous SNVs over subsequent passages, the tumors appeared to

reach a plateau, not surpassing 6000 total non-synonymous mutations per exome, unlike those

tumors that were also unexposed to immune surveillance but passaged intracranially. Moreover,

tumors passaged intracranially in immunocompetent animals initially increased in non-

synonymous SNVs but then decreased by the third passage, again appearing to reach a similar

plateau to those passaged subcutaneously in immunocompromised animals. While the plateau in

mutations observed in immunocompromised subcutaneously passaged tumors was unexpected, the

striking differences between tumors passaged intracranially vs subcutaneously in syngeneic

animals suggests that, indeed, less immune surveillance occurs in the brain than outside of the

brain.

To more closely examine changes in non-synonymous mutations over time, we quantified the

number of mutations conserved between the primary tumor and specific tumors which were

serially passaged. Few mutations were conserved from the primary tumor across all three

subcutaneously passaged tumors in syngeneic immunocompetent animals (2.1% of non-

synonymous SNVs conserved). This was unlike tumors passaged in immunocompromised

animals, regardless of anatomical location of transplant, where greater than a third of mutations

were conserved across passages (39.6% for subcutaneously passaged tumors; 33.5% for

intracranially passaged tumors) (Figure 5.3-4). Interestingly, tumors passaged intracranially in

syngeneic animals maintained an intermediate number of mutations (between those passaged in

immunocompromised animals and those passaged subcutaneously in syngeneic animals) across

serial passages (15.5% of non-synonymous mutations).

Furthermore, after the initial loss of most non-synonymous mutations, tumors serially passaged

subcutaneously in syngeneic mice seemed to remain static in the non-synonymous mutations

accumulated in the first passage (Figure 5.3-4). This suggests that new mutations that would have

accumulated during the cell divisions of subsequent passages were potentially targeted and

removed via immunoediting. This was again, unlike the evolution of mutations in tumors serially

passaged in the other three environments where mutation accumulation and loss were much more

dynamic.

133

Collectively, these genomic data demonstrate how anatomical location of a tumor can influence

the extent to which a tumor undergoes immune surveillance. Using both syngeneic and

immunocompromised animals we showed how the initial striking loss and subsequent static

landscape of mutations in tumors transplanted extracranially was not mimicked intracranially.

Together, tumor growth kinetics, survival data, immunofluorescence imaging, and genomic data

suggest that the brain is unique in its decreased surveillance of malignant tumors.

134

Figure 5.3-2 Schematic of serial transplantation experiment

Freshly dissociated primary RRD brain tumors were implanted subcutaneously and intracranially in both C57BL/6J

and NSG animals (n=5 mice per generation). One tumor from each serial transplant generation was dissociated and

transplanted into another five animals.

Figure 5.3-3 Conservation of mutations between primary and transplanted RRD brain tumor

Transplanted tumors from subcutaneously or orthotopically transplanted NSG and C57BL/6J animals underwent WES

(n=3 tumors per group). Variant prediction for both nonsynonymous and indel mutations was conducted and the

fraction of mutations conserved between primary tumor and first tumor transplant passage was determined.

Significance were determined by two-way ANOVA and Tukey's multiple comparisons test.

135

136

Figure 5.3-4 Differences in mutation accumulation and changes to mutational landscape between mechanisms

of serial transplantation of RRD brain tumors

Changes in TMB and specific mutations are shown for primary RRD brain tumors transplanted via C57BL/6J flank

(top right), C57BL/6J intracranial (top left), NSG flank (bottom right), and NSG intracranial (bottom left). Bar plots

demonstrate the TMB for primary tumor (cyan; same tumor in all bar plots) and three generations of serially passaged

tumors (n=3 tumors/passage). Venn diagrams depict the overlap in mutations across serial passages of specifically

those tumors which were passaged to the subsequent generation. Tumors are displayed proportional to their number

of nonsynonymous mutations. Primary tumor is the same size across all four Venn diagrams. UpSet plots report

quantification of each Venn diagrams’ fractions of private and shared nonsynonymous mutations.

5.4 Discussion

Despite the promise of ICI therapy for MMRD and hypermutant cancers, only a subset of patients

display durable, long term response to therapy (Topalian et al. 2019, Le et al. 2020). In particular,

trials assessing ICI for childhood solid tumors (Davis et al. 2020), and adult HGGs (Reardon et al.

2020), have not found PD-1 blockade effective in the treatment of these cancers. While childhood

brain tumors driven by RRD may prove an exception to this rule (Bouffet et al. 2016), preliminary

data from our consortium demonstrates that despite their high mutational burden, childhood RRD

brain tumors respond less well to PD-1 blockade than extracranial solid tumors (Figure 5.1-1).

The reasons for this are not clear. Here we demonstrate that survival differences between

immunocompromised and syngeneic immunocompetent animals transplanted with RRD MBTs

subcutaneously are not seen when tumors are transplanted intracranially (Figure 5.3-1), suggesting

that the anatomical location of these tumors preclude widespread immune surveillance.

While endpoints symptoms for tumors transplanted subcutaneously differ from intracranial tumor

endpoints (size of flank tumor vs welfare-related neurological symptoms, respectively),

intracranially implanted tumors appeared the same size upon necropsy and histological

examination regardless of the immune status of the animal transplanted (Figure 5.3-1 c). This

suggests that tumors grew similarly in both immunocompetent and immunocompromised animals,

although MRI or bioluminescence imaging will be useful to better quantify tumor growth kinetics

for intracranially implanted tumors in both the immunocompetent and immunocompromised

animals. While immunofluorescence data demonstrates immune infiltration occurs in both the

intracranial and flank setting, further experiments explicitly quantifying and characterizing

immune infiltrates are ongoing.

137

Although mutations are presumed to continually accumulate through each cell division due to

underlying repair deficiencies in this unique group of childhood tumors, we do not observe

increased mutational burdens from primary to relapsed RRD brain tumors. Indeed, often the

mutational burden of relapsed tumors decreases (Figure 5.1-1). We had originally postulated this

was due to cell autonomous influences, which limited the mutational loads to preserve cancer cell

function (Shlien et al. 2015). However, genomic data in this study suggests that at least a

combination of cell autonomous and cell non-autonomous factors may be responsible for

controlling the mutational burdens of RRD hypermutant cancers over time. We observed that

subcutaneously transplanted tumors in syngeneic immunocompetent animals displayed a gradual

decrease in non-synonymous mutations over serial passages and a striking loss of most mutations

from the primary tumor. This was not observed when tumors were transplanted intracranially nor

in immunocompromised animals regardless of anatomical location (Figure 5.3-3 and Figure 5.3-

4). Moreover, tumors transplanted intracranially in syngeneic animals did appear to reach a

mutational ceiling, unlike those transplanted in immunocompromised animals (Figure 5.3-4),

suggesting that immune surveillance may still contribute to these perceived mutational maximums

by targeting immunogenic mutations.

Further ongoing experiments include:

1. Better characterization of infiltrating CD8+ T cells in these tumors using flow cytometry to

identify other immune infiltrates important for tumor surveillance, and determine whether

the immune infiltrates present are sufficient to stimulate an anti-tumor response within the

brain upon ICI or combinatorial therapy administration.

2. Moreover, since data from human tumors suggest that mutational burden predicts response

to ICI (Figure 5.1-1), we directly test this observation by rechallenging syngeneic

immunocompetent mice with serially transplanted tumors with the highest and lowest

mutational burdens to determine whether there are significant differences in tumor growth

kinetics, survival, and response to ICI in vivo.

138

Chapter 6 General Discussion

General Discussion

6.1 Summary & Significance of Findings

Preclinical models of cancer are necessary to study the de novo development, evolution, and

treatment of disease. At the onset of this thesis, few preclinical models of RRD were available

outside of established human and mouse cell lines or animal models that did not accurately

recapitulate human RRD syndromes (Goldsby et al. 2002, Albertson et al. 2009, Lee et al. 2016).

Collectively, the results outlined in this thesis demonstrate the utility of various preclinical RRD

models for validation of therapeutic targets and offer new in vivo platforms that will aid in future

RRD studies.

Childhood RRD syndromes are relatively rare and, therefore, understudied. Prior to work

conducted in this thesis, many of the insights into the molecular basis for these diseases and their

therapeutic resistances and vulnerabilities came from clinical observations (Vasen et al. 2014,

Wimmer et al. 2014, Bodo et al. 2015, Durno et al. 2015), genomic studies made feasible by

international consortia efforts (Shlien et al. 2015, Campbell et al. 2017), or studies in adult RRD

cancers (Le et al. 2015, Le et al. 2017). The second chapter of this thesis utilized a rational,

translational approach to detect and validate therapeutic targets for RRD cancers. Utilizing a multi-

omics approach, we identified the RAS/MAPK pathway as frequently altered across hypermutant

pediatric cancers that were enriched for RRD and demonstrated that this pathway was universally

upregulated in pediatric RRD cancers (Chapter 2). As a complement to this in silico approach,

we utilized traditional in vitro and in vivo models including established and patient-derived cell

lines as well as PDX models to validate the susceptibility of RRD cells to MEK inhibition. These

studies culminated in the treatment of two CMMRD patients with HGGs using MEK inhibitors.

Both patients demonstrated striking responses to therapy despite repeated relapse to previous

treatments. While targeted therapies for RRD cancers are unlikely to provide clinical benefit due

to the tumors’ ability to acquire new mutations and mechanisms of resistance, approaches such as

139

ours that capitalize on the tumors’ oncogenic addiction to specific pathways will enable the

treatment of these hypermutant cancers despite the difficulty associated with an overabundance of

tumor specific variants. Recently uncovered synthetic lethal targets unique to MMRD/MSI-high

cancers, including WRN inhibition and neddylation inhibition, are also promising therapeutic

vulnerabilities (Chan et al. 2019, Lieb et al. 2019, McGrail et al. 2020). It is unclear, however,

whether these approaches would also significantly affect normal cells in CMMRD individuals, as

they are also MMRD (unlike LS), and therefore may be overly susceptible to cell death following

treatment with these inhibitors.

Finally, in one patient that received combinatorial trametinib and nivolumab, trametinib appeared

to synergize with nivolumab and improve the patient’s response to ICI. This was evidenced by the

spike in total and proliferating T lymphocytes detected in the patient’s blood, which corresponded

with the addition of trametinib to the treatment regime. These exciting findings suggest that in

addition to controlling the growth kinetics of RRD tumors directly, MEK inhibitors may allow for

antigen spreading within a tumor, thus enabling greater CD8+ T lymphocyte infiltration and attack.

These results have provided the foundational preclinical data to support an upcoming clinical trial

examining the efficacy of combinatorial ICI and MEK inhibition. Although promising, the

preclinical models used will not be useful for further exploration of mechanisms of synergy

between MEK inhibitors and ICI, which rely on the presence of a competent immune system.

In parallel to the first study reported in this thesis which utilized classical preclinical models of

RRD cancers, we worked to develop robust animal models that would better recapitulate the

human RRD syndromes. At the onset of this thesis, our group had discovered that CMMRD brain

tumors frequently harboured somatic mutations in the exonuclease domains of either POLE or

POLD1 (Shlien et al. 2015). Moreover, the significance of polymerase mutations in cancer as

germline and somatic driver events was just emerging (Palles et al. 2013). Yet, animal models

harboring human cancer-associated polymerase mutations did not exist. This meant that inferences

about the variability of polymerase exonuclease mutations in tumorigenesis were left to studies in

yeast (Barbari et al. 2018, Xing et al. 2019), artificial systems (Shinbrot et al. 2014), and clinical

mapping of disease and mutation co-occurrence across families.

140

The only in vivo models that had been generated were achieved by double alanine substitutions of

active site catalytic residues, which have never been observed in human cancer. Moreover, these

models did not recapitulate the human syndrome since heterozygous mutants (Poleexo/+) did not

experience reduced survival (Goldsby et al. 2002, Albertson et al. 2009). We reasoned that

assessment of polymerase cancer variants should be conducted directly in vivo to study their

evolution and characteristics and ultimately to aid in the future generation of combined RRD

models. We focused our efforts on POLE, since exonuclease domain mutations in POLE are more

common than those in POLD1 (Rayner et al. 2016). We therefore generated and characterized two

novel germline Pole exonuclease mutant models based on two commonly occurring mutations in

human cancer: S459F and P286R (Chapter 3). Strikingly, these mutants demonstrated distinct

and rapid time to tumor initiation, far faster than those exhibited in Poleexo mice. Our findings led

us to examine genotype/phenotype correlation more closely in reported cases of germline and

somatic POLE mutations in cancer (Figure 3.4-3 b). It is likely that both S459F and P286R will

not occur as germline mutations in PPAP due to their aggressive phenotypes in vivo. Finally, good

concordance was found between survival findings between our germline PoleP286R mice and LSL-

PoleP286R mice (Li et al. 2018), which were reported during preparation of this thesis. Both models,

which were established on different genetic backgrounds, reported embryonic lethality of

homozygous mutants, suggesting the mutational burden conferred by P286R homozygosity was

incompatible with life. PoleS459F/S459F mutant mice were viable and occurred in expected mendelian

ratios, albeit with a shorter lifespan than their heterozygous littermates. While these findings were

initially surprising since S459F appeared to confer a greater decrease in exonuclease function in

vitro (Figure 3.4-3 c), recent reports assessing the mutator phenotype rather than exonuclease

capability of POLE mutants demonstrated P286R actually caused the highest levels of

hypermutation (Barbari et al. 2018). This paradoxical finding was recently explained by the

observation that P286R confers hyperactivity on the protein’s polymerase function (Xing et al.

2019).

Although our polymerase mutant models develop primarily lymphomas and not tumors commonly

found in PPAP such as colon and endometrial cancers, we could still utilize them to examine the

evolution and characteristics of POLE driven cancers. Unsurprisingly, all tumors exhibited

ultramutator phenotypes and mutational signatures supporting polymerase mediated mutagenesis

141

as the underlying cause of cancer. We exploited the infiltration of lymphomas across multiple

organs to sequence them as multi-site cancers and reconstruct tumor evolutionary trees (Figure

3.4-7). Together with our CyTOF findings, these experiments revealed tumors were

phenotypically heterogenous and developed mutations continuously overtime, leading to

evolutionary trees with prominent “branches” unlike many other hypermutant cancers where

“truncal” mutations predominate due to the cell-extrinsic exposure to mutagens compared to cell-

intrinsic hypermutation process conferred by RRD (von Loga et al. 2020).

Moreover, these models enabled the testing of prophylactic ICI for RRD driven hematological

cancers. Given these animals developed primarily T-cell lymphomas, it was perhaps unsurprising

that our treatments failed to extend lifespan in PoleS459F/S459F mice. However, individuals with

CMMRD develop more T-cell lymphomas than B-cell lymphomas (Figure 1.3-1) (Wimmer et al.

2014), in contrast to non-RRD associated pediatric lymphomas which skew towards B-cell

malignancies. Therefore, our results underscore the need to approach childhood RRD driven

hematological malignancies with caution. Indeed, as previously mentioned, the hypermutation

phenotype can result in rapid resistance to targeted immune-based therapies as well (Oshrine et al.

2019). These data, together with the knowledge that synchronous cancers are not uncommon in

RRD syndromes, highlight the importance of finding novel therapies and strategies in the treatment

of such cancers.

This thesis reports, for the first time, combined models of RRD childhood tumors including GI

cancers and MBTs (Chapter 4). One motivation for the development of polymerase mutant mouse

models was to utilize such a model in the study of cancers driven by combined RRD. Given the

embryonic lethality of combined PPD and Mlh1-/- mice reported by Albertson et al., we reasoned

that a conditional model of MMRD would be necessary to achieve our aims (Kucherlapati et al.

2010). We focused our efforts on models of GI cancer and MBTs as these have been found to

harbour combined RRD (Campbell et al. 2017) and are common in CMMRD patients (Shlien et

al. 2015), respectively.

Strikingly, NC+Msh2L/LPoleSF/+ mice developed MBTs within the first 8-13 weeks of life, resulting

in a rapid, robust model that may be used in more detailed studies of disease progression.

Moreover, our WES analyses revealed that these tumors exhibited ultrahypermutation (>100

142

mut/Mb), with mutational loads expected of cancers driven by combined RRD. They additionally

displayed a high contribution of signature 14, which is associated with combined MMRD and

POLE exonuclease deficiency in human cancers (Alexandrov et al. 2020). These exciting findings

together with the tumor penetrance of the model makes it attractive for pursuing additional studies.

We additionally discovered that brain specific MMRD alone was insufficient to drive brain

tumorigenesis in vivo. Given that not all CMMRD brain tumors evolve somatic PPD, further study

into additional secondary drivers of these brain tumors is necessary (see section 6.2 below). One

additional limitation of the model is that it reflects an inversion of the genetic events occurring in

individuals with CMMRD. Most combined RRD brain tumors result from germline MMRD and

acquire secondary somatic PPD whereas our model relies on early inactivation of MMRD (E11.5)

on a PPD background. It is unlikely that this difference would significantly change downstream

genetic events leading to malignant transformation given that mutations should still occur

randomly, however, if we wish to utilize these models to simulate treatment of RRD cancers on

the background of MMRD, models that more accurately recapitulate the CMMRD syndrome will

be necessary (see 6.2).

The paradoxical finding that NC+Msh2L/LPoleSF/SF mice can live longer than NC+Msh2L/LPoleSF/+

mice, suggests that mutational loads may exhibit a “goldilocks effect”, whereby an overabundance

of mutations can increase the chance of acquiring mutations that confer selective growth advantage

for cells, but too many mutations may be detrimental for normal cellular processes and therefore

be less amenable to transformation. This was somewhat supported by the lack of advanced disease

in both VC+Msh2L/LPoleSF/+ and VC+Msh2L/LPoleSF/SF mice, and the inability of crypts derived

from VC+Msh2L/LPoleSF/+ small bowels to generate intestinal organoids.

Unfortunately, as previously discussed, GI specific ablation of both MMRD and PPD did not result

in advanced disease, and instead led to tract-wide diffuse disease resulting in early endpoint. Multi-

focal primary tumor burden is a known challenge with GEMMs since they require euthanasia

before macroscopic metastases (which accounts for most cancer-related deaths in humans) can

develop. Alternative model designs that stand a greater chance at ameliorating the widespread low-

grade disease were discussed previously (see section 4.4).

143

Finally, this thesis culminates in the application of our novel RRD brain tumor model to elucidate

the relationship between tumor mutational burden and immune surveillance. Biomarkers including

high mutational burdens, and CD8+ T cell infiltration are associated with better responses to ICI

therapy. Despite these predictive biomarkers, less than a third of patients respond long-term to

ICIs, the reasons for which are unclear. Brain tumors from CMMRD patients are known to exhibit

less immune infiltration than their GI cancers. Moreover, many RRD cancers appear to exhibit a

plateau in mutational loads, and do not surpass specific ceilings (Shlien et al. 2015). Data from our

international consortium indicates that relapsed combined RRD tumors do not continue to

accumulate mutations (indeed, many of them display decreased TMB at relapse) (Figure 5.1-1).

We had previously postulated this was due to cell autonomous mechanisms, whereby further

increases in mutational loads would not be compatible with cancer cell viability. An alternative

explanation is that an abundance of non-synonymous mutations within the tumor result in

neoepitopes, increasing immune surveillance and restraining the upward accumulation of

mutations. Few studies have assessed the accumulation of mutations in an immune competent

setting over time. Our mouse models provided a unique opportunity to assess the effects of immune

surveillance and anatomical location on the mutational landscape of RRD brain tumors. Our

preliminary genomic data suggests that the brain may be unique in that it may provide tumors with

greater immune escape than extracranial lesions. Moreover, tumors that could establish

subcutaneously in syngeneic animals had relatively low TMBs suggesting that tumors may avoid

additional immune surveillance by maintaining a lower TMB (and potentially actionable

neoantigens). Current experiments involve conducting neoantigen prediction using NetMHCpan

4.0 to verify that the observed changes and levels of non-synonymous SNVs are reflected in

quantity of predicted neoantigens. Further, we aim to better characterize and quantify the immune

infiltrate differences between brain and flank implanted tumors through flow cytometry to

determine whether greater infiltrates are present in the flank compared to the brain and whether

these cells are activated, for example by quantifying the number of CD69+ T-cells. Additionally,

we are currently rechallenging mice with tumors displaying the highest and lowest TMB. We

hypothesize that higher TMB tumors will result in greater immune surveillance, slower growth

kinetics, and improved response to ICI therapy. Finally, we are analysing data from our

international consortia on relapsed RRD tumors to compare with serial passaging data from mouse

experiments and see whether any commonalities exist in retained mutations.

144

The findings in this thesis have raised new questions and opportunities for the study of RRD

syndromes and cancers. Outside of the immediate studies being conducted in completion of

Chapter 5, I propose the following as future work that has already commenced or could be

undertaken.

6.2 Future Directions

6.2.1 In vitro small molecule screens & in vivo combinatorial therapies

As a collaborative effort in the Tabori lab, 20 cell lines were established from 20 independent

NC+Msh2L/LPoleSF/+ mouse brain tumors to assist with more rapid assessment of small molecule

library screens and in vivo combinatorial therapy efficacy. These cell lines have undergone WES

and genotyping to confirm they are indeed combined RRD and that they maintained hypermutation

and signature 14 after in vitro passaging (Figure 6.2-1). Moreover, they demonstrate good growth

kinetics in immunocompetent animals, and several have exhibited sensitivity to anti-PD-1

blockade (Figure 6.2-1). These cell lines provide an excellent platform to perform drug screens

for RRD brain tumors as well as test combinatorial therapies in vivo. This is important since, as

previously discussed, only a third of patients respond to ICI. Several in vivo combinations that

could be tested include anti-PD-1 with anti-CTLA-4, anti-PD-1 with MEK inhibition, anti-PD-1

with anti-LAG3, or radiation following initial treatment with anti-PD-1. All of these combinations

may help to induce antigen spreading and revive the immune response.

145

Figure 6.2-1 Established RRD brain tumor cell lines genomically resemble human and mouse tumors and can

be used for in vivo testing of therapies

RRD brain tumors were harvested and established as cell lines in vitro. All cell lines were sequenced and mutational

signatures for each were obtained (top). A representative signatures contribution pie chart is displayed. Signature 14

= combined MMRD + POLE proofreading loss; Signature 10 = POLE proofreading loss; Signature 28 = POLE

proofreading loss; Signature 21 = MMRD. One cell line was used to test the efficacy of anti-PD-1 therapy in vivo

(bottom). Treatment commenced at 100mm3.

0

100

200

300

400

500

600

700

SN

Vs/M

b

Established cell lines

Signature.10

Signature.14

Signature.21Signature.28

unknown

0

100

200

300

400

500

600

700

SN

Vs/M

b

Established cell lines

Signature.10

Signature.14

Signature.21Signature.28

unknown

146

6.2.2 Ablation of combined RRD in oligodendrocyte precursors for a more clinically relevant model of brain tumorigenesis

Although posterior fossa cancers occur in children with RRD, the majority of MBTs are high grade

gliomas. We therefore wanted to utilize a Cre-driver strain which would more specifically ablate

MMR in the mouse cerebrum rather than the entire CNS, and potentially better recapitulate the

diversity of tumor types observed in CMMRD. One avenue of current work that I have initiated is

to substitute Nestin-Cre for Olig2-Cre to generate Olig2-Cre+/Msh2LoxP/LoxP (abbreviated

OC+Msh2L/L), Olig2-Cre+/Msh2LoxP/LoxP/PoleS459F/+ (abbreviated OC+Msh2L/LPoleSF/+), and Olig2-

Cre+/Msh2LoxP/LoxP/PoleS459F/S459F (abbreviated mice OC+Msh2L/LPoleSF/SF) mice. Olig2-Cre is a

transgenic mouse line which is expressed in oligodendrocyte-lineage cells and motor neurons

(Zawadzka et al. 2010). These experiments are ongoing and provide promising preliminary data.

As with NC+Msh2L/L animals, OC+Msh2L/L mice demonstrate normal lifespans comparable to Cre

negative controls. However, in contrast to NC+Msh2L/LPoleSF/+ mice, OC+Msh2L/LPoleSF/+ mice

demonstrate significantly longer survival and develop CNS tumors with less penetrance (Figure

6.2-2). Indeed, necropsy and histology findings demonstrate that these mice succumb to either

lymphomas or MBTs due to the presence of the germline heterozygous polymerase mutation.

