Strategies of Cancer Immunotherapy: Model of Triple Negative Breast Cancer

Strategies of Cancer Immunotherapy : Model of Triple Negative Breast Cancer Masae Kishi

To cite this version:

Masae Kishi. Strategies of Cancer Immunotherapy : Model of Triple Negative Breast Cancer. Cancer. Université Paris Saclay (COmUE), 2019. English. NNT : 2019SACLS070. tel-02527172

HAL Id: tel-02527172 https://tel.archives-ouvertes.fr/tel-02527172 Submitted on 1 Apr 2020

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés.

Strategies of Cancer Immunotherapy: Model of Triple Negative Breast Cancer

NNT: 2019SACLS070 NNT:

Thèse de doctorat de l'Université Paris-Saclay Préparée à l’Université Paris-Sud

École doctorale n°582 : cancérologie : biologie - médecine - santé (CBMS) Spécialité de doctorat : Sciences de la vie et de la santé

Thèse présentée et soutenue à Villejuif, le 15 mars 2019, par

Masae KISHI HERONT

Composition du Jury :

Fawzia LOUACHE Professeur, Université Paris-Saclay (INSERM U1170) Président

Anne CAIGNARD Directeur de Recherche, Institut Cochin (INSERM U1016) Rapporteur

Christelle MONVILLE Professeur, I-STEM (INSERM / UEVE U861) Rapporteur

Franck PAGES Professeur, Université Paris Descartes (INSERM U1138) Examinateur

Virginie DANGLES-MARIE Maître de Conférences, Institut Curie Examinateur

Frank GRISCELLI Professeur, Université Paris Descartes (INSERM U935) Directeur de thèse

Annelise BENNACEUR GRISCELLI Professeur, Université Paris Sud / Paris Saclay (INSERM U935) Co-Directeur de thèse

TABLE OF CONTENTS

Abstract ··········································································································· a List of figures ··································································································· c List of tables ···································································································· e List of annexes ··································································································· g Abbreviations ·································································································· i Acknowledgements ···························································································· k

INTRODUCTION ······························································································ 1 1 Chapter 1: Triple Negative Breast Cancer ······························································ 1 1.1 Human Mammary gland ·········································································· 1 1.2 Epidemiology of breast cancer ··································································· 2 1.3 Classification of breast cancer ··································································· 3 1.3.1 Molecular subtypes of breast cancer ······················································ 5 1.3.2 Triple negative breast cancer (TNBC) and Basal-like subtype ························ 9 1.3.3 Immune system and TNBC ······························································· 10 1.4 Breast cancer stem cells ········································································· 13 1.4.1 Cancer Stem Cell (CSC) ·································································· 13 1.4.2 Embryonic stem cell-like gene signature in TNBC ··································· 15 1.4.3 Stemness molecular profile in breast cancer ··········································· 19 1.4.4 Breast cancer stem cells ··································································· 24 1.4.5 Tumor evolution of TNBC ······························································· 25 1.5 Treatment for TNBC············································································· 28

2 Chapter 2: Immunity and cancer ······································································ 32 2.1 Tumor microenvironment (TME) ······························································ 34 2.1.1 Overview ···················································································· 34 2.1.2 Immune context in breast cancer ······················································· 37 2.2 Inflammation and cancer ········································································ 38 2.3 Innate immunity ·················································································· 39 2.3.1 Macrophages ··············································································· 41 2.3.2 Dendritic cells ·············································································· 41 2.3.3 Myeloid-derived suppressor cells (MDSC) ············································ 42 2.3.3.1 Murine MDSCs ···································································· 44 2.3.3.2 Human MDSCs ···································································· 44 2.3.3.3 MDSC infiltration in tumor ······················································ 44 2.3.3.4 Mechanisms of immunomodulatory functions by MDSCs ·················· 45 2.3.3.5 MDSC triggers tumor metastasis ················································ 46

2.3.3.6 Re-education potential of MDSCs ·············································· 46 2.3.4 NK cells ····················································································· 47 2.4 Adaptive immunity ·············································································· 48 2.4.1 T cells ························································································ 49 2.4.1.1 Cytotoxic T lymphocyte (CTL) ·················································· 51 2.4.1.2 Regulatory T cells ································································· 53 2.4.1.2.1 Differentiation of Tregs ··················································· 53 2.4.1.2.2 Function of Tregs ·························································· 54 2.4.1.2.3 nTregs and iTregs in cancer ·············································· 55 2.4.1.3 Helper T cells ······································································ 56 2.4.1.3.1 Type 1 helper T (Th1) cells ·············································· 56 2.4.1.3.2 Type 2 helper T (Th2) cells ·············································· 57 2.4.1.3.3 T helper 9 (Th9) and T helper 17 (Th17) cells ························ 58 2.4.1.3.4 Plasticity of helper T cells ················································ 58 2.4.1.3.5 TCR signal strength in mature T cell differentiation ················· 58 2.4.1.4 T cell memory ······································································ 59 2.4.1.5 Cross reaction, allo/xeno reaction by T cells ·································· 60 1.1.1.1.1 Reactions against allo/xeno products ··································· 60 1.1.1.1.2 Structural evidences ······················································· 61 2.4.2 B cells ······················································································· 62 2.5 Cytokines ·························································································· 63 2.6 Chemokines ······················································································· 64

3 Chapter 3: Cancer immunotherapy ···································································· 67 3.1 History ····························································································· 67 3.2 Therapeutic approaches ········································································· 69 3.2.1 Cytokines and Vaccines ··································································· 69 3.2.1.1 Protein/peptide-based vaccines ·················································· 70 3.2.1.2 DNA/RNA based vaccines ······················································· 71 3.2.1.3 Viral based vaccines ······························································· 71 3.2.1.4 Immunological perspective behind cancer vaccine ··························· 71 3.2.2 Adoptive cell therapy ······································································ 72 3.2.2.1 Chimeric antigen receptor T (CAR T) cells ···································· 73 3.2.3 Immune check point inhibitors ··························································· 74 3.2.3.1 Anti-CTLA4 antibodies ··························································· 79 3.2.3.2 Anti-PD1/PDL1 antibodies ······················································· 79 3.3 Cancer antigens ·················································································· 80 3.3.1 Tumor-associated antigens (TAAs)······················································ 81 3.3.2 Differentiation antigens ··································································· 81 3.3.3 Cancer/testis antigens (CTAs) ···························································· 81 3.3.4 Oncofetal antigens ········································································· 81 3.3.5 Tumor specific antigens (TSAs) ························································· 84 3.3.6 Neo-antigens (mutant peptides or protein) ············································· 84

3.4 Vaccine adjuvants and immune response modifiers ········································· 85 3.4.1 Epigenetic modulation of immune functions ·········································· 87 3.4.1.1 Histone modification ······························································ 87 3.4.1.2 The role of HDACs in immune system and the effects of HDAC inhibitors (HDACis) ······························································· 88 3.5 Immunotherapy in breast cancer treatment ················································· 89

4 Chapter 4: Immunotherapy strategies using PSCs ·················································· 94 4.1 Characteristics of Pluripotent stem cells ······················································ 94 4.2 iPS cells ··························································································· 95 4.2.1 Generation of iPS cells ···································································· 95 4.2.2 Characterization of ESC and iPSC ······················································ 95 4.3 Immunotherapy strategies using PSCs ························································ 97 4.3.1 PSC as a source of cancer immunotherapy ············································· 97 4.3.1.1 Generation of DCs from iPSCs ·················································· 97 4.3.1.2 Generation of T cells from iPSCs ··············································· 98 4.3.1.3 Generation of NK cells from iPSCs ············································· 99 4.3.1.4 Generation of NKT cells from iPSCs ··········································· 99 4.3.2 Cancer vaccines using pluripotent stem cell ·········································· 97

RESULTS ······································································································ 103 A- In vitro characterization of TNBC CSCs from 4T1 cell line ································· 105 B- Establishment of TNBC model ····································································· 115 C- Generation of transgene-free iPS cells from Balb/c mouse ··································· 125 D- Article: Pluripotent Stem Cell-Based Cancer Vaccines Targeting Tumor Microenvironment and Metastatic Spread ············································ 132

DISCUSSION Major findings and perspective ········································································ 196 General conclusion ······················································································ 199

ANNEXES ··································································································· 203 REFERENCES ····························································································· 232

ABSTRACT

Most of Triple Negative Breast cancers (TNBC) are basal-like, aggressive, and high grade. They are often at grade III at diagnosis and possess mesenchymal or embryonic-type gene signature. TNBCs are difficult to eradicate due in part to the presence of a rare population called "cancer stem cells (CSCs)" showing extensive capacity of self-renewal representing a property sharing with stem cells. Pluripotent stem cells (PSCs) and CSCs have been shown to share antigenic determinants and several teams have shown that it is possible to use PSCs as a vaccine material to induce a cellular and/or humoral immune response against CSCs. PSC-based vaccine (PSCV) approach has been previously evaluated on murine models of lung, colon and ovarian cancer.

My thesis work consisted in setting up a metastatic TNBC model in mice (syngeneic Balb/c) and evaluating the antitumor effect of PSCV in this model. 4T1 immunocompetent murine model was used as a model which closely mimic metastatic human TNBC. In relation to this research, the following subjects are discussed in this thesis: the description of the tumor microenvironment, the relationship between cancer and stem cells, and the cellular and molecular mechanisms of tumor immunosuppression in the context of TNBC. Knowledge of the mechanisms of immunosurveillance is a major element to understand before proposing any new approaches to immunotherapies.

We have demonstrated that PSCV in combination with a histone deacetylase inhibitor prevents the establishment of 4T1 tumors by developing a strong anti-tumor immune response. We have shown that this anti-tumor effect is associated with a reduction of regulatory T cells and Myeloid-derived suppressor cells, and an increase of cytotoxic CD8+ T cells in the tumor and spleen. In addition, the anti-tumor response was associated with a drastic reduction in metastatic spread and an improved survival rate in 4T1 breast cancer model. This active immunotherapy strategy was also effective in inhibiting the establishment of CSCs, with inducing a major modification of the tumor microenvironment.

By analysis of gene expression by transcriptome and RT-qPCR, we were able to show the presence of a significant increase of CXCL9, CXCL10 and CXCL13 transcripts in the tumors of mice that received the combined treatment. The chemokine ligand CXCL9 and CXCL10 are chemo-attractants for CD8+ T cells, Th1 cells and NK cells. CXCL13 plays an important role in attracting lymphocytes to the tumor site and forming tertiary lymphoid structures. These results reveal that a part of the underlying mechanism of the observed

antitumor effect was by polarizing the tumor microenvironment to the attraction of T cells to the tumor and turning it to an active "hot" state. with the increased presence of effector immune cells.

We have also shown that the candidate vaccine is able to exert long-term effect, which was demonstrated by inhibiting tumor establishment at 6 and 9 months after vaccination with PSCV, showing the possibility of inducing long-lasting adaptive immunity with activation of the immune system and inhibition of suppressor cells.

We also observed the emergence of a small population of CSC within 4T1 tumors once implanted to the mammary fat pad of mice, underlying that the clinical strategy must consider the influence of microenvironment on cancer cells. Further, the increase of CSC in tumors is time-dependent after tumor implantation, emphasizing the importance of intervening as soon as possible to eradicate a possible presence of CSC in tumors.

This work proposes PSCV to be used for prophylactic purpose, or as a prevention of recurrence in patients with TNBC after the surgery.

LIST OF FIGURES

Figure 1: Anatomical and histological illustrations of human mammary gland ························ 1 Figure 2: Differentiation hierarchy model from mammary stem cells to the differentiated cells. ·· 2 Figure 3: Breast cancer incidents and mortality. ····························································· 3 Figure 4: Histological stratification of breast cancer. ························································ 4 Figure 5: Molecular intrinsic subtypes of breast cancer. Gene clustering, characters and prognosis ············································································································ 7 Figure 6: Distribution of dinstinct molecular subtype in ··················································· 9 Figure 7: The prevalence of somatic mutations across human cancer types ························· 11 Figure 8: CSC model and Clonal evolution model ··························································· 13 Figure 10: Poorly differentiated breast cancers show an ES-like enrichment pattern. ·············· 17 Figure 11: ES like gene signature in breast cancers ························································ 19 Figure 13: Diversity and existence of multi-subtypes within DCIS tumor ······························ 26 Figure 14: Copy number Evolution in Clonal Extinction and Resistant Patients ····················· 27 Figure 15: Hallmarks of cancer: The revised perspective ················································· 33 Figure 16: The microenvironment supports metastatic dissemination and colonization at secondary sites ······················································································· 34 Figure 17: Changes that occur in myeloid cells in cancer ················································· 43 Figure 19: Peripheral tolerance ·················································································· 50 Figure 20: Relationship between TCR affinity and T cell activity for CD8 and CD4 T cells. ······· 51 Figure 21: Interaction of a cytotoxic lymphocyte with a target cell ······································· 52 Figure 22: CD4+ thymocyte fate and regulatory T cell differentiation by TCR signal strength ··· 54 Figure 23: T helper-cell differentiation to Th1, Th2 and Th17 ············································· 57 Figure 24: The promotion of tumor immunity by chemokines ········································· 65 Figure 25: History of cancer immunology and immunotherapy. ·········································· 68 Figure 26: Engineered T cells: design of TCR versus CAR T cells. ····································· 73 Figure 27: Multiple co-stimulatory and inhibitory interactions regulate T cell responses ··········· 75 Figure 28: Customizing a patient-specific cancer vaccine. ················································ 84 Figure 29: Morphology change after 4 days culture with cytokines.····································· 109 Figure 30: The representative photos of the mammosphere culture. ··································· 110 Figure 31: Mammosphere forming efficiency (MFE) and large sphere count ························· 111 Figure 32: Representative photo of MF culture of MF (N) and MFptt + TNF+TGF (N) ············· 111 Figure 33: CD44+CD24-/low population of 4T1 cells in MF and MF ptt+TNF+TGF cultures ······· 112 Figure 34: ALDH+ of 4T1 cells in MF and MF ptt + TNF + TGF cultures ······························ 113

Figure 35: MHC I expression in bulk 4T1 and 4T1 mammosphere······································ 114 Figure 36: Tumor volume in 4T1 tumorigenesis study ······················································ 117 Figure 37: Pathological observations in primary tumors and spleen ···································· 117 Figure 38: GFP intensity measurement by FACS ···························································· 118 Figure 39: Cloning in 98 well plate: One 4T1-GFP-Luc cell in a well under (a) microscope and (b) fluorescent microscope. (c, d) The cell formed a small colony 4 days after seeding. 119 Figure 40: GFP expressions of 4T1-GFP-Luc clones (above) and subclones (below). ············ 120 Figure 41: GFP positive rate and median value of each clones. ········································· 121 Figure 42: Analysis of the clones by bioluminescent imaging by IVIS. ································· 122 Figure 43: Tumor volume, tumor weight and bioluminescent images of 4T1-GFP-Luc clones ··· 123 Figure 44: Metastasis to the organs analyzed by bioluminescence detection of 4T1-GFP-Luc clones ···································································································· 124 Figure 45: Reprogramming of primary fibroblasts obtained from Balb/c mouse. ··················· 128 Figure 46: iPSC colonies observed after the infection of Adenovirus expressing Cre recombinase and GFP. ································································································ 129 Figure 47: PCR results after the process of transgene deletion ········································· 130 Figure 48: Characterization of iPSC clones. ·································································· 131 Figure 49: CSC marker CD44+CD24low/- expression in 4T1 cells in vitro and in vivo ············· 221

LIST OF TABLES

Table 1: TNM staging ································································································ 4 Table 2: Scoring criteria for histological grade determination of breast cancer by Nottingham Histological Score system. ··············································································· 5 Table 3: Character of molecular subtypes of breast cancer ················································ 8 Table 4: Gene analysis on mutational load and TIL according the molecular subtypes ············· 12 Table 5: Important molecular pathways and transcription programs in breast CSC ·················· 20 Table 6: Systemic treatment recommendations for early breast cancer subtypes ···················· 29 Table 7: Immune cell population in TME ········································································ 35 Table 8: Toll like receptors and their ligands ··································································· 40 Table 9: Regulation of MDSCs ···················································································· 45 Table 10: Summary of the biological and molecular functions of T-cell costimulatory molecules · 76 Table 11: Summary of FDA approvals for immune checkpoint blockers ································ 78 Table 12: Types of tumor antigens and their advantages and disadvantages as therapeutic targets ·················································································································· 83 Table 13: TLR ligands used for cancer immunotherapy ····················································· 86 Table 14: Clinical trials of immune check point inhibitors in breast cancer ····························· 90 Table 15: Clinical trials of vaccine for the prevention of breast cancer recurrence ··················· 91 Table 16: Clinical trials of therapeutic cancer vaccine in breast cancer ································· 92 Table 17: Different methods of reprogramming factor delivery to generate iPSC ··················· 96 Table 18: Pluripotent stem cell- based vaccine studies in animal models ······························ 102 Table 19: Results of RT-PCR performed on different cancer cell lines demonstrating the expression of pluripotent markers. ···················································································· 107 Table 20: Frequency (%) of MHC II positive DCs and level of the expression ························· 226

LIST OF ANNEXES

Annex 1: Other factors consists of immune tumor microenvironment ··································· 203 Annex 2: Immunological effect of conventional chemotherapeutics ······································ 212 Annex 3: Immunological effect of targeted anticancer agents ············································· 215 Annex 4: CSC, stem cell or mesenchymal stem cell markers tested and those descriptions ······ 217 Annex 5: Primer sequences used for RT-PCR ································································ 219 Annex 6: List of the antibodies used in the experiments ···················································· 220 Annex 7: Supplementary data 1 (CSC marker CD44+CD24low/- expression in 4T1) ·················· 221 Annex 8: Supplementary data 2 (Dendritic cell vaccination loaded with PSC antigens) ············ 222 Annex 9: Résumé substantiel ···················································································· 232

ABBREVIATIONS

ADCC : antibody-dependent cell-mediated cytotoxicity AFP : α-fetoprotein AJCC : American Joint Committee on Cancer ALDH : Aldehyde dehydrogenases 1 APC : antigen-presenting cell ARG1 : arginase 1 BC : Breast cancer BLBC : Basal-like breast cancer BRCA1/2 : breast cancer susceptibility gene 1/2 CAF : cancer associated fibroblast CART : Chimeric Antigen Receptor T cells. CCL : chemokine CEA : carcinoembryonic antigen CK : cytokeratin CSC : Cancer stem cell CSF-1 : colony-stimulating factor 1 CTLA4 : cytotoxic T-lymphocyte–associated antigen 4 CXCL : chemokine (C-X-C motif) ligand CXCL9 : CXC-chemokine ligand 9 CXCR : CXC-chemokine receptor DAMP : Damage-associated molecular pattern DC : dendritic cells DCIS : Ductal carcinoma in situ EGFR : epidermal growth factor receptor EMT : epithelial mesenchymal transition ER : Estrogen receptor ESC : embryonic stem cell ESMO : European Society for Medical Oncology FACS : fluorescence-activated cell sorting FDA : Food and Drug Administration FGF2 : fibroblast growth factor 2 GM-CSF : granulocyte–macrophage colony-stimulating factor HER2 : human epidermal growth factor receptor 2 HLA : Human Leukocyte Antigen CPI : Immune check point inhibitor IDC : invasive ductal carcinoma IFNγ : interferon-γ IL : Interleukin ILC : Innate lymphoid cell iNKT : invariant natural killer T cell iPSC : induced pluripotent stem cell iTreg : induced regulatory T cell LIF : leukemia inhibitory factor

MAIT : mucosal-associated invariant T cells M-CSF : macrophage-colony-stimulating factor MDSC : Myeloid-derived suppressor cell MHC : major histocompatibility complex MICA : MHC class I-related chain A MIF : migration inhibitory factor NK : Natural killer nTreg : natural regulatory T Oct4 : octamer-binding transcription factor 4 OS : overall survival PAMP : Pathogen Associated Molecular pattern pCR : Pathological Complete Response PD-1 : Programmed cell death 1 PD-L1 : programmed cell death 1 ligand 1 PFS : progression-free survival PI3K : phosphatidylinositol 3-kinase pMHC : peptide-MHC complex PR : Progesterone receptor PRC : Polycomb-group repressive protein complex PRRs : Pattern recognition receptors PSCVs : pluripotent stem cell-based vaccines G-MDSC : granulocytic-Myeloid-derived suppressor cell siRNA : small interfering RNA Mo-MDSC : monocytic-Myeloid-derived suppressor cell Sox2 : SRY (sex determining region Y) -box 2 STAT1 : signal transducer and activator of transcription factor-1 TAM : Tumor-Associated Macrophages TCR : T-cell receptor TDLU : Terminal ductal lobular unites Tfh : T follicular helper cells TGF-β : transforming growth factor-β Th : helper T cell TIDCs : tumor infiltrating DCs TILs : tumor-infiltrating lymphocytes TIM3 : T cell immunoglobulin and mucin domain-containing protein 3 TLRs : Toll-like receptors TLSs : tertiary lymphoid structures TME : tumor microenvironment TNBC : Triple negative breast cancer TNF-α : Tumor necrosis factor-α TNM : tumor node metastasis Treg : regulatory T cell

TRM : tissue-resident memory T cells VEGF : vascular endothelial growth factor

ACKNOWLEDGMENTS

First of all, I would like to express my greatest appreciation to Pr. Frank Griscelli for his guidance and continuous support during M2 study and my thesis. His strong knwoledge and intellectual support were precious for the experimental experiences obtained during my work. I express deep gratitude to Pr. Annelise Bennaceur Griscelli for accepting me in the Unit despite of my age, for her global support and for coaching me about the important and innovative ideas and feedbacks.

I am honor to present my thesis to an internationally recognized expert jury in the field of immunology, stem cells and experimental animal cancer models. I am grateful to Dr. Anne Caignard and Pr. Christelle Monville for accepting to analyse my work and for their precious guidances, Pr. Franck Pagès and Dr. Virginie Dangles-Marie for accepting to evaluate my work, and Pr. Fawzia Louache for accepting to be the chairwomen of the jury.

I would like to express great appreciation to my collaborators for this work: Special appreciation to Dr. Afag Asgarova for her huge helps and collaboration works. I enjoyed very much working and sharing the time together, even when the experiments were busy. Dr. Marie-Ghislaine de Goër for her kindness and supports to perform immunological experiments and analysis. Dr. Christophe Desterke for his important expertise in bioinformatics. Dr. Diana Chaker for her kind help in the experiments.

I thank very much Pr. Ali Turhan to welcome me in the Unit and for his precious advises during thesis. I am very thankful to all current and past members of the Team Unit, for their warm encouragements and sharing discussions during my thesis and scientific meetings. Dr. Jérôme Artus for taking his time to help me in stem cell culture for the experiments. Dr. Adlen Foudi, Dr. Herve Aloque for their knowledge in experimental protocols. Olivier Féraud, Dominique Divers and Theodoros Latsis from Stem Cell Bank of the Unit for Embryonic and iPS cell maintenance.

My heartful appreciation to all of my collegues. Lucie, Eva, Jinwook, Sarah, Lucas, Gladys, Patricia, Albert, Sandra, Sabrina, Estelle, Delia, Lara, Tony, Mathieu, Isabelle, Alexandre, Cécile. It was really great to share the time together, giving me always positive energies and wonderfully colorized my memories. I would also like to express my thanks to Dr. Xiao Mei Li, Dr. Yunhua Chang-Marchand and Dr. Sandrine Dulong from team 3 of the Unit and Dr. Ibrahim Casal and Gaëlle Duvallet from animal facility.

I would like to express my appreciation to the mentors, previous colleagues and friends in Japan. Dr. Kaori Mohr. My motivation for studying science is not like now without her. It was really lucky of me to meet her. She taught me important principles in the works that I will not forget. Special thanks to Dr. Masayuki Tanaka and Dr. Kiyoshi Kobayashi for their support that made it possible for me to study in France. Dr. Aoki Emiko. Thank you for being there. With big love. Many thoughts and appreciation to my friends in Japan. Goli, Akiko, Ai, Chika, Nobuko and Waka.

Finally, I would like to express my sincere appreciation to my family. My parents for their tremendous supports. My grandmother, my brother and his wife and children. My uncles. Lastly, I would like to express my deepest and utmost appreciation to my husband and my son for their presences and supports.

INTRODUCTION 1. Chapter 1: Triple negative breast cancer

1.1 Human Mammary gland

The mammary gland is the organ responsible for lactation and one of the few organs which undergoes continued development throughout the lifetime of a woman. Human mammary gland consists of a network of ducts that ends in small ductules constituting the terminal ductal lobular units (TDLU). TDLU consists of terminal ductule and lobule, functional unit of the breast. It is believed that the majority of breast cancers arise in the TDLUs.

The TDLU contains two major types of cells: Inner luminal cells consists of ductal luminal cells, which line the inside of the ducts, and alveolar luminal cells, which secrete milk during lactation and myoepithelial cells, which contract to help excrete milk during lactation (Figure 1) (1). The stroma is composed of adipocytes, fibroblasts and immune cells.

Figure 1: Anatomical and histological illustrations of human mammary gland (Adopted from Dimri 2005) (2)

The mammary epithelial cells develop from mammary stem cells. While the exact differentiation hierarchy is still not completely understood, it is believed that the mammary stem cells give rise to bi-potent progenitor cells, which can differentiate into luminal or, myoepithelial lineage-specific progenitors. The myoepithelial progenitors differentiate into myoepithelial (basal) cells, while the luminal progenitors differentiate

into luminal cells (Figure 2). The biological and molecular characteristics of these cell types consists the basis for the molecular classification of breast cancers into multiple subtypes and will be discussed in more detail later.

Figure 2: Differentiation hierarchy model from mammary stem cells to the differentiated cells. The described cell surface markers are not exhaustive and those of progenitor cells are still in debates. (Modified from Polyak. 2007. Keller. 2010 (3, 4))

1.2 Epidemiology of breast cancer

Breast cancer is the most frequently diagnosed cancer and the most frequent cause of cancer death among women, representing 25% of all cancers in women and 15% of all cancer death in women in the world. Breast cancer is the most common cancer diagnosis in women in 140 countries and the most frequent cause of cancer mortality in 101 countries. Age standardized incidence rates are highest in Western Europe and lowest in East Asia (Figure 3).

In France, breast cancer is the most frequent cancer in the females and 80% of them are diagnosed after 50 years old. The mortality rate is 14.6 in 100,000 females and breast cancer is the first ranked of cancer death in the females in France, followed by lung cancer and colorectal cancer (5). Median age of diagnosis is at 63 years old and the median age

of the death is 73 years old. One out of eight women in France develops this disease in her life.

Mortality rates have been declining in a number of highly developed countries since the late 1980s and early 1990s, a result of a combination of improved detection and earlier diagnosis and more effective treatment regimens (6).

Figure 3: Breast cancer incidents and mortality. Adopted from GLOBOCAN 2012. (6)

1.3 Classifications of breast cancer

Breast cancer is a heterogeneous disease with different histopathological and biological characteristics with variable prognoses and responses to therapies. Breast carcinoma is classified as either non-invasive (carcinoma in situ) or invasive, depending on whether or not the tumor has invaded to grow outside the basal membrane. Invasive carcinomas are cancers in which the altered cells diffuse to surrounding connective tissues and metastasize to distant organs of the body. Ductal carcinoma in situ (DCIS) and lobular carcinoma in situ are the most common forms of precursor lesions. Similarly, invasive ductal carcinoma (IDC) and invasive lobular carcinoma are the most common forms of invasive breast carcinoma. IDCs account for approximately 55% of all breast cancers (Figure 4).

Figure 4: Histological stratification of breast cancer.

The pathological status of breast cancer is made by TNM scores (Table 1) which describe pathological stages including the assessment of primary tumor (T), regional lymph nodes (N) and distant metastasis (M). This classification system provides a basis for survival prognosis, choice of initial therapeutic approaches and evaluation of therapeutic results.

Table 1: TNM staging (AJCC criteria)

Historical Grade represents aggressive potential of the tumor, with the amount of gland formation (differentiation level), the nuclear features (pleomorphism) and the mitotic activity (proliferation level), which classified into from Grade I (low) to III (high). Grade I tumors have a total score of 3-5, Grade II tumors have a total score of 6-7, Grade III tumors have a total score of 8-9 according to the following scoring criteria (Table 2).

Glandular/Tubular Differentiation Nuclear Pleomorphism Mitotic Count* Uniform cells with small nuclei similar in size to normal < 7 mitoses per 10 high Score 1 >75% of tumor forms glands breast epithelial cells power fields Cells larger than normal with open vesicular nuclei, visible 8-15 mitoses per 10 high Score 2 10% to 75% of tumor forms glands nucleoli, and moderate variability in size and shape power fields Cells with vesicular nuclei, prominent nucleoli, marked > 16 mitoses per 10 high Score 3 <10% of tumor forms glands variation in size and shap power fields Table 2: Scoring criteria for histological grade determination of breast cancer by Nottingham Histological Score system. (Johns Hopkins medicine pathology. https://pathology.jhu.edu) *The mitotic count score criteria vary depending on the field diameter of the microscope. The criteria above use a high-power field diameter of 0.52 mm.

The expression of hormonal receptors, such as estrogen receptor (ER) and progesterone receptor (PR), has been used to classify tumors into hormone receptor positive and negative groups. Since the discovery of the role of amplification or overexpression of human epidermal growth factor receptor 2 (HER2 or ERBB2) in a subset of breast tumors, measurement of HER2 status has been added to clinical diagnosis of breast cancer. Thus, based on the immunohistochemical staining of ER, PR, and HER2, breast tumors are classified into three subtypes: hormone receptor positive (ER+, PR+), HER2 positive (HER2+) and triple negative breast cancer (TNBC) (ER-, PR-, HER2-).

1.3.1 Molecular subtypes of breast cancer

The classification of molecular subtypes of breast cancers was proposed in 2000, based on the microarray analysis and its genetic clustering of the tumor specimens from the patients (Figure 5) (7). The subtype was divided into four different subtypes, which are ER+ luminal and three ER- groups including HER2 overexpressing group (HER2+ subtype), basal-like group, and normal breast-like group. The following year, the luminal subtype was divided into at least two groups (luminal A and luminal B)(8). These subtypes have been subsequently confirmed by multiple groups and are now widely accepted in the field.

More recently, another intrinsic subtype termed Claudin-low type has been identified (9) (Figure 5). This type is characterized by the low gene expression of tight junction proteins, claudin 3, 4 and 7 and E-cadherin, and high expression of EMT related genes. For this subtype, the analysis of genomic signature of tumor initiating cells derived from CD44+/CD24-/low-sorted cells and mammospheres obtained from primary human breast tumors was found to be enriched by gene expression in the claudin-low subtype, indicating claudin-low subtype resembles the mammary epithelial stem cell (9).

The characters of each molecular subtype are described in Table 3. To note, Normal breast-like subgroup shows a gene expression pattern similar to normal breast tissue including epithelial cells, non-epithelial cells, and adipose tissue. Since they do not show expression of many proliferation genes, it has been speculated that this subtype could potentially be the consequence of normal tissue contamination (10).

Luminal subtype present luminal cytokeratins, epithelial gene expression whereas Basal-like subtype has myoepithelial (basal) cytokeratin expression. Luminal A subtype tumors are usually hormone receptor positive and have relatively good prognosis. Luminal B subtype tumor are positive for hormone receptors but with low level and tend to show high proliferation profile and worse prognosis than luminal A. HER2-enriched, basal-like and claudin-low subtype show often bad prognosis though emergence of Trastuzumab (anti-HER2 antibody) therapy targeting HER2 is improving the prognosis of HER2-enriched BC patients.

(a) (b)

Prat et al., 2010

Figure 5: Molecular intrinsic subtypes of breast cancer. Gene clustering, characters and prognosis (a) First analysis and proposal by Perou et al. (b) Addition of Luminal B (and C) subtype (c) Addition of Claudin-low subtype. (d) Clinical and pathological characteristics and prognosis of all intrinsic subtypes across three independent breast cancer data sets of UNC337, NKI295 and MDACC133. pCR: Pathological complete response rate after anthracycline/taxane-based chemotherapy. (Adopted from Perou et al., 2000, 2001, 2011)(7-9)

Table 3: Character of molecular subtypes of breast cancer

General Prevalence ER/PR HER2 Description Mutation Treatment/Reaction Grade Luminal A 50-60% + (high) - 1 or 2 low Ki-67, low mitotic count Lowest mutational rate in subtypes Good response to endocrine therapies High luminal cytokeratins KRT8, KRT18 PIK3CA (45%), MAP3K1, TP53, Less responsive to chemotherapy Luminal signature (ESR1, GATA3, FOXA1, GATA3, CDH1, and MAP2K4 XBP1, and MYB) Good prognosis Luminal B 15-20% + (low) - (70- 2 High Ki-67, high mitotic count Mutations in PIK3CA and GATA Do not respond as well to 80%) Worse prognosis than Luminal A Higher frequency of mutations in hormonal intervention therapies + (20- TP53 and frequent mutations in RB1, compared to luminal A, though they 30%) RUNX1, and MALAT1 and high have a better response to expression of MYC neoadjuvant chemotherapy

HER2 10-20% +/- + 2 or 3 Tumors are dependent on HER2 signaling; Often show mutations in TP53 and Good response to certain enriched upon ligand binding, HER proteins undergo PIK3CA but lack many other chemotherapeutic drugs and anti-HER2 dimerization and activation of tyrosine kinase mutations present in luminal tumors targeted therapies such as anti11 activity and downstream signal transduction HER2 antibodies and EGFR/HER2 High proliferation rate tyrosine kinase inhibitors Often show aneuploidy Basal-like 10-20% - - 3 Gene characteristic of myoepithelial cells Mutations in TP53 Better response to chemotherapy (majority) (majority) Basal cytokeratins (5/6, 14, 17), p-cadherins, Overall mutation rate was much compared to luminal subtypes caveolins 1 and 2, nestin, laminin y1, and higher than luminal subtypes. However, many patients annexin A8 PIK3CA pathway including PTEN and relapse in the first five years, and the Often overexpress EGFR INPP4B. ones who do not respond have a Mostly aneuploid and display high histological RB1 mutations compared to other dismal outcome grade, are very proliferative with high mitotic subtypes indices Germline BRCA1 mutation Claudin low 7-14% - - 2 or 3 Low expression of genes involved in cell-cell Poorly described about the Intermediate response to (majority) (majority) Adhesions and tight junctions including claudin mutations in Claudin-low subtype chemotherapy 3,4,7, cingulin, and occluding Immune cell infiltration Enriched for EMT markers Stem cell-like features High ALDH1 8

1.3.2 Triple Negative Breast Cancer (TNBC) and Basal-like subtype

Breast cancer categorized as TNBC is negative for ER, PR and HER2, and consists 12-17% of breast cancer patients (11). 95% of TNBCs are invasive ductal carcinoma (IDC). TNBC contains molecular subtypes of Basal-like and Claudin-low groups. Approximately 80% of Basal-like subtype and 60-70% of Claudin-low subtype are TNBC. In addition, 80-90% of breast cancers harboring BRCA1 and BRCS2 germline mutation are classified into basal like breast cancer (BLBCs) based on the genetic expression profile. Majority of them belongs to TNBC (12). Figure 6 shows ratio of 5 or 6 molecular subtypes in BC with or without claudin-low subtype.

Figure 6: Distribution of dinstinct molecular subtype in breast cancer (Modified from Prat and Perou. 2011, 2013) (13, 14) Left: Ratio of 5 subtypes without claudin-low Right: Ratio of 6 subtypes with claudin-low

TNBCs share many epidemiological, morphological, histopathological, genomic, and clinical features of BLBCs, including high prevalence in younger and African American women, and aggressive tumors with high histological grade. BLBCs are characterized by the expression of genes characteristic of myoepithelial cell including the cytokeratins (5/6, 14, 17), p-cadherins and often overexpress EGFR. BLBCs are mostly aneuploid, display high histological grade, are very proliferative with high mitotic indices, and mostly clinically aggressive. BLBCs are more prevalent in women under the age of 50 and women of African American or Hispanic ethnicity. Some other factors associated with higher risk of BLBC are higher body mass index during premenopausal years, earlier age at menarche, higher parity and shorter duration of breastfeeding.

Genomic studies reported that 80% of BLBCs harbor mutations in TP53. Loss of function from genes involved in DNA damage repair including TP53, RB1, and BRCA is a hallmark of BLBCs (15). The overall mutation rate is much higher in BLBCs compared to luminal subtypes. However, mutations are present in fewer genes and interestingly, mutations present in luminal subtypes are almost absent in the BLBCs except for PIK3CA, which is present in 9% of the cases.

Claudin-low sub-type tumors are characterized by a low expression of genes involved in cell-cell adhesions and tight junctions including claudin 3,4,7, cingulin, and occluding. Similar to basal-like subtype, these tumors display low expression of luminal cytokeratins, ER, HER2, and other luminal specific genes; however, these tumors show relatively low expression of proliferation-associated genes. These tumors express a highly number of immune system response genes from B- and T- lymphoid cells, and suggest a high immune cell infiltration. Other important characteristics of Claudin-low tumors are a stem cell-like gene signatures including Aldehyde dehydrogenases 1 (ALDH1) expression and EMT markers.

TNBC is still a heterogenous group of breast cancers. It was further subtyped by genetic cluster analysis into 6 groups: Basal-Like 1 (BL1) and Basal-Like 2 (BL2) with a high expression of cell proliferation, cell cycle, DNA repair. genes; Immunomodulatory (IM) with a high expression of immune reaction related genes; Mesenchymal (M); Mesenchymal Stem-Like (MSL) with a high expression of TGF-β, EMT genes, proliferation factors, Wnt/β-catenin signaling pathway and Stem cell related genes; Luminal Androgen Receptor (LAR) with a high expression of AR and luminal related genes (16). The IM and MSL subtypes were speculated mostly defined by a high expression of genes likely coming from tumor microenvironment (TME) including immune cells and stromal cells.

1.3.3 Immune system and TNBC

Breast cancers were considered to be relatively less immunogenic than other type of cancers. High mutational load is one of the important determinants that can predict higher spontaneous immunogenicity of the tumor. Breast cancer harbor less mutational load compared to melanoma or lung cancer (17) (Figure 7).

Genomic mutations generate peptides that don’t exist in the repertoire as self- antigens. These new peptides can be recognized by adaptive immune cells as foreign

proteins presented by MHC molecule on the cell surface. Clinical trials using immune check point inhibitor (CPI) therapies in non–small cell lung cancer patients or colo rectal cancer with MSI (Microsatellite Instable) status, revealed neoantigen-specific T cell reactivity. A higher efficacy and response were reported in sub-group of cancer with higher neoantigens and mutation burden (18).

However other determinants are also important to consider such as the cellular phenotypes and metabolism functions of the TME, the amount of tumor infiltrating activated lymphocytes, the expression level of suppressive immunological ligands and the molecular signaling pathways in tumor cells.

Figure 7: The prevalence of somatic mutations across human cancer types Relatively low mutational load in breast cancer compared to Melanoma or Lung squamous (Adopted from Alexandrov. 2013. (17))

Globally, breast cancer harbors a low mutational load and is less immunogenic than lung, melanoma or bladder cancers. However, TNBCs have higher mutational burden than other subtypes (Table 4) and are more immunogenic with a higher number of tumor- infiltrating lymphocytes (TILs) (18). A higher expression of programmed cell death 1 ligand 1 (PD-L1) was reported in TNBC Basal-like subtype compared to non-TNBC luminal subtype (19) (respectively 39% vs 4%). PD-L1 expression is significantly associated with the presence of TILs. High expression of PD-1 and/or PD-L1 is associated with a higher OS (Overall survival) and higher pCR (pathological Complete Response) rate in TNBC (20).

HER2- Luminal A Luminal B enriched Basal-like Average mutational load (mutations per exomic Mb) 25.2 41.4 61.5 50.4 7 10 15 25 Median% stromal TIL (IQR) (5–15) (5–20) (10–30) (14.38–36.25) Table 4: Gene analysis on mutational load and TIL according the molecular subtypes (Adopted from Loi 2013(21))

Other determinants inducing immunogenicity of tumor cells are aberrant expression of proteins by the tumor cells. Some self-epitopes could be “neo-antigens” even without genetic mutation. For exemple, overexpression of proteins such as HER2 can also induce a specific immunogenicity. The amount of HER2 specific T-cells and anti-HER2 antibodies is correlated with the HER2 protein level expression. The detection of HER2 specific antibodies in serum of patients is associated with a significantly improved recurrence free survival compared to patients without detectable HER2 specific antibodies (22). In normal cells, MUC-1 is fully shielded in the fully glycosylated state. Aberrant processing of MUC-1 glycoprotein in tumor cells can triggers their immunogenicity.

In TNBC, high level of immune infiltration is a prognostic factor and predicts a good survival even in patients who haven’t received systemic adjuvant therapies (23). Several studies demonstrated significantly higher pCR rates after chemotherapy in immune-rich as compared to immune-poor TNBC (24). Among TNBC patients who received adjuvant chemotherapies, TIL counts are strongly predictive of cancer-free survival; each 10% increase in TIL count is associated with 18% of reduction risk of distant recurrence (8, 9). At diagnosis, approximately 5% to 15% of TNBCs are classified as lymphocyte predominant with abundant lymphocytes in the stroma. 15% to 20% of TNBC have no lymphocytic infiltration, whereas the majority (65%–80%) harbor low to moderate level of immune cells. Both stromal lymphocytes (residing in the stroma without direct contact with neoplastic cells) and intratumoral lymphocytes provide prognostic and predictive information. Stromal TILs are more abundant and, therefore, can be quantified more reliably (25).

Regarding lymphocyte recruitment into the tumor site, the role of tertiary lymphoid structures (TLSs) has recently gained attentions. TLSs are developed at the periphery or within tumor mass to limit disease progression or as a consequence of effective treatment intervention. They are also the sites of intense activity with mature dendritic cells in

contact with T cells, and germinal-like centers with the expansion of proliferating B cells. In breast cancer, TLSs are sometimes observed within the tumor core proximal to the stroma, and within an individual lymphoid structure (26, 27).

1.4 Breast cancer stem cells

1.4.1 Cancer stem cell (CSC)

Two major theories have been described to explain the expansion and evolution of cancer: The CSC model and Clonal evolution model. Those both theories have been a topic of scientific debates, during the past decades. The CSC model proposes that the growth and progression of cancers are driven by a rare subpopulation of CSC at the origin of a hierarchical organization. The clonal evolution model proposes that multiple genetic and epigenetic modification occur over time in individual cancer cell and in the microenvironment, confer a selective advantage allowing individual tumor clones to out- compete other clones, like Darwin’s theory evolution (Figure 8).

Figure 8: CSC model and Clonal evolution model (Adopted form Adams. 2008(28) )

Debates between both models were already intense between 1950’s and 1970’s (Figure 9). Several experiments using single cell transplantation in animal models succeeded often using ascites liquids. Based on these observations, the concept that the tumor arises from CSC emerged in 1950 (called in this period “Stemline”) (29, 30). Clonal evolution model was proposed by Nowell in 1976 (31).

CSC theory

Clonal evolution theory

Figure 9: Horizon of CSC theory and Clonal evolution theory: (a) Single cell transplantation of murine sarcoma from ascites (b) Single cell transplantation of murine breast cancer and passaging history in mice (c) CSC hypothesis in 1956 mentioned by Makino (Adopted from Ishibashi 1956, Makino 1956, Nowell 1976 (29-31))

The CSC theory was also well documented by John Dick’s group in human Acute Myeloid Leukemia confirming the existence of a primitive hematopoietic leukemic stem cell (32), at the origin of a leukemic hierarchy organization. Many experimental attempts were made after this initial report to identify such CSCs in various epithelial cancers as well as in breast cancer.

CSCs were functionally defined based on experimental models by limiting dilution transplantation of selected tumor compartment cell using fluorescence-activated cell sorting (FACS), into an orthotopic site in immunocompromised mice. CSCs population were thus characterized by their ability to initiate tumorigenesis, undergoing self-renewal and differentiation after iterative transplantation, whereas the remaining majority of non- CSCs lacked these in vivo proliferation properties and were supposed to more “committed or engage” into a differentiation lineage.

In breast cancer, CSCs or Tumor Initiating Cells was confirmed by Al Hajj et al., in 2003 (33). They identified CD44+CD24-/low sub-population cells that are able to form new tumor in fad pat transplantation studies after serially passages.

It has been shown that CSCs are de- or undifferentiated cells expressing stem cell- like genes and signaling pathways similar to normal stem cell counterparts, including the genes related to drug resistance, EMT markers and dormancy. Intensive interest on CSC research has been based on the issues of cancer recurrence and metastasis in patients after primary treatment, even after a long period of clinical apparent remission. The principal reason of cancer relapse and metastasis is the persistence of CSC resistant to cytotoxic chemotherapy and/or radiation, and indeed therapies targeting specifically CSCs remain a crucial challenge.

1.4.2 Embryonic stem cell-like gene signature in TNBC

Tumor cells lost gradually specific and appropriate differentiation pattern, they are able to progressively dedifferentiate and will acquire autonomous proliferation, migration, and survival. Those dedifferentiated tumor cells were shown to share molecular and epigenetic profile and features of normal stem cells. Indeed, accumulated evidence has shown that some cancer epithelial cells could re-express very immature state – like embryonic genes like Oct4 and Sox2 (34).

Embryonic-like genetic signature was indeed observed in poorly differentiated human tumors of advanced stage (35). Chang’s and Weinberg’s groups observed

independently in 2008 that several aggressive cancers harbor from gene expression profiles of large-scale data set, an embryonic stem cell-like gene signature (Figure 10) (35, 36). They observed gene expression of ESC-like signature in Grade 3 breast cancer, Invasive hepatocellular carcinoma, diffuse large B-cell lymphoma, high grade gliomas and bladder carcinomas (34, 37, 38). Following years, other groups verified these observations in many other tumor types.

Chang’s group analyzed gene expressions of mouse ESC, differentiated cells and adult stem cells from the various organs. The genetic clustering revealed the two distinct clustering of ESC-like and another adult stem cell specific mapping including mammary stem cells. They proposed that the activation of an ES-cell-like transcriptional program via Myc may induce the characteristics of CSCs.

Weinberg group showed ES-like gene expressions in several classifications of breast cancer mostly in basal-like subtype from Grade 3. Additionally, the repression of PRC target genes was observed. Tumors with larger size at the time of diagnosis (more than 2 cm diameter) were more likely to possess the ES signature compared to smaller tumors.

Kaplan-Meier analyses of OS using both overexpression of the ES expression module and underexpression of the PRC2 targets module showed that individuals bearing tumors with this signature had worse survival rates. It was also shown that the ESC-like signature expression in primary human breast cancers is significantly correlated with expression of CSC-enriched CD44+CD24−/low subpopulation phenotype. This suggests that ESC-like signature may be activated in the tumor initiating cells / CSCs. Collectively, ESC-like signature correlates with aggressive breast tumor behavior (Figure 10).

(a)

(b)

(d) (c)

Figure 10: Poorly differentiated breast cancers show an ES-like enrichment pattern. (a) Mammary stem cell was clustered within adult tissue classification whereas Grade 3 breast carcinoma expressed ES like gene module. (b) Enrichment pattern of indicated gene sets (rows) across 1,211 breast cancer samples (columns) included in six profiling studies. Red and green indicate significantly over- or underexpressed gene sets, respectively. Shown are 1,089 tumors for which both ER status and grade were available. Brown bars (bottom) indicate individual tumor annotations for grade, ER

status and tumor size (T Sz), where available. S, tumor smaller than 2 cm across (pathological T1); L, tumor larger than 2 cm (pathological T2 or T3). (c) Kaplan-Meier analyses of disease-specific survival of individuals with breast cancer included in three of the five studies analyzed. (d) Correlation of Pearson value to CD44+CD24−/low tumorigenic breast cancer cell signature with average log2 expression value of ESC-like module in the primary human breast cancers (Adopted from Wong. Ben-Porath 2008.(35) (36))

An active ES-cell-like expression program has been observed upon inactivation of p53 in breast and lung cancer. Interestingly, the p53 tumour suppressor pathway is closely linked to somatic cell reprogramming (39). Reduction activity of p53 in hESCs promote spontaneous sub-chromosomal abnormalities and epigenetic changes similar to those found in embryonic carcinomas and aggressive teratocarcinomas (40). TP53 inactivation and stemness features are closely associated in breast cancer (42) particularly in basal- like subtype and HER-2 enriched subtypes (Figure 11). Given these facts, it is speculated that dysfunction of TP53 in BLBC have critical influence on the induction of de- differentiation machinery into ES-like status.

There are still open questions. Is p53 mutation and p53 dysfunction the primary cause of TNBC, basal-like and then ES-like gene signature? Or, are the genetic instability caused by basal-like or ES-like cancer cells more susceptible to p53 mutation leading to more aggressive behavior? How exactly tumor microenvironment involves this process? Answers to those questions should clarify clearer mechanism of cancer initiation and CSC function in tumor cell population in the future.

(a) (b)

(c)

Figure 11: ES like gene signature in breast cancers (a) Basal-like and HER2 subtypes show ESC signature (b) High scores for the ESC signature correlated highly with the tumor's p53 mutational status. (c) Mutated gene analysis using Tumor samples of luminal A (n = 225), luminal B (n = 126), HER2E (n = 57) and basal-like (n = 93). 80% of basal-like subtype has TP53 mutation. (Modified from Mizuno 2010, The cancer genome altas network 2012) (41, 42)

1.4.3 Stemness molecular profile in breast cancer

Molecular circuits like ES or stem cells in breast cancers further support the background of ES-like signature of advanced TNBCs. There are many evidences that specific pathways and transcription factors (TFs) are shared between stem cells and TNBCs. The critical molecular pathways and transcription programs in breast CSCs are summarized in Table 5.

Table 5: Important molecular pathways and transcription programs in breast CSC

Transcription factors Oct4 Oct4 inhibition decreased CSC and tumor growth in breast and lung cancer Hu 2008 model (43) Ectopic expression of Oct4 into normal primary breast epithelial preparations Beltran generated cell lines forming triple-negative breast carcinomas in nude mice 2011(44) Sox2 Expression in early stage breast cancer Leis 2012 (45) Nanog BC patients with strong expression of Nanog had significantly lower disease- Nagata free survival and OS rates than those with weak expression of Nanog 2014(46) Down-regulation/knockdown of Nanog reduced cell proliferation, Han 2012 (47) expression levels of cyclinD1 and c-Myc in BC cells. This blocked the cell cycle at G0/G1 phases c-Myc c-Myc was sufficient to reactivate the ESC-like program in normal and cancer Wong 2008 cells (36) Identification of Myc-centered regulatory network in ES cells. Kim 2010 (37) Myc rather than core pluripotency module accounts for the shared signatures of embryonic stem and cancer cells Pathways Hedgehog EMT programs promote basal mammary stem cell and tumor-initiating cell Guen 2017 stemness by inducing primary ciliogenesis and Hedgehog signaling (48) Hedgehog signaling components PTCH1, Gli1, and Gli2 are highly Liu 2006 expressed in normal human mammary stem / progenitor cells cultured as (49) mammospheres and that these genes are down-regulated when cells are induced to differentiate. Overexpression of Gli2 in mammosphere-initiating cells results in the production of ductal hyperplasia, and modulation of Bmi-1 expression in mammosphere-initiating cells alters mammary development. Wnt/β- Induction of Mammary gland tumorigenesis Michaelson catenin 2001 (50) IL-6/JAK2 Preferentially active in CD44+CD24- breast cancer cells Tsukamoto /Stat3 1988(51) Notch Notch4 signaling activity was 8-fold higher in breast cancer stem cell- Harrison 2010 enriched cell populations. (52) TGF-β Increase CSC population in SMAD4 dependent manner. TGF-β–responsive Bhola 2013 gene signature score was significantly enriched by chemotherapy in biopsies (53) from TNBC patients. miRNAs miRNA-200 miRNA-200c strongly suppressed the ability of normal Shimono family mammary stem cells to form mammary ducts and tumor formation driven by 2009(54) human breast CSCs in vivo Let-7 let-7 miRNAs were markedly reduced in breast tumor initiating cells and Yu 2007(55) increased with differentiation Viswanathan Lin28 as a negative regulator of miRNA biogenesis and suggest that Lin28 2008 (56) may play a central role in blocking miRNA-mediated differentiation in stem Iliopoulos cells and in certain cancers. 2009(57) An inflammatory response mediated by NF-κB that directly activates Lin28 transcription and rapidly reduces let-7 microRNA levels.

The main core of pluripotent TFs (Oct4, Sox2 and NANOG) and c-Myc have been found to have critical roles in breast cancer. Those TFs are inactive in mammary gland cells except during lactation phase, and can be activated in malignancy. Strong expression of Nanog in breast cancer correlated with dismal prognosis with a significantly lower disease-free survival and OS rates (46). Down-regulation or knockdown of Nanog reduced cell proliferation, expression levels of cyclinD1 and c-Myc in breast cancer cells which are blocked in G0/G1 phases of cell cycle (47). c-Myc was sufficient to reactivate the ESC-like program in normal and cancer cells (36). Oct4 inhibition decreased CSC and tumor growth in breast and lung cancer model (43). Ectopic expression of Oct4 into normal primary breast epithelial cells generated breast carcinomas in nude mice (44).

In ES cells, transcriptional regulation can be subdivided into functionally separable regulatory units (37). Modules for the core pluripotency factors (Core module: Oct4, Nanog, Sox2, Smad1, Stat3, Klf4, Nac1, Zfp281, and Dax1), the Polycomb complex factors (PRC module: Suz12, Eed, Phc1, and Rnf2), and the Myc-related factors (Myc module: Myc, Max, nMyc, Dmap1, E2F1, E2F4, and Zfx). Myc module was functionally separated from core module in regulating aspects of ES cell identity. Myc modules covered more target genes than the ES cell core factors and were suggested to have more global roles in their target gene regulation.

The molecular pathways that support maintenance and metabolism of stem cells have also critical role in CSCs. Hedgehog signaling components PTCH1, Gli1, and Gli2 are highly expressed in normal human mammary stem / progenitor cells cultured as mammospheres. These genes are down-regulated when mammary cells are induced to differentiate. Overexpression of Gli2 in mammosphere-initiating cells results in the production of ductal hyperplasia. Modulation of Bmi-1 expression in mammosphere- initiating cells alters mammary development (49). EMT programs promote basal mammary stem cell and tumor-initiating cell stemness by inducing primary ciliogenesis and Hedgehog signaling (48). Wnt/β-catenin pathway were shown to induce mammary gland tumorigenesis (50). IL-6/JAK2/Stat3 pathway was preferentially active in CD44+CD24- breast CSCs (51). Notch4 signaling activity was 8-fold higher in breast CSCs-enriched cell populations (52). TGF-β–responsive gene signature score was significantly enriched by chemotherapy in TNBC specimens and the increase of breast CSC population was SMAD4 dependent manner (53)

In addition to these genes and signaling pathways, critical role of micro RNA (miRNA) in maintenance of normal stem cells and CSCs are being revealed recently. miR-200c strongly suppressed the ability of normal mammary stem cells to form mammary ducts and tumor formation driven by human breast CSCs in vivo (54). Let-7 miRNAs were markedly reduced in breast tumor initiating cells and increased with differentiation. Lin28, as a negative regulator of let-7 miRNA biogenesis, may play a central role in blocking miRNA-mediated differentiation in stem cells and some cancers. Inflammatory response mediated by NF-κB can directly activates Lin28 transcription and rapidly reduces let-7 miRNA levels and maintain CSCs (55, 57, 58).

The pathways and transcriptional programs that are involved in CSC transformation in TNBCs are summarized in Figure 12.

(a)

(b)

Figure 12: Schematic model of CSC transformation (a) Molecular circuits, transcription factors, miRNAs that can influence CSC phenotype (b) TNBC CSC model. P53, miR-200, let-7 are inactive or down-regulated, whereas molecular pathways including TGFβ /SMAD, WNT/β -catenin, Hedgehog and JAK/STAT pathway are activated. Pluripotent TFs and EMT TFs are overexpressed. Altogether, those molecular changes lead the cells to lost their identity and to promote cell transformation into ES-cell like CSCs with high proliferation, drug resistance, EMT phenotype and invasion.

1.4.4 Breast Cancer Stem Cells

The presence of CSCs in breast tumors is likely one of the main reasons why current conventional therapies are not sufficiently effective to eradicate all of the tumor population which often lead to relapse and metastasis. Al Hajj et al. found in 2003 that CD44+CD24−/low population was significantly enriched in tumor-initiating cells with self- renewal capacity (33).

Most of the breast tumors that possess more than 10% of CD44+CD24−/low cells express basal markers CK5, P-cadherin, CK14 and vimentin. 76.5% of basal-like subtype expressed ≥10% of CD44+CD24−/low cells (59). Another group also observed that 69% of luminal A, 70% of luminal B, 52% of HER2, and 100% of basal-like tumors contained some CD44+CD24−/low cells. The relative frequency of CD44+ and CD24+ cells in breast tumors was correlated with distant metastasis-free survival (60). CD44+CD24−/low population was also shown to be chemoresistance. Hedgehog signaling pathway is highly expressed in this population (61). Alternatively, a small ubiquitin-like modifier (SUMO) pathway has been shown to correlate with this population and its invasion (62). SUMO (particularly SUMO1) is also known to involve ES self-renewal.

Aldehyde dehydrogenases (ALDH1) has been shown to be a marker of normal and malignant human mammary stem cells (63). In 179 cases of breast cancer, the positive expression rate of ALDH1 was 16.7% in luminal A subtype, 21.4% in luminal B subtype, 54.5% in Her2-enriched subtype, 33.3% in basal-like subtype (64). Another study showed that ALDH1+ breast cancers were significantly associated with the phenotype of hormone receptor negative and HER2 positive BC (65). Thus, ALDH1 expression is seen highest in HER2-enriched subtype whereas CD44+CD24−/low prevalence was highest in basal-like subtype.

Recent report showed a molecular link between normal breast stem cells, CD44+CD24−/low breast cancer cells, and embryonic carcinoma cells that commonly down-regulate miR-200c-141, miR-200b-200a-429, and miR-183-96-182. Expression of BMI1, a well-known self-renewal regulator in stem cell, is modulated by miR-200c. miR- 200c inhibited the clonal expansion of breast cancer cells and suppressed the growth of embryonic carcinoma cells. The components of EMT pathways including SNAI2 are highest in the CD44+CD24−/lowlineage− breast cancer cells. miR-200 family miRNAs were strongly suppressed in CD44+CD24−/low breast cancer cells. The miR-200 family targets multiple genes such as ZEB1 that serve as EMT inducers.

Vimentin expression is one of the characteristics of epithelial mesenchymal transition (EMT) that may play a role in migration and metastatic invasion. Basal-like breast tumors were reported to show high expression of multiple EMT markers and were reported to have intrinsic phenotypic plasticity for mesenchymal transition. Thus, there is a link between stem cell–like phenotype, EMT, and basal-like subtype of breast cancer. Claudin low breast cancer type shows EMT markers. However, not all basal-like tumors are uniformly vimentin positive, suggesting a heterogeneity among this tumor subtype.

Despite the growing list of CSC markers, several of these are not uniformly useful in identifying CSCs. For example, although the CD44+CD24−/low profile was used in early studies of breast CSCs, the authors report that not all breast cancer cell populations could be stratified using this set of markers.

1.4.5 Tumor evolution of TNBC

To discuss about CSC as tumor initiating cell, my central question has been how TNBC initiate in the mammary tissue and how it evolved towards an invasive cancer. Most of TNBCs are diagnosed at invasive stage (IDC). In contrast, ductal hyperplasia (ADH) or, DCIS are mostly revealed in mammography screening without any clinical signs. It has been revealed that those tumors have already intratumoral heterogenicity including histological grade, IHC staining and molecular subtypes. Allred et al. described the diversity and existence of multi-subtypes of cells in DCIS tumor (66). Figure 13 shows the specimen from one patient including different status of histological grade, ER and HER2 expression depending on the microlocation of the tumor. Clustering of gene expression showed that proportion of luminal, basal, and HER2-enriched intrinsic subtypes observed in DCIS is similar to the previous studies in invasive breast cancers (i.e 44%, 8%, and 28% respectively from 25 DCIS samples). In 120 cases of DCIS, histological grading showed 45.8% of diversity including 30.0% with grades 1 and 2, 6.6% with grades 2 and 3, and 9.2% with grades 1, 2, and 3. This result indicate that DCIS represent a pre-invasive stage already heterogenous, and that genomic analysis become crucial since the early stage of the disease.

Figure 13: Diversity and existence of multi-subtypes within DCIS tumor (a) IHC photos of one DCIS from a patient including different types of cancer cells (b) Gene clustering of molecular subtype showed presence of multi-subtypes. The red curve lines in the genetic clustering of the molecular subtype show presence of some different subtypes in one specimen. (Adopted from Allured et al., 2008 (66))

Kim et al., showed important insight in chemoresistance mechanism and clonal selection by CNA (Copy number aberration) analysis of twenty TNBC samples (67) (Figure 14). In responder (extinction) cases, two or three major clones were eliminated after chemotherapy. In two resistant cases, two minor clones (7.7% and 18.6% respectively) initialy present at pre-treatment became the majority (71.8% and 100%) after treatment. Another resistant case showed 67.4% to 87% of dominant clone before and after therapy. In some resistant tumor, new clone emerged after middle of chemotherapy and became 37%, with other driving clone selected under chemoptherapy from 5.7% to 41.4%. Classification of molecular subtype by single cell RNA analysis revealed co-existence of multi-subclones and the proportion of molecular changes after therapy.

(a)

(b)

(c)

Figure 14: Copy number Evolution in Clonal Extinction and Resistant Patients (a) Consensus integer copy-number profiles of the clonal subpopulations Consensus line profiles show annotated cancer genes and subpopulation-specific differences indicated with gray bars. (b-c) Classification of single tumor cells into the five breast cancer subtypes (luminal A, luminal B, her2+, basal-like or normal-like) based on PAM50 gene expression and ordered in pentagram graphs where each dot represents a single cell.

Another rare case reported by Yetes et al. was a germline BRCA1 mutation carrier who was diagnosed with a triple-negative cancer in left breast, and over the next 10 years treated for two apparent local tumor relapses and a distant metastasis in the contralateral breast (68). Genomic analysis revealed that the three lesions in left breast were clonally unrelated, completely independent primary cancers with the second of them seeding the contralateral breast metastasis. Again, it emphasizes the importance of close monitoring and accurate verification of the subtype by understanding each case.

Though it would be essential to verify those results with more samples, these results implied several important findings. 1) Rare population pre-existed before chemotherapy

can resist during therapy, expand and able to become dominant (CSC model). 2) Dominant clone before therapy can continue to be dominant if they resist during therapy. 3) New distinct clone that didn’t exist before therapy can emerge during the therapy (Acquired resistance, Clonal evolution model). 4) Co-existence of multiple molecular subtypes in one tumor. 5) Switch of molecular subtype before and after therapy is possible with selective dominant clone (Clonal evolution model). Taken together, both CSC model and clonal evolution model co-exist in breast tumor. It is crucial that treatments take into account this heterogeneity and evolutional dynamics of cancer populations. Clonal heterogeneity and molecular landscape of TNBCs will now be explored by single cell RNA sequencing analysis in future investigation to monitor the resistant population clones. Genetic analysis combining CSCs, EMT and drug resistant markers should provide further information of molecular mechanism of resistance, progression and invasion of those clones. Importantly investigations analyzing in addition the Tumor Immune microenviroment (TIME) landscape will be crucial to explore the progressive stage from immune surveillance to immune evasion.

1.5 Treatment for TBNC

TNBCs are associated with poor prognosis with high risk of relapse, short PFS (Progression-Free Survival) and OS. Neoadjuvant chemotherapy before surgery results in higher rates of pCR, disease-free survival and OS in TNBC than in hormone receptor positive, Her2-negative type (28–30% vs. 6.7%). However, cancer relapse occures in 50% of patients diagnosed with early-stage TNBC, and 37% die in the first 5 years after surgery. Similarly, patients with metastatic triple-negative breast cancer have short PFS after failure of first-line chemotherapy (median PFS, 3 to 4 months), indicating the pressing need for innovative therapy options (69).

According to the ESMO (European Society for Medical Oncology) guideline for primary breast cancer treatment, Table 6 shows the current standards of care and treatment according cancer subtype defined by hormone receptors (ER, PR) and HER2 status. One study, using a panel of ER, HER2, EGFR, and CK5/6 for IHC to identify basal-like BC, showed a sensitivity of 76% and specificity of 100%. However, there is still no consensus regarding clinical histopathological definition of basal-like subtype.

Subtype Recommended therapy Comments

ET alone in the majority of Consider ChT if: high tumor burden (four or more positive LN, T3 Luminal A-like cases or higher), or, grade 3

Luminal B-like ET + ChT for the majority

(HER2-negative) of cases

Luminal B-like ChT + anti-HER2 + ET for If contraindications for the use of ChT, one may consider ET + (HER2-positive) all patients anti-HER2 therapy, although no randomized data exist.

HER2-positive ChT + anti-HER2 (non-luminal)

Triple-negative ChT + RT

Table 6: Systemic treatment recommendations for early breast cancer subtypes (ESMO guideline. 2015 (70)) ET, endocrine therapy; ChT, chemotherapy; RT, Radiotherapy; LN, lymph node

As seen in Table 6, TNBC are typically treated with a combination of surgery, radiation therapy, and chemotherapy.

The most frequently used regimens contain anthracyclines and taxanes, although in selected patients, cyclophosphamide / methotrexate / fluorouracil (CMF) may still be used. Four cycles of doxorubicin and cyclophosphamide (AC) are considered equal to six cycles of CMF. The addition of platinum compound (carboplatin) to neoadjuvant chemotherapy allows for an increase in the pCR rate in triple-negative tumors, particularly in those carrying deleterious BRCA 1/2 or RAD mutations. But the effect of those compounds on long-term outcomes is unknown. Chemotherapy is usually administered for 12–24 weeks (four to eight cycles), depending on the individual recurrence risk and the selected regimen. The use of dose-dense schedules is sometimes used particularly in highly proliferative tumors.

In contrary to the struggling situation of TNBC therapy, HER2-positive BC treatment has made considerable improvement with Trastuzumab. Trastuzumab combined with chemotherapy in patients with HER2 overexpression/ amplification approximately halves the recurrence risk, compared with chemotherapy alone, translating into a 10% absolute improvement in long-term DFS and 9% increase in 10-year OS, emphasizing the effectiveness of the targeted therapy.

PARP inhibitors:

In January 2018, FDA approved Olaparib which is an oral PARP (poly adenosine diphosphate–ribose polymerase) inhibitor that has antitumor activity in patients with metastatic breast cancer with somatic or germline BRCA mutation. Olaparib provided a significant benefit over standard therapy; median PFS was 2.8 months longer and the risk of disease progression or death was 42% lower with Olaparib monotherapy than with standard therapy. Combination regimen with platinum-based agents like carboplatin or cisplatin is being evaluated for safety and efficacy for BRCA mutation carrier. It was accompanied with the approval of Myriad’s BRCAnalysis CDx, a companion diagnostic to Olaparib, to detect BRCA mutations from BC patirnts’ blood samples. ESMO has encouraged for including this genetic testing in the advanced BC and may also be considered for all patients with TNBC.

Recommendation for BRCA mutation carrier:

The estimated prevalence of BRCA1 and BRCA2 mutations is dependent on the population and can vary between 1 in 300 and 1 in 800, respectively. Even among those with no family history of cancer, the lifetime risk of developing breast or ovarian cancer by the age of 80 is up to 83% (±7%) in the presence of BRCA1 and 76% (±13%) in the presence of a BRCA2 mutation. Risk-reducing surgery (with prophylactic bilateral mastectomy and reconstruction) may be offered to women at very high risk. With bilateral mastectomy, the risk for both subsequent breast cancer incidence and mortality is reduced by 90%–95%. Limited data are available about the use of selective ER receptor modulators (tamoxifen, raloxifene) and aromatase inhibitors as primary prevention among BRCA1/2 mutations carriers. Use of tamoxifen may be considered; however, the level of evidence is weak.

Immunotherapy:

Exploring immunotherapy approaches for breast cancer is at intense attention. Despite much anticipation, phase Ib trial of PD-L1 antibody with 32 TNBC patients with PD-L1 positive tumors, reported in 2016, showed the overall response rate was 18.5%. 15.6% patients had high grade toxicity. Considerable possibility should be remaining to improve

the treatment by immunotherapy approaches by understanding the rapidly unraveling TNBC pathology and combinational regimes.

Clinical trials of Phase I/IIa IIb including 28 diverse schedule combining immune therapy (immune check point inhibitors, vaccine strategies) and chemotherapy are currently on investigation. More details of clinical trials of immunotherapy for breast cancers are discussed in Chapter 3.

2. Chapter 2: Immunity and cancer

In this chapter, the physiological state of immunity and the modifications in the immune state during tumor development are discussed. The complexity in cancer environment is represented with immune tolerance, continuously changing stroma and patient’s physiological conditions. In the vaccinology field for infectious diseases, antibodies was the first immune effector that were extensively researched and its titer level has been used as surrogate for quality control of many commercialized vaccines for long time. T cells has been then evaluated for its relationship to the correlate of protection of the vaccines. Meanwhile, in the field of cancer immunotherapy, T cells and NK cells have been the main target to use as principle effectors against the tumor cells.

Great expansion of recent insight regarding the immunity and tumor microenvironment revealed the presence of huge diverse of participants which can influence activation or suppression of the immune status. Even with rapid expanding knowledge of immunology and oncoimmunology, there might be yet unknown population or combinational effect of the several participants. Therefore, it is critical to understand the immune actors and substances that have been known so far and estimate the immune consequence for the successful immunotherapy strategy.

In the following chapters of innate and adaptive immunity, I focused on the discussion on macrophages, DCs, MDSCs, NK cells, T cells and B cells to describe their roles in cancer. Meanwhile, the roles of neutrophils, eosinophils and mast cells are still largely obscure. However, it has been documented that neutrophils and eosinophils can act both tumor promoting- and tumor suppressive function depending on the immune environment. Mast cells principally contribute to tumor progression. In the TME, recent evidences demonstrated cancer-associated fibroblasts and exosomes exert considerable immune-modulating functions as well as preparation of the metastatic niche. Those factos which are not the principle research topic of my experimental works are separately summarized in Annex 1.

Hanahan and Weinberg issued the revision of the hallmarks of cancer that was originally proposed in 2000 and was re-proposed in 2011 (Figure 15) (71, 72). In 2011, the critical enrolment of cancer immunity became accepted as one of the cores of the cancer pathology, which represents massive progress of recent onco-immunology research.

Proposition in 2000

Proposition in 2011

Figure 15: Hallmarks of cancer: The revised perspective Adopted from Hanahan and Weinberg, 2000. 2011. (71, 72)

2.1 Tumor microenvironment (TME)

2.1.1 Overview

The stromal environment around the tumor gives influence to the fate of cancer cells. When transformation of somatic cells occurs and the cells start to develop dysplasia and cancer, dynamic modification in the surrounding microenvironment, termed tumor microenvironment (TME), happens. TME gradually lose its homeostasis. Immune suppression, angiogenesis, inflammation and metabolic modification of the TME by altered cancer cells and cancer cells by altered TME represent a consequence of a bidirectional cross talk. TME contains a diversity of cells with lymphocytes, various type of myeloid cells, fibroblast, mesenchymal cells, endothelial cells, and molecules with cytokines, chemokines, RNAs, exosomes, reactive oxygen species, extracellular matrix , various soluble agents and metabolite. The dynamic change in the stromal environment in cancers has been first mentioned to the primary tumor lesion but recent evidence indicates biological communication between the local TME and the distant metastatic sites (Figure 16), therefore TME is also a part of the systemic pathology of the malignancy. The immune cells in TME are summarized in Table 7.

Figure 16: The microenvironment supports metastatic dissemination and colonization at secondary sites Adopted from Quail and Joyce 2013. (73)

Table 7: Immune cell population in TME (Quoted from Quail and Joyce. 2013 (73)) Cell type Markers (human) Markers (mouse) Function CD11b+ CD11b+GR1−CD68+ CD68+ CSF1R+ Classically activated M1 macrophages are proinflammatory and anti-tumorigenic and secrete TH1 cytokines. Alternatively, activated M2 macrophages are anti-inflammatory and pro-tumorigenic and secrete TH2 TAM CSF1R+ F4/80+ cytokines. TAMs frequently exhibit an M2 phenotype; their presence in tumors supports angiogenesis and CD163+ invasion. EMR1+

CD11c+ CD83+ CD11c+ CD83+ DCs are monocytic antigen-presenting cells that are derived from the bone marrow. DCs presenting tumor- DC specific antigens are being developed as vaccines to induce both innate and adaptive immune responses to CD123+ CD123+ regress tumors and prevent relapse.

+ + + CD11b CD66b CD11b Neutrophils are the most abundant circulating leukocyte in humans and are phenotypically plastic. Similar to Neutrophil CD63+ GR1+ TAMs, neutrophils have been shown to have opposing functions in regulating cancer progression and 7/4+ metastasis, indicating that they have context-dependent roles within the TME. CD11b− CD11b− CD49d+ CD49d+ CD117+ Mast cells are best known for their role during allergies and autoimmunity. Mast cells are recruited to tumors, Mast cell CD117+ CD203c+ where they release factors that enhance proliferation of endothelial cells to promote tumor angiogenesis. CD203c+ CD11b+ CD11b+ CD33+ GR1+ MDSCs are immunosuppressive precursors of dendritic cells, macrophages and granulocytes. In cancer, − − + MDSC HLA-DR Ly6G Ly6C (monocytic) their main function is to disrupt tumor immunosurveillance by interfering with T cell activation, cytotoxic + CD14 (monocytic) − activity, antigen presentation and cell polarization. Ly6G+Ly6C (granulocytic) CD14− CD15+(granulocytic)

Lymphoid lineage

NK cells are CTLs that can kill stressed cells in the absence of antigen presentation. NK cells detect and kill NK cell CD56+CD16+ CD335+NK1.1+ tumor cells through 'missing-self' activation (loss of healthy cell markers) or 'stress-induced' activation (gain of stressed cell markers).

+ CD4 TH cells can be divided into TH1 and TH2 lineages. TH1 cells secrete proinflammatory cytokines and + + + + TH cell CD3 CD4 CD3 CD4 can be anti-tumorigenic. TH2 cells secrete anti-inflammatory cytokines and can be pro-tumorigenic. The ratio of TH1 to TH2 cells in cancer correlates with tumor stage and grade.

CD4+ CD25+ CD4+CD25+ Treg cells have primarily pro-tumorigenic roles by suppressing immunosurveillance; however, their presence + + + FOXP3 FOXP3 CTLA-4 in tumors is positively correlated with OS in certain cancer types. These divergent roles may be attributed to Treg cell CTLA-4+ CD103+ context-dependent functions or distinct subpopulations that are challenging to identify at present using conventional markers. CD45RA+

+ + + + + CD8 cytotoxic T (TC) cells are effector cells of the adaptive immune system. They specifically recognize and TC cell CD3 CD8 CD3 CD8 destroy cancer cells through perforin- and granzyme-mediated apoptosis.

B lymphocytes are important mediators of humoral immunity. In cancer, they can promote disease progression by secreting pro-tumorigenic cytokines and altering TH1-to-TH2 ratios. Their importance in B cell CD19+CD20+ B220+CD19+CD22+ supporting tumor growth is evident in B cell–deficient mice, which exhibit resistance to engraftment of certain syngeneic tumors.

2.1.2 Immune context in breast cancer

Elements that impact antitumor immunologic activity include the composition of immune cell type as well as their location, density and function. The density of the immune cells can be impacted by the presence of tertiary lymphoid structures (TLS) (74) and chemokines.

The composition of infiltrating CD45+ cells in breast cancers consists of a majority of T cells (60 – 90%) together with B cells (<20%), monocytes (<10%), NK cells (<5%), or NKT cells (75). Traditionally, CD8+ cytotoxic lymphocytes (CTLs) have been considered as a key component of effective antitumor immunity. Breast tumors with higher levels of infiltrating CD8+ T cells have been associated with a better patient survival (76, 77) like most of the other tumor types. However, studies have also shown that CD8+ T cells frequently fail to fully function in vivo, if there is a lack of adequate CD4+ T cell help (78).

The current consensus is that IFNγ–producing CD4+ Th1 and CD8+ T cells, along with mature DCs, NK cells, M1 macrophages, and type 1 NKT cells can generate an effective, although frequently attenuated, antitumor responses, while CD4+ Th2 cells and type 2 NKT cells in cooperation with Tregs, myeloid-derived suppressor cells, immature DCs, or M2 macrophages suppress antitumor immunity and can also promote tumor progression (75).

However, this generalization is not an absolute point at the evaluation of prognosis in breast cancer patients. The association of cell type can be variable: Th1 cells to be good or ‘no correlation’, Th2 cells to be good or ‘no correlation’, and Tregs to be ‘no correlation’ or poor (74). The tumor progress and immune state are likely the consequences of the various immune contexture even though there are still apparent tendencies represented by specific cell types. Recently, CXCL13-producing CD4+ Tfh cells were shown to distinguish extensive immune infiltrates, principally located in tertiary lymphoid structure germinal center. Tfh along with the CXCL13 were associated with better prognosis of breast cancer patients (75).

Tregs are suppressive actors so that poor prognosis looks logical. However, in some cases, even presence of Tregs indicates the presence of infiltrating immune cells. The tumors with abundant of TILs are called “Hot” tumor and those with poor or absent of TILs, “Cold” tumor. Hot tumors are considered to have better prognosis than cold tumor but again, it depends on the total contexture of various elements. Combinational effects

of the existing factors in the TME together shape the consequence of the tumor progression.

Briefly, successful immunotherapy should convert cold tumor to hot tumor by increasing functional CTLs at the tumor site accompanied with functional quality, by recruiting active antitumoral immune cells and by reprograming TME into an immune active status without suppressive polarization.

2.2 Inflammation and cancer

As a reaction to infection or other harmful stimuli, inflammation is often observed. Inflammation is composed of complex biological responses and its classical signs are heat, pain, redness, swelling and loss of function. Acute inflammation is initiated by the resident immune cells already present in the involved tissue, mainly resident macrophages, dendritic cells, histiocytes, Kupffer cells and mast cells.

These cells sense the abnormality through Pattern recognition receptors (PRRs), release cytokines, histamine and serotonin, eicosanoids such as prostaglandin and leukotriene to mediate a rapid vasodilatation and a permeabilization of the blood vessels.

Matzinger in 1994 postulated “Danger Theory” for the release of alarmins or danger signals (79), later termed as “Damage-associated molecular patterns (DAMPs), which are the molecules that are secreted, released or surface exposed by dying or stressed cells.

Carcinogenesis is the result of the interplay of multiple processes, which include genomic instability, proliferative abnormality, reprogramming of the stromal environment, and aberrant differentiation. Acute inflammation contributed to the regression of cancer but chronic inflammatory diseases are frequently associated with increased risk of cancers (80). The examples are inflammatory bowel disease in association with colorectal cancer, gastric Helicobacter pylori in association with adenocarcinoma and mucosa-associated lymphoid tissue lymphoma and hepatitis B and C viruses in association with hepatocellular carcinoma. The risk of esophageal cancer, pancreatic cancer, and bladder cancer may be increased by inflammatory diseases, such as esophagitis. Chronic inflammation damages the tissue and the tissue gradually increases its proliferative activity, becoming tissue atrophy, metaplasia then dysplasia. Dysplastic changes are often found adjacent foci of cancer.

2.3 Innate immunity

In a situation of infection, at the entrance of microbes to the body through mucosa, germline–encoded PRRs are responsible for sensing the presence of the microbes by recognizing Pathogen associated molecular patterns (PAMPs) like Toll like receptor (TLR) ligands (Table 8) or, endogenous molecules released from damaged cells (DAMPs). It is well known that PRRs are expressed in macrophages and DCs but other cells of immune cells as well as non-immune cells express PRRs. Majority of the cases, ligand- bindings to PRRs leads to production of cytokines or, caspase-1. Stimulation to those receptors lead to accumulate DCs and macrophages. DCs and macrophages digest the microbe and present the antigen on their surface, then enter local blood vessels to immigrate to lymph nodes. It reacts immediately by virtue of the germline-encoded receptors, the innate immune cells and the cascades of immune activating molecules.

Adaptive immune response is a reaction that is acquired specifically responding to the experienced pathogen or other antigens, which enable the host to eliminate them more rapidly and effectively after the second exposure to the same antigen. Innate and adaptive immunity had been considered as lower and higher levels of immune systems respectively. The adaptive immunity had attracted more academic interests, until Hoffman reported the existence of spätzle/Toll/cactus/dorsal was essential if the fly was to mount a potent antifungal response (81). Innate immunity is currently recognized as sophisticated control system as well and is indispensable for establishing effective adaptive immunity. Moreover, some of innate immune cells were revealed to have some extent of adaptation to the experienced antigens by the epigenetic memory.

Another way of recognizing the structure is through the situation termed “missing self-recognition” mainly by NK cells to detect infected or stressed cells (82). Missing- self theory was first proposed by Kärre in 1986, which suggested NK cells are effector cells in a defense system geared to detect the deleted or reduced expression of self-MHC. This theory was verified partially correct based on the discovery of suppressing receptor on NK cell bind to MHC I. However, it is now known that the cancer cells which express MHC I can be also the target of NK cells with the combinational recognition of various suppressing and activating receptors. The details of NK cell recognition will be discussed later in this chapter.

Table 8: Toll like receptors and their ligands

TLR Localization PAMPs Origin of the ligand (PAMPs) DAMPs TLR1 Plasma membrane Triacyl lipoprotein Bacteria β-defensin-3 TLR2 Plasma membrane Lipoprotein Bacteria, viruses, parasites, self Monosodium urate, Pancreatic adeno-carcinoma up-regulated factor (PAUF), Serum amyloid A, neutrophil elastase, HSP60, HSP70, gp96, surfactants A/D, eosinophil-derived neurotoxin, biglycan, hyaluronic acid, HMBG1, MMP-2 TLR3 Endolysosome dsRNA Virus Tumor derived dsRNA and siRNAs TLR4 Plasma membrane LPS Bacteria, viruses, self HMGB1, gp96, HSP22, HSP60, HSP70, HSP72, HSP90, hyaluronan, heparin sulfate, fibrinogen, monosodium urate, peroxiredoxin, biglycan, neutrophil elastase, serum amyloid A, oxidized LDL, fibronectin EDA, fibrinogen, tenascin-C, lactoferrin, β-defensin-2, saturated fatty acids, surfactant protein-A, HMGN1 TLR5 Plasma membrane Flagellin Bacteria TLR6 Plasma membrane Diacyl lipoprotein Bacteria, viruses TLR7 Endolysosome ssRNA Bacteria, viruses, self Tumor derived ssRNA and siRNAs, anti- (human TLR8) phospholipid antibodies, miRNAs TLR9 Endolysosome CpG-DNA Bacteria, viruses, protozoa, self Tumor mtDNA, HMGB1, IgG-chromatin TLR10 Endolysosome Unknown Unknown TLR11 Plasma membrane Profilin-like molecule Protozoa

2.3.1 Macrophages

Significant data have shown a causal role for macrophages in cancer initiation or promotion because of their central status as mediators of inflammation. The macrophages at the tumor site often show distinct phenotype and are termed Tumor Associated Macrophages (TAM). Like Th1 and Th2 polarization, two distinct states of polarized activation for macrophages have been recognized: the classically activated (M1) macrophage phenotype and the alternatively activated (M2) macrophage phenotype. M1 expresses Th1 cell-attracting chemokines and M2 expresses Th2 cell-attracting chemokines (83). Yet, there are intermediate state of M1 and M2, emphasizing the heterogeneity of macrophage functional states (84).

Tumor promoting mechanism by macrophage includes angiogenesis (85), inhibition of the activation of NK cells and T cells. TAMs also recruit natural regulatory T (nTreg) cells to the tumor microenvironment, as well as by inducing the CD4+ regulatory fraction (iTreg) cells and sustaining their survival. Altogether, they have an important role in recruitment and activation of Treg cells and the suppression of effector cells in the tumor microenvironment. TAM-secreted cathepsins B and S are critical for promoting pancreatic tumor growth, angiogenesis, and invasion (86). STAT3 and STAT6 were shown to synergistically promote cathepsin secretion by macrophages, thereby enhancing tumor invasion and metastasis (87). M2 promote EMT in pancreatic cancer cells, partially through the TLR4/IL-10 signaling pathway (88). TAMs also promote CSC- like properties via TGF-β1-induced EMT (89).

2.3.2 Dendritic cells

Dendritic cells (DCs) are called professional antigen-presenting cells (APC) which links innate immunity and adaptive immunity. Once DCs recognize pathogens or damages, they change into mature DCs, which are characterized by high level expression of MHC II and increased expressions of costimulatory molecules such as CD40, CD80 and CD86, changes in morphology (bigger dendrites) and reprogram chemokine receptor expressions. Those changes make mature DCs move to the T cell area of local draining lymph nodes and meet with Ag-specific-T cells and induce their activation and differentiation into effector cells (90)

Antigen presentation through MHC II starts with endocytosis of exogenous proteins. The protein that was internalized into endosome by endocytosis is digested by protease

into peptides. Antigen presentation through MHC I can depend on two possible mechanism. One is the digested proteins produced in the cytoplasm. For example, DC which was infected with virus presents part of virus particles in MHC I present to CD8+ T cells. Second mechanism called cross presentation is that some DC phenotypes can internalize exogenous antigens and present to MHC I. In both cases, effective activation of naïve T cells requires costimulatory molecule expressions like CD80 and CD86 in addition to the antigen presentation by MHC molecule, otherwise it results in anergy.

Tumor infiltrating DCs (TIDCs) have been found in the TME in many cancer types. Its association with prognosis may not be evaluated only by the amount of dendritic cells but depends more on the phenotype and polarization capability of the dendritic cells (91). TIDCs often express immune suppressive markers like PD-1 (92) and TIM-3 (93). Furthermore, TIDCs participate creating suppressive immune TME by releasing IDO (93). IL-10, TGF-β, Arginase released from tumor cells may change DCs into suppressive phenotype (94).

2.3.3 Myeloid-derived suppressor cells (MDSC)

One of the most prevalent mechanisms of immune evasion in cancer patients is through the suppressive activity of MDSCs. In the early 1900s, an increase in extramedullary hematopoiesis and neutrophilia, was originally described as one of tumor progression (95), which was later shown to result in immune evasion and tumor vascularization. Classically, this occurs within the host macroenvironment with an increased level of hematopoietic colony-stimulating activity in serum (96) and abnormal myeloid cell differentiation (Figure 17). That results to a bidirectional molecular crosstalk between tumor cells and myeloid progenitor cells. These abnormal myeloid cells were described as veto cells, null cells or as natural suppressor cells (97).

Figure 17: Changes that occur in myeloid cells in cancer Aberrant differentiation of myeloid lineage cells. The dotted lines show the normal pathways of myeloid cell differentiation from immature myeloid precursor cells to DCs, macrophages and granulocytes. The solid bold lines indicate the aberrant pathways of myeloid cell differentiation that occur in cancer. The dotted thick line depicts a pathway of cell differentiation that has been suggested but has not yet been confirmed. (Adopted from Gabrilovich et al. (98))

These cells lacked membrane markers for mature T cells, B cells and NK cells, as well as macrophages, resulting in the nomenclature of null cells. In recent years, the concept of MDSCs was introduced to reflect the abnormal nature of myelopoiesis in cancer. MDSCs are functionally defined as immunosuppressive, immature myeloid cells (99) that maintain normal tissue homeostasis in response to various systemic insults, including infection and traumatic stress.

2.3.3.1 Murine MDSCs

Murine MDSCs have recently been subdivided by the expression of lymphocyte antigens Ly-6C and Ly-6G. CD11b+Ly-6GlowLy-6Chi cell preferentially express iNOS, increase T cell suppressive activity and are termed monocytic-MDSCs (Mo- MDSCs)(100). This contrasts with CD11b+Ly-6G+Ly-6Clow cells that express high levels of arginase 1 (ARG1) and are termed as granulocytic-MDSCs (G-MDSCs). These cells have a polymorphonuclear (PMN) morphology and, therefore, the term PMN-MDSC is used interchangeably with the term G-MDSC.

G-MDSC subset is the predominant MDSC population in tumor-bearing mice (100). This subset distinction can also be clarified by CD11b and GR1 expression (101). Two cellular fractions are generally recognized, a GR1bright subset that is mainly composed of G-MDSCs, and a GR1int subset that encompasses Mo-MDSCs.

TAMs are a cellular population that can be histologically confused with MDSCs, but that are defined as mature, differentiated macrophages (102). Distinguishing MDSCs from macrophages and granulocytes can be technically challenging. Murine Mo-MDSCs express low levels of F4/80 and higher levels of GR1 than TAMs (103). Mo-MDSCs are a mixture of myeloid progenitor cells in varying stages of differentiation (104), which can differentiate into macrophages, DCs or granulocytes.

2.3.3.2 Human MDSCs

Human MDSCs were initially defined as HLA-DR−CD33+ (105) or CD14−CD11b+ cells (72), with both phenotypes identifying cell populations with T cell suppressive activity. Recently, more rigorous markers have defined MDSC subsets, including Mo- MDSC expression of CD14+/dull (106) and G-MDSC expression of CD15+, as primarily observed in patients with advanced renal cell carcinoma (106). Similar to murine MDSCs, human MDSCs are hematopoietic progenitors that can differentiate into not only granulocytes and monocytes, but also into endothelial cells and osteoclasts.

2.3.3.3 MDSC infiltration in tumor

MDSCs within tumors have been shown, in 4T1 breast tumor model, to be derived from splenic and hepatic extramedullary hematopoiesis before tumor infiltration (107). Cells with a natural suppressor phenotype and function were also identified at sites of intense hematopoiesis and found to be increased in number by tumor secretion or by

exogenous administration of hematopoietic growth factors, including granulocyte– macrophage colony-stimulating factor (GM-CSF) and granulocyte-CSF (G-CSF) (108). The regulation of MDSC is summarized in Table 9.

Process or Mechanism Examples of mediators

MDSC proliferation in the bone marrow C-CSF, GM-CSF and SCF

Impaired differentiation of MDSC TNF and VEGFA

Mobilization of MDSCs CXCL8, G-CSF and GM-CSF

Chemotaxis of MDSCs to an inflammatory or tumor site CCL and CXCL12 Differentiation of MDSCs into osteoclasts, monocytes, ATRA and vitamin D granulocytes, endothelial cells or dendritic cells Table 9: Regulation of MDSCs (Adopted from Talmadge and Gabrilovich. 2013 (99))

2.3.3.4 Mechanisms of immunomodulatory functions by MDSCs

The mechanisms of immunomodulatory functions can be generally grouped into four classes (98). The first type of mechanism is the depletion of nutrients required by lymphocytes specifically, L-arginine through ARG1-dependent consumption (109) and L-cysteine deprivation (110). The depletion of these amino acids causes down-regulation TCR complex and proliferative arrest of activated T cells.

The second type of mechanism is the generation of oxidative stress, which is caused by the production of ROS and reactive nitrogen species by MDSCs. These reactive species drive several molecular blocks in T cells leading to the desensitization of the TCR (111).

The third type of mechanism interferes with lymphocyte trafficking and viability. Expression of ADAM17 (disintegrin and metalloproteinase domain-containing protein 17) at the plasma membrane of MDSCs decreases CD62L expression on the surface of naive CD4+ and CD8+ T cells, thereby limiting T cell recirculation to lymph nodes (112). MDSCs express galectin 9, which binds to T cell immunoglobulin and mucin domain- containing protein 3 (TIM3) on lymphocytes and induces T cell apoptosis (113). MDSCs also decrease the number and inhibit the function of NK cells (114).

The fourth type of mechanism is the activation and expansion of Treg cell populations. MDSCs promote the clonal expansion of antigen-specific natural Treg cells and also induce the conversion of naive CD4+ T cells into induced Treg cells.

2.3.3.5 MDSC triggers tumor metastasis

Increase of IDO-expressing MDSC in breast cancer tissue correlated with increased lymph node metastasis in patients (115). In patients with melanoma, the development of metastases and poor survival was associated with increase of both PMN-MDSC and M- MDSC (116).

Serum amyloid A (SAA) 3 directly attracted MDSC to pre-metastatic lungs, stimulated NF-κB signaling and facilitated metastasis. MSDC also promote angiogenesis (116). In 4T1 breast cancer model, accumulation of PMN-MDSC was correlated with increased bone metastasis. Co-injection of MDSC and 4T1 cells led to increased lung metastasis. MDSCs in 4T1 tumors up-regulated the expression of several MMPs, which was critical in mediating invasiveness (117).

PMN-MDSC produced HGF and TGF-β and induced EMT of primary melanoma cells. The depletion of PMN-MDSC led to decreased EMT and fewer metastatic lesions in mice (118). In ovarian cancer patients, accumulation of MDSC correlated with poor survival in metastatic and non-metastatic disease. MDSC directly interacted with ovarian tumor cells and induced their stemness. This effect was mediated by up-regulation of microRNA-101 in ovarian cancer cells, which in turn targeted CtBP2, a co-repressor of stem cell genes (119).

2.3.3.6 Re-education potential of MDSCs

It has been shown in animal models that monocytic MDSCs can be reprogrammed to adopt an anti-tumorigenic phenotype in response to bacteria-mimmicking-stimulation of the immune system (120). This transition is accompanied by an increase in proinflammatory T helper type 1 (Th1) cytokines, a reduction in T cell–suppressive factors (for example, arginase-1 or nitric oxide) and differentiation of MDSCs into M1- like macrophages.

2.3.4 NK cells

NK cells are differentiated from common lymphoid progenitor in bone marrow and circulating blood. Cytotoxicity mechanism is similar as that of cytotoxic T cells; Lytic granule is released to the surface of the target cells. Effector proteins pass through cell membrane and lead to programmed cell death.

Cytotoxicity activity is controlled by the balance of two signals of activating and suppressing receptors. NK cells engage in surveillance for transformed or virus-infected cells that have down-regulated expression of MHC class I. It was suggested that NK cell activation cannot be achieved by single activation receptor but realized by combination of several activation receptors and only when threshold of signaling that exceeds the counterbalancing influence of the inhibitory receptors (121) (Figure 18).

Figure 18: NK cell surface receptors and possible outcome at encounters between NK cells and potential target cells. Cell surface receptors expressed by NK cells (Right) indicating the activating, inhibitory and cytokine receptors. Possible outcome at encounters between NK cells and potential target cells (Left). Modified from Lanier. 2005 and Choissone. 2018 (122, 123)

Other important mechanism that NK cell’s surveillance of abnormality is to sense the stressed cells. One example is the activation of NK cells via activating receptor

NKG2D. NKG2D interacts with self-molecules that are selectively up-regulated on stressed cells, such as tumor cells (124). The ligands for NKG2D are stress-inducible proteins, namely: MICA (MHC class I polypeptide-related sequence A), MICB and members of the ULBP (also known as RAET1) family in humans (125, 126), which may be up-regulated at DNA damage and immune response.

Evidences for NK cell role in tumor surveillance have been addressed in correlative studies. In an 11-year follow-up study, it was found that a low NK-like cytotoxicity of peripheral blood lymphocytes correlates with an increased risk for cancer (127). However, in established tumors, there are often only a few infiltrating NK cells. Low NK cell numbers in tumors are likely due to their inefficient homing into malignant tissues.

Rejection of NKG2DL-expressing tumor cells was due to the activity of NK cells and CD8+ T cells and depending on functional NKG2D (128). Constitutive exposure of NKG2DL leads systemic down-regulation of NKG2D and deficiencies in NKG2D- mediated NK cell function. A broad spectrum of tumors also often up-regulates expression of the non-classical MHC class I molecule HLA-G which dampens NK cell responses (129). Many sera of patients with epithelial and hematopoietic malignancies contain elevated levels of soluble MICA and MICB. Soluble MICA has been shown to systemically down-regulate NKG2D. Down-regulation of NKG2D is also mediated by TGF-β and 2,3-dioxygenase (IDO) (130).

2.4 Adaptive immunity

When the tumor cells appear, how it’s seen by the immune cells?

Self/Non-self theory is one of the principles for the cases of infection. It is different when we think about the immunity against the tumor, it is developed from self except the tumor caused by the infection of cancer inducing viruses. One proposed self/altered self theory for cancer.

Cancer rejection epitopes may be derived from two classes of antigens. A first class of potential cancer rejection antigens is formed by nonmutated proteins to which T cell tolerance is incomplete. A second class of potential cancer rejection antigens is formed by peptides that are entirely absent from the normal human genome, so-called neoantigens. For the large group of human tumors without a viral etiology, such neo-epitopes are solely created by tumor-specific DNA alterations that result in the formation of novel protein

sequences. For virus-associated tumors, such as cervical cancer and a subset of head and neck cancers, epitopes derived from viral open reading frames also contribute to the pool of neoantigens.

Investigation revealed that the vast majority of mutations within expressed genes do not lead to the formation of neoantigens that are recognized by autologous T cells (131, 132) Although such T-cell reactivity only involves a small fraction of the total mutanome (around 0.5% of mutated peptides) in individual subjects, CD4+ T-cell reactivity against the autologous mutanome was nevertheless observed (133)

How could T cells be triggered to reject tumors expressing weak self antigens?

The answer is based on how the repertoire of host T cells is selected during development. Any immature T cell with high-affinity TCRs for a self-antigen is deleted during its early development in the thymus; this process leads to destruction of more than 90% of immature T cells. Elimination of these high-affinity T cells is necessary to avoid the development of autoimmunity. However, despite this stringent selection, the immune repertoire is replete with mature, naïve, self-reactive T cells that have TCRs with relatively low affinity for self-antigens. These weak self antigens are incapable of inducing immune responses. However, if T cells against these apparently ineffective self antigens can somehow be activated through vaccination or other means, then these T cells can reject tumors presenting these antigens (134, 135).

2.4.1 T cells

At T cell development, T cell progenitor population moves into thymus from bone marrow and the differentiation to T cells start with interaction with thymic stroma cells. The antigen-recognizing-molecule of T cells are called TCR consists of variable (V) region and constant (C) region. TCR structure is similar to immunoglobulin but TCR recognizes antigen-MHC complex.

MHC is highly polymorphic and the major difference present in the peptide-binding groove. The peptides which binds to the MHC I is usually constituted with 8-10 amino acids which binds to the both ends of the peptide-binding groove though longer peptides can bind to MHC I when it can bind to the binding sites; two anchors on the both sides of MHC molecule. In contrary to MHC I, there is no theoretical restriction of peptide length for MHC II for antigen binding (however, majority of the cases are 13-17 amino acids).

When CD4+CD8+ T cells recognize antigen-MHC complex presented on cortical thymic epithelial cells (cTECs), they receive critical survival signals and differentiate into CD4- and CD8-single-positive T cells according to their affinity to MHC II and I, respectively (136). This process of T cell selection and acquisition of MHC restriction is called ‘Positive selection’ followed by ‘Negative selection’, which eliminates majority of the autoreactive T cells. Medullary thymic epithelial cells (mTECs) express a great many tissue-restricted antigens (TRAs) (137), a process which allow the deletion of autoreactive T cells specific for the antigens that would otherwise only be encountered in the periphery. Although most autoreactive CD4+ T cells are negatively eliminated in the medulla, a portion of them differentiate into Treg cells, which are characterized by Foxp3 expression and specialized in the control of peripheral immune tolerance (138).

High-affinity TCRs specific for tumor-associated (self) antigens (TAAs) are rarely present in the human T-cell repertoire, because tolerance mechanisms select against T cells with TCRs that have a high affinity for self-antigens. Although central-tolerance mechanisms are efficient, they cannot eliminate all self-reactive lymphocytes, in part because not all self-antigens are expressed at thymus. Therefore, peripheral-tolerance mechanisms exist, and these are crucial to control tolerance of lymphocytes that first encounter their cognate self-antigens outside of the thymus—such as food antigens, developmental antigens, and antigens displayed during chronic infection. Both anergy and deletion of self-reactive T cells can occur in the periphery (139) (Figure 19).

Figure 19: Peripheral tolerance T-cell anergy is induced by inhibiting mTOR pathways or can be induced by tolerogenic DCs. Lymph node stromal cells express tissue-specific antigens and can mediate the deletion of self- reactive naive T cells.(140)

Although some peripheral T cells against tumor-associated self-peptides escape negative selection, it has been shown that their TCRs exhibit lower average affinities than typical foreign antigens (mean Kd values of about 100 μM, compared to mean Kd values of about 10 μM for viral antigens) (140). TCRs with even modestly higher affinity can yield significant increases in potency (141). CD8+ T cells are more sensitive at weaker affinities (Figure 20).

Figure 20: Relationship between TCR affinity and T cell activity for CD8 and CD4 T cells. TCR affinity for pep/MHC impacts T cell activity for CD8 (top) and CD4 (bottom) T cells. (Adopted from Stone et al. 2015 (139))

2.4.1.1 Cytotoxic T lymphocyte (CTL)

Lymphocyte-mediated killing could be confined to two pathways, the perforin/granzyme-mediated and the Fas-mediated pathways. The perforin-dependent pathway is dominant in CD8+ cytotoxic T lymphocyte (CTL) and NK cells (142). In NK cells, granules are pre-formed, although its activity can be increased by cytokines like IL- 2 and IFNγ. NK cells are thus constitutively armed and can kill within minutes of the first stimulation of activating receptors, but they generally do not proliferate significantly in response to these stimulations.

In contrast, naïve CD8+ CTL precursors have no cytotoxic capability and must undergo an activation process requiring 1–3 days for maximal activity. This activation process requires TCR-stimulated induction of cytokine receptors, which then induce the expression of granule components, including perforin and granzymes. The same signals that activate CD8+ cells also stimulate their proliferation. antigen-specific CD8+ can expand several orders of magnitude (143).

When T cells receive specific signals to activate via the TCR, transcriptional mechanisms are activated that lead to the production of cytotoxic granules and their

constituent proteins (144). Upon target-cell-identification and conjugation, specific signals are generated in the effector lymphocytes that cause the granules to migrate to the site of contact. At the cell surface, the granule fuses with the plasma cell membrane, and its contents are secreted into the tight intracellular junction formed between the two cells.

Figure 21: Interaction of a cytotoxic lymphocyte with a target cell (Adopted from Voskoboinik et al. 2015 (145)) The cytotoxic lymphocyte recognizes its target cell and forms an immunological synapse (a). The microtubule-organizing center of the cytotoxic lymphocyte polarizes and secretory granules traffic toward the presynaptic membrane (b). The secretory granules then fuse with the presynaptic membrane and release perforin and granzymes into the synaptic cleft (c and d). At the postsynaptic membrane, perforin forms large transmembrane pores that enable the diffusion of granzymes into the target cell cytosol. Granzymes then initiate apoptosis of the target cell, and the cytotoxic lymphocyte detaches from the dying cell (indicated by the arrow; e) and can interact with another target cell to carry out serial killing.

Once conjugated to a target cell, the cytotoxic secretory granules traffic to the immunological synapse and release a cargo of deadly proteins including perforin, granzyme and granulysin1 into the synaptic cleft (Figure 21). Granzyme B is the most

powerful pro-apoptotic granzyme, as its ability to cleave target cell proteins at sites after selected aspartate residues mimics the caspases. This 'imposed' death is rapid and effective; the exposed target cells die within 5–8 minutes by apoptosis (146).

2.4.1.2 Regulatory T cells

2.4.1.2.1 Differentiation of Tregs

In 1971, Gershon and Kondo identified so-called “suppressor” cells when they transferred antigen-specific tolerance to naive animals by transferring antigen- experienced T cells (147). Sakaguchi et al. proposed in 1995, a terminology of “regulatory” T cells (Tregs) by identifying a population of CD4+ T cells highly expressing CD25 and preventing autoimmunity in a murine model (148).

Negative selection of self-reactive T cells in the thymus leads to elimination of self- reactive T cell clones and is likely responsible for neutralization of most high-affinity T cells recognizing self. In the periphery, chronic engagement of TCRs by self antigens induces anergy. The differentiation of Treg cells in the thymus is promoted by increased affinity interactions with self-peptide-MHC complexes, whereas differentiation of peripheral iTreg cells likely occur more environmental factors like cytokines and antigens. High-affinity TCR signaling together with suboptimal co-stimulation (increased CTLA- 4 and decreased CD28 signaling) favors Foxp3 induction and iTreg cell generation. Induction of Foxp3 expression in peripheral naive CD4+ T cells can be also facilitated by high amounts of TGF-β (149). Foxp3 represses production of proinflammatory cytokines (IL-2, TNF-α, IFNγ, IL-17, and IL-4) (150).

Figure 22: CD4+ thymocyte fate and regulatory T cell differentiation by TCR signal strength (Adopted from Josefowicx et al., 2012 (150) ) Immature CD4+ thymocytes receive TCR signals of varied strength via interactions with peptide- MHC on APCs. The strength of TCR signals, or functional avidity, peptide-MHC abundance and their duration determine its fate.

T-bet, a key transcription factor in Th1 effector cell differentiation, can be induced in Foxp3+Treg cells enables them to migrate, and accumulate at the sites of Th1 responses. This cell trafficking was through the expression of CXCR3 and led to specifically suppress Th1 reaction (151). IRF4, transcription factor to engage Th2, can form complexe with Foxp3, which suppress Th2 responses (152). Foxp3 is a forkhead box transcription factor whose expression is restricted to Treg cells, although under some circumstances conventional T cells can transiently express Foxp3 (153).

2.4.1.2.2 Function of Tregs

Under normal physiological conditions, Tregs regulate the expansion and activation of T and B cells and have a critical role in maintaining the homeostasis of innate CTLs (154). However, in some tumor types including breast cancer and hepatocellular carcinoma, increased numbers of Tregs correlate with reduced OS (155, 156), whereas in other types, such as colorectal cancer, Treg cells are associated with improved survival.

CD25 is a good marker for Tregs in experimental animals that are held under pathogen-free conditions. However, humans are constantly exposed to foreign antigens,

leading to a significant fraction of recently activated CD25+ effector T cells. In search of more specific Treg markers, the transcription factor Foxp3 has been identified as uniquely expressed in Tregs (157) and expression has been proposed as a lineage marker.

Characteristics of CD4+CD25highFoxp3+ Tregs are their ability to actively inhibit CD4+CD25– T cells, CD8+ T cells, DCs, NK cells, natural killer T (NKT) cells, and B cells in a cell-to-cell contact and dose-dependent manner (158). CTLA-4 and glucocorticoid-induced TNFR-related protein (GITR) are the most prominent molecules among the cell-surface markers associated with Treg phenotype and function. (159, 160).

Murine fibrosarcoma model showed that the majority of TILs at late stage of tumor progression were Tregs. Their depletion during the effector phase successfully enhanced antitumor immunity. Furthermore, local depletion of CD4+ T cells inside the tumor led to eradication of well-established tumors and development of long-term antitumor memory. Suppression of antitumor immunity by Tregs occurs predominantly at the tumor site and that local reversal of suppression, even late during tumor development, can be an effective treatment (161). The tumors actively promotes increase of Tregs through activation of naturally occurring Tregs as well as conversion of non-Treg cells into Treg cells (162).

Tumor-derived prostaglandin E2 resulted in an increase of Treg activity and Foxp3 expression.

2.4.1.2.3 nTregs and iTregs in cancer

Treg cell populations can be divided into two major groups: the thymus-derived Treg cells, known as natural Treg (nTreg) cells, and those that are extrathymically derived, known as induced Treg (iTreg) cells (163). An early experiment using transplanted tumors showed extensive proliferation of Treg cells in the tumor bed and draining lymph nodes in TGF-β dependent manner (164). The tumor may provide an environment in which iTreg cells are generated, in which nTreg cells proliferate, or into which nTreg cells are attracted. Another study postulates that preexisting human Treg cells (likely nTreg cells) migrated into the tumors due to attraction by CCL22 present in the tumors (165) . TGF-β was shown not only to induce iTreg cells, but also to expand preexisting Treg cells (nTreg cells) (166).

2.4.1.3 Helper T cells

A subset of the CD4+cells, including natural regulatory cells and NK T cells (NKT cells), are already distinct differentiated cells on release from the thymus (167). In addition to classical T-helper 1 (Th1) and T-helper 2 (Th2) cells, the distinct cell types have been identified, including T-helper 17 (Th17), follicular helper T cell (Tfh), the regulatory type 1 cells (Tr1) and the potentially distinct T-helper 9 (Th9). The differentiation of the different lineages depends on the complex network of specific cytokine signaling and TFs followed by epigenetic modifications (Figure 23).

The initial step of differentiation of the naïve cells is the antigenic stimulation as a result of interaction of TCR and CD4 as co-receptor with antigen-MHC II complex, presented by APCs. TCR coupled with CD3 activation consequently induces a network of downstream signaling pathways that eventually lead to naïve cell proliferation and differentiation into specific effector cells. Lineage-specific differentiation depends on the cytokine milieu of the microenvironment, as well as on the concentration of antigens, type of APCs, and costimulatory molecules (168).

2.4.1.3.1 Type 1 helper T (Th1) cells

IL-12 and IFNγ are the critical cytokines initiating the downstream signaling cascade to develop Th1 cells (169). IL-12 is secreted in large amounts by APCs after their activation through the PRRs (170). Several TFs in coordination induce full differentiation of the Th1 cells. T-bet is the principal transcription factor, as it significantly enhances the production of IFNγ, and plays important role in suppressing the development of Th2 and Th17 (171, 172). T-bet further induces IFNγ production by the differentiating cells, thereby amplifying T-bet expression and upregulating the expression of IL-12Rβ2. The latter cells can then be selected by the abundant IL-12 from the APCs, thus ensuring selective expansion of the differentiating Th1 cells (173).