OC+Msh2L/LPoleSF/SF animals never developed MBTs and only succumbed to lymphomas, with

survival comparable to PoleS459F/S459F mice (Figure 3.4-3 a). Interestingly, brain tumors that

developed from OC+Msh2L/LPoleSF/+ animals were a mixture of hind and forebrain tumors,

suggesting that this model may more accurately recapitulate the heterogeneity of the human

syndrome (Figure 6.2-2).

Current work includes establishing cell lines from tumors arising in these animals as well as

genomic characterization of these Olig2-Cre-driven brain tumors for comparison with those

tumors derived from Nestin-Cre mice. We have submitted 4 tumors from these mice for WES and

expect them to display similar genomic features to NC+Msh2L/LPoleSF/+ tumors. Due to the

prevalence of lymphomas in these animals, we have additionally acquired LSL-PoleP286R mice in

collaboration with the Castrillon Lab to substitute the germline polymerase mutant model for an

inducible model and prevent lymphoma-related death which is likely masking the full extent of

the brain tumor phenotype in this model. These models can then be used to assess differences in

tumor evolution by imaging and sacrificing animals at different timepoints to sequence early

147

lesions for additional driver events beyond polymerase mutations since MMRD and PPD are not

cancer causing in and of themselves.

Figure 6.2-2 OC+/Msh2L/L/PoleSF/+ mutant mice display delayed brain tumor onset and greater incidence of

glioma-like tumors

Survival findings for OC+/Msh2L/L/PoleSF/+ and OC+/Msh2L/L/PoleSF/SF mutant mice (upper). Lymphoma-free survival

and brain tumor-free survival are plotted separately for OC+/Msh2L/L/PoleSF/+ mice. HE and IHC staining are shown

for a representative Olig2-Cre driven RRD brain tumor which developed in the cerebrum (bottom). Note that hindbrain

tumors were observed as well. Images were taken at 40x magnification (black scale bar indicates 50 μm; white bar

indicates 20 μm) and either 5x magnification (HE inlet; scale bar indicates 500 μm) or 1.2x magnification (IHC inlets;

scale bar indicates 1000 μm).

6.2.3 Emerging secondary driver events in CMMRD brain tumors

Comprehensive analyses to elucidate molecular subgroups within CMMRD brain tumors are

ongoing. While combined RRD brain tumors constitute roughly a third of these cancers, two thirds

of CMMRD brain tumors do not harbor somatic polymerase mutations (data not shown). Within

CMMRD HGGs specifically, three subgroups are emerging: MMRD with somatic PPD, MMRD

tumors harbouring TP53 mutations, and finally MMRD tumors harboring mutations in both TP53

0 5 10 150

50

100

Age (months)

Su

rviv

al(%

)

Log Rank

p < 0.0001

Olig2Cre+/Msh2fl/fl/PoleS459F/+ (BT)

Olig2Cre+/Msh2fl/fl/PoleS459F/+ (Lymph)

Olig2Cre+/Msh2fl/fl/PoleS459F/S459F (Lymph)

HE Synaptophysin Gfap Olig2

148

and isocitrate dehydrogenases 1 and 2 (IHD1/2). IDH mutations are common in adult HGGs and

in CMMRD appear to result in worse outcomes than PPD tumors, in particular in response to ICI

therapy. Whether this is due to their lower mutational burdens or other tumor-intrinsic mechanisms

contributing to immune evasion is unknown. Our ongoing work is aimed at developing appropriate

animal models to capture these distinct molecular groups. We are currently developing combined

mouse models with brain specific MMRD and Trp53 loss as well as triple mutant mice harboring

brain specific MMRD, Trp53 loss, and an inducible Idh1 mutation. This triple mutant model will

be important as IDH1 mutant gliomas are notoriously difficult to study as they tend not to establish

well as PDX models. For preclinical modeling of response to ICI therapy, both of these new models

will be vital as they may prove less responsive to ICI than MMRD/PPD mouse brain tumors and

can therefore more accurately model CMMRD human brain tumors that do not respond to

checkpoint inhibition, thus providing a platform to test novel and combinatorial therapies for these

cancers.

149

References

Aaltonen, L. A., P. Peltomäki, F. S. Leach, P. Sistonen, L. Pylkkänen, J. P. Mecklin, H. Järvinen, S. M.

Powell, J. Jen, S. R. Hamilton and et al. (1993). "Clues to the pathogenesis of familial colorectal cancer."

Science 260(5109): 812-816.

Adzhubei, I., D. M. Jordan and S. R. Sunyaev (2013). "Predicting functional effect of human missense

mutations using PolyPhen-2." Curr Protoc Hum Genet 76: 7.20.21-27.20.41.

Aebi, S., B. Kurdi-Haidar, R. Gordon, B. Cenni, H. Zheng, D. Fink, R. D. Christen, C. R. Boland, M.

Koi, R. Fishel and S. B. Howell (1996). "Loss of DNA mismatch repair in acquired resistance to

cisplatin." Cancer Res 56(13): 3087-3090.

Albertson, T. M., M. Ogawa, J. M. Bugni, L. E. Hays, Y. Chen, Y. Wang, P. M. Treuting, J. A. Heddle,

R. E. Goldsby and B. D. Preston (2009). "DNA polymerase epsilon and delta proofreading suppress

discrete mutator and cancer phenotypes in mice." Proc Natl Acad Sci U S A 106(40): 17101-17104.

Alexandrov, L. B., J. Kim, N. J. Haradhvala, M. N. Huang, A. W. Ng, A. Boot, K. R. Covington, D. A.

Gordenin, E. Bergstrom, N. Lopez-Bigas, L. J. Klimczak, J. R. McPherson, S. Morganella, R.

Sabarinathan, D. A. Wheeler, V. Mustonen, G. Getz, S. G. Rozen and M. R. Stratton (2018). "The

Repertoire of Mutational Signatures in Human Cancer." bioRxiv: 322859.

Alexandrov, L. B., J. Kim, N. J. Haradhvala, M. N. Huang, A. W. Tian Ng, Y. Wu, A. Boot, K. R.

Covington, D. A. Gordenin, E. N. Bergstrom, S. M. A. Islam, N. Lopez-Bigas, L. J. Klimczak, J. R.

McPherson, S. Morganella, R. Sabarinathan, D. A. Wheeler, V. Mustonen, G. Getz, S. G. Rozen and M.

R. Stratton (2020). "The repertoire of mutational signatures in human cancer." Nature 578(7793): 94-101.

Alexandrov, L. B., S. Nik-Zainal, D. C. Wedge, S. A. Aparicio, S. Behjati, A. V. Biankin, G. R. Bignell,

N. Bolli, A. Borg, A. L. Borresen-Dale, S. Boyault, B. Burkhardt, A. P. Butler, C. Caldas, H. R. Davies,

C. Desmedt, R. Eils, J. E. Eyfjord, J. A. Foekens, M. Greaves, F. Hosoda, B. Hutter, T. Ilicic, S. Imbeaud,

M. Imielinski, N. Jager, D. T. Jones, D. Jones, S. Knappskog, M. Kool, S. R. Lakhani, C. Lopez-Otin, S.

Martin, N. C. Munshi, H. Nakamura, P. A. Northcott, M. Pajic, E. Papaemmanuil, A. Paradiso, J. V.

Pearson, X. S. Puente, K. Raine, M. Ramakrishna, A. L. Richardson, J. Richter, P. Rosenstiel, M.

Schlesner, T. N. Schumacher, P. N. Span, J. W. Teague, Y. Totoki, A. N. Tutt, R. Valdes-Mas, M. M. van

Buuren, L. van 't Veer, A. Vincent-Salomon, N. Waddell, L. R. Yates, I. Australian Pancreatic Cancer

Genome, I. B. C. Consortium, I. M.-S. Consortium, I. PedBrain, J. Zucman-Rossi, P. A. Futreal, U.

McDermott, P. Lichter, M. Meyerson, S. M. Grimmond, R. Siebert, E. Campo, T. Shibata, S. M. Pfister,

P. J. Campbell and M. R. Stratton (2013). "Signatures of mutational processes in human cancer." Nature

500(7463): 415-421.

Alexandrov, L. B., S. Nik-Zainal, D. C. Wedge, S. A. J. R. Aparicio, S. Behjati, A. V. Biankin, G. R.

Bignell, N. Bolli, A. Borg, A.-L. Borresen-Dale, S. Boyault, B. Burkhardt, A. P. Butler, C. Caldas, H. R.

Davies, C. Desmedt, R. Eils, J. E. Eyfjord, J. A. Foekens, M. Greaves, F. Hosoda, B. Hutter, T. Ilicic, S.

Imbeaud, M. Imielinsk, N. Jager, D. T. W. Jones, D. Jones, S. Knappskog, M. Kool, S. R. Lakhani, C.

Lopez-Otin, S. Martin, N. C. Munshi, H. Nakamura, P. A. Northcott, M. Pajic, E. Papaemmanuil, A.

Paradiso, J. V. Pearson, X. S. Puente, K. Raine, M. Ramakrishna, A. L. Richardson, J. Richter, P.

Rosenstiel, M. Schlesner, T. N. Schumacher, P. N. Span, J. W. Teague, Y. Totoki, A. N. J. Tutt, R.

Valdes-Mas, M. M. van Buuren, L. van /'t Veer, A. Vincent-Salomon, N. Waddell, L. R. Yates, I.

Australian Pancreatic Cancer Genome, I. B. C. Consortium, I. M.-S. Consortium, I. PedBrain, J. Zucman-

Rossi, P. Andrew Futreal, U. McDermott, P. Lichter, M. Meyerson, S. M. Grimmond, R. Siebert, E.

Campo, T. Shibata, S. M. Pfister, P. J. Campbell and M. R. Stratton (2013). "Signatures of mutational

processes in human cancer." Nature 500(7463): 415-421.

150

Amayiri, N., M. Al-Hussaini, M. Swaidan, I. Jaradat, M. Qandeel, U. Tabori, C. Hawkins, A.

Musharbash, K. Alsaad and E. Bouffet (2016). "Synchronous glioblastoma and medulloblastoma in a

child with mismatch repair mutation." Childs Nerv Syst 32(3): 553-557.

Amayiri, N., U. Tabori, B. Campbell, D. Bakry, M. Aronson, C. Durno, P. Rakopoulos, D. Malkin, I.

Qaddoumi, A. Musharbash, M. Swaidan, E. Bouffet, C. Hawkins and M. Al-Hussaini (2016). "High

frequency of mismatch repair deficiency among pediatric high grade gliomas in Jordan." Int J Cancer

138(2): 380-385.

Amir el, A. D., K. L. Davis, M. D. Tadmor, E. F. Simonds, J. H. Levine, S. C. Bendall, D. K. Shenfeld, S.

Krishnaswamy, G. P. Nolan and D. Pe'er (2013). "viSNE enables visualization of high dimensional

single-cell data and reveals phenotypic heterogeneity of leukemia." Nat Biotechnol 31(6): 545-552.

An, Y., J. R. Adams, D. P. Hollern, A. Zhao, S. G. Chang, M. S. Gams, P. E. D. Chung, X. He, R. Jangra,

J. S. Shah, J. Yang, L. A. Beck, N. Raghuram, K. J. Kozma, A. J. Loch, W. Wang, C. Fan, S. J. Done, E.

Zacksenhaus, C. J. Guidos, C. M. Perou and S. E. Egan (2018). "Cdh1 and Pik3ca Mutations Cooperate to

Induce Immune-Related Invasive Lobular Carcinoma of the Breast." Cell Rep 25(3): 702-714.e706.

Ansell, S. M., A. M. Lesokhin, I. Borrello, A. Halwani, E. C. Scott, M. Gutierrez, S. J. Schuster, M. M.

Millenson, D. Cattry, G. J. Freeman, S. J. Rodig, B. Chapuy, A. H. Ligon, L. Zhu, J. F. Grosso, S. Y.

Kim, J. M. Timmerman, M. A. Shipp and P. Armand (2015). "PD-1 blockade with nivolumab in relapsed

or refractory Hodgkin's lymphoma." N Engl J Med 372(4): 311-319.

Antonia, S. J., J. A. López-Martin, J. Bendell, P. A. Ott, M. Taylor, J. P. Eder, D. Jäger, M. C. Pietanza,

D. T. Le, F. de Braud, M. A. Morse, P. A. Ascierto, L. Horn, A. Amin, R. N. Pillai, J. Evans, I. Chau, P.

Bono, A. Atmaca, P. Sharma, C. T. Harbison, C. S. Lin, O. Christensen and E. Calvo (2016). "Nivolumab

alone and nivolumab plus ipilimumab in recurrent small-cell lung cancer (CheckMate 032): a multicentre,

open-label, phase 1/2 trial." Lancet Oncol 17(7): 883-895.

Aoude, L. G., E. Heitzer, P. Johansson, M. Gartside, K. Wadt, A. L. Pritchard, J. M. Palmer, J. Symmons,

A. M. Gerdes, G. W. Montgomery, N. G. Martin, I. Tomlinson, S. Kearsey and N. K. Hayward (2015).

"POLE mutations in families predisposed to cutaneous melanoma." Fam Cancer 14(4): 621-628.

Bakry, D., M. Aronson, C. Durno, H. Rimawi, R. Farah, Q. K. Alharbi, M. Alharbi, A. Shamvil, S. Ben-

Shachar, M. Mistry, S. Constantini, R. Dvir, I. Qaddoumi, S. Gallinger, J. Lerner-Ellis, A. Pollett, D.

Stephens, S. Kelies, E. Chao, D. Malkin, E. Bouffet, C. Hawkins and U. Tabori (2014). "Genetic and

clinical determinants of constitutional mismatch repair deficiency syndrome: report from the

constitutional mismatch repair deficiency consortium." Eur J Cancer 50(5): 987-996.

Barbari, S. R., D. P. Kane, E. A. Moore and P. V. Shcherbakova (2018). "Functional Analysis of Cancer-

Associated DNA Polymerase ε Variants in <em>Saccharomyces cerevisiae</em>." G3:

Genes|Genomes|Genetics 8(3): 1019-1029.

Barbari, S. R. and P. V. Shcherbakova (2017). "Replicative DNA polymerase defects in human cancers:

Consequences, mechanisms, and implications for therapy." DNA Repair (Amst) 56: 16-25.

Barber, D. L., E. J. Wherry, D. Masopust, B. Zhu, J. P. Allison, A. H. Sharpe, G. J. Freeman and R.

Ahmed (2006). "Restoring function in exhausted CD8 T cells during chronic viral infection." Nature

439(7077): 682-687.

Bargou, R., E. Leo, G. Zugmaier, M. Klinger, M. Goebeler, S. Knop, R. Noppeney, A. Viardot, G. Hess,

M. Schuler, H. Einsele, C. Brandl, A. Wolf, P. Kirchinger, P. Klappers, M. Schmidt, G. Riethmüller, C.

Reinhardt, P. A. Baeuerle and P. Kufer (2008). "Tumor regression in cancer patients by very low doses of

a T cell-engaging antibody." Science 321(5891): 974-977.

151

Barker, N., J. H. van Es, J. Kuipers, P. Kujala, M. van den Born, M. Cozijnsen, A. Haegebarth, J.

Korving, H. Begthel, P. J. Peters and H. Clevers (2007). "Identification of stem cells in small intestine

and colon by marker gene Lgr5." Nature 449(7165): 1003-1007.

Bartholomeusz, C., T. Oishi, H. Saso, U. Akar, P. Liu, K. Kondo, A. Kazansky, S. Krishnamurthy, J. Lee,

F. J. Esteva, J. Kigawa and N. T. Ueno (2012). "MEK1/2 inhibitor selumetinib (AZD6244) inhibits

growth of ovarian clear cell carcinoma in a PEA-15-dependent manner in a mouse xenograft model." Mol

Cancer Ther 11(2): 360-369.

Bebenek, K. and T. A. Kunkel (2004). "Functions of DNA polymerases." Adv Protein Chem 69: 137-165.

Beese, L. S. and T. A. Steitz (1991). "Structural basis for the 3'-5' exonuclease activity of Escherichia coli

DNA polymerase I: a two metal ion mechanism." Embo j 10(1): 25-33.

Behan, F. M., F. Iorio, G. Picco, E. Gonçalves, C. M. Beaver, G. Migliardi, R. Santos, Y. Rao, F. Sassi,

M. Pinnelli, R. Ansari, S. Harper, D. A. Jackson, R. McRae, R. Pooley, P. Wilkinson, D. van der Meer, D.

Dow, C. Buser-Doepner, A. Bertotti, L. Trusolino, E. A. Stronach, J. Saez-Rodriguez, K. Yusa and M. J.

Garnett (2019). "Prioritization of cancer therapeutic targets using CRISPR-Cas9 screens." Nature

568(7753): 511-516.

Bellido, F., M. Pineda, G. Aiza, R. Valdés-Mas, M. Navarro, D. A. Puente, T. Pons, S. González, S.

Iglesias, E. Darder, V. Piñol, J. L. Soto, A. Valencia, I. Blanco, M. Urioste, J. Brunet, C. Lázaro, G.

Capellá, X. S. Puente and L. Valle (2016). "POLE and POLD1 mutations in 529 kindred with familial

colorectal cancer and/or polyposis: review of reported cases and recommendations for genetic testing and

surveillance." Genetics in Medicine 18(4): 325-332.

Birkeland, E., S. Zhang, D. Poduval, J. Geisler, S. Nakken, D. Vodak, L. A. Meza-Zepeda, E. Hovig, O.

Myklebost, S. Knappskog and P. E. Lønning (2018). "Patterns of genomic evolution in advanced

melanoma." Nature Communications 9(1): 2665.

Bodo, S., C. Colas, O. Buhard, A. Collura, J. Tinat, N. Lavoine, A. Guilloux, A. Chalastanis, P. Lafitte, F.

Coulet, M. P. Buisine, D. Ilencikova, C. Ruiz-Ponte, M. Kinzel, S. Grandjouan, H. Brems, S. Lejeune, H.

Blanché, Q. Wang, O. Caron, O. Cabaret, M. Svrcek, D. Vidaud, B. Parfait, A. Verloes, U. J. Knappe, F.

Soubrier, I. Mortemousque, A. Leis, J. Auclair-Perrossier, T. Frébourg, J. F. Fléjou, N. Entz-Werle, J.

Leclerc, D. Malka, O. Cohen-Haguenauer, Y. Goldberg, A. M. Gerdes, F. Fedhila, M. Mathieu-Dramard,

R. Hamelin, B. Wafaa, M. Gauthier-Villars, F. Bourdeaut, E. Sheridan, H. Vasen, L. Brugières, K.

Wimmer, M. Muleris and A. Duval (2015). "Diagnosis of Constitutional Mismatch Repair-Deficiency

Syndrome Based on Microsatellite Instability and Lymphocyte Tolerance to Methylating Agents."

Gastroenterology 149(4): 1017-1029.e1013.

Boland, C. R. and F. J. Troncale (1984). "Familial Colonic Cancer Without Antecedent Polyposis."

Annals of Internal Medicine 100(5): 700-701.

Bonadona, V., B. Bonaiti, S. Olschwang, S. Grandjouan, L. Huiart, M. Longy, R. Guimbaud, B. Buecher,

Y. J. Bignon, O. Caron, C. Colas, C. Nogues, S. Lejeune-Dumoulin, L. Olivier-Faivre, F. Polycarpe-

Osaer, T. D. Nguyen, F. Desseigne, J. C. Saurin, P. Berthet, D. Leroux, J. Duffour, S. Manouvrier, T.

Frebourg, H. Sobol, C. Lasset, C. Bonaiti-Pellie and N. French Cancer Genetics (2011). "Cancer risks

associated with germline mutations in MLH1, MSH2, and MSH6 genes in Lynch syndrome." JAMA

305(22): 2304-2310.

Boudil, A., I. R. Matei, H. Y. Shih, G. Bogdanoski, J. S. Yuan, S. G. Chang, B. Montpellier, P. E.

Kowalski, V. Voisin, S. Bashir, G. D. Bader, M. S. Krangel and C. J. Guidos (2015). "IL-7 coordinates

proliferation, differentiation and Tcra recombination during thymocyte β-selection." Nat Immunol 16(4):

397-405.

152

Bouffet, E., V. Larouche, B. B. Campbell, D. Merico, R. de Borja, M. Aronson, C. Durno, J. Krueger, V.

Cabric, V. Ramaswamy, N. Zhukova, G. Mason, R. Farah, S. Afzal, M. Yalon, G. Rechavi, V.

Magimairajan, M. F. Walsh, S. Constantini, R. Dvir, R. Elhasid, A. Reddy, M. Osborn, M. Sullivan, J.

Hansford, A. Dodgshun, N. Klauber-Demore, L. Peterson, S. Patel, S. Lindhorst, J. Atkinson, Z. Cohen,

R. Laframboise, P. Dirks, M. Taylor, D. Malkin, S. Albrecht, R. W. Dudley, N. Jabado, C. E. Hawkins,

A. Shlien and U. Tabori (2016). "Immune Checkpoint Inhibition for Hypermutant Glioblastoma

Multiforme Resulting From Germline Biallelic Mismatch Repair Deficiency." J Clin Oncol 34(19): 2206-

2211.

Bouffet, E., V. Larouche, B. B. Campbell, D. Merico, R. de Borja, M. Aronson, C. Durno, J. Krueger, V.

Cabric, V. Ramaswamy, N. Zhukova, G. Mason, R. Farah, S. Afzal, M. Yalon, G. Rechavi, V.

Magimairajan, M. F. Walsh, S. Constantini, R. Dvir, R. Elhasid, A. Reddy, M. Osborn, M. Sullivan, J.

Hansford, A. Dodgshun, N. Klauber-Demore, L. Peterson, S. Patel, S. Lindhorst, J. Atkinson, Z. Cohen,

R. Laframboise, P. Dirks, M. Taylor, D. Malkin, S. Albrecht, R. W. R. Dudley, N. Jabado, C. E. Hawkins,

A. Shlien and U. Tabori (2016). "Immune Checkpoint Inhibition for Hypermutant Glioblastoma

Multiforme Resulting From Germline Biallelic Mismatch Repair Deficiency." Journal of Clinical

Oncology.

Brahmer, J., K. L. Reckamp, P. Baas, L. Crinò, W. E. Eberhardt, E. Poddubskaya, S. Antonia, A.

Pluzanski, E. E. Vokes, E. Holgado, D. Waterhouse, N. Ready, J. Gainor, O. Arén Frontera, L. Havel, M.

Steins, M. C. Garassino, J. G. Aerts, M. Domine, L. Paz-Ares, M. Reck, C. Baudelet, C. T. Harbison, B.

Lestini and D. R. Spigel (2015). "Nivolumab versus Docetaxel in Advanced Squamous-Cell Non-Small-

Cell Lung Cancer." N Engl J Med 373(2): 123-135.

Branch, P., G. Aquilina, M. Bignami and P. Karran (1993). "Defective mismatch binding and a mutator

phenotype in cells tolerant to DNA damage." Nature 362(6421): 652-654.

Briggs, S. and I. Tomlinson (2013). "Germline and somatic polymerase epsilon and delta mutations

define a new class of hypermutated colorectal and endometrial cancers." J Pathol 230(2): 148-153.

Bronner, C. E., S. M. Baker, P. T. Morrison, G. Warren, L. G. Smith, M. K. Lescoe, M. Kane, C.

Earabino, J. Lipford, A. Lindblom and et al. (1994). "Mutation in the DNA mismatch repair gene

homologue hMLH1 is associated with hereditary non-polyposis colon cancer." Nature 368(6468): 258-

261.

Brown, K. D., A. Rathi, R. Kamath, D. I. Beardsley, Q. Zhan, J. L. Mannino and R. Baskaran (2003).

"The mismatch repair system is required for S-phase checkpoint activation." Nat Genet 33(1): 80-84.

Buchanan, D. D., J. R. Stewart, M. Clendenning, C. Rosty, K. Mahmood, B. J. Pope, M. A. Jenkins, J. L.

Hopper, M. C. Southey, F. A. Macrae, I. M. Winship and A. K. Win (2018). "Risk of colorectal cancer for

carriers of a germ-line mutation in POLE or POLD1." Genetics in Medicine 20(8): 890-895.

Burgers, P. M. (2009). "Polymerase dynamics at the eukaryotic DNA replication fork." J Biol Chem

284(7): 4041-4045.

Burgers, P. M. J. and T. A. Kunkel (2017). "Eukaryotic DNA Replication Fork." Annu Rev Biochem 86:

417-438.

Bush, L., M. Aronson, U. Tabori, B. B. Campbell, R. B. Bedgood and K. Jasperson (2019). "Delineating a

new feature of constitutional mismatch repair deficiency (CMMRD) syndrome: breast cancer." Fam

Cancer 18(1): 105-108.

Cahill, D. P., K. K. Levine, R. A. Betensky, P. J. Codd, C. A. Romany, L. B. Reavie, T. T. Batchelor, P.

A. Futreal, M. R. Stratton, W. T. Curry, A. J. Iafrate and D. N. Louis (2007). "Loss of the Mismatch

Repair Protein MSH6 in Human Glioblastomas Is Associated with Tumor Progression during

Temozolomide Treatment." Clinical Cancer Research 13(7): 2038-2045.

153

Campbell, B. B., N. Light, D. Fabrizio, M. Zatzman, F. Fuligni, R. de Borja, S. Davidson, M. Edwards, J.

A. Elvin, K. P. Hodel, W. J. Zahurancik, Z. Suo, T. Lipman, K. Wimmer, C. P. Kratz, D. C. Bowers, T.

W. Laetsch, G. P. Dunn, T. M. Johanns, M. R. Grimmer, I. V. Smirnov, V. Larouche, D. Samuel, A.

Bronsema, M. Osborn, D. Stearns, P. Raman, K. A. Cole, P. B. Storm, M. Yalon, E. Opocher, G. Mason,

G. A. Thomas, M. Sabel, B. George, D. S. Ziegler, S. Lindhorst, V. M. Issai, S. Constantini, H. Toledano,

R. Elhasid, R. Farah, R. Dvir, P. Dirks, A. Huang, M. A. Galati, J. Chung, V. Ramaswamy, M. S. Irwin,

M. Aronson, C. Durno, M. D. Taylor, G. Rechavi, J. M. Maris, E. Bouffet, C. Hawkins, J. F. Costello, M.

S. Meyn, Z. F. Pursell, D. Malkin, U. Tabori and A. Shlien (2017). "Comprehensive Analysis of

Hypermutation in Human Cancer." Cell 171(5): 1042-1056.e1010.

Canale, F. P., M. C. Ramello, N. Núñez, C. L. Araujo Furlan, S. N. Bossio, M. Gorosito Serrán, J. Tosello

Boari, A. Del Castillo, M. Ledesma, C. Sedlik, E. Piaggio, A. Gruppi, E. A. Acosta Rodríguez and C. L.

Montes (2018). "CD39 Expression Defines Cell Exhaustion in Tumor-Infiltrating CD8(+) T Cells."

Cancer Res 78(1): 115-128.

Cancer Genome Atlas, N. (2015). "Genomic Classification of Cutaneous Melanoma." Cell 161(7): 1681-

1696.

Cancer Genome Atlas Network, T. (2012). "Comprehensive molecular characterization of human colon

and rectal cancer." Nature 487(7407): 330-337.

Carethers, J. M., E. J. Smith, C. A. Behling, L. Nguyen, A. Tajima, R. T. Doctolero, B. L. Cabrera, A.

Goel, C. A. Arnold, K. Miyai and C. R. Boland (2004). "Use of 5-fluorouracil and survival in patients

with microsatellite-unstable colorectal cancer." Gastroenterology 126(2): 394-401.

Cejka, P., L. Stojic, G. Marra and J. Jiricny (2004). "Is mismatch repair really required for ionizing

radiation-induced DNA damage signaling?" Nat Genet 36(5): 432-433; author reply 434.

Chan, E. M., T. Shibue, J. M. McFarland, B. Gaeta, M. Ghandi, N. Dumont, A. Gonzalez, J. S.

McPartlan, T. Li, Y. Zhang, J. Bin Liu, J. B. Lazaro, P. Gu, C. G. Piett, A. Apffel, S. O. Ali, R. Deasy, P.

Keskula, R. W. S. Ng, E. A. Roberts, E. Reznichenko, L. Leung, M. Alimova, M. Schenone, M. Islam, Y.

E. Maruvka, Y. Liu, J. Roper, S. Raghavan, M. Giannakis, Y. Y. Tseng, Z. D. Nagel, A. D'Andrea, D. E.

Root, J. S. Boehm, G. Getz, S. Chang, T. R. Golub, A. Tsherniak, F. Vazquez and A. J. Bass (2019).

"WRN helicase is a synthetic lethal target in microsatellite unstable cancers." Nature 568(7753): 551-556.

Chang, M. T., T. S. Bhattarai, A. M. Schram, C. M. Bielski, M. T. A. Donoghue, P. Jonsson, D.

Chakravarty, S. Phillips, C. Kandoth, A. Penson, A. Gorelick, T. Shamu, S. Patel, C. Harris, J. Gao, S. O.

Sumer, R. Kundra, P. Razavi, B. T. Li, D. N. Reales, N. D. Socci, G. Jayakumaran, A. Zehir, R. Benayed,

M. E. Arcila, S. Chandarlapaty, M. Ladanyi, N. Schultz, J. Baselga, M. F. Berger, N. Rosen, D. B. Solit,

D. M. Hyman and B. S. Taylor (2018). "Accelerating Discovery of Functional Mutant Alleles in Cancer."

Cancer Discov 8(2): 174-183.

Chen, L. and D. B. Flies (2013). "Molecular mechanisms of T cell co-stimulation and co-inhibition." Nat

Rev Immunol 13(4): 227-242.

Chu, W. K. and I. D. Hickson (2009). "RecQ helicases: multifunctional genome caretakers." Nat Rev

Cancer 9(9): 644-654.

Chung, H. C., W. Ros, J. P. Delord, R. Perets, A. Italiano, R. Shapira-Frommer, L. Manzuk, S. A. Piha-

Paul, L. Xu, S. Zeigenfuss, S. K. Pruitt and A. Leary (2019). "Efficacy and Safety of Pembrolizumab in

Previously Treated Advanced Cervical Cancer: Results From the Phase II KEYNOTE-158 Study." J Clin

Oncol 37(17): 1470-1478.

Chung, J., Y. E. Maruvka, S. Sudhaman, J. Kelly, N. J. Haradhvala, V. Bianchi, M. Edwards, V. J.

Forster, N. M. Nunes, M. A. Galati, M. Komosa, S. Deshmukh, V. Cabric, S. Davidson, M. Zatzman, N.

Light, R. Hayes, L. Brunga, N. D. Anderson, B. Ho, K. P. Hodel, R. Siddaway, A. S. Morrissy, D. C.

154

Bowers, V. Larouche, A. Bronsema, M. Osborn, K. A. Cole, E. Opocher, G. Mason, G. A. Thomas, B.

George, D. S. Ziegler, S. Lindhorst, M. Vanan, M. Yalon-Oren, A. T. Reddy, M. Massimino, P. Tomboc,

A. Van Damme, A. Lossos, C. Durno, M. Aronson, D. A. Morgenstern, E. Bouffet, A. Huang, M. D.

Taylor, A. Villani, D. Malkin, C. E. Hawkins, Z. F. Pursell, A. Shlien, T. A. Kunkel, G. Getz and U.

Tabori (2020). "DNA polymerase and mismatch repair exert distinct microsatellite instability signatures

in normal and malignant human cells." Cancer Discov.

Church, D. N., S. E. Briggs, C. Palles, E. Domingo, S. J. Kearsey, J. M. Grimes, M. Gorman, L. Martin,

K. M. Howarth, S. V. Hodgson, K. Kaur, J. Taylor and I. P. Tomlinson (2013). "DNA polymerase epsilon

and delta exonuclease domain mutations in endometrial cancer." Hum Mol Genet 22(14): 2820-2828.

Cortez, D. (2019). "Replication-Coupled DNA Repair." Molecular Cell 74(5): 866-876.

Crotty, S. (2019). "T Follicular Helper Cell Biology: A Decade of Discovery and Diseases." Immunity

50(5): 1132-1148.

D'Angelo, F., M. Ceccarelli, Tala, L. Garofano, J. Zhang, V. Frattini, F. P. Caruso, G. Lewis, K. D.

Alfaro, L. Bauchet, G. Berzero, D. Cachia, M. Cangiano, L. Capelle, J. de Groot, F. DiMeco, F. Ducray,

W. Farah, G. Finocchiaro, S. Goutagny, C. Kamiya-Matsuoka, C. Lavarino, H. Loiseau, V. Lorgis, C. E.

Marras, I. McCutcheon, D.-H. Nam, S. Ronchi, V. Saletti, R. Seizeur, J. Slopis, M. Suñol, F. Vandenbos,

P. Varlet, D. Vidaud, C. Watts, V. Tabar, D. E. Reuss, S.-K. Kim, D. Meyronet, K. Mokhtari, H.

Salvador, K. P. Bhat, M. Eoli, M. Sanson, A. Lasorella and A. Iavarone (2019). "The molecular landscape

of glioma in patients with Neurofibromatosis 1." Nature medicine 25(1): 176-187.

da Costa, L. T., B. Liu, W. el-Deiry, S. R. Hamilton, K. W. Kinzler, B. Vogelstein, S. Markowitz, J. K.

Willson, A. de la Chapelle, K. M. Downey and et al. (1995). "Polymerase delta variants in RER colorectal

tumours." Nat Genet 9(1): 10-11.

Danecek, P., A. Auton, G. Abecasis, C. A. Albers, E. Banks, M. A. DePristo, R. E. Handsaker, G. Lunter,

G. T. Marth, S. T. Sherry, G. McVean and R. Durbin (2011). "The variant call format and VCFtools."

Bioinformatics 27(15): 2156-2158.

Danska, J. S., F. Pflumio, C. J. Williams, O. Huner, J. E. Dick and C. J. Guidos (1994). "Rescue of T cell-

specific V(D)J recombination in SCID mice by DNA-damaging agents." Science 266(5184): 450-455.

Davis, K. L., E. Fox, M. S. Merchant, J. M. Reid, R. A. Kudgus, X. Liu, C. G. Minard, S. Voss, S. L.

Berg, B. J. Weigel and C. L. Mackall (2020). "Nivolumab in children and young adults with relapsed or

refractory solid tumours or lymphoma (ADVL1412): a multicentre, open-label, single-arm, phase 1–2

trial." The Lancet Oncology 21(4): 541-550.

de Wind, N., M. Dekker, A. Berns, M. Radman and H. te Riele (1995). "Inactivation of the mouse Msh2

gene results in mismatch repair deficiency, methylation tolerance, hyperrecombination, and predisposition

to cancer." Cell 82(2): 321-330.

DeWeese, T. L., J. M. Shipman, N. A. Larrier, N. M. Buckley, L. R. Kidd, J. D. Groopman, R. G. Cutler,

H. te Riele and W. G. Nelson (1998). "Mouse embryonic stem cells carrying one or two defective Msh2

alleles respond abnormally to oxidative stress inflicted by low-level radiation." Proc Natl Acad Sci U S A

95(20): 11915-11920.

DiJoseph, J. F., D. C. Armellino, E. R. Boghaert, K. Khandke, M. M. Dougher, L. Sridharan, A. Kunz, P.

R. Hamann, B. Gorovits, C. Udata, J. K. Moran, A. G. Popplewell, S. Stephens, P. Frost and N. K. Damle

(2004). "Antibody-targeted chemotherapy with CMC-544: a CD22-targeted immunoconjugate of

calicheamicin for the treatment of B-lymphoid malignancies." Blood 103(5): 1807-1814.

Diouf, B., Q. Cheng, N. F. Krynetskaia, W. Yang, M. Cheok, D. Pei, Y. Fan, C. Cheng, E. Y. Krynetskiy,

H. Geng, S. Chen, W. E. Thierfelder, C. G. Mullighan, J. R. Downing, P. Hsieh, C. H. Pui, M. V. Relling

155

and W. E. Evans (2011). "Somatic deletions of genes regulating MSH2 protein stability cause DNA

mismatch repair deficiency and drug resistance in human leukemia cells." Nat Med 17(10): 1298-1303.

Dodgshun, A. J., K. Fukuoka, M. Edwards, V. J. Bianchi, A. Das, A. Sexton-Oates, V. Larouche, M. I.

Vanan, S. Lindhorst, M. Yalon, G. Mason, B. Crooks, S. Constantini, M. Massimino, S. Chiaravalli, J.

Ramdas, W. Mason, S. Ashraf, R. Farah, A. Van Damme, E. Opocher, S. A. Hamid, D. S. Ziegler, D.

Samuel, K. A. Cole, P. Tomboc, D. Stearns, G. A. Thomas, A. Lossos, M. Sullivan, J. R. Hansford, A.

Mackay, C. Jones, D. T. W. Jones, V. Ramaswamy, C. Hawkins, E. Bouffet and U. Tabori (2020).

"Germline-driven replication repair-deficient high-grade gliomas exhibit unique hypomethylation

patterns." Acta Neuropathol 140(5): 765-776.

Domingo, E., L. Freeman-Mills, E. Rayner, M. Glaire, S. Briggs, L. Vermeulen, E. Fessler, J. P. Medema,

A. Boot, H. Morreau, T. van Wezel, G. J. Liefers, R. A. Lothe, S. A. Danielsen, A. Sveen, A. Nesbakken,

I. Zlobec, A. Lugli, V. H. Koelzer, M. D. Berger, S. Castellví-Bel, J. Muñoz, M. de Bruyn, H. W. Nijman,

M. Novelli, K. Lawson, D. Oukrif, E. Frangou, P. Dutton, S. Tejpar, M. Delorenzi, R. Kerr, D. Kerr, I.

Tomlinson and D. N. Church (2016). "Somatic POLE proofreading domain mutation, immune response,

and prognosis in colorectal cancer: a retrospective, pooled biomarker study." Lancet Gastroenterol

Hepatol 1(3): 207-216.

Dong, H., S. E. Strome, D. R. Salomao, H. Tamura, F. Hirano, D. B. Flies, P. C. Roche, J. Lu, G. Zhu, K.

Tamada, V. A. Lennon, E. Celis and L. Chen (2002). "Tumor-associated B7-H1 promotes T-cell

apoptosis: a potential mechanism of immune evasion." Nat Med 8(8): 793-800.

Drake, J. W., B. Charlesworth, D. Charlesworth and J. F. Crow (1998). "Rates of spontaneous mutation."

Genetics 148(4): 1667-1686.

Drummond, J. T., A. Anthoney, R. Brown and P. Modrich (1996). "Cisplatin and adriamycin resistance

are associated with MutLalpha and mismatch repair deficiency in an ovarian tumor cell line." J Biol

Chem 271(33): 19645-19648.

Dry, J. R., S. Pavey, C. A. Pratilas, C. Harbron, S. Runswick, D. Hodgson, C. Chresta, R. McCormack, N.

Byrne, M. Cockerill, A. Graham, G. Beran, A. Cassidy, C. Haggerty, H. Brown, G. Ellison, J. Dering, B.

S. Taylor, M. Stark, V. Bonazzi, S. Ravishankar, L. Packer, F. Xing, D. B. Solit, R. S. Finn, N. Rosen, N.

K. Hayward, T. French and P. D. Smith (2010). "Transcriptional pathway signatures predict MEK

addiction and response to selumetinib (AZD6244)." Cancer Res 70(6): 2264-2273.

Duhen, T., R. Duhen, R. Montler, J. Moses, T. Moudgil, N. F. de Miranda, C. P. Goodall, T. C. Blair, B.

A. Fox, J. E. McDermott, S. C. Chang, G. Grunkemeier, R. Leidner, R. B. Bell and A. D. Weinberg

(2018). "Co-expression of CD39 and CD103 identifies tumor-reactive CD8 T cells in human solid

tumors." Nat Commun 9(1): 2724.

Durno, C. A., P. M. Sherman, M. Aronson, D. Malkin, C. Hawkins, D. Bakry, E. Bouffet, S. Gallinger, A.

Pollett, B. Campbell, U. Tabori and B. C. International (2015). "Phenotypic and genotypic

characterisation of biallelic mismatch repair deficiency (BMMR-D) syndrome." Eur J Cancer 51(8): 977-

983.

Edelmann, W., K. Yang, M. Kuraguchi, J. Heyer, M. Lia, B. Kneitz, K. Fan, A. M. Brown, M. Lipkin and

R. Kucherlapati (1999). "Tumorigenesis in Mlh1 and Mlh1/Apc1638N mutant mice." Cancer Res 59(6):

1301-1307.

Edelmann, W., K. Yang, A. Umar, J. Heyer, K. Lau, K. Fan, W. Liedtke, P. E. Cohen, M. F. Kane, J. R.

Lipford, N. Yu, G. F. Crouse, J. W. Pollard, T. Kunkel, M. Lipkin, R. Kolodner and R. Kucherlapati

(1997). "Mutation in the mismatch repair gene Msh6 causes cancer susceptibility." Cell 91(4): 467-477.

El-Khoueiry, A. B., B. Sangro, T. Yau, T. S. Crocenzi, M. Kudo, C. Hsu, T. Y. Kim, S. P. Choo, J.

Trojan, T. H. R. Welling, T. Meyer, Y. K. Kang, W. Yeo, A. Chopra, J. Anderson, C. Dela Cruz, L. Lang,

156

J. Neely, H. Tang, H. B. Dastani and I. Melero (2017). "Nivolumab in patients with advanced

hepatocellular carcinoma (CheckMate 040): an open-label, non-comparative, phase 1/2 dose escalation

and expansion trial." Lancet 389(10088): 2492-2502.

El Marjou, F., K.-P. Janssen, B. Hung-Junn Chang, M. Li, V. Hindie, L. Chan, D. Louvard, P. Chambon,

D. Metzger and S. Robine (2004). "Tissue-specific and inducible Cre-mediated recombination in the gut

epithelium." genesis 39(3): 186-193.

Eleveld, T. F., D. A. Oldridge, V. Bernard, J. Koster, L. C. Daage, S. J. Diskin, L. Schild, N. B. Bentahar,

A. Bellini, M. Chicard, E. Lapouble, V. Combaret, P. Legoix-Ne, J. Michon, T. J. Pugh, L. S. Hart, J.

Rader, E. F. Attiyeh, J. S. Wei, S. Zhang, A. Naranjo, J. M. Gastier-Foster, M. D. Hogarty, S.

Asgharzadeh, M. A. Smith, J. M. Guidry Auvil, T. B. Watkins, D. A. Zwijnenburg, M. E. Ebus, P. van

Sluis, A. Hakkert, E. van Wezel, C. E. van der Schoot, E. M. Westerhout, J. H. Schulte, G. A. Tytgat, M.

E. Dolman, I. Janoueix-Lerosey, D. S. Gerhard, H. N. Caron, O. Delattre, J. Khan, R. Versteeg, G.

Schleiermacher, J. J. Molenaar and J. M. Maris (2015). "Relapsed neuroblastomas show frequent RAS-

MAPK pathway mutations." Nat Genet 47(8): 864-871.

Elsayed, F. A., C. M. Kets, D. Ruano, B. van den Akker, A. R. Mensenkamp, M. Schrumpf, M. Nielsen,

J. T. Wijnen, C. M. Tops, M. J. Ligtenberg, H. F. A. Vasen, F. J. Hes, H. Morreau and T. van Wezel

(2015). "Germline variants in POLE are associated with early onset mismatch repair deficient colorectal

cancer." European Journal of Human Genetics 23(8): 1080-1084.

Eng, C., T. W. Kim, J. Bendell, G. Argilés, N. C. Tebbutt, M. Di Bartolomeo, A. Falcone, M. Fakih, M.

Kozloff, N. H. Segal, A. Sobrero, Y. Yan, I. Chang, A. Uyei, L. Roberts, F. Ciardiello, J. B. Ahn, J.

Asselah, S. Badarinath, S. Baijal, S. Begbie, S. Berry, J. L. Canon, R. G. Carbone, A. Cervantes, Y. J.

Cha, K. Chang, A. Chaudhry, E. Chmielowska, S. H. Cho, D. Chu, F. Couture, J. Cultrera, D.

Cunningham, E. Van Cutsem, P. J. Cuyle, J. Davies, S. Dowden, M. Dvorkin, V. Ganju, R. V. Garcia, R.

Kerr, T. Y. Kim, K. King, J. Kortmansky, M. Kozloff, K. O. Lam, J. Lee, A. S. Lee, B. Lesperance, G.

Luppi, B. Ma, E. Maiello, R. Mandanas, J. Marshall, G. Marx, S. Mullamitha, M. Nechaeva, J. O. Park,

N. Pavlakis, C. G. Ponce, P. Potemski, S. Raouf, J. Reeves, N. Segal, S. Siena, A. Smolin, J. O. Streb, A.

Strickland, E. Szutowicz-Zielinska, J. M. Tabernero, B. Tan, J. S. Valera, M. Van den Eynde, P.

Vergauwe, M. Vickers, M. Womack, M. Wroblewska and R. Young (2019). "Atezolizumab with or

without cobimetinib versus regorafenib in previously treated metastatic colorectal cancer (IMblaze370): a

multicentre, open-label, phase 3, randomised, controlled trial." The Lancet Oncology 20(6): 849-861.

Errico, A. (2015). "Melanoma: CheckMate 067--frontline nivolumab improves PFS alone or in

combination with ipilimumab." Nat Rev Clin Oncol 12(8): 435.

Fedele, C., H. Ran, B. Diskin, W. Wei, J. Jen, M. J. Geer, K. Araki, U. Ozerdem, D. M. Simeone, G.

Miller, B. G. Neel and K. H. Tang (2018). "SHP2 Inhibition Prevents Adaptive Resistance to MEK

Inhibitors in Multiple Cancer Models." Cancer Discov 8(10): 1237-1249.

Felsberg, J., N. Thon, S. Eigenbrod, B. Hentschel, M. C. Sabel, M. Westphal, G. Schackert, F. W. Kreth,

T. Pietsch, M. Löffler, M. Weller, G. Reifenberger and J. C. Tonn (2011). "Promoter methylation and

expression of MGMT and the DNA mismatch repair genes MLH1, MSH2, MSH6 and PMS2 in paired

primary and recurrent glioblastomas." Int J Cancer 129(3): 659-670.

Ferris, R. L., G. Blumenschein, Jr., J. Fayette, J. Guigay, A. D. Colevas, L. Licitra, K. Harrington, S.

Kasper, E. E. Vokes, C. Even, F. Worden, N. F. Saba, L. C. Iglesias Docampo, R. Haddad, T. Rordorf, N.

Kiyota, M. Tahara, M. Monga, M. Lynch, W. J. Geese, J. Kopit, J. W. Shaw and M. L. Gillison (2016).

"Nivolumab for Recurrent Squamous-Cell Carcinoma of the Head and Neck." N Engl J Med 375(19):

1856-1867.

Fishel, R. (2001). "The selection for mismatch repair defects in hereditary nonpolyposis colorectal cancer:

revising the mutator hypothesis." Cancer Res 61(20): 7369-7374.

157

Fishel, R., M. K. Lescoe, M. R. Rao, N. G. Copeland, N. A. Jenkins, J. Garber, M. Kane and R. Kolodner

(1993). "The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon

cancer." Cell 75(5): 1027-1038.

Flohr, T., J. C. Dai, J. Büttner, O. Popanda, E. Hagmüller and H. W. Thielmann (1999). "Detection of

mutations in the DNA polymerase delta gene of human sporadic colorectal cancers and colon cancer cell

lines." Int J Cancer 80(6): 919-929.

Fortune, J. M., Y. I. Pavlov, C. M. Welch, E. Johansson, P. M. Burgers and T. A. Kunkel (2005).

"Saccharomyces cerevisiae DNA polymerase delta: high fidelity for base substitutions but lower fidelity

for single- and multi-base deletions." J Biol Chem 280(33): 29980-29987.

Fortune, J. M., C. M. Stith, G. E. Kissling, P. M. Burgers and T. A. Kunkel (2006). "RPA and PCNA

suppress formation of large deletion errors by yeast DNA polymerase delta." Nucleic Acids Res 34(16):

4335-4341.