Figure 23: T helper-cell differentiation to Th1, Th2 and Th17 (Adopted from (174))

2.4.1.3.2 Type 2 helper T (Th2) cells

IL-4 and IL-2 are critical for Th2 differentiation. The major transcription factor involved in Th2 lineage differentiation includes the IL4-induced STAT6, which upregulates the expression of the master regulator GATA3 (GATA-binding protein) (175). Enhanced Th2 cytokine production, selective proliferation of Th2 cells through recruitment of Gfi-1, and inhibition of Th1 differentiation presumably by interacting with T-bet (176). Moreover, GATA3 was found to suppress Th1 differentiation by downregulating STAT4 (177). Recent studies showed that GATA3 by itself cannot regulate all the Th2-specific genes, but instead needed the collaboration of STAT6 (178). Th2 cell differentiation involves several other transcriptional factors activated downstream to several cytokines, including IL-2, IL-6, and IL-21.

2.4.1.3.3 T helper 9 (Th9) and T helper 17 (Th17) cells

Ongoing researches tend to classify IL9 secreting-Th9 cells as a distinct subset of CD4+ T cells. TGF-β in combination with IL 4 directly induces the differentiation of Th9 cells (179) though more research is necessary to get more insights about the Th9 cells.

IL-6, IL-21, IL-23, and TGF-β are the major signaling cytokines involved in Th17 cells differentiation, and retinoic acid receptor-related orphan receptor gamma-T (RORγt) is the master regulator. The differentiation process can be split into 3 stages, including the differentiation stage mediated by TGF-β and IL-6, the self-amplification stage by IL-21, and the stabilization stage by IL-23. TGF-β alone, at high concentration, can divert lineage differentiation toward iTreg development, through the induction of FOXP3 (149, 180). However, at low concentration and in the presence of IL-6, TGF-β induces Th17 differentiation, production of IL-21 and upregulates expression of IL-23R (180).

2.4.1.3.4 Plasticity of helper T cells

TGF-β caused Th2 cells to convert into Th9 cells (179). Th17 in the presence of IL- 12 switched to Th1 phenotype, and interaction with IL-4 led to the differentiation into Th2 cells (181, 182). Treg showed tendency to convert to Th17 and Tfh. In the presence of IL-6, CD4+CD25+FoxP3+cells upon activation reprogrammed into Th17 (183). FoxP3+Treg in Peyer’s patches differentiated into Tfh, with subsequent interaction with B cells and production of IgA (184). IRF4 inactivation in Foxp3+ cells resulted in Th2 development and increased germinal center formation (152).

2.4.1.3.5 TCR signal strength in mature T cell differentiation

Signal strength from TCR in T cells also plays an important role in T cell differentiation into distinct T cell subsets (185). Strong TCR signals lead to differentiation of naive T cells into Th1/IFNγ-expressing cells, whereas weaker signaling promotes development toward the IL-4/IL-5-producing Th2 lineage (186, 187). Another insight on this point is that during the response to microbial infection, even very weak TCR–ligand interactions are sufficient to activate naive T cells, induce rapid initial proliferation and generate effector and memory cells however, strong TCR ligation was required to sustain T-cell expansion.

2.4.1.4 T cell memory

Memory T cell population have following characters to exert immune memory compared to naïve T cells; an increased frequency of antigen-specific precursors (up to 1000-fold), changes in gene-expression profile that allow faster responses, and homeostatic proliferation leading to longevity.

The duration and the strength of the immunological protection are explained by primary the dose of the pathogen and immune stimulating signals involving signal 1 (TCR signaling), signal 2 (co-stimulation) and signal 3 (inflammatory cytokine, e.g. IFN-α/β, IFNγ, IL-2, IL-12, IL-21, IL-33, TNF-α), which lead to the quantity and quality of the memory T cells to be formed. After vast expansion of effector T cells, memory T cells remain to keep the memory function.

hi hi Central memory T (TCM) cells are mainly CD62L CCR7 and home to secondary lymphoid organs and bone marrow. Effector memory T (TEM) cells are defined based on a CD62LlowCCR7low phenotype and are most commonly found in non-lymphoid tissues.

Functionally, there are some notable differences: TCM cells tend to mount more robust + recall responses and produce IL-2, whereas CD4 TEM cells are immediate producers of + cytokines such as IFNγ and tumor necrosis factor and CD8 TEM cells are immediate producers of cytotoxic proteins. TRM cells are CD103hiCD69hiCD27low and persists in peripheral tissues and does not recirculate (188).

TCM stimulated in vitro appear to be heterogeneous in their ability to differentiate. Some cells can be propagated in a non-effector state by stimulation under neutral conditions and can be induced to differentiate to Th1 and Th2 upon stimulation in the presence of IL-12 or IL-4, respectively. When compared with naïve T cells, TCM have higher sensitivity to antigenic stimulation, are less dependent on co-stimulation, and upregulate CD40L to a greater extent, thus providing more effective stimulatory feedback to DC and B cells. Following TCR triggering, TCM produce mainly IL-2, but after proliferation they efficiently differentiate to effector cells and produce large amounts of IFNγ or IL-4.

In the tumor and chronic viral infection, the generation of functioning memory T cell population is disturbed, primarily because the cognate antigens are not eliminated which results in dysfunction of both effector T cells and memory T cells. The presence of high levels of infiltrating memory CD45RO+ effector-memory CD8+ T cells, correlated with the absence of signs of early metastatic invasion, a less advanced pathological stage, and increased survival (189). Recent report using chronic infection model, transcription

factor IRF4 promoted CD8+ T cell exhaustion and limits the development of memory- like T cells during chronic infection (190). The formation of effector-memory CD8+ T cells producing the effector cytokines IFN-l and TNF predicts the degree of therapeutic efficacy of these vaccines (191).

Additionally, tissue resident memory T cells appear to possess important components in tumor immune surveillance. It was shown that Tissue-resident memory T cells of natural TRM cells or cancer-vaccine-induced TRM directly control tumor growth.

TRM infiltration into various human cancers, including lung cancer, are correlated with better clinical outcome independently of CD8+ T cells. TRM cells also predominantly express checkpoint receptors such as PD-1, CTLA-4 and TIM-3. Blockade of PD-1 with neutralizing antibodies on TRM cells isolated from human lung cancer promotes cytolytic activity toward autologous tumor cells (192, 193).

2.4.1.5 Cross reaction, allo/xeno reaction by T cells

2.4.1.5.1 Reactions against allo/xeno products

As we used allo-derived and xeno-derived cells for the vaccination, I would like to discuss about the immune reactions that can be expected when we used those products. As a core of the adaptive response and especially one of the important parts of the vaccination strategy rely on the correct recognitions of the antigens by T cells, which leads to the development of the immune memory against those antigens. The antigenic peptides that are placed on the MHC molecule are subject to be censored by TCR. In the classical view, TCR recognize peptide-MHC complex (pMHC) which is constituted with only self-MHC by positive selection in thymus. However, it is estimated that 1–10% of TCR react to non-self MHC molecules (alloreactivity or xenoreactivity) (194).

The ability of a single TCR to recognize two distinct peptide/MHC complexes, termed T cell cross reactivity, occurs frequently and has been suggested to be critical to recognition of foreign antigen (195). It had been proposed that T cell cross reactivity was the result of similarities in the charge and/or shape of the antigenic surface presented to the TCR, an idea termed molecular mimicry. However, recent structural studies of cross- reactive TCR ligands demonstrate that cross reactivity is not necessarily due to any obvious resemblance between pMHC surfaces (196). AHIII 12.2 CD8+ T cells was originally generated by injection of a human B cell into a C57BL/6 mouse. Previous studies with AHIII 12.2 T cells demonstrate that this T cell clone is xenogeneic and

recognizes both human HLA-A2 in complex with the p1049 (ALWGFFPYL) peptide, and murine H-2 Db in complex with the p1058 (FAPGFFPVL) peptide (197, 198). CD8 independence of xenoreactive p1049/A2 and CD8 dependence of p1058/Db is seen by multimer binding (199).

The study by Birnbaum et al. is an elegant attempt to quantify the cross-reactivity of a given TCR (200). Using five different CD4 TCR clones (three from mouse origin and two from human origin), high throughput screening of yeast libraries and deep sequencing, the authors demonstrate that a single TCR can interact with more than 100 different peptides. Crystal structures of allo-pMHC complexes such as 2C TCR with allogeneic H-2Kbm3 presenting dEV8/Kbm3 (201) or BM3.3 TCR with allogeneic pBM1–H-2Kb (202) have shown that alloreactive TCRs interact with allogeneic MHC in a similar fashion as with syngeneic MHC.

2.4.1.5.2 Structural evidences

The mechanism of T cell allo- and xeno-reactivity remains one of the paradoxes of modern immunology. The most popular model proposed to answer this paradox is that the molecular surfaces of the allo- or xeno- and syngeneic MHC antigens are similar in shape or/and charge regardless of their derivation.

It was suggested that TCR cross-recognition of these class I/peptide epitopes is not due to simple shape and/or charge mimicry (202). These observations can be best interpreted using a functional mimicry model, where TCR makes contacts with both common and different features on the two class I/peptide complexes and the mimicry is functional instead of structural. The common features between the two complexes mostly come from conserved residues and may help to steer TCR into a common orientation on different class I/peptide complexes (203).

Speir et al. (204) hypothesized a modified molecular mimicry model, in which a critical local charge mimicry (formed by residues SerP7 and AspA77 in Kb/dEV8 and AspP8 and AsnA77 in Ld/QL9) contributes to the recognition of these two complexes by the same TCR. The functional mimicry model implies that TCR cross-reactivity, like antibody cross-reactivity, is almost unpredictable within the current knowledge.

2.4.2 B cells

B cell development encompasses a continuum of stages that begin in primary lymphoid tissue (fetal liver and fetal/adult marrow), with subsequent functional maturation in secondary lymphoid tissue (lymph nodes and spleen). The functional end point is antibody production by terminally differentiated plasma cells (205).

Initial priming of naive B cells and subsequent cognate contact with Tfh cells initiate immunoglobulin class switching and the differentiation of some B cells into plasma cells. Priming naive B cells can be occurred by soluble antigens that freely diffuse into lymphoid follicles or by that are transported through the lymphoid system of conduits, subcapsular sinus macrophages or DCs.

In breast cancer, improved metastasis-free/progression-free survival was correlated with B cell-gene expression signatures, which were observed mainly to the basal-like and HER2-enriched subtypes with high expression of a low-diversity population of BCR gene segments (206). Three cohorts of untreated, node-negative breast cancer patients showed that the humoral immune system plays a pivotal role in metastasis-free survival of breast cancer (207).

Screening analysis of autoantibodies (AAbs) developed in basal-like BC identified 13 AAbs (CTAG1B, CTAG2, TP53, RNF216, PPHLN1, PIP4K2C, ZBTB16, TAS2R8, WBP2NL, DOK2, PSRC1, MN1, TRIM21) that distinguished from controls with 33% sensitivity and 98% specificity. Moreover, a strong association of TP53 AAbs with its protein expression (P =0.009) in BLBC patients and MN1 and TP53 AAbs were associated with worse survival (208). Autoantibodies directed at cytokeratin 5/6 and 14 were elevated in plasmas of pre-diagnosed human TNBC but not in luminal type of breast cancer (209). Moreover, AAbs targeting CTAG1B, CTAG2, and TP53 were significantly higher in BLBC patients' plasma compared to the other subtypes of BCs. Using TCGA breast cancer data they found CTAG1B, RNF216, and PSRC1 showing significantly elevated mRNA levels in BLBC compared with other subtypes.

In a survey of naturally occurring CD4+ T cell responses against NY-ESO-1 in melanoma patients, no patients that tested seronegative for NY-ESO-1 had detectable CD4+ T cell responses. On the contrary, 11 of 13 cancer patients with serum antibodies to NY-ESO-1 had polyclonal CD4+ T cell responses directed against various known and previously undescribed NY-ESO-1 epitopes. NY-ESO-1 peptide 80–109 was the most immunogenic, with 10 of 11 patients responding to this peptide (210). Majority of patients

with advanced cancers developing NY-ESO-1 specific antibodies and CD8+ T cells in response to their own tumors. Another survey of anti-HER2 immunity in breast cancer patients has shown that some patients have both existing CD4+ helper/inducer T-cell immunity and antibody-mediated immunity to HER2 protein. (211).

2.5 Cytokines

TGF-β

TGF-β has been suggested to be the principal immune-suppressive factor secreted by tumor cells. TGF-β suppresses IL-12 substantially and inhibits IL-2 and interleukin-2- induced proliferation in T cells (212). In CD8+ cytotoxic T and NK cells, TGF-β is a strong antagonist of IFNγ production. TGF-β has a negative effect on B-cell proliferation and differentiation (213). TGF-β is needed for the differentiation of both Th17 and iTregs. In breast cancer patients, raised TGF-β1 serum concentrations were independently detected and were associated with poor prognosis and with metastases.

Il-10

TGF-β also upregulates immunosuppressive cytokine IL-10, whereas IL-10 enhances the expression of TGF-β in a positive feedback (214). IL-10 inhibits antigen presentation, MHC class II expression, and the upregulation of costimulatory molecules CD80 and CD86. IL-10 prevents the production of Th1-associated cytokines IL-2 and IFNγ from APCs. Physiologically, IL-10 significantly suppresses the major inflammatory cytokines IL-1, IL-6, IL-12, and TNF-α. As a consequence, IL-10 was initially classified among the immunosuppressive cytokines with Th2 polarization in the termination of the T-cell-mediated response.

IL-2

IL-2 is a growth factor for antigen-stimulated T cells and is responsible for T-cell clonal expansion after antigen recognition in adaptive immunity. IL-2 is produced primarily by activated CD4+ T cells and by naive CD8+ T cells and DCs (215). IL-2 stimulates proliferation and differentiation of NK cells. Activated cytotoxic T cells need IL-2 as a growth factor at late stages of the immune reaction. IL-2 is also crucial for

development and peripheral expansion of Treg cells. In the absence of IL-2, TCR engagement results in anergy (214).

IFNγ

IL-12 is essential for the production of IFNγ, which in turn has an essential role in MHC expression. IFNγ is the principal macrophage-activating cytokine. The major sources of IFNγ are CD8+ T cells, CD4+ T cells and NK cells. Activated CD8+ T cells produce IFNγ after antigen stimulation. IFNγ inhibits the production of the immunosuppressive factors TGF-β and prostaglandin E2, whereas negative regulators of IFNγ production include IL-4, IL-10, and TGF-β. TGF-β inhibits IFNγ production by suppression of T-bet that is essential for Th1 differentiation of CD4+ T cells and for IFNγ production.

Tumor necrosis factor (TNF)-α

Macrophages respond to inflammatory stimuli by immediate production of TNF-α, IL-1, and other chemokines. TNF-α and IL-1 are described as acute-response cytokines. The principal function of TNF-α is to stimulate the activation and recruitment of neutrophils and monocytes to sites of inflammation. Malignant cells constitutively produce small amounts of TNF-α, causing hyperpermeability of blood vessels. This effect promoted pleural effusion in a lung-cancer model (216). Increased serum concentrations of TNF-α were described in eight independent cancer types including breast cancer. In established tumors, TNF-α contributes to the maintenance of a proinflammatory environment.

2.6 Chemokines Chemokines are small, secreted proteins that are best known for their roles in mediating immune cell trafficking and lymphoid tissue development (217). The chemokines are the largest subfamily of cytokines and can be further subdivided into four main classes depending on the location of the first two cysteine (C) residues in their protein sequence: namely, the CC-chemokines, the CXC-chemokines, C-chemokines and CX3C-chemokines (218). There is an important degree of redundancy in the chemokine superfamily, with many ligands binding different receptors and vice versa. The numbers

and types of cell that make up the leukocyte infiltrate in solid tumors are related to the local production of chemokines by both the tumor cells and stromal cells (219).

Recruitment of T cells and NK cells

Effector CD8+ T cells, Th1 cells and NK cells express CXCR3, which is the receptor for the Th1-type chemokines CXCL9 and CXCL10, and they can migrate into tumors in response to these chemokines (Figure 24). Increased levels of CXCL9 and CXCL10 are associated with increased numbers of tumor-infiltrating CD8+ T cells, and correlate with decreased levels of cancer metastasis and improved survival in patients with ovarian cancer and colon cancer (189, 220).

Figure 24: The promotion of tumor immunity by chemokines (Adopted from Zou. 2017 (221))

Recruitment of Tregs

Tregs express CCR4 and are recruited in response to CCL22, which is produced mainly by macrophages and tumor cells (165). Tregs express CCR10 and migrate in response to the CCL28 that is found in hypoxic regions (222). High frequencies of Treg cells are found in the bone marrow, via the CXCL12–CXCR4 signaling pathway, which supports metastasis to bone marrow (223).

Chemokine expressions on tumor cells, migration and metastasis

Infiltrating leukocytes are not the only cells that respond to chemokine gradients in cancers; cancer cells themselves can express chemokine receptors and respond to chemokine gradients. Organ-specific metastasis might be governed, in part, by interactions between chemokine receptors on cancer cells with metastatic potential and chemokine gradients in target organs.

CXCR4 expression is generally a characteristic of the malignant epithelial cells. Vascular endothelial growth factor (VEGF) can induce the expression of CXCR4 (224). Furthermore, in breast cancer, the transcription factor NF-κB which controls cell motility, upregulates CXCR4 (225). High CXCR4-expressing breast tumors also produced more extensive nodal metastasis compared with low CXCR4-expressing tumors. Sub- populations of breast cancer cells with increased metastatic abilities in nude mice are characterized by the overexpression of four genes including CXCR4 (225). In another breast cancer mouse model, daily treatment of tumor-bearing mice with an antagonist of CCR1 and CCR5 (receptors of CCL5) led to anticancer effects at doses similar to those that have activity in animal models of inflammatory disease (219).

3. Chapter 3: Cancer immunotherapy

3.1 History

While the recent major success of cancer immunotherapy, the relationship between cancer and immune system has been recognized since long time. Some physicians have remarked that there are more numbers of cancer developments in immune deficient patients and there are some reports of individuals whose tumors spontaneously regressed after infections by pathogens etc.

In late 1800s, Busch and Fehleisen independently observed the regressions of the tumors after spontaneous erysipelas infection, followed by the identification of Streptococcus pyogenes (226, 227). Coley injected heat-inactivated bacteria (Coley’s toxins) to inoperable patients in 1891 and has seen many cases of the regressions or cure of the tumors (228). After certain period of less interest to this type of cancer treatment because of main stream by chemotherapy and radiation therapy, Morales et al. showed effectiveness of bacterium BCG for the treatment of bladder cancer in 1976 (229). Old et al. discovered TNF-α based on the extensive cancer immunology research (230).

Here, we see that huge attempts of cancer immunotherapy and the discoveries of molecular mechanisms behind tumor immunity brought the great advances to the cancer immunotherapy field. Each findings including, immunological tolerance (1956) (231), immune rejection of transplanted mice (1957) (232), immune surveillance theory (1959) (233), TCR (1982) (234), human tumor antigen recognized by T cells (1991) (235), immune check points and those inhibitors such CTLA-4 (1996) (236), PD1 (1992) (237) and LAG3 (1990) (238) provide critical insights to improve cancer immunotherapy. Allison and Honjo received the Nobel prize in 2018 for their discovery of cancer therapy by inhibition of negative immune regulation.

In this recent 10 years, efforts and results in research and development of cancer immunotherapy has been tremendous and donated strong options to cure cancers. FDA has approved autologous cell-based cancer vaccine (sipuleucel-T) for the treatment of stage IV prostate cancer in 2010, anti-CTLA (ipilimumab) for the treatment of stage IV melanoma in 2011, anti–PD-1 (pembrolizumab) for the treatment of melanoma and anti– PD-L1 (atezolizumab) for the treatment of bladder cancer in 2016 and many clinical trials are on-going with promising results currently.

(Year) (Event) First report of an intentional infection of a cancer patient with erysipelas by 1868 Wilhelm Busch, with notable shrinkage of the tumor

William Coley injected his first of many cancer patients with bacteria, reporting 1891 tumor regressions in many of them

1901-08 Rejection of transplanted tumors in mice, reported by Carl Jensen & Leo Loeb

Discovery of acquired immunological tolerance by R. Billingham. Brent & 1956 P.Medawar. Nobel Prize awarded to Medawar & F. Burnet in 1960

Immune rejection of transplanted syngeneic tumors (i.e., each tumor is 1957 antigenically unique) Reported by Richmond Prehn & Joan Main.

1959 Immune surveillance of cancer theory by Lewis Thomas & F. Macfarlane Burnet BCG shown to have anti-tumor effect in mouse model by Lloyd Old, Donald Clark & Baruj Benacerraf

1982 Discovery of the TCR in 1982, reported by James Allison, B. McIntyre, & D. Bloch

First report of a human tumor antigen recognized by T-cells, reported by Pierre 1991 van der Bruggen, C. Traversari, P. Chomez, et al.

First autologous cell-based cancer vaccine (sipuleucel-T) is approved by the FDA 2010 for the treatment of metastatic, asymptomatic stage IV prostate cancer First successful use of gene-edited T-cells for the treatment of hematologic malignancies in humans, reported by W.Qasim, H. Zhan, S. Samarasinghe et al. Anti-CTLA-4 (ipilimumab), is the first inhibitory checkpoint inhibitor (CPI) 2011 approved by the FDA for treatment of stage IV melanoma

A second class of CPIs, anti–PD-1 (pembrolizumab), is approved for the 2016 treatment of melanoma

A third class of CPIs, PD-L1(atezolizumab), is approved for treatment of bladder 2016 cancer

Figure 25: History of cancer immunology and immunotherapy. (Modified from Oiseth and Aziz (239))

3.2 Therapeutic approaches

3.2.1 Cytokines and Vaccines

Cytokines like IL-2 and IFN-α have been also used for cancer treatment but those are relatively non-specific actions when delivered to the patients. High-dose IL-2 for metastatic renal cell carcinoma and metastatic melanoma was first approved by FDA in 1992 and 1998, respectively (240). Despite its known effect to the cancer, the toxicity accompanied with high dose IL-2 treatment is not negligible. Capillary leak syndrome is a life-threatening toxicity resembling septic shock that may occur with intravenous high- dose IL-2. The increased IL-2 in the circulation and immune stimulation may cause massive cytokine release and inflammatory reaction. Side effects include hypotension, gastrointestinal symptoms, renal failure, neurological side effects etc. and the management by the experienced physician is essential (241).

Sipuleucel-T (Provenge) is the first and only cancer vaccine that has been approved by FDA. Dendritic cells from the patient are exposed to prostatic acid phosphatase and GM-CSF and reinfused into the patient, which results in a 4-month increase in median survival (242). Vaccine using autologous tumor cells processed by irradiation for example, is one of the earliest attempt (243) as cancer vaccine, which often combined with stimulating adjuvant. Major advantage of whole tumor cell vaccines is its potential to present the entire spectrum of own tumor-associated antigens to the patient's immune system. Autologous tumor cells were also transduced with cytokines, co-stimulatory molecules or virus. Newcastle disease virus -infected autologous tumor cells were shown to induce tumor protective immunity in animal tumor models (244). Transduction with IL-12 resulted in strong tumor suppression in mice accompanied by high IFNγ production and increased activation of CTL and NK cells (245).

GM-CSF-transduced autologous tumor cell vaccines (GVAX) have been shown anti-tumor effect (246) in pancreatic cancer, prostate cancer and melanoma in the clinical trials, which recruits dendritic cells (DCs) for presentation of tumor antigens and priming of CD8+ T cells and also promotes maturation of DCs by upregulating B7-1 expression (247)

Allogeneic whole tumor cell vaccine is another strategy using whole tumor cells, which typically contains two or three established human tumor cell lines, which can realize limitless sources of tumor antigens and standardized and large-scale vaccine production (248). “Canvaxin” is an allogeneic whole-cell cancer vaccine consisting of

three melanoma lines combined with BCG as an adjuvant, showing superior OS in phase II trial though it couldn’t achieve the determination of efficacy in the phase III trial (249).

Dendritic cell vaccine is a concept to present the desired antigen(s) to dendritic APCs or, prosecutor cells and/or to make them mature with immune stimulating phenotype. As mentioned above, one of DC vaccines has been approved by FDA. Within this approach, the presented antigens include; tumor-derived proteins or peptides, whole tumor cells and viruses. The stimulatory agents or modification of DCs includes; GM- CSF, CD40L, CD70, GITRL, 4-1BBL, OX40L, IL-12p70, IL-2 and IL-18. Majority of them are intended to polarize DCs to Th1 and CTL responses (248).

3.2.1.1 Protein/peptide-based vaccines

The discoveries of tumor associated antigens like MAGE, NY-ESO-1 led to seek the targeted treatment. Properties of the different tumor antigens are discussed in the different chapter. Protein/peptide-based vaccines target the tumor cells which presents specific antigens or shared antigens and are designed to eliminate those cells by immunologic cytotoxicity and/or potentially antibodies. Peptide-based vaccine often targets T cell epitopes. Tumor associated antigens, tumor specific antigens or neoantigens are used for vaccine products.

The vaccines are generally administered with immune stimulatory adjuvants for effective immune response. Adaptive immune responses are preceded by, and partially dependent on, innate immunity receptors triggered by microbial components at infection (250) as mentioned earlier. Aluminum salts is a traditional adjuvant for the vaccines against infectious diseases which are known to induce humoral immunity. For the adjuvant for cancer vaccine should be effective also cell-mediated immunity. Each adjuvant has the own character of its immune stimulating pathway and it is important to choose the adjuvant that have the desired immune stimulating pathway to match the pathological state of each cancer or immune status of the patients. The characteristics of the different adjuvants are discussed in Chapter 3.4. Main advantage of whole protein vaccine compared to the peptide vaccine is a large reservoir of potential peptides that match with own MHCs and increase the chance of polyepitope-directed T- and B-cell responses.

3.2.1.2 DNA/RNA based vaccines

DNA vaccines are bacterial plasmids that are designed to deliver tumor antigens for generating cellular and humoral immunity (251). It is not fully understood how DNA- uptake and antigen-presentation occur but DNAs are thought to be internalized into the cells and the processed antigens by the cells are presented to antigen presenting cells.

RNA vaccines are composed of RNAs, which are translated into protein antigens. Two major types are non-replicating mRNA and virally-derived, self-amplifying RNA. Naked mRNA is quickly degraded by extracellular RNases and is not internalized efficiently. Therefore, a great variety of transfection reagents have been developed that facilitate cellular uptake of mRNA and protect it from degradation. Once the mRNA is internalized to cytosol, the cellular translation machinery produces protein that undergoes post-translational modifications, resulting in a properly folded, fully functional protein, which is the functionally target of RNA vaccines.

One recent successful example is RNA personalized vaccine targeting mutations in stage III and IV melanoma patients after exome and RNA-sequencing of the tumor. Mutations encoding for peptide variants were predicted to bind to HLA class I and II. Following vaccination, all 13 patients showed significantly reduced cumulative recurrent metastatic events and sustained PFS (252).

3.2.1.3 Viral based vaccines

The rationale for using viruses as immunization vehicles is based on the phenomenon that viral infection often results in the presentation of MHC class I/II restricted. The viral vectors with low disease-causing potential and low intrinsic immunogenicity are engineered to encode TAAs (248). Viral based vectors in cancer vaccines include Poxviridae family, such as vaccinia, modified vaccinia strain Ankara (MVA), and the avipoxviruses (fowlpox and canarypox; ALVAC), adenovirus and Herpes simplex virus type 1 (HSV-1) (253)

3.2.1.4 Immunological perspective behind cancer vaccine

To date, the therapeutic cancer vaccines that have been applied to tumor-bearing patients have shown little success. Clinical responses occur in only 5–10% of the patients, whereas 80% of the patients do not show any sign of tumor regression (254). The initial view of the process occurring in the responding patients had been that the CTLs that were

activated by the vaccine would have to move in large numbers to the tumor sites and directly destroy the tumor cells. However, the findings so far add and correct this viewpoint.

First, large numbers of T cells directed against various tumor-specific antigens are already present in the blood and the tumor before vaccination (255). Second, in most responding patients, very low numbers of anti-vaccine T cells are observed in blood and into the tumors. The frequency of these T cells in regressing tumors can be as low as 1 per 106 tumor cells (256). Third, in the regressing tumor, we observe a considerable expansion of pre-existing and new T cells against various antigens expressed by the tumor, so that the frequency of these antitumoral T cells can reach 1 per 100 tumor cells (257).

It is now clear that most patients with melanoma produce a spontaneous T cell response against their tumor (258). In many patients, this attempt at eliminating the tumor evidently fails and a large number of antitumor T cells remain in an inactive state. Inside the tumors, this anergy seems to be reinforced by an immunosuppressive environment. For the vaccinated patients who show tumor regression, it is also speculated that vaccination produces a spark that reactivates the anergic memory T cells that recognize tumor antigens

This revised view has several consequences. First, even if only a fraction of the tumor cells expresses the vaccine antigen, this should not prevent a rejection response because the sparking effect results in the activation of CTLs that are directed against other antigens (254). Immunosuppressive tumor microenvironment could prevent the sparking effects of vaccine and T cell response (259).

3.2.2 Adoptive cell therapy

Adoptive cell therapy mostly involves the isolation and ex vivo expansion of tumor- specific T-cells, followed by infusion back into the cancer patient or, intratumoral injection. As a molecular form, it includes TILs, TCR T cells and Chimeric Antigen Receptor T (CAR T) cells. T cells are obtained directly from peripheral blood or from tumor; isolating and expanding particular T-cell clones.

TCR-based therapy uses T cell which has genetically engineered TCR to target specific tumor antigens. First clinical trial has used TCR that bound an HLA-A2– restricted peptide from a melanocytic differentiation antigen (260) in which 2 of 13 patients experienced tumor regression upon adoptive transfer of autologous T cells

engineered to express a MART-1-reactive TCR. A follow-up study utilizing TCR with higher-avidity targeting MART-1 epitope showed an improved response (6 of 20). Unfortunately, severe on-target off-tumor toxicity, mostly affecting normal melanocytes in the skin, eye, and ear highlighted the need to target antigens (nearly) absent in critical normal tissues in TCR-based therapy.

3.2.2.1 Chimeric antigen receptor T (CAR T) cells

A CAR combines antigen-binding domains (a single-chain variable fragment (scFv) derived from the variable domains of antibodies) with the signaling domains and additional costimulatory domains (261). Very recently, autologous CAR T-cells specific for CD19 present in B lymphocytes have recently been approved by FDA for treatment of refractory pre-B cell acute lymphoblastic leukemia and diffuse large B cell lymphoma.

Figure 26: Engineered T cells: design of TCR versus CAR T cells. (Quoted from June et al. (261) )

CARs overcome some limitations of engineered TCRs, such as the need for MHC expression, MHC identity and co-stimulation (262). This character gives an advantage of CAR T therapy as the tumor cells often reduce or loose the expression of MHC I.

The molecular structure of CAR T cells has been evolved which have delivered improved results. Second-generation CAR T cells targeting CD19 and encoding costimulatory domains brought unexpected outstanding success. CD19 is often highly expressed in B cell malignancies and it is not expressed outside of the B cell lineage. CD19-specific CAR T cell treatment yielded 70% response rates in patients with B-cell acute lymphoblastic leukemias. However, therapy resistance was also emerged

caused by the loss of the antigenic epitope on CD19. The underlying mechanism was that the selection for preexisting alternatively spliced CD19 isoforms (263).

Despite its remarkable results in clinical trials, some deaths have occurred in the trial due to massive cytokine release (cytokine storm) and cerebral edema (261). To provide enhanced specificity toward tumors, combined sensing approaches are increasingly being developed that target two or more antigens. Kloss et al., demonstrated that co-transduced T cells destroy tumors that express both prostate tumor antigens (PSMA and PSCA) but do not affect tumors expressing either antigen alone (264).

3.2.3 Immune check point inhibitors (CPIs)

Under normal physiological conditions, immune checkpoints are crucial for the maintenance of self-tolerance and also to protect tissues from damage when the immune system is responding to pathogenic infection. However, the expression of immune checkpoint proteins can be dysregulated by tumors as an important immune resistance mechanism. The antibodies that block immune checkpoints target immune receptors or their ligands in order to enhance endogenous antitumor activity.

Various ligand–receptor interactions between T cells and APCs that regulate the T cell response have been found. These responses can occur at the initiation of T cell responses. In general, T cells do not respond to these ligand–receptor interactions unless they first recognize their cognate antigen through the TCR. Many of the ligands bind to multiple receptors, some of which deliver co-stimulatory signals and others deliver inhibitory signals (Figure 27).

In 1996, CTLA-4-blocking monoclonal antibodies (Mabs) was shown to exert anti- tumor immunity in animal models (236). Currently, Mabs against PD-1, PD-L1 and CTLA-4 have been approved by FDA and rapidly introduced to the primary care (Table 11). Anti-PD-1 and anti–PD-L1 antibody treatments are currently the most investigated because they have shown less severe toxicity (or, high-grade “immune-related adverse effects”) than anti-CTLA-4 antibody treatments (5-20% compared to 10-40% respectively) (265)

Figure 27: Multiple co-stimulatory and inhibitory interactions regulate T cell responses (Adopted from (266))

Table 10: Summary of the biological and molecular functions of T-cell costimulatory molecules (Adopted from Wei 2018(267)) Molecule Ligand(s) Receptor expression pattern Biological function Molecular function

CTLA4 B7-1 (CD80), Activated T cells, Treg Negative T-cell co-stimulation (primarily at priming); prevent Competitive inhibition of CD28 co-stimulation B7-2 (CD86) tonic signaling and/or attenuate high-affinity clones (binding of B7-1 and B7-2)

PD-1 PD-L1, PD-L2 Activated T cells, NK cells, NKT Negative T-cell co-stimulation (primarily in periphery); attenuate Attenuate proximal TCR signaling, attenuate cells, B cells, macrophages, peripheral activity, preserve T-cell function in the context of CD28 signaling subsets of DC; as a result of chronic antigen inflammation

PD-L1 PD-1, B7-1 Inducible in DC, monocytes, Attenuate T-cell activity in inflamed peripheral tissues PD-1 ligation; cell-intrinsic mechanism unclear (CD80) macrophages, mast cells, T cells, B cells, NK cells

LAG3 MHC-II, Activated CD4 and CD8 T cells, Negative regulator of T-cell expansion; control T-cell Competitive binding to MHC-II; proximal LSECtin NK cells, Treg homeostasis; DC activation LSECtin mechanism unknown

TIM3 Galectin-9, Th1 CD4 and Tc1 CD8, Treg, DC, Negative regulation of Type 1 immunity; maintain peripheral Negative regulation of proximal TCR PtdSer, NK cells, monocytes tolerance components; differences between ligands HMGB1, unclear CEACAM-1

TIGIT PVR (CD155), CD4 and CD8, Treg, Tfh, NK cells Negative regulation of T-cell activity; DC tolerization Competitive inhibition of DNAM1 (CD226) co- PVRL2 stimulation (binding of PVR), binding of DNAM1 (CD112) in cis; cell-intrinsic ITIM-negative signaling

VISTA Counter- T cells and activated Treg, Negative regulation of T-cell activity; suppression of CD4 T cells Increase threshold for TCR signaling, induce receptor myeloid cells, mature APC FOXP3 synthesis; proximal signaling unknown unknown

ICOS ICOSL Activated T cells, B cells, ILC2 Positive co-stimulation; Type I and II immune responses; Treg p50 PI3K recruitment (AKT signaling); enhance maintenance; Tfh differentiation calcium signaling (PLCγ)

OX40 OX40L Activated T cells, Treg, NK cells, Sustain and enhance CD4 T-cell responses; role in CD8 T cells Regulation of BCL2/XL (survival); enhance NKT cells, neutrophils and Tregs PI3K/AKT signaling

Molecule Ligand(s) Receptor expression pattern Biological function Molecular function

GITR GITRL Activated T cells, Treg, B cells, NK Inhibition of Tregs; co-stimulation of activated T cells, NK cell Signal through TRAF5 cells, macrophages activation

4-1BB 4-1BBL Activated T cells, Treg, NK cells, Positive T-cell co-stimulation; DC activation Signal through TRAF1, TRAF2 (CD137) monocytes, DC, B cells

CD40 CD40L APCs, B cells, monocytes, APC licensing Signal through TRAF2, 3, 5, 6; TRAF- nonhematopoietic cells (e.g., independent mechanisms? fibroblasts, endothelial cells)

CD27 CD70 CD4 and CD8 T cells, B cells, NK Lymphocyte and NK cell co-stimulation; generation of T-cell Signal through TRAF2, TRAF5 cells memory

Table 11: Summary of FDA approvals for immune checkpoint blockers

Tumor type Therapeutic agent FDA approval year Melanoma Ipilimumab (anti-CTLA4) 2011 Melanoma Nivolumab (anti–PD-1) 2014 Melanoma Pembrolizumab(anti–PD-1) 2014 Non-small-cell lung cancer Nivolumab (anti–PD-1) 2015 Non-small-cell lung cancer Pembrolizumab(anti–PD-1) 2015 Ipilimumab (anti-CTLA4) + Melanoma (BRAF wild-type) Nivolumab (anti–PD-1) 2015 Melanoma (adjuvant) Ipilimumab (anti-CTLA4) 2015 Renal cell carcinoma Nivolumab (anti–PD-1) 2015 Hodgkin lymphoma Nivolumab (anti–PD-1) 2016 Urothelial carcinoma Atezolizumab (anti–PD-L1) 2016 Head and neck squamous cell carcinoma Nivolumab (anti–PD-1) 2016 Head and neck squamous cell carcinoma Pembrolizumab(anti–PD-1) 2016 Ipilimumab (anti-CTLA4) + Melanoma (any BRAF status) nivolumab 2016 Non-small-cell lung cancer Atezolizumab (anti–PD-L1) 2016 Hodgkin lymphoma Pembrolizumab(anti–PD-1) 2017 Merkel cell carcinoma Avelumab (anti–PD-L1) 2017 Urothelial carcinoma Avelumab (anti–PD-L1) 2017 Urothelial carcinoma Durvalumab (anti–PD-L1) 2017 Urothelial carcinoma Nivolumab (anti–PD-1) 2017 Urothelial carcinoma Pembrolizumab(anti–PD-1) 2017 MSI-high or MMR-deficient solid tumors of any histology Pembrolizumab(anti–PD-1) 2017 MSI-high, MMR-deficient metastatic colorectal cancer Nivolumab (anti–PD-1) 2017 Pediatric melanoma Ipilimumab (anti-CTLA4) 2017 Hepatocellular carcinoma Nivolumab (anti–PD-1) 2017 Gastric and gastroesophageal carcinoma Pembrolizumab(anti–PD-1) 2017 Non-small-cell lung cancer Durvalumab (anti–PD-L1) 2018 Ipilimumab (anti-CTLA4) + Renal cell carcinoma Nivolumab (anti–PD-1) 2018 Abbreviations: MSI, microsatellite instability; MMR, mismatch repair.

3.2.3.1 Anti-CTLA4 antibodies

CTLA4 is expressed exclusively on T cells where it regulates the amplitude of the early stages of T cell activation. CTLA4 counteracts the activity of the T cell costimulatory receptor, CD28 (268). Once the antigen recognition occurs, CD28 signaling strongly amplifies TCR signaling to activate T cells. CD28 and CTLA4 share identical ligands: CD80 and CD86 (269). It has been proposed that CTLA4 expression dampens the activation of T cells by outcompeting CD28 in binding CD80 and CD86, as well as actively delivering inhibitory signals to the T cell (270, 271).

CTLA4 have major roles of downmodulation of helper T cell activity and enhancement of regulatory T cell immunosuppressive activity (272)(273). The initial studies demonstrated significant antitumor responses when mice bearing partially immunogenic tumors were treated with CTLA4 antibodies as single agents. Poorly immunogenic tumors did not respond to anti-CTLA4 as a single agent but did respond when anti-CTLA4 was combined with a GM-CSF-transduced cellular vaccine (274). Ipilimumab was the first CPI therapy approved by FDA for the treatment of advanced melanoma in 2010 by demonstrating a survival benefit for patients with metastatic melanoma.

3.2.3.2 Anti-PD1/PDL1 antibodies

The major role of PD1 is to limit the activity of T cells in peripheral tissues at the time of an inflammatory response. Similarly to CTLA4, PD1 is highly expressed on Treg cells, where it may enhance their proliferation in the presence of ligand; PDL1 and PDL2 (275, 276). PD1 is more broadly expressed than CTLA4: it is induced on other activated non-T lymphocyte subsets, including B cells and NK cells which limits their lytic activity (277, 278). Chronic antigen exposure, such as occurs with chronic viral infection and cancer, can lead to high levels of persistent PD1 expression, which induces a state of exhaustion or anergy (279).

Like PD1 is highly expressed on TILs in many cancers, PD1 ligands are also commonly upregulated on the tumor cell surface from many different human tumors (280). PDL1 is also expressed on myeloid cells in the tumor microenvironment (281). Forced expression of PDL1 on mouse tumor cells inhibit local antitumor T cell-mediated responses (282). High PD1 expression on CD8+ TILs showed decreased cytokine production in TILs from melanoma samples from patients (280).

Two general mechanisms for the regulation of PDL1 by tumor cells have been explained: innate immune resistance and adaptive immune resistance.

Innate immune resistance: Constitutive oncogenic signaling can upregulate PDL1 expression on all tumor cells, independently of inflammatory signals in the tumor microenvironment. Activation of the AKT and signal transducer and activator of transcription 3 (STAT3) pathways has been reported to drive PDL1 expression.

Adaptive immune resistance: PDL1 is not constitutively expressed, but rather it is induced in response to inflammatory signals that are produced by an active antitumor immune response. The non-uniform expression of PDL1, which is commonly restricted to regions of the tumor that have TILs, suggests that PDL1 is adaptively induced as a consequence of immune responses within the tumor microenvironment. It is considered that there is a negative feedback loop whereby IFNγ induces PDL1 expression, which in turn suppresses the activity of PD1+ T cells (266).

Two anti-PD1 antibodies (pembrolizumab and nivolumab) and one anti–PD-L1 antibody (atorolimumab) have been approaved for the treatment of many types of cancers (Table 11). Immune related toxicities are observed but with generally lower frequency than that of anti-CTLA-4 antibody, possibly because PD-1 and PD-L1 checkpoint may act later in the T cell response, resulting in a more restricted T cell reactivity toward tumor cells (283). Phase Ib using PD-1 in TNBC showed an overall response rate of 18.5%. Yet, 15.6% of patients with at least one grade 3 to 5 adverse event during the trial (284).

3.3 Cancer antigens

Certain antigens are over expressed on the tumor cells as a result of cell transformation. Those antigens are targets for immunotherapy and represented by Tumor- associated Ags (TAAs), cancer-germline or cancer/testis Ags (CTAs), and tumor-specific Ags (TSAs). The probability of attack of normal tissues by the antitumoral T cells clearly increases for antigens with low tumoral specificity, such as differentiation antigens or overexpressed antigens, as experienced with carcinoembryonic antigen (285) and carbonic anhydrase IX (286). Classification of cancer antigen is shown in Table 12.

3.3.1 Tumor-associated antigens (TAAs)

TAAs are normal host proteins that demonstrate distinct expression profiles between host and tumor cells. In general, the dysregulation of gene pathways as a result of mutations within the tumor cells results in the atypical expression of unmutated proteins that would otherwise be expressed at relatively lower levels, or not at all, in normal cells of that tissue type in its current developmental state (287).

3.3.2 Differentiation antigens

Differentiation antigens (Table 12) are associated with the proteins displaying a cell lineage-specific pattern of expression. Differentiation antigens are expressed only in the tumor cells and in the normal tissue of origin (254).

3.3.3 Cancer/testis antigens (CTAs)

CTAs are the antigens that are expressed in testes, fetal ovaries, or trophoblasts, but are otherwise absent in healthy somatic cells. The attractiveness in targeting CTAs can be attributed to: 1) their disrupted gene regulation in various tumor types, 2) their limited expression in normal tissues, 3) their lack of presentation in germline and trophoblastic cells, which do not display MHC class I molecules on their surface, and 4) their immunogenic potential (287). Besides germline cells, a low level of expression of MAGEA12 in brain cells has recently been reported.

3.3.4 Oncofetal antigens

Oncofetal antigens are proteins which are produced by tumors and by fetal tissues but are absent or much lower concentration by adult tissues. These antigens can either be present on the cell surface binding to cellular membrane, or intracellular proteins that can be presented on MHC I molecule, or secrete soluble membrane. This includes α- fetoprotein (AFP), cancer antigen 125 (CA125), CA19-9 and prostate-specific antigen (PSA).

To note, carcinoembryonic antigen (CEA) was first described in 1965 by Gold and Freedman (288), when they identified an antigen that was present in both fetal colon and colon adenocarcinoma but that was absent from the healthy adult colon, hence its name,

carcinoembryonic antigen. Subsequent work showed that CEA was also present in certain healthy tissues, although concentrations in tumors were on average 60-fold higher than in the nonmalignant tissues (289). Now CEA gene is classified as a member of the immunoglobulin supergene family (290).

Alpha-fetoprotein (AFP)

AFP is a major fetal serum protein that is synthesized mainly by the yolk sac and liver. After birth, circulating levels of AFP drop sharply, virtually disappearing from the blood of normal adult individuals (291). It has been documented that up to 60% of hepatocellular carcinoma patients have elevated serum AFP (292). A subclass of AFP- positive HCCs have a unique gene expression signature and a poor survival outcome (293).

Immature laminin receptor protein (OFA-iLRP)

Experimental studies indicate that OFA-iLRP acts as a receptor for the uptake of prion protein into eukaryotic cells (294). Although the mature 67-kDa form is present on many healthy cells, 37-kDa OFA-iLRP is abundantly expressed in many types of human tumors, including breast, lung, ovary, prostate, renal cancer, and lymphoma (295). OFA- iLRP expression was detected in embryos/early fetuses but not in neonate, or adult differentiated tissues. It has been documented that OFA-iLRP is an immunogenic protein that can specifically activate both T and B lymphocytes (296-298).

Table 12: Types of tumor antigens and their advantages and disadvantages as therapeutic targets (Adopted from Zamora 2018 (287))

Class of Tumor Ag Description of Ag Examples of Ag Type Advantages of Targeting Disadvantages of Targeting Tumor associated antigens Associated with proteins Antigenic targets shared Antigenic expression on normal displaying a cell lineage- between patients healthy tissues CD19, MART-1, gp100, TRP2, Differentiation Ag specific pattern of expression or CEA Development of off-the-shelf Potential for on-target, off- present during specific treatments tumor toxicity developmental stages. Difficult to determine relative Antigenic targets shared Normal cellular proteins abundance on tumor compared WT1, ERBB2, PRAME, RAGE- between patients Overexpressed Ag expressed in greater with normal cells 1, mesothelin abundance in cancerous cells. Development of off-the-shelf Potential for on-target, off- treatments tumor toxicity Expressed in testes, fetal Antigenic targets shared Are not ubiquitously expressed ovaries, or trophoblasts, but between patients with the same MAGE, BAGE, GAGE, NY- across cancer types Cancer/testis (germline) Ag absent in healthy somatic cells. cancer type ESO-1 Selectively expressed by Development of off-the-shelf Potential for unanticipated on- specific types of cancer. treatments target, off-tumor toxicity Tumor specific antigens Gene mutations resulting in the Decreased likelihood of on- Antigenic targets are not expression of new peptides target, off-tumor toxicity shared between patients (from point mutations, altering CDK4, KRAS, BRCA1/2, p53, Mutated Ag Potential sharing of the phase of a gene's reading TGF-βRII driver/fusion mutations between Requirement for patient-specific frame, or chromosomal patients with the same cancer treatments translocations). type Abnormal proteins expressed Viral Ags shared between by cells infected with patients with cancers with viral Relatively low frequency of HPV E6/E7, EBV Oncogenic viral Ag oncoviruses that can be at the etiology cancer types with known viral EBNA1/LMP1/LMP2 origin of several types of Development of off-the-shelf etiology cancers. treatments

3.3.5 Tumor specific antigens (TSAs)

TSAs generally arise from tumor-specific mutations, which result in the exclusive expression of neoantigens in tumors and, by definition, their absence in normal cells. Another source of TSAs are the viral proteins expressed by the cells infected with oncoviruses (Table 12).

3.3.6 Neoantigens (mutant peptides or protein)

When cancer arises due to an accumulation of genetic alterations, which leads to the production and processing of mutant proteins that are otherwise absent from host cells. The sequence-altered proteins are termed “neoantigens” and their mutated epitopes that are recognized by T cells are called “neoepitopes”.

Next-generation sequencing allows rapid sequencing of genomes at low cost. Together with dedicated bioinformatics tools, NGS and Whole Genome or exome Sequencing Aanalysis enables comprehensive mapping of all mutations in a cancer (called the “mutanome”) (299). Neoantigens from exome and transcriptome analysis, are identify followed by in silico epitope prediction and large-scale immunogenicity assays by tumor-infiltrating T cells (300), mass spectrometry (301), or predictictive algorythms of MHC molecule–binding neoepitopes.

Figure 28: Customizing a patient-specific cancer vaccine. (Adopted from Sahin and Türeci. 2018 (299)) Patient tumor biopsies and healthy tissue (e.g., peripheral blood white blood cells) are subjected to next-generation sequencing. By comparing the sequences obtained from tumor and normal DNA, tumor-specific nonsynonymous single-nucleotide variations or short indels in protein-coding genes are identified. A computational pipeline is used to examine the mutant peptide regions for binding to the patient’s HLA alleles (based on predicted affinity) and other features of the mutated protein deemed relevant for prioritization of potential vaccine targets. These data can facilitate selection of multiple mutations to design unique neoepitope vaccines.

Very recently, two studies targeting neoantigens showed benefit in the patients with high mutational load by successfully inducing T cell responses. One was RNA mutanome vaccine using RNA encoding for 5 selected peptide variants based on exome- sequencing and RNA-sequencing of melanoma patient’s tumor samples (252). The other one was Peptide vaccine containing long peptides encoding tumor-specific somatic mutations and selected 13–20 mutations per patient (302). Interestingly, immune response by CD4+T cells were observed higher than that by CD8+ T cells even in the latter study designed for HLA class I binding peptides.

The important aspect of neoantigen-based vaccine is anticipating to broaden the repertoire of neoantigen-specific T cells beyond what is induced spontaneously. Many years, tumor rejection has mainly been considered to be attributed to cytotoxic CD8+ T cells, and vaccine approaches have relied on potential MHC class I epitopes. There is a growing body of data, however, indicating that specific CD4+ T cells contribute critically to the effectiveness of cancer immunotherapy (133). Whereas the primary function of CD8+ effectors is cytotoxic killing of cells presenting their cognate antigen, CD4+ effectors have a wider range of functions, including orchestration of various cell types and cytokines of the adaptive and innate immune system (303).

One of the critical challenges for personalized cancer vaccines is to accurately map the cancer mutanome. Mutations are detected by comparing exome sequencing data generated by NGS from tumor tissue and a matched healthy tissue sample like the patient’s blood cells, so that it can prevent incorrect classification of germline variants as neoepitopes. NGS analysis is being continuously improved. Another challenge is that small biopsy from a single tumor lesion collected for routine diagnostics, and thus sequence data may not be representative of the tumor’s full clonal spectrum. The processing and presentation of antigens is a complex. Only a portion of mutated sequences are presented on MHC at levels sufficient to trigger an effective T cell response. Technical feasibility and costs limit the number of mutations that can be incorporated into a drug product. “Hit rates” of finding a tetramer positive or peptide-reactive T cell population in patients have been as low as 0.5–2% of screened Ags (287).

3.4 Vaccine adjuvants and immune response modifiers

A vaccine adjuvant is a component that can improve the effectiveness of vaccines by inducing robust immune responses. They were initially developed to be used with inactivated vaccines for infectious disease. The purpose is to stimulate the local injection

site and recruit the immune cells effectively to enhance the efficacy of the inactivated vaccines which are often less immunogenic than live attenuated vaccines.

In the cancer immunotherapy, vaccine adjuvants can also be used by mixing with the vaccine product. However, the selection of the adjuvant is different between the vaccines against infectious disease and against cancer. While the adjuvants which stimulate humoral response are chosen for the former, stimulating cellular immunity is principally important for the latter.

The agents that are combined with cancer vaccine are diverse including cytokines, TLRs, classical vaccine adjuvants, the epigenetic molecules and immune modulating chemical compounds. The most frequently used adjuvant for cancer vaccine is GM-CSF. GM-CSF plays a critical role in development and maturation of DCs and proliferation and activation of T cells, linking the innate and acquired immune response (304). TLRs are also used for cancer immunotherapy (Table 13). They are a family of PRRs that function as primary sensors of the innate immune system to recognize microbial pathogens. TLR can be expressed on various members of the innate and adaptive immune system (DCs, macrophages, granulocytes, T cells, B cells, NK cells and mast cells) as well as by endothelial and epithelial cells (305).

Table 13: TLR ligands used for cancer immunotherapy

Conventional chemotherapeutics and the target therapy agents also have been found to have the immune modulatory effect (Annex 2 and Annex 3) (306). Epigenetic agents have also been shown to exert immune modulatory effect, which is discussed in the next chapter.

3.4.1 Epigenetic modulation of immune functions

Epigenetic alterations affect gene expression by modifying chromatin structure without changing the sequence of DNA. DNA methylation, histone modification and non- coding RNA have been found as molecular mechanism underlying behind the epigenetic control.

In the tumor microenvironment, the cancer cells often escape from the immune elimination. Epigenetic modification has critical role to create immune suppressive environment around the tumor cells. This fact proposes possibility of therapeutic intervention by reversing the epigenetic restricts. Indeed, very recent works reported that epigenetic modulators like 5-aza-2′-deoxycytidine (5-AZA-dC), DNA methyltransferase (DNMT) inhibitor, and inhibitor of EZH2 methyltransferase activity, removes the repression in the tumor environment via Th1-type chemokines CXCL9 and CXCL10, improved effector T-cell trafficking and increased TILs, which led to slow down tumor progression and improved efficacy of the checkpoint blockage (307).

Histone deacetylase inhibitors (HDACis) have also been shown various immune modulating effects. Since we tested the impact of HDACi valproic acid (VPA) in stem cell vaccine strategy, I focus the topic on HDACi and their immune modulation effect.

3.4.1.1 Histone modification A nucleosome contains a protein core made of eight histone molecules. 146 base pairs of DNAs wrap a histone octamer, consisting of 2 each of the core histones H2A, H2B, H3, and H4. Histone modification and DNA methylation can be closely linked between their activities. Yet, DNA methylation is generally thought to be more rigid as system to maintain the epigenetic modification whereas histone modification is more plastic that enable to react the rapid environmental change or developmental needs.

In the resting cell, DNA is wound tightly around these basic core histones, excluding the binding of the enzyme RNA polymerase II. This chromatin structure is

described as “closed” associated with the suppression of gene expression. Gene transcription only occurs when the chromatin structure is opened up, with unwinding of DNA so that RNA polymerase II then basal transcription complexes can bind to the naked DNA to initiate transcription (308).