Frampton, G. M., A. Fichtenholtz, G. A. Otto, K. Wang, S. R. Downing, J. He, M. Schnall-Levin, J.

White, E. M. Sanford, P. An, J. Sun, F. Juhn, K. Brennan, K. Iwanik, A. Maillet, J. Buell, E. White, M.

Zhao, S. Balasubramanian, S. Terzic, T. Richards, V. Banning, L. Garcia, K. Mahoney, Z. Zwirko, A.

Donahue, H. Beltran, J. M. Mosquera, M. A. Rubin, S. Dogan, C. V. Hedvat, M. F. Berger, L. Pusztai, M.

Lechner, C. Boshoff, M. Jarosz, C. Vietz, A. Parker, V. A. Miller, J. S. Ross, J. Curran, M. T. Cronin, P.

J. Stephens, D. Lipson and R. Yelensky (2013). "Development and validation of a clinical cancer

genomic profiling test based on massively parallel DNA sequencing." Nat Biotechnol 31(11): 1023-1031.

Franchitto, A., P. Pichierri, R. Piergentili, M. Crescenzi, M. Bignami and F. Palitti (2003). "The

mammalian mismatch repair protein MSH2 is required for correct MRE11 and RAD51 relocalization and

for efficient cell cycle arrest induced by ionizing radiation in G2 phase." Oncogene 22(14): 2110-2120.

Fridman, W. H., F. Pagès, C. Sautès-Fridman and J. Galon (2012). "The immune contexture in human

tumours: impact on clinical outcome." Nat Rev Cancer 12(4): 298-306.

Fritzell, J. A., L. Narayanan, S. M. Baker, C. E. Bronner, S. E. Andrew, T. A. Prolla, A. Bradley, F. R.

Jirik, R. M. Liskay and P. M. Glazer (1997). "Role of DNA mismatch repair in the cytotoxicity of

ionizing radiation." Cancer Res 57(22): 5143-5147.

Fuchs, C. S., T. Doi, R. W. Jang, K. Muro, T. Satoh, M. Machado, W. Sun, S. I. Jalal, M. A. Shah, J. P.

Metges, M. Garrido, T. Golan, M. Mandala, Z. A. Wainberg, D. V. Catenacci, A. Ohtsu, K. Shitara, R.

Geva, J. Bleeker, A. H. Ko, G. Ku, P. Philip, P. C. Enzinger, Y. J. Bang, D. Levitan, J. Wang, M. Rosales,

R. P. Dalal and H. H. Yoon (2018). "Safety and Efficacy of Pembrolizumab Monotherapy in Patients

With Previously Treated Advanced Gastric and Gastroesophageal Junction Cancer: Phase 2 Clinical

KEYNOTE-059 Trial." JAMA Oncol 4(5): e180013.

Ganai, Rais A. and E. Johansson (2016). "DNA Replication—A Matter of Fidelity." Molecular Cell

62(5): 745-755.

Garbacz, M. A., S. A. Lujan, A. B. Burkholder, P. B. Cox, Q. Wu, Z. X. Zhou, J. E. Haber and T. A.

Kunkel (2018). "Evidence that DNA polymerase δ contributes to initiating leading strand DNA

replication in Saccharomyces cerevisiae." Nat Commun 9(1): 858.

Gargiulo, P., C. Della Pepa, S. Berardi, D. Califano, S. Scala, L. Buonaguro, G. Ciliberto, P. Brauchli and

S. Pignata (2016). "Tumor genotype and immune microenvironment in POLE-ultramutated and MSI-

hypermutated Endometrial Cancers: New candidates for checkpoint blockade immunotherapy?" Cancer

Treat Rev 48: 61-68.

Garon, E. B., N. A. Rizvi, R. Hui, N. Leighl, A. S. Balmanoukian, J. P. Eder, A. Patnaik, C. Aggarwal,

M. Gubens, L. Horn, E. Carcereny, M. J. Ahn, E. Felip, J. S. Lee, M. D. Hellmann, O. Hamid, J. W.

Goldman, J. C. Soria, M. Dolled-Filhart, R. Z. Rutledge, J. Zhang, J. K. Lunceford, R. Rangwala, G. M.

158

Lubiniecki, C. Roach, K. Emancipator and L. Gandhi (2015). "Pembrolizumab for the treatment of non-

small-cell lung cancer." N Engl J Med 372(21): 2018-2028.

Gengenbacher, N., M. Singhal and H. G. Augustin (2017). "Preclinical mouse solid tumour models: status

quo, challenges and perspectives." Nat Rev Cancer 17(12): 751-765.

Geoerger, B., H. J. Kang, M. Yalon-Oren, L. V. Marshall, C. Vezina, A. Pappo, T. W. Laetsch, A. S.

Petrilli, M. Ebinger, J. Toporski, J. Glade-Bender, W. Nicholls, E. Fox, S. G. DuBois, M. E. Macy, S. L.

Cohn, K. Pathiraja, S. J. Diede, S. Ebbinghaus and N. Pinto (2020). "Pembrolizumab in paediatric

patients with advanced melanoma or a PD-L1-positive, advanced, relapsed, or refractory solid tumour or

lymphoma (KEYNOTE-051): interim analysis of an open-label, single-arm, phase 1-2 trial." Lancet

Oncol 21(1): 121-133.

Germano, G., S. Lamba, G. Rospo, L. Barault, A. Magrì, F. Maione, M. Russo, G. Crisafulli, A. Bartolini,

G. Lerda, G. Siravegna, B. Mussolin, R. Frapolli, M. Montone, F. Morano, F. de Braud, N. Amirouchene-

Angelozzi, S. Marsoni, M. D'Incalci, A. Orlandi, E. Giraudo, A. Sartore-Bianchi, S. Siena, F.

Pietrantonio, F. Di Nicolantonio and A. Bardelli (2017). "Inactivation of DNA repair triggers neoantigen

generation and impairs tumour growth." Nature 552(7683): 116-120.

Gertsenstein, M. and L. M. J. Nutter (2018). "Engineering Point Mutant and Epitope-Tagged Alleles in

Mice Using Cas9 RNA-Guided Nuclease." Current Protocols in Mouse Biology 8(1): 28-53.

Goldsby, R. E., L. E. Hays, X. Chen, E. A. Olmsted, W. B. Slayton, G. J. Spangrude and B. D. Preston

(2002). "High incidence of epithelial cancers in mice deficient for DNA polymerase delta proofreading."

Proc Natl Acad Sci U S A 99(24): 15560-15565.

Goldsby, R. E., N. A. Lawrence, L. E. Hays, E. A. Olmsted, X. Chen, M. Singh and B. D. Preston (2001).

"Defective DNA polymerase-delta proofreading causes cancer susceptibility in mice." Nat Med 7(6): 638-

639.

Grilley, M., K. M. Welsh, S. S. Su and P. Modrich (1989). "Isolation and characterization of the

Escherichia coli mutL gene product." J Biol Chem 264(2): 1000-1004.

Guan, J., C. Lu, Q. Jin, H. Lu, X. Chen, L. Tian, Y. Zhang, J. Ortega, J. Zhang, S. Siteni, M. Chen, L. Gu,

J. W. Shay, A. J. Davis, Z. J. Chen, Y. X. Fu and G. M. Li (2021). "MLH1 Deficiency-Triggered DNA

Hyperexcision by Exonuclease 1 Activates the cGAS-STING Pathway." Cancer Cell 39(1): 109-

121.e105.

Gubin, M. M. and R. D. Schreiber (2015). "CANCER. The odds of immunotherapy success." Science

350(6257): 158-159.

Guerreiro Stucklin, A. S., S. Ryall, K. Fukuoka, M. Zapotocky, A. Lassaletta, C. Li, T. Bridge, B. Kim,

A. Arnoldo, P. E. Kowalski, Y. Zhong, M. Johnson, C. Li, A. K. Ramani, R. Siddaway, L. F. Nobre, P. de

Antonellis, C. Dunham, S. Cheng, D. R. Boué, J. L. Finlay, S. L. Coven, I. de Prada, M. Perez-Somarriba,

C. C. Faria, M. A. Grotzer, E. Rushing, D. Sumerauer, J. Zamecnik, L. Krskova, M. Garcia Ariza, O.

Cruz, A. Morales La Madrid, P. Solano, K. Terashima, Y. Nakano, K. Ichimura, M. Nagane, H.

Sakamoto, M. J. Gil-da-Costa, R. Silva, D. L. Johnston, J. Michaud, B. Wilson, F. K. H. van Landeghem,

A. Oviedo, P. D. McNeely, B. Crooks, I. Fried, N. Zhukova, J. R. Hansford, A. Nageswararao, L. Garzia,

M. Shago, M. Brudno, M. S. Irwin, U. Bartels, V. Ramaswamy, E. Bouffet, M. D. Taylor, U. Tabori and

C. Hawkins (2019). "Alterations in ALK/ROS1/NTRK/MET drive a group of infantile hemispheric

gliomas." Nat Commun 10(1): 4343.

Guidos, C. J., I. L. Weissman and B. Adkins (1989). "Intrathymic maturation of murine T lymphocytes

from CD8+ precursors." Proc Natl Acad Sci U S A 86(19): 7542-7546.

Hamzaoui, N., F. Alarcon, N. Leulliot, R. Guimbaud, B. Buecher, C. Colas, C. Corsini, G. Nuel, B.

Terris, P. Laurent-Puig, S. Chaussade, M. Dhooge, C. Madru and E. Clauser (2020). "Genetic, structural,

159

and functional characterization of POLE polymerase proofreading variants allows cancer risk prediction."

Genetics in Medicine 22(9): 1533-1541.

Hanahan, D. and R. A. Weinberg (2011). "Hallmarks of cancer: the next generation." Cell 144(5): 646-

674.

Haradhvala, N. J., J. Kim, Y. E. Maruvka, P. Polak, D. Rosebrock, D. Livitz, J. M. Hess, I. Leshchiner, A.

Kamburov, K. W. Mouw, M. S. Lawrence and G. Getz (2018). "Distinct mutational signatures

characterize concurrent loss of polymerase proofreading and mismatch repair." Nature Communications

9(1): 1746.

Haraldsdottir, S., H. Hampel, J. Tomsic, W. L. Frankel, R. Pearlman, A. de la Chapelle and C. C.

Pritchard (2014). "Colon and Endometrial Cancers With Mismatch Repair Deficiency Can Arise From

Somatic, Rather Than Germline, Mutations." Gastroenterology 147(6): 1308-1316.e1301.

Hellmann, M. D., T. E. Ciuleanu, A. Pluzanski, J. S. Lee, G. A. Otterson, C. Audigier-Valette, E.

Minenza, H. Linardou, S. Burgers, P. Salman, H. Borghaei, S. S. Ramalingam, J. Brahmer, M. Reck, K. J.

O'Byrne, W. J. Geese, G. Green, H. Chang, J. Szustakowski, P. Bhagavatheeswaran, D. Healey, Y. Fu, F.

Nathan and L. Paz-Ares (2018). "Nivolumab plus Ipilimumab in Lung Cancer with a High Tumor

Mutational Burden." N Engl J Med 378(22): 2093-2104.

Hellmann, M. D., T. W. Kim, C. B. Lee, B. C. Goh, W. H. Miller, Jr., D. Y. Oh, R. Jamal, C. E. Chee, L.

Q. M. Chow, J. F. Gainor, J. Desai, B. J. Solomon, M. Das Thakur, B. Pitcher, P. Foster, G. Hernandez,

M. J. Wongchenko, E. Cha, Y. J. Bang, L. L. Siu and J. Bendell (2019). "Phase Ib study of atezolizumab

combined with cobimetinib in patients with solid tumors." Ann Oncol 30(7): 1134-1142.

Hemminki, A., P. Peltomäki, J. P. Mecklin, H. Järvinen, R. Salovaara, M. Nyström-Lahti, A. de la

Chapelle and L. A. Aaltonen (1994). "Loss of the wild type MLH1 gene is a feature of hereditary

nonpolyposis colorectal cancer." Nat Genet 8(4): 405-410.

Henninger, E. E. and Z. F. Pursell (2014). "DNA polymerase ε and its roles in genome stability." IUBMB

Life 66(5): 339-351.

Herbst, R. S., P. Baas, D. W. Kim, E. Felip, J. L. Pérez-Gracia, J. Y. Han, J. Molina, J. H. Kim, C. D.

Arvis, M. J. Ahn, M. Majem, M. J. Fidler, G. de Castro, Jr., M. Garrido, G. M. Lubiniecki, Y. Shentu, E.

Im, M. Dolled-Filhart and E. B. Garon (2016). "Pembrolizumab versus docetaxel for previously treated,

PD-L1-positive, advanced non-small-cell lung cancer (KEYNOTE-010): a randomised controlled trial."

Lancet 387(10027): 1540-1550.

Herr, A. J., S. R. Kennedy, G. M. Knowels, E. M. Schultz and B. D. Preston (2014). "DNA Replication

Error-Induced Extinction of Diploid Yeast." Genetics 196(3): 677-691.

Hidalgo, M., F. Amant, A. V. Biankin, E. Budinská, A. T. Byrne, C. Caldas, R. B. Clarke, S. de Jong, J.

Jonkers, G. M. Mælandsmo, S. Roman-Roman, J. Seoane, L. Trusolino and A. Villanueva (2014).

"Patient-derived xenograft models: an emerging platform for translational cancer research." Cancer

Discov 4(9): 998-1013.

Hitchins, M., R. Williams, K. Cheong, N. Halani, V. A. Lin, D. Packham, S. Ku, A. Buckle, N. Hawkins,

J. Burn, S. Gallinger, J. Goldblatt, J. Kirk, I. Tomlinson, R. Scott, A. Spigelman, C. Suter, D. Martin, G.

Suthers and R. Ward (2005). "MLH1 germline epimutations as a factor in hereditary nonpolyposis

colorectal cancer." Gastroenterology 129(5): 1392-1399.

Hitchins, M. P., R. W. Rapkins, C. T. Kwok, S. Srivastava, J. J. Wong, L. M. Khachigian, P. Polly, J.

Goldblatt and R. L. Ward (2011). "Dominantly inherited constitutional epigenetic silencing of MLH1 in a

cancer-affected family is linked to a single nucleotide variant within the 5'UTR." Cancer Cell 20(2): 200-

213.

160

Hodel, K. P., R. de Borja, E. E. Henninger, B. B. Campbell, N. Ungerleider, N. Light, T. Wu, K. G.

LeCompte, A. Y. Goksenin, B. A. Bunnell, U. Tabori, A. Shlien and Z. F. Pursell (2018). "Explosive

mutation accumulation triggered by heterozygous human Pol epsilon proofreading-deficiency is driven by

suppression of mismatch repair." Elife 7.

Hodi, F. S., S. J. O'Day, D. F. McDermott, R. W. Weber, J. A. Sosman, J. B. Haanen, R. Gonzalez, C.

Robert, D. Schadendorf, J. C. Hassel, W. Akerley, A. J. van den Eertwegh, J. Lutzky, P. Lorigan, J. M.

Vaubel, G. P. Linette, D. Hogg, C. H. Ottensmeier, C. Lebbé, C. Peschel, I. Quirt, J. I. Clark, J. D.

Wolchok, J. S. Weber, J. Tian, M. J. Yellin, G. M. Nichol, A. Hoos and W. J. Urba (2010). "Improved

survival with ipilimumab in patients with metastatic melanoma." N Engl J Med 363(8): 711-723.

Hogg, M., P. Osterman, G. O. Bylund, R. A. Ganai, E. B. Lundstrom, A. E. Sauer-Eriksson and E.

Johansson (2014). "Structural basis for processive DNA synthesis by yeast DNA polymerase varepsilon."

Nat Struct Mol Biol 21(1): 49-55.

Humbert, O., S. Fiumicino, G. Aquilina, P. Branch, S. Oda, A. Zijno, P. Karran and M. Bignami (1999).

"Mismatch repair and differential sensitivity of mouse and human cells to methylating agents."

Carcinogenesis 20(2): 205-214.

Hunter, C., R. Smith, D. P. Cahill, P. Stephens, C. Stevens, J. Teague, C. Greenman, S. Edkins, G.

Bignell, H. Davies, S. O'Meara, A. Parker, T. Avis, S. Barthorpe, L. Brackenbury, G. Buck, A. Butler, J.

Clements, J. Cole, E. Dicks, S. Forbes, M. Gorton, K. Gray, K. Halliday, R. Harrison, K. Hills, J. Hinton,

A. Jenkinson, D. Jones, V. Kosmidou, R. Laman, R. Lugg, A. Menzies, J. Perry, R. Petty, K. Raine, D.

Richardson, R. Shepherd, A. Small, H. Solomon, C. Tofts, J. Varian, S. West, S. Widaa, A. Yates, D. F.

Easton, G. Riggins, J. E. Roy, K. K. Levine, W. Mueller, T. T. Batchelor, D. N. Louis, M. R. Stratton, P.

A. Futreal and R. Wooster (2006). "A hypermutation phenotype and somatic MSH6 mutations in

recurrent human malignant gliomas after alkylator chemotherapy." Cancer Res 66(8): 3987-3991.

Itoh, H. and K. Ohsato (1985). "Turcot syndrome and its characteristic colonic manifestations." Dis Colon

Rectum 28(6): 399-402.

Jabbour, E., J. Düll, M. Yilmaz, J. D. Khoury, F. Ravandi, N. Jain, H. Einsele, G. Garcia-Manero, M.

Konopleva, N. J. Short, P. A. Thompson, W. Wierda, N. Daver, J. Cortes, S. O'Brien, H. Kantarjian and

M. S. Topp (2018). "Outcome of patients with relapsed/refractory acute lymphoblastic leukemia after

blinatumomab failure: No change in the level of CD19 expression." Am J Hematol 93(3): 371-374.

Johanns, T. M., C. A. Miller, I. G. Dorward, C. Tsien, E. Chang, A. Perry, R. Uppaluri, C. Ferguson, R.

E. Schmidt, S. Dahiya, G. Ansstas, E. R. Mardis and G. P. Dunn (2016). "Immunogenomics of

Hypermutated Glioblastoma: A Patient with Germline <em>POLE</em> Deficiency Treated with

Checkpoint Blockade Immunotherapy." Cancer Discovery 6(11): 1230-1236.

Johanns, T. M., C. A. Miller, I. G. Dorward, C. Tsien, E. Chang, A. Perry, R. Uppaluri, C. Ferguson, R.

E. Schmidt, S. Dahiya, G. Ansstas, E. R. Mardis and G. P. Dunn (2016). "Immunogenomics of

Hypermutated Glioblastoma: A Patient with Germline POLE Deficiency Treated with Checkpoint

Blockade Immunotherapy." Cancer Discov 6(11): 1230-1236.

Kadyrov, F. A., L. Dzantiev, N. Constantin and P. Modrich (2006). "Endonucleolytic function of

MutLalpha in human mismatch repair." Cell 126(2): 297-308.

Kandoth, C., M. D. McLellan, F. Vandin, K. Ye, B. Niu, C. Lu, M. Xie, Q. Zhang, J. F. McMichael, M.

A. Wyczalkowski, M. D. M. Leiserson, C. A. Miller, J. S. Welch, M. J. Walter, M. C. Wendl, T. J. Ley,

R. K. Wilson, B. J. Raphael and L. Ding (2013). "Mutational landscape and significance across 12 major

cancer types." Nature 502(7471): 333-339.

Kandoth, C., N. Schultz, A. D. Cherniack, R. Akbani, Y. Liu, H. Shen, A. G. Robertson, I. Pashtan, R.

Shen, C. C. Benz, C. Yau, P. W. Laird, L. Ding, W. Zhang, G. B. Mills, R. Kucherlapati, E. R. Mardis

161

and D. A. Levine (2013). "Integrated genomic characterization of endometrial carcinoma." Nature

497(7447): 67-73.

Karran, P. and M. G. Marinus (1982). "Mismatch correction at O6-methylguanine residues in E. coli

DNA." Nature 296(5860): 868-869.

Kategaya, L., S. K. Perumal, J. H. Hager and L. D. Belmont (2019). "Werner Syndrome Helicase Is

Required for the Survival of Cancer Cells with Microsatellite Instability." iScience 13: 488-497.

Kaufman, H. L., J. Russell, O. Hamid, S. Bhatia, P. Terheyden, S. P. D'Angelo, K. C. Shih, C. Lebbé, G.

P. Linette, M. Milella, I. Brownell, K. D. Lewis, J. H. Lorch, K. Chin, L. Mahnke, A. von Heydebreck, J.

M. Cuillerot and P. Nghiem (2016). "Avelumab in patients with chemotherapy-refractory metastatic

Merkel cell carcinoma: a multicentre, single-group, open-label, phase 2 trial." Lancet Oncol 17(10): 1374-

1385.

Keysselt, K., T. Kreutzmann, K. Rother, C. Kerner, K. Krohn, J. Przybilla, P. Buske, H. Löffler-Wirth,

M. Loeffler, J. Galle and G. Aust (2017). "Different in vivo and in vitro transformation of intestinal stem

cells in mismatch repair deficiency." Oncogene 36(19): 2750-2761.

Klinger, M., C. Brandl, G. Zugmaier, Y. Hijazi, R. C. Bargou, M. S. Topp, N. Gökbuget, S. Neumann, M.

Goebeler, A. Viardot, M. Stelljes, M. Brüggemann, D. Hoelzer, E. Degenhard, D. Nagorsen, P. A.

Baeuerle, A. Wolf and P. Kufer (2012). "Immunopharmacologic response of patients with B-lineage acute

lymphoblastic leukemia to continuous infusion of T cell-engaging CD19/CD3-bispecific BiTE antibody

blinatumomab." Blood 119(26): 6226-6233.

Korona, D. A., K. G. Lecompte and Z. F. Pursell (2011). "The high fidelity and unique error signature of

human DNA polymerase epsilon." Nucleic Acids Res 39(5): 1763-1773.

Kovalenko, M., E. Dragileva, J. St Claire, T. Gillis, J. R. Guide, J. New, H. Dong, R. Kucherlapati, M. H.

Kucherlapati, M. E. Ehrlich, J. M. Lee and V. C. Wheeler (2012). "Msh2 acts in medium-spiny striatal

neurons as an enhancer of CAG instability and mutant huntingtin phenotypes in Huntington's disease

knock-in mice." PLoS One 7(9): e44273.

Krummel, M. F. and J. P. Allison (1995). "CD28 and CTLA-4 have opposing effects on the response of T

cells to stimulation." J Exp Med 182(2): 459-465.

Kucherlapati, M. H., S. Esfahani, P. Habibollahi, J. Wang, E. R. Still, R. T. Bronson, U. Mahmood and R.

S. Kucherlapati (2013). "Genotype directed therapy in murine mismatch repair deficient tumors." PLoS

One 8(7): e68817.

Kucherlapati, M. H., K. Lee, A. A. Nguyen, A. B. Clark, H. Hou, Jr., A. Rosulek, H. Li, K. Yang, K. Fan,

M. Lipkin, R. T. Bronson, L. Jelicks, T. A. Kunkel, R. Kucherlapati and W. Edelmann (2010). "An Msh2

conditional knockout mouse for studying intestinal cancer and testing anticancer agents."

Gastroenterology 138(3): 993-1002.e1001.

Kunkel, T. A. and D. A. Erie (2015). "Eukaryotic Mismatch Repair in Relation to DNA Replication."

Annu Rev Genet 49: 291-313.

Kunkel, T. A. and L. A. Loeb (1981). "Fidelity of mammalian DNA polymerases." Science 213(4509):

765-767.

Kyi, C. and M. A. Postow (2016). "Immune checkpoint inhibitor combinations in solid tumors:

opportunities and challenges." Immunotherapy 8(7): 821-837.

Lahue, R. S., K. G. Au and P. Modrich (1989). "DNA mismatch correction in a defined system." Science

245(4914): 160-164.

162

Lamers, M. H., A. Perrakis, J. H. Enzlin, H. H. Winterwerp, N. de Wind and T. K. Sixma (2000). "The

crystal structure of DNA mismatch repair protein MutS binding to a G x T mismatch." Nature 407(6805):

711-717.

Lange, S. S., K. Takata and R. D. Wood (2011). "DNA polymerases and cancer." Nat Rev Cancer 11(2):

96-110.

Längle-Rouault, F., G. Maenhaut-Michel and M. Radman (1987). "GATC sequences, DNA nicks and the

MutH function in Escherichia coli mismatch repair." Embo j 6(4): 1121-1127.

Larkin, J., V. Chiarion-Sileni, R. Gonzalez, J. J. Grob, C. L. Cowey, C. D. Lao, D. Schadendorf, R.

Dummer, M. Smylie, P. Rutkowski, P. F. Ferrucci, A. Hill, J. Wagstaff, M. S. Carlino, J. B. Haanen, M.

Maio, I. Marquez-Rodas, G. A. McArthur, P. A. Ascierto, G. V. Long, M. K. Callahan, M. A. Postow, K.

Grossmann, M. Sznol, B. Dreno, L. Bastholt, A. Yang, L. M. Rollin, C. Horak, F. S. Hodi and J. D.

Wolchok (2015). "Combined Nivolumab and Ipilimumab or Monotherapy in Untreated Melanoma." N

Engl J Med 373(1): 23-34.

Lavoine, N., C. Colas, M. Muleris, S. Bodo, A. Duval, N. Entz-Werle, F. Coulet, O. Cabaret, F.

Andreiuolo, C. Charpy, G. Sebille, Q. Wang, S. Lejeune, M. P. Buisine, D. Leroux, G. Couillault, G.

Leverger, J. P. Fricker, R. Guimbaud, M. Mathieu-Dramard, G. Jedraszak, O. Cohen-Hagenauer, L.

Guerrini-Rousseau, F. Bourdeaut, J. Grill, O. Caron, S. Baert-Dusermont, J. Tinat, G. Bougeard, T.

Frébourg and L. Brugières (2015). "Constitutional mismatch repair deficiency syndrome: clinical

description in a French cohort." J Med Genet 52(11): 770-778.

Lawrence, M. S., P. Stojanov, C. H. Mermel, J. T. Robinson, L. A. Garraway, T. R. Golub, M. Meyerson,

S. B. Gabriel, E. S. Lander and G. Getz (2014). "Discovery and saturation analysis of cancer genes across

21 tumour types." Nature 505(7484): 495-501.

Le, D. T., J. N. Durham, K. N. Smith, H. Wang, B. R. Bartlett, L. K. Aulakh, S. Lu, H. Kemberling, C.

Wilt, B. S. Luber, F. Wong, N. S. Azad, A. A. Rucki, D. Laheru, R. Donehower, A. Zaheer, G. A. Fisher,

T. S. Crocenzi, J. J. Lee, T. F. Greten, A. G. Duffy, K. K. Ciombor, A. D. Eyring, B. H. Lam, A. Joe, S.

P. Kang, M. Holdhoff, L. Danilova, L. Cope, C. Meyer, S. Zhou, R. M. Goldberg, D. K. Armstrong, K.

M. Bever, A. N. Fader, J. Taube, F. Housseau, D. Spetzler, N. Xiao, D. M. Pardoll, N. Papadopoulos, K.

W. Kinzler, J. R. Eshleman, B. Vogelstein, R. A. Anders and L. A. Diaz, Jr. (2017). "Mismatch repair

deficiency predicts response of solid tumors to PD-1 blockade." Science 357(6349): 409-413.

Le, D. T., T. W. Kim, E. V. Cutsem, R. Geva, D. Jäger, H. Hara, M. Burge, B. O’Neil, P. Kavan, T.

Yoshino, R. Guimbaud, H. Taniguchi, E. Elez, S.-E. Al-Batran, P. M. Boland, T. Crocenzi, C. E. Atreya,

Y. Cui, T. Dai, P. Marinello, L. A. D. Jr and T. André (2020). "Phase II Open-Label Study of

Pembrolizumab in Treatment-Refractory, Microsatellite Instability–High/Mismatch Repair–Deficient

Metastatic Colorectal Cancer: KEYNOTE-164." Journal of Clinical Oncology 38(1): 11-19.

Le, D. T., J. N. Uram, H. Wang, B. R. Bartlett, H. Kemberling, A. D. Eyring, A. D. Skora, B. S. Luber, N.

S. Azad, D. Laheru, B. Biedrzycki, R. C. Donehower, A. Zaheer, G. A. Fisher, T. S. Crocenzi, J. J. Lee,

S. M. Duffy, R. M. Goldberg, A. de la Chapelle, M. Koshiji, F. Bhaijee, T. Huebner, R. H. Hruban, L. D.

Wood, N. Cuka, D. M. Pardoll, N. Papadopoulos, K. W. Kinzler, S. Zhou, T. C. Cornish, J. M. Taube, R.

A. Anders, J. R. Eshleman, B. Vogelstein and L. A. Diaz, Jr. (2015). "PD-1 Blockade in Tumors with

Mismatch-Repair Deficiency." N Engl J Med 372(26): 2509-2520.

Leach, F. S., N. C. Nicolaides, N. Papadopoulos, B. Liu, J. Jen, R. Parsons, P. Peltomäki, P. Sistonen, L.

A. Aaltonen, M. Nyström-Lahti and et al. (1993). "Mutations of a mutS homolog in hereditary

nonpolyposis colorectal cancer." Cell 75(6): 1215-1225.

Lee, K., E. Tosti and W. Edelmann (2016). "Mouse models of DNA mismatch repair in cancer research."

DNA Repair (Amst) 38: 140-146.

163

Lehman, I. R., M. J. Bessman, E. S. Simms and A. Kornberg (1958). "Enzymatic synthesis of

deoxyribonucleic acid. I. Preparation of substrates and partial purification of an enzyme from Escherichia

coli." J Biol Chem 233(1): 163-170.

León-Castillo, A., H. Britton, M. K. McConechy, J. N. McAlpine, R. Nout, S. Kommoss, S. Y. Brucker,

J. W. Carlson, E. Epstein, T. T. Rau, T. Bosse, D. N. Church and C. B. Gilks (2020). "Interpretation of

somatic POLE mutations in endometrial carcinoma." The Journal of Pathology 250(3): 323-335.

Levati, L., G. Marra, T. Lettieri, S. D'Atri, P. Vernole, L. Tentori, P. M. Lacal, E. Pagani, E. Bonmassar,

J. Jiricny and G. Graziani (1998). "Mutation of the mismatch repair gene hMSH2 and hMSH6 in a human

T-cell leukemia line tolerant to methylating agents." Genes Chromosomes Cancer 23(2): 159-166.

Levine, D. A., N. The Cancer Genome Atlas Research, G. Getz, S. B. Gabriel, K. Cibulskis, E. Lander, A.

Sivachenko, C. Sougnez, M. Lawrence, C. Kandoth, D. Dooling, R. Fulton, L. Fulton, J. Kalicki-Veizer,

M. D. McLellan, M. O’Laughlin, H. Schmidt, R. K. Wilson, K. Ye, L. Ding, E. R. Mardis, A. Ally, M.

Balasundaram, I. Birol, Y. S. N. Butterfield, R. Carlsen, C. Carter, A. Chu, E. Chuah, H.-J. E. Chun, N.

Dhalla, R. Guin, C. Hirst, R. A. Holt, S. J. M. Jones, D. Lee, H. I. Li, M. A. Marra, M. Mayo, R. A.

Moore, A. J. Mungall, P. Plettner, J. E. Schein, P. Sipahimalani, A. Tam, R. J. Varhol, A. Gordon

Robertson, A. D. Cherniack, I. Pashtan, G. Saksena, R. C. Onofrio, S. E. Schumacher, B. Tabak, S. L.

Carter, B. Hernandez, J. Gentry, H. B. Salvesen, K. Ardlie, G. Getz, W. Winckler, R. Beroukhim, S. B.

Gabriel, M. Meyerson, A. Hadjipanayis, S. Lee, H. S. Mahadeshwar, P. Park, A. Protopopov, X. Ren, S.

Seth, X. Song, J. Tang, R. Xi, L. Yang, D. Zeng, R. Kucherlapati, L. Chin, J. Zhang, J. Todd Auman, S.

Balu, T. Bodenheimer, E. Buda, D. Neil Hayes, A. P. Hoyle, S. R. Jefferys, C. D. Jones, S. Meng, P. A.

Mieczkowski, L. E. Mose, J. S. Parker, C. M. Perou, J. Roach, Y. Shi, J. V. Simons, M. G. Soloway, D.

Tan, M. D. Topal, S. Waring, J. Wu, K. A. Hoadley, S. B. Baylin, M. S. Bootwalla, P. H. Lai, T. J. Triche

Jr, D. J. Van Den Berg, D. J. Weisenberger, P. W. Laird, H. Shen, L. Chin, J. Zhang, G. Getz, J. Cho, D.

DiCara, S. Frazer, D. Heiman, R. Jing, P. Lin, W. Mallard, P. Stojanov, D. Voet, H. Zhang, L. Zou, M.

Noble, M. Lawrence, S. M. Reynolds, I. Shmulevich, B. Arman Aksoy, Y. Antipin, G. Ciriello, G.

Dresdner, J. Gao, B. Gross, A. Jacobsen, M. Ladanyi, B. Reva, C. Sander, R. Sinha, S. Onur Sumer, B. S.

Taylor, E. Cerami, N. Weinhold, N. Schultz, R. Shen, S. Benz, T. Goldstein, D. Haussler, S. Ng, C. Szeto,

J. Stuart, C. C. Benz, C. Yau, W. Zhang, M. Annala, B. M. Broom, T. D. Casasent, Z. Ju, H. Liang, G.

Liu, Y. Lu, A. K. Unruh, C. Wakefield, J. N. Weinstein, N. Zhang, Y. Liu, R. Broaddus, R. Akbani, G. B.

Mills, C. Adams, T. Barr, A. D. Black, J. Bowen, J. Deardurff, J. Frick, J. M. Gastier-Foster, T.

Grossman, H. A. Harper, M. Hart-Kothari, C. Helsel, A. Hobensack, H. Kuck, K. Kneile, K. M. Leraas,

T. M. Lichtenberg, C. McAllister, R. E. Pyatt, N. C. Ramirez, T. R. Tabler, N. Vanhoose, P. White, L.

Wise, E. Zmuda, N. Barnabas, C. Berry-Green, V. Blanc, L. Boice, M. Button, A. Farkas, A. Green, J.

MacKenzie, D. Nicholson, S. E. Kalloger, C. Blake Gilks, B. Y. Karlan, J. Lester, S. Orsulic, M.

Borowsky, M. Cadungog, C. Czerwinski, L. Huelsenbeck-Dill, M. Iacocca, N. Petrelli, B. Rabeno, G.

Witkin, E. Nemirovich-Danchenko, O. Potapova, D. Rotin, A. Berchuck, M. Birrer, P. DiSaia, L.

Monovich, E. Curley, J. Gardner, D. Mallery, R. Penny, S. C. Dowdy, B. Winterhoff, L. Dao, B. Gostout,

A. Meuter, A. Teoman, F. Dao, N. Olvera, F. Bogomolniy, K. Garg, R. A. Soslow, D. A. Levine, M.

Abramov, J. M. S. Bartlett, S. Kodeeswaran, J. Parfitt, F. Moiseenko, B. A. Clarke, M. T. Goodman, M.

E. Carney, R. K. Matsuno, J. Fisher, M. Huang, W. Kimryn Rathmell, L. Thorne, L. Van Le, R. Dhir, R.

Edwards, E. Elishaev, K. Zorn, R. Broaddus, P. J. Goodfellow, D. Mutch, N. Schultz, Y. Liu, R. Akbani,

A. D. Cherniack, E. Cerami, N. Weinhold, H. Shen, K. A. Hoadley, A. B. Kahn, D. W. Bell, P. M.

Pollock, C. Wang, D. A.Wheeler, E. Shinbrot, B. Y. Karlan, A. Berchuck, S. C. Dowdy, B. Winterhoff,

M. T. Goodman, A. Gordon Robertson, R. Beroukhim, I. Pashtan, H. B. Salvesen, P. W. Laird, M. Noble,

J. Stuart, L. Ding, C. Kandoth, C. Blake Gilks, R. A. Soslow, P. J. Goodfellow, D. Mutch, R. Broaddus,

W. Zhang, G. B. Mills, R. Kucherlapati, E. R. Mardis, D. A. Levine, B. Ayala, A. L. Chu, M. A. Jensen,

P. Kothiyal, T. D. Pihl, J. Pontius, D. A. Pot, E. E. Snyder, D. Srinivasan, A. B. Kahn, K. R. Mills Shaw,

M. Sheth, T. Davidsen, G. Eley Martin L. Ferguson, J. A. Demchok, L. Yang, M. S. Guyer, B. A.

Ozenberger, H. J. Sofia, C. Kandoth, N. Schultz, A. D. Cherniack, R. Akbani, Y. Liu, H. Shen, A. Gordon

164

Robertson, I. Pashtan, R. Shen, C. C. Benz, C. Yau, P. W. Laird, L. Ding, W. Zhang, G. B. Mills, R.

Kucherlapati, E. R. Mardis and D. A. Levine (2013). "Integrated genomic characterization of endometrial

carcinoma." Nature 497: 67.

Levinson, G. and G. A. Gutman (1987). "High frequencies of short frameshifts in poly-CA/TG tandem

repeats borne by bacteriophage M13 in Escherichia coli K-12." Nucleic Acids Res 15(13): 5323-5338.

Li, H.-D., C. Lu, H. Zhang, Q. Hu, J. Zhang, I. C. Cuevas, S. S. Sahoo, M. Aguilar, E. G. Maurais, S.

Zhang, X. Wang, E. A. Akbay, G.-M. Li, B. Li, P. Koduru, P. Ly, Y.-X. Fu and D. H. Castrillon (2020).

"A PoleP286R mouse model of endometrial cancer recapitulates high mutational burden and

immunotherapy response." JCI Insight 5(14).

Li, H. D., I. Cuevas, M. Zhang, C. Lu, M. M. Alam, Y. X. Fu, M. J. You, E. A. Akbay, H. Zhang and D.

H. Castrillon (2018). "Polymerase-mediated ultramutagenesis in mice produces diverse cancers with high

mutational load." J Clin Invest 128(9): 4179-4191.

Liberzon, A., C. Birger, H. Thorvaldsdóttir, M. Ghandi, Jill P. Mesirov and P. Tamayo (2015). "The

Molecular Signatures Database Hallmark Gene Set Collection." Cell Systems 1(6): 417-425.

Lieb, S., S. Blaha-Ostermann, E. Kamper, J. Rippka, C. Schwarz, K. Ehrenhöfer-Wölfer, A. Schlattl, A.

Wernitznig, J. J. Lipp, K. Nagasaka, P. van der Lelij, G. Bader, M. Koi, A. Goel, R. A. Neumüller, J. M.

Peters, N. Kraut, M. A. Pearson, M. Petronczki and S. Wöhrle (2019). "Werner syndrome helicase is a

selective vulnerability of microsatellite instability-high tumor cells." Elife 8.

Ligtenberg, M. J., R. P. Kuiper, T. L. Chan, M. Goossens, K. M. Hebeda, M. Voorendt, T. Y. Lee, D.

Bodmer, E. Hoenselaar, S. J. Hendriks-Cornelissen, W. Y. Tsui, C. K. Kong, H. G. Brunner, A. G. van

Kessel, S. T. Yuen, J. H. van Krieken, S. Y. Leung and N. Hoogerbrugge (2009). "Heritable somatic

methylation and inactivation of MSH2 in families with Lynch syndrome due to deletion of the 3' exons of

TACSTD1." Nat Genet 41(1): 112-117.

Lin, G. L., K. M. Wilson, M. Ceribelli, B. Z. Stanton, P. J. Woo, S. Kreimer, E. Y. Qin, X. Zhang, J.

Lennon, S. Nagaraja, P. J. Morris, M. Quezada, S. M. Gillespie, D. Y. Duveau, A. M. Michalowski, P.

Shinn, R. Guha, M. Ferrer, C. Klumpp-Thomas, S. Michael, C. McKnight, P. Minhas, Z. Itkin, E. H.

Raabe, L. Chen, R. Ghanem, A. C. Geraghty, L. Ni, K. I. Andreasson, N. A. Vitanza, K. E. Warren, C. J.

Thomas and M. Monje (2019). "Therapeutic strategies for diffuse midline glioma from high-throughput

combination drug screening." Sci Transl Med 11(519): eaaw0064.

Lin, L., A. J. Sabnis, E. Chan, V. Olivas, L. Cade, E. Pazarentzos, S. Asthana, D. Neel, J. J. Yan, X. Lu,

L. Pham, M. M. Wang, N. Karachaliou, M. G. Cao, J. L. Manzano, J. L. Ramirez, J. M. Torres, F.

Buttitta, C. M. Rudin, E. A. Collisson, A. Algazi, E. Robinson, I. Osman, E. Muñoz-Couselo, J. Cortes,

D. T. Frederick, Z. A. Cooper, M. McMahon, A. Marchetti, R. Rosell, K. T. Flaherty, J. A. Wargo and T.

G. Bivona (2015). "The Hippo effector YAP promotes resistance to RAF- and MEK-targeted cancer

therapies." Nat Genet 47(3): 250-256.

Lindahl, T. and R. D. Wood (1999). "Quality control by DNA repair." Science 286(5446): 1897-1905.

Loeb, L. A., C. F. Springgate and N. Battula (1974). "Errors in DNA replication as a basis of malignant

changes." Cancer Res 34(9): 2311-2321.

Loeb, L. A., C. F. Springgate and N. Battula (1974). "Errors in DNA Replication as a Basis of Malignant

Changes." Cancer Research 34(9): 2311-2321.

Longley, M. J., A. J. Pierce and P. Modrich (1997). "DNA polymerase delta is required for human

mismatch repair in vitro." J Biol Chem 272(16): 10917-10921.

Lu, A. L., S. Clark and P. Modrich (1983). "Methyl-directed repair of DNA base-pair mismatches in

vitro." Proc Natl Acad Sci U S A 80(15): 4639-4643.

165

Lu, C., J. Guan, S. Lu, Q. Jin, B. Rousseau, T. Lu, D. Stephens, H. Zhang, J. Zhu, M. Yang, Z. Ren, Y.

Liang, Z. Liu, C. Han, L. Liu, X. Cao, A. Zhang, J. Qiao, K. Batten, M. Chen, D. H. Castrillon, T. Wang,

B. Li, L. A. Diaz, Jr., G. M. Li and Y. X. Fu (2021). "DNA Sensing in Mismatch Repair-Deficient Tumor

Cells Is Essential for Anti-tumor Immunity." Cancer Cell 39(1): 96-108.e106.

Lujan, S. A., A. R. Clausen, A. B. Clark, H. K. MacAlpine, D. M. MacAlpine, E. P. Malc, P. A.

Mieczkowski, A. B. Burkholder, D. C. Fargo, D. A. Gordenin and T. A. Kunkel (2014). "Heterogeneous

polymerase fidelity and mismatch repair bias genome variation and composition." Genome Res 24(11):

1751-1764.

Lynch, H. T. and A. de la Chapelle (2003). "Hereditary colorectal cancer." N Engl J Med 348(10): 919-

932.

Lynch, H. T., P. M. Lynch, J. Pester and R. M. Fusaro (1981). "The cancer family syndrome. Rare

cutaneous phenotypic linkage of Torre's syndrome." Arch Intern Med 141(5): 607-611.

Lynch, H. T., M. W. Shaw, C. W. Magnuson, A. L. Larsen and A. J. Krush (1966). "Hereditary factors in

cancer. Study of two large midwestern kindreds." Arch Intern Med 117(2): 206-212.

Maertens, O., R. Kuzmickas, H. E. Manchester, C. E. Emerson, A. G. Gavin, C. J. Guild, T. C. Wong, T.

De Raedt, C. Bowman-Colin, E. Hatchi, L. A. Garraway, K. T. Flaherty, S. Pathania, S. J. Elledge and K.

Cichowski (2019). "MAPK Pathway Suppression Unmasks Latent DNA Repair Defects and Confers a

Chemical Synthetic Vulnerability in <em>BRAF-, NRAS</em>-, and <em>NF1</em>-Mutant

Melanomas." Cancer Discovery 9(4): 526-545.

Mahoney, K. M., G. J. Freeman and D. F. McDermott (2015). "The Next Immune-Checkpoint Inhibitors:

PD-1/PD-L1 Blockade in Melanoma." Clin Ther 37(4): 764-782.

Mandal, R., R. M. Samstein, K.-W. Lee, J. J. Havel, H. Wang, C. Krishna, E. Y. Sabio, V. Makarov, F.

Kuo, P. Blecua, A. T. Ramaswamy, J. N. Durham, B. Bartlett, X. Ma, R. Srivastava, S. Middha, A. Zehir,

J. F. Hechtman, L. G. T. Morris, N. Weinhold, N. Riaz, D. T. Le, L. A. Diaz and T. A. Chan (2019).

"Genetic diversity of tumors with mismatch repair deficiency influences anti–PD-1 immunotherapy

response." Science 364(6439): 485.

Marabelle, A., M. Fakih, J. Lopez, M. Shah, R. Shapira-Frommer, K. Nakagawa, H. C. Chung, H. L.

Kindler, J. A. Lopez-Martin, W. H. Miller, Jr., A. Italiano, S. Kao, S. A. Piha-Paul, J. P. Delord, R. R.

McWilliams, D. A. Fabrizio, D. Aurora-Garg, L. Xu, F. Jin, K. Norwood and Y. J. Bang (2020).

"Association of tumour mutational burden with outcomes in patients with advanced solid tumours treated

with pembrolizumab: prospective biomarker analysis of the multicohort, open-label, phase 2 KEYNOTE-

158 study." Lancet Oncol 21(10): 1353-1365.

Marabelle, A., D. T. Le, P. A. Ascierto, A. M. D. Giacomo, A. D. Jesus-Acosta, J.-P. Delord, R. Geva, M.

Gottfried, N. Penel, A. R. Hansen, S. A. Piha-Paul, T. Doi, B. Gao, H. C. Chung, J. Lopez-Martin, Y.-J.

Bang, R. S. Frommer, M. Shah, R. Ghori, A. K. Joe, S. K. Pruitt and L. A. D. Jr (2020). "Efficacy of

Pembrolizumab in Patients With Noncolorectal High Microsatellite Instability/Mismatch Repair–

Deficient Cancer: Results From the Phase II KEYNOTE-158 Study." Journal of Clinical Oncology 38(1):

1-10.

Martin, L., B. Marples, M. Coffey, M. Lawler, D. Hollywood and L. Marignol (2009). "Recognition of

O6MeG lesions by MGMT and mismatch repair proficiency may be a prerequisite for low-dose radiation

hypersensitivity." Radiat Res 172(4): 405-413.

Martincorena, I., K. M. Raine, M. Gerstung, K. J. Dawson, K. Haase, P. Van Loo, H. Davies, M. R.

Stratton and P. J. Campbell (2017). "Universal Patterns of Selection in Cancer and Somatic Tissues." Cell

171(5): 1029-1041 e1021.

166

Massard, C., M. S. Gordon, S. Sharma, S. Rafii, Z. A. Wainberg, J. Luke, T. J. Curiel, G. Colon-Otero, O.

Hamid, R. E. Sanborn, P. H. O'Donnell, A. Drakaki, W. Tan, J. F. Kurland, M. C. Rebelatto, X. Jin, J. A.

Blake-Haskins, A. Gupta and N. H. Segal (2016). "Safety and Efficacy of Durvalumab (MEDI4736), an

Anti-Programmed Cell Death Ligand-1 Immune Checkpoint Inhibitor, in Patients With Advanced

Urothelial Bladder Cancer." J Clin Oncol 34(26): 3119-3125.

Matei, I. R., C. J. Guidos and J. S. Danska (2006). "ATM-dependent DNA damage surveillance in T-cell

development and leukemogenesis: the DSB connection." Immunol Rev 209: 142-158.

McCulloch, S. D. and T. A. Kunkel (2008). "The fidelity of DNA synthesis by eukaryotic replicative and

translesion synthesis polymerases." Cell Res 18(1): 148-161.

McFaline-Figueroa, J. L., C. J. Braun, M. Stanciu, Z. D. Nagel, P. Mazzucato, D. Sangaraju, E.

Cerniauskas, K. Barford, A. Vargas, Y. Chen, N. Tretyakova, J. A. Lees, M. T. Hemann, F. M. White and

L. D. Samson (2015). "Minor Changes in Expression of the Mismatch Repair Protein MSH2 Exert a

Major Impact on Glioblastoma Response to Temozolomide." Cancer Res 75(15): 3127-3138.