The acetylation state is controlled by antagonizing histone modifying enzymes, histone acetyltransferases (HATs) and deacetylases (HDACs), which add or remove acetyl groups to/from target histones respectively. HAT including GCN5, PCAF, CBP, p300, Tip60 and MOF (309) involves acetylation which is generally associated with transcriptional activation. HDACs involve histone deacetylation which is generally associated with transcriptional repression. 18 HDACs in humans can be grouped into four classes: class I (HDAC1, 2, 3, and 8 with a homology to Rpd3), class II (HDAC4, 5, 6, 7, 9, and 10 with a homology to Hda1), class III (Sirt1, 2, 3, 4, 5, 6, and 7 with a homology to Sir2), and class IV (HDAC11)(310).

3.4.1.2 The role of HDACs in immune system and the effects of HDAC inhibitors (HDACis) HDACs have a role in innate immunity through the regulation of TLR pathways and IFN signaling (311), and in the adaptive immunity through the regulation of antigen presentation and B and T lymphocyte development and function (312) (313). For example, HDAC7 positively regulates TLR-inducible pro-inflammatory gene expression in inflammatory macrophages (314). HDAC9 is essential for maintaining Treg cell homeostasis (315). HDAC1 and HDAC2 appear to negatively regulate antigen presentation through inhibition of the MHC class II transactivator (CIITA).

Exhausted T-cells are characterized by profound changes in their epigenetic landscape compared to memory T-cells (316). Treatment of exhausted T-cells with HDAC inhibitors reversed the exhausted functional state (317). MDSC build-up can be reduced by HDAC inhibitor entinostat, currently in clinical trials in combination with checkpoint inhibitors. Despite the possibility of HDAC inhibitors having a global effect on histone modification, studies have surprisingly found that cancer cells exhibit altered expression in fewer than 10% of genes following HDAC inhibition (318-320) We would have a question, how HDAC inhibitors can have tumor-cell-selective effects. One of the suggested theories is that this is due to multi-tiered and semi-redundant epigenetic regulation in normal cells (321) whereas the transformed tumor cells often rely on the abnormal expression of a set of key genes, which are more limited compared to the normal

cells. By this way, HDACi treatment may lead only the transformed tumor cells and tumor environment to change dramatically.

Despite many trials of cancer vaccines, results in activation of immune response is not sufficient because of immune evasion mediated by TME. Approach overcoming the immune tolerance should enhance cancer vaccine efficacy. HDACi has been shown to increase tumor antigen expression, increase perforin in T cells (322). Valproic acid (VPA) increased the antigenic recognition by CTL in cervical cancer cells (323). VPA mediates recognition of cancer cells by CTLs via NKG2D(324). HDACi can increase the immunogenicity of tumor cells by up-regulating expression of MHC and co-stimulatory molecules such as CD40, CD80, CD86 and intercellular adhesion molecule 1 (ICAM-1) as well as MICA and MICB molecules (325, 326). Class IIa HDAC inhibitors led macrophages to instruct the IFNγ axis to alter the tumor microenvironment and activate a functional adaptive anti-tumor immune response in breast cancer model (327).

3.5 Immunotherapy in breast cancer treatment

In the immunotherapy field, breast cancer has not been the first selection as the target because of the thought to be less immunogenic compared to melanoma or lung cancer. Remarkable success of cancer immunotherapy of other cancers led intensive effort to deliver immunotherapy to breast cancer as well. The most promising among the strategies had been thought to be anti–PD-1 and anti–PD-L1 blocking antibodies for the treatment of TNBC. However, objective response rate (ORR) was modest as 19% for both cases (Table 14). CTLA-4–blocking antibody tremelimumab in combination with exemestane (aromatase inhibitor) in 26 patients with metastatic ER+ breast cancer but found only limited signs of clinical activity. LAG-3 combined with chemotherapy (Paclitaxel) showed relatively good outcome. LAG-3, PD-1 and PD-L1 blockage trials are currently on-going with larger sample size.

Immune check point inhibitors

Patient Molecule Sample size Clinical outcome Immune monitoring population

Increased activated APCs LAG-3 Ig (IMP321) 6-month PFS 90% Metastatic BC 30 Higher% NK + Paclitaxel (historical control 50%) Higher% CD8+ EM T cells

Increased ICOS+ T cells Metastatic Anti- CTLA4 + Increased ICOS+ T cells/ hormone 26 SD> 12 weeks 42% exemestane Foxp3+ Treg ratio in responsive BC peripheral blood

ORR 18.5% Anti-PD-1 32 (27 for Metastatic TNBC 1CR and 2 PRs NR (pembrolizumab) efficacy) 6-month PFS 23.3%

Increased number of ORR 19% Anti-PD-L1 52 (21 for activated proliferating CD8+ T Metastatic TNBC 2CR and 2 PRs (MPDL3280A) efficacy) cells in peripheral blood 6-month PFS 27% Increased serum IL-18 Table 14: Clinical trials of immune check point inhibitors in breast cancer (Modified from Chimino-Mathews et al., 2015 (328))

Various cancer vaccine trials are ongoing and none of them have been approved yet. Those cancer vaccines can be design in two clinical approach: adjuvant cancer vaccine to prevent recurrence of breast cancer after primary treatment and curative cancer vaccine.

The prevention setting against the recurrence is also clinically meaningful settings especially because the breast cancer treatment after recurrence often face lower rate of therapeutic success compared to the primary tumor. HER2 peptide loaded DC vaccine in DCIS showed non-evidence of disease progression after surgery in 18.5% of patients.

Vaccine for the prevention of breast cancer recurrence

Patient Immune Vaccine Sample size Clinical outcome Note population monitoring

5-year DFS 89.7% Eligibility HER2 1+-3+ Early HER2+ BC HER2 specific HER2 E75 peptide 108 5-year DFS 94.5% at Node positive or high NED after memory T cells vaccine +GM-CSF vaccinated optimized dose risk negative adjuvant Decreased T regs adjuvant 79 control (p=0.05 versus Toxicity was minimum therapy Decreased TGF-β control) Intradermal injection

HER2 specific T cells HER2 DC 18.5% with NED at loss of HER2 DCIS 27 Eligibility HER2 2+-3+ intranodal surgery expression in 50% of residual DCIS at surgery Table 15: Clinical trials of vaccine for the prevention of breast cancer recurrence (Modified from Chimino-Mathews et al., 2015 (328))

HER2 E75 peptide vaccine + GM-CSF adjuvant is targeting HER2+ patients. The interesting point is that this trial contained also “low to intermediate” HER2 expression holders who are usually excluded from the application of Trastuzumab (anti-HER2 antibody). The criteria of Trastuzumab application is HER2 3+ in IHC score while 0 or 1+ is classified as “negative” and 2+ as “equivocal” in current IHC interpretation. Interestingly, the patients with HER2-low-expressing tumors had the most robust immune responses and the largest decrease in mortality was seen in IHC 1+ patients (P = 0.05). Constant with this observation, the interim results of the therapeutic vaccine using the same vaccine reported promising results. HER2 peptide + GM-CSF combined with trastuzumab is showing significant results for TNBC patients who consists of 36% of this trial. The current trial with node-negative as inclusion criteria for TNBC patents is showing better results compared to previous trial with advanced (stage 4) BCs (Table 16). Whereas another trial with Trastuzumab alone targeting BC patients who have HER2 1+ and 2+ at IHC failed to show any efficacy. The investigator of HER2-vaccine trial considered the synergetic effect of Trastuzumab and cancer vaccine.

Table 16: Clinical trials of therapeutic cancer vaccine in breast cancer (Modified from Chimino-Mathews et al., 2015 (328)) Therapeutic cancer vaccine Sample Clinical Immune Vaccine Patient population Note size outcome monitoring HER2 specific T HER2 peptide + GM- Stage4 HER2+ BC Median PFS cells Eligibility HER2 2+- CSF adjuvant and in CR or SD on 22 17.7 months Decreased serum 3+ trastuzumab trastuzumab TFG-β Eligibility HER2 1+- node-positive 2+ (or node-negative if 36% was TNBC HER2 peptide + GM- negative for both ER and (Currently 275 HER2 specific T Study Start: CSF adjuvant and PR) BC with HER2 1+ and study is in (phase 2) cells January 2013 trastuzumab 2+ who are disease-free active) Estimated Study after standard of care Completion Date: therapy. June 2020 HER2+BC of stage 2 (with at least one positive Total 61 HER2 protein +AS15 DFS 73% at lymph node) or stage3 in (15 for HER2 specific Ab adjuvant and highest dose remission with no clinical highest and T cell response lapatinib group (500ug) evidence of metastatic dose) disease No difference STn-KLH+CY vs 1028 New vaccine Metastatic BC compared to KLH+CY (phase 3) specific Ab control Improved New TILs post- hTERT survival vaccine; Functional peptide+montanide+ Metastatic BC 19 associated hTERT-specific GM-CSF with hTERT peripheral CD8+ T immunity cells Possible clinical benefit CEA-MUC-1-TRICOM Metastatic BC 26 in patients Inconsistent poxvirus with minimal disease 1/3 of patients New p53-specific p53-loaded DCs Metastatic BC 6 with stable T-cell responses disease New p53-specific T-cell responses in 38% of vaccinated 42% with p53-loaded DCs Metastatic BC 26 patients, serum stable disease YKL-40 and IL-6 associated with response Increased HER-2 NR; factorial specific DTH and response antibody Allogeneic, GM-CSF- surface design responses; best secreting breast Metastatic BC 28 of various chemotherapy tumor cells chemotherapy doses CY 200 doses mg/m2+DOX 35 mg/m2

In contrary to the unexpected good response to HER2 peptide vaccine in the patients with low (1+ or 2+) HER2 expression, the patients with strong (3+) HER2 expression didn’t respond well. We can speculate the reason behind this is 1) In Her2 over amplifying patients may already have considerable amounts of anti-Her2 autoantibody, which may neutralize the peptide vaccine after injection before establishment of vaccine mediate immunity. 2) High degree of tolerance may have been already established in those patients that can’t be overcome by cancer vaccine.

In summary, it is important to optimize the targeted population based on deep understanding of pathological and immunological background of the breast cancers. Most of the immunotherapy trials for breast cancer so far have been done with metastatic breast cancers. Recent examples of good outcome in clinical activity, however, imply better outcome of cancer vaccines in early breast cancer setting or prevention of recurrence setting. Furthermore, combination of cancer vaccine with immune check point inhibitor are currently under investigation in many trials.

4. Chapter 4: Immunotherapy strategies using PSCs

4.1 Characteristics of Pluripotent stem cells (PSCs)

PSCs are undifferentiated cells with self-renewal proprietary, are able by definition to differentiate into the three germ layers of an embryo and are able to form all tissues and produce all of the diversed adult cells. The first PSCs, embryonic stem cells (ESCs), were derived from murine blastocysts (329) and in human from the inner cell mass of pre-implantation embryos that destined for destruction at in vitro fertilization clinics (330). Among the alternative methods which followed to generate PSCs, the most revolutionary method was the conversion of somatic cells isolated from adult cells (initially from murine and human fibroblast) into induced pluripotent stem cells (iPSCs) by genetic manipulation accomplished by the seminal work of Yamanaka, receiving Nobel prize in 2012 (331, 332). This somatic reprogrammed technology required transient ectopic expression of key transcription factors (TFs) involved in pluripotency (331).

PSCs are characterized by molecular (bivalent domain) and epigenetic (miRNA) landscape actively maintained by LIF-dependent (in mice) and bFGF- dependent (in human) signaling pathways that sustain self-renewal and repress differentiation lineage programs (333). Oct4 and Sox2 are both key TFs required for the acquisition and maintenance of pluripotency in mice and human PSCs. Oct4 and Sox2 are defined as ‘‘core pluripotency factors’’ with a mutual auto-loop regulation. Loss or repress of Oct4 and Sox2 expression elicits lineage commitment and specification into lineage differentiation. Nanog is another a key TF which stabilize Oct4 and Sox2 and is crucial for the maintenance of pluripotency (334).

Functional assessment of pluripotency includes different methodologies (1) in vitro differentiation in three germ layers; (2) teratoma formation; (3) chimera formation; (4) germline transmission; (5) tetraploid complementation; and (6) single-cell chimera formation.

4.2 iPS cells

4.2.1 Generation of iPS cells

iPSCs are derived from non-pluripotent cells, typically from fibroblast, blood cells or epithelial cells, after ectopic expression of pluripotency genes.

Takahashi and Yamanaka introduced 24 candidates, known as pluripotency- associated genes expressed in ESC, into adult cells and induced PSCs grown in permissive embryonic culture conditions. Among these 24 candidate, he identified a minimal of genes with four factors: Oct3/4, Sox2, Klf4, and c-Myc (331). The group of Thomson combined Oct3/4 and Sox2 with Lin28 and Nanog to generate human iPSCs (335).

Different strategies and methods were used to deliver the reprogramming factors into somatic cells. The advantage and disadvantages of each methods are summarized in Table 17.

4.2.2 Characterization of ESC and iPSC

Comparative transcriptome analyses using microarrays of ESCs and iPSCs have been compared in human and mice. Whole-genome expression profile of iPSCs showed very similar profile with ESC lines (336). Takahashi and colleagues compared the global gene expression profile of human iPSCs with human ESCs for 32,266 transcripts. 1,267 (approximately 4%) of the genes were detected with more than five-fold difference in up- or downregulation between them (337). Transcriptional profiles of human iPSC lines after the removal of Cre-recombinase excisable exogenous viral sequences (factor-free human iPSCs) were studied in comparison with those of human iPSCs before transgene excision (338). Transcriptomes of factor-free human iPSCs resembled more closely to those of human ESCs than those of parental human iPSCs with integrated viral sequences.

Globally, biological and functional properties of iPSCs are similar or indistinguishable to those of ESCs. However, some difference by careful analysis has been revealed, for example, miRNA expression level residually expressed from adult cells.

Table 17: Different methods of reprogramming factor delivery to generate iPSC (331, 339-350) Delivery method Efficiency Advantage Disadvantage Viral-based Methods Integrative virus Retrovirus High • Lasting expression of transgenes • Insertional mutagenesis • Residual expression of reprogramming factors • Titer limited tropism • Do not infect nondividing cells • Premature silencing • More immunogenic Lentivirus High • Lasting expression of transgenes • Insertional mutagenesis • Infect also nondividing cells • Residual expression of reprogramming factors Non-integrative virus Adenovirus Low • Nonintegrating • High spontaneous integration • Inefficient • High virus titer required • Transient expression Sendai virus Very high • Nonintegrating • Immunogenic • No premature silencing • Fusogenic • Broad tropism • Expensive • Virus curing process needed • Screening for integrations may be required Non-viral-based Methods piggyBac vector Low • Excisable integration • Insertional mutagenesis • No foot-print • Extra excision step • Less immunogenic • Inefficient reprogramming Recombinant proteins Very low • Absolute nongenomic integration • Inefficient Controllable factor stimulation • Multiple transductions • Affected by quality of recombinant proteins MicroRNAs Low • Nonintegrating • Inefficient • Easy synthesis • Transient expression • Multiple transfections

Expressions of pluripotent TFs can vary depending on the ESC culture conditions and it remains unclear that this is due to the culture conditions or truly the difference based on the genomic background of the cells. International Stem Cell Initiative has compared various hESC from different laboratories and described a short list of genes that showed a high correlation with that of Nanog. A pluripotency assay based on these gene expressions (Pluritest) is now used to qualify the pluripotency of PSCs.

iPSCs were also shown to be highly similar to ESCs for epigenetic status based on the presence of H3K4me3 and H3K27me3 in promoter regions of 16,500 genes (351). Quantitative differences in promoter methylation of pluripotency-related genes including Oct4, Sox2 and Nanog can be observed between iPSCs and ESCs (352), probably due to the persistence of original somatic cells-derived epigenetic “footprint” (353).

4.3 Immunotherapy strategies using PSCs

4.3.1 PSCs as a source of cancer immunotherapy

4.3.1.1 Generation of DCs from iPSCs

Generation of Dendritic cells (DC) was performed from ESC (354, 355). The described protocol from Senju et al., started by the hematopoietic differentiation method using OP9 cells as feeder cells. After a first step of hematopoietic commitment, cells were further cultured on OP9 and GM-CSF followed by a third period of feeder-free culture with TNF-α, IL-4 and CD40 or LPS. DCs were obtained from human iPS cells (356).

The iPS cell-derived DCs have similar characteristics of adult DCs including the capability of T-cell stimulation, processing and presenting antigens, and the capability of producing cytokines. While using the OP9 culture system is the main method for generating DCs from iPSC, xeno-free culture systems are now available to generate iPSC- DCs for clinical perspective (357). Choi et al. generated myelomonocytic cells including DCs from human iPSCs (358) and DCs from murine iPSCs (359). However, some technical issues remains to be solved before used for clinical applications : apoptosis and limitation of cell growth of iPSC-derived DCs, scale-up production, and problems of cost (360).

4.3.1.2 Generation of T cells from iPSCs

It has been shown that ESC can differentiate into lymphocyte lineage using in vitro OP9 coculture system expressing a Notch ligand, delta-like 1 (361). Lei et al. generated iPSCs from mouse embryonic fibroblasts through retroviral transfection of Oct3/4, Sox2, Klf4, and c-Myc (Yamanaka factors) and differentiate them into T-cell lineages by culturing it on OP9-DL1 cell system supplemented with Flt-3 ligand and IL-7. Adaptive transfer of these T cells to Rag1-deficient mice (mice lacking mature T-cells) enabled them to reconstitute T-cell pool by generation of CD4+ and CD8+ T lymphocytes in lymph nodes and spleen (362). Research team of the Unit has described a CD7+CD34+ Hematopoietic progenitors able to produce T lymphocytes (CD4+ and CD8+) using the technology of Fetal Thymic Orthoptic Culture system (363).

An interesting approach to generate antigen-specific CTLs is to obtain iPSC from blood T-cells and redifferentiate them into T-cells. CD3+ T-cells are selected from PBMCs and expanded by IL-2 and anti-CD3 antibody. T-cell-derived iPSCs (TiPS) were generated from activated T-cells after retroviral transduction of reprogramming factors (364). These T-iPSCs express the original TCR gene rearrangements, so they can be used as a source of T cells with the endogenous specific TCR. Antigen specific T-cells from an iPSC were also generated from CTL specific for particular epitope. CTLs were transduced with Sendai virus bearing Yamanaka factors (Klf4, Sox2, Oct4, c-Myc), miR-302 target sequence and SV40 (large T antigen). iPSCs derived CD8+T cells were able to retain their original antigen specificity, against MART-1 (melanoma epitope) (365) and pp65 antigen (cytomegalovirus)(366). Furthermore, stimulation of CTL-iPSC-CTL cells with their specific antigens led to IFNγ secretion and degranulation. CTL-iPSC-CTL cells have some advantages to parent CD8+ T-cells with elongated telomeres with a high proliferation potential and survival.

Additionally, some of them display central memory T-cell (Tcm) and stem-cell memory T-cells (Tscm) phenotypes which was associated with increasing expression of CCR7, CD27, and CD28 markers. Several lines of evidence show that Tcm and Tscm have superior antitumor immunity for Adoptive transfer cell-based immunotherapy because of their characters such as the resistance to apoptosis, potent response to homeostatic cytokines, self-renewal, and efficient generation of other T-cells’ population (360, 365-367). Combination of iPS generation technology with transduction of tumor antigen-specific TCRs or CARs showed successful generation of tumor-specific transgene T-cells (368). However, despite this proof of concept, major limitation resid to

a very small amount of IPSC-derived T cells and robust protocols of differentiation and expansion.

4.3.1.3 Generation of NK cells from iPS cells

In 2005, Woll and colleagues used a two-step process to differentiate human ESCs into NK cells in vitro. These cells had the ability to lyse human tumor cells deficient in MHC class I expression (369). The Research Team of the Unit has generated ESC-derived NK cells from murine stromal cells MS5 expressing HoxB4 to improve in vitro expansion of NK cells, with a high purity (>95%) of mature NK cells and the capacity to kill K562 cells. HOXB4 homeoprotein are known to promotes the expansion of human adult HSCs but also myeloid and lymphoid progenitors. (370).

Subsequently, NK cells were successfully differentiated from human iPS cells, using a similar two-step culture system, which Stromal cell-mediated differentiation of hematopoietic progenitors on M210-B4 cells followed by differentiation into NK cells on EL0801D2 stromal cells with the presence of NK itiniating cytokines (371), the cells obtained representing a pure population that did not require cell sorting or co-culture with xenogeneic stromal cells. Moreover, sufficient cytotoxic NK cells could be differentiated from 250,000 iPS cells to treat a single patient. iPSCs could provide a scalable source of NK in the future (372).

4.3.1.4 Generation of NKTs from iPSCs

NKT cells are characterized by the expression of an invariant TCR encoded by Vα24–Jα18 in humans and Vα14–Jα18 in mice recognize a CD1d1-restricted self- glycolipid antigen (373). These cells share the properties of both NK cells and T cells and are thought to play an important role in cancer immune surveillance (374). It has recently proven to derive NKT cells from iPSCs. NKT cells differentiated from mouse iPSCs were shown to secrete large quantities of IFNγ and actively suppress tumor growth in vivo (375). Yamada et al., has demonstrated successful regeneration of Vα24+NKT cells in vitro by two-step culture (reprogramming to iPSC and redifferentiation to iNKT cells) which allows sufficient expansion (376).

4.3.2 Pluripotent stem cell-based vaccines against cancer

Similarity in gene expressions between ESC and various cancers from transcriptome analysis, led to the hypothesis that ESCs could be used as cancer vaccine. While the first report using PSC-based vaccines (PSCV) as cancer vaccine in 2009, the anticancer activity with embryonic material has been documented since a century ago. In 1906, Georg Schöne observed that the prior injection of fetal tissue into mice led to rejection of transplanted tumors.

Stonehill and Bendich made the observation in 1970 that a rabbit antiserum raised against antigens of mouse embryos (and adsorbed against a mixture of adult murine tissues) cross-reacted with all of 72 tumors arising from 12 different tissues. Several independent groups also demonstrated the anti-tumoral effects of fetal materials in animal models afterwards (377).

In some cases, no anti-tumor effects were found when immunization was performed with older embryonic or neonatal cells. Indeed, this is one possible source of confusion in the earlier work in this field; it appeared that the age of the embryonic material used for vaccination was critical. In the case of mice (term pregnancy= 21 days), embryos of <12–14 days appeared most effective in causing tumor rejection.

In recent years, the experimental settings in this field now use ESCs or iPSCs as a source of embryonic TAAs instead of fetal tissue. The first report of PSC-based cancer vaccine was by Li et al. in 2009 which demonstrated the hESCs showed anti-tumor effect in mouse colon cancer model (378). They also showed that the immune memory lasts at least 4 weeks after the last vaccination. After this first report, the anti-tumor immunity of PSCVs were verified in lung and ovarian cancer animal models by several independent groups (379-382) (Table 18). Yaddanapudi et al., used PSCVs combined with GM-CSF- expressing-fibloblast (PSCV/GM-CSF) that also showed reduction of the tumor, longer survival and lower metastasis load in the vaccinated group compared to the control group. Very recently, Kooreman and colleagues demonstrated a protective anti-tumor effect in the murine tumor models of non-metastatic breast cancer, mesothelioma and melanoma using autologous iPSCs (383).

These studies demonstrated prophylactic anti-tumoral effects by performing tumor challenges to the immunized animals. In general, immune monitoring showed increases of CD4+, CD8+ T cells, pro-imflammatory cytokines such as IFNγ and IL-2 with a Th1 polarization and reduction of MDSCs and Tregs. CD8+ T cell depletion completely abrogated the protective effect of PSCV, verifying that CD8+ T cells are the principle

100

effectors (382). Induction of humoral immunity was also demonstrated with the observation of cross reacting antibodies against tumor cells and ESC. The rate of serum IgG binding to iPSC and tumor cells was significantly higher in the treated group than in control. Experiments in lung cancer model reported by Dong et al. (429) showed also therapeutic effect.

TCR sequencing revealed that the TCRs in the PSCV-treated group were more diverse among different mice whereas TCRs in the control group revealed an overlap in T cell clones that are commonly present in thymus and spleen. There was one TCR clone that was shared majority of the treated mice but was not present in any of the other groups; this clone was also extremely rare in naive mice. This result implies PSCV immunization expanded TCR repertoires and likely created TCRs that reacts specifically to the epitopes on PSCV.

One of the issues with other immunotherapy such as CAR therapy is organ toxicity from cytokine storms as mentioned earlier. It was shown that, systemic cytokine levels of PSCV vaccinated mice were low. In addition, tissue analysis of our mice at different time points after vaccination did not show any increases in immune cells within heart and kidney tissues compared to the control groups, nor were elevated levels of anti-nuclear antigen IgG seen in serum from PSCVs vaccinated mice. These evidences indicate a constant safety profile of PSCV strategy.

101

Table 18: Pluripotent stem cell- based vaccine studies in animal models

Li et al., 2009 Dong et al., 2010 Zhang et al., 2012 Yaddanapudi et al.,2012 Kooreman et al., 2018

Xenogeneic ESCs(H9) in Autologous ESC in C57 Xenogeneic ESC (H9) in Allogeneic ESC (D3) + GM- Autologous iPSC + CpG Balb/c murine colon cancer BL/6 lung cancer model ovarian rat cancer model CSF-expressing STO adjuvant in murine Vaccine murine embryonic melanoma (B16F0), hESC (CT26, CT2) fibroblasts: in C57BL/6 lung mesothelioma (AC29), hiPSC (TZ1) cancer model (LLC) breast cancer (DB7) model

Reduction of tumor in Reduction of tumor in both Reduction of tumor and Reduction of tumor and Reduction of tumor in Anti-tumor prophylactic setting prophylactic and metastasis in prophylactic prolongation of survival in prophylactic setting, effect hESC CT2 was superior therapeutic setting setting prophylactic setting Inhibition of recurrence in

than hiPS TZ1 preventing setting (B16F0)

Reduction of MDSC Increase of CD4+ and Increase of IFN-gamma, Increase of Increase of IFN gamma CD8+ T cells TNF-α, IL-2 producing effector/memory helper T, Immune Increase of IL-2 (in serum), CD8+ T cells IFN gamma monitoring IFN gamma (in serum) Decreased MDSCs Reduction of MDSC, Th17 Tregs

No significant autoimmunity No significant hematology No significant difference in by immunofluorescence for toxicity and side effect by serum anti-nuclear antinuclear antibodies testing of CBC, antibody titers Safety GOT, GPT, Cre No inflammatory observation in heart and kidney

Cross reactive antibody by Immunohistochemistry of Mice depleted of CD4+ Adoptive T cell transfer western blotting H9: nm23 (+++), p53 (++), were at least partially reduced tumor Other (sera for total cell lysate of c-Myc (++), HER-2 (+) protected TCR analysis detected remarks CT26 and undifferentiated Cross reactive antibody by CD8+ depletion completely specific rare TCR in the H9) western blotting (sera abrogated the protective vaccinated mice NuTu-19 vs H9) effect

102

RESULTS

Despite much progress of breast cancer biology, TNBC patients remain to have worse prognosis and an unmet medical need. It has been shown that TNBC possess CD44+CD24-/low CSCs population more frequently than other molecular subtypes. Therapeutic strategies remain a crucial challenge in order to target and eradicate specifically CSCs. Transcriptome meta-analysis revealed that advanced TNBCs showed embryonic stem cell-like gene signature.

The Research Unit, INSERM UMR 935 has a strong background on human and mice ESC and iPSC research in the field of oncology. Their main goals are to model the cancers and to establish immunotherapy strategies using PSCs and derivatives. Considering that TNBCs share some molecular genes with pluripotent stem cells, we explore the hypothesis; T cells can target breast CSC by using a pluripotent stem cell- based vaccine strategy.

Research objectives of my thesis was:

(1) To characterize TNBC 4T1-cancer stem cell compartment in vitro and in vivo.

(2) To established a syngeneic TNBC 4T1 immuno-competent mouse model.

(3) To validate the methods of immunological monitoring analysis

(4) To evaluate the anti-tumoral effect of immune response by a pluripotent stem cell- based vaccine strategy.

Before starting this work, previous studies used ESC to induce an immune response against different cancer type. Li et al. demonstrated for the first time in 2009 the anti- tumoral effect using xenogeneic hESC in murine colon cancer model, followed by Dong et al., in lung cancer, Zhang et al. in rat ovarian cancer and Yaddanapudi et al., in lung cancer.

We explore for the first time the immune response against cancer stem cell compartment and the metastatic lesions by using both ESC and iPSC in a breast TN aggressive cancer model.

103

EXPERIMENTAL DATA OF MY THESIS

Spheroid-forming culture has been widely used to isolate CSCs in various type of cancers. CSCs were obtained using free-floating spheroid culture in human glioblastoma (called neurosphere), in breast (called mammosphere), in ovary tumors (384) and in most epithelial cancers like gastric, lung, hepatocarcinoma prostate etc.

Transplantation of CSCs enriched from spheroid-forming culture methods into mice gives rise to more aggressive tumor with rapid metastasis compared to the tumor obtained using an equal number of cancer cells grown in adherent monolayer culture condition (385). The ability to form in vitro mammospheres is in fact dependent on the presence of the cells with extensive self-renewal capacity such as gland-reconstituting stem cells within the population (49, 386).

Thus, spheroid-forming culture is used as functional assay to induce and enrich CSC with elevated self-renewal property and with the capacity to seed the tumor and to drive metastasis.

The team of the research unit has previously showed that this spheroid culture condition is permissible to maintain a mesenchymal profile with a high expression of pluripotent TFs such as Oct4 Sox 2 and Nanog and stem cell markers such as CD133 in glioma stem cells (387, 388).

TFs Oct4, Sox2 and Nanog defined as ‘‘core factors of pluripotency” are crucial for the acquisition of pluripotent state (333). These pluripotent TFs play a relevant role in carcinogenesis. In breast cancer, downregulation of Nanog inhibit clonal expansion and tumor development of MCF7 breast cancer cell (389). Downregulation of Sox2 using si- RNA in MCF7 cells has been shown to inhibit cell proliferation and tumor growth (390). High levels of Nanog and Oct4 were associated with aggressive tumor biology and poor prognosis in patients with breast cancer (391). Their co-expression was associated with lymph-node metastasis, mesenchymal marker expression and increased invasiveness of CSCs. Nanog expression has been associated with high proliferation (47) and resistance to chemotherapeutic agents (392). Furthermore, patients with stronger expression of Nanog have reduced disease-free and OS rates (46).

For the project, it was important to first optimize the best in vitro culture conditions and to analyze the CSC before the development of the in vivo immunotherapy experimental CSC model.

104

A- In vitro characterization of TNBC CSCs from 4T1 cell line

We aimed to investigate the expression of pluripotent TFs and various markers related to CSC in 4T1 and other cancer cell lines by RT-PCR. Given the genomic plasticity in cancer cells in different conditions, we explored and characterized the “cancer stemness” profile in 2D (adherent) versus 3D (spheroid) culture conditions with various cytokines.

Material and Methods

Cancer cell lines

Cell line Type of cancer Description Human cell line MCF7 Breast cancer o Sourced from metastasis lesion, pleural effusion o Subtype is luminal A o ER and progesterone receptor expressed. MDA-MB-231 Breast cancer o Sourced from metastasis lesion, pleural effusion. o Subtype is basal, TNBC PC3 Prostate cancer o Sourced from metastasis legion of bone. o Grade IV adenocarcinoma. TT Thyroid cancer o Sourced from thyroid medullary carcinoma. Murine cell line 4T1 Breast cancer o Derived from BALB/c mouse. o TNBC o The tumor growth and metastatic spread of 4T1 cells in BALB/c mice very closely mimic human breast cancer. o This tumor is an animal model for stage IV human breast cancer. B16-F10 Melanoma o Derived from C57BL/6J LL/2 Lung cancer o Derived from C57 BL/6J. o Lewis lung carcinoma.

Culture conditions and Mammosphere conditions

4T1, B16-F10, LL/2, MCF 7, MDA-MB 231 and PC3 cancer cell lines were obtained from ATCC and cultured in Dulbecco’s modiﬁed Eagle's/F12 medium (DMEM) supplemented with 10% Bovine fetal serum, 1% penicillin and streptomycin (Invitrogen). Medium are changed 2-3 times per week. The passage of the cells was done approximately once or twice a week depending on growth speed of the cells by treatment with TrypLE (Gibco). Medium for spheroid culture was prepared as reported from mouse embryonic fibroblast–conditioned medium supplemented with 4 ng/mL of basic fibroblast growth factor. 5 x 104 cancer cells /well were inoculated to 2 mL of media in 6

105

well ultra-low attachment plate. 1 ml of media was added to each well every other day. Spheroid were trypsinized and passaged every 7-10 days.

Cell Culture was performed in normoxia or hypoxia (5% O2). The experiment included following conditions (Pre-treatment: 4 days, MF culture: 5 days).

#1: MF media at Normoxia (N) (MF N) #2: MF media at Normoxia (N) after pretreatment (ptt) with TNF+TGF (MFptt N) #3: MF media + TNF+TGF at Normoxia after pretreatment (ptt) with TNF+TGF (MFptt N + TNF+TGF) #4: MF media at Hypoxia (H) after pretreatment (ptt) with TNF+TGF (MFptt H)

Expression of pluripotent TFs by Reverse Transcription Polymerase Chain Reaction (RT-PCR)

Total RNA was extracted using the Trizol reagent (Invitrogen). The Optic Density (OD) was then measured to estimate the concentration of the extracted RNA. cDNA was then produced using Revert Aid First stand cDNA synthesis kit (Fermentas) starting from 2 µg of RNA. PCR was performed using Fast start Taq Polymerase (Roche). All PCR experiments started with the activation of the Taq polymerase at 95°C for 4 minutes. The amplification cycles were repeated 40 times as following: 45 seconds at 94°C, 30 seconds at defined annealing temperature and concluded with 1 minutes at 72°C. After the last amplification cycle, the final elongation step which is performed at 72°C for 7 minutes. The PCR products were then analyzed by electrophoresis. Primer sequences are shown in Annex 5.

ALDH Staining

The Aldefluor kit (StemCell Technologies) was used to profile cells with ALDH enzymatic activity. The experiments were undertaken according to the manufacturer’s instruction. Briefly, 1×106 cells were incubated in Aldefluor buffer containing the ALDH protein substrate (BAAA, BODIPY-amino acetaldehyde, 1 mmol/L) for 30 minutes at 37°C. Cells that could catalyze BAAA to its fluorescent product (BAA) were considered ALDH+ Sorting gates for FACS were drawn relative to cell baseline fluorescence, which was determined by the addition of the ALDH-specific inhibitor diethylamino benzaldehyde (DEAB) during the incubation. 7-AAD (eBioscience) was used to exclude dead cells.

106

Flow cytometry analysis

Cells were treated with Accutase (Stemcell technologies) or TrypLE (Gibco) to detach from the plate, then washed and resuspended in Phosphate Buffered Saline 1X (PBS) (Life Technologies) and 5% FBS (Gibco) and counted. A number of 105 cells was reacted with antibodies. Isotype control or non-staining control were used as control. The cells were placed at 4℃ for 30 minutes. After wash with PBS, the cells were resuspended in 200 µL of DPBS. FACS was performed using MACS Quant Analyzer. The details of the antibodies used in the experiments are listed in Annex 6.

RESULTS

1) Pluripotent TFs, Stemness and EMT markers in different culture conditions.

Breast cancer cell lines from mouse (4T1) and human (MCF7 and MDA-MB-231) showed various expression patterns of Oct4, Sox2, Nanog and c-Myc (Table 19).

Cell line Oct4 Sox2 Nanog c-Myc Mouse 4T1 - + + + B16 - - + + LL/2 - - - + Human MCF7 + + + - MDA-MB-231 + + + + PC3 - + - - TT + + + +

Table 19: Results of RT-PCR performed on different cancer cell lines demonstrating the expression of pluripotent markers. (-) Absence of amplification (+) Amplification observed.

LIN28 which is also related to the pluripotency was shown to be associated with high grade malignancies in human and experimental models. LIN28 induce the repression of the let-7 miRNAs family (58). Thus LIN28 could be contributed to malignancy in

107

which levels of all miRNAs in the let-7 family are coordinately repressed, therefore promoting oncogenesis by derepressing target genes involved in tumorigenesis (393).

We show that 4T1 express positive stem cell-like EMT markers and resistant drug resistance markers. CD146 up-regulate mesenchymal markers and significantly promote cell migration and invasion by EMT (394).

PTFs Stemness, EMT and invasion Drug resistance

LIN28 Klf4 CD44 CD133 CD146 CD166 Sca1 CD105 ABCG2 ABCB5

4T1 + - + + + + - + + +

(-) Absence of amplification (+) Amplification.

2) Plasticity of 4T1 regarding the culture conditions in adherent culture and spheroid culture methods

In TME, cytokines play a critical role to modulate the behavior and propriety of tumor cells. TGF-β is a strong inducer of EMT. Recently, prolonged exposure of tumor cells to TGF-β and TNF-α induces EMT and induced a stable breast CSC phenotype which is shown by self-renewal capacity and increased tumorigenicity (395). Hypoxia and IL-6 signaling are in addition two conditions that play important roles in cancer progression and stemness (396).

After 4 days culture with TNF-α and TGF-β as well as the culture with TNF-α, TGF-β, IL-6, GM-CSF and IFNγ, 4T1 cells showed more mesenchymal like morphology (Figure 29).

108

No treatment

TNF-α+TGF-β

TNF-α+TGF-β+IL-6+GM-CSF+IFN-γ

Figure 29: Morphology change after 4 days culture with cytokines. Detachment of the individual cells from the colony was observed in cytokine treated cells whereas no treated cells showed very epithelial colonies.

109

2-1) Mammosphere culture conditions

Representative 4T1mammosphere after 3 days and 5 days of culture are illustrated in figure below:

Day 3 Day5

Figure 30: The representative photos of the mammosphere culture. The photo shows the cultures at day 3 and day5 after the seeding of the cells.

2-2) Mammosphere forming efficiency (MFE)

In each MF culture conditions (N= Normoxia, H= Hypoxia, and cytokines (TNF, TGF)) we estimated two parameters: The MFE calculated by the formula as follows. MFE= mammosphere count at Day 5 in one well / number of inoculated cells x 100 (Figure 31a), The number of large sphere count >200 μm (Figure 31b) Among the different MF culture, the conditions MF ptt +TNF+TGF at normoxia showed both highest MFE and large sphere (>200um) counts.

110

(a) (b)

Figure 31: Mammosphere forming efficiency (MFE) and large sphere count (a) MFE, (b) Large sphere count (>200um)

Representative mammosphere in two conditions: MF (N) and MFptt + TNF+TGF (N)

MF (N) MF ptt +TNF+TGF (N)

Figure 32: Representative photo of MF culture of MF (N) and MFptt + TNF+TGF (N)

111

2-3) Stem cell markers in 4T1 mammosphere We performed immunostaining of CSC markers represented by CD44+CD24low/- by flow cytometry and ALDH expression by Aldefluor test in 4T1 mammosphere from the different conditions. Only the condition of MFptt+TNF+TGF showed 7.6% CD44+CD24low/- whereas other condition didn’t increase CD44+CD24low/- population.

(a) (b)

Figure 33: CD44+CD24-/low population of 4T1 cells in MF and MF ptt+TNF+TGF cultures Comparison between the different culture conditions (b) Dot plot images of CD44CD24 staining

112

Regarding ALDH+ expression, ALDH activity was higher in mammosphere than in adherent 4T1 cells and pre-treatment with TNF+TGF (N and H) showed a highest rate.

(a) (b)

Figure 34: ALDH+ of 4T1 cells in MF and MF ptt + TNF + TGF cultures

113

2-4) MHC Class I expression

MHC I expression on tumor surface is often downregulated which limits effective cytotoxic CD8+ T cell reaction. We evaluated the expression level of MHC I in 4T1 Mammosphere compared to adherent 4T1 cancer cells. As shown in Figure 35, MHC I expression was diminished in all 4T1 mammosphere in MF culture compared to bulk 4T1 in adherent culture.

Figure 35: MHC I expression in bulk 4T1 and 4T1 mammosphere

114

B- Establishment of TNBC model

Adequate experimental model is important to evaluate the new medical approaches. The model should closely mimic the pathology of the human disease and should have reproducibility to assure the robust results and analysis. Inadequate preclinical setting may lead to the unexpected failures or surprises in the following clinical trial in human afterwards.

The orthotopic implantation of tumor cells into syngeneic immunocompetent mice is considered as one of the best breast cancer models. It reproduces the entire pathogenic cycle, including formation of the primary tumor, involvement of regional lymph nodes, and development of metastases (397), including metastasis in bones (398). A representative of these models is a rapidly growing and highly metastatic murine adenocarcinoma 4T1 derived from a spontaneous mammary tumor in a BALB/c mouse (399). After injected into the fat pad of a mammary gland, TNBC 4T1 cells metastasize to lung, liver, bone, and brain. Subclones of 4T1 cells stably transfected with reporter genes have been developed and applied for optical monitoring of primary tumors, metastases and circulating tumor cells (400-402)

In order to evaluate the vaccine experiments in accurate manner by monitor in vivo the live tumor cell distribution in the body including micro metastasis, we optimize the 4T1 cell model by transfecting green fluorescent protein (GFP)-Luc reporter genes.

Material and Methods

Retroviral Vector Production and Transduction

We used a retrovirus expressing the luciferase gene (pMEGIX-Luc vector) carrying an expression cassette encoding the firefly luciferase and GFP genes. The Luc-GFP virus were produced by transfection of 293 EBNA cells by lipofection using Lipofectamine 2000 (Thermo fisher). Briefly, subconfluent 293 EBNA cells were co-transfected with pMEGIX, gag-pol and VSV with lipofectamine with opti-MEM. The next day, the culture media was changed and the recombinant retrovirus was harvested every day for 3 days starting from 24 hours after the change of media. The intensity of GFP expression was measured by flow cytometer MACS Quant Analyzer.

115

Bioluminescent imaging

In vitro measurements were performed in cell culture fluid of Luc-expressing 4T1 cells grown in a 96-well cell culture plate. Growth medium was removed and wells were filled with 100 μl luciferin solution (150 μg/mL; Promega) in the complete DMEM containing Penicillin-Streptomycin (Sigma-Aldrich) and 10% FBS (Gibco). The bioluminescent intensity was measured on black well plate (Perkin Elmer) by IVIS Spectrum CT. In vivo measurements were performed by injecting luciferin solution (150 mg/kg) to Balb/c mice intra-peritoneally 10 minutes before imaging. At 3 minutes after luciferin injection, mice were anesthetized with isoflurane.

RESULTS

1) Tumorigenicity of 4T1 cells in Balb/c mice

First, we have checked the tumorigenesis property of 4T1 using limited numbers (n=2/group) of mice. Two mice per group have received 3 doses of 4T1 cells: 103, 104 and 105 cells injected subcutaneously. All 3 doses were shown to be tumorigenic in the immunocompetent mice (Fig 7).

In addition, one mouse of highest dose group was dead on day 20 after injection due to organ dysfunction at metastatic legions and one another mouse of the same group and two mice having received 104 cells were euthanized at day 21 and day 25 after the injection each considering clinical condition severity due to metastasis (Figure 36). The experiment was completed on day 33 after the injection.

116

(a) (b)

Injected Death or euthanized date cell count 105 cells #1: Dead (day 20), #2: Euthanized (day 21) 104 cells #3 and 4: Euthanized (day 25) 103 cells #5 and 6: Euthanized (day 33)

Figure 36: Tumor volume in 4T1 tumorigenesis study (a) Tumor volume. Data shown are average + SE, which are shown until 18 days post tumor injection due to drop out of mouse. (b) Dead or euthanized mice log. # shows mouse ID and the date shows days post injection.

Highest dose group showed the largest tumor volume until day 21. At day 25, 104 cells mice group was shown larger volume of tumor compared to mice having received 103 cells group. Histopathology analysis performed for the primary tumor tissue and metastatic legions showed highly metastatic and invasive property of 4T1 cancer cell line (Figure 37).

Splenomegaly Tumor Tumor x100

Liver x100 Lung x100 Lymph node x100

Figure 37: Pathological observations in primary tumors and spleen

117

(compared to normal size of spleen in photo) and histopathological observations of mouse received 4T1 cells. Tumors were observed in primary tumor as well as metastatic legion of spleen, ganglion, liver and lung.

Based on this result, we decided to adopt 5 x 104 cells per mouse for challenge the mice and test the immune therapy approach.

2) Establishment of 4T1-GFP-Luc C3 model

2-1) Transfection of pMEGIX to 4T1 cells

GFP expression in transduced 4T1 cells was confirmed by fluorescent microscopy and its intensity measured by flow cytometry (Figure 38). At day 2, 36% of 4T1 were GFP positive and were sorted. The sorted cells were then processed for cloning.

(a)

Figure 38: GFP intensity measurement by FACS

2-2) Cloning of 4T1-GFP-Luc cells

The sorted GFP positive cells were seeded by limiting dilution to 98 well plate. All of the wells were carefully checked for the number of the cells. When there was only single cell in a well and formed one colony, it was termed “clone” (Figure 39). When the well was not completely verified to have only one cell (or, presence of two cells), it was termed “subclone”. We have obtained 3 clones termed, C1, C2 and C3. and 12 subclones from SC1 to SC12.

118

Figure 39: Cloning in 98 well plate: One 4T1-GFP-Luc cell in a well under (a) microscope and (b) fluorescent microscope. (c, d) The cell formed a small colony 4 days after seeding.

2-3) Selection of the clone in vitro and in vivo

All of the clone C1, C2 and C3 showed highly positive rate of more than 90% with only one peak, indicating successful cloning of the cells carrying GFP-Luc. In subclones, some of them showed high GFP positive rate as more than 90% whereas part of the subclones like SC2, SC7, SC8 and SC10 had two population (peaks) including GFP negative population (Figure 40). The GFP positive rate and GFP median value of each clone are shown in Figure 41.

119

Figure 40: GFP expressions of 4T1-GFP-Luc clones (above) and subclones (below). C1, C2, C3, SC9 and SC12 showed highly GFP positive homogenous population whereas SC2 contained GFP negative population and SC10 contained low positive different population too.

120

Figure 41: GFP positive rate and median value of each clones.

We chose 7 clones of C1, C2, C3, SC5, SC6, SC9 and SC11 for further analysis to select the best clone. Those clones were analyzed for luciferase expressions by in vitro bioluminescent imaging. The clones were performed serial dilution in 98 well plate in luciferin solution and measured for emission of bioluminescent signals by IVIS imaging. p/s per 1 cell and limit of detection were calculated and compared as well (Figure 41). The bioluminescent imaging of 4T1-GFP-Luc C3 shows highly sensitive and bright (Figure 41).

121

(a) (b)

Figure 42: Analysis of the clones by bioluminescent imaging by IVIS. (a) Comparison of the emitted photon by each clone, (b) p/s per 1 cell (c) Bioluminescent image of 4T1-GFP-Luc C3 cells in 98 well black plate, indicating highly sensitive detection ability

At the end of selection of the clones, we analyzed the selected clones by testing in vivo bioluminescent imaging, tumorigenicity, metastatic property in Balb/c mice. 5 x 104 cells of C1, C3, SC5, SC6, SC9 and GFP positive bulk (not sorted) were implanted to each 2 female Balb/c mice. Tumor volume and clinical body condition were monitored weekly. In vivo bioluminescent imaging was performed every 6-10 days. At 47 days, all mice were sacrificed. At sacrifice, the tumor weight was measured and the extent of metastasis was analyzed by direct bioluminescent imaging of lung, liver, spleen, bones (femurs) and kidney/adrenals.

Among the injected clones, C3 and bulk showed tumor growth curve relatively similar to that of parental cells (Figure 43a). Other clones didn’t grow like parental cells. C1 showed much less tumor weight compared to C3. One mouse from SC6 and two mice from SC9 didn’t develop any tumor. SC5 showed only very small tumor.

In vivo bioluminescent images of C3 showed constant growth of the living tumor cells. At the late phase of the experiment, necrosis was observed in the center of the tumor. At 47 days, lung metastasis was observed at dorsal position (Figure 43c).

122

(a) (b)

(c)

Figure 43: Tumor volume, tumor weight and bioluminescent images of 4T1-GFP-Luc clones (a) Tumor volume, (b) Tumor weight at day 47 post tumor challenge, (c) Sequential bioluminescent images of 4T1-GFP-Luc C3 injected mice. Image at Day 0 was taken just after the tumor injection to verify the successful implantation.

123

Extent of metastasis was analyzed. C1 or C3 showed highly metastatic property and metastasis was seen in almost all of the analyzed organs, whereas SC6 and bulk didn’t metastasize to the bones (Figure 44). SC5 didn’t show metastasis at all.

Given the high tumorigenicity, metastasis and high sensitivity in bioluminescence imaging of 4T1-GFP-Luc C3 like parental 4T1 cells, we choose 4T1-GFP-Luc C3 cells to use in the vaccination study.

(a)

Figure 44: Metastasis to the organs analyzed by bioluminescence detection of 4T1-GFP- Luc clones (a) Extent of metastasis per group (n=2) (b) Bioluminescent images of 4T1-GFP-Luc C3 injected mice (ID 9 and ID 10) showing broad spread of the living 4T1-GFP-Luc C3 in the various organs

124

C- Generation of transgene-free iPS cells from Balb/c mouse

To obtain our therapeutic cell source, we generated a murine iPSC line from Balb/c mouse (the same genetical background as 4T1) that are used for the following vaccination study.

Material and Methods

Isolation and culture of fibroblast cells from Balb/c mouse

A part of tail was collected from one Balb/c female mouse. During this operation, the mouse was deeply anesthetized with high dose isoflurane at 4% with oxygen, then euthanized. The collected ear and tail were disinfected with alcohol rapidly and rinsed with PBS then cut to small pieces by scalpel. The pieces of tissues were then enzymatically digested with 5 mL dissociation medium (40 mL: 40mg of BSA fraction V, 20 mg of collagenase, and 30 mg trypsin (1.2 mL of 2.5% trypsin)) for 1 hour at 37℃ with constant pipetting. Pipetting and digesting with 18G syringe. Centrifuge at 1200 rpm for 5 minutes. Add 2 mL each well of 6 well plate. The culture was passaged and expanded before confluent. Mouse fibroblast cells were cultured in DMEM glutamax including 10% FBS (biowest), 1% penicillin streptomycin, 1% sodium pyruvate, 0.1% 2- Mercaptoethanol (MEF media).

Reprogramming mouse fibroblast

The reprogramming mouse fibroblasts to pluripotent stem cells was done utilizing STEMCCA Cre-Excisable Constitutive Polycistronic (OKSM) Lentivirus Reprogramming Kit (Millipore). 1 x 105 cells of fibroblasts were plated to gelatin-coated 6 well plate in MEF media. At day1, the media was changed with fresh 2 mL MEF media and 1 uL polybrene and 6.6 uL of EF1α-STEMCCA-LoxP (OKSM) Lentivirus were added. At day 2, MEF media was exchanged with 3 ml of ES media (DMEM + glutamax including 15% serum (Eurobio), 1% penicillin streptomycin, 0.1% 2-Mercaptoethanol, 1% Sodium pyruvate and 1000 unit/ml of mLIF). After that, mES media was changed every other day. The clonal passage was done starting day 9.

125

Clonal expansion of mouse iPS colonies

The plate with grown iPS colonies was washed with PBS. DMEM glutamax without FBS was added to the plate and the iPS colonies were picked by the pipette cone under microscope and dissociated with trypsin in 98-well plate, then the number of the cells were counted. 500 cells were added onto the MEF (mitomycin treated) coated 6 cm culture plate in DMEM glutamax including FBS (eurobio), 1% penicillin streptomycin, 1% sodium pyruvate, 250 uL 2-Mercaptoethanol 0.1% and LIF 1000 unit/ml (mES media). Starting 3 days after the first passage, the colony was appearing. The colonies that have the typical domical shape were picked, dissociated with trypsin and replated onto MEF coated 24-well plate for cloning and expansion. When the colonies became sufficient for passage, the plate were washed with PBS and trypsin was added. The dissociated cells were replated onto MEF-coated 6-well plate in mESC media. The passage was repeated as the same manner to 6cm plate for expansion, then the cells were frozen in FBS containing DMSO.

Excision of viral transgenes

Adenovirus expressing Cre recombinase and GFP (Vector Biolabs) was diluted in mESC media in Eppendorf tube. The iPSCs were dissociated into single cells with trypsin and put into the prepared Eppendorf tube with the adenovirus then incubated at 37℃ for 6 hours. After the incubation, the cells were plated onto the MEF-coated 6cm plate for cloning. The cloning and expansion were done like the same manner as those after reprogramming. After the expansion, the cells were analyzed for the deletion of the transgene by PCR analysis. The expanded cells were frozen in FBS containing DMSO.

Extraction of genomic DNA

The individual colonies were picked into 96-well plates, split into duplicate sets of plates and one set were frozen with freezing medium (80% FBS and 20% DMSO) for later recovery. When the majority of the colonies in the wells became confluent, the wells were rinsed twice with PBS and 50 uL of lysis buffer (10 mM Tris, pH 7.5, 25 mM EDTA, 100 mM NaCl, 0.5% SDS. Just before use, proteinase K (Sigma) was added to a final concentration of 0.1 mg/mL) was added to the each well. After the incubation of the sealed plate overnight, add 100 ul per well of mixture if NaCl and ethanol (150 ul of 5M NaCl mixed with 10 mL of cold absolute ethanol). The plates were agitated with pipetting and

126

kept at room temperature for 2 hours. The plates were inverted to decant the supernatant and rinsed with 150 uL of 70% ethanol 3 times and decanted each time on paper towel. The plates were dried and then added with 150 uL of 70% ethanol. The plates were sealed with Parafilm and kept at 4℃.

PCR analysis of genomic DNA

Genomic DNA was isolated from the mouse iPSC colonies that have undergone the excision process. As a control, also isolate genomic DNA from mouse iPS colonies that have not undergone the excision process. WPRE which is specific to the viral genome, and housekeeping gene GAPDH were tested using the primers detailed in Annex 5. PCR cycling was done according to the manufacturer’s protocol as follows; Initial denaturation at 94°C for 2 minutes, 30 cycles of: 95°C 30 seconds, 65°C 45 seconds and 72°C 45 seconds. Final extension at 72°C for 10 minutes. The PCR products were loaded on 1.5% agarose gel containing ethidium bromide for electrophoresis.

Immunostaining of iPSCs on the plate

The cells were fixed for 10 minutes at room temperature with 4% paraformaldehyde and then permeabilized for 10 minutes at room temperature with 0.25% Triton X-100. Incubations with primary and secondary antibodies were performed in phosphate- buffered saline (PBS), 0.1% Tween-20, and 10% fetal calf serum. The following antibodies were used: anti-Oct-3/4 antibody (1/300, BD Biosciences), anti-NANOG antibody (1/100, Cell Signaling). The cells were incubated at room temperature for 90 minutes with primary antibodies and for 1 hour with secondary antibodies. Nuclei were counterstained with DAPI (4′,6-diamidino-2-phenylindole). Fluorescent images were obtained using fluorescence microscope equipped with a digital camera.