McGrail, D. J., J. Garnett, J. Yin, H. Dai, D. J. H. Shih, T. N. A. Lam, Y. Li, C. Sun, Y. Li, R. Schmandt,

J. Y. Wu, L. Hu, Y. Liang, G. Peng, E. Jonasch, D. Menter, M. S. Yates, S. Kopetz, K. H. Lu, R.

Broaddus, G. B. Mills, N. Sahni and S. Y. Lin (2020). "Proteome Instability Is a Therapeutic

Vulnerability in Mismatch Repair-Deficient Cancer." Cancer Cell 37(3): 371-386.e312.

McGranahan, N., A. J. Furness, R. Rosenthal, S. Ramskov, R. Lyngaa, S. K. Saini, M. Jamal-Hanjani, G.

A. Wilson, N. J. Birkbak, C. T. Hiley, T. B. Watkins, S. Shafi, N. Murugaesu, R. Mitter, A. U. Akarca, J.

Linares, T. Marafioti, J. Y. Henry, E. M. Van Allen, D. Miao, B. Schilling, D. Schadendorf, L. A.

Garraway, V. Makarov, N. A. Rizvi, A. Snyder, M. D. Hellmann, T. Merghoub, J. D. Wolchok, S. A.

Shukla, C. J. Wu, K. S. Peggs, T. A. Chan, S. R. Hadrup, S. A. Quezada and C. Swanton (2016). "Clonal

neoantigens elicit T cell immunoreactivity and sensitivity to immune checkpoint blockade." Science

351(6280): 1463-1469.

Meyers, M., M. W. Wagner, A. Mazurek, C. Schmutte, R. Fishel and D. A. Boothman (2005). "DNA

mismatch repair-dependent response to fluoropyrimidine-generated damage." J Biol Chem 280(7): 5516-

5526.

Miller, C. A., B. S. White, N. D. Dees, M. Griffith, J. S. Welch, O. L. Griffith, R. Vij, M. H. Tomasson,

T. A. Graubert, M. J. Walter, M. J. Ellis, W. Schierding, J. F. DiPersio, T. J. Ley, E. R. Mardis, R. K.

Wilson and L. Ding (2014). "SciClone: inferring clonal architecture and tracking the spatial and temporal

patterns of tumor evolution." PLoS Comput Biol 10(8): e1003665.

Miyabe, I., T. A. Kunkel and A. M. Carr (2011). "The major roles of DNA polymerases epsilon and delta

at the eukaryotic replication fork are evolutionarily conserved." PLoS Genet 7(12): e1002407.

Miyaki, M., M. Konishi, K. Tanaka, R. Kikuchi-Yanoshita, M. Muraoka, M. Yasuno, T. Igari, M. Koike,

M. Chiba and T. Mori (1997). "Germline mutation of MSH6 as the cause of hereditary nonpolyposis

colorectal cancer." Nat Genet 17(3): 271-272.

Miyaki, M., J. Nishio, M. Konishi, R. Kikuchi-Yanoshita, K. Tanaka, M. Muraoka, M. Nagato, J. M.

Chong, M. Koike, T. Terada, Y. Kawahara, A. Fukutome, J. Tomiyama, Y. Chuganji, M. Momoi and J.

Utsunomiya (1997). "Drastic genetic instability of tumors and normal tissues in Turcot syndrome."

Oncogene 15(23): 2877-2881.

Mlecnik, B., G. Bindea, H. K. Angell, P. Maby, M. Angelova, D. Tougeron, S. E. Church, L. Lafontaine,

M. Fischer, T. Fredriksen, M. Sasso, A. M. Bilocq, A. Kirilovsky, A. C. Obenauf, M. Hamieh, A. Berger,

P. Bruneval, J. J. Tuech, J. C. Sabourin, F. Le Pessot, J. Mauillon, A. Rafii, P. Laurent-Puig, M. R.

Speicher, Z. Trajanoski, P. Michel, R. Sesboüe, T. Frebourg, F. Pagès, V. Valge-Archer, J. B. Latouche

167

and J. Galon (2016). "Integrative Analyses of Colorectal Cancer Show Immunoscore Is a Stronger

Predictor of Patient Survival Than Microsatellite Instability." Immunity 44(3): 698-711.

Modrich, P. (1991). "Mechanisms and biological effects of mismatch repair." Annu Rev Genet 25: 229-

253.

Modrich, P. (2006). "Mechanisms in eukaryotic mismatch repair." J Biol Chem 281(41): 30305-30309.

Mojas, N., M. Lopes and J. Jiricny (2007). "Mismatch repair-dependent processing of methylation

damage gives rise to persistent single-stranded gaps in newly replicated DNA." Genes & Development

21(24): 3342-3355.

Montagut, C. and J. Settleman (2009). "Targeting the RAF-MEK-ERK pathway in cancer therapy."

Cancer Lett 283(2): 125-134.

Morrison, A., H. Araki, A. B. Clark, R. K. Hamatake and A. Sugino (1990). "A third essential DNA

polymerase in S. cerevisiae." Cell 62(6): 1143-1151.

Morrison, A., J. B. Bell, T. A. Kunkel and A. Sugino (1991). "Eukaryotic DNA polymerase amino acid

sequence required for 3'----5' exonuclease activity." Proc Natl Acad Sci U S A 88(21): 9473-9477.

Motzer, R. J., B. Escudier, D. F. McDermott, S. George, H. J. Hammers, S. Srinivas, S. S. Tykodi, J. A.

Sosman, G. Procopio, E. R. Plimack, D. Castellano, T. K. Choueiri, H. Gurney, F. Donskov, P. Bono, J.

Wagstaff, T. C. Gauler, T. Ueda, Y. Tomita, F. A. Schutz, C. Kollmannsberger, J. Larkin, A. Ravaud, J.

S. Simon, L. A. Xu, I. M. Waxman and P. Sharma (2015). "Nivolumab versus Everolimus in Advanced

Renal-Cell Carcinoma." N Engl J Med 373(19): 1803-1813.

Murphy, K., H. Darmawan, A. Schultz, E. Fidalgo da Silva and L. J. Reha-Krantz (2006). "A method to

select for mutator DNA polymerase deltas in Saccharomyces cerevisiae." Genome 49(4): 403-410.

Nagel, Z. D., G. J. Kitange, S. K. Gupta, B. A. Joughin, I. A. Chaim, P. Mazzucato, D. A. Lauffenburger,

J. N. Sarkaria and L. D. Samson (2017). "DNA Repair Capacity in Multiple Pathways Predicts

Chemoresistance in Glioblastoma Multiforme." Cancer Research 77(1): 198-206.

Nghiem, P., S. Bhatia, E. J. Lipson, W. H. Sharfman, R. R. Kudchadkar, A. S. Brohl, P. A. Friedlander,

A. Daud, H. M. Kluger, S. A. Reddy, B. C. Boulmay, A. I. Riker, M. A. Burgess, B. A. Hanks, T.

Olencki, K. Margolin, L. M. Lundgren, A. Soni, N. Ramchurren, C. Church, S. Y. Park, M. M. Shinohara,

B. Salim, J. M. Taube, S. R. Bird, N. Ibrahim, S. P. Fling, B. Homet Moreno, E. Sharon, M. A. Cheever

and S. L. Topalian (2019). "Durable Tumor Regression and Overall Survival in Patients With Advanced

Merkel Cell Carcinoma Receiving Pembrolizumab as First-Line Therapy." J Clin Oncol 37(9): 693-702.

Nguyen, S. A., O. D. M. Stechishin, H. A. Luchman, X. Q. Lun, D. L. Senger, S. M. Robbins, J. G.

Cairncross and S. Weiss (2014). "Novel <em>MSH6</em> Mutations in Treatment-Naïve Glioblastoma

and Anaplastic Oligodendroglioma Contribute to Temozolomide Resistance Independently of

<em>MGMT</em> Promoter Methylation." Clinical Cancer Research 20(18): 4894-4903.

Nick McElhinny, S. A., D. A. Gordenin, C. M. Stith, P. M. Burgers and T. A. Kunkel (2008). "Division

of labor at the eukaryotic replication fork." Mol Cell 30(2): 137-144.

Nicolaides, N. C., N. Papadopoulos, B. Liu, Y. F. Wei, K. C. Carter, S. M. Ruben, C. A. Rosen, W. A.

Haseltine, R. D. Fleischmann, C. M. Fraser and et al. (1994). "Mutations of two PMS homologues in

hereditary nonpolyposis colon cancer." Nature 371(6492): 75-80.

Northcott, P. A., S. M. Pfister and D. T. Jones (2015). "Next-generation (epi)genetic drivers of childhood

brain tumours and the outlook for targeted therapies." Lancet Oncol 16(6): e293-302.

Nowell, P. C. (1976). "The clonal evolution of tumor cell populations." Science 194(4260): 23-28.

168

Obmolova, G., C. Ban, P. Hsieh and W. Yang (2000). "Crystal structures of mismatch repair protein

MutS and its complex with a substrate DNA." Nature 407(6805): 703-710.

Offman, J., G. Opelz, B. Doehler, D. Cummins, O. Halil, N. R. Banner, M. M. Burke, D. Sullivan, P.

Macpherson and P. Karran (2004). "Defective DNA mismatch repair in acute myeloid

leukemia/myelodysplastic syndrome after organ transplantation." Blood 104(3): 822-828.

Okada, Y., G. Streisinger, J. E. Owen, J. Newton, A. Tsugita and M. Inouye (1972). "Molecular basis of a

mutational hot spot in the lysozyme gene of bacteriophage T4." Nature 236(5346): 338-341.

Okazaki, R., T. Okazaki, K. Sakabe, K. Sugimoto and A. Sugino (1968). "Mechanism of DNA chain

growth. I. Possible discontinuity and unusual secondary structure of newly synthesized chains."

Proceedings of the National Academy of Sciences of the United States of America 59(2): 598-605.

Oshrine, B., N. Grana, C. Moore, J. Nguyen, M. Crenshaw, M. Edwards, S. Sudhaman, V. J. Forster and

U. Tabori (2019). "B-cell acute lymphoblastic leukemia with high mutation burden presenting in a child

with constitutional mismatch repair deficiency." Blood Adv 3(12): 1795-1798.

Overman, M. J., S. Lonardi, K. Y. M. Wong, H.-J. Lenz, F. Gelsomino, M. Aglietta, M. A. Morse, E. V.

Cutsem, R. McDermott, A. Hill, M. B. Sawyer, A. Hendlisz, B. Neyns, M. Svrcek, R. A. Moss, J.-M.

Ledeine, Z. A. Cao, S. Kamble, S. Kopetz and T. André (2018). "Durable Clinical Benefit With

Nivolumab Plus Ipilimumab in DNA Mismatch Repair–Deficient/Microsatellite Instability–High

Metastatic Colorectal Cancer." Journal of Clinical Oncology 36(8): 773-779.

Overman, M. J., R. McDermott, J. L. Leach, S. Lonardi, H. J. Lenz, M. A. Morse, J. Desai, A. Hill, M.

Axelson, R. A. Moss, M. V. Goldberg, Z. A. Cao, J. M. Ledeine, G. A. Maglinte, S. Kopetz and T. Andre

(2017). "Nivolumab in patients with metastatic DNA mismatch repair-deficient or microsatellite

instability-high colorectal cancer (CheckMate 142): an open-label, multicentre, phase 2 study." Lancet

Oncol 18(9): 1182-1191.

Palles, C., J. B. Cazier, K. M. Howarth, E. Domingo, A. M. Jones, P. Broderick, Z. Kemp, S. L. Spain, E.

Guarino, I. Salguero, A. Sherborne, D. Chubb, L. G. Carvajal-Carmona, Y. Ma, K. Kaur, S. Dobbins, E.

Barclay, M. Gorman, L. Martin, M. B. Kovac, S. Humphray, C. Consortium, W. G. S. Consortium, A.

Lucassen, C. C. Holmes, D. Bentley, P. Donnelly, J. Taylor, C. Petridis, R. Roylance, E. J. Sawyer, D. J.

Kerr, S. Clark, J. Grimes, S. E. Kearsey, H. J. Thomas, G. McVean, R. S. Houlston and I. Tomlinson

(2013). "Germline mutations affecting the proofreading domains of POLE and POLD1 predispose to

colorectal adenomas and carcinomas." Nat Genet 45(2): 136-144.

Palles, C., J. B. Cazier, K. M. Howarth, E. Domingo, A. M. Jones, P. Broderick, Z. Kemp, S. L. Spain, E.

Guarino, I. Salguero, A. Sherborne, D. Chubb, L. G. Carvajal-Carmona, Y. Ma, K. Kaur, S. Dobbins, E.

Barclay, M. Gorman, L. Martin, M. B. Kovac, S. Humphray, A. Lucassen, C. C. Holmes, D. Bentley, P.

Donnelly, J. Taylor, C. Petridis, R. Roylance, E. J. Sawyer, D. J. Kerr, S. Clark, J. Grimes, S. E. Kearsey,

H. J. Thomas, G. McVean, R. S. Houlston and I. Tomlinson (2013). "Germline mutations affecting the

proofreading domains of POLE and POLD1 predispose to colorectal adenomas and carcinomas." Nat

Genet 45(2): 136-144.

Papadopoulos, N., N. C. Nicolaides, Y. F. Wei, S. M. Ruben, K. C. Carter, C. A. Rosen, W. A. Haseltine,

R. D. Fleischmann, C. M. Fraser, M. D. Adams and et al. (1994). "Mutation of a mutL homolog in

hereditary colon cancer." Science 263(5153): 1625-1629.

Parsons, R., G. M. Li, M. J. Longley, W. H. Fang, N. Papadopoulos, J. Jen, A. de la Chapelle, K. W.

Kinzler, B. Vogelstein and P. Modrich (1993). "Hypermutability and mismatch repair deficiency in RER+

tumor cells." Cell 75(6): 1227-1236.

Pavlov, Y. I., A. S. Zhuk and E. I. Stepchenkova (2020). "DNA Polymerases at the Eukaryotic

Replication Fork Thirty Years after: Connection to Cancer." Cancers (Basel) 12(12).

169

Pellegrini, L. (2012). "The Pol α-primase complex." Subcell Biochem 62: 157-169.

Petljak, M., L. B. Alexandrov, J. S. Brammeld, S. Price, D. C. Wedge, S. Grossmann, K. J. Dawson, Y. S.

Ju, F. Iorio, J. M. C. Tubio, C. C. Koh, I. Georgakopoulos-Soares, B. Rodriguez-Martin, B. Otlu, S.

O'Meara, A. P. Butler, A. Menzies, S. G. Bhosle, K. Raine, D. R. Jones, J. W. Teague, K. Beal, C.

Latimer, L. O'Neill, J. Zamora, E. Anderson, N. Patel, M. Maddison, B. L. Ng, J. Graham, M. J. Garnett,

U. McDermott, S. Nik-Zainal, P. J. Campbell and M. R. Stratton (2019). "Characterizing Mutational

Signatures in Human Cancer Cell Lines Reveals Episodic APOBEC Mutagenesis." Cell 176(6): 1282-

1294.e1220.

Pfeifer, G. P., Y. H. You and A. Besaratinia (2005). "Mutations induced by ultraviolet light." Mutat Res

571(1-2): 19-31.

Plazzer, J. P., R. H. Sijmons, M. O. Woods, P. Peltomäki, B. Thompson, J. T. Den Dunnen and F. Macrae

(2013). "The InSiGHT database: utilizing 100 years of insights into Lynch syndrome." Fam Cancer 12(2):

175-180.

Pleasance, E. D., P. J. Stephens, S. O'Meara, D. J. McBride, A. Meynert, D. Jones, M. L. Lin, D. Beare,

K. W. Lau, C. Greenman, I. Varela, S. Nik-Zainal, H. R. Davies, G. R. Ordonez, L. J. Mudie, C. Latimer,

S. Edkins, L. Stebbings, L. Chen, M. Jia, C. Leroy, J. Marshall, A. Menzies, A. Butler, J. W. Teague, J.

Mangion, Y. A. Sun, S. F. McLaughlin, H. E. Peckham, E. F. Tsung, G. L. Costa, C. C. Lee, J. D. Minna,

A. Gazdar, E. Birney, M. D. Rhodes, K. J. McKernan, M. R. Stratton, P. A. Futreal and P. J. Campbell

(2010). "A small-cell lung cancer genome with complex signatures of tobacco exposure." Nature

463(7278): 184-190.

Poon, S. L., S. T. Pang, J. R. McPherson, W. Yu, K. K. Huang, P. Guan, W. H. Weng, E. Y. Siew, Y. Liu,

H. L. Heng, S. C. Chong, A. Gan, S. T. Tay, W. K. Lim, I. Cutcutache, D. Huang, L. D. Ler, M. L.

Nairismagi, M. H. Lee, Y. H. Chang, K. J. Yu, W. Chan-On, B. K. Li, Y. F. Yuan, C. N. Qian, K. F. Ng,

C. F. Wu, C. L. Hsu, R. M. Bunte, M. R. Stratton, P. A. Futreal, W. K. Sung, C. K. Chuang, C. K. Ong, S.

G. Rozen, P. Tan and B. T. Teh (2013). "Genome-wide mutational signatures of aristolochic acid and its

application as a screening tool." Sci Transl Med 5(197): 197ra101.

Postow, M. A., J. Chesney, A. C. Pavlick, C. Robert, K. Grossmann, D. McDermott, G. P. Linette, N.

Meyer, J. K. Giguere, S. S. Agarwala, M. Shaheen, M. S. Ernstoff, D. Minor, A. K. Salama, M. Taylor, P.

A. Ott, L. M. Rollin, C. Horak, P. Gagnier, J. D. Wolchok and F. S. Hodi (2015). "Nivolumab and

ipilimumab versus ipilimumab in untreated melanoma." N Engl J Med 372(21): 2006-2017.

Prasad, V., V. Kaestner and S. Mailankody (2018). "Cancer Drugs Approved Based on Biomarkers and

Not Tumor Type-FDA Approval of Pembrolizumab for Mismatch Repair-Deficient Solid Cancers."

JAMA Oncol 4(2): 157-158.

Preston, B. D., T. M. Albertson and A. J. Herr (2010). "DNA replication fidelity and cancer." Semin

Cancer Biol 20(5): 281-293.

Pritchard, C. C., C. Morrissey, A. Kumar, X. Zhang, C. Smith, I. Coleman, S. J. Salipante, J. Milbank, M.

Yu, W. M. Grady, J. F. Tait, E. Corey, R. L. Vessella, T. Walsh, J. Shendure and P. S. Nelson (2014).

"Complex MSH2 and MSH6 mutations in hypermutated microsatellite unstable advanced prostate

cancer." Nat Commun 5: 4988.

Prolla, T. A., S. M. Baker, A. C. Harris, J. L. Tsao, X. Yao, C. E. Bronner, B. Zheng, M. Gordon, J.

Reneker, N. Arnheim, D. Shibata, A. Bradley and R. M. Liskay (1998). "Tumour susceptibility and

spontaneous mutation in mice deficient in Mlh1, Pms1 and Pms2 DNA mismatch repair." Nat Genet

18(3): 276-279.

Pursell, Z. F., I. Isoz, E. B. Lundstrom, E. Johansson and T. A. Kunkel (2007). "Yeast DNA polymerase

epsilon participates in leading-strand DNA replication." Science 317(5834): 127-130.

170

Quiros, S., W. P. Roos and B. Kaina (2010). "Processing of O6-methylguanine into DNA double-strand

breaks requires two rounds of replication whereas apoptosis is also induced in subsequent cell cycles."

Cell Cycle 9(1): 168-178.

Rayner, E., I. C. van Gool, C. Palles, S. E. Kearsey, T. Bosse, I. Tomlinson and D. N. Church (2016). "A

panoply of errors: polymerase proofreading domain mutations in cancer." Nat Rev Cancer 16(2): 71-81.

Reardon, D. A., A. A. Brandes, A. Omuro, P. Mulholland, M. Lim, A. Wick, J. Baehring, M. S.

Ahluwalia, P. Roth, O. Bähr, S. Phuphanich, J. M. Sepulveda, P. De Souza, S. Sahebjam, M. Carleton, K.

Tatsuoka, C. Taitt, R. Zwirtes, J. Sampson and M. Weller (2020). "Effect of Nivolumab vs Bevacizumab

in Patients With Recurrent Glioblastoma: The CheckMate 143 Phase 3 Randomized Clinical Trial."

JAMA Oncol 6(7): 1003-1010.

Reck, M., T. S. K. Mok, M. Nishio, R. M. Jotte, F. Cappuzzo, F. Orlandi, D. Stroyakovskiy, N. Nogami,

D. Rodríguez-Abreu, D. Moro-Sibilot, C. A. Thomas, F. Barlesi, G. Finley, A. Lee, S. Coleman, Y. Deng,

M. Kowanetz, G. Shankar, W. Lin and M. A. Socinski (2019). "Atezolizumab plus bevacizumab and

chemotherapy in non-small-cell lung cancer (IMpower150): key subgroup analyses of patients with

EGFR mutations or baseline liver metastases in a randomised, open-label phase 3 trial." Lancet Respir

Med 7(5): 387-401.

Reiss, C., T. Haneke, H. U. Volker, M. Spahn, A. Rosenwald, W. Edelmann and B. Kneitz (2010).

"Conditional inactivation of MLH1 in thymic and naive T-cells in mice leads to a limited incidence of

lymphoblastic T-cell lymphomas." Leuk Lymphoma 51(10): 1875-1886.

Reitmair, A. H., R. Schmits, A. Ewel, B. Bapat, M. Redston, A. Mitri, P. Waterhouse, H. W. Mittrucker,

A. Wakeham, B. Liu and et al. (1995). "MSH2 deficient mice are viable and susceptible to lymphoid

tumours." Nat Genet 11(1): 64-70.

Riaz, N., J. J. Havel, V. Makarov, A. Desrichard, W. J. Urba, J. S. Sims, F. S. Hodi, S. Martín-Algarra, R.

Mandal, W. H. Sharfman, S. Bhatia, W. J. Hwu, T. F. Gajewski, C. L. Slingluff, Jr., D. Chowell, S. M.

Kendall, H. Chang, R. Shah, F. Kuo, L. G. T. Morris, J. W. Sidhom, J. P. Schneck, C. E. Horak, N.

Weinhold and T. A. Chan (2017). "Tumor and Microenvironment Evolution during Immunotherapy with

Nivolumab." Cell 171(4): 934-949.e916.

Ribas, A., O. Hamid, A. Daud, F. S. Hodi, J. D. Wolchok, R. Kefford, A. M. Joshua, A. Patnaik, W. J.

Hwu, J. S. Weber, T. C. Gangadhar, P. Hersey, R. Dronca, R. W. Joseph, H. Zarour, B. Chmielowski, D.

P. Lawrence, A. Algazi, N. A. Rizvi, B. Hoffner, C. Mateus, K. Gergich, J. A. Lindia, M. Giannotti, X. N.

Li, S. Ebbinghaus, S. P. Kang and C. Robert (2016). "Association of Pembrolizumab With Tumor

Response and Survival Among Patients With Advanced Melanoma." Jama 315(15): 1600-1609.

Ribas, A., I. Puzanov, R. Dummer, D. Schadendorf, O. Hamid, C. Robert, F. S. Hodi, J. Schachter, A. C.

Pavlick, K. D. Lewis, L. D. Cranmer, C. U. Blank, S. J. O'Day, P. A. Ascierto, A. K. Salama, K. A.

Margolin, C. Loquai, T. K. Eigentler, T. C. Gangadhar, M. S. Carlino, S. S. Agarwala, S. J. Moschos, J.

A. Sosman, S. M. Goldinger, R. Shapira-Frommer, R. Gonzalez, J. M. Kirkwood, J. D. Wolchok, A.

Eggermont, X. N. Li, W. Zhou, A. M. Zernhelt, J. Lis, S. Ebbinghaus, S. P. Kang and A. Daud (2015).

"Pembrolizumab versus investigator-choice chemotherapy for ipilimumab-refractory melanoma

(KEYNOTE-002): a randomised, controlled, phase 2 trial." Lancet Oncol 16(8): 908-918.

Ribic, C. M., D. J. Sargent, M. J. Moore, S. N. Thibodeau, A. J. French, R. M. Goldberg, S. R. Hamilton,

P. Laurent-Puig, R. Gryfe, L. E. Shepherd, D. Tu, M. Redston and S. Gallinger (2003). "Tumor

microsatellite-instability status as a predictor of benefit from fluorouracil-based adjuvant chemotherapy

for colon cancer." N Engl J Med 349(3): 247-257.

Ricciardone, M. D., T. Ozçelik, B. Cevher, H. Ozdağ, M. Tuncer, A. Gürgey, O. Uzunalimoğlu, H.

Cetinkaya, A. Tanyeli, E. Erken and M. Oztürk (1999). "Human MLH1 deficiency predisposes to

hematological malignancy and neurofibromatosis type 1." Cancer Res 59(2): 290-293.

171

Rizvi, N. A., M. D. Hellmann, A. Snyder, P. Kvistborg, V. Makarov, J. J. Havel, W. Lee, J. Yuan, P.

Wong, T. S. Ho, M. L. Miller, N. Rekhtman, A. L. Moreira, F. Ibrahim, C. Bruggeman, B. Gasmi, R.

Zappasodi, Y. Maeda, C. Sander, E. B. Garon, T. Merghoub, J. D. Wolchok, T. N. Schumacher and T. A.

Chan (2015). "Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung

cancer." Science 348(6230): 124-128.

Robert, C., J. Schachter, G. V. Long, A. Arance, J. J. Grob, L. Mortier, A. Daud, M. S. Carlino, C.

McNeil, M. Lotem, J. Larkin, P. Lorigan, B. Neyns, C. U. Blank, O. Hamid, C. Mateus, R. Shapira-

Frommer, M. Kosh, H. Zhou, N. Ibrahim, S. Ebbinghaus and A. Ribas (2015). "Pembrolizumab versus

Ipilimumab in Advanced Melanoma." N Engl J Med 372(26): 2521-2532.

Rohlin, A., T. Zagoras, S. Nilsson, U. Lundstam, J. Wahlstrom, L. Hulten, T. Martinsson, G. B. Karlsson

and M. Nordling (2014). "A mutation in POLE predisposing to a multi-tumour phenotype." Int J Oncol

45(1): 77-81.

Rooney, M. S., S. A. Shukla, C. J. Wu, G. Getz and N. Hacohen (2015). "Molecular and genetic

properties of tumors associated with local immune cytolytic activity." Cell 160(1-2): 48-61.

Rosenthal, R., N. McGranahan, J. Herrero, B. S. Taylor and C. Swanton (2016). "deconstructSigs:

delineating mutational processes in single tumors distinguishes DNA repair deficiencies and patterns of

carcinoma evolution." Genome Biology 17(1): 31.

Rosner, G., R. Elya, D. Bercovich, E. Santo, Z. Halpern and R. Kariv (2015). "291 Mutations in DNA

Polymerase Genes (POLD1 & POLE) in Individuals Having Early-Onset Colorectal Cancer and/or

Multiple Adenomas." Gastroenterology 148(4, Supplement 1): S-63.

Russo, M., G. Crisafulli, A. Sogari, N. M. Reilly, S. Arena, S. Lamba, A. Bartolini, V. Amodio, A. Magrì,

L. Novara, I. Sarotto, Z. D. Nagel, C. G. Piett, A. Amatu, A. Sartore-Bianchi, S. Siena, A. Bertotti, L.

Trusolino, M. Corigliano, M. Gherardi, M. C. Lagomarsino, F. Di Nicolantonio and A. Bardelli (2019).

"Adaptive mutability of colorectal cancers in response to targeted therapies." Science 366(6472): 1473.

Ryall, S., R. Krishnatry, A. Arnoldo, P. Buczkowicz, M. Mistry, R. Siddaway, C. Ling, S. Pajovic, M.

Yu, J. B. Rubin, J. Hukin, P. Steinbok, U. Bartels, E. Bouffet, U. Tabori and C. Hawkins (2016).

"Targeted detection of genetic alterations reveal the prognostic impact of H3K27M and MAPK pathway

aberrations in paediatric thalamic glioma." Acta Neuropathol Commun 4(1): 93.

Sade-Feldman, M., K. Yizhak, S. L. Bjorgaard, J. P. Ray, C. G. de Boer, R. W. Jenkins, D. J. Lieb, J. H.

Chen, D. T. Frederick, M. Barzily-Rokni, S. S. Freeman, A. Reuben, P. J. Hoover, A. C. Villani, E.

Ivanova, A. Portell, P. H. Lizotte, A. R. Aref, J. P. Eliane, M. R. Hammond, H. Vitzthum, S. M.

Blackmon, B. Li, V. Gopalakrishnan, S. M. Reddy, Z. A. Cooper, C. P. Paweletz, D. A. Barbie, A.

Stemmer-Rachamimov, K. T. Flaherty, J. A. Wargo, G. M. Boland, R. J. Sullivan, G. Getz and N.

Hacohen (2018). "Defining T Cell States Associated with Response to Checkpoint Immunotherapy in

Melanoma." Cell 175(4): 998-1013.e1020.

Sage, P. T., A. M. Paterson, S. B. Lovitch and A. H. Sharpe (2014). "The coinhibitory receptor CTLA-4

controls B cell responses by modulating T follicular helper, T follicular regulatory, and T regulatory

cells." Immunity 41(6): 1026-1039.

Santin, A. D., S. Bellone, N. Buza, J. Choi, P. E. Schwartz, J. Schlessinger and R. P. Lifton (2016).

"Regression of Chemotherapy-Resistant Polymerase epsilon (POLE) Ultra-Mutated and MSH6 Hyper-

Mutated Endometrial Tumors with Nivolumab." Clin Cancer Res 22(23): 5682-5687.

Schadendorf, D., F. S. Hodi, C. Robert, J. S. Weber, K. Margolin, O. Hamid, D. Patt, T. T. Chen, D. M.

Berman and J. D. Wolchok (2015). "Pooled Analysis of Long-Term Survival Data From Phase II and

Phase III Trials of Ipilimumab in Unresectable or Metastatic Melanoma." J Clin Oncol 33(17): 1889-

1894.

172

Schubert, M., B. Klinger, M. Klünemann, A. Sieber, F. Uhlitz, S. Sauer, M. J. Garnett, N. Blüthgen and J.

Saez-Rodriguez (2018). "Perturbation-response genes reveal signaling footprints in cancer gene

expression." Nature Communications 9(1): 20.

Sebolt-Leopold, J. S. and R. Herrera (2004). "Targeting the mitogen-activated protein kinase cascade to

treat cancer." Nat Rev Cancer 4(12): 937-947.

Shamoo, Y. and T. A. Steitz (1999). "Building a replisome from interacting pieces: sliding clamp

complexed to a peptide from DNA polymerase and a polymerase editing complex." Cell 99(2): 155-166.

Shankaran, V., H. Ikeda, A. T. Bruce, J. M. White, P. E. Swanson, L. J. Old and R. D. Schreiber (2001).

"IFNgamma and lymphocytes prevent primary tumour development and shape tumour immunogenicity."

Nature 410(6832): 1107-1111.

Shannon, S., D. Jia, I. Entersz, P. Beelen, M. Yu, C. Carcione, J. Carcione, A. Mahtabfar, C. Vaca, M.

Weaver, D. Shreiber, J. D. Zahn, L. Liu, H. Lin and R. A. Foty (2017). "Inhibition of glioblastoma

dispersal by the MEK inhibitor PD0325901." BMC Cancer 17(1): 121.

Sharma, P. and J. P. Allison (2015). "Immune checkpoint targeting in cancer therapy: toward combination

strategies with curative potential." Cell 161(2): 205-214.

Sharma, P., S. Hu-Lieskovan, J. A. Wargo and A. Ribas (2017). "Primary, Adaptive, and Acquired

Resistance to Cancer Immunotherapy." Cell 168(4): 707-723.

Sharma, P., M. Retz, A. Siefker-Radtke, A. Baron, A. Necchi, J. Bedke, E. R. Plimack, D. Vaena, M. O.

Grimm, S. Bracarda, J. Arranz, S. Pal, C. Ohyama, A. Saci, X. Qu, A. Lambert, S. Krishnan, A.

Azrilevich and M. D. Galsky (2017). "Nivolumab in metastatic urothelial carcinoma after platinum

therapy (CheckMate 275): a multicentre, single-arm, phase 2 trial." Lancet Oncol 18(3): 312-322.

Shcherbakova, P. V. and Y. I. Pavlov (1996). "3'-->5' exonucleases of DNA polymerases epsilon and

delta correct base analog induced DNA replication errors on opposite DNA strands in Saccharomyces

cerevisiae." Genetics 142(3): 717-726.

Shcherbakova, P. V., Y. I. Pavlov, O. Chilkova, I. B. Rogozin, E. Johansson and T. A. Kunkel (2003).

"Unique error signature of the four-subunit yeast DNA polymerase epsilon." J Biol Chem 278(44):

43770-43780.

Shi, J., S. Hou, Q. Fang, X. Liu, X. Liu and H. Qi (2018). "PD-1 Controls Follicular T Helper Cell

Positioning and Function." Immunity 49(2): 264-274.e264.

Shinbrot, E., E. E. Henninger, N. Weinhold, K. R. Covington, A. Y. Goksenin, N. Schultz, H. Chao, H.

Doddapaneni, D. M. Muzny, R. A. Gibbs, C. Sander, Z. F. Pursell and D. A. Wheeler (2014).

"Exonuclease mutations in DNA polymerase epsilon reveal replication strand specific mutation patterns

and human origins of replication." Genome Res 24(11): 1740-1750.

Shlien, A., B. B. Campbell, R. de Borja, L. B. Alexandrov, D. Merico, D. Wedge, P. Van Loo, P. S.

Tarpey, P. Coupland, S. Behjati, A. Pollett, T. Lipman, A. Heidari, S. Deshmukh, N. Avitzur, B. Meier,

M. Gerstung, Y. Hong, D. M. Merino, M. Ramakrishna, M. Remke, R. Arnold, G. B. Panigrahi, N. P.

Thakkar, K. P. Hodel, E. E. Henninger, A. Y. Goksenin, D. Bakry, G. S. Charames, H. Druker, J. Lerner-

Ellis, M. Mistry, R. Dvir, R. Grant, R. Elhasid, R. Farah, G. P. Taylor, P. C. Nathan, S. Alexander, S.

Ben-Shachar, S. C. Ling, S. Gallinger, S. Constantini, P. Dirks, A. Huang, S. W. Scherer, R. G. Grundy,

C. Durno, M. Aronson, A. Gartner, M. S. Meyn, M. D. Taylor, Z. F. Pursell, C. E. Pearson, D. Malkin, P.

A. Futreal, M. R. Stratton, E. Bouffet, C. Hawkins, P. J. Campbell and U. Tabori (2015). "Combined

hereditary and somatic mutations of replication error repair genes result in rapid onset of ultra-

hypermutated cancers." Nat Genet 47(3): 257-262.

173

Snyder, A., V. Makarov, T. Merghoub, J. Yuan, J. M. Zaretsky, A. Desrichard, L. A. Walsh, M. A.

Postow, P. Wong, T. S. Ho, T. J. Hollmann, C. Bruggeman, K. Kannan, Y. Li, C. Elipenahli, C. Liu, C. T.

Harbison, L. Wang, A. Ribas, J. D. Wolchok and T. A. Chan (2014). "Genetic basis for clinical response

to CTLA-4 blockade in melanoma." N Engl J Med 371(23): 2189-2199.

Spier, I., S. Holzapfel, J. Altmuller, B. Zhao, S. Horpaopan, S. Vogt, S. Chen, M. Morak, S. Raeder, K.

Kayser, D. Stienen, R. Adam, P. Nurnberg, G. Plotz, E. Holinski-Feder, R. P. Lifton, H. Thiele, P.

Hoffmann, V. Steinke and S. Aretz (2015). "Frequency and phenotypic spectrum of germline mutations in

POLE and seven other polymerase genes in 266 patients with colorectal adenomas and carcinomas." Int J

Cancer 137(2): 320-331.

Strand, M., T. A. Prolla, R. M. Liskay and T. D. Petes (1993). "Destabilization of tracts of simple

repetitive DNA in yeast by mutations affecting DNA mismatch repair." Nature 365(6443): 274-276.

Streisinger, G. and J. Owen (1985). "Mechanisms of spontaneous and induced frameshift mutation in

bacteriophage T4." Genetics 109(4): 633-659.

Stupp, R., W. P. Mason, M. J. van den Bent, M. Weller, B. Fisher, M. J. Taphoorn, K. Belanger, A. A.

Brandes, C. Marosi, U. Bogdahn, J. Curschmann, R. C. Janzer, S. K. Ludwin, T. Gorlia, A. Allgeier, D.

Lacombe, J. G. Cairncross, E. Eisenhauer and R. O. Mirimanoff (2005). "Radiotherapy plus concomitant

and adjuvant temozolomide for glioblastoma." N Engl J Med 352(10): 987-996.

Sun, C., Y. Fang, J. Yin, J. Chen, Z. Ju, D. Zhang, X. Chen, C. P. Vellano, K. J. Jeong, P. K.-S. Ng, A. K.

B. Eterovic, N. H. Bhola, Y. Lu, S. N. Westin, J. R. Grandis, S.-Y. Lin, K. L. Scott, G. Peng, J. Brugge

and G. B. Mills (2017). "Rational combination therapy with PARP and MEK inhibitors capitalizes on

therapeutic liabilities in <em>RAS</em> mutant cancers." Science Translational Medicine 9(392):

eaal5148.

Swann, P. F., T. R. Waters, D. C. Moulton, Y. Z. Xu, Q. Zheng, M. Edwards and R. Mace (1996). "Role

of postreplicative DNA mismatch repair in the cytotoxic action of thioguanine." Science 273(5278):

1109-1111.

Tabori, U., J. R. Hansford, M. I. Achatz, C. P. Kratz, S. E. Plon, T. Frebourg and L. Brugières (2017).

"Clinical Management and Tumor Surveillance Recommendations of Inherited Mismatch Repair

Deficiency in Childhood." Clinical Cancer Research 23(11): e32-e37.

Takagi, Y., M. Takahashi, M. Sanada, R. Ito, M. Yamaizumi and M. Sekiguchi (2003). "Roles of MGMT

and MLH1 proteins in alkylation-induced apoptosis and mutagenesis." DNA Repair (Amst) 2(10): 1135-

1146.

Tempero, M. A., M. M. Jacobs, H. T. Lynch, C. L. Graham and A. J. Blotcky (1984). "Serum and hair

selenium levels in hereditary nonpolyposis colorectal cancer." Biol Trace Elem Res 6(1): 51-55.

The Cancer Genome Atlas, N., D. M. Muzny, M. N. Bainbridge, K. Chang, H. H. Dinh, J. A. Drummond,

G. Fowler, C. L. Kovar, L. R. Lewis, M. B. Morgan, I. F. Newsham, J. G. Reid, J. Santibanez, E.

Shinbrot, L. R. Trevino, Y.-Q. Wu, M. Wang, P. Gunaratne, L. A. Donehower, C. J. Creighton, D. A.

Wheeler, R. A. Gibbs, M. S. Lawrence, D. Voet, R. Jing, K. Cibulskis, A. Sivachenko, P. Stojanov, A.

McKenna, E. S. Lander, S. Gabriel, G. Getz, L. Ding, R. S. Fulton, D. C. Koboldt, T. Wylie, J. Walker,

D. J. Dooling, L. Fulton, K. D. Delehaunty, C. C. Fronick, R. Demeter, E. R. Mardis, R. K. Wilson, A.

Chu, H.-J. E. Chun, A. J. Mungall, E. Pleasance, A. Gordon Robertson, D. Stoll, M. Balasundaram, I.

Birol, Y. S. N. Butterfield, E. Chuah, R. J. N. Coope, N. Dhalla, R. Guin, C. Hirst, M. Hirst, R. A. Holt,

D. Lee, H. I. Li, M. Mayo, R. A. Moore, J. E. Schein, J. R. Slobodan, A. Tam, N. Thiessen, R. Varhol, T.

Zeng, Y. Zhao, S. J. M. Jones, M. A. Marra, A. J. Bass, A. H. Ramos, G. Saksena, A. D. Cherniack, S. E.

Schumacher, B. Tabak, S. L. Carter, N. H. Pho, H. Nguyen, R. C. Onofrio, A. Crenshaw, K. Ardlie, R.

Beroukhim, W. Winckler, G. Getz, M. Meyerson, A. Protopopov, J. Zhang, A. Hadjipanayis, E. Lee, R.

Xi, L. Yang, X. Ren, H. Zhang, N. Sathiamoorthy, S. Shukla, P.-C. Chen, P. Haseley, Y. Xiao, S. Lee, J.

174

Seidman, L. Chin, P. J. Park, R. Kucherlapati, J. Todd Auman, K. A. Hoadley, Y. Du, M. D. Wilkerson,

Y. Shi, C. Liquori, S. Meng, L. Li, Y. J. Turman, M. D. Topal, D. Tan, S. Waring, E. Buda, J. Walsh, C.

D. Jones, P. A. Mieczkowski, D. Singh, J. Wu, A. Gulabani, P. Dolina, T. Bodenheimer, A. P. Hoyle, J.

V. Simons, M. Soloway, L. E. Mose, S. R. Jefferys, S. Balu, B. D. O’Connor, J. F. Prins, D. Y. Chiang,

D. Neil Hayes, C. M. Perou, T. Hinoue, D. J. Weisenberger, D. T. Maglinte, F. Pan, B. P. Berman, D. J.

Van Den Berg, H. Shen, T. Triche Jr, S. B. Baylin, P. W. Laird, G. Getz, M. Noble, D. Voet, G. Saksena,

N. Gehlenborg, D. DiCara, J. Zhang, H. Zhang, C.-J. Wu, S. Yingchun Liu, S. Shukla, M. S. Lawrence,

L. Zhou, A. Sivachenko, P. Lin, P. Stojanov, R. Jing, R. W. Park, M.-D. Nazaire, J. Robinson, H.

Thorvaldsdottir, J. Mesirov, P. J. Park, L. Chin, V. Thorsson, S. M. Reynolds, B. Bernard, R. Kreisberg,

J. Lin, L. Iype, R. Bressler, T. Erkkilä, M. Gundapuneni, Y. Liu, A. Norberg, T. Robinson, D. Yang, W.

Zhang, I. Shmulevich, J. J. de Ronde, N. Schultz, E. Cerami, G. Ciriello, A. P. Goldberg, B. Gross, A.

Jacobsen, J. Gao, B. Kaczkowski, R. Sinha, B. Arman Aksoy, Y. Antipin, B. Reva, R. Shen, B. S. Taylor,

T. A. Chan, M. Ladanyi, C. Sander, R. Akbani, N. Zhang, B. M. Broom, T. Casasent, A. Unruh, C.

Wakefield, S. R. Hamilton, R. Craig Cason, K. A. Baggerly, J. N. Weinstein, D. Haussler, C. C. Benz, J.

M. Stuart, S. C. Benz, J. Zachary Sanborn, C. J. Vaske, J. Zhu, C. Szeto, G. K. Scott, C. Yau, S. Ng, T.

Goldstein, K. Ellrott, E. Collisson, A. E. Cozen, D. Zerbino, C. Wilks, B. Craft, P. Spellman, R. Penny, T.

Shelton, M. Hatfield, S. Morris, P. Yena, C. Shelton, M. Sherman, J. Paulauskis, J. M. Gastier-Foster, J.

Bowen, N. C. Ramirez, A. Black, R. Pyatt, L. Wise, P. White, M. Bertagnolli, J. Brown, T. A. Chan, G.

C. Chu, C. Czerwinski, F. Denstman, R. Dhir, A. Dörner, C. S. Fuchs, J. G. Guillem, M. Iacocca, H. Juhl,

A. Kaufman, B. Kohl Iii, X. Van Le, M. C. Mariano, E. N. Medina, M. Meyers, G. M. Nash, P. B. Paty,

N. Petrelli, B. Rabeno, W. G. Richards, D. Solit, P. Swanson, L. Temple, J. E. Tepper, R. Thorp, E.

Vakiani, M. R. Weiser, J. E. Willis, G. Witkin, Z. Zeng, M. J. Zinner, C. Zornig, M. A. Jensen, R. Sfeir,

A. B. Kahn, A. L. Chu, P. Kothiyal, Z. Wang, E. E. Snyder, J. Pontius, T. D. Pihl, B. Ayala, M. Backus, J.

Walton, J. Whitmore, J. Baboud, D. L. Berton, M. C. Nicholls, D. Srinivasan, R. Raman, S. Girshik, P. A.

Kigonya, S. Alonso, R. N. Sanbhadti, S. P. Barletta, J. M. Greene, D. A. Pot, K. R. Mills Shaw, L. A. L.

Dillon, K. Buetow, T. Davidsen, J. A. Demchok, G. Eley, M. Ferguson, P. Fielding, C. Schaefer, M.

Sheth, L. Yang, M. S. Guyer, B. A. Ozenberger, J. D. Palchik, J. Peterson, H. J. Sofia and E. Thomson

(2012). "Comprehensive molecular characterization of human colon and rectal cancer." Nature 487: 330.

Thommen, D. S., V. H. Koelzer, P. Herzig, A. Roller, M. Trefny, S. Dimeloe, A. Kiialainen, J. Hanhart,

C. Schill, C. Hess, S. Savic Prince, M. Wiese, D. Lardinois, P. C. Ho, C. Klein, V. Karanikas, K. D.

Mertz, T. N. Schumacher and A. Zippelius (2018). "A transcriptionally and functionally distinct PD-1(+)

CD8(+) T cell pool with predictive potential in non-small-cell lung cancer treated with PD-1 blockade."

Nat Med 24(7): 994-1004.

Tomasetti, C., L. Li and B. Vogelstein (2017). "Stem cell divisions, somatic mutations, cancer etiology,

and cancer prevention." Science 355(6331): 1330-1334.

Topalian, S. L., F. S. Hodi, J. R. Brahmer, S. N. Gettinger, D. C. Smith, D. F. McDermott, J. D.

Powderly, R. D. Carvajal, J. A. Sosman, M. B. Atkins, P. D. Leming, D. R. Spigel, S. J. Antonia, L. Horn,

C. G. Drake, D. M. Pardoll, L. Chen, W. H. Sharfman, R. A. Anders, J. M. Taube, T. L. McMiller, H. Xu,

A. J. Korman, M. Jure-Kunkel, S. Agrawal, D. McDonald, G. D. Kollia, A. Gupta, J. M. Wigginton and

M. Sznol (2012). "Safety, activity, and immune correlates of anti-PD-1 antibody in cancer." N Engl J

Med 366(26): 2443-2454.

Topalian, S. L., F. S. Hodi, J. R. Brahmer, S. N. Gettinger, D. C. Smith, D. F. McDermott, J. D.

Powderly, J. A. Sosman, M. B. Atkins, P. D. Leming, D. R. Spigel, S. J. Antonia, A. Drilon, J. D.

Wolchok, R. D. Carvajal, M. B. McHenry, F. Hosein, C. T. Harbison, J. F. Grosso and M. Sznol (2019).

"Five-Year Survival and Correlates Among Patients With Advanced Melanoma, Renal Cell Carcinoma,

or Non-Small Cell Lung Cancer Treated With Nivolumab." JAMA Oncol 5(10): 1411-1420.

Touat, M., Y. Y. Li, A. N. Boynton, L. F. Spurr, J. B. Iorgulescu, C. L. Bohrson, I. Cortes-Ciriano, C.

Birzu, J. E. Geduldig, K. Pelton, M. J. Lim-Fat, S. Pal, R. Ferrer-Luna, S. H. Ramkissoon, F. Dubois, C.

175

Bellamy, N. Currimjee, J. Bonardi, K. Qian, P. Ho, S. Malinowski, L. Taquet, R. E. Jones, A. Shetty, K.-

H. Chow, R. Sharaf, D. Pavlick, L. A. Albacker, N. Younan, C. Baldini, M. Verreault, M. Giry, E.

Guillerm, S. Ammari, F. Beuvon, K. Mokhtari, A. Alentorn, C. Dehais, C. Houillier, F. Laigle-Donadey,

D. Psimaras, E. Q. Lee, L. Nayak, J. R. McFaline-Figueroa, A. Carpentier, P. Cornu, L. Capelle, B.

Mathon, J. S. Barnholtz-Sloan, A. Chakravarti, W. L. Bi, E. A. Chiocca, K. P. Fehnel, S. Alexandrescu, S.