Teratoma assay

The cells were harvested by collagenase IV treatment, collected into tubes, and centrifuged, and the pellets were suspended in Matrigel dissolved in PBS. 106 cells were injected subcutaneously to dorsal flank of a NOD/SCID mouse. Nine weeks after injection, tumors were dissected and fixed with PBS containing 4% paraformaldehyde. Paraffin-embedded tissue was sliced and stained with hematoxylin and eosin.

127

RESULTS

1) Reprogramming of somatic fibroblast to pluripotent stem cell

Primary mouse fibroblast was obtained from female Balb/c mouse. The fibroblast was typical mesenchymal shape (Figure 45). EF1α-STEMCCA-LoxP (OKSM) Lentivirus including reprogramming factors Oct4, Klf4, Sox2 and c-Myc were infected with the subconfluent fibroblasts. Approximately 1 week after the infection, domal shaped iPSC colonies appeared on the plate (Figure 45). We have selected only the isolated colonies which were not combined with other colony and the colonies with domal shape for further expansion. We selected and picked 12 colonies and the colonies were termed, B1, B2, …B12.

Figure 45: Reprogramming of primary fibroblasts obtained from Balb/c mouse. Representative photos from Day 0 to Day 9. Typical domical colony of miPSCs was observed at Day 9.

128

2) Deletion of transgene

We chose the miPSC clone B9 for proceeding with the deletion of the transgene. This process was done using transient infection of Adenovirus expressing Cre recombinase and GFP. After virus infection, cloning and expansion of the cells were performed and 96 clones, termed B9D1, B9D2,…B9D96, were obtained (Figure 46). The genomic DNAs of all of the clones were tested to verify the absence of WPRE sequence which is specific to the transgene. At the end, 12 clones were verified negative for WPRE by repeated 2-3 rounds of PCR tests for each clone to be sure, indicating successful deletion of the transgenes.

Figure 46: iPSC colonies observed after the infection of Adenovirus expressing Cre recombinase and GFP.

129

(a)

(b)

Figure 47: PCR results after the process of transgene deletion (a) First round of PCR testing of all of Adenovirus applied clones. B9C15, B9C16, B9C19, B9C89 and B8C90 showed presence of amplification. NDC: Non-deleted colony (positive control) (b) Final verification of absence of WPRE by PCR analysis.

130

3) Pluripotency assay

miPSC clones B9D7, B9D29 and B9D38 were tested for expressions of pluripotent TFs. All of the three clones were confirmed to express Oct4 and Nanog (Figure 48). B9D7 was then injected to NGS mouse for teratoma assay. The teratoma of B9D7 showed differentiation of 3 germ layers, ectoderm, mesoderm and endoderm, indicating pluripotency of B9D7 (Figure 48).

(a)

(b)

Figure 48: Characterization of iPSC clones. (a) Immunostaining of miPSC clones B9D7, B9D29 and B9D38 for Oct4, Nanog. (b) Teratoma derived from B9D7 shows 3 germ layers

131

D- Article: Pluripotent Stem Cell-Based Cancer Vaccines Targeting Tumor Microenvironment and Metastatic Spread

In this study, anti-tumor and anti-metastasis effects of PSC-based vaccines were evaluated.

The similarity in gene expression between PSCs and advanced TNBC indicated the presence of TAAs between those cells including unknown antigens.

Despite many efforts using specific peptides or proteins as cancer vaccine, the clinical trials haven’t shown constant positive results. The cause of insufficient immune activation against the cancer cells can be due to 1) intratumoral heterogenicity of tumor cells, 2) continuous evolution of the cancer cells which lead to the loss of the specific antigens, 3) emergence of new antigens due to the genetic and epigenetic changes in the clones, and 4) development of tolerance afterwards. In this regard, originality of using PSCs is, once immunized, the ability to give immunological memory against a variety of potential TAAs and can prepare for the emergence of CSCs with PSC gene signature.

We show here that PSC-based vaccines strikingly emitted anti-tumor and anti- metastasis effects with successful induction of Th1 type immune reaction with long and robust immune memory.

132

[Submitted article]

Pluripotent Stem Cell-Based Cancer Vaccines Targeting Tumor Microenvironment and Metastatic Spread

Masae Kishi Heront1*, Afag Asgarova1*, Diana Chaker1, Christophe Desterke1,2, Mathis Soubeyrand1, Theodoros Latsis1, Jérôme Artus1,2, Marie-Ghislaine de Goër de Herve3, Ali G Turhan1,2,4*, Annelise Bennaceur-Griscelli1,2,4*, Frank Griscelli1,2,5*

1 Institut National de la Santé et de la Recherche Médicale (INSERM) U935-Human Embryonic Stem Cell Core Facility, ESTeam Paris Sud-INGESTEM, Villejuif, France 2 Université Paris-Sud/ Paris Saclay, Faculté de Médecine, Kremlin Bicêtre, France 3 Institut National de la Santé et de la Recherche Médicale (INSERM) U1184, Immunology of Viral Infection and Auto-immune Diseases, Normal and Pathological T Memory, Faculté de Médecine Paris-Sud, Kremlin-Bicêtre, France 4 Service d’Hématologie Biologique, Hôpital Universitaire Paris Sud, (AP-HP), Kremlin Bicêtre, France 5 Université Paris Descartes, Sorbonne Paris Cité, Faculté des Sciences Pharmaceutiques et Biologiques, Paris, France,

* : Equal contribution

Running title: Pluripotent Stem Cells as a Breast Cancer Vaccine

Keywords: Breast, cancer, vaccine, PSCs

To whom correspondence should be addressed: Professor Frank Griscelli INSERM 935, University Paris Sud/ Paris Saclay. 7 rue Guy Moquet 94802 Villejuif, France Tel : 33 1 42 11 51 93

133

Graphical abstract

134

ABSTRACT

Extensive recent data have now established the heterogeneity of cancer which is maintained by the activity of a rare population of self-renewing “cancer stem cells” (CSCs) difficult to eradicate due to their resistance to conventional therapies. Pluripotent stem cells (PSCs) and CSCs share several antigenic determinants and due to the rarity of latter, PSCs could be an attractive way to elicit anti-tumor responses against CSCs. We show here that PSCs immunologically primed with the use of HDAC inhibitor valproic acid, are able to elicit major anti-tumor responses in a highly aggressive breast cancer model. This active immunotherapy strategy is efficient to prevent the tumor establishment and to efficiently target CSCs-derived breast cancer cells by inducing a major modification of the tumor microenvironment. This anti-tumoral effect was associated with reduction of Tregs and MDSC populations and increase of cytotoxic CD8+ T cells within the tumor and the spleen. In addition, the anti-tumor response was associated with a drastic reduction of the metastatic dissemination and to improve the survival rate in the highly aggressive 4T1 breast cancer model. These results show for the first time the possibility of using PSC as an anticancer vaccine in combination with HDACi in the allogeneic setting and establish the basis for future clinically applicable active and universal immunotherapy strategies.

135

INTRODUCTION During the last decade, the concept tumor heterogeneity has been extensively explored in solid tumors, leading to the identification of a rare population of cells designed under the terms of “cancer stem cells” (CSC) or “tumor initiating cells” (TIC) in several types of cancers (Reya et al. 2001; Clarke et al. 2006; Al-Hajj et al. 2003). These cells represent within the bulk of the tumor, a minor population with defined cellular and molecular characteristics which have been analyzed based on their ability to reinitiate tumor in immunodeficient mice suggesting their self-renewal capacity (Chen et al. 2012; Driessens et al. 2012) and, on their specific transcriptome (Puram et al. 2017; Galardi et al. 2016). Current evidence indicates, indeed, that in addition to c-MYC, which is a well-established oncogene, some of the key regulators of ESCs such OCT4, SOX2, and NANOG are also expressed in CSC (Ben-Porath et al. 2008). These factors are part of a highly integrated network including the c-MYC and polycomb networks that use the epigenetic machinery to remodel the chromatin through histone modifications and DNA methylations. Their ability to induce major epigenetic modifications has been recently demonstrated by the groundbreaking experiments of Yamanaka’s and Thomson’s teams showing the induction of embryonic features in somatic cells by overexpression of limited set of pluripotency genes, OCT4, SOX2, C-MYC, KLF4 NANOG, LIN28) (Takahashi et al. 2006; Yu et al. 2007). The induced pluripotent stem cells (iPSCs) obtained by the reprogramming technology are closely linked to ESC by the expression of the same auto regulatory circuitries (Robinton et al. 2012). Genomic characteristics of IPSCs has shown multiple genetic abnormalities that could be due either to the genomic status of initial somatic target cell (Hussein et al. 2011; Schlaeger et al. 2015) or to their secondary appearance during their expansion in vitro (Hussein et al. 2011). From this regard, IPCs present several genomic characteristics identified in CSC in which, random mutations in driver genes or outside of functional genome have been shown to be present in addition to the expression of pluripotency genes (Ben-Porath et al. 2008). CSC present a down regulation of their antigen processing capacities leading to low expression of the MHC-I molecules making them difficult to be detected by the host immune system (Prager et al 2019). Furthermore, this effect is enhanced by their location in a highly immunosuppressive tumor microenvironment (Prager et al 2019). Thus, their escape from the surveillance of an efficient immune system is inevitable. The embryonic stem cell (ES)-like gene expression programs have been identified in several cancers with a “stemness” profile that are related to mesenchymal traits on carcinoma cells with Epithelial-Mesenchymal Transition “EMT” markers and low levels of MHC-I expression. Tumor cells undergoing EMT become CSCs with the capacity to migrate very early, to form metastasis, and to persist in a dormant stage for long periods of time. Those cancers correlate with aggressive poorly differentiated tumors histology and invasive tumors, and with very adverse outcomes

136

(Ben-Porath et al. 2008; Glinsky et al. 2008; Schoenhals et al. 2009). These observations have led to the idea of using ESCs or iPSCs as source of tumor associated antigens (TAA) with the goal to promote an anti-tumor response. The concept of using embryonic cells as a cancer vaccine have been reported in a preventive context using animal models of transplantable colon, lung, ovarian and breast cancers (Li et al. 2009; Yaddanapudi et al. 2012; Zhang et al. 2013). More recently autologous anti-tumor vaccines were developed by using iPSCs in combination to TRL9 as adjuvant in a prophylactic setting, in a non-metastatic syngeneic murine breast cancer as well as in mesothelioma, and melanoma models (Kooreman et al. 2018). Nevertheless, all these reports have not evaluated the anti-metastatic potential of these anti-tumor vaccines and have not evaluated whether these products have the capacity to target CSCs and to modify their microenvironment. This question remains crucial since anti-tumor vaccines targeting only the bulk and transit amplifying cells without eradication of CSCs are doomed to be unsuccessful. Here, we show, that a treatment combining modified pluripotent stem cells (ESC or IPSC) with histone deacetylase inhibitors (HDACi) prevent the establishment of CSC-enriched tumors and the development of lung metastases in an aggressive triple negative breast cancer model.

137

RESULTS

Poorly differentiated murine 4T1 breast tumors display an ESC-like expression signature in vivo We first asked whether embryonic stem cell (ESC)-associated genes were enriched in 4T1 murine breast cancer cell line. For this purpose, we generated a compendium of data permitting to compare different gene sets in different contexts including adherent 4T1 cells cultured in vitro and 4T1 injected in vivo into the mammary fat pads of BALB/c mice. RNAs from the murine ESCs (D3) and that from micro-dissected mammary glands were used as positive and negative controls, respectively. Transcriptome meta-analysis by ANOVA allowed us to identify 1304 different genes between the four experimental groups (data not shown). Nested DEG analysis performed between implanted 4T1 and micro-dissected mammary glands allowed to identify 85 genes that are up regulated in 4T1 implanted in vivo. Unsupervised classification on corrected matrix with all the samples identified two major clusters including a cluster with non-tumoral mammary gland sample and 4T1 harvest in 2D culture in vitro and a second cluster with 4T1 cells implanted in BALB/c mice and mESCs (Figure 1A). These results suggested that in the 4T1 cells, the transcriptional activation of embryonic genes shared with ES cells occurred only in the in vivo context, after establishment of a tumor microenvironment. Functional enrichment of these stem-like cells are mainly linked to important functionalities implicated in breast tumor development such as in: aromatic metabolism, cell adhesion, cell growth, cell division and cell cycle (Figure 1B). In order to confirm the stemness signature of implanted 4T1 cells, we quantified by FACS analysis the expression of breast cancer stem cell markers CD44 and CD24 (Al-Hajj et al. 2003) on 4T1 cells recovered in vitro and in vivo 12 and 28 days after having injected them into mammary fat pads. A CD44+/CD24-/low population with a frequency up to 32% (Figure 1C) could be identified in 4T1 cells in vivo, suggesting strongly the emergence of a high proportion of cancer stem cell-like cells in this metastatic breast cancer model, The demonstration of this phenotype suggested that the antigens shared with ESC could be used as tumor vaccines in this context.

ESC vaccination Elicits Anti-tumor Activity Associated with a Tumor Cell-specific CD8- dependent Cytotoxic T-lymphocyte Response. To study whether ESCs could stimulate the immune system to recognize embryonic antigens shared with tumor cells and to confer tumor protection, we performed experiments using the previously described 4T1 model known to metastasize spontaneously and rapidly to lungs after implantation (Pulaski et al. 2001). Since ESCs could trigger different immunological pathways

138

with different impacts of immune response to them, we used as tumor vaccines, xenogeneic, allogeneic pluripotent stem cells as compared to syngeneic tumor cells (Figure 2A) including lethally irradiated human ESCs (H9) murine ESCs (D3) and the 4T1 cell line. Preliminary experiments have shown that both human and murine ESCs used, expressed pluripotency markers (Figure S1A, Figure S1B) and were able to form teratomas in vivo (Figure S1C). The protocol included the immunization of naïve immune competent BALB/c mice twice (1 week apart) with irradiated cells followed by a challenge with 5x104 4T1 cells transplanted at mammary fat pad (Figure 2A). The animals were examined weekly for their vital characteristics and the tumor size. After 28 days, in contrast to control non-immunized mice, all mice primed with hESCs, mESCs and 4T1 cells showed significant (p<0.0001) reduction of their tumor volumes (Figure 2B, 2C, 2D). The strongest reduction of tumor volume was achieved by using xenogeneic materials, (Figure 2E) hESCs exhibiting 73% of tumor reduction compared to the use of allogenic mESCs (52% of reduction) and 4T1 cells (56% of reduction). At this time point, all mice were sacrificed to analyze the phenotype of immune cells present in tumor and spleen. We found a significant negative correlation (p=0.05) between tumor volume and the amount of cytotoxic of CD8+ T cells in spleen (Figure 2F) as well as, the amount of CD4+ T cell (p=0.0039) infiltrates in the tumor (Figures 2G). Moreover, a significant positive correlation between tumor volume and frequency of Tregs in CD4+ T cells in tumor was observed (p=0.049) (Figure 2H). These results suggested that the anti-tumor effect observed by vaccination induced a cytotoxic T cell response as well as the reduction of the immune-repressive Treg cells in the tumor microenvironment. We then asked whether we could modulate the immune response effect of the vaccine by epigenetic modulators such as Histone deacetylase inhibitors (HDCAi), being clinically investigated.

Valproic acid (VPA) Modulate Immune-related genes in 4T1 cells To determine a potential immunogenic effect of VPA on 4T1 cells, we first performed a transcriptome analysis on 4T1 treated in vitro with 0.5 mM of VPA for 10 days compared to non- VPA treated cells. These analyses allowed the identification of 117 immune-related genes implicated in the TNF-α signaling and in the response to IFN-α and IFN-γ (Figure 3A). These results were confirmed by gene set enrichment analysis (GSEA) showing a significant enrichment of three immune gene sets (Figure 3B). In addition, the use of SAM algorithm allowed to discriminate 44 immune-related genes between the VPA-treated samples and their control counterparts (Figure 3C). These were validated by principal component analysis (Figure 3D, p- value=3.3x10-4). Among these 44 immune-related genes, CD74, CCL2 and TNFRSF9 were found to be overexpressed with a fold change greater than 2 (Figure 3E). More importantly, 4T1 cells treated with VPA expressed an increased level of MHC I in a dose-

139

dependent manner (Figure S2A) highlighting that VPA might enhance anti-tumor immune response by improving antigen recognition by T cells. This higher RNA expression was confirmed by flow cytometry. Measurement of Relative Fluorescence Intensity (RFI) revealed a 2.1-and 2.7- fold increase of MHC I expression in respectively bulk 4T1 cells and mammosphere-derived 4T1 cells after treatment with 2 mM of VPA (Figure 2SB). These results prompted us to explore the efficacy of combinatorial treatment approach, using pluripotent cell based immune therapy and HDACi, such as VPA.

VPA Potentiates ESC-based Vaccination Elicit a Decrease of Tumor and Lung Metastasis by Modulating the Immune Tumor Microenvironment. To potential synergistic interaction of VPA with hESC-based vaccine was evaluated in 4T1-GFP- Luc model allowing the quantification of viable tumor cells at metastatic sites by the use of a sensitive luminescent reporter system and IVIS imaging. VPA was added in the drinking water at the dose 4mg/ml at the day of tumor challenge (Figure 3SA). After optimizing of the experimental schedule (Figure S3A), vaccination protocol included 4 groups of mice: immunized mice with hESCs alone, VPA alone or hESC+VPA and a control group treated with PBS (Figure 2A). Tumor size were monitored in each group. Lungs were dissected in order to quantify metastasis by measuring the level of total flux emitted by using bioluminescent imaging. Forty for days after tumor implantation, the highest reduction of tumor growth was observed in mice treated with VPA+hESC with 59% of statistically significant reduction as compared to control (p<0.0001) (Figures S3C and S4) with a small residual tumor (Figure 4B). Mice treated by VPA or primed with hESCs showed a partial tumor regression (24.5 and 25% respectively) (Figures 4A, and Figure S4). hESC+VPA induced a highest significant reduction of pulmonary metastatic spread (more than 12-fold) as compared in mice of control group (Figure 4 C) and of hESC group (Figure S3 D). In these experiments, the tumor volume was found to be associated with total flux in lung and this was highly significant (p=0062) (Figure 4D). Immune contents in primary tumor and spleens were measured by an extensive immunophenotyping analysis. As shown in Figure 4, mice primed with hESC+VPA showed an increase frequency of CD4+ and CD8+ T cells in the tumor mass (Figure 4E) that was highly significant in the spleen (p=0.0179 and p=0.0186 respectively for CD4+T and CD8+T cell) (Figure 4F). Interestingly, programmed cell death 1 (PD-1) expression on both CD4+ and CD8+ T cells from spleen decreased significantly (p=0.012 and p=0.0171 respectively) in hESC+VPA treated mice (Figure 4G and 4H). These results show that hESC+VPA combinatory regimen potentiate the immune system by enhancing T cell infiltration with probably less T-cell exhaustion. In the same experiments, we also have compared the effect of VPA with an classical adjuvant such as CpG ODN, well known to stimulate the Th1 based immune pathway by supporting CD8+

140

T cell responses (Bode et al., 2011). These experiments did not show any significant effect of the CpG ODNs and hESC combination on the tumor volume and metastasis, as compared to hESC alone and control groups (Figures S3B S3C and S3D).

ESC+VPA Combo Therapy Induces an Anti-tumor and Anti-metastatic Potential by Tackling CSC Compartment. The tumor protective effect of hESC + VPA combo therapy was next evaluated in 4T1-derived mammosphere model (MS). For this purpose, we first produced MS known to contain a high subset of CSC-like cells in two permissive 3D culture conditions with or without addition of TGF- β/TNF-α in the medium. Both protocols were effective to produce MS after 5 days of culture (Figure S5A) expressing Aldehyde Dehydrogenase 1 (ALDH1) (Figure S5B). The amount of ALDH1 activity evaluated by flow cytometry was 1.7 higher in MS without TGF-β/TNF-α as compared to 2D-adherent 4T1 cells. Addition of TGF-β/TNF-α during MS development enhance the ALDH1 activity at 3 fold compared to 2D-adherent 4T1 cells (Figure S5C). Both cytokines induced EMT with generation of CSCs phenotype with the capacity of higher colony formation ability and higher MS formation efficacy (MFE) (Figure S5D). associated with the highest expression of embryonic transcription factors NANOG and SOX2 (Figure S5E). MS and TGF- β/TNF-α exhibited a higher tumorigenicity in vivo suggestion a higher rate of proliferation (Figure S6A). Tumor protective effect of the combination hESC and VPA was explored after implantation in fad pad, MS from 4T1-GFP-Luc cells obtained in both conditions. After 27 days post-challenge, mice treated with hESC+VPA and mESC+VPA showed significant (p<0.0001 and p=0.0008 respectively) delays in tumor growth and reduction of tumor size (Figure 5A and 5B). Xenogeneic hESCs were found to be more effective compared to allogeneic mESCs to prevent the development of 4T1-derived MS. Indeed, tumors size analyzed by bioluminescent imaging in hESC+VPA treated groups appeared much smaller compared to controls (Figure 5C). Importantly, mice treated with hESC+VPA exhibited significantly (p<0.05) reduced lung metastases in comparison with controls (Figure 5D, 5E). A drastic reduction of tumor development was similarly observed by immunization with hESCs+VPA or mESC+VPA in engrafted mice with aggressive MS TGF-β+TNF-α (Figure S6B and S6C). A reduction of cell dissemination and migration into the lung was also observed in this model (Figure S6D).

Immunological monitoring showed a significant (p<0.05) increase of intra-tumoral CD4+ and CD8+ T cells frequency (Figure 5F, 5G) and a significant decrease (p<0.05) of Gr1+ CD11b+ MDSCs frequency in primary tumor from immunized mice with hESC+VPA compared to controls group (Figure 5H and 5I).

141

These results confirm that hESC+VPA allowed the reprograming of TME by modifying the balance of an immune-repressive state to higher immune-active state. The tumor specificity of these cytotoxic T cells was confirmed by increased secretion of interferon-gamma (IFN- γ) by TIL isolated from vaccinated mice in response to 4T1 tumor cells. There was a significant negative correlation (p=0.0083) between the secretion of IFN-γ by the total population of TIL and tumor weight (Figure 5J). Indeed, mean induced IFN-γ values were significantly higher in hESC+VPA as compared to control group and to the mice primed with mESCs (Figure 5K).

Comparative Anti-tumor Effects of Syngeneic or Allogeneic iPSCs Combined with VPA. To assess the effectiveness of this combo therapy by using other source of PSC, we compared the anti-tumor activity elicited by allogenic (C57BL/6) or syngeneic (BALB/c ) iPSC. For this purpose, we have generated murine iPSC from BALB/c and C57BL/6 fibroblasts with lentiviral or retroviral vectors respectively, expressing OCT4, KLF4, SOX2, c-MYC. Both miPSCs exhibited nearly identical gene expression and surface markers profiles with expression of SSEA1 (Figure S7A), OCT4, NANOG (Figure S7B). They were able to generate teratomas in vivo (Figure S7C) confirming their pluripotency. We performed tumor vaccination experiments following the same protocol by using both murine IPSC in presence of VPA or not (Figure S3A). At day+20, mice treated with syngeneic BALB/c- iPSCs had significantly (p=0.0058) reduced the tumor sizes by 35%, whereas mice vaccinated with C57BL/6 iPSC have shown a lower reduction of the tumor by 21% (Figure 6A and 6B). Interestingly, the adjunction of VPA highly increased the anti-tumoral response in both vaccination strategy allowing a drastic inhibition (61%) of tumors burdens with allogeneic miPSC and 48% with syngeneic mIPSC. (Figure 6A and 6B). Inhibition of tumor burden was significantly higher (p<0.0001) by allogeneic miPSCs compared to syngeneic PSCs (p=0.0018). Mice were not sacrificed and kept for a long-term survival study. A significant improvement in median survival rate was observed after vaccination with both types of miPSCs, compared to control groups (Figure 6C and 6D). Importantly, VPA had a significant benefit in median survival rate using both vaccine products (Figure 6E and 6F).

We then performed an additional experiment consisting to vaccine 8 mice with two injections of 2x106 C57BL/6-derived iPSCs and to compare the tumor burden to 8 non-vaccinated mice. All mice were challenged with 5x104 4T1 cancer cells using the same previous protocol (Figure 3SA). Tumor size was monitored every two days until day+20 (Figure 6G). After 6 days, all mice from both groups presented similar tumors that have strongly progressed in 7 out 8 mice from the control group. In contrast, all treated mice showed significant reduction in tumor volumes that have regressed in 4 out 8 C57BL/6-derived iPSCs+VPA-treated mice, one tumor having

142

completely regressed and three having partially regressed with a tumor volume of less than 35 mm3.

Generation of an Effective Memory Immune Response after Vaccination with miPSCs To investigate the potential T cell memory effects induced by miPSCs vaccination, 2x106 irradiated BALB/c-derived miPSCs were inoculated 6 times over a period of 180 or 270 days. During this successive vaccination period, all mice remained healthy throughout the study. After tumor challenge with 5x104 4T1-GFP-Luc cells, VPA was added in the drinking water at the dose 4mg/ml and tumor growth was monitored during 26 - 28 days (Figure 7A). In two independent experiments, the vaccination reduced the tumor burden by 64% (Figure 7B) and 46% (Figure 7C) resulting to a significant tumor growth inhibition as compared to the control mice after 26 (1968+96 versus 695+102 mm3,, p=0.0059) and 28 (600+40 versus 320+23 mm3, p<0.0001) days, respectively.

The analysis of metastasis dissemination revealed a major reduction of lung metastases after 28 days (Figure 7D) with a significant correlation between tumor burden and metastatic spread (Figure 7E). This protection was correlated with a significant increase of frequency of CD4, CD8 T cells (Figure S8A) in spleen and with a significant decrease of the frequency of PD1 in CD4 and CD8 T cells and of the frequency of T Reg population (Figure S8B) indicating that long-term immune memory has been well established by the injections of miPSC vaccine leading to the activation of an efficient anti-tumor immunity. To evaluate further the tumor protective effect induced by iPSC vaccination, we performed an experiment over a period of 28 days consisting to challenge the mice after 270 days with 5x105 4T1 cells or MS-derived 4T1 (Figure 7A). Mice immunized with miPSCs were able to reject significantly (p<0.0001) and efficiently MS-derived 4T1 with a reduction of the tumor burden by 83% (Figure 7F), with highly reduced tumor sizes as compared to controls (Figure 7G). This long-term immune memory also significantly protected the mice from developing lung metastases (Figure 7H) and a significant correlation between tumor weight and the metastatic dissemination was observed (Figure 7I). To evaluate the effect of the vaccine and to explore the mechanisms of the long-term immune protection observed, we performed a transcriptome analysis on tumors from non-treated mice and from mice primed with 6 boosts of miPSCs and VPA. Differential Expressed Gene (DEG) analysis was evaluated between the 2 conditions and 206 genes were found to present significant differential expression (Figure 8A). We found 98 genes that were up regulated and implicated in immune response to cytokines and lymphocyte chemotaxis (Figure 8B). Immune module infiltration showed a significant enrichment of monocyte derived dendritic cell (Normalized Enriched Score (NES =1.66, p-value<0.001), of T cells (NES=1.54, p-value<0.001) and of B cells

143

(NES=1.33, p-value<0.001) (Figure 8C). Molecular immune network highlighted the major implication of B cells and T cells by the importance of molecules implicated in these respective infiltrations (Figure 8D). Interestingly, a significant (p-value=5.1787803E-5) up regulation of chemokine (C-X-C motif) ligand 13 was observed (by 8.8-fold) in post-vaccination tumors. This result was confirmed by RT-qPCR showing a strong up-regulation of CXCL13, CXCL9 and CXCL10 (Figure S9). These results suggested strongly that the vaccine acts essentially by recruitments of B and T lymphocytes to the tumor sites in the vaccinated animals by CXCL9, CXCL10 and CXCL13 with an active immune action.

144

MATERIALS AND METHODS

Pluripotent stem cell lines and breast cancer cell lines hESC line (H9) was obtained from the University of Wisconsin (Thomson et al. 1998) and used under the agreements # RE07-008R provided by the French Biomedical Agency. They were maintained on mitomycin-treated mouse embryonic fibroblasts (MEF) in Dulbecco’s modiﬁed Eagle's/F12 medium (DMEM) supplemented with 20% Knockout serum replacement, 1% penicillin and streptomycin (Invitrogen), 1% non-essential amino acids, 1mM 2-mercaptoethanol (Sigma) and 12.5µg/mL of basic fibroblast growth factor (bFGF). Murine iPSC cell line was generated by transduction of fibroblasts isolated from primary cells of BALB/c with Cre- Excisable Constitutive Polycistronic Lentivirus expressing OCT4, SOX2, CMYC, KLF4 (EF1α- STEMCCA-LoxP backbone from Millipore). Mouse iPS cell line from C57BL/6 was derived from fibroblasts by using retroviral vectors expressing OCT4, SOX2, CMYC, KLF4 transgenes (ALSTEM Richmond, CA). Murine ESCs (D3) and murine iPSC cell line were maintained on mitomycin-treated mouse embryonic fibroblasts (MEF) in DMEM glutamax (Gibco) containing 15% fetal bovine serum (Eurobio), 1% penicillin and streptomycin (Invitrogen), 1mM 2- mercaptoethanol (Sigma) and LIF 1000 unit/mL. Prior to vaccine preparation, hESCs were cultured on Geltrex (Thermo fisher) in Essential 8 media (Thermo fisher) and D3 cells and murine iPSC were cultured on gelatin (Sigma) in the same media as indicated above. For the teratoma assays iPSCs were resuspended in Matrigel (BD Biosciences) and injected to the flank of immunodeficient mice (NOD-SCID) and sacrificed for histological analysis. All PSCs and 4T1 cells were irradiated with a lethal dose prior to injection. The breast cancer line 4T1 was obtained from ATCC (CRL-2539) and grown in DMEM, 10% FBS and 1% penicillin and streptomycin under normal culture conditions. For in vitro induction of mammospheres (MS) the 4T1 cells were seeded in low attachment plates at a density of 100.000 cells per well in 6-well plates with the cocktail of MEF conditioned medium (3/4 MEF conditioned medium + 1/4 mES medium + 4ng/ml bFGF) with or without adding of cytokines TNF-α (20ng/ml, Cell Signaling Technology) and TGF-β1 (10ng/ml, Cell Signaling Technology). 4T1 cell line was transduced using retroviral vector pMEGIX encoding the green fluorescent protein (GFP) gene and the luciferase gene. Stable clones were isolated and selected by GFP expression by flow cytometry.

Transcriptome meta-analysis of 4T1 cells In order to evaluate the stem cell signature of 4T1 murine breast cancer cell line, we compared 4T1 cells transcriptome experiments in different contexts to that of murine embryonic stem cell line (D3 from GEO dataset GSE51782) and to that of micro-dissected mammary glands samples

145

were analyzed (from GEO dataset GSE14202) (Padovani et al. 2009). Microarray samples of 4T1 generated from in vitro culture were analyzed from GEO data accession number: GSE73296 and microarray samples of 4T1 taken in the context of murine transplantation were analyzed from GEO data accession GSE69006. Normalized matrix respectively annotated matrix were merged via their gene symbol annotation to be process for a batch correction with Stanford algorithm in R (Tibshirani et al. 2002). One way analysis of variance (ANOVA) with 500 permutations was performed between the four experimental groups on the corrected matrix. Nested differential expressed genes (DEG) analysis was performed between 4T1-transplanted and mammary gland sample conditions with Significance Analysis for Microarray (SAM) with a False Discovery Rate (FDR) Threshold under 5 percent (Tusher et al. 2001). Subsequently, functional enrichment was performed on Gene Ontology Database with Go-elite standalone application (Zambon et al. 2012). Heatmap on DEG was performed with made4 Bioconductor R-package (Culhane et al. 2005, 4) in R version 3.4.1.

Transcriptome analysis on 4T1 treated with valproic acid. Total RNA was extracted from 4T1 cells cultured with and without VPA at the dose of 0.5mM following the instructions of the manufacturer (TRIzol, Life Technologies). Concentration of the total RNA was measured on Nanodrop and quality of the extracted nucleic acid was assessed on Bio-analyzer 2100 (Agilent technologies, CA). Microarray probes were synthetized by following one cycle of amplified RNA during which molecules were labeled in Affymetrix microarray station (Affymetrix, CA). Labeled microarray probes were hybridized on Mouse ClariomS (mm10) microarray (Thermo Fisher Scientific): CEL files of microarray scanned on Affymetrix platform were normalized by RMA method in Affymetrix Expression Console software (Affymetrix, CA). Gene set enrichment analysis was performed with online java module of GSEA software version 3.0. Network enriched gene set was performed with Cytoscape software version 3.6.0. Bioinformatics analysis was performed in R software environment version 3.4.1 with R- package made4 to performed expression heatmap on Euclidean distances and Ward method and with FactoMineR R-package to performed unsupervised principal component analysis. Genes with significantly differential expression were selected with Significance Analysis for Microarray (SAM) algorithm by adjusting the false discovery rate threshold under 5 percent.

Animal model Wild-type female 10 weeks old BALB/c mice were purchased from Janvier Laboratory and maintained by the animal core facility using standard guidelines. The animal experiments were carried out using a protocol approved by the Animal Care Committee of the Val de Marne. Irradiated ESCs or iPSCs were suspended in 100 μl of PBS and implanted subcutaneously in the

146

right flank of anesthetized BALB/c mice using 2% isoflurane in 100% oxygen. In vivo tumor studies were performed using 5 to 10 BALB/c mice in each group otherwise defined elsewhere. Mice were immunized subcutaneously with 2x106 irradiated cells twice at a 1-week interval. 5x104 4T1GFP-Luc cancer cell lines were resuspended in 100 μL PBS and injected at the 4th mammary fat-pad using tuberculin syringe. For H9+CpG vaccine, the irradiated H9 cells were suspended in 100uL PBS containing 50ug of CpG (Invivogen). After tumor implantation, oral treatment with VPA was started by adding in the drinking water VPA at the dose 4mg/ml in the vaccinated mice group. VPA (Sigma) was dissolved in drinking water at concentration of 0.4% w/v and administered to mice using feeding bottles ad libitum, a dosage leading to approximate concentrations of 0.4 mM in the plasma, as previously reported (Shabbeer et al., 2007). Tumor growth was measured using a caliper. In vivo bioluminescent imaging was performed using IVIS Spectrum (Perkin Elmer) and images were analyzed and quantified with Living Image software (Perkin Elmer). For direct imaging of lungs at sacrifice, lungs were isolated and incubated with 150ug/mL luciferin in 12-well plates and imaged. After sacrifice, tumors were dissociated using Mouse Tumor dissociation kit (Miltenyi) without enzyme R (Tumor dissociation kit, Miltenyi Biotec) and Dissociation machine (GentleMACS dissociator, Miltenyi Biotec). Spleens were dissociated using cell strainer (Fisher) followed by RBC removal using RBC Lysis Buffer (eBioscience). After dissociation, the cell suspension was washed with PBS and used for subsequent analysis. To study immunological memory induction ability of the vaccine product, BALB/c mice were immunized subcutaneously with 2x106 irradiated miPSCs in 100µl PBS solution 6 times with 3 day intervals in the presence of control group receiving PBS. At Month +6, mice from both groups were injected with 2.5x104 4T1GFP-Luc cancer cell lines at the 4th mammary fat-pad. Drinking water treatment with VPA (Sigma, at the dose 4mg/ml) was also included in the vaccinated mice group after tumor implantation.

Staining of immune cells and tumor cells for FACS analysis and ELISA assay Cells isolated from spleen and tumor were resuspended in PBS (Gibco) containing 1% FBS. Cell surface marker staining were done with the antibodies of CD44, CD24, CD45, CD8a, CD25, CD279 (PD-1), MHC class I (eBioscience) and CD11b, Gr-1, CD3, CD4, CD279 (PD-1), CD45, CD44, CD24 (Miltenyi Biotec). 7-aminoactinomycin D (7-AAD, eBioscience) and zombie violet (BioLegend) were used for exclusion of dead cells. FACS analysis was done using MACS quant (Miltenyi Biotec). Immune stimulation was performed using 50ng/ml PMA (Sigma) and 500ng/ml Ionomycine b (Sigma) in RPMI 1640 media (Gibco) containing 10% bovine fetal serum (Gibco) and 1% penicillin and streptomycin. IFN- was quantified using IFN-  kit (eBioscience) according to the manufacturer’s protocol.

147

Aldefluor assay The Aldefluor kit (Stem Cell Technologies) was used to profile 4T1 cells with ALDH activity as previously described (Pearce et al. 2005). Cells were incubated in Aldefluor assay buffer containing the ALDH substrate, BODIPY-aminoacetaldehyde at 37°C. The enzymatic activity of ALDH was blocked by a specific inhibitor, DEAB. The gates for FACS analysis were drawn relative to baseline fluorescence, which was determined by DEAB-treated samples. FACS analysis was done using MACS quant (Miltenyi Biotec).

Cytotoxicity assays 4T1 target cells were incubated with CFSE (eBioscience) dissolved with PBS at final concentration of 2uM for 9 min. Positive selection was performed to isolate CD8+ T cells from primary tumors and spleens using CD8a microbeads (Miltenyi Biotec), LS column and MACS separator (Miltenyi Biotec) according to the manufacture’s protocol. 10,000 CFSE labeled 4T1 cells and 10,000 CD8+ sorted cells from primary tumor or spleen in RMPI, 10% FBS and 1% penicillin and streptomycin (Invitrogen) were put in the wells of 98 well plates then incubated at 37°C overnight. Next day, the incubated cells were recovered and stained with 7-AAD for viability followed by flow cytometry analysis to evaluate the fraction of live CSFE positive cells. Cytotoxicity rate (%) was calculated by the formula T/C ratio = % target cells / % control cells (none immunized cells), survival rate (%) by the formula = {T/C ratio of sample}/{T/C ratio of none treated BALB/c mouse} and cytotoxicity rate (%) by the formula= 100 – survival rate.

Transcriptome analysis on 4T1-derived tumors from mice vaccinated with miPSCs and VPA and RT-qPCR Total RNA was extracted from 4T1 tumors from mice that were vaccinated with miPSCs and VPA and from 4T1 tumors from control mice according the instructions of the manufacturer (TRIzol, Life Technologies). Concentration of the total RNA was measured on Nanodrop technology and quality of the extracted nucleic acid was assessed on Bio-analyzer 2100 (Agilent technologies, CA). Microarray probes was synthetized by following one cycle of amplified RNA during which molecules were labeled in Affymetrix microarray station (Affymetrix, CA). Labeled microarray probes were hybridized on MouseClariomS (mm10) microarray (Thermo Fisher Scientific): CEL files of microarray scanned on Affymetrix platform were normalized by RMA method in Affymetrix Expression Console software (Affymetrix, CA) (Irizarry et al., 2003). Gene modules of specific immune cell expression profile were built as defined by Lyons YA et al. for a cell specific identification in the context of immune cell profiling in cancer (Lyons et al., 2017). With these specific immune modules, gene set enrichment analysis was performed with online java

148

module of GSEA software version 3.0 (Subramanian et al., 2005). Network enriched gene set was performed with Cytoscape software version 3.6.0 (Cline et al., 2007). Differential expressed genes were identified with ranking product analysis. Functional analysis performed on up regulated genes in vaccine condition was done with Gene ontology biological process database and DAVID application from NIH website (Huang et al., 2009). Bioinformatic analysis was performed in R software environment version 3.4.1 with R-package made4 to performed expression heatmap on Euclidean distances and Ward method (Culhane et al., 2005). For RT-qPCR total RNA was converted to cDNA using MultiScribe Reverse Transcriptase (Applied Biosystems, #4319983) and qPCR was performed in duplicate using the Strategene Mx3005P Real-Time PCR system (Agilent Technologies) and FastStart Universal SYBR Green Master Mix (Rox) (Roche, #04913914001), according to manufacturer’s instructions. Relative gene expression of CXCL9 (Fw: CCATGAAGTCCGCTGTTCTT, Rv: TGAGGGATTTGTAGTGGATCG), CXCL10 (Fw: ATCAGCACCATGAACCCAAG, Rv: TTCCCTATGGCCCTCATTCT) and CXCL13 (Fw: ATGAGGCTCAGCACAGCA, Rv: ATGGGCTTCCAGAATACCG) was normalized to that of glyceraldehyde-3- phosphate dehydrogenase (Fw: AAGGAGTAAGAAACCCTGGACCAC, Rv: GAAATTGTGAGGGAGATGCTCAGT).

Quantification and Statistical Analyses All values were expressed as mean ± s.e. as indicated. Intergroup differences were appropriately assessed by either unpaired two-tailed Student’s t test or one-way/two-way analysis of variance (ANOVA) using PRISM GraphPad software or, Microsoft office excel software. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

149

DISCUSSION Although the concept of the origin of some cancers from embryonic tissues was reported as early as 19th century (Review Brewer BG Exp Mol Path), the experimental data testing the idea of using embryonic tissues as cancer vaccines have been produced almost 100 years later using immunization of mice with fetal material (Review Brewer BG Exp Mol Path) or antisera developed against 9-day old mouse embryos (Stonehill et al. 1970) showing rejection of chemically induced tumors in these animals. With the availability of murine and human ESC, the ideas of using PSC as cancer vaccines is reemerged during the last decade. Several groups have suggested to use PSCs for priming the immune system in targeting different types of cancer (Li et al., 2009; Yaddanapudi et al., 2012, Kooreman et al. 2018;). These experiments were based on the demonstration that cancer cells and embryonic tissues share a number of cellular and molecular properties and an important number of TAA (de Almeida et al., 2014; Zhao et al., 2011). The major discovery of the possibility of reprogramming somatic cells to pluripotency rendered possible to test the possibility of using iPSC as source of TAA. This approach was tested in an autologous setting in mice (Kooreman al 2018). These experiments, combined with the previous data showing the used of human ESC as tumor antigens (Li et al. 2009; Dong et al. 2010) led to envision the possibility of using PSC in clinical immunotherapy settings. To date different ESCs and iPSCs vaccine candidates were evaluated but only based on their ability to induce tumor reduction in mice without evalution of anti-metastatic potential of these vaccine strategies. Moreover, previously reported work did not study the potential of these vaccine to target CSCs, at the origin of resistance, relapses and metastatic spead.

In this study, we evaluated for the first time the anti-tumoral potential of PSCs combined with the HDACi, Valproic Acid (VPA), for its multiple immune stimulatory functions and its capacity to modify the tumor microenvironment. We show that combined HDACi with IPSC/ESC based vaccine identifies a novel adjuvant therapeutic strategy to prevent risk of relapse and metastasis in TNBC. In particular, we show that VPA potentiates PSC vaccine and immune response against tumor bulk and CSCs compartment, leading to the prevention of metastatic spread. VPA increases significantly the expression of MHC1 on 4T1 cells and on 4T1-derived MS, as well as, MHC2 (CD74), chemokine (CCL2), and TNFRSF9, a TNF-receptor superfamily known to contribute to development of T cells and to regulate CD28 co-stimulation to promote Th1 cell responses. VPA was also shown to promote the production in vivo of different chemokines such as CXCL 9, 10 and 13, turning “cold” to “hot” tumor by significantly recruiting T, B and Dendritic Cells into the tumor. In addition, it modifies the immunosuppressive microenvironment with reduction of T reg and MDSC in the primary tumor.

150

We evaluated several PSCs vaccine candidates in combination with VPA on an aggressive, poorly differentiated, triple negative breast cancer model (TNBC) 4T1 cell line deriving from a spontaneously arising BALB/c mammary tumor (Tao et al. 2008; Aslakson et al. 1992). In this model, as few as 5x104 tumor cells given into the mammary fat pad of the mice lead to a rapid local growth of the tumor and distant metastases occurring via a hematogenous route mainly in lung and brain, closely resembling metastatic breast cancer in humans (Morecki et al. 1998, Wagenblast et al. 2015). This model contrasts with that used by Kooreman et al. (Kooreman et al. 2018) who evaluated the anti-tumor potential of autologous FVB-derived murine iPSC combined with a CpG ODN 1826 adjuvant on a DB7 breast tumor model. Indeed, DB7 exhibits a very low metastatic potential and does not allow the evaluation of anti-metastatic properties of iPSCs- derived vaccines. This work, using the aggressive 4T1 cell line, reports, not only the tumor protective capacity of PSC, but also their major anti-metastatic potential. We have also evaluated the potential generation of an immunity towards more primitive CSC-like mammosphere cells. It is in fact well established, in several types of cancers that CSCs are the cause of resistances towards standard therapies. Their long term persistence are at the origin of relapses (Holohan et al. 2013; Diehn et al. 2009; Bao et al. 2006). Secondly, CSCs participate to the generation of metastatic dissemination, which can occur even at diagnosis, with presence of infra-clinical metastatic cells which remain in dormancy (Plaks et al. 2015). Besides, it is also well known that a subpopulation of CSCs could undergo EMT (Bronsert et al. 2014) with acquisition of a migratory phenotype, conferring resistance to anti-proliferative drugs (Creighton et al. 2009). Third, CSCs which are resistant to conventional therapies (Shafee et al. 2008; Yamauchi et al. 2008) may limit the efficiency of immunotherapy approaches. Different cancer vaccine approaches so far, use mainly, TAA isolated from differentiated cancer cells, rather than, TAA from CSCs. Today, there is no evidence whether immunotherapy strategies using PSCs exhibit the capacity to target CSCs and/or the tumor microenvironment in TNBC. The 4T1 breast model is well suited to answer to this question, as this cell line present cellular plasticity and stemness phenotype in permissive conditions (Wagenblast et al. 2015, Sally et al. 2007). For this purpose, we developed a protocol allowing to induce via EMT process 4T1-derived mammospheres (MS), expressing ALDH1 (Ginestier et al. 2007) with some CSC-like characteristics. 4T1-derived MS have higher colony formation ability and expressed embryonic stem cell-associated factors such as NANOG and SOX2, shown to be associated with a poor prognosis in breast cancer (Nagata et al. 2014, Liu et al. 2017). Our combinatory regimen using PSCs vaccines along with VPA in this model, has generated a major enhancement of the anti-tumor effect as compared to PSCs vaccination alone, with a highly

151

significant reduction of lung metastases (12-fold). VPA significantly modify the immunosuppressive microenvironment within the primary tumor, suggesting strongly, that it allows the reduction of tumor cells with EMT/CSC phenotype, able to migrate to lungs and brain. This anti-metastatic effect was associated with an increase of CD4+ and CD8+ T cells in tumor and in spleen, with a significant decrease of PD1+ in CD4+ and CD8+ T cells proportions in spleen only in mice treated by combined PSC vaccine and VPA . One of the major findings of our study consisted also, on the demonstration that combined treatment allows to efficiently induce an anti-tumor immunity against 4T1-derived CSCs/MS. As a matter of fact, the combinatory regimen has allowed a highly efficient prevention of the establishment of 4T1-derived MS with an increases of CD4+, CD8+ CTLs and IFN- induction associated with a reduction of Gr1+CD11b+ MDSCs within the tumor. We next evaluated iPSCs as easier source of TAA to immunize naïve mice against lethal dose of 4T1 breast carcinoma cells. For this purpose, we compare two sources of fibroblast-derived iPSCs generated from BALB/c and C57BL/6 strain mice to prime BALB/c mice, and evaluated the anti- tumor immunity generated by the use of syngeneic or allogeneic iPSCs. Mice treated with both type of miPSCs, presented with significantly reduced tumor sizes. The co-injection of VPA and iPSC-based vaccine, highly enhance the anti-tumoral response. A significant improvement in median survival rate was observed especially in group of mice primed with allogeneic material as compared to syngeneic. A striking observation, was the demonstration of an effective generation of memory immune response against breast 4T1 breast carcinoma. Mice treated with 6 boost of vaccination, and challenged with 5x104 4T1-GFP-Luc cells, were more prompt to reject CSC/MS-derived 4T1 cells 180 or 270 days later. This finding strongly suggested that long-term immune memory had been well established is this context via the activation of an efficient anti-tumor immunity mainly against CSCs. Overall, these results, demonstrate the possibility inducing immunization by VPA+iPSC as an adjuvant cancer therapy to prevent risk of relapse and metastasis. Long-term vaccination strategy seems to be safe, without any side effects observed after 6 and 9 months of treatment (weight, hair and musculature were normal, no weakness, without colitis nor cytopenia), and without any clinical evidence of autoimmune diseases.

Importantly, allogeneic miPSCs appeared to be more efficient than syngeneic miPSCs to prevent the establishment of the 4T1 carcinoma cells suggesting a future advantage of this strategy in term of future clinical applications in patients with TNBC. Indeed, allogeneic iPSCs will provide an “off-the-shelf” material compared to autologous iPSCs. Reprogramming of autologous somatic cells require several months for cell characterization and amplification in a clinical GMP

152

condition, making the autologous approaches highly unpractical. Therefore, autologous iPSCs, as future cancer therapy approach, would be extremely difficult to implement at large scale and, to be performed in randomized setting.

Taken together, our data show the feasibility of creating tumor immunity against cancer by combining HDACi and PSCs-based vaccine, that can be easily applied to patients to permit to decrease the occurrence of metastases or distant relapses, by targeting CSCs. This combinatory regimen could increase the survival of some patients whose tumors present stemness phenotype, associated with resistance to current chekpoint immunotherapies. This approach could generate a major benefit as invasion and metastasis account for more than 90% of mortality (Bronsert et al. 2014; Sleeman, et al. 2010; Lazebnik et al. 2010). In addition, combined HDACi and PSCs-based vaccine modulate significantly the tumor microenvironment by recruiting T and B cells and modify immunosuppressive cellular components, such as the myeloid-derived suppressor cells. The latter have been shown to regulate tumor plasticity facilitating the dissemination of tumor cells from the primary site by inducing an EMT/CSC phenotype (Ouzounova et al. 2017). Moreover, tumor associated macrophages (TAMs) have been shown to produce pro-invasive cytokines that not only affect invasion directly, but also sustain the cancer-associated mesenchymal phenotype (Noy et al. 2016). These beneficial properties reported by this combinatory regimen against these immunosuppressive cells represent a novel concept of immunotherapy. This could be applied to several clinical contexts, in particular, to eradicate minimal residual disease, after conventional therapies or in combination with check point inhibitors or adoptive T/NK cell therapies.

153

ACKNOWLEDGMENTS We thank, Olivia Bawa for technical assistances, Olivier Feraud for PSC maintenances. We also thank Ibrahim Casal for animal care, Paule Opolon for anatomopathological analysis. This work was performed with grants from ANR Programme d’Investissement d’Avenir of INGESTEM National Infrastructure (ANR-11-INBS-0009-INGESTEM), INSERM, University Paris Sud and Vaincre le Cancer NRB.

AUTHOR CONTRIBUTION Study concept, direction and funding: A.B.G, F.G., A.T. Generation of miPSCs production and characterization: M.H.K. Molecular analysis: D.C., C.D. A.A. Mice experiments: M.H.K, A.A. D.C. F.G. Cell profiles: M.H.K, A.A, D.C., M-G.H. Flow cytometry analysis: M.H.K, A.A, D.C, M-G.G.H. Microarray: C.D. RT-qPCR: A.A.

154

FIGURES

Figure 1. 4T1 breast tumors display an ESC-like expression signature in vitro and in vivo (A) Transcriptome heatmap showing overexpressed genes found in 4T1 in context of transplantation and shared with D3 murine embryonic stemcell (Euclidean Distances, method complete). (B) Functional enrichment performed with 85 up regulated genes found in 4T1 in context of transplantation and shared with D3 murine embryonic stemcell (Gene Ontology database), bars represent negative log10 of enriched p-values. (C) Quantification of CD44/CD24 markers on 4T1 cells by flow cytometry in vitro and in vivo 12 and 28 days after implantation into the fad pat of BALB/c mice.

155

Color Key

A B

locomotion 2.60 Gene Ontology (Z Scores) extracellular region 2.80

reproduction 2.86

-2 0 1 2 cell cycle 3.18 Value growth Skp2 3.20 Sugt1 Brix1 Cct5 Rad51 chromosome 3.59 Cdc25c Usp11 Crip1 Cacng7 Ppp1r1a cell division 3.89 Pecam1 Lbr Hmgb2 Fam169a biological adhesion Psph 3.92 Grb10 Rad1 Dock6 Slain1 cellular aromatic… 4.45 Cenpv Dcbld1 Aspm Spry 2 Tcea3 nitrogen compound… 4.45 Csrp2 Akap12 Trap1a Nid1 heterocycle metabolic… Zf p532 4.45 Etv 5 Tet1 Ctnnal1 Brca2 Kif 2c Msh2 Trim28 Tmef f 1 Sirt1 C Isotype In vivo (12 days) Zf p202 Rcn1 Sf rp1 Sorl1 0.13 % 18 % Spry 4 Cdc25a Phlda2 Nutf 2 Arhgap8 Ptk7 Etl4 Ntpcr Epcam Fndc3a Kank3 St14 Spint1 Sulf 2 Ddah2 Zbtb8a Mest In vitro (2D) In vivo (28 days) Lamc1 Sy ngr1 Itm2a Aspn 0.02 % 32 % Tert Pmv k Gpa33 Hunk Cth Car13 Hap1 Tiam1 Cldn4 Kcnk1 Esy t3 44 CD Plekha7 H2-Ea-ps Prss22 Adamdec1 Mmp9 Ccl24 CD 24 Ibsp Tubgcp2 p=0.0005 Tslp Gpr97 Moxd1 40 p=0.0076

MG 4T1 MG 4T1 MG mESC 4T1

D3_3 D3_1 D3_2

MG_8 MG_6 MG_7 MG_5 MG_4 MG_1 MG_2 MG_9 MG_3

MG_10

X4T1_1 X4T1_3

X4T1_2 (D3) (Transplant) 30

XENO_4T1_4 XENO_4T1_1 XENO_4T1_2 XENO_4T1_3 20

10 CD44+/CD24 low (%) low CD44+/CD24 0 in vitro in vivo in vivo 2D 12 days 28 days

Figure 1

156

Figure 2. Embryonic stem cell vaccination elicits anti-tumoral effect in vivo (A) Vaccination and challenge protocol: the irradiated stem cells or 4T1 cells were vaccinated to BALB/c mice twice with one week interval then were challenged with 5x104 4T1 cells. (B-D) hESC, mESC and 4T1 vaccinated groups showed significant (p<0.0001) reduction of the tumor volume compared to the control group (n=5). (E) Macroscopic views of tumors from hESC-treated and non-treated groups are shown 28 days showing hESC vaccination reduced tumor size compared to the control group. (F) Significant negative correlation between frequency of cytotoxicity rate (%) of CD8+ T cells in tumor and tumor volume (mm3) in the mean values of each treatment group (p< 0.05). (G) Significant negative correlation between frequency of CD4+ T cells in CD3+ T cells (%) in tumor and tumor volume (mm3). (H) Significant negative correlation between frequencies of infiltrated CD4+ CD25+ Treg cells (%) in tumor and tumor volume (mm3).