N. Chi, D. Haas-Kogan, T. T. Batchelor, G. M. Frampton, B. M. Alexander, R. Y. Huang, A. H. Ligon, F.

Coulet, J.-Y. Delattre, K. Hoang-Xuan, D. M. Meredith, S. Santagata, A. Duval, M. Sanson, A. D.

Cherniack, P. Y. Wen, D. A. Reardon, A. Marabelle, P. J. Park, A. Idbaih, R. Beroukhim, P.

Bandopadhayay, F. Bielle and K. L. Ligon (2020). "Mechanisms and therapeutic implications of

hypermutation in gliomas." Nature 580(7804): 517-523.

Troiani, T., L. Vecchione, E. Martinelli, A. Capasso, S. Costantino, L. P. Ciuffreda, F. Morgillo, D.

Vitagliano, E. D'Aiuto, R. De Palma, S. Tejpar, E. Van Cutsem, M. De Lorenzi, M. Caraglia, L. Berrino

and F. Ciardiello (2012). "Intrinsic resistance to selumetinib, a selective inhibitor of MEK1/2, by cAMP-

dependent protein kinase A activation in human lung and colorectal cancer cells." Br J Cancer 106(10):

1648-1659.

Tronche, F., C. Kellendonk, O. Kretz, P. Gass, K. Anlag, P. C. Orban, R. Bock, R. Klein and G. Schütz

(1999). "Disruption of the glucocorticoid receptor gene in the nervous system results in reduced anxiety."

Nature Genetics 23(1): 99-103.

Turcot, J., J. P. Despres and F. St Pierre (1959). "Malignant tumors of the central nervous system

associated with familial polyposis of the colon: report of two cases." Dis Colon Rectum 2: 465-468.

Valle, L., E. Hernández-Illán, F. Bellido, G. Aiza, A. Castillejo, M. I. Castillejo, M. Navarro, N. Seguí, G.

Vargas, C. Guarinos, M. Juarez, X. Sanjuán, S. Iglesias, C. Alenda, C. Egoavil, Á. Segura, M. J. Juan, M.

Rodriguez-Soler, J. Brunet, S. González, R. Jover, C. Lázaro, G. Capellá, M. Pineda, J. L. Soto and I.

Blanco (2014). "New insights into POLE and POLD1 germline mutations in familial colorectal cancer

and polyposis." Hum Mol Genet 23(13): 3506-3512.

Vallois, D., M. P. Dobay, R. D. Morin, F. Lemonnier, E. Missiaglia, M. Juilland, J. Iwaszkiewicz, V.

Fataccioli, B. Bisig, A. Roberti, J. Grewal, J. Bruneau, B. Fabiani, A. Martin, C. Bonnet, O. Michielin, J.

P. Jais, M. Figeac, O. A. Bernard, M. Delorenzi, C. Haioun, O. Tournilhac, M. Thome, R. D. Gascoyne,

P. Gaulard and L. de Leval (2016). "Activating mutations in genes related to TCR signaling in

angioimmunoblastic and other follicular helper T-cell-derived lymphomas." Blood 128(11): 1490-1502.

Van Allen, E. M., D. Miao, B. Schilling, S. A. Shukla, C. Blank, L. Zimmer, A. Sucker, U. Hillen, M. H.

Foppen, S. M. Goldinger, J. Utikal, J. C. Hassel, B. Weide, K. C. Kaehler, C. Loquai, P. Mohr, R.

Gutzmer, R. Dummer, S. Gabriel, C. J. Wu, D. Schadendorf and L. A. Garraway (2015). "Genomic

correlates of response to CTLA-4 blockade in metastatic melanoma." Science 350(6257): 207-211.

Van Gassen, S., B. Callebaut, M. J. Van Helden, B. N. Lambrecht, P. Demeester, T. Dhaene and Y. Saeys

(2015). "FlowSOM: Using self-organizing maps for visualization and interpretation of cytometry data."

Cytometry A 87(7): 636-645.

van Thuijl, H. F., T. Mazor, B. E. Johnson, S. D. Fouse, K. Aihara, C. Hong, A. Malmström, M.

Hallbeck, J. J. Heimans, J. J. Kloezeman, M. Stenmark-Askmalm, M. L. M. Lamfers, N. Saito, H.

Aburatani, A. Mukasa, M. S. Berger, P. Söderkvist, B. S. Taylor, A. M. Molinaro, P. Wesseling, J. C.

Reijneveld, S. M. Chang, B. Ylstra and J. F. Costello (2015). "Evolution of DNA repair defects during

malignant progression of low-grade gliomas after temozolomide treatment." Acta Neuropathologica

129(4): 597-607.

van Wietmarschen, N., S. Sridharan, W. J. Nathan, A. Tubbs, E. M. Chan, E. Callen, W. Wu, F. Belinky,

V. Tripathi, N. Wong, K. Foster, J. Noorbakhsh, K. Garimella, A. Cruz-Migoni, J. A. Sommers, Y.

Huang, A. A. Borah, J. T. Smith, J. Kalfon, N. Kesten, K. Fugger, R. L. Walker, E. Dolzhenko, M. A.

176

Eberle, B. E. Hayward, K. Usdin, C. H. Freudenreich, R. M. Brosh, Jr., S. C. West, P. J. McHugh, P. S.

Meltzer, A. J. Bass and A. Nussenzweig (2020). "Repeat expansions confer WRN dependence in

microsatellite-unstable cancers." Nature 586(7828): 292-298.

Vandeven, N., C. W. Lewis, V. Makarov, N. Riaz, K. G. Paulson, D. Hippe, A. Bestick, R. Doumani, T.

Marx, S. Takagishi, T. A. Chan, J. Choi and P. Nghiem (2018). "Merkel Cell Carcinoma Patients

Presenting Without a Primary Lesion Have Elevated Markers of Immunity, Higher Tumor Mutation

Burden, and Improved Survival." Clin Cancer Res 24(4): 963-971.

Vasen, H. F., Z. Ghorbanoghli, F. Bourdeaut, O. Cabaret, O. Caron, A. Duval, N. Entz-Werle, Y.

Goldberg, D. Ilencikova, C. P. Kratz, N. Lavoine, J. Loeffen, F. H. Menko, M. Muleris, G. Sebille, C.

Colas, B. Burkhardt, L. Brugieres and K. Wimmer (2014). "Guidelines for surveillance of individuals

with constitutional mismatch repair-deficiency proposed by the European Consortium "Care for CMMR-

D" (C4CMMR-D)." J Med Genet 51(5): 283-293.

Visvader, J. E. (2011). "Cells of origin in cancer." Nature 469(7330): 314-322.

von Loga, K., A. Woolston, M. Punta, L. J. Barber, B. Griffiths, M. Semiannikova, G. Spain, B.

Challoner, K. Fenwick, R. Simon, A. Marx, G. Sauter, S. Lise, N. Matthews and M. Gerlinger (2020).

"Extreme intratumour heterogeneity and driver evolution in mismatch repair deficient gastro-oesophageal

cancer." Nature Communications 11(1): 139.

Walunas, T. L., D. J. Lenschow, C. Y. Bakker, P. S. Linsley, G. J. Freeman, J. M. Green, C. B. Thompson

and J. A. Bluestone (1994). "CTLA-4 can function as a negative regulator of T cell activation." Immunity

1(5): 405-413.

Wang, Q., C. Lasset, F. Desseigne, D. Frappaz, C. Bergeron, C. Navarro, E. Ruano and A. Puisieux

(1999). "Neurofibromatosis and early onset of cancers in hMLH1-deficient children." Cancer Res 59(2):

294-297.

Wartewig, T., Z. Kurgyis, S. Keppler, K. Pechloff, E. Hameister, R. Ollinger, R. Maresch, T. Buch, K.

Steiger, C. Winter, R. Rad and J. Ruland (2017). "PD-1 is a haploinsufficient suppressor of T cell

lymphomagenesis." Nature.

Warthin, A. S. (1913). "HEREDITY WITH REFERENCE TO CARCINOMA: AS SHOWN BY THE

STUDY OF THE CASES EXAMINED IN THE PATHOLOGICAL LABORATORY OF THE

UNIVERSITY OF MICHIGAN, 1895-1913." Archives of Internal Medicine XII(5): 546-555.

Watson, J. D. and F. H. Crick (1953). "Molecular structure of nucleic acids; a structure for deoxyribose

nucleic acid." Nature 171(4356): 737-738.

Watson, P., H. F. A. Vasen, J. P. Mecklin, I. Bernstein, M. Aarnio, H. J. Järvinen, T. Myrhøj, L. Sunde, J.

T. Wijnen and H. T. Lynch (2008). "The risk of extra-colonic, extra-endometrial cancer in the Lynch

syndrome." Int J Cancer 123(2): 444-449.

Weber, J., M. Mandala, M. Del Vecchio, H. J. Gogas, A. M. Arance, C. L. Cowey, S. Dalle, M. Schenker,

V. Chiarion-Sileni, I. Marquez-Rodas, J. J. Grob, M. O. Butler, M. R. Middleton, M. Maio, V. Atkinson,

P. Queirolo, R. Gonzalez, R. R. Kudchadkar, M. Smylie, N. Meyer, L. Mortier, M. B. Atkins, G. V. Long,

S. Bhatia, C. Lebbé, P. Rutkowski, K. Yokota, N. Yamazaki, T. M. Kim, V. de Pril, J. Sabater, A.

Qureshi, J. Larkin and P. A. Ascierto (2017). "Adjuvant Nivolumab versus Ipilimumab in Resected Stage

III or IV Melanoma." N Engl J Med 377(19): 1824-1835.

Wei, S. C., J. H. Levine, A. P. Cogdill, Y. Zhao, N.-A. A. S. Anang, M. C. Andrews, P. Sharma, J. Wang,

J. A. Wargo, D. Pe’er and J. P. Allison (2017). "Distinct Cellular Mechanisms Underlie Anti-CTLA-4 and

Anti-PD-1 Checkpoint Blockade." Cell 170(6): 1120-1133.e1117.

177

Wei, S. C., J. H. Levine, A. P. Cogdill, Y. Zhao, N. A. S. Anang, M. C. Andrews, P. Sharma, J. Wang, J.

A. Wargo, D. Pe'er and J. P. Allison (2017). "Distinct Cellular Mechanisms Underlie Anti-CTLA-4 and

Anti-PD-1 Checkpoint Blockade." Cell 170(6): 1120-1133.e1117.

Wimmer, K., C. P. Kratz, H. F. A. Vasen, O. Caron, C. Colas, N. Entz-Werle, A.-M. Gerdes, Y.

Goldberg, D. Ilencikova, M. Muleris, A. Duval, N. Lavoine, C. Ruiz-Ponte, I. Slavc, B. Burkhardt and L.

Brugieres (2014). "Diagnostic criteria for constitutional mismatch repair deficiency syndrome:

suggestions of the European consortium ‘Care for CMMRD’ (C4CMMRD)." Journal of Medical Genetics

51(6): 355-365.

Wojciechowicz, K., E. Cantelli, B. Van Gerwen, M. Plug, A. Van Der Wal, E. Delzenne-Goette, J. Y.

Song, S. De Vries, M. Dekker and H. Te Riele (2014). "Temozolomide increases the number of mismatch

repair-deficient intestinal crypts and accelerates tumorigenesis in a mouse model of Lynch syndrome."

Gastroenterology 147(5): 1064-1072.e1065.

Xing, X., D. P. Kane, C. R. Bulock, E. A. Moore, S. Sharma, A. Chabes and P. V. Shcherbakova (2019).

"A recurrent cancer-associated substitution in DNA polymerase ε produces a hyperactive enzyme." Nat

Commun 10(1): 374.

Yaeger, R., Z. Yao, D. M. Hyman, J. F. Hechtman, E. Vakiani, H. Zhao, W. Su, L. Wang, A. Joelson, A.

Cercek, J. Baselga, E. de Stanchina, L. Saltz, M. F. Berger, D. B. Solit and N. Rosen (2017).

"Mechanisms of Acquired Resistance to BRAF V600E Inhibition in Colon Cancers Converge on RAF

Dimerization and Are Sensitive to Its Inhibition." Cancer Res 77(23): 6513-6523.

Yan, T., J. E. Schupp, H. S. Hwang, M. W. Wagner, S. E. Berry, S. Strickfaden, M. L. Veigl, W. D.

Sedwick, D. A. Boothman and T. J. Kinsella (2001). "Loss of DNA mismatch repair imparts defective

cdc2 signaling and G(2) arrest responses without altering survival after ionizing radiation." Cancer Res

61(22): 8290-8297.

Yao, Z., R. Yaeger, V. S. Rodrik-Outmezguine, A. Tao, N. M. Torres, M. T. Chang, M. Drosten, H.

Zhao, F. Cecchi, T. Hembrough, J. Michels, H. Baumert, L. Miles, N. M. Campbell, E. de Stanchina, D.

B. Solit, M. Barbacid, B. S. Taylor and N. Rosen (2017). "Tumours with class 3 BRAF mutants are

sensitive to the inhibition of activated RAS." Nature 548(7666): 234-238.

Yarchoan, M., A. Hopkins and E. M. Jaffee (2017). "Tumor Mutational Burden and Response Rate to

PD-1 Inhibition." N Engl J Med 377(25): 2500-2501.

Yip, S., J. Miao, D. P. Cahill, A. J. Iafrate, K. Aldape, C. L. Nutt and D. N. Louis (2009).

"<em>MSH6</em> Mutations Arise in Glioblastomas during Temozolomide Therapy and Mediate

Temozolomide Resistance." Clinical Cancer Research 15(14): 4622-4629.

Yoshida, R., K. Miyashita, M. Inoue, A. Shimamoto, Z. Yan, A. Egashira, E. Oki, Y. Kakeji, S. Oda and

Y. Maehara (2011). "Concurrent genetic alterations in DNA polymerase proofreading and mismatch

repair in human colorectal cancer." Eur J Hum Genet 19(3): 320-325.

Zahurancik, W. J., A. G. Baranovskiy, T. H. Tahirov and Z. Suo (2015). "Comparison of the kinetic

parameters of the truncated catalytic subunit and holoenzyme of human DNA polymerase varepsilon."

DNA Repair (Amst) 29: 16-22.

Zahurancik, W. J., S. J. Klein and Z. Suo (2014). "Significant contribution of the 3'-->5' exonuclease

activity to the high fidelity of nucleotide incorporation catalyzed by human DNA polymerase." Nucleic

Acids Res 42(22): 13853-13860.

Zappasodi, R., T. Merghoub and J. D. Wolchok (2018). "Emerging Concepts for Immune Checkpoint

Blockade-Based Combination Therapies." Cancer Cell 34(4): 690.

178

Zawadzka, M., L. E. Rivers, S. P. J. Fancy, C. Zhao, R. Tripathi, F. Jamen, K. Young, A. Goncharevich,

H. Pohl, M. Rizzi, D. H. Rowitch, N. Kessaris, U. Suter, W. D. Richardson and R. J. M. Franklin (2010).

"CNS-Resident Glial Progenitor/Stem Cells Produce Schwann Cells as well as Oligodendrocytes during

Repair of CNS Demyelination." Cell Stem Cell 6(6): 578-590.

Zehir, A., R. Benayed, R. H. Shah, A. Syed, S. Middha, H. R. Kim, P. Srinivasan, J. Gao, D. Chakravarty,

S. M. Devlin, M. D. Hellmann, D. A. Barron, A. M. Schram, M. Hameed, S. Dogan, D. S. Ross, J. F.

Hechtman, D. F. DeLair, J. Yao, D. L. Mandelker, D. T. Cheng, R. Chandramohan, A. S. Mohanty, R. N.

Ptashkin, G. Jayakumaran, M. Prasad, M. H. Syed, A. B. Rema, Z. Y. Liu, K. Nafa, L. Borsu, J.

Sadowska, J. Casanova, R. Bacares, I. J. Kiecka, A. Razumova, J. B. Son, L. Stewart, T. Baldi, K. A.

Mullaney, H. Al-Ahmadie, E. Vakiani, A. A. Abeshouse, A. V. Penson, P. Jonsson, N. Camacho, M. T.

Chang, H. H. Won, B. E. Gross, R. Kundra, Z. J. Heins, H. W. Chen, S. Phillips, H. Zhang, J. Wang, A.

Ochoa, J. Wills, M. Eubank, S. B. Thomas, S. M. Gardos, D. N. Reales, J. Galle, R. Durany, R. Cambria,

W. Abida, A. Cercek, D. R. Feldman, M. M. Gounder, A. A. Hakimi, J. J. Harding, G. Iyer, Y. Y.

Janjigian, E. J. Jordan, C. M. Kelly, M. A. Lowery, L. G. T. Morris, A. M. Omuro, N. Raj, P. Razavi, A.

N. Shoushtari, N. Shukla, T. E. Soumerai, A. M. Varghese, R. Yaeger, J. Coleman, B. Bochner, G. J.

Riely, L. B. Saltz, H. I. Scher, P. J. Sabbatini, M. E. Robson, D. S. Klimstra, B. S. Taylor, J. Baselga, N.

Schultz, D. M. Hyman, M. E. Arcila, D. B. Solit, M. Ladanyi and M. F. Berger (2017). "Mutational

landscape of metastatic cancer revealed from prospective clinical sequencing of 10,000 patients." Nat

Med 23(6): 703-713.

Zhang, J., G. Wu, C. P. Miller, R. G. Tatevossian, J. D. Dalton, B. Tang, W. Orisme, C. Punchihewa, M.

Parker, I. Qaddoumi, F. A. Boop, C. Lu, C. Kandoth, L. Ding, R. Lee, R. Huether, X. Chen, E. Hedlund,

P. Nagahawatte, M. Rusch, K. Boggs, J. Cheng, J. Becksfort, J. Ma, G. Song, Y. Li, L. Wei, J. Wang, S.

Shurtleff, J. Easton, D. Zhao, R. S. Fulton, L. L. Fulton, D. J. Dooling, B. Vadodaria, H. L. Mulder, C.

Tang, K. Ochoa, C. G. Mullighan, A. Gajjar, R. Kriwacki, D. Sheer, R. J. Gilbertson, E. R. Mardis, R. K.

Wilson, J. R. Downing, S. J. Baker and D. W. Ellison (2013). "Whole-genome sequencing identifies

genetic alterations in pediatric low-grade gliomas." Nat Genet 45(6): 602-612.

Zinzani, P., C. Thieblemont, V. Melnichenko, D. Osmanov, K. Bouabdallah, J. Walewski, A. Majlis, L.

Fogliatto, M. D. Caballero Barrigón, B. Christian, Z. Gulbas, M. Özcan, G. A. Salles, M. A. Shipp, A.

Balakumaran, S. Chlosta, A. Chatterjee and P. Armand (2017). "Efficacy and safety of pembrolizumab in

relapsed/refractory primary mediastinal large B-cell lymphoma (rrPMBCL): interim analysis of the

KEYNOTE-170 phase 2 trial." Hematological Oncology 35(S2): 62-63.

179

Appendices

Chapter 2 Supplementary Tables and Figure:

Supplementary Table S1: Summary of likely pathogenic mutations detected in Foundation

Medicine cohort of 1215 pediatric cancers.

Supplementary Table S2: Summary of RAS/MAPK pathway mutations and consequences

detected in 48 RRD cancers.

Supplementary Tables S1 and S2 can be found here:

https://cancerdiscovery.aacrjournals.org/highwire/filestream/49836/field_highwire_adjunct_files/

0/247438_2_supp_6862881_qnb5yg.xlsx

Supplementary Figure S1: Phospho-ERK immunohistochemistry staining of RRD cancers

Immunohistochemical staining of 11 hypermutant replication repair deficient gliomas and 2 hypermutant replication

repair deficient colorectal cancers. Germline status is shown, along with various mutations in RAS/MAPK pathway

genes and their variant allele fractions. Black bars indicate 100μm; while scale bars indicate 50μm.

180

181

182

183

10

11

184

185

Chapter 3 Supplementary Tables:

Supplementary Table S3: Mass cytometry panel

TAG MARKER CLONE VENDOR IDENTIFIER

89 Y CD45 30-F11 Fluidigm 3089005B

115 In CD8b HK1.4 BioLegend 128002

141 Pr CD44 IM7 BioLegend 103002

143 Nd CD4 E50-2440 BD Biosciences

552125

144 Nd MHC class II M5/114.15.2 BioLegend 107602

146 Nd Eomes Dan11mag Thermo Fisher 14-4875-82

147 Sm BCL6 K112-91 BD Biosciences

561520

148 Nd CD11b M1/70 BioLegend 101202

150 Nd IgD 11-26C.2A BioLegend 405737

151 Eu IgM II/41 BD Biosciences

553435

153 Eu ICOS 7E.17G9 BD Biosciences

552437

154 Sm CD22 Cy34.1 In-house In-house

155 Gd CD103 M290 BD Biosciences

553699

156 Gd CD45 30-G12 In-house In-house

158 Gd FOXP3 FJK-16s Thermo Fisher 14-5773-82

159 Tb Sirpa P84 BioLegend 144002

161 Dy CD122 TM-b1 Thermo Fisher 14-1222-85

162 Dy PD-1 RMP1-30 Thermo Fisher 14-9981-85

163 Dy Tbet O4-46 BD Biosciences

561263

164 Dy CD62L MEL-14 BioLegend 104443

166 Er CD19 1D3 BD Biosciences

553783

167 Er CD150 1A8 Fluidigm 3167004B

168 Er KLRG1 2F1 BD Biosciences

562190

169 Tm TCRb H57-597 BioLegend 109202

170 Er CD25 PC61.5 BioLegend 102002

172 Yb Itgb7 FIB504 BioLegend 321218

173 Yb CD95 Jo2 BD Biosciences

554255

174 Yb Lag3 C9B7W Fluidigm 3174019B

175 Lu CD127 A7R34 BioLegend 135002

176 Yb B220 RA3-6B2 BioLegend 103202

186

191 Ir Iridium Fluidigm 201192B

193 Ir Iridium Fluidigm 201192B

195 Pt Cisplatin BioVision 1550-1000

Supplementary Table S4: Summary of POLE germline mutations in human cancer

MUTATION TUMOR TYPE (AGE) SOURCE

E277G Oligodendroglioma (23); endometrial cancer (30)

(Rosner et al., 2018)

T278K CRC (54) (Castellsague et al., 2019)

P282S Melanoma (55) (Aoude et al., 2015)

D287E Melanoma (57) (Aoude et al., 2015)

S297F Adenocarcinoma (20); Glioblastoma (30) This paper (IRRDC)

W347C Melanoma (14) (Aoude et al., 2015)

Q352P Melanoma (54) (Aoude et al., 2015)

N363K Glioblastoma () (Vande Perre et al., 2019, Rohlin et al., 2014)

Glioblastoma ()

V411L CRC (14) (Wimmer et al., 2017)

L424V Glioblastoma (30) This paper (IRRDC)

K425R Melanoma (55) (Aoude et al., 2015)

P436R Astrocytoma (17) This paper (IRRDC)

Adenocarcinoma (17)

P436S CRC (31) (Spier et al., 2015)

R446Q Endometrial () (Church et al., 2013)

A456P Medulloblastoma (5) (Lindsay et al., 2019)

Y458F CRC (38) (Hansen et al., 2015)

Y458N Adenocarcinoma (25) This paper (IRRDC)

V460M Melanoma (29) (Aoude et al., 2015)

V474I CRC (55) (Esteban-Jurado et al., 2017)

I515M Melanoma (36) (Aoude et al., 2015)

Q520R CRC (45) (Aoude et al., 2015)

A895T CRC (29) (Kayser et al., 2018)

187

Copyright Acknowledgements

Figure 1.1-1 (A) – modified from article in Seminars in Cancer Biology

188

Figure 1.1-1 (B) modified from article in Cell Research

189

Figure 1.3-2 reproduced from article in Nature Reviews Cancer

190

Figure 1.5-1 (A) reproduced from PNAS

191

Figure 1.5-1 (B) reproduced from American Society for clinical Investigation