157

B A Day -14 -7 0

1e6 PSC/4T1 1e6 PSC/4T1 Challenge

5e4 4T1

) 3

Irradiation Irradiation Tumor volume (mm volume Tumor

C Day post challenge

) )

3 3

Tumor volume (mm volume Tumor Tumor volume (mm volume Tumor

Day post challenge Day post challenge E F hESC Control P<0.05

Tumor volume (mm3)

G H

p=0.0429

(%) p=0.0039

(%)

tumor

in in

. of CD4+CD25+ in in (% CD4+CD25+ of .CD4

. of CD4 in in CD4of .

Freq Freq

Tumor volume (mm3) Tumor volume (mm3)

Figure 2

158

Figure 3. 4T1 cells treated with valproic acid induced an upregulation of the immune response in vitro (A) Immune Network induced by VPA on 4T1 treated in vitro: octagon represent enriched immune module, circle genes connected to enriched module by blue edges (NES: normalized enriched score). (B) Immune Gene Sets significantly enriched in 4T1 cells treated with VPA as compared to untreated 4T1 control condition (NES: normalized enrichment score, p-value and FDR were obtained by hypergeometric test performed on MSigDB 6.1 Hallmarks database). (C) Expression heatmap performed immune genes found significantly overexpression by VPA on 4T1 (algorithm SAM, and unsupervised classification on Euclidean distances with Ward method). (D) Unsupervised principal component analysis performed with immune up regulated genes by VPA in 4T1. (E) boxplot of immune genes regulated which were found significantly overexpressed with a fold change greater than 2.

159

A B Interferon a response

NES = + 1.82 P-value<0.001 FDR<0.001

4T1+VPA 4T1 Interferon  response

NES = + 1.76 P-value<0.001 FDR<0.001

4T1-VPA 4T1

TNFa signaling

NES = + 1.52 P-value<0.001 FDR=0.03

1.3 NES 2.0 4T1-VPA 4T1

C D

Color Key 4

2 4T1

4T1 +VPA 0

-2 -1 0 1 2

Value 2 -

SP110

SERPINB2 (19.55%) Dim2 MSC 4 CCL20 - EGR1 -4 CD274 p=3.3x10 IL10RA CCL7 NCOA7 KLF9 -5 0 5 ATP2B1 TNFRSF9 ZBTB10 Dim 1 (48.62%) LY6E E AREG CCRL2 DHX58 BST2 Fold change= 8.5 STAT1 Fold change= UBE2L6 8 2.75 PSMB9 2.07 IL15 8.0 CD74 CASP4 RNF213 7

NLRC5 7.5 IRS2

SOD2 CD74

DRAM1

6 CCL2

REL 7.0 ISOC1 CSF1 IFIT2

TUBB2A

5 6.5 SLC2A6 IL15RA PLK2 IL7 4T1 4T1+VPA CCL2 4T1 4T1+VPA SAMD9L IFIH1 CEBPD

TDRD7 8.5 HERC6 Fold change=

8.0 2.15

X4T1

X4T1M X4T1_VPA

X4T1M_VPA 4T1+VPA 4T1

7.5

X4T1_TNF_TGF

X4T1M_TNF_TGF

X4T1_VPA_TNF_TGF

X4T1M_VPA_TNF_TGF

7.0

TNFRSF9

6.5 6.0 4T1 4T1+VPA Figure 3

160

Figure 4. hESC vaccination with VPA administration showed synergic effect against primary tumor and metastasis (A) Tumor volume of mice treated with VPA, hESC and hESC+VPA (n=5-6) groups showing a significant reduction of the tumor volume (p<0.01, p<0.0001, p<0.0001 respectively). hESC+VPA showed synergic anti-tumoral effect. (B) Bioluminescent imaging by IVIS showing reduction of the tumor region by hESC+VPA compared to the control tumor. The images on the left shows the time-dependent changes of one mouse from each group. The images on the right shows comparison of the tumor at day 44 post tumor challenge. (C) Direct bioluminescent imaging of the explanted lungs at day 44 post tumor challenge showing the reduction of metastasis load by mESC+VPA compared to the control mice. (D) Total flux (photon/second) in the explanted lungs measured by IVIS showing significant positive correlation with tumor burden. (E) Frequency of CD4+ and CD8+ T cells in tumor were higher than those of the other groups. (F) Frequency of CD4+ and CD8+ T cell in spleen were higher in hESC+VPA treatment group than that of the control group significantly. (G) PD-1 expression on CD4+ T cells and CD8+ T cells were lower in hESC+VPA treatment group than that of the control group significantly. (H) High PD-1 expression on CD4+ T cells in the spleen of the control group whereas lower expression of that of hESC+VPA group.

161

A B

Day 14 24 35 44 44 44

) 3

Control volume (mm volume

Tumor hESC + VPA

Day post challenge C D E

Control (%)

p=0.0062 CD4 of

tumor

in in Frequence

hESC + VPA Control hESC hESC+ VPA

VPA

(%)

of CD8 CD8 of

Total flux in lung (photon/second) lung in flux Total

tumor in in

Tumor weight (g) Frequence

Control hESC hESC+ VPA VPA F G H

of CD4 CD4 of FMO control

ｙ 0.36%

in spleen (%) spleen in

cells

in spleen (%) spleen in

Frequenc

PD1 expression on on expression PD1 CD4+ CD4+ control Control hESC hESC+ VPA Control hESC hESC+ VPA

VPA VPA 60%

CD4 of CD8 CD8 of hESC+ VPA

23%

in spleen (%) spleen in

Frequency cells

PD1 expression on on expression PD1 PD-1

Control hESC hESC+ VPA CD8+ VPA Control hESC hESC+ VPA VPA

Figure 4

162

Figure 5. Evaluation of the anti-tumor and anti-metastatic potential of ESC and VPA combination in the context of tumor stem cells (A) hESC+VPA treatment group showed lower tumor volume (mm3) compared to the control group in the mammosphere challenge model at day 27 post tumor challenge (n=5 per group). (B) mESC+VPA treatment group showed lower tumor volume (mm3) compared to the control group in the mammosphere challenge model at day 27 post tumor challenge (n=5 per group) (C) Bio luminescent imaging showed comparison of the tumor of the control group and hESC+VPA treatment group. (D) hESC+VPA reduced lung metastasis load when measured by direct bioluminescent imaging for total flux (photon/second) in the explanted lungs. (E) Bio luminescent images of the explanted lungs of the control and hESC+VPA groups. (F) Quantification of CD4+ and CD8+ T cells in the tumor from hESC+VPA and control groups. (G) Dot plot images of flow cytometry analysis, showing the increase of CD4 and CD8 T cells in mice treated with hESC+VPA compared to the control group. (H) Quantification of GR1+CD11b+ cell populations in the tumor from mice treated with hESC+VPA compared to the control group. (I) Dot plot images of flow cytometry analysis of the tumors, showing the double Gr1+CD11b+ cell populations in the control group and in mice treated with hESC+VPA. (J) Negative correlation between IFN-γ(pg/mL) in tumor and tumor burden. (K) Quantification of IFN-γ(pg/mL) in tumor from treated and control mice.

163

A 500 B 500 Control Control 450 450 mESC+VPA

hESC+VPA

) ) 3 3 400 400 350 350 ** 300 300

250 250

volume (mm volume (mm volume 200 *** 200

150 Tumor Tumor 150 100 100 50 50 0 0 0 3 6 10 14 16 21 23 27 0 3 6 10 14 16 21 23 27 Day post challenge Day post challenge D C Control hESC+VPA E Control **

hESC+VPA

Total flux inlung (p/s)

hESC mESC control +VPA +VPA F G Control hESC +VPA ** **

0.44% 11%

of CD8 of

of CD4 of

(%)

tumor

CD4

Frequency in

Frequency in 0.51% 10.2% Control hESC mESC Control hESC mESC +VPA +VPA +VPA +VPA CD8 H I FMO control Control hESC+VPA **

0% 41% 12%

(%)

GR1+CD11b+

tumor

CD11b

in in Frequency Control hESC mESC +VPA +VPA Gr-1 J K ** NS

p=0.0083

/ml)

(



(



IFN IFN

Control hESC mESC Tumor weight (g) +VPA +VPA Figure 5

164

Figure 6. Evaluation of anti-tumor effects of murine iPSCs derived from BALB/c and C57Bl/6 strain mice. (A) Immunization with murine BALB/c-derived iPSCs leads to inhibition of tumor growth. (B) Immunization with murine C57BL/6-derived iPSCs leads to inhibition of tumor growth. (C) Effect of allogeneic miPSCs treatment on the survival of mice challenged with 4T1 cells.. (D) Effect of autologous miPSCs treatment on the survival of mice challenged with 4T1 cells. (E) Effect of allogeneic miPSCs+VPA treatment on the survival of mice challenged with 4T1 cells. (F) Effect of autologous miPSCs+VPA treatment on the survival of mice challenged with 4T1 cells. (G) Tumor volume of mice treated allogeneic miPSCs and VPA compare to the non-treated. The data represents the mean + se of tumor volumes (8 mice per group).

165

A B control 1600 control autologous miPSC 1600

) allogeneic miPSC

1400 autologous miPSC + VPA 3

1400 allogeneic miPSC + VPA ) 3 1200 1200 1000 1000 ** 800 (mm volume 800

volume (mm volume ** 600 600 ***

400 Tumor 400 Tumor 200 200 0 0 9 12 15 20 9 12 15 20 Day post challenge Day post challenge 4 T 1 4 T 1 4 T 1 + m iP S C fro m C 5 7 B L /6 D 4 T +control m4 TiP S1C f r o m B A L B /c C control4 T 1 autologous4 T + m iP SmIPSCC fro m B A L B /c allogeneic4 T 1 + m mIPSCiP S C fro m C 5 7 B L /6 1 0 0 1 0 0

l p=0.0084 a

l P=0.0442

v v

i 1 0 0

survival 1 0 0

s survival

5 0

t a

5 0 e

Percent

u 50 0 n

s 0

t 3 5 4 0 4 5 5 0 5 5 6 0 3 5 4 0 4 5 5 0 5 5 6 0

5 0 c n r Day postD a y pchallengeo s t c h a lle n g e e D a y p o s t c h a lle n g e

Day post challenge e 4 T 1

c P r 4 T 1 4 T 1 + m iP S C fro m B A L B /c + V P A E e 4controlT 1

4 T 1 +control m iP S C fr o m C 5 7 B L /6 + V P A F P allogeneic mIPSC + VPA 4autologousT 1 + m iP mIPSCS C fro+ mVPA B A L B /c + V P A

1 0 0 1 0 0 0

l a

0 p=0.0026 a 3 5p=0.0912 4 0 4 5 5 0 5 5 6 0

v r

3 5 4 0 4 5 5 0 5 5r 1 0 0 6 0 D a y p o s t c h a lle n g e

t survival

D a y p o s t c h a lle n g e survival n

5 0 n 5 0

Percent

0 n 05 0 e

3 5 4 0 4 5 5 0 5 5 6 0 c 3 5 4 0 4 5 5 0 5 5 6 0 r

D a y p o s t c h a lle n g e e DayD aposty p o challenges t c h a lle n g e P Day post challenge4 T1

4 T 1 + m iP S C fro m B a lb /c + V P A + T L R 3 )

3 3 0 0 G 4 T1 0 p=0.0091 p=0.0184

m control4 T1

p=0.01433 5 p= 0.00974 0 4 5 * 5 0 5 5 6 0 m

) allogeneic4 TmiPSCs1 + m+ VPAiP S C fro m B a lb /c + V P A + T L R 3 ( 3 * ** ** D a y p o s t c h a lle n g e

e 4 T 1 + m iP S C fro m B a lb /c + V P A + T L R 3

) m

3 3 0 0

u 2 0 0

l p=0.0446 o

m 3 0 0 v *

volume (mm volume

o m

( **

m * **

u *

1 0 0

Tumor e T NS

) ** 3 m * **

u 2 0 0 l m *

o 0

m 0 2 4 6 0 3 6 8

v 1 1 * 1 1

(

0 3 6 8 D 10a y s 12 14 20 r

e 2 0 0

o Day

u u l *

1 0 0 o

T NS

v Figure 6

m * u 1 0 0

T NS 0 0 2 4 6 0 3 6 8 166 1 1 1 1 D a y s

0 3 6 8 0 2 4 6 1 1 1 1 D a y s

Figure 7. Effective Memory Immune Response after Vaccination with miPSCs (A) In vivo memory model schema consists to six subcutaneous injection of 15Gy irradiated 2x106 miPSC cells in the flank of BALB/c mice. After 6 or 9 months mice were challenged with the 2.5x104 4T1GFP-Luc cells or 4T1GFP-Luc MS-derived cells. (B) Significant reduction of breast tumor sizes (n=5 per group) in mice challenged with 2.5x104 4T1GFP-Luc cells 6 months after having receiving 6 boots of miPSCs compared to non- vaccinated mice. (C) Significant reduction of breast tumor sizes at day 28 (n=6 per group) in mice challenged with 2.5x104 4T1GFP-Luc cells 9 months after having receiving 6 boots of miPSCs compared to non- vaccinated mice (D) Image of lungs isolated from vaccinated and control mice 28 days after having challenged the mice with 2.5x104 4T1GFP-Luc cells. (E) Significant correlation between tumor burden and metastatic spread at day 28 for mice injected with 2.5x104 4T1GFP-Luc cells. (F) Significant reduction of breast tumor sizes (n=6 per group) in mice challenged with 2.5x104 4T1GFP-Luc MS-derived cells 9 months after having receiving 6 boots of miPSCs compared to non-vaccinated mice. (H) Image of lungs isolated from vaccinated and control mice 28 days after having challenged the mice with 2.5x104 4T1GFP-Luc MS-derived cells. (I) Significant correlation between tumor burden and metastatic spread at day 28 for mice injected with 2.5x104 4T1GFP-Luc MS-derived cells.

167

Day -180 -150 -120 -90 -60 -30 0 +90

2e6 2e6 2e6 2e6 2e6 2e6 miPSCs miPSCs miPSCs miPSCs miPSCs miPSCs

or 15 gy 15 gy 15 gy 15 gy 15 gy 15 gy

Challenge Challenge 2.5e4 4T1 2.5e4 4T1 MS +VPA +VPA

B C D

2500 700 control control miPSC+VPA 600 miPS + VPA Control

2000

)

3 )

3 500 1500 miPSC +VPA 400 *** E volume (mm volume 300 1000 *** (mm volume

Tumor 200 Tumor 500 *** 100

0 0 0 5 8 15 19 22 26 7 9 13 16 20 23 28 Day post challenge Day post challenge F G H 600 control Control Control 500 miPS + VPA

) miPSC +VPA 3 400 I 300 volume (mm volume miPSC +VPA 200

Tumor *** 100

0 7 9 13 16 20 23 28 Day post challenge

Figure 7

168

Figure 8. Lymphoid immune profile infiltration in 4T1 tumors in mice primed with miPSCs. (A) Differential expressed genes (DEG) regulated between control and vaccinate mice transplanted with 4T1, expression heatmap was processed with unsupervised classification algorithm (Euclidean distances, ward method). (B) Barplot of functional enrichment on Gene Ontology Biological Process performed with overexpressed genes in vaccine condition. (C) Immune profiling performed by gene set enrichment analysis on transcriptome of 4T1- transplanted mice (NES: normalized enriched score) (D) Molecular Immune network enriched in 4T1 tumors from mice that were vaccinated with miPSC+VPA.

169

Color Key A B

-1 0 1 Value

EdnrbCxcl14Ctsk Atp6v0d2Gabrb2Gm20878 Hba-a1Hbb-bt Hbb-bsHba-a2 NgpH2-Ke6Clca3a2 Slitrk1Gsg1l2 Zmynd8Nts Card14Pm20d1 Tacr1Gm2310Gm11060 Ccl24Pgap1 Pvrl3Ibsp Fgf3Abcb1b Wfdc18Wscd2Lum Clca3a1Gm5878 FcgbpCbr3 Tff3Cd55 Gm10663Serpina16 ManealClec12bIl1a LtfPla2g2dCcdc141 Sim2Kcnh7 Gm13304Crlf1 4930432M17RikAdamdec1 108 Rsad2Bst2Isoc2a Parp14Ifit3b Stfa3Ifit3 C1s2Tmem140 Ifit1IfngGdap10 Slfn4Zfp874b Nos2Atp8b3 Calr4AA467197 H2-T22Gm18853Sepw1 DOWN C1s1Gbp10 Gbp4Irgm1 Apol6Igtp Gbp8Gbp2b Irf1H2-T23Gm4070 Tmprss6Psmb10Cd274 Gbp2Gm12250 Iigp1Stat1 Gm4841Gimap9 Ube2l6Irf7Cxcl9 Gm8979Gm8989Cxcl11 Xaf1F830016B08Rik Gbp9Gm4951 Gvin1Gbp3Gm12185 Tgtp2Gbp11 Ssx2ipIrgm2 Txnrd3Syt13 Cd46Gm5595Pmp22 CdonStard3 Lmo7Slx4ip Nmrk1Zfp595 Ccr2Gfra2 Olfr452Ica1Rnf223 Gm1818SordSmtnl1 MestNipal3 Ncoa7Fxyd1Inmt Krt14Ppp1r3fHhip Sod3Wasf1 BlkKlk1b11

206DEG Ndufa4l2 Serpinb2Msmp PigwMir675Cldn9 Smr2Gm17482 Pgr15lRasl12 9230104L09RikEntpd5 Dpep1Atp13a5Kdm1b Pbx4Ano2 98 UP 1700011H14RikEnpp2 Olfr822 Ccl22Olfr384Olfr1258 CtseCd72Bin2 Tc2nCxcl13 Clec12aCpa3Klra17 Csf2raGm2573Trp53i11 Cd300eLy96 HnmtFhitSlc6a8 Gjb4Ms4a7Aldh2 Mtfr2Pid1 Naf1Aim2Etv3 Egln3Olfml2bTuft1 Sh2d4aClec4a1 Smpd4Fam49aGhr Il20rbRassf9Shroom3 Vps33bNupl2 Psmg1Ttll1Hormad2 Ms4a14Pou4f1Fbln5

Control Treated

X4t1_control_3 X4t1_control_4

X4t1_vaccin_13 X4t1_vaccin_14

NES=1.66 NES=1.54 NES=1.33 C P-value<0.001 P-value<0.001 P-value<0.001 FDR<0.001 FDR=0.083 FDR=0.016

1.2 NES 2

Figure 8

170

Supplementary figure 1. Characterization of D3 murine embryonic stem cells. (A) Murine iPSCs exhibited typical markers of pluripotency as expression of SSEA1 by flow cytometry and expanded as typical iPSC colonies. (B) Expression of key pluripotency markers NANOG, OCT4 by immunoﬂuorescence. (C) The differentiation capacity into the three germ layers was confirmed by teratoma formation assays showing a differentiation into ectodermal, endodermal and mesodermal tissues.

171

A B C

mESC (D3) DAPI NANOG Mesoderm (cartilaginous) Endoderm (intestinal E.)

SSEA1

>90%

DAPI OCT4 Ectoderm (neural crest)

Supplementary Figure 1

172

Supplementary figure 2. VPA treatment increases MHC I expression (A) Increase of MHC I expression on 4T1 cells surface by flow cytometry during treatment with different doses of VPA (0, 0.2 and 2 mM). (B) Variation in the expression of MHC I during VPA treatment at 2 mM on 4T1 adherent cells and on 4T1-derived mammospheres.

173

A Isotype VPA (mM): 0 0.2 2

MHC I on 4T1 cells

B Without VPA With VPA (2mM)

MHC I on 4T1 cells

RFI =351 RFI =753

MHC I on 4T1 mammospheres

RFI=172 RFI=467

Supplementary Figure 2

174

Supplementary figure 3. Impact of VPA and CpG with hESC vaccination to tumor formation and lung metastasis (A) Vaccination and VPA administration protocol: 2x106 irradiated hESC cells were vaccinated twice with one week interval then 4T1-GFP-Luc cells were implanted to the mammary fat pad. VPA was administered by drinking water from the day of tumor challenge. (B) Mice vaccinated with hESC and hESC+VPA showed significant reduction of tumor volume compared to the control group in contrast to mice having been vaccinated with hESC+CpG. (C) Mice vaccinated with hESC+VPA showed significantly lower tumor weight compared to the control mice. (D) Quantification (Total flux) of luciferase by bioluminescent imaging in lung after dissection.

175

A Day -14 -7 0

Challenge 2e6 PSC + CpG 2e6 PSC + CpG 5e4 4T1-GFP-Luc + VPA

Irradiation Irradiation

B C NS

** NS

Supplementary figure 3

176

Supplementary figure 4. hESC vaccination combined with VPA administration resulted in drastic reduction of tumor volume Tumor volume at day 44 post tumor challenge from mice vaccinated with hESC, VPA and hESC+VPA showing a significant decrease of the tumor volume compared to mice from control group.

177

p<0.0001

p<0.001

p<0.0001 1,600 P<0.05

1,400

1,200

) 3 1,000

800 volume (mm volume

600 Tumor

400

200

0 Control hESC VPA hESC +VPA

Supplementary figure 4

178

Supplementary figure 5. Production of Mammosphere from 4T1 cell line. (A) Images of the morphology of adherent 4T1 cells, of mammosphere-derived 4T1 amplified in 3D culture conditions without or with TGFβ +TNFα. (Magnification x20). (B) Quantification of ALDH1 activity by flow cytometry in adherent 4T1 cells and in mammosphere-derived 4T1 amplified in 3D culture conditions without or with TGFβ +TNFα. (C) Percentage of ALDH1 positive cells in 3 independent experiments in adherent 4T1 cells and in mammosphere-derived 4T1 amplified in 3D culture conditions without or with TGFβ +TNFα. (D) Mammosphere Forming Efficiently (MEF) quantified by using the Cell Selector Software permitting to quantify the mammospheres sizes in the presence or absence of TGF-β+TNFα. (E) Quantification of NANOG, OCT4 and SOX2 expression at the mRNA levels by qRT-PCR in 4T1 adherent cells and in mammosphere-derived 4T1 amplified in 3D culture conditions without or with TGFβ +TNFα.

179

A B Phase ALDH+DEAB ALDH

Control 0.7% 12.7% (2D) Control (2D)

Mammosphere (3D) 0.11% 21.4% Mammosphere (3D)

Mammosphere + TGFβ+TNFα (3D) 46.1% Mammosphere +

0.36% TGFβ+TNFα (3D) SSC

ALDH

C D E 250 x90 50 x3 2500 x7 x32 x1.3 200 x1.5 40 2000 x1.7 x1.4 150 30 1500

100 ALDH (%) ALDH 20 1000

MFE (%) x1.2

10 500 50 Relative Relative quantity of mRNA 0 0 0 NANOG OCT4 SOX2

Control (2D) Mammosphere (3D) Control (2D) Mammosphere (3D) Mammosphere + Mammosphere (3D) Mammosphere + TGFβ+TNFα (3D) Mammosphere + TGFβ+TNFα (3D) TGFβ+TNFα (3D)

Supplementary figure 5

180

Supplementary figure 6. Evaluation of the anti-tumor and anti-metastatic potential of human and murine ESC and VPA combination on mice challenged with mammosphere- derived 4T1 produced in 3D culture condition with TGFβ +TNFα. (A) Tumor growth showing significantly higher tumor volume when 2.5 104 mammosphere- derived 4T1 were injected compared to the mice receiving non-mammasphere 4T1 cells that were harvested in 2D culture condition. (B) hESC+VPA and mESC+VPA showed significant reduction (p<0.001) of tumor volume at day 23 post tumor challenge (injection of 2.5x104 mammosphere-derived 4T1 produced with TGFβ +TNFα). (C) Bioluminescent images showing smaller tumor for mice vaccinated with hESC+VPA compared to the non-vaccinated control mice. (D) Direct bioluminescent images of the dissected lungs showing a reduction or absence of the metastasis load in the lungs of mice treated with hESC+VPA and challenged 2.5x104 mammosphere-derived 4T1 produced with TGFβ +TNFα compared to non-vaccinated control mice.

181

A Control (2D) B Mammosphere + 250 TGFb+TNFa (3D) control hESC +VPA

140 )

3 +VPA p<0.01 200 mESC

120 p<0.01

) 3 100 150

80 (mm volume 100

60 p<0.001 Tumor p<0.001 40 50

Tumor volume (mm volume Tumor 20

0 0 0 3 6 10 14 16 0 3 6 10 14 16 21 23 Day post challenge Day C Control hESC+ VPA

D Control hESC+ VPA

Supplementary figure 6

182

Supplementary figure 7. Characterization of murine induced Pluripotent Stem Cells derived from BALB/c fibroblasts (A) Membranous expression of SSEA1 by flow cytometry and morphology of the miPSCs expanded on MEF. (B) Expression of key pluripotency markers NANOG, OCT4 expression by immune-fluorescence. (C) Teratoma formation assays showing a differentiation into ectodermal, endodermal and mesodermal tissues.

183

A B C

miPS (BALB/c) DAPI NANOG Mesoderm (bone) Endoderm (intestinal E.) SSEA1

>90%

DAPI OCT4 Ectoderm (nerve)

Supplementary figure 7

184

Supplementary figure 8. Immune cell profiling in mice vaccinated with miPSCs and VPA (A) Percentage of CD4+, CD8+T cells in spleen (B) Percentage of PD1+ in CD4+ and CD8+ T cells and CD4+, CD25+, FoxP3+ T Reg in spleen.

185

p=0.0005 p=0.0012

A 5 1 .5

1 .0

of CD4 of of CD8 of

2 0 .5

Frequency in spleen (%) spleen in

Frequency (%) spleen in

0 0 .0 Control miPSC+ VPA Control miPSC +VPA

B 7 5 1 .5 p=0.0060 p=0.0450

7 0

1 .0 Treg

6 5 of of CD4+PD1+ CD4+PD1+ of

0 .5

6 0

Frequency (%) spleen in

Frequency In spleen (%) spleen In 5 5 0 .0 Control miPSC +VPA Control miPSC +VPA

Supplementary figure 8

186

Supplementary figure 9. Quantification of CXCL9, CXCL10 and CXCL13 mRNA by RT-qPCR in 4T1-derived tumors from mice vaccinated with miPSCs and VPA and in non-treated mice.

187

4500000 CXCL9 7000000 CXCL10 CXCL13 4000000 1200000000 6000000 3500000 1000000000 x 3864 5000000 x 26244 x 3403 3000000 800000000 2500000 4000000 600000000 2000000 3000000 1500000 2000000 400000000 1000000 200000000 500000 1000000

0 0 0 control mIPS+VPA control mIPS+VPA control mIPS+VPA

Supplementary figure 9

188

REFERENCES

Al-Hajj, M., Wicha, M.S., Benito-Hernandez, A., Morrison, S.J., Clarke, M.F. (2003). Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A 100, 3983–3988.

Aslakson, C.J., Miller, F.R.(1992). Selective events in the metastatic process defined by analysis of the sequential dissemination of subpopulations of a mouse mammary tumor. Cancer Res. 52, 1399–1405.

Bao, S., Wu, Q., McLendon, R.E., Hao, Y., Shi, Q., Hjelmeland, A.B., Dewhirst, M.W., Bigner, D.D., Rich, J.N. (2006). Glioma stem cells promote radioresistance by preferential activation of the DNA damage response. Nature 444, 756–760.

Ben-Porath, I., Thomson, M.W., Carey, V.J., Ge, R., Bell, G.W., Regev, A., Weinberg, R.A. (2008). An embryonic stem cell-like gene expression signature in poorly differentiated aggressive human tumors. Nat. Genet. 40, 499-507.

Bode, C., Zhao, G., Steinhagen, F., Kinjo, T. and Klinman, D. M. (2011). CpG DNA as a vaccine adjuvant. Expert Rev Vaccines 10 ,499-511.

Brewer, B.G., Mitchell, R.A., Harandi, A., Eaton, J.W. (2009). Embryonic vaccines against cancer: an early history. Exp Mol Pathol. 86, 192-7.

Bronsert, P., Enderle-Ammour, K., Bader, M., Timme, S., Kuehs, M., Csanadi, A., Kayser, G., Kohler, I., Bausch, D., Hoeppner, J., Hopt, U.T., Keck, T., Stickeler, E., Passlick, B., Schilling, O., Reiss, C.P., Vashist, Y., Brabletz, T., Berger, J., Lotz, J., Olesch, J., Werner, M., Wellner, U.F. (2014). Cancer cell invasion and EMT marker expression: a three-dimensional study of the human cancer–host interface. J Pathol. 234, 410–422.

Chen, J., Li, Y., Yu, T.-S., McKay, R.M., Burns, D.K., Kernie, S.G., and Parada, L.F. (2012). A restricted cell population propagates glioblastoma growth after chemotherapy. Nature 488, 522– 526.

Clarke, M.F., Dick, J.E., Dirks, P.B., et al. (2006). Cancer stem cells--perspectives on current status and future directions: AACR Workshop on cancer stem cells. Cancer Res 66, 9339–9344.

189

Cline, M.S., Smoot, M., Cerami, E., Kuchinsky, A., Landys, N., Workman, C., Christmas, R., Avila-Campilo, I., Creech, M., Gross, B., et al. (2007). Integration of biological networks and gene expression data using Cytoscape. Nat. Protoc. 2, 2366–2382.

Creighton, C.J., Li, X., Landis, M., Dixon, J.M., Neumeister, V.M., Sjolund, A., Rimm, D.L., Wong, H., Rodriguez, A., Herschkowitz, J.I., Fan, C., Zhang, X., He, X., Pavlick, A., Gutierrez, M.C., Renshaw, L., Larionov, A.A., Faratian, D., Hilsenbeck, S.G., Perou, C.M., Lewis, M.T., Rosen, J.M., Chang, J.C. (2009). Residual breast cancers after conventional therapy display mesenchymal as well as tumor-initiating features. Proc. Natl. Acad. Sci. USA 106, 13820–13825.

Culhane, A.C., Thioulouse, J., Perrière, G., Higgins, D.G. (2005). MADE4: An R Package for Multivariate Analysis of Gene Expression Data. Bioinformatics (Oxford, England) 21, 2789-90.

De Almeida, P.E., Meyer, E.H., Kooreman, N.G., Diecke, S., Dey, D., Sanchez- Freire, V., Hu, S., Ebert, A., Odegaard, J., Mordwinkin, N.M., et al. (2014). Transplanted terminally differentiated induced pluripotent stem cells are accepted by immune mechanisms similar to self-tolerance. Nat. Commun. 5, 3903.

Diehn, M., Cho, R.W., Lobo, N.A., Kalisky, T., Dorie, M.J., Kulp, A.N., Qian, D., Lam, J.S., Ailles, L.E., Wong, M., Joshua, B., Kaplan, M.J., Wapnir, I., Dirbas, F.M, Somlo, G., Garberoglio, C., Paz, B., Shen, J., Lau, S.K., Quake, S.R., Brown, J.M., Weissman, I.L., Clarke, M.F. (2009). Association of reactive oxygen species levels and radioresistance in cancer stem cells. Nature 458, 780–783.

Driessens, G., Beck, B., Caauwe, A., Simons, B.D., and Blanpain, C. (2012). Defining the mode of tumour growth by clonal analysis. Nature 488, 527–530.

Dong, W., Du, J., Shen, H., Gao, D., Li, Z., Wang, G., Mu, X., Liu, Q. (2010). Administration of embryonic stem cells generates effective antitumor immunity in mice with minor and heavy tumor load Cancer Immunol Immunother. 59, 1697–1705.

DuPré, S.A., Redelman, D., Hunter, K.W. Jr. (2007). The mouse mammary carcinoma 4T1: characterization of the cellular landscape of primary tumours and metastatic tumour foci. Int J Exp Pathol.88, 351–360.

Galardi, S., Savino, M., Scagnoli, F., Pellegatta, S., Pisati, F., Zambelli, F., Illi, B., Annibali, D.,

190

Beji, S., Orecchini, E., Alberelli, M.A., Apicella, C., Fontanella, R.A., Michienzi, A., Finocchiaro, G., Farace, M.G., Pavesi, G., Ciafrè, S.A., Nasi, S.0. (2016). Resetting cancer stem cell regulatory nodes upon MYC inhibition. EMBO Rep 17, 1872-1889.

Ginestier, C., Hur, M.H., Charafe-Jauffret, E., Monville, F., Dutcher, J., Brown, M., Jacquemier, J., Viens, P., Kleer, C.G., Liu, S., Schott, A., Hayes, D., Birnbaum, D., Wicha, M.S., Dontu, G. (2007). ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome. Cell Stem Cell 1, 555–67.

Glinsky, G.V. (2008). “Stemness” genomics law governs clinical behavior of human cancer: implications for decision making in disease management. J Clin Oncol. 26, 2846–2853.

Holohan, C., Van Schaeybroeck, S., Longley, D.B. & Johnston, P.G. (2013). Cancer drug resistance: an evolving paradigm. Nat. Rev. Cancer 13, 714–726.

Huang, D.W., Sherman, B.T., and Lempicki, R.A. (2009). Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat. Protoc. 4, 44–57.

Hussein, S.M., Batada, N.N., Vuoristo, S., Ching, R.W., Autio, R., Närvä, E., Ng, S., Sourour, M., Hämäläinen, R., Olsson, C., Lundin, K., Mikkola, M., Trokovic, R., Peitz, M., Brüstle, O., Bazett- Jones, D.P., Alitalo, K., Lahesmaa, R., Nagy, A., Otonkoski, T. (2011). Copy number variation and selection during reprogramming to pluripotency. Nature.471, 58-62.

Irizarry, R.A., Bolstad, B.M., Collin, F., Cope, L.M., Hobbs, B., and Speed, T.P. (2003). Summaries of Affymetrix GeneChip probe level data. Nucleic Acids Res. 31, e15.

Jonsdottir, G.A., Ruotti, V., Stewart, R., Slukvin, I.I., Thomson, J.A. (2007). Induced pluripotent stem cell lines derived from human somatic cells. Science 318, 1917-1920.

Kooreman, N.G., Kim, Y., de Almeida, P.E., Termglinchan, V., Diecke, S., Shao, N.Y., Wei, T.T.,Yi, H., Dey, D., Nelakanti, R., Brouwer, T.P., Paik, D.T., Sagiv-Barfi, I., Han, A., Quax, P.H.A., Hamming, J.F., Levy, R., Davis, M.M., Wu, J.C. (2018). Autologous iPSC-Based Vaccines Elicit Anti-tumor Responses In Vivo. Cell Stem Cell 22, 501-513.

Lazebnik, Y. (2010) What are the hallmarks of cancer? Nat. Rev. Cancer 10, 232–233.

191

Li, Y., Zeng, H., Xu, R.H., Liu, B., and Li, Z. (2009). Vaccination with human pluripotent stem cells generates a broad spectrum of immunological and clinical responses against colon cancer. Stem Cells 27, 3103–3111.

Liu, K., Xie, F., Gao, A., Zhang, R., Zhang, L., Xiao, Z., Hu, Q., Huang, W., Huang, Q., Lin, B., Zhu, J., Wang, H., Que, J., Lan, X. (2017). SOX2 regulates multiple malignant processes of breast cancer development through the SOX2/miR-181a-5p, miR-30e-5p/TUSC3 axis. Mol Cancer 16:62.

Lyons, Y.A., Wu, S.Y., Overwijk, W.W., Baggerly, K.A., and Sood, A.K. (2017). Immune cell profiling in cancer: molecular approaches to cell-specific identification. NPJ Precis. Oncol. 1, 26.

Morecki, S., Yacovlev, E., Diab, A., Slavin, S. (1998). Allogeneic cell therapy for a murine mammary carcinoma. Cancer Res. 58, 3891–3895.

Nagata, T., Shimada, Y., Sekine, S., Hori, R., Matsui, K., Okumura, T., Sawada, S., Fukuoka, J., Tsukada, K. (2014). Prognostic significance of NANOG and KLF4 for breast cancer. Breast Cancer 21, 96-101.

Noy, R. and Pollard, J.W. (2016) Tumor-associated macrophages: from mechanisms to therapy. Immunity 41, 49–61.

Ouzounova, M., Lee, E., Piranlioglu, R., E.l., Andaloussi, A., Kolhe, R., Demirci, M.F., Marasco, D., Asm, I., Chadli, A., Hassan, K.A., Thangaraju, M., Zhou, G., Arbab, A.S., Cowell, J.K., Korkaya, H. (2017). Monocytic and granulocytic myeloid derived suppressor cells differentially regulate spatiotemporal tumour plasticity during metastatic cascade. Nat Commun. 8, 14979.

Padovani, M., Lavigne, J.A., Chandramouli, G.V., Perkins, S.N., Barrett, J.C., Hursting, S.D., Bennett, L.M., Berrigan, D. (2009). Distinct Effects of Calorie Restriction and Exercise on Mammary Gland Gene Expression in C57BL/6 Mice. Cancer Prevention Research 2, 1076-87.

Pearce, D.J., Taussig, D., Simpson, C., Allen, K., Rohatiner, A.Z., Lister, T.A., et al. (2005). Characterization of cells with a high aldehyde dehydrogenase activity from cord blood and acute myeloid leukemia samples. Stem Cells 23, 752-60.

Plaks, V., Kong N, Werb Z. (2015). The Cancer Stem Cell Niche: How Essential is the Niche in

192

Regulating Stemness of Tumor Cells? Cell Stem Cell 16, 225–238.

Prager, B.C., Xie, Q., Bao, S., Rich, J.N. (2019). Cancer Stem Cells: The Architects of the Tumor Ecosystem. Cell Stem Cell 24, 41-53.

Pulaski, B.A., Ostrand-Rosenberg, S. (2001). Mouse 4T1 breast tumor model. Curr Protoc Immunol. 2001;Chapter 20:Unit 20.2

Puram, S.V., Tirosh, I., Parikh, A.S., Patel, A.P., Yizhak, K., Gillespie, S., Rodman, C., Luo, C.L., Mroz, E.A., Emerick, K.S., et al. (2017). Single-Cell Transcriptomic Analysis of Primary and Metastatic Tumor Ecosystems in Head and Neck Cancer. Cell 171, 1611–1624..

Reya, T., Morrison, S.J., Clarke, M.F., Weissman, IL. (2001). Stem cells, cancer, and cancer stem cells. Nature 414, 105–111.

Robinton, D.A., Daley, G.Q. (2012). The promise of induced pluripotent stem cells in research and therapy. Nature 481, 295–305.

Schlaeger, T.M., Daheron, L., Brickler, T.R., Entwisle, S., Chan, K., Cianci, A., DeVine, A., Ettenger, A., Fitzgerald, K., Godfrey, M., Gupta, D., McPherson, J., Malwadkar, P., Gupta, M., Bell, B., Doi, A., Jung, N., Li, X., Lynes, M.S., Brookes, E., Cherry, A.B., Demirbas, D., Tsankov, A.M., Zon, LI., Rubin, L.L., Feinberg, A.P., Meissner, A., Cowan, C.A., Daley, G.Q. (2015). A comparison of non-integrating reprogramming methods. Nat. Biotechnol. 33: 58-63.

Schoenhals, M., Kassambara, A., De Vos, J., Hose, D., Moreaux, J., Klein, B. (2009). Embryonic stem cell markers expression in cancers. Biochem Biophys Res Commun. 383, 157–162.

Shabbeer, S., Kortenhorst, M. S., Kachhap, S., Galloway, N., Rodriguez, R. and Carducci, M. A. (2007). Multiple Molecular pathways explain the anti-proliferative effect of valproic acid on prostate cancer cells in vitro and in vivo. Prostate 67, 1099-110.

Shafee N, Smith CR, Wei S, Kim Y, Mills GB, Hortobagyi GN, Stanbridge, E.J., Lee, E.Y. (2008). Cancer stem cells contribute to cisplatin resistance in Brca1/p53-mediated mouse mammary tumors. Cancer Res. 68, 3243–50.

Sleeman, J. and Steeg, P.S. (2010) Cancer metastasis as a therapeutic target. Eur. J. Cancer 46,

193

1177–1180.

Sondheimer, N., and Lindquist, S. (2000). Rnq1: an epigenetic modifier of protein function in yeast. Mol. Cell 5, 163–172 Stonehill, E.H., Bendich, A. (1970). Retrogenetic expression: the reappearance of embryonal antigens in cancer cells. Nature 228, 370-2.

Subramanian, A., Tamayo, P., Mootha, V.K., Mukherjee, S., Ebert, B.L., Gillette, M.A., Paulovich, A., Pomeroy, S.L., Golub, T.R., Lander, E.S., et al. (2005). Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl. Acad. Sci. U. S. A. 102, 15545–15550.

Takahashi, K., and Yamanaka, S. (2006). Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 126, 663-676.

Tao, K., Fang, M., Alroy, J., Sahagian, G.G. (2008). Imagable 4T1 model for the study of late stage breast cancer. BMC Cancer 8, 228.

Thomson, J.A., Itskovitz-Eldor, J., Shapiro, SS, Waknitz MA, Swiergiel JJ, Marshall VS, Jones JM. (1998). Embryonic stem cell lines derived from human blastocysts. Science 282, 1145-7.

Tibshirani, R., Hastie, T., Narasimhan, B., Chu, G. (2002). Diagnosis of Multiple Cancer Types by Shrunken Centroids of Gene Expression . Proc Natl Acad Sci U S A. 99, 6567-72.

Tusher, V. G., R. Tibshirani, et G. Chu. (2001). Significance Analysis of Microarrays Applied to the Ionizing Radiation Response . Proc Natl Acad Sci U S A 98, 5116-21.

Wagenblast, E., Soto, M., Gutierrez-Angel, S., Hartl, C.A., Gable, A.L., Maceli, A.R., Erard, N., Williams, A.M., Kim, S.Y., Dickopf, S., et al. (2015). A model of breast cancer heterogeneity reveals vascular mimicry as a driver of metastasis. Nature 520, 358–362.

Yaddanapudi, K., Mitchell, R.A., Putty, K., Willer, S., Sharma, R.K., Yan, J., Bodduluri, H., and Eaton, J.W. (2012). Vaccination with embryonic stem cells protects against lung cancer: is a broad-spectrum prophylactic vaccine against cancer possible? PLoS ONE 7, e42289.

Yamauchi, K., Yang, M., Hayashi, K., Jiang, P., Yamamoto, N., Tsuchiya, H., Tomita, K., Moossa,

194

A.R., Bouvet, M., Hoffman, R.M. (2008). Induction of cancer metastasis by cyclophosphamide pretreatment of host mice: an opposite effect of chemotherapy. Cancer Res. 68, 516–20.

Yu, J., Vodyanik, M.A., Smuga-Otto, K., Antosiewicz-Bourget, J., Frane, J.L., Tian, S., Nie, J.,

Zambon, A.C., Gaj, S., Ho, I., Hanspers, K., Vranizan,K., Evelo, C.T., Conklin, B.R., Pico, A.R., Salomonis, N. (2012). GO-Elite: A Flexible Solution for Pathway and Ontology over- Representation. Bioinformatics (Oxford, England) 28, 2209-10.

Zhang, Z., Chen, X., Chang, X., Ye, X., Li, Y., Cui, H. (2013). Vaccination with embryonic stem cells generates effective antitumor immunity against ovarian cancer. Int J Mol Med. 31, 147-53.

Zhao, T., Zhang, Z.N., Rong, Z., and Xu, Y. (2011). Immunogenicity of induced pluripotent stem cells. Nature 474, 212–215.

195

MAJOR FINDINGS AND PERSPECTIVE

We are the first having demonstrated that whole lethally inactivated pluripotent stem cells can be used to protect a 4T1 TNBC in a syngeneic mice model. This Pluripotent Stem cell-based vaccine (PSCV) have succeeded to significantly reduce the breast tumor burden that was correlated with the recruitment of CD4+ and CD8+ T cells to the tumor. In addition, we observed an increase of CD4+ and CD8+ T cells also in spleen showing that the mice have acquired a systemic adaptive immune memory against the shared antigens between PSC and 4T1 tumors cells. Additionally, the long-term effect of our PSCV offers attractive strategy in clinical settings for the prevention and the recurrence of breast cancer after surgery of TNBC patients.

We also have demonstrated that CSC population marked with CD44+CD24-/low increased in vivo in time dependent manner after the tumor implantation. 4T1 gene signature analyzed with transcriptome analysis revealed that the gene expression of in vitro (DMEM) cultured 4T1 cells switched to the ES-like signature once transplanted into the mammary fat pad in mouse. Both observations indicate critical influence of TME in cellular transformation into CSC phenotype.

HDAC inhibitor, such as VPA, showed synergic anti-tumor effect with the PSCV when administrated at the currently approved oral dose for epilepsy patients. Moreover, VPA has shown to partially rescued the transformation of 4T1 cells to CD44+CD24-/low CSC phenotype (Figure 49 in Annex 7).

In our experimental settings, we have used auto, allo or xeno materials to vaccinate the mice. Interestingly the used of human ESC have shown a high reduction of mammary tumor burden as well a drastic reduction of lung metastasis implying that cross reactivity either by T and/or B cells have occurred against the homolog peptides or protein products shared between human and mouse materials. It is also apparent that innate immune reaction should be different. The difference occurred by the different ligation to PRR or TLRs on the immune cells by the auto, allo or xeno materials may lead different immunological consequences, which remains to be investigated.

It is noteworthy that we didn’t see any safety issues accompanied with PSCV, which is constant observation with the other studies using PSCVs. It has been shown (36) that neural stem cell and retinal stem cells were clustered similarly with ESC whereas other

196

adult stem cells including mammary stem cells have showed distanced clustering. This may provide caution for the presence of shared antigens between neural tissues and ESCs. However, during our vaccination programs we didn’t have seen any neurological or muscular disorders even after more than 9 mouths post-vaccination meaning that PSCVs are quite safe. As TNBCs show relatively high PD-L1 expression compared to the other BC subtypes, using immune check point inhibitor (CPI) can been expected to be an attractive approach to cure those patients. Despite much anticipation, phase 1b clinical trial using anti–PD-L1 antibody showed only a modest effect (18% ORR). This may tell us that the limitation of natural recognition of the tumor antigens by immune cells, thereby only blocking the PD1/PDL1 interactions didn’t work well. Combination of a cancer vaccine approach with an CPI may solve this problem by triggered a synergy effect on the immune system, permitting to expand T and B cell repertoires by the vaccine, to block the emergence of immune tolerance by CPI and to consequently activate anti-tumor immunity.

In contrast to the use of PSV-derived differentiated cells for regeneration medicine our objective is to immunize the mice with PSCV and allogeneic materials may be easily used. Even though there are several reports that autologous, allogeneic or xenogeneic PSC can be used as vaccine products in different animal models, more evidence should be necessary to say that it would be preferable to use a vaccine product of the same species: human in human or mice in mice. Additionally, it may also depend, on the regulatory requirement in each country, cultural and religious background at clinical trials, either ESC versus iPSC, or either autologous, allogeneic versus xenogeneic material.

The discoveries of hESC generation in 1998 (330) and hiPSC generation in 2007 (332) have delivered an exciting new era for the fields of stem cell, regenerative medicine, disease modelling and therapy discovery. The first clinical study to evaluate human iPSC-derived cells was initiated in 2014, which is only 7 years after the discovery of iPSC technology. The study used human iPSC-derived retinal pigment epithelial cells to treat macular degeneration and the treatment was reported to improve the patient's vision (403). Many clinical trials using iPSCs/ESCs have been performed or started. Those trials are the therapy targeting retinal diseases, spinal cord injury, heart failure, Parkinson’s disease etc. Immunotherapy trial using allogeneic ESC- derived DCs that are engineered to express hTERT (human telomerase reverse transcriptase) for the treatment of lung cancer is currently on-going. Banking PSCs for allogeneic use has the potential to reduce costs because one production may be used for multiple patients.

197

Despite encouraging results in cancer animal models, PSCV concept has never proceeded beyond pre-clinical phase. With the current availability of well-established quality control of PSCs and vastly improved techniques to investigate immune responses, it is about time to bring this proof-of-concept to the clinical stage in order to verify the efficacy and safety in human. In the safety perspective, one of the obstacles in transferring iPSC technology to the patients has been the potential risk of tumorigenesis caused by genetical modifications or mutations in iPSC cells. However, there is no such a concern in case of PSCVs because the PSCs used for the vaccine are lethally irradiated. More relevant concern in vaccination of PSCs is auto-immune related adverse effect like other cancer immunotherapy. PSC vaccines haven’t shown any adverse events in rodent models so far. Safety assessment of PSCVs in non-human primate may further assure the safety of this product and may smoothen the clinical transfer to the human in the future. In the efficacy perspective, needless to say, the quality control of clinical-grade PSCs should be the most important aspect. On this aspect, the methods to assure the pluripotency have been established and continuously improved with various different approaches including Immunophenotyping by flow cytometry (positive for OCT4, TRA-1-60, TRA-1-81, SSEA-3, SSEA-4, Sox2, Nanog), Embryoid body formation, Molecular-, mRNA array- or RNA-Seq-based gene expression assays (‘Pluritest’ and ‘hPSC ScoreCard’ are commercially available) (404). Furthermore, manufacturing of qualified PSCs with feeder-free cell system assures the purity in clinical-grade PSCs.

Lastly, the use of pluripotent stem cells itself for cancer therapies is receiving much attention largely because recent insights of molecular and genetic expression similarity between CSC and PSC. Pluripotent stem cell vaccination against cancer is providing new strategies in the field of cancer immunotherapy. One of the advantages of the PSC strategy is that whole PSC may share many specific tumor antigens including unknown antigens. One of the biggest problems of TNBC treatment is linked to a possible recurrence and to the chemo-resistance of the CSCs. If the CSC population can be recognized and eliminated by the immune system, ta complete remission would be thus plausible with a definitive eradication of the tumor. Therefore, this is one of the advantages by vaccine approaches only by which the immune memory in the patients can be established. Moreover, PSC-based vaccine should be relatively easy in clinical transfer from technical and production perspective because it doesn’t need to differentiate the cells. PSC vaccine can be frozen or lyophilized so that it can be ready immediately when it becomes necessary.

198

GENERAL CONCLUSION

Stem cell-like signature represents a specific character of TNBCs. Studies have shown that ES-like gene signature, CSC-like molecular pattern and TP53 mutation correlate with aggressive behavior of this disease. Those molecular characteristics involve the malignant aspects by promotion of self-renewal and invasion, inhibition of cellular differentiation and drug resistance. The similarity in the genetic expressions between grade 3 TNBCs and ESC implies the presence of the shared antigens between TNBCs and pluripotent stem cells. Since CSCs have also shown to be more resistant to the chemotherapy, its presence can cause the residual disease. However, the therapy for TNBCs is currently limited to only chemotherapy and radiotherapy, there is no targeted therapy available. Therefore, the development of CSCs targeted therapy for the treatment of TNBCs is in critical need.

In this work, we have shown the prophylactic anti-tumor effect of the pluripotent stem cell-based vaccines (PSCV) in a TNBC mouse model. PSCV immunization combined with epigenetic agent VPA significantly reduced the primary tumor volume and lung metastasis load. We used in order to analyze the metastasis sites, a 4T1-GFP-Luc tracing reporter system that have permitted to evaluate for the first time the anti-metastasis potential of the PSCV approach. In particular, we were able to observe 12.8-fold reduction of lung metastasis of mice treated with the combined protocol (PSCV+ VPA) compared to the control mice. Moreover, we have shown a significant anti-tumor effect in CSC- enriched TNBC model. Immune monitoring has demonstrated the recruitment of CD4+ and CD8+ T cells in the tumor and spleen with a reduction of MDSCs and Tregs.

Interestingly, our gene expression analysis by transcriptome and RT-qPCR have revealed a significant increase of CXCL9, CXCL10 and CXCL13 mRNA in the tumors of the mice that have received the vaccine. Chemokine ligand CXCL9 and CXCL10 are chemoattractant for CD8+ T cells, Th1 cells and NK cells (221). CXCL13 has known to have important role in attracting lymphocytes to the tumor site and formation of Tertiary lymphoid structures. Thus, this result strongly supports that, at least part of, the underlying mechanism of the anti-tumor effect is by promoting T cell trafficking to the tumor and reversing the tumor microenvironment to active “hot” state with the increased presence of immune effector cells. Further, intratumoral Th1 type chemokines have been shown to be associated with positive responses to blockage of PD-1 and PD-L1, which

199

leaves further potential of PSC vaccine strategy as the therapeutic approach in combination with CPIs.

We also showed the long duration of PSCV potency in the study with subsequent immunization followed by the tumor challenge after 6 and 9 months. Taken together, PSCVs elicit long lasting adaptive immunity, activation of the immune system and inhibition of the immune suppression which leads to anti-tumor and anti-metastasis effects.

We have observed that CSC feature of 4T1 cells were further induced in the tumor microenvironment in vivo, which emphasize that the medical strategy needs to take into account of the environmental influence to the cancer cells. Furthermore, increase of CSC marker and immune suppressive ligand on 4T1 was time dependent manner after the tumor implantation. Given that the intermediate immune suppression is more susceptible to be recovered compared to the severely suppressed status, those results also indicate that it may be important to treat breast cancers with immunotherapy at earlier stage before the tumor establish the severe immune suppression.

This work proposes PSCV to be utilized as prophylactic purpose or, prevention of recurrence after the surgery in TNBC patients and even offers a potential of therapeutic approach for TNBCs in combination with VPA or CPIs for a maximal clinical response.

200

ANNEXES

201

202

Annex 1: Other factors consists of immune tumor microenvironment

• Neutrophils • Eosinophil • Mast cell • Cancer-associated fibroblasts • Exosomes

Neutrophils

In humans, neutrophils are the most abundant immune cell population, representing 50–70% of all leukocytes (1). More than 1011 neutrophils may be produced per day, and tumors can further increase this number. G-CSF is the master regulator of neutrophil generation and differentiation (2).

In tumor microenvironment, TGF-β and G-CSF activate a tumor- and metastasis- promoting program, by regulating the TFs inhibitor of DNA binding 1 (ID1), retinoblastoma 1 (RB1) and interferon regulatory factor 8 (IRF8) that control the immunosuppressive functions of neutrophils (3). In contrary, IFN-β acts as a negative regulator of the pro-tumorigenic phenotype of neutrophils (4, 5) They also promote tumor growth by neutrophils is the induction of angiogenesis.

Antitumoral function of neutrophils has been also described. Mechanism involved with this includes presumption that is through their cytotoxic effects mediated by H2O2 (3). There are contradictory results regarding neutrophil function using the same transplantable cell lines. The timing of neutrophil depletion experiments may be crucial for the interpretation of the data, as neutrophil function evolves from antitumoral to pro- tumoral in mice bearing transplantable cancer cell lines (3). Experimental metastasis models bypass several initial steps of the metastatic cascade, including exit from the primary tumor, intravasation and priming of the premetastatic niche. The genetic loss of TGF-β receptor 2 or TGFβ signaling blockade in neutrophils decreased lung metastasis in the 4T1 mammary tumor model and activation of regulatory B cells were seen (6).

Eosinophil

Eosinophil, basophil and mast cell are known to have important roles at parasite infection and allergy. They have Fcε receptor which binds to IgE antibodies. At big

203

parasites infection, the pathogens are covered with IgE then those immune cells can attack it by directly releasing toxic granules to the target.

Tumor infiltrating eosinophils have been reported for a variety of solid tumors however, its role in the clinical cancer setting is controversial since tissue eosinophilia (also termed tumor-associated tissue eosinophils) and peripheral blood eosinophilia have been associated with both good and poor prognoses (7). As pro-tumorigenic activities of eosinophils, eosinophils facilitated the colonization of the metastatic niche. It was demonstrated that IL-5 enabled the formation of lung metastases via recruitment of eosinophils that secrete CCL22, which recruits regulatory T cells to the lungs (8).

As antitumorigenic activities of eosinophils, direct killing of tumor cells in vitro by cytotoxic mechanism have been shown while evidence demonstrating direct eosinophil- mediated tumor extermination in vivo is largely missing (7). Eosinophils have been shown to orchestrate antitumor immunity via indirect mechanisms. Using models of melanoma that were depleted of regulatory T cells (Tregs), eosinophils promoted the recruitment of CD8+ T cells, improved vascular healing, and polarized macrophages into a proinflammatory phenotype, which is usually associated with antitumorigenic properties though the promotion or suppression of CD8+ T cells by eosinophils may depend on the mediators.

Mast cells

Mast cell modulation of the immune response through the release of histamine, IL- 10, and TNF-α, contributes to tumor growth and Mast cells also secrete proteases such as MMP-2 and MMP-9, which digest ECM and, together with heparin, stimulate heparin- binding pro-angiogenic factors in the tumor microenvironment, thus influencing tumor progression and metastasis (9-11).

Cancer-associated fibroblasts

Cancer associated fibroblast (CAF) includes at least two distinct cell types. One is that the cells with similarities to the fibroblasts that create the structural foundation supporting most normal epithelial tissues and another one is the myofibroblasts, whose biological roles and properties differ from those of tissue-derived fibroblasts (12). Fibroblasts are usually quiescent and become activated in a wound healing response.

204

Character of fibroblasts, in particular those that are activated, also depend on the origin of the precursor fibroblasts. CAFs can originate from a variety of tissue types through a number of different cellular processes (Figure A). These include EMT and EndMT (Endothelial to Mesenchymal Transition) from tumor epithelial and endothelial cells, respectively. They can also derive from pericytes, adipocytes and circulating mesenchymal stem cells originating from the bone marrow. Increased levels of TGF-β in the microenvironment can also cause resident tissue fibroblasts to acquire a CAF phenotype.

Figure A: Cellular origins of cancer associated fibroblasts and their impact on cancer progression. (Adopted from (13)) CAFs can originate from a variety of tissue types through a number of different cellular processes. These include EMT and EndMT from tumor epithelia and endothelial cells, respectively. They can also derive from pericytes, adipocytes and circulating mesenchymal stem cells. Increased levels of TGF-β in the microenvironment can also cause resident tissue fibroblasts to acquire a CAF phenotype. Once present in TME, they can affect tumor growth, survival, progression and dissemination through the processes stated.

205

The majority of CAF appears to originate from the activation of resident fibroblasts and are activated through similar processes to those observed during wound healing.

However, unlike wound healing, the activated phenotype of CAFs is not reversible and is long lasting (13-15). Activated fibroblast can transdifferentiate into chondrocytes, myocytes, adipocytes and endothelial cells. Persistent emergence and accumulation of cancer cells in a given tissue represents an ongoing tissue injury, initiating a chronic wound healing response toward the cancer cells. This results in a chronic host repair response in tumors that is known as cancer fibrosis. The recruitment of stromal fibroblasts to the tumor is largely governed by the growth factors released by the cancer cells and the infiltrating immune cells.

TGF-β, PDGF and fibroblast growth factor 2 (FGF2) are key mediators of fibroblast activation in tissue damage. The recruitment of activated fibroblasts in many cancers is dependent on TGF-β. Generally, CAFs are considered to promote an immunosuppressive TME (16) though it depends on the specificity of the CAF. Secretion of cytokines, chemokines and pro-angiogenic factors by CAFs in established tumors, including, but not limited to, IL-6, IL-4, IL-8, IL-10, TNF, TGF-β, CCL2, CCL5 (also known as RANTES), CXCL9, CXCL10, SDF1, prostaglandin E2 (PGE2), nitric oxide (NO), HGF and human leukocyte antigen G (HLAG) may have direct and/or indirect implications for tumor immunity (17-19). Fibroblast-derived IL-6 also redirects monocytes toward differentiation into a macrophage lineage rather than DC differentiation and recruits and activates mast cells (20, 21).

Exosomes

Exosomes are homogenous membrane vesicles (40–150 nm diameter), derived from the exocytosis of intraluminal vesicles and released into the extracellular space when fused with the plasma membrane (22). Most cell types release exosomes through this mechanism. The precise mechanism of exosome internalization is still unclear, however, receptor-mediated endocytosis, phagocytosis, and direct plasma membrane fusion have been proposed (23). Low pH of the tumor microenvironment was shown to be essential for exosome uptake by human metastatic melanoma in vitro, that is suggested to be related to elevated stability and lipid/cholesterol content of exosomal membranes in an acidic environment (24). Immune modulation role of exosome both for immunosuppressive and immunostimulatory function have been reported.

206

The suppressor activity involves induction of the generation, expansion, and suppressive function of Tregs and the molecules expressed on exosomes like FasL, TRAIL, membrane-bound TGF-β, and galectins (25). Tumor-derived exosomes can also suppress immune-cell responses by inhibiting the cytotoxicity of CD8+ T cells and NK cells predominantly through NKG2D down-regulation by MHC class I-related chain (MIC) A (MICA) (26) and TGF-β (27) on the tumor-derived exosome surface. The inhibition of cytotoxic T-cell function has also been suggested to be induced in part by the increased T-cell ROS content mediated by melanoma-derived exosomes (28). Murine TS/A and 4T1 breast tumor cell-derived exosomes re-taken up by bone marrow myeloid cells and switched to MDSC phenotype (29). Exosomes also involve in antigen presentation. B lymphoblast-derived exosomes that bear MHC class II–peptide complexes were capable of activating human and mouse antigen specific T cell clones (30).

Exosomes derived from DCs that were pulsed with antigenic peptides induced anti- tumor CD8+ T cell responses in murine tumor models (31) via presentation of MHC– peptide complexes to naïve T lymphocytes. Also, it has been shown that exosomes were attributed to cross-priming (32), stimulation of macrophages and neutrophils to secrete pro-inflammatory mediators, including TNF-α and RANTES (33).

Various studies have shown that exosomes are composed of mRNAs, miRs, DNAs, proteins and lipid components as well (34). The delivered RNA molecules are functional, and mRNAs can be translated, while miRs target host mRNAs to modulate translation in the recipient cell (35). Oncogene miR-21 and -29a secreted from lung cancer cell-derived exosomes were shown to bind TLRs to murine (TLR7) and human (TLR8), leading to TLR-mediated NF-κB activation and secretion of pro-metastatic inflammatory cytokines TNF-α and IL-6 (36). Exosomes derived from heat-stressed ACE-positive tumor cells have enhanced immunogenicity via increased HSP70 and MHC-I (37).

207

Figure B: Tumor exosomes direct organ-specific metastasis via integrins. (Adopted from Liu and Cao. 2016) (38) Tumor-derived exosomes transport proteins, nucleic acids and lipids to specific organ and fuse with resident cells, which can prepare distant organ site as pre-metastatic niche. ITGα6β4- and ITGα6β1-expressing exosomes preferentially interact with fibroblasts and epithelial cells in lung, and ITGαvβ5-expressing exosomes preferentially fuse with Kupffer cells in liver. Once uptaken, tumor exosomes induce cellular changes in the target organ, thus promoting cancer cell colonization and organ-specific metastasis.

Exosomes also participate to create metastatic niche. Melanoma cells secrete exosomes that might induce vascular leakiness and inflammation during the formation of pre- metastatic niches (39). Similarly, MIF-containing exosomes that are released by pancreatic cancer cells increase liver metastasis by inducing TGF-β secretion, stimulating the production of the glycoprotein fibronectin in hepatic stellate cells and recruiting bone-marrow-derived cells to the liver (40). Integrins were proposed to target cancer-cell-derived exosomes to specific organs to unload their cargo and prepare the organ for the arrival of tumor cells (38, 41) (Figure B).

208

References

1. Dancey JT, Deubelbeiss KA, Harker LA, Finch CA. Neutrophil kinetics in man. J Clin Invest. 1976;58:705-15. 2. Lieschke GJ. Mice lacking granulocyte colony-stimulating factor have chronic neutropenia, granulocyte and macrophage progenitor cell deficiency, and impaired neutrophil mobilization. Blood. 1994;84:1737-46. 3. Coffelt SB, Wellenstein MD, de Visser KE. Neutrophils in cancer: neutral no more. Nat Rev Cancer. 2016;16(7):431-46. 4. Jablonska J, Leschner S, Westphal K, Lienenklaus S, Weiss S. Neutrophils responsive to endogenous IFN-β regulate tumor angiogenesis and growth in a mouse tumor model. J Clin Invest. 2010;120:1151- 64. 5. Wu CF. The lack of type I interferon induces neutrophil-mediated pre-metastatic niche formation in the mouse lung. Int J Cancer. 2015;137:837-47. 6. Bodogai M. Immunosuppressive and prometastatic functions of myeloid-derived suppressive cells rely upon education from tumor-associated B cells. Cancer Res. 2015;75:3456-65. 7. Reichman H, Karo-Atar D, Munitz A. Emerging Roles for Eosinophils in the Tumor Microenvironment. Trends Cancer. 2016;2(11):664-75. 8. Zaynagetdinov R, Sherrill TP, Gleaves LA, McLoed AG, Saxon JA, Habermann AC, et al. Interleukin- 5 facilitates lung metastasis by modulating the immune microenvironment. Cancer Res. 2015;75(8):1624-34. 9. Coussens LM, Raymond WW, Bergers G, Laig-Webster M, Behrendtsen O, Werb Z, et al. Inflammatory mast cells up-regulate angiogenesis during squamous epithelial carcinogenesis. Genes Dev. 1999;13(11):1382-97. 10. Baram D, Vaday GG, Salamon P, Drucker I, Hershkoviz R, Mekori YA. Human mast cells release metalloproteinase-9 on contact with activated T cells: juxtacrine regulation by TNF-alpha. J Immunol. 2001;167(7):4008-16. 11. Norrby K. Mast cells and angiogenesis. APMIS. 2002;110(5):355-71. 12. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144:646-74. 13. Harper J, Sainson RC. Regulation of the anti-tumour immune response by cancer-associated fibroblasts. Semin Cancer Biol. 2014;25:69-77. 14. Kalluri R, Zeisberg M. Fibroblasts in cancer. Nat Rev Cancer. 2006;6:392-401. 15. Kalluri R. The biology and function of fibroblasts in cancer. Nat Rev Cancer. 2016;16(9):582-98. 16. Raffaghello L, Dazzi F. Classification and biology of tumour associated stromal cells. Immunol Lett. 2015;168:175-82. 17. Patel R, Filer A, Barone F, Buckley CD. Stroma: fertile soil for inflammation. Best practice and

209

research. Clin Rheumatol. 2014;28:565-76. 18. Poggi A, Musso A, Dapino I, Zocchi MR. Mechanisms of tumor escape from immune system: role of mesenchymal stromal cells. Immunol Lett. 2014;159:55-72. 19. Soleymaninejadian E, Pramanik K, Samadian E. Immunomodulatory properties of mesenchymal stem cells: cytokines and factors. Am J Reprod Immunol. 2012;67:1-8. 20. Chomarat P, Banchereau J, Davoust J, Palucka AK. IL-6 switches the differentiation of monocytes from dendritic cells to macrophages. Nat Immunol. 2000;1:510-4. 21. Hugo HJ. Contribution of fibroblast and mast cell (afferent) and tumor (efferent) IL-6 effects within the tumor microenvironment. Cancer Microenviron. 2012;5:83-93. 22. Théry C, Zitvogel L, Amigorena S. Exosomes: composition, biogenesis and function. Nat Rev Immunol. 2002;2(8):569-79. 23. Théry C, Ostrowski M, Segura E. Membrane vesicles as conveyors of immune responses. Nat Rev Immunol. 2009;9(8):581-93. 24. Parolini I, Federici C, Raggi C, Lugini L, Palleschi S, De Milito A, et al. Microenvironmental pH is a key factor for exosome traffic in tumor cells. J Biol Chem. 2009;284(49):34211-22. 25. Greening DW, Gopal SK, Xu R, Simpson RJ, Chen W. Exosomes and their roles in immune regulation and cancer. Semin Cell Dev Biol. 2015;40:72-81. 26. Ashiru O, Boutet P, Fernández-Messina L, Agüera-González S, Skepper JN, Valés-Gómez M, et al. Natural killer cell cytotoxicity is suppressed by exposure to the human NKG2D ligand MICA*008 that is shed by tumor cells in exosomes. Cancer Res. 2010;70(2):481-9. 27. Szczepanski MJ, Szajnik M, Welsh A, Whiteside TL, Boyiadzis M. Blast-derived microvesicles in sera from patients with acute myeloid leukemia suppress natural killer cell function via membrane- associated transforming growth factor-beta1. Haematologica. 2011;96(9):1302-9. 28. Clayton A, Mitchell JP, Court J, Mason MD, Tabi Z. Human tumor-derived exosomes selectively impair lymphocyte responses to interleukin-2. Cancer Res. 2007;67(15):7458-66. 29. Xiang X, Poliakov A, Liu C, Liu Y, Deng ZB, Wang J, et al. Induction of myeloid-derived suppressor cells by tumor exosomes. Int J Cancer. 2009;124(11):2621-33. 30. Denzer K, Kleijmeer MJ, Heijnen HF, Stoorvogel W, Geuze HJ. Exosome: from internal vesicle of the multivesicular body to intercellular signaling device. J Cell Sci. 2000;113 Pt 19:3365-74. 31. Zitvogel L, Regnault A, Lozier A, Wolfers J, Flament C, Tenza D, et al. Eradication of established murine tumors using a novel cell-free vaccine: dendritic cell-derived exosomes. Nat Med. 1998;4(5):594-600. 32. Wolfers J, Lozier A, Raposo G, Regnault A, Théry C, Masurier C, et al. Tumor-derived exosomes are a source of shared tumor rejection antigens for CTL cross-priming. Nat Med. 2001;7(3):297-303. 33. Bhatnagar S, Shinagawa K, Castellino FJ, Schorey JS. Exosomes released from macrophages infected with intracellular pathogens stimulate a proinflammatory response in vitro and in vivo. Blood.

210

2007;110(9):3234-44. 34. Théry C. Exosomes: secreted vesicles and intercellular communications. F1000 Biol Rep. 2011;3:15. 35. Valadi H, Ekström K, Bossios A, Sjöstrand M, Lee JJ, Lötvall JO. Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells. Nat Cell Biol. 2007;9(6):654-9. 36. Valenti R, Huber V, Filipazzi P, Pilla L, Sovena G, Villa A, et al. Human tumor-released microvesicles promote the differentiation of myeloid cells with transforming growth factor-beta-mediated suppressive activity on T lymphocytes. Cancer Res. 2006;66(18):9290-8. 37. Dai S, Wan T, Wang B, Zhou X, Xiu F, Chen T, et al. More efficient induction of HLA-A*0201- restricted and carcinoembryonic antigen (CEA)-specific CTL response by immunization with exosomes prepared from heat-stressed CEA-positive tumor cells. Clin Cancer Res. 2005;11(20):7554- 63. 38. Liu Y, Cao X. Organotropic metastasis: role of tumor exosomes. Cell Res. 2016;26(2):149-50. 39. Peinado H. Melanoma exosomes educate bone marrow progenitor cells toward a pro-metastatic phenotype through MET. Nat Med. 2012;18:883-91. 40. Costa-Silva B. Pancreatic cancer exosomes initiate pre-metastatic niche formation in the liver. Nature Cell Biol. 2015;17:816-26. 41. Hoshino A. Tumour exosome integrins determine organotropic metastasis. Nature. 2015;527:329-35.

211

Annex 2: Immunological effect of conventional chemotherapeutics (Galluzzi et al., 2015(306))

Agent Setting Effect Notes transplantable murine also stimulates the expansion of Bleomycin promotes ICD CRC T REG cells associated with heat shock protein Bortezomib human MM cells promotes ICD exposure on the cell surface Carboplatin secondary to STAT6 human DCs downregulate PD-L2 Cisplatin dephosphorylation associated with the restoration of advanced cancer depletes circulating T REG cells NK and T cell functions mechanism involving the transplantable murine favors T H17 and T H1 memory translocation of some components CRC responses of the gut microbiota to SLOs transplantable murine promotes ICD, favors the expansion of NK associated with expansion of the glioma cells and depletes T REG cells macrophage compartment Cyclophospha results in PD-1/PD-L1-dependent favors the expansion of MDSCs mide transplantable murine immunosuppression lymphoma promotes ICD and favors the expansion of therapeutic effects boosted by Cd8a + DCs recombinant type I IFN enables tumor lysates and CpG- transplantable murine promotes CTL-dependent immune containing oligonucleotides to NB responses mediate therapeutic effects enables immunotherapy to mediate transplantable rat CRC depletes T REG cells complete tumor regression important for the clinical Dacarbazine melanoma induces lymphopenia management of melanoma

breast carcinoma indirectly favors immunosurveillance upon polyploidization

breast and prostate associated with a reduced activity Docetaxel increases the CTL/T REG cell ratio carcinoma of T REG cells involves different T REG cell NSCLC depletes circulating T REG cells subsets transplantable murine similar activity exhibited by other promotes ICD CRC anthracyclines, like epirubicin Doxorubicin transplantable murine results in PD-1/PD-L1-dependent favors the expansion of MDSCs lymphoma immunosuppression increases the frequency of tumor- in combination with radiation rectal cancer infiltrating CTLs therapy orthotopic murine upregulates NKG2D ligands and MHC improves NK cell recruitment pancreatic carcinoma class I on cancer cells transplantable murine favors MDSC differentiation, relieving in the context of FOLFIRI CRC immunosuppression 5-Fluorouracil results in increased IFNγ secretion depletes MDSCs transplantable murine by CTLs lymphoma promotes CTL-dependent immune in combination with a TAA-targeting responses vaccine murine MDSCs and dependent on IL-1β secretion by favors IL-17 secretion by CD4 + cells CD4 +cells MDSCs 5-Fluorouracil supports the expansion of circulating CRC in the context of FOLFIRI Irinotecan MDSCs 5-Fluorouracil CRC depletes circulating MDSCs in the context of FOLFOX Oxaliplatin

212

increases circulating CD11C + DCs and associated with an increase in CD14 + monocytes CD123 +plasmacytoid DCs pancreatic cancer no effects observed on other depletes circulating T REG cells circulating immune cell subsets orthotopic murine inhibition of TAMs synergizes with promotes TAM accumulation pancreatic carcinoma gemcitabine transplantable murine depletes circulating MDSCs restores type I IFN signaling CRC transplantable murine boost the therapeutic activity of depletes circulating MDSCs and nodal fibrosarcoma and TNF-α targeted to the tumor T REG cells osteosarcoma vessels combined with melphalan transplantable murine enables therapeutically relevant depletes circulating MDSCs lung carcinoma tumor-targeting immune responses enables DC-based vaccine to exert depletes circulating MDSCs Gemcitabine transplantable murine robust therapeutic effects lymphoma results in increased IFNγ secretion depletes MDSCs by CTLs transplantable murine results in therapeutic TAA-specific restores defective cross-presentation mesothelioma immune responses transplantable murine negatively affects CTL proliferation irrespectively, boosts the pancreatic carcinoma induced by DC-based vaccination therapeutic effects of vaccination transplantable murine accompanied by the restoration of depletes circulating MDSCs tumors NK- and T cell functions upregulates NKG2D ligands and MHC human cancer cells improves NK cell recruitment class I on cancer cells reprogram TAMs toward an important for the management of human TAMs immunostimulatory activity pancreatic carcinoma murine MDSCs and dependent on IL-1β secretion by favors IL-17 secretion by CD4 + cells CD4 +cells MDSCs Gemcitabine important for the clinical NSCLC depletes circulating T REG cells (plus cisplatin) management of NSCLC transplantable murine favors MDSC accumulation in the tumor Irinotecan in the context of FOLFIRI CRC microenvironment transplantable murine results in PD-1/PD-L1-dependent Melphalan favors the expansion of MDSCs lymphoma immunosuppression transgenic and promotes CTL-dependent immune associated with expansion of transplantable murine responses immunosuppressive plasma cells prostate cancers associated with CALR exposure on promotes ICD transplantable murine the surface of dying cells Oxaliplatin CRC increases the CTL/T REG cell ratio and in combination with recombinant IL- depleted MDSCs 12 transplantable murine promotes the activity of neutrophils and mechanism involving an intact CRC, lymphoma and macrophages commensal microbiota melanoma

indirectly favors immunosurveillance upon polyploidization

breast carcinoma favors tumor infiltration by tumor infiltration post-therapy CD68 +macrophages negatively correlates with response Paclitaxel favors tumor infiltration by NK cells and tumor infiltration post-therapy CTLs correlates with clinical response transgenic murine boosts DC maturation and cross-priming TLR4-dependent effect breast carcinoma transgenic murine results in the therapeutically depletes circulating MDSCs melanoma relevant restoration of CTL activity

213

positive effect on NK cell functions activates IFNγ-producing NK cells but Pemetrexed pancreatic cancer lost by the co-administration of depletes CD45RO + memory T cells gemcitabine Vinorelbine associated with a reduced activity NSCLC increases the CTL/T REG cell ratio (plus cisplatin) of T REG cells

214

Annex 3: Immunological effect of targeted anticancer agents (Galluzzi et al., 2015(306))

Agent Setting Effect Notes transplantable Bevacizumab depletes T REG cells mechanism involving KDR murine CRC favors the expansion of circulating correlates with molecular response to CML CTLs and NK cells chemotherapy Dasatinib depletes tumor-infiltrating T REG cells transplantable and MDSCs, promotes the expansion combined with DC-based vaccination murine melanoma of T cell repertoire transplantable increases sensitivity to CTLA4-targeting triggers a type I IFN response murine melanoma mAbs Decitabine upregulates antigen presentation and results in increased antigenicity and human cancer cells triggers type I IFN responses viral mimicry

transplantable Decitabine enables tumor eradication by PD-1- or murine breast depletes circulating MDSCs CTLA4-targeting mAbs (plus entinostat) carcinoma and CRC transplantable EGFR-targeting unclear whether other EGFR-targeting murine melanoma promotes ICD mAb mAbs have similar effects and lung carcinoma transplantable ERBB2-targeting triggers type I IFN signaling, NK cell drives an immune response based on murine breast mAb and CD8 + DC activation IFNγ-secreting CTLs carcinoma results in increased susceptibility to NK human cancer cells upregulates NKG2D ligands cell cytotoxicity Erlotinib immortalized human may favor the accumulation of MDSCs inhibits differentiation monocytes in cancer patients upregulates NKG2D ligands and MHC results in increased susceptibility to NK Gefitinib human cancer cells class I on cancer cells cell cytotoxicity favors tumor infiltration by CD4 + cells correlates with reduction in tumor size GSK2118436 melanoma and CTLs and degree of necrosis transplantable in combination with intratumoral promotes ICD murine lymphoma injection of a TLR9 agonist Ibrutinib immortalized human robustly inhibits T H2 polarization ITK-dependent effect T lymphocytes promotes the expansion of circulating persisting as long as 3 years after NK cells discontinuation CML promotes the accumulation of BCR- effect persisting after discontinuation ABL1-specific CTLs activates peripheral NK cells to correlates with clinical efficacy of produce IFNγ treatment promotes tumor infiltration by CTLs CTLs and NK cells distribute to different Imatinib and NK cells tumor areas GIST promotes an activatory crosstalk results in therapeutically relevant between DCs and NK cells secretion of IFNγ by NK cells favors the relative accumulation of M2 KIT-dependent effect TAMs favors ADCC mediated by EGFR- human cancer cells ABL1-dependent effect targeting mAb transplantable depletes circulating MDSCs and KDR-targeting at low (vasculature-normalizing), not murine breast reprograms TAMs toward an mAb high (antiangiogenic), doses carcinoma immunostimulatory phenotype

215

transgenic murine depletes TAMs and promotes tumor Lapatinib in combination with doxorubicin breast carcinoma infiltration by IFNγ-secreting CTLs human melanoma upregulates expression of MHC class I MAPK inhibitors results in increased antigenicity cells molecules MAPK inhibitors (plus melanoma increases clonality of preexisting TILs correlates with improved OS vemurafenib) human breast may exert checkpoint-blocking functions MK-2206 decreases PD-L1 expression carcinoma cells in patients depletes T REG cells and MDSCs, transgenic murine results in a generalized relieve of tumor- and upregulates CD40LG and IFNγ by melanoma induced immunosuppression PLX4720 TILs transplantable favors recruitment and activation of NK may be combined with adoptive NK cell murine melanoma cells transfer in the clinic MTORC1 is required for T REG cell C57BL/6J mice may deplete T REG cells expansion and suppressive activity human breast may exert checkpoint-blocking functions Rapamycin decreases PD-L1 expression carcinoma cells in patients metabolic effect relying on limited murine DCs expands DC lifespan production of NO favors the accumulation of peripheral melanoma correlates with improved OS CD4 +NKG2D+ T cells possible basis for the development of Sorafenib renal cell carcinoma depletes tumor-infiltrating T REG cells combinatorial regimens transplantable depletes circulating T REG cells and maximizes therapeutic activity of murine lymphoma MDSCs adoptively transferred CTLs promotes an increase in the renal cell carcinoma not seen with other therapeutic options CTL/T REG cell ratio depletes T REG cells mechanism involving KDR transplantable Sunitinib decreases levels of exhaustion results in restored IFNγ secretion by murine CRC markers among TILs tumor-infiltrating CTLs transplantable favors the recruitment of CD3 + T cells dependent on secretion of chemotactic murine melanoma to the tumor bed factors by tumor vessels transplantable improves the therapeutic effect of PD-1- Trametinib increases tumor infiltration by CTLs murine CRC or CTLA4-targeting mAbs transplantable prevents the upregulation of FASLG improves the therapeutic activity of Trichostatin A murine melanoma on TILs CTLA4-targeting mAbs transgenic murine expands IFNγ-producing CTLs and both lymphocytes populations underlie Triciribine breast carcinoma CD4 + T cells therapeutic efficacy VEGFA-targeting transplantable decreases levels of exhaustion results in restored IFNγ secretion by mAb murine CRC markers among TILs tumor-infiltrating CTLs favors tumor infiltration by CD4 + cells correlates with reduction in tumor size Vemurafenib melanoma and CTLs and degree of necrosis favors tumor infiltration by CTLs and Vemurafenib increases the clonality of preexisting correlates with improved OS (plus MAPK melanoma TILs inhibitors) promotes TAA expression and tumor associated with an increased in T cell infiltration by CTLs exhaustion markers transplantable promotes expansion of IFNγ-producing similar effects observed with Vorinostat murine lymphoma CTLs and CD4 + T cells, and panobinostat and CRC phagocytosis by DCs

216

Annex 4: CSC, stem cell or mesenchymal stem cell markers tested and those descriptions

Previously suggested Marker Description Normal expression example correlations to cancer ABCG2 ATP-binding cassette sub-family G member Placenta Breast Cancer Resistance 2 is a protein that in humans is encoded by apical membrane of the intestine, Protein, this protein the ABCG2 gene. and at the blood-testis barrier, the functions as a xenobiotic The membrane-associated protein encoded blood–brain barrier and the transporter which may play a by this gene is included in the superfamily membranes of hematopoietic role in multi-drug resistance of ATP-binding cassette (ABC) transporters. progenitor and other stem cells. to chemotherapeutic agents ABC proteins transport various molecules At the apical membranes of the including mitoxantrone and across extra- and intra-cellular membranes. liver and kidney, it enhances camptothecin analogues. excretion of xenobiotics. In the lactating mammary gland, it has a role in excreting vitamins such as riboflavin and biotin into milk. ABCB5 ABC sub-family B member 5 also known as Physiological skin ABCB5 has been suggested P-glycoprotein ABCB5 is a plasma to regulate skin progenitor membrane-spanning protein that in humans cell fusion and mediate is encoded by the ABCB5 gene. ABCB5 is chemotherapeutic drug an ABC transporter and P-glycoprotein resistance in stem–like family member. tumor cell subpopulations in human malignant melanoma. It is commonly over-expressed on circulating melanoma tumor cells. CD44 A cell-surface glycoprotein involved in cell– a large number of mammalian CD44+/CD24- expression is cell interactions, cell adhesion and cell types. commonly used as a marker migration. In humans, the CD44 antigen is for breast CSCs and is used encoded by the CD44 gene on to sort breast cancer cells Chromosome 11. into a population enriched in cells with stem-like characteristics. CD133 CD133 is a glycoprotein also known in Hematopoietic stem cells, Recent studies in brain humans and rodents as Prominin 1 endothelial progenitor cells, tumors have identified a (PROM1). glioblastoma, neuronal and glial CD133+ cell population Currently the function of CD133 is unknown. stem cells. thought to be a CSC It is a member of pentaspan population transmembrane glycoproteins (5- transmembrane, 5-TM), which specifically localize to cellular protrusions.

217

Previously suggested Marker Description Normal expression example correlations to cancer CD90 Thy-1 or CD90 (Cluster of Differentiation thymocytes (precursor of T cells circulating metastatic 90) is a 25–37 kDa heavily N-glycosylated, in the thymus) & CD34(+) melanoma cells glycophosphatidylinositol (GPI) anchored prothymocytes; neurons, conserved cell surface protein with a single mesenchymal stem cells, V-like immunoglobulin domain. Thy-1 can hematopoietic stem cells, NK be used as a marker for a variety of stem cells, murine T-cells, endothelium cells and for the axonal processes of (mainly in high endothelial mature neurons. venules or HEVs where diapedesis takes place), renal glomerular mesangial cells, follicular dendritic cells (FDC), a fraction of fibroblasts and myofibroblasts. CD105 Endoglin is a type I membrane glycoprotein Endoglin has a role in the In breast cancer, for located on cell surfaces and is part of the development of the example, the reduction of TGF beta receptor complex. It is also cardiovascular system and in the full form of endoglin, and commonly referred to as CD105, END, vascular remodeling. Its the increase of the soluble FLJ41744, HHT1, ORW and ORW1. It has expression is regulated during form of endoglin correlate a crucial role in angiogenesis, therefore, heart development. with metastasis of cancer making it an important protein for tumor cells. growth, survival and metastasis of cancer cells to other locations in the body The expression of the endoglin gene is usually low in resting endothelial cells. This, however, changes once neoangiogenesis begins and endothelial cells become active in places like tumor vessels, inflamed tissues, skin with psoriasis, vascular injury and during embryogenesis. CD166 100-105 kD type I transmembrane activated T cells, activated melanoma glycoprotein that is a member of the monocytes, epithelial cells, immunoglobulin superfamily of proteins. In fibroblasts, neurons, and also in humans it is encoded by the ALCAM gene. sweat and sebaceous glands CD146 113kDa cell adhesion molecule currently CD146 has been seen as a used as a marker for endothelial cell lineage marker for mesenchymal stem cells isolated from multiple adult and fetal organs, and its expression may be linked to multipotency A small subset of T and B lymphocytes in the peripheral blood. Sca1 [stem cell antigen-1] This murine cell Immature hematopoietic surface antigen is called also Ly6A/E or progenitor cells and, together with Ly6D. The gene encoding this antigen has other markers, defines been cloned independently as TAP [T-cell hematopoietic stem cells. activating protein].

218

Annex 5: Primer sequences used for RT-PCR

Genes Primers (5'-3') mOCT4 Forward: CTGAGGGCCAGGCAGGAGCACGAG Reverse: CTGTAGGGAGGGCTTCGGGCACTT mSOX2 Forward: GGTTACCTCTTCCTCCCACTCCAG Reverse: TCACATGTGCGACAGGGGCAG mLIN28 Forward: AGACCAACCATTTGGAGTGC Reverse: AATCGAAACCCGTGAGACAC mNanog Forward: AGGGTCTGCTACTGAGATGCTCTG Reverse: CAACCACTGGTTTTTCTGCCACCG mcMyc Forward: CAGAGGAGGAACGAGCTGAAGCGC Reverse: TTATGCACCAGAGTTTCGAAGCTGTTCG mKlf4 Forward: CACCATGGACCCGGGCGTGGCTGCCAGAAA Reverse: TTAGGCTGTTCTTTTCCGGGGCCACGA mABCG2 Forward: GCCGCTGGAATGCAAAATAG Reverse: TGTTGCTACAGACACCACAC mABCB5 Forward: CAAAGCTGTCTCTTGGAGCA Reverse: ACCTTTCAGAACCTTGGCAG mCD44 Forward: CCTTCTTTATCCGGAGCACC Reverse: GTGGTCCTGTCCTGTTCAAG mCD133 Forward: GCCCTCTATGAGATCGGAGT Reverse: CTTGGTGTTGGTGTACTGCT hOCT4 Forward: CTTCCCTCCAACCAGTTGCCCCAAAC Reverse: GCACTGTACTCCTCGGTCCCTTTCCC hSOX2 Forward: TTGCGTGAGTGTGGATGGGATTGGTG Reverse: GGGAAATGGGAGGGGTGCAAAAGAGG hNanog Forward: TGGAAGGTTCCCAGTCGGGTTCACC Reverse: CAGCCCTGATTCTTCCACCAGTCCC hcMyc Forward: CAGCGAGGATATCTGGAAGA Reverse: CTTCTCTGAGACGAGCTTGG mCD105 Forward: TTCATCGACATCAACCACAG Reverse: ATGACCTGATTGCCACACTT mCD166 Forward: GCCTTGGATGGTACACTGTC Reverse: TTTCAAGAAAGGGTGCTTTG mCD146 Forward: CCCCAGAGGAACCAACTATT Reverse: TCTGCAGGGTAGAAAACAGG Sca1 Forward: ATCCACCCTCAATGACAAGA Reverse: AGCCTCTGGGTTGAAGTTCT GAPDH Forward: AAGGAGTAAGAAACCCTGGACCAC Reverse: GAAATTGTGAGGGAGATGCTCAGT

219

Annex 6: List of the antibodies used in the experiments

Antibodies Source

Anti-Mouse CD44-APC-Vio770 Miltenyi Biotec Anti-Mouse CD24-VioBlue Miltenyi Biotec Anti-Mouse MHC Class I (H-2Kd/H-2Dd) eFluor 450 eBioscience Anti-Mouse MHC Class I (H-2Kd/H-2Dd) PerCP-eFluor eBioscience Viobility 405/520 Fixable Dye Militenyi 7-AAD Viability Staining Solution eBioscience Zombie Violet Fixable Viability Kit BioLegend Anti-Mouse CD274 (PD-L1, B7-H1) PE eBioscience Anti-Mouse CD273 (PD-L2)-PE-Vio770 Miltenyi Biotec Anti-Mouse CD45-APC-Vio770 Miltenyi Biotec Anti-Mouse CD45 PE-Cyanine7 eBioscience Anti-Mouse CD86 (B7-2) APC eBioscience Anti-Mouse CD40 PE eBioscience Anti-Mouse CD11c eFluor 450 eBioscience Anti-Mouse MHC Class II (I-A/I-E) FITC eBioscience Anti-Mouse CD83 eFluor 660 eBioscience

220

Annex 7: Supplementary data 1 (CSC marker CD44+CD24low/- expression in 4T1 cells)

Figure 49: CSC marker CD44+CD24low/- expression in 4T1 cells in vitro and in vivo (a) Comparison of in vitro (DMEM), early tumor (11-13 days) and late tumor (4 weeks) in vivo (b) Comparison of in vitro (DMEM), late tumor (4 weeks) without treatment, late tumor (4 weeks) with VPA treatment in vivo (c) Representative dot plot images

221

Annex 8: Supplementary data 2 (Dendritic cell vaccination loaded with PSC antigens)

Dendritic cell vaccination that loaded with pluripotent stem cell antigens

Dendritic cell vaccination is one of powerful strategy in cancer immunotherapy. DCs as most potent APC, can greatly impact the TME. For the effective anti-tumor immunity, ideal situation is that DCs are well stimulated by the stimulatory signals for the maturation, present the tumor antigens to CD8 CTLs and CD4 helper T cells and release cytokines, chemokines or other soluble agents that promote anti-tumor immunity. However, in TME, DCs are often immature or inactivated status under the immune suppressive environment and doesn’t react well to eliminate the tumor cells. Therefore, the aim of the DC vaccinations is to remodel the TME by powerful stimulation by the DCs and to effectively perform the antigen presentation to the T cells, leading to the tumor cell elimination. We have conducted the whole cell antigen approach to deliver the pluripotent stem cells antigens. As another delivery method of vaccination, we aimed to load the pluripotent stem cell antigens to DCs and inject to mice to evaluate anti-tumor effect. First, we began with generating DCs from mouse bone marrow for DC vaccination. Then, we planned to vaccinate mice after loading the PSC antigen to the DCs and the maturation. A potential limitation of classical DC vaccination is that the availability of high amount of DCs largely depends on the patients. In the clinic, the cancer patients often develop suppressed immunity due to the prior history of chemotherapy treatments or, impaired general condition etc. To breakthrough this limitation, generating DCs from autologous iPSC from the patients is a strong alternative way to have sufficient amount of DCs. In this aspect, we have attempted to generation of DCs from autologous iPSCs in mouse model.

Unfortunately, we haven’t completed all of the experimental steps of the studies of DC vaccination approach and haven’t yet performed the injections of DCs in mice during my thesis. However, for the possibility of future continuation in the laboratory, I precise the theoretical processes and the processes that I have already validated or achieved so far in the following chapters.

222

DC vaccination utilizing DCs generated from mouse bone marrow

Methods and Materials

Bone marrow collection from mouse

Femurs were collected from euthanized female Balb/c mice. After removing the surrounding muscles, the bones were disinfected with 70% ethanol then washed with PBS. Both bone ends were cut with surgical scissors and the marrow was flushed with RPMI- 1640 (Gibco) complete media containing FBS and penicillin streptomycin using a syringe with 21G needle. The cell solution was pipetting gently to dissociate the clumps.

DC culture

The principle method for generating DC was adapted from Lutz et al. (405) The culture media was RPMI-1640 supplemented with FBS, Penicillin streptomycin, 2 mL of glutamin and 10% FBS.

223

At day 0, BM leukocytes were seeded at 2x106 cells per 100 mm dish in 10 mL medium containing 20 ng/mL GM-CSF. At day 3 another 10 mL R10 medium containing 20 ng/mL GM-CSF were added to the plates. At days 6 and 8, half of the culture supernatant was collected, centrifuged, and the cell pellet resuspended in 10 mL fresh media containing 20 ng/mL GM-CSF, and given back into the original plate. At day 10, non-adherent cells were collected by gentle pipetting, centrifuged at 300 xg for 5 minutes.

Preparation of cell lysate

The cultured cells were trypsinized and harvested from the culture plate then the complete media was added to neutralize trypsin. Dilute the cells at 2 x 107 cells/mL in cold RPMI-1640 media. Freeze the cells rapidly in dry ice and ethanol in the safety cabinet. Once it was frozen, thaw it in RT. Repeat the freeze-thaw process total 5 times. The total cell disruption was verified under the microscope. The solution was sonicated for 1 minute at a power of about 180 watts (in rounds of 10 seconds sonication/10 seconds rest for each cycle) in the ice. Centrifuge at 10,000 x g for 20 minutes at 4°C to pellet cell debris, and then transfer the supernatant to a new tube without disturbing the pellet. Cell lysate is filtered by 0.22 μm PES filter. The protein concentration of the lysate was determined by Bradford protein assay. The lysate was stored at -80℃.

Antigen loading and maturation

The antigens for loading DCs were obtained from osmotic shock of the target cells for apoptotic cell bodies or, cell lysate solution as described below. The DCs cultured for 10 days were resuspended in media containing 20 ng/mL GM-CSF and 1ug/ml LPS or 10 ng/mL TNF-α and cultured 1-2 days. 2 x 106 cells were put in one culture plate.

Phenotyping by FACS

At the end of maturation, the supernatant of the culture was collected and washed with PBS. Antibodies were added and incubated for 30 minutes at 4℃. PBS including 2% FCS were added and measured by MACSquant.

224

To generate DCs from bone marrow (BM-DCs), we followed the protocol described by Lutz et al., (405). The method is to culture BM cells with media including GM-CSF and separate the cell population by the level of the adherence in the bacterial dish (not cell culture plate). In the late phase of the DC culture, loosely adherent or floating cells which have veils. After 9-10 days of DC culture, the DCs were replanted to the culture plate in media containing GM-CSF and LPS or TNF for complete maturation. After maturation, DCs with large veils were verified by the microscopy observation (Figure 50).

Figure 50: Representative photos of DC culture

To verify the maturation of DCs, LPS or TNF-α were added in the culture as maturation signal. We have tried two cell numbers for seeding to evaluate the difference to assure high cell number of DCs can be cultured at the same time for future in vivo experiments. MHC II, CD11c, CD86 and CD40 were used to stain DCs to measure the expressions of those markers in flow cytometry.

225

MHC II expression on the matured DCs of different culture conditions are shown in Table 20.

Positive rate of MHC II for both LPS treated and TNF treated cells showed almost 100% and even without LPS or TNF treatment, the cells were 96.4% positive. However, when the level of expression was compared between the different conditions, LPS treated cells showed slightly higher rate than that of TNF treated cells and much higher rate than that of the control when comparing expression of MHC II high (Table 20 and Figure 51). There was no significant difference between the different cell input.

Sample Positive High Low Control (before complete 96.4 67.5 32.5 maturation) LPS (2 x 106 cells/plate) 99.9 98.4 1.6 LPS (8 x 106 cells/plate) 99.8 99.4 0.6 TNF-α (2 x 106 cells/plate) 99.9 95.0 5.0 TNF-α (8 x 106 cells/plate) 99.9 95.7 4.3 Table 20: Frequency (%) of MHC II positive DCs and level of the expression

226

Figure 51: FACS dot plot images of MHC II expressions on DCs

Expressions of CD11b was slightly decreased after activation except LPS treated cells with 1 x 106 cells. LPS treated cells showed almost 100% positive for CD80 and highly positive for CD40. TNF treated cells showed more than 90% positive for CD80 and 40- 46% positive for CD40.

227

Figure 52: CD11c, CD86, CD40 expression on MHC II positive cells

4.1 DC vaccination utilizing the DCs generated from iPSCs

The experimental model of vaccination using DCs derived from autologous iPSCs started from the reprogramming of fibroblasts from Balb/c mouse to iPSCs. As described in the previous chapter, this step has been successfully completed (Figure XX).

Figure 53: Progress of the experimental steps for iPSC-DCs during my thesis

228

The next step was to differentiate iPSCs to DCs. There are several protocols of differentiation of ESC or iPSC to DCs. The summary of the process is described as below. Among these protocols, we decided to try the co-culture method of iPSCs with OP9 cells proposed by Senju (406).

1. Fairchild (2000) (355) (407) ES-DCs • Feeder free culture of ESC (on gelatin + mLIF) • Formation of EB for 14 days (Petri dish without mLIF) • Transfer to cell culture plate (DMEM + FBS + GM-CSF + IL-3 for 2weeks 2. Senju (2003) (354) ES-DCs • Suspend ESC in α-MEM+20%FBS on OP9 cell layered plate • Day 3: Change media • Day 5: Replate on OP9 cells • Day 10: Transfer the cells to Petri dish without feeder cells in RPMI1640 + FBS + GM-CSF + 2ME) • Day 17-19: Harvest ES-DCs (when more than half of cells become adherent after Day 12: Transfer the floating cells to fresh dishes.) 3. Senju (2009) (406) iPS-DCs • Suspend ESC in α-MEM+FBS on OP9 cell layered plate • Day 6-7: Replate to α-MEM+FBS+GM-CSF+2-ME on freshly prepared OP9 cells • Day 12-13: Collect the floating cells by pipetting. • Transfer the cells to Petri dish without feeder cells in RPMI1640 + FBS + GM-CSF + 2ME • Culture for 10 – 14 days

After plating OP9 cells on gelatin coated culture plate, OP9 cells were cultured for 3 days to have confluent layers or, 7 days (with changing the media on day 4 after seeding) to have overconfluent layers to evaluate the impact to the differentiation. After 3 or 7 days, B9D7 iPSCs were plated on the OP9 cell layers. At 7 days after the seeding of iPSCs on OP9 cells, iPSC-derived colonies appeared. The morphology of the colonies was heterogeneous, some of them looked flat and epithelial like and some of them looks small round cell-based colony. At 7 days after the seeding of iPSCs on OP9 cells, the cells were trypsinized and replated onto the fresh OP9 layers. After the replating, the cell population which resemble iPSC derived cells formed again the similar colonies like those before replating however, floating cells or loosely adherent cells that are considered to differentiate to DCs didn’t appear.

229

As nest step, to investigate the cause that we didn’t observe DC like cells after the culture and to try the alternative protocol like that by Fairchild should find the solution to differentiate the iPSC to DCs. The possible factors in culture which may impact the differentiation consequence are as follows.

▪ It is known that the quality and culture history of OP9 cells can have impact to the co-culture differentiation process. During the culture, we observed emergence of some adipocytic cells in the OP9 culture. It may be important to find the optimized cell density which doesn’t allow emergence of too much adipocytic cells as well as to support iPSC differentiation to hematopoietic lineage.

230

▪ To validate the differentiation method using ESC before using iPSC clone should eliminate the potential issues in case that the differentiation problem came from the specific iPS clone ▪ Using the lineage markers to ensure differentiation progress in each culture phase should detect the culture point that may interfere with the differentiation. For example, decrease or loss of pluripotent stem cell marker Nanog and increase of Flk-1 receptor tyrosine kinase, as hematopoietic prosecutor marker, and CD45 as hematopoietic cell marker can be monitored before and after the replating of iPSC/OP9 co-culture.

231

Annex 9: Résumé substantiel Résumé substantiel

La plupart des cancers du sein triple négatif (CSTN) sont de type basal, agressives, et de haut grade. Ils sont le plus souvent à un grade III au diagnostic et possèdent le plus souvent une signature génique de type mésenchymateux ou de type embryonnaire. Les CSTN sont difficiles à éradiquer due en partie à la présence au sein de ces tumeurs d’une rare population appelée « cellules souches cancéreuses (CSC) », présentant une capacité d'auto-renouvellement, une propriété qu’elles partagent avec les cellules souches dites pluripotentes. Il a été montré que les cellules souches pluripotentes (CSP) et les CSC partagent plusieurs centaines de déterminants antigéniques et plusieurs équipes ont montré qu’il était possible d’utiliser des CSP comme agent vaccinal pour induire une réponse immunitaire cellulaire et/ou humorale dirigée contre les CSC. Ces travaux ont étés réalisés sur des modèles murins de cancer du poumon, du colon et de l’ovaire.

Mon travail de thèse a consisté à mettre en place un modèle de CSTN métastatique chez la souris (syngénique Balb/c) et d'évaluer l'effet anti-tumorale d’un vaccin à base de cellules souches pluripotentes (VCSP) sur ce modèle. Le modèle murin immunocompétent 4T1 a été utilisé présentant un modèle proche du CSTN humain métastatique. En lien avec cette recherche, sont discutés dans cette thèse : la description du microenvironnement tumoral, la relation entre le cancer et les cellules souches, et les mécanismes cellulaires et moléculaires de l’immunosuppression tumoral dans le cadre des CSTN. La connaissance des mécanismes de l’immunosurveillance est un élément majeur à comprendre avant de proposer toutes nouvelles approches d’immunothérapies.

Nous avons démontré que des vaccins à base de CSP en combinaison avec un inhibiteur des histones désacétylases a permis de prévenir l’établissement de tumeurs 4T1 grâce au développement d’une forte réponse immunitaire antitumorale. Nous montrons que cet effet anti-tumorale est associé à une réduction s des lymphocytes T régulateurs et des cellules myéloïdes suppressives, et à une augmentation des cellules T CD8 + cytotoxiques dans la tumeur et la rate. De plus, la réponse anti-tumorale est associée à une réduction drastique de la dissémination métastatique et à une amélioration du taux de survie dans ce modèle de cancer du sein 4T1, Cette stratégie d'immunothérapie active a également été efficace pour inhiber l'établissement de CSCs issues de cellules 4T1, en induisant une modification majeure du microenvironnement tumoral.

232

Par une analyse de l'expression des gènes par transcriptome et RT-qPCR nous avons pu montrer la présence d’une augmentation significative des transcrits de CXCL9, CXCL10 et CXCL13 dans les tumeurs des souris ayant reçu le traitement combiné. Le ligand de chimiokine CXCL9 et CXCL10 sont des agents chimioattractifs pour les cellules T CD8 +, les cellules Th1 et les cellules NK. CXCL13 joue un rôle important dans l’attraction des lymphocytes vers le site tumoral et la formation de structures lymphoïdes tertiaires. Ces résultats révèlent qu’une partie du mécanisme sous-jacent de l'effet antitumoral observé a été de favoriser dans le microenvironnement tumoral l’attraction de cellules T vers la tumeur, avec une modification du microenvironnement tumoral en favorisant un état « chaud » actif avec la présence accrue de cellules immunitaires effectrices.

Nous avons également montré que le vaccin candidat permet d’avoir des effets à long terme en inhibant l’établissement de la tumeur 4T1, 6 et 9 mois après une vaccination par des VCSP, montrant la possibilité d’induire une immunité adaptative de longue durée avec l'activation du système immunitaire et l'inhibition des effecteurs suppresseurs.

Nous avons également observé après l’implantation des cellules 4T1 l’émergence qu’une petite population de CSC au sein des tumeurs in vivo ce qui souligne que la stratégie médicale doit prendre en compte l'influence du microenvironnement sur les cellules cancéreuses. De plus, l’augmentation des CSC au sein des tumeurs est dépendante du temps après l’implantation de la tumeur soulignant l’importance d’intervenir le plus tôt possible pour éradiquer une possible présence de CSC au sein des tumeurs avant même l’établissement d’une immunosuppression sévère et irréversible.

Ce travail propose que le VCSP soit utilisé dans un but prophylactique, ou comme prévention des récidives après chirurgie sur des patients atteints de CSTN.

233

REFERENCES

1. Tiede B, Kang Y. From milk to malignancy: the role of mammary stem cells in development, pregnancy and breast cancer. Cell Res. 2011;21(2):245-57. 2. Dimri G, Band H, Band V. Mammary epithelial cell transformation: insights from cell culture and mouse models. Breast Cancer Res. 2005;7(4):171-9. 3. Polyak K. Breast cancer: origins and evolution. J Clin Invest. 2007;117(11):3155-63. 4. Keller PJ, Lin AF, Arendt LM, Klebba I, Jones AD, Rudnick JA, et al. Mapping the cellular and molecular heterogeneity of normal and malignant breast tissues and cultured cell lines. Breast Cancer Res. 2010;12(5):R87. 5. Jéhannin-Ligier K DE, Bossard N, Molinié F, Defossez G, Daubisse-Marliac L, Delafosse P, Remontet L, Uhry Z. Projection de l’incidence et de la mortalité par cancer en France métropolitaine en 2017. 2018. 6. Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer. 2015;136(5):E359-86. 7. Perou CM, Sørlie T, Eisen MB, van de Rijn M, Jeffrey SS, Rees CA, et al. Molecular portraits of human breast tumours. Nature. 2000;406(6797):747-52. 8. Sørlie T, Perou CM, Tibshirani R, Aas T, Geisler S, Johnsen H, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci U S A. 2001;98(19):10869-74. 9. Prat A, Parker JS, Karginova O, Fan C, Livasy C, Herschkowitz JI, et al. Phenotypic and molecular characterization of the claudin-low intrinsic subtype of breast cancer. Breast Cancer Res. 2010;12(5):R68. 10. Bors W, Michel C, Saran M, Lengfelder E. Kinetic investigations of the autoxidation of adrenalin. Z Naturforsch C. 1978;33(11-12):891-6. 11. Foulkes WD, Smith IE, Reis-Filho JS. Triple-negative breast cancer. N Engl J Med. 2010;363(20):1938-48. 12. Sorlie T, Tibshirani R, Parker J, Hastie T, Marron JS, Nobel A, et al. Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc Natl Acad Sci U S A. 2003;100(14):8418-23. 13. Prat A, Adamo B, Cheang MC, Anders CK, Carey LA, Perou CM. Molecular characterization of basal-like and non-basal-like triple-negative breast cancer. Oncologist. 2013;18(2):123-33. 14. Prat A, Perou CM. Deconstructing the molecular portraits of breast cancer. Mol Oncol. 2011;5(1):5-23.

234

15. Turner N, Tutt A, Ashworth A. Hallmarks of 'BRCAness' in sporadic cancers. Nat Rev Cancer. 2004;4(10):814-9. 16. Lehmann BD, Bauer JA, Chen X, Sanders ME, Chakravarthy AB, Shyr Y, et al. Identification of human triple-negative breast cancer subtypes and preclinical models for selection of targeted therapies. J Clin Invest. 2011;121(7):2750-67. 17. Alexandrov LB. Signatures of mutational processes in human cancer. Nature. 2013;500:415-21. 18. Loi S, Sirtaine N, Piette F, Salgado R, Viale G, Van Eenoo F, et al. Prognostic and predictive value of tumor-infiltrating lymphocytes in a phase III randomized adjuvant breast cancer trial in node-positive breast cancer comparing the addition of docetaxel to doxorubicin with doxorubicin-based chemotherapy: BIG 02-98. J Clin Oncol. 2013;31(7):860-7. 19. Mittendorf EA, Philips AV, Meric-Bernstam F, Qiao N, Wu Y, Harrington S, et al. PD-L1 expression in triple-negative breast cancer. Cancer Immunol Res. 2014;2(4):361-70. 20. Pusztai L, Karn T, Safonov A, Abu-Khalaf MM, Bianchini G. New Strategies in Breast Cancer: Immunotherapy. Clin Cancer Res. 2016;22(9):2105-10. 21. Luen S, Virassamy B, Savas P, Salgado R, Loi S. The genomic landscape of breast cancer and its interaction with host immunity. Breast. 2016;29:241-50. 22. Disis ML, Stanton SE. Immunotherapy in breast cancer: An introduction. Breast. 2018;37:196-9. 23. Rody A, Karn T, Liedtke C, Pusztai L, Ruckhaeberle E, Hanker L, et al. A clinically relevant gene signature in triple negative and basal-like breast cancer. Breast Cancer Res. 2011;13(5):R97. 24. Denkert C, von Minckwitz G, Brase JC, Sinn BV, Gade S, Kronenwett R, et al. Tumor-infiltrating lymphocytes and response to neoadjuvant chemotherapy with or without carboplatin in human epidermal growth factor receptor 2-positive and triple-negative primary breast cancers. J Clin Oncol. 2015;33(9):983-91. 25. Salgado R, Denkert C, Demaria S, Sirtaine N, Klauschen F, Pruneri G, et al. The evaluation of tumor-infiltrating lymphocytes (TILs) in breast cancer: recommendations by an International TILs Working Group 2014. Ann Oncol. 2015;26(2):259-71. 26. Fridman WH, Galon J, Dieu-Nosjean MC, Cremer I, Fisson S, Damotte D, et al. Immune infiltration in human cancer: prognostic significance and disease control. Curr Top Microbiol Immunol. 2011;344:1-24. 27. Engelhard VH, Rodriguez AB, Mauldin IS, Woods AN, Peske JD, Slingluff CL. Immune Cell Infiltration and Tertiary Lymphoid Structures as Determinants of Antitumor Immunity. J Immunol. 2018;200(2):432-42. 28. Adams JM, Strasser A. Is tumor growth sustained by rare cancer stem cells or

235

dominant clones? Cancer Res. 2008;68(11):4018-21. 29. MAKINO S. Further evidence favoring the concept of the stem cell in ascites tumors of rats. Ann N Y Acad Sci. 1956;63(5):818-30. 30. ISHIBASHI K. Studies on the number of cells necessary for the transplantation of Yoshida sarcoma; transmission of the tumor with a single cell. Gan. 1950;41(1):1-14. 31. Nowell PC. The clonal evolution of tumor cell populations. Science. 1976;194:23-8. 32. Lapidot T, Sirard C, Vormoor J, Murdoch B, Hoang T, Caceres-Cortes J, et al. A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature. 1994;367(6464):645-8. 33. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF. Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A. 2003;100(7):3983- 8. 34. Schoenhals M, Kassambara A, De Vos J, Hose D, Moreaux J, Klein B. Embryonic stem cell markers expression in cancers. Biochem Biophys Res Commun. 2009;383(2):157-62. 35. Ben-Porath I, Thomson MW, Carey VJ, Ge R, Bell GW, Regev A, et al. An embryonic stem cell-like gene expression signature in poorly differentiated aggressive human tumors. Nat Genet. 2008;40(5):499-507. 36. Wong DJ, Liu H, Ridky TW, Cassarino D, Segal E, Chang HY. Module map of stem cell genes guides creation of epithelial cancer stem cells. Cell Stem Cell. 2008;2(4):333-44. 37. Kim J, Woo AJ, Chu J, Snow JW, Fujiwara Y, Kim CG, et al. A Myc network accounts for similarities between embryonic stem and cancer cell transcription programs. Cell. 2010;143(2):313-24. 38. Shats I, Gatza ML, Chang JT, Mori S, Wang J, Rich J, et al. Using a stem cell-based signature to guide therapeutic selection in cancer. Cancer Res. 2011;71(5):1772-80. 39. Kawamura T, Suzuki J, Wang YV, Menendez S, Morera LB, Raya A, et al. Linking the p53 tumour suppressor pathway to somatic cell reprogramming. Nature. 2009;460(7259):1140-4. 40. Blum B, Benvenisty N. The tumorigenicity of diploid and aneuploid human pluripotent stem cells. Cell Cycle. 2009;8(23):3822-30. 41. Mizuno H, Spike BT, Wahl GM, Levine AJ. Inactivation of p53 in breast cancers correlates with stem cell transcriptional signatures. Proc Natl Acad Sci U S A. 2010;107(52):22745-50. 42. Network CGA. Comprehensive molecular portraits of human breast tumours. Nature. 2012;490(7418):61-70. 43. Hu T, Liu S, Breiter DR, Wang F, Tang Y, Sun S. Octamer 4 small interfering RNA results in cancer stem cell-like cell apoptosis. Cancer Res. 2008;68(16):6533-40.

236

44. Beltran AS, Rivenbark AG, Richardson BT, Yuan X, Quian H, Hunt JP, et al. Generation of tumor-initiating cells by exogenous delivery of OCT4 transcription factor. Breast Cancer Res. 2011;13(5):R94. 45. Leis O, Eguiara A, Lopez-Arribillaga E, Alberdi MJ, Hernandez-Garcia S, Elorriaga K, et al. Sox2 expression in breast tumours and activation in breast cancer stem cells. Oncogene. 2012;31(11):1354-65. 46. Nagata T, Shimada Y, Sekine S, Hori R, Matsui K, Okumura T, et al. Prognostic significance of NANOG and KLF4 for breast cancer. Breast Cancer. 2014;21(1):96-101. 47. Han J, Zhang F, Yu M, Zhao P, Ji W, Zhang H, et al. RNA interference-mediated silencing of NANOG reduces cell proliferation and induces G0/G1 cell cycle arrest in breast cancer cells. Cancer Lett. 2012;321(1):80-8. 48. Guen VJ, Chavarria TE, Kröger C, Ye X, Weinberg RA, Lees JA. EMT programs promote basal mammary stem cell and tumor-initiating cell stemness by inducing primary ciliogenesis and Hedgehog signaling. Proc Natl Acad Sci U S A. 2017;114(49):E10532-E9. 49. Liu S, Dontu G, Mantle ID, Patel S, Ahn NS, Jackson KW, et al. Hedgehog signaling and Bmi-1 regulate self-renewal of normal and malignant human mammary stem cells. Cancer Res. 2006;66(12):6063-71. 50. Michaelson JS, Leder P. beta-catenin is a downstream effector of Wnt-mediated tumorigenesis in the mammary gland. Oncogene. 2001;20(37):5093-9. 51. Tsukamoto AS, Grosschedl R, Guzman RC, Parslow T, Varmus HE. Expression of the int-1 gene in transgenic mice is associated with mammary gland hyperplasia and adenocarcinomas in male and female mice. Cell. 1988;55(4):619-25. 52. Harrison H, Farnie G, Howell SJ, Rock RE, Stylianou S, Brennan KR, et al. Regulation of breast cancer stem cell activity by signaling through the Notch4 receptor. Cancer Res. 2010;70(2):709-18. 53. Bhola NE, Balko JM, Dugger TC, Kuba MG, Sánchez V, Sanders M, et al. TGF-β inhibition enhances chemotherapy action against triple-negative breast cancer. J Clin Invest. 2013;123(3):1348-58. 54. Shimono Y, Zabala M, Cho RW, Lobo N, Dalerba P, Qian D, et al. Downregulation of miRNA-200c links breast cancer stem cells with normal stem cells. Cell. 2009;138(3):592-603. 55. Yu F. let-7 regulates self renewal and tumorigenicity of breast cancer cells. Cell. 2007;131:1109-23. 56. Viswanathan SR, Daley GQ, Gregory RI. Selective blockade of microRNA processing by Lin28. Science. 2008;320(5872):97-100. 57. Iliopoulos D, Hirsch HA, Struhl K. An epigenetic switch involving NF-kappaB, Lin28, Let-7 MicroRNA, and IL6 links inflammation to cell transformation. Cell.

237

2009;139(4):693-706. 58. Viswanathan SR, Powers JT, Einhorn W, Hoshida Y, Ng TL, Toffanin S, et al. Lin28 promotes transformation and is associated with advanced human malignancies. Nat Genet. 2009;41(7):843-8. 59. Ricardo S, Vieira AF, Gerhard R, Leitão D, Pinto R, Cameselle-Teijeiro JF, et al. Breast cancer stem cell markers CD44, CD24 and ALDH1: expression distribution within intrinsic molecular subtype. J Clin Pathol. 2011;64(11):937-46. 60. Park SY, Lee HE, Li H, Shipitsin M, Gelman R, Polyak K. Heterogeneity for stem cell-related markers according to tumor subtype and histologic stage in breast cancer. Clin Cancer Res. 2010;16(3):876-87. 61. Tanaka H, Nakamura M, Kameda C, Kubo M, Sato N, Kuroki S, et al. The Hedgehog signaling pathway plays an essential role in maintaining the CD44+CD24-/low subpopulation and the side population of breast cancer cells. Anticancer Res. 2009;29(6):2147-57. 62. Bogachek MV, Park JM, De Andrade JP, Lorenzen AW, Kulak MV, White JR, et al. Inhibiting the SUMO Pathway Represses the Cancer Stem Cell Population in Breast and Colorectal Carcinomas. Stem Cell Reports. 2016;7(6):1140-51. 63. Ginestier C. ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome. Cell Stem Cell. 2007;1:555-67. 64. Bi X, Wu C, Han M, Cai J. Correlations of ALDH1 expression with molecular subtypes and ABCG2 in breast cancer. Gland Surg. 2012;1(1):12-9. 65. Morimoto K, Kim SJ, Tanei T, Shimazu K, Tanji Y, Taguchi T, et al. Stem cell marker aldehyde dehydrogenase 1-positive breast cancers are characterized by negative estrogen receptor, positive human epidermal growth factor receptor type 2, and high Ki67 expression. Cancer Sci. 2009;100(6):1062-8. 66. Allred DC, Wu Y, Mao S, Nagtegaal ID, Lee S, Perou CM, et al. Ductal carcinoma in situ and the emergence of diversity during breast cancer evolution. Clin Cancer Res. 2008;14(2):370-8. 67. Kim C, Gao R, Sei E, Brandt R, Hartman J, Hatschek T, et al. Chemoresistance Evolution in Triple-Negative Breast Cancer Delineated by Single-Cell Sequencing. Cell. 2018;173(4):879-93.e13. 68. Yates LR, Knappskog S, Wedge D, Farmery JHR, Gonzalez S, Martincorena I, et al. Genomic Evolution of Breast Cancer Metastasis and Relapse. Cancer Cell. 2017;32(2):169- 84.e7. 69. Lebert JM, Lester R, Powell E, Seal M, McCarthy J. Advances in the systemic treatment of triple-negative breast cancer. Curr Oncol. 2018;25(Suppl 1):S142-S50.

238

70. Senkus E, Kyriakides S, Ohno S, Penault-Llorca F, Poortmans P, Rutgers E, et al. Primary breast cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2015;26 Suppl 5:v8-30. 71. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell. 2000;100(1):57-70. 72. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144:646-74. 73. Quail DF, Joyce JA. Microenvironmental regulation of tumor progression and metastasis. Nat Med. 2013;19:1423-37. 74. Fridman WH, Pages F, Sautes-Fridman C, Galon J. The immune contexture in human tumours: impact on clinical outcome. Nat Rev Cancer. 2012;12:298-306. 75. Gu-Trantien C, Loi S, Garaud S, Equeter C, Libin M, de Wind A, et al. CD4⁺ follicular helper T cell infiltration predicts breast cancer survival. J Clin Invest. 2013;123(7):2873-92. 76. Mahmoud SM, Paish EC, Powe DG, Macmillan RD, Grainge MJ, Lee AH, et al. Tumor-infiltrating CD8+ lymphocytes predict clinical outcome in breast cancer. J Clin Oncol. 2011;29(15):1949-55. 77. Mahmoud S, Lee A, Ellis I, Green A. CD8(+) T lymphocytes infiltrating breast cancer: A promising new prognostic marker? Oncoimmunology. 2012;1(3):364-5. 78. Bos R, Marquardt KL, Cheung J, Sherman LA. Functional differences between low- and high-affinity CD8(+) T cells in the tumor environment. Oncoimmunology. 2012;1(8):1239- 47. 79. Matzinger P. Tolerance, danger, and the extended family. Annu Rev Immunol. 1994;12:991-1045. 80. Coussens LM, Werb Z. Inflammation and cancer. Nature. 2002;420(6917):860-7. 81. Lemaitre B, Nicolas E, Michaut L, Reichhart JM, Hoffmann JA. Pillars article: the dorsoventral regulatory gene cassette spätzle/Toll/cactus controls the potent antifungal response in Drosophila adults. Cell. 1996. 86: 973-983. J Immunol. 2012;188(11):5210-20. 82. Kärre K, Ljunggren HG, Piontek G, Kiessling R. Selective rejection of H-2-deficient lymphoma variants suggests alternative immune defence strategy. Nature. 1986;319(6055):675-8. 83. Martinez FO, Gordon S, Locati M, Mantovani A. Transcriptional profiling of the human monocyte-to-macrophage differentiation and polarization: new molecules and patterns of gene expression. J Immunol. 2006;177:7303-11. 84. Biswas SK, Mantovani A. Macrophage plasticity and interaction with lymphocyte subsets: cancer as a paradigm. Nat Immunol. 2010;11:889-96. 85. Yeo EJ, Cassetta L, Qian BZ, Lewkowich I, Li JF, Stefater JA, et al. Myeloid WNT7b

239

mediates the angiogenic switch and metastasis in breast cancer. Cancer Res. 2014;74(11):2962-73. 86. Gocheva V, Wang HW, Gadea BB, Shree T, Hunter KE, Garfall AL, et al. IL-4 induces cathepsin protease activity in tumor-associated macrophages to promote cancer growth and invasion. Genes Dev. 2010;24(3):241-55. 87. Yan D, Wang HW, Bowman RL, Joyce JA. STAT3 and STAT6 Signaling Pathways Synergize to Promote Cathepsin Secretion from Macrophages via IRE1α Activation. Cell Rep. 2016;16(11):2914-27. 88. Liu CY, Xu JY, Shi XY, Huang W, Ruan TY, Xie P, et al. M2-polarized tumor- associated macrophages promoted epithelial-mesenchymal transition in pancreatic cancer cells, partially through TLR4/IL-10 signaling pathway. Lab Invest. 2013;93(7):844-54. 89. Fan QM, Jing YY, Yu GF, Kou XR, Ye F, Gao L, et al. Tumor-associated macrophages promote cancer stem cell-like properties via transforming growth factor-beta1-induced epithelial-mesenchymal transition in hepatocellular carcinoma. Cancer Lett. 2014;352(2):160-8. 90. Sato K, Fujita S. Dendritic cells: nature and classification. Allergol Int. 2007;56(3):183-91. 91. Tran Janco JM, Lamichhane P, Karyampudi L, Knutson KL. Tumor-infiltrating dendritic cells in cancer pathogenesis. J Immunol. 2015;194(7):2985-91. 92. Krempski J, Karyampudi L, Behrens MD, Erskine CL, Hartmann L, Dong H, et al. Tumor-infiltrating programmed death receptor-1+ dendritic cells mediate immune suppression in ovarian cancer. J Immunol. 2011;186(12):6905-13. 93. Chiba S, Baghdadi M, Akiba H, Yoshiyama H, Kinoshita I, Dosaka-Akita H, et al. Tumor-infiltrating DCs suppress nucleic acid-mediated innate immune responses through interactions between the receptor TIM-3 and the alarmin HMGB1. Nat Immunol. 2012;13(9):832-42. 94. Kusmartsev S, Nefedova Y, Yoder D, Gabrilovich DI. Antigen-specific inhibition of CD8+ T cell response by immature myeloid cells in cancer is mediated by reactive oxygen species. J Immunol. 2004;172(2):989-99. 95. Sonnenfeld A. Leukamische reaktiones bei carcinoma. Zeitschrift F Klin Med. 1929;111:108. 96. Robinson WA. Granulocytosis in Neoplasia. Ann Ny Acad Sci. 1965;230:212-8. 97. Bennett JA, Rao VS, Mitchell MS. Systemic bacillus Calmette-Guerin (BCG) activates natural suppressor cells. Proc Natl Acad Sci Usa. 1978;75:5142-4. 98. Gabrilovich DI, Ostrand-Rosenberg S, Bronte V. Coordinated regulation of myeloid cells by tumours. Nat Rev Immunol. 2012;12:253-68.

240

99. Talmadge JE, Gabrilovich DI. History of myeloid-derived suppressor cells. Nat Rev Cancer. 2013;13:739-52. 100. Youn JI, Nagaraj S, Collazo M, Gabrilovich DI. Subsets of myeloid-derived suppressor cells in tumor-bearing mice. J Immunol. 2008;181:5791-802. 101. Karsunky H, Merad M, Cozzio A, Weissman IL, Manz MG. Flt3 ligand regulates dendritic cell development from Flt3+ lymphoid and myeloid-committed progenitors to Flt3+ dendritic cells in vivo. J Exp Med. 2003;198:305-13. 102. Movahedi K, Laoui D, Gysemans C, Baeten M, Stangé G, Van den Bossche J, et al. Different tumor microenvironments contain functionally distinct subsets of macrophages derived from Ly6C(high) monocytes. Cancer Res. 2010;70(14):5728-39. 103. Gabrilovich DI, Nagaraj S. Myeloid-derived suppressor cells as regulators of the immune system. Nature Rev Immunol. 2009;9:162-74. 104. Sawanobori Y. Chemokine-mediated rapid turnover of myeloid-derived suppressor cells in tumor-bearing mice. Blood. 2008;111:5457-66. 105. Almand B. Increased production of immature myeloid cells in cancer patients: a mechanism of immunosuppression in cancer. J Immunol. 2001;166:678-89. 106. Poschke I, Mougiakakos D, Hansson J, Masucci GV, Kiessling R. Immature immunosuppressive CD14+HLA-DR-/low cells in melanoma patients are Stat3hi and overexpress CD80, CD83, and DC-sign. Cancer Res. 2010;70:4335-45. 107. Younos IH, Dafferner AJ, Gulen D, Britton HC, Talmadge JE. Tumor regulation of myeloid-derived suppressor cell proliferation and trafficking. Int Immunopharmacol. 2012;13:245-56. 108. Serafini P. High-dose granulocyte-macrophage colony-stimulating factor-producing vaccines impair the immune response through the recruitment of myeloid suppressor cells. Cancer Res. 2004;64:6337-43. 109. Rodriguez PC. Arginase I production in the tumor microenvironment by mature myeloid cells inhibits T-cell receptor expression and antigen-specific T-cell responses. Cancer Res. 2004;64:5839-49. 110. Srivastava MK. Myeloid suppressor cell depletion augments antitumor activity in lung cancer. P Lo S One. 2012;7:e40677. 111. Schmielau J, Finn OJ. Activated granulocytes and granulocyte-derived hydrogen peroxide are the underlying mechanism of suppression of t-cell function in advanced cancer patients. Cancer Res. 2001;61(12):4756-60. 112. Hanson EM, Clements VK, Sinha P, Ilkovitch D, Ostrand-Rosenberg S. Myeloid- derived suppressor cells down-regulate L-selectin expression on CD4+ and CD8+ T cells. J Immunol. 2009;183(2):937-44.

241

113. Sakuishi K, Jayaraman P, Behar SM, Anderson AC, Kuchroo VK. Emerging Tim-3 functions in antimicrobial and tumor immunity. Trends Immunol. 2011;32(8):345-9. 114. Li H, Han Y, Guo Q, Zhang M, Cao X. Cancer-Expanded Myeloid-Derived Suppressor Cells Induce Anergy of NK Cells through Membrane-Bound TGF-{beta}1. J Immunol. 2009;182:240-9. 115. Yu J, Du W, Yan F, Wang Y, Li H, Cao S, et al. Myeloid-derived suppressor cells suppress antitumor immune responses through IDO expression and correlate with lymph node metastasis in patients with breast cancer. J Immunol. 2013;190(7):3783-97. 116. Condamine T, Ramachandran I, Youn JI, Gabrilovich DI. Regulation of tumor metastasis by myeloid-derived suppressor cells. Annu Rev Med. 2015;66:97-110. 117. Yang L, Huang J, Ren X, Gorska AE, Chytil A, Aakre M, et al. Abrogation of TGF beta signaling in mammary carcinomas recruits Gr-1+CD11b+ myeloid cells that promote metastasis. Cancer Cell. 2008;13(1):23-35. 118. Toh B, Wang X, Keeble J, Sim WJ, Khoo K, Wong WC, et al. Mesenchymal transition and dissemination of cancer cells is driven by myeloid-derived suppressor cells infiltrating the primary tumor. PLoS Biol. 2011;9(9):e1001162. 119. Cui TX, Kryczek I, Zhao L, Zhao E, Kuick R, Roh MH, et al. Myeloid-derived suppressor cells enhance stemness of cancer cells by inducing microRNA101 and suppressing the corepressor CtBP2. Immunity. 2013;39(3):611-21. 120. Shirota Y, Shirota H, Klinman DM. Intratumoral injection of CpG oligonucleotides induces the differentiation and reduces the immunosuppressive activity of myeloid-derived suppressor cells. J Immunol. 2012;188:1592-9. 121. Bryceson YT, March ME, Ljunggren HG, Long EO. Synergy among receptors on resting NK cells for the activation of natural cytotoxicity and cytokine secretion. Blood. 2006;107(1):159-66. 122. Lanier LL. NK cell recognition. Annu Rev Immunol. 2005;23:225-74. 123. Chiossone L, Dumas PY, Vienne M, Vivier E. Natural killer cells and other innate lymphoid cells in cancer. Nat Rev Immunol. 2018. 124. Diefenbach A, Jamieson AM, Liu SD, Shastri N, Raulet DH. Ligands for the murine NKG2D receptor: expression by tumor cells and activation of NK cells and macrophages. Nat Immunol. 2000;1(2):119-26. 125. Bauer S. Activation of NK cells and T cells by NKG2D, a receptor for stress- inducible MICA. Science. 1999;285:727-9. 126. Cosman D. ULBPs, novel MHC class I-related molecules, bind to CMV glycoprotein UL16 and stimulate NK cytotoxicity through the NKG2D receptor. Immunity. 2001;14:123- 33.

242

127. Imai K, Matsuyama S, Miyake S, Suga K, Nakachi K. Natural cytotoxic activity of peripheral-blood lymphocytes and cancer incidence: an 11-year follow-up study of a general population. Lancet. 2000;356:1795-9. 128. Oppenheim DE, Roberts SJ, Clarke SL, Filler R, Lewis JM, Tigelaar RE, et al. Sustained localized expression of ligand for the activating NKG2D receptor impairs natural cytotoxicity in vivo and reduces tumor immunosurveillance. Nat Immunol. 2005;6(9):928-37. 129. Urosevic M, Dummer R. Human leukocyte antigen-G and cancer immunoediting. Cancer Res. 2008;68(3):627-30. 130. Waldhauer I, Steinle A. NK cells and cancer immunosurveillance. Oncogene. 2008;27(45):5932-43. 131. Linnemann C, van Buuren MM, Bies L, Verdegaal EM, Schotte R, Calis JJ, et al. High-throughput epitope discovery reveals frequent recognition of neo-antigens by CD4+ T cells in human melanoma. Nat Med. 2015;21(1):81-5. 132. Lu YC, Yao X, Crystal JS, Li YF, El-Gamil M, Gross C, et al. Efficient identification of mutated cancer antigens recognized by T cells associated with durable tumor regressions. Clin Cancer Res. 2014;20(13):3401-10. 133. Tran E, Turcotte S, Gros A, Robbins PF, Lu YC, Dudley ME, et al. Cancer immunotherapy based on mutation-specific CD4+ T cells in a patient with epithelial cancer. Science. 2014;344(6184):641-5. 134. Naftzger C, Takechi Y, Kohda H, Hara I, Vijayasaradhi S, Houghton AN. Immune response to a differentiation antigen induced by altered antigen: a study of tumor rejection and autoimmunity. Proc Natl Acad Sci U S A. 1996;93(25):14809-14. 135. Bowne WB, Srinivasan R, Wolchok JD, Hawkins WG, Blachere NE, Dyall R, et al. Coupling and uncoupling of tumor immunity and autoimmunity. J Exp Med. 1999;190(11):1717-22. 136. Takada K, Takahama Y. Positive-selection-inducing self-peptides displayed by cortical thymic epithelial cells. Adv Immunol. 2015;125:87-110. 137. Derbinski J, Schulte A, Kyewski B, Klein L. Promiscuous gene expression in medullary thymic epithelial cells mirrors the peripheral self. Nat Immunol. 2001;2(11):1032- 9. 138. Sakaguchi S. Regulatory T cells: key controllers of immunologic self-tolerance. Cell. 2000;101(5):455-8. 139. Stone JD, Harris DT, Kranz DM. TCR affinity for p/MHC formed by tumor antigens that are self-proteins: impact on efficacy and toxicity. Curr Opin Immunol. 2015;33:16-22. 140. Xing Y, Hogquist KA. T-cell tolerance: central and peripheral. Cold Spring Harb Perspect Biol. 2012;4(6).

243

141. Johnson LA, Morgan RA, Dudley ME, Cassard L, Yang JC, Hughes MS, et al. Gene therapy with human and mouse T-cell receptors mediates cancer regression and targets normal tissues expressing cognate antigen. Blood. 2009;114(3):535-46. 142. Kägi D, Ledermann B, Bürki K, Zinkernagel RM, Hengartner H. Molecular mechanisms of lymphocyte-mediated cytotoxicity and their role in immunological protection and pathogenesis in vivo. Annu Rev Immunol. 1996;14:207-32. 143. Doherty PC, Christensen JP. Accessing complexity: the dynamics of virus-specific T cell responses. Annu Rev Immunol. 2000;18:561-92. 144. Russell JH, Ley TJ. Lymphocyte-mediated cytotoxicity. Annu Rev Immunol. 2002;20:323-70. 145. Voskoboinik I, Whisstock JC, Trapani JA. Perforin and granzymes: function, dysfunction and human pathology. Nat Rev Immunol. 2015;15(6):388-400. 146. Sutton VR, Wowk ME, Cancilla M, Trapani JA. Caspase activation by granzyme B is indirect, and caspase autoprocessing requires the release of proapoptotic mitochondrial factors. Immunity. 2003;18:319-29. 147. Gershon RK, Kondo K. Infectious immunological tolerance. Immunology. 1971;21(6):903-14. 148. Sakaguchi S, Sakaguchi N, Asano M, Itoh M, Toda M. Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor alpha-chains (CD25). Breakdown of a single mechanism of self-tolerance causes various autoimmune diseases. J Immunol. 1995;155(3):1151-64. 149. Chen W. Conversion of peripheral CD4+CD25− naive T cells to CD4+CD25+ regulatory T cells by TGF-β induction of transcription factor Foxp3. J Exp Med. 2003;198:1875-86. 150. Josefowicz SZ, Lu LF, Rudensky AY. Regulatory T cells: mechanisms of differentiation and function. Annu Rev Immunol. 2012;30:531-64. 151. Koch MA, Tucker-Heard G, Perdue NR, Killebrew JR, Urdahl KB, Campbell DJ. The transcription factor T-bet controls regulatory T cell homeostasis and function during type 1 inflammation. Nat Immunol. 2009;10(6):595-602. 152. Zheng Y, Chaudhry A, Kas A, deRoos P, Kim JM, Chu TT, et al. Regulatory T-cell suppressor program co-opts transcription factor IRF4 to control T(H)2 responses. Nature. 2009;458(7236):351-6. 153. Walker MR, Kasprowicz DJ, Gersuk VH, Benard A, Van Landeghen M, Buckner JH, et al. Induction of FoxP3 and acquisition of T regulatory activity by stimulated human CD4+CD25- T cells. J Clin Invest. 2003;112(9):1437-43. 154. Gasteiger G. IL-2–dependent tuning of NK cell sensitivity for target cells is

244

controlled by regulatory T cells. J Exp Med. 2013;210:1179-87. 155. Bates GJ. Quantification of regulatory T cells enables the identification of high-risk breast cancer patients and those at risk of late relapse. J Clin Oncol. 2006;24:5373-80. 156. Fu J. Increased regulatory T cells correlate with CD8 T-cell impairment and poor survival in hepatocellular carcinoma patients. Gastroenterology. 2007;132:2328-39. 157. Hori S, Nomura T, Sakaguchi S. Control of regulatory T cell development by the transcription factor Foxp3. Science. 2003;299:1057-61. 158. Murakami M, Sakamoto A, Bender J, Kappler J, Marrack P. CD25+CD4+ T cells contribute to the control of memory CD8+ T cells. Proc Natl Acad Sci U S A. 2002;99(13):8832- 7. 159. Shimizu J, Yamazaki S, Takahashi T, Ishida Y, Sakaguchi S. Stimulation of CD25(+)CD4(+) regulatory T cells through GITR breaks immunological self-tolerance. Nat Immunol. 2002;3(2):135-42. 160. Read S, Malmström V, Powrie F. Cytotoxic T lymphocyte-associated antigen 4 plays an essential role in the function of CD25(+)CD4(+) regulatory cells that control intestinal inflammation. J Exp Med. 2000;192(2):295-302. 161. Yu P, Lee Y, Liu W, Krausz T, Chong A, Schreiber H, et al. Intratumor depletion of CD4+ cells unmasks tumor immunogenicity leading to the rejection of late-stage tumors. J Exp Med. 2005;201(5):779-91. 162. Linehan DC, Goedegebuure PS. CD25+ CD4+ regulatory T-cells in cancer. Immunol Res. 2005;32(1-3):155-68. 163. Bilate AM, Lafaille JJ. Induced CD4+Foxp3+ regulatory T cells in immune tolerance. Annu Rev Immunol. 2012;30:733-58. 164. Ghiringhelli F, Puig PE, Roux S, Parcellier A, Schmitt E, Solary E, et al. Tumor cells convert immature myeloid dendritic cells into TGF-beta-secreting cells inducing CD4+CD25+ regulatory T cell proliferation. J Exp Med. 2005;202(7):919-29. 165. Curiel TJ. Specific recruitment of regulatory T cells in ovarian carcinoma fosters immune privilege and predicts reduced survival. Nature Med. 2004;10:942-9. 166. Peng Y, Laouar Y, Li MO, Green EA, Flavell RA. TGF-β regulates in vivo expansion of Foxp3-expressing CD4+CD25+ regulatory T cells responsible for protection against diabetes. Proc Natl Acad Sci Usa. 2004;101:4572-7. 167. Luckheeram RV, Zhou R, Verma AD, Xia B. CD4⁺T cells: differentiation and functions. Clin Dev Immunol. 2012;2012:925135. 168. Tao X, Constant S, Jorritsma P, Bottomly K. Strength of TCR signal determines the costimulatory requirements for Th1 and Th2 CD4+ T cell differentiation. J Immunol. 1997;159(12):5956-63.

245

169. Trinchieri G, Pflanz S, Kastelein RA. The IL-12 family of heterodimeric cytokines: new players in the regulation of T cell responses. Immunity. 2003;19(5):641-4. 170. Schnare M, Barton GM, Holt AC, Takeda K, Akira S, Medzhitov R. Toll-like receptors control activation of adaptive immune responses. Nat Immunol. 2001;2(10):947-50. 171. Lugo-Villarino G, Maldonado-Lopez R, Possemato R, Penaranda C, Glimcher LH. T- bet is required for optimal production of IFN-gamma and antigen-specific T cell activation by dendritic cells. Proc Natl Acad Sci U S A. 2003;100(13):7749-54. 172. Lazarevic V, Chen X, Shim JH, Hwang ES, Jang E, Bolm AN, et al. T-bet represses T(H)17 differentiation by preventing Runx1-mediated activation of the gene encoding RORγt. Nat Immunol. 2011;12(1):96-104. 173. Afkarian M, Sedy JR, Yang J, Jacobson NG, Cereb N, Yang SY, et al. T-bet is a STAT1-induced regulator of IL-12R expression in naïve CD4+ T cells. Nat Immunol. 2002;3(6):549-57. 174. Dong C. Diversification of T-helper-cell lineages: finding the family root of IL-17- producing cells. Nat Rev Immunol. 2006;6(4):329-33. 175. Kaplan MH, Schindler U, Smiley ST, Grusby MJ. Stat6 is required for mediating responses to IL-4 and for development of Th2 cells. Immunity. 1996;4(3):313-9. 176. Zhu J, Yamane H, Cote-Sierra J, Guo L, Paul WE. GATA-3 promotes Th2 responses through three different mechanisms: induction of Th2 cytokine production, selective growth of Th2 cells and inhibition of Th1 cell-specific factors. Cell Res. 2006;16(1):3-10. 177. Usui T, Nishikomori R, Kitani A, Strober W. GATA-3 suppresses Th1 development by downregulation of Stat4 and not through effects on IL-12Rbeta2 chain or T-bet. Immunity. 2003;18(3):415-28. 178. Horiuchi S, Onodera A, Hosokawa H, Watanabe Y, Tanaka T, Sugano S, et al. Genome-wide analysis reveals unique regulation of transcription of Th2-specific genes by GATA3. J Immunol. 2011;186(11):6378-89. 179. Veldhoen M, Uyttenhove C, van Snick J, Helmby H, Westendorf A, Buer J, et al. Transforming growth factor-beta 'reprograms' the differentiation of T helper 2 cells and promotes an interleukin 9-producing subset. Nat Immunol. 2008;9(12):1341-6. 180. Zhou L, Lopes JE, Chong MM, Ivanov II, Min R, Victora GD, et al. TGF-beta-induced Foxp3 inhibits T(H)17 cell differentiation by antagonizing RORgammat function. Nature. 2008;453(7192):236-40. 181. Bending D, De la Peña H, Veldhoen M, Phillips JM, Uyttenhove C, Stockinger B, et al. Highly purified Th17 cells from BDC2.5NOD mice convert into Th1-like cells in NOD/SCID recipient mice. J Clin Invest. 2009;119(3):565-72. 182. Lee YK, Turner H, Maynard CL, Oliver JR, Chen D, Elson CO, et al. Late

246

developmental plasticity in the T helper 17 lineage. Immunity. 2009;30(1):92-107. 183. Xu L, Kitani A, Fuss I, Strober W. Cutting edge: regulatory T cells induce CD4+CD25-Foxp3- T cells or are self-induced to become Th17 cells in the absence of exogenous TGF-beta. J Immunol. 2007;178(11):6725-9. 184. Tsuji M, Komatsu N, Kawamoto S, Suzuki K, Kanagawa O, Honjo T, et al. Preferential generation of follicular B helper T cells from Foxp3+ T cells in gut Peyer's patches. Science. 2009;323(5920):1488-92. 185. Gascoigne NR, Rybakin V, Acuto O, Brzostek J. TCR Signal Strength and T Cell Development. Annu Rev Cell Dev Biol. 2016;32:327-48. 186. Constant SL, Bottomly K. Induction of Th1 and Th2 CD4+ T cell responses: the alternative approaches. Annu Rev Immunol. 1997;15:297-322. 187. van Panhuys N, Klauschen F, Germain RN. T-cell-receptor-dependent signal intensity dominantly controls CD4(+) T cell polarization In Vivo. Immunity. 2014;41(1):63-74. 188. Kaech SM, Wherry EJ, Ahmed R. Effector and memory T-cell differentiation: implications for vaccine development. Nat Rev Immunol. 2002;2(4):251-62. 189. Pages F. Effector memory T cells, early metastasis, and survival in colorectal cancer. N Engl J Med. 2005;353:2654-66. 190. Man K, Gabriel SS, Liao Y, Gloury R, Preston S, Henstridge DC, et al. Transcription Factor IRF4 Promotes CD8. Immunity. 2017;47(6):1129-41.e5. 191. van Duikeren S, Fransen MF, Redeker A, Wieles B, Platenburg G, Krebber WJ, et al. Vaccine-induced effector-memory CD8+ T cell responses predict therapeutic efficacy against tumors. J Immunol. 2012;189(7):3397-403. 192. Granier C, Blanc C, Karaki S, Tran T, Roussel H, Tartour E. Tissue-resident memory T cells play a key role in the efficacy of cancer vaccines. Oncoimmunology. 2017;6(11):e1358841. 193. Mami-Chouaib F, Blanc C, Corgnac S, Hans S, Malenica I, Granier C, et al. Resident memory T cells, critical components in tumor immunology. J Immunother Cancer. 2018;6(1):87. 194. Buslepp J, Wang H, Biddison WE, Appella E, Collins EJ. A correlation between TCR Valpha docking on MHC and CD8 dependence: implications for T cell selection. Immunity. 2003;19(4):595-606. 195. Mason D. A very high level of crossreactivity is an essential feature of the T-cell receptor. Immunol Today. 1998;19(9):395-404. 196. Reiser JB, Darnault C, Grégoire C, Mosser T, Mazza G, Kearney A, et al. CDR3 loop flexibility contributes to the degeneracy of TCR recognition. Nat Immunol. 2003;4(3):241-7. 197. Loftus DJ, Chen Y, Covell DG, Engelhard VH, Appella E. Differential contact of

247

disparate class I/peptide complexes as the basis for epitope cross-recognition by a single T cell receptor. J Immunol. 1997;158(8):3651-8. 198. Henderson RA, Cox AL, Sakaguchi K, Appella E, Shabanowitz J, Hunt DF, et al. Direct identification of an endogenous peptide recognized by multiple HLA-A2.1-specific cytotoxic T cells. Proc Natl Acad Sci U S A. 1993;90(21):10275-9. 199. Buslepp J, Kerry SE, Loftus D, Frelinger JA, Appella E, Collins EJ. High affinity xenoreactive TCR:MHC interaction recruits CD8 in absence of binding to MHC. J Immunol. 2003;170(1):373-83. 200. Birnbaum ME, Mendoza JL, Sethi DK, Dong S, Glanville J, Dobbins J, et al. Deconstructing the peptide-MHC specificity of T cell recognition. Cell. 2014;157(5):1073-87. 201. Luz JG, Huang M, Garcia KC, Rudolph MG, Apostolopoulos V, Teyton L, et al. Structural comparison of allogeneic and syngeneic T cell receptor-peptide-major histocompatibility complex complexes: a buried alloreactive mutation subtly alters peptide presentation substantially increasing V(beta) Interactions. J Exp Med. 2002;195(9):1175-86. 202. Reiser JB, Darnault C, Guimezanes A, Grégoire C, Mosser T, Schmitt-Verhulst AM, et al. Crystal structure of a T cell receptor bound to an allogeneic MHC molecule. Nat Immunol. 2000;1(4):291-7. 203. Zhao R, Loftus DJ, Appella E, Collins EJ. Structural evidence of T cell xenoreactivity in the absence of molecular mimicry. J Exp Med. 1999;189(2):359-70. 204. Speir JA, Garcia KC, Brunmark A, Degano M, Peterson PA, Teyton L, et al. Structural basis of 2C TCR allorecognition of H-2Ld peptide complexes. Immunity. 1998;8(5):553-62. 205. LeBien TW, Tedder TF. B lymphocytes: how they develop and function. Blood. 2008;112(5):1570-80. 206. Iglesia MD, Vincent BG, Parker JS, Hoadley KA, Carey LA, Perou CM, et al. Prognostic B-cell signatures using mRNA-seq in patients with subtype-specific breast and ovarian cancer. Clin Cancer Res. 2014;20(14):3818-29. 207. Schmidt M, Böhm D, von Törne C, Steiner E, Puhl A, Pilch H, et al. The humoral immune system has a key prognostic impact in node-negative breast cancer. Cancer Res. 2008;68(13):5405-13. 208. Wang J, Figueroa JD, Wallstrom G, Barker K, Park JG, Demirkan G, et al. Plasma Autoantibodies Associated with Basal-like Breast Cancers. Cancer Epidemiol Biomarkers Prev. 2015;24(9):1332-40. 209. Stovgaard ES, Nielsen D, Hogdall E, Balslev E. Triple negative breast cancer - prognostic role of immune-related factors: a systematic review. Acta Oncol. 2018;57(1):74-82. 210. Gnjatic S, Atanackovic D, Jäger E, Matsuo M, Selvakumar A, Altorki NK, et al.

248

Survey of naturally occurring CD4+ T cell responses against NY-ESO-1 in cancer patients: correlation with antibody responses. Proc Natl Acad Sci U S A. 2003;100(15):8862-7. 211. Disis ML, Calenoff E, McLaughlin G, Murphy AE, Chen W, Groner B, et al. Existent T-cell and antibody immunity to HER-2/neu protein in patients with breast cancer. Cancer Res. 1994;54(1):16-20. 212. Kehrl JH, Wakefield LM, Roberts AB, Jakowlew S, Alvarez-Mon M, Derynck R, et al. Production of transforming growth factor beta by human T lymphocytes and its potential role in the regulation of T cell growth. J Exp Med. 1986;163(5):1037-50. 213. Lebman DA, Edmiston JS. The role of TGF-beta in growth, differentiation, and maturation of B lymphocytes. Microbes Infect. 1999;1(15):1297-304. 214. Mocellin S, Marincola FM, Young HA. Interleukin-10 and the immune response against cancer: a counterpoint. J Leukoc Biol. 2005;78(5):1043-51. 215. Nelson BH. IL-2, regulatory T cells, and tolerance. J Immunol. 2004;172(7):3983-8. 216. Stathopoulos GT, Kollintza A, Moschos C, Psallidas I, Sherrill TP, Pitsinos EN, et al. Tumor necrosis factor-alpha promotes malignant pleural effusion. Cancer Res. 2007;67(20):9825-34. 217. Rot A, von Andrian UH. Chemokines in innate and adaptive host defense: basic chemokinese grammar for immune cells. Annu Rev Immunol. 2004;22:891-928. 218. Griffith JW, Sokol CL, Luster AD. Chemokines and chemokine receptors: positioning cells for host defense and immunity. Annu Rev Immunol. 2014;32:659-702. 219. Balkwill F. Cancer and the chemokine network. Nat Rev Cancer. 2004;4(7):540-50. 220. Galon J. Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science. 2006;313:1960-4. 221. Nagarsheth N, Wicha MS, Zou W. Chemokines in the cancer microenvironment and their relevance in cancer immunotherapy. Nat Rev Immunol. 2017;17(9):559-72. 222. Facciabene A. Tumour hypoxia promotes tolerance and angiogenesis via CCL28 and Treg cells. Nature. 2011;475:226-30. 223. Zou L. Bone marrow is a reservoir for CD4+CD25+ regulatory T cells that traffic through CXCL12/CXCR4 signals. Cancer Res. 2004;64:8451-5. 224. Balkwill F, Mantovani A. Inflammation and cancer: back to Virchow? Lancet. 2001;357(9255):539-45. 225. Helbig G. NF-κB promotes breast cancer cell migration and metastasis by inducing the expression of the chemokine receptor CXCR4. J Biol Chem. 2003;278:21631-8. 226. Busch W. Aus der sitzung der medicinischen section vom 13. November 1867. Berliner Klinische Wochenschrift. 1868;5:137. 227. F. F. Ueber die Züchtung der Erysipelkokken auf künstlichem

249

Nährboden und ihre Übertragbarkeit auf den Menschen.; 1882. 228. Coley WB. The treatment of malignant tumors by repeated inoculations of erysipelas. With a report of ten original cases. 1893. Clin Orthop Relat Res. 1991(262):3-11. 229. Morales A, Eidinger D, Bruce AW. Intracavitary Bacillus Calmette-Guerin in the treatment of superficial bladder tumors. J Urol. 1976;116(2):180-3. 230. OLD LJ, CLARKE DA, BENACERRAF B. Effect of Bacillus Calmette-Guerin infection on transplanted tumours in the mouse. Nature. 1959;184(Suppl 5):291-2. 231. Simpson E. Medawar's legacy to cellular immunology and clinical transplantation: a commentary on Billingham, Brent and Medawar (1956) 'Quantitative studies on tissue transplantation immunity. III. Actively acquired tolerance'. Philos Trans R Soc Lond B Biol Sci. 2015;370(1666). 232. PREHN RT, MAIN JM. Immunity to methylcholanthrene-induced sarcomas. J Natl Cancer Inst. 1957;18(6):769-78. 233. Burnet FM. The concept of immunological surveillance. Prog Exp Tumor Res. 1970;13:1-27. 234. Allison JP, McIntyre BW, Bloch D. Tumor-specific antigen of murine T-lymphoma defined with monoclonal antibody. J Immunol. 1982;129(5):2293-300. 235. van der Bruggen P, Traversari C, Chomez P, Lurquin C, De Plaen E, Van den Eynde B, et al. A gene encoding an antigen recognized by cytolytic T lymphocytes on a human melanoma. Science. 1991;254(5038):1643-7. 236. Leach DR, Krummel MF, Allison JP. Enhancement of antitumor immunity by CTLA-4 blockade. Science. 1996;271(5256):1734-6. 237. Ishida Y, Agata Y, Shibahara K, Honjo T. Induced expression of PD-1, a novel member of the immunoglobulin gene superfamily, upon programmed cell death. Embo J. 1992;11:3887-95. 238. Triebel F. LAG-3, a novel lymphocyte activation gene closely related to CD4. J Exp Med. 1990;171:1393-405. 239. Oiseth SJ, Aziz MS. Cancer immunotherapy: a brief review of the history, possibilities, and challenges ahead. Journal of Cancer Metastasis and Treatment. 2017;3(10):250-61. 240. Spanknebel K, Cheung KY, Stoutenburg J, Hurst-Wicker K, Hesdorffer C, Rn GD, et al. Initial clinical response predicts outcome and is associated with dose schedule in metastatic melanoma and renal cell carcinoma patients treated with high-dose interleukin 2. Ann Surg Oncol. 2005;12(5):381-90. 241. Pachella LA, Madsen LT, Dains JE. The Toxicity and Benefit of Various Dosing Strategies for Interleukin-2 in Metastatic Melanoma and Renal Cell Carcinoma. J Adv Pract

250

Oncol. 2015;6(3):212-21. 242. Gardner TA, Elzey BD, Hahn NM. Sipuleucel-T (Provenge) autologous vaccine approved for treatment of men with asymptomatic or minimally symptomatic castrate- resistant metastatic prostate cancer. Hum Vaccin Immunother. 2012;8(4):534-9. 243. Hanna MG, Peters LC. Specific immunotherapy of established visceral micrometastases by BCG-tumor cell vaccine alone or as an adjunct to surgery. Cancer. 1978;42(6):2613-25. 244. Plaksin D, Porgador A, Vadai E, Feldman M, Schirrmacher V, Eisenbach L. Effective anti-metastatic melanoma vaccination with tumor cells transfected with MHC genes and/or infected with Newcastle disease virus (NDV). Int J Cancer. 1994;59(6):796-801. 245. Asada H, Kishida T, Hirai H, Satoh E, Ohashi S, Takeuchi M, et al. Significant antitumor effects obtained by autologous tumor cell vaccine engineered to secrete interleukin (IL)-12 and IL-18 by means of the EBV/lipoplex. Mol Ther. 2002;5(5 Pt 1):609-16. 246. Armstrong CA, Botella R, Galloway TH, Murray N, Kramp JM, Song IS, et al. Antitumor effects of granulocyte-macrophage colony-stimulating factor production by melanoma cells. Cancer Res. 1996;56(9):2191-8. 247. Dranoff G. GM-CSF-based cancer vaccines. Immunol Rev. 2002;188:147-54. 248. Guo C, Manjili MH, Subjeck JR, Sarkar D, Fisher PB, Wang XY. Therapeutic cancer vaccines: past, present, and future. Adv Cancer Res. 2013;119:421-75. 249. Sondak VK, Sabel MS, Mulé JJ. Allogeneic and autologous melanoma vaccines: where have we been and where are we going? Clin Cancer Res. 2006;12(7 Pt 2):2337s-41s. 250. Janeway CA. The immune system evolved to discriminate infectious nonself from noninfectious self. Immunol Today. 1992;13(1):11-6. 251. Liu MA. DNA vaccines: an historical perspective and view to the future. Immunol Rev. 2011;239(1):62-84. 252. Sahin U, Derhovanessian E, Miller M, Kloke BP, Simon P, Löwer M, et al. Personalized RNA mutanome vaccines mobilize poly-specific therapeutic immunity against cancer. Nature. 2017;547(7662):222-6. 253. Walsh SR, Dolin R. Vaccinia viruses: vaccines against smallpox and vectors against infectious diseases and tumors. Expert Rev Vaccines. 2011;10(8):1221-40. 254. Coulie PG, Van den Eynde BJ, van der Bruggen P, Boon T. Tumour antigens recognized by T lymphocytes: at the core of cancer immunotherapy. Nat Rev Cancer. 2014;14(2):135-46. 255. Germeau C. High frequency of antitumor T cells in the blood of melanoma patients before and after vaccination with tumor antigens. J Exp Med. 2005;201:241-8. 256. Lurquin C. Contrasting frequencies of antitumor and anti-vaccine T cells in

251

metastases of a melanoma patient vaccinated with a MAGE tumor antigen. J Exp Med. 2005;201:249-57. 257. Carrasco J. Vaccination of a melanoma patient with mature dendritic cells pulsed with MAGE-3 peptides triggers the activity of nonvaccine anti-tumor cells. J Immunol. 2008;180:3585-93. 258. Coulie PG. Precursor frequency analysis of human cytolytic T lymphocytes directed against autologous melanoma cells. Int J Cancer. 1992;50:289-97. 259. Gajewski TF. Immune resistance orchestrated by the tumor microenvironment. Immunol Rev. 2006;213:131-45. 260. Morgan RA, Dudley ME, Wunderlich JR, Hughes MS, Yang JC, Sherry RM, et al. Cancer regression in patients after transfer of genetically engineered lymphocytes. Science. 2006;314(5796):126-9. 261. June CH, O'Connor RS, Kawalekar OU, Ghassemi S, Milone MC. CAR T cell immunotherapy for human cancer. Science. 2018;359(6382):1361-5. 262. Kuwana Y, Asakura Y, Utsunomiya N, Nakanishi M, Arata Y, Itoh S, et al. Expression of chimeric receptor composed of immunoglobulin-derived V regions and T-cell receptor-derived C regions. Biochem Biophys Res Commun. 1987;149(3):960-8. 263. Sotillo E, Barrett DM, Black KL, Bagashev A, Oldridge D, Wu G, et al. Convergence of Acquired Mutations and Alternative Splicing of CD19 Enables Resistance to CART-19 Immunotherapy. Cancer Discov. 2015;5(12):1282-95. 264. Kloss CC, Condomines M, Cartellieri M, Bachmann M, Sadelain M. Combinatorial antigen recognition with balanced signaling promotes selective tumor eradication by engineered T cells. Nat Biotechnol. 2013;31(1):71-5. 265. Haanen JBAG, Carbonnel F, Robert C, Kerr KM, Peters S, Larkin J, et al. Management of toxicities from immunotherapy: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2017;28(suppl_4):iv119-iv42. 266. Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer. 2012;12(4):252-64. 267. Wei SC, Duffy CR, Allison JP. Fundamental Mechanisms of Immune Checkpoint Blockade Therapy. Cancer Discov. 2018. 268. Schwartz RH. Costimulation of T lymphocytes: the role of CD28, CTLA-4, and B7/BB1 in interleukin-2 production and immunotherapy. Cell. 1992;71:1065-8. 269. Hathcock KS. Identification of an alternative CTLA-4 ligand costimulatory for T cell activation. Science. 1993;262:905-7. 270. Linsley PS. Human B7–1 (CD80) and B7–2 (CD86) bind with similar avidities but distinct kinetics to CD28 and CTLA-4 receptors. Immunity. 1994;1:793-801.

252

271. Schneider H. Reversal of the TCR stop signal by CTLA-4. Science. 2006;313:1972-5. 272. Wing K. CTLA-4 control over Foxp3+ regulatory T cell function. Science. 2008;322:271-5. 273. Lenschow DJ, Walunas TL, Bluestone JA. CD28/B7 system of T cell costimulation. Annu Rev Immunol. 1996;14:233-58. 274. van Elsas A, Hurwitz AA, Allison JP. Combination immunotherapy of B16 melanoma using anti-cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) and granulocyte/macrophage colony-stimulating factor (GM-CSF)-producing vaccines induces rejection of subcutaneous and metastatic tumors accompanied by autoimmune depigmentation. J Exp Med. 1999;190:355-66. 275. Freeman GJ. Engagement of the PD-1 immunoinhibitory receptor by a novel B7 family member leads to negative regulation of lymphocyte activation. J Exp Med. 2000;192:1027-34. 276. Latchman Y. PD-L2 is a second ligand for PD-1 and inhibits T cell activation. Nature Immunol. 2001;2:261-8. 277. Terme M. IL-18 induces PD-1-dependent immunosuppression in cancer. Cancer Res. 2011;71:5393-9. 278. Fanoni D. New monoclonal antibodies against B-cell antigens: possible new strategies for diagnosis of primary cutaneous B-cell lymphomas. Immunol Lett. 2011;134:157-60. 279. Barber DL. Restoring function in exhausted CD8 T cells during chronic viral infection. Nature. 2006;439:682-7. 280. Ahmadzadeh M. Tumor antigen-specific CD8 T cells infiltrating the tumor express high levels of PD-1 and are functionally impaired. Blood. 2009;114:1537-44. 281. Konishi J. B7-H1 expression on non-small cell lung cancer cells and its relationship with tumor-infiltrating lymphocytes and their PD-1 expression. Clin Cancer Res. 2004;10:5094-100. 282. Dong H. Tumor-associated B7-H1 promotes T-cell apoptosis: a potential mechanism of immune evasion. Nature Med. 2002;8:793-800. 283. Larkin J, Lao CD, Urba WJ, McDermott DF, Horak C, Jiang J, et al. Efficacy and Safety of Nivolumab in Patients With BRAF V600 Mutant and BRAF Wild-Type Advanced Melanoma: A Pooled Analysis of 4 Clinical Trials. JAMA Oncol. 2015;1(4):433-40. 284. Nanda R, Chow LQ, Dees EC, Berger R, Gupta S, Geva R, et al. Pembrolizumab in Patients With Advanced Triple-Negative Breast Cancer: Phase Ib KEYNOTE-012 Study. J Clin Oncol. 2016;34(21):2460-7. 285. Parkhurst MR. T cells targeting carcinoembryonic antigen can mediate regression

253

of metastatic colorectal cancer but induce severe transient colitis. Mol Ther. 2011;19:620-6. 286. Lamers CH. Treatment of metastatic renal cell carcinoma with autologous T- lymphocytes genetically retargeted against carbonic anhydrase IX: first clinical experience. J Clin Oncol. 2006;24:e20-e2. 287. Zamora AE, Crawford JC, Thomas PG. Hitting the Target: How T Cells Detect and Eliminate Tumors. J Immunol. 2018;200(2):392-9. 288. GOLD P, FREEDMAN SO. DEMONSTRATION OF TUMOR-SPECIFIC ANTIGENS IN HUMAN COLONIC CARCINOMATA BY IMMUNOLOGICAL TOLERANCE AND ABSORPTION TECHNIQUES. J Exp Med. 1965;121:439-62. 289. Boucher D, Cournoyer D, Stanners CP, Fuks A. Studies on the control of gene expression of the carcinoembryonic antigen family in human tissue. Cancer Res. 1989;49(4):847-52. 290. Thompson JA, Grunert F, Zimmermann W. Carcinoembryonic antigen gene family: molecular biology and clinical perspectives. J Clin Lab Anal. 1991;5(5):344-66. 291. Deutsch HF. Chemistry and biology of alpha-fetoprotein. Adv Cancer Res. 1991;56:253-312. 292. Taketa K. Alpha-fetoprotein: reevaluation in hepatology. Hepatology. 1990;12(6):1420-32. 293. Lee JS, Heo J, Libbrecht L, Chu IS, Kaposi-Novak P, Calvisi DF, et al. A novel prognostic subtype of human hepatocellular carcinoma derived from hepatic progenitor cells. Nat Med. 2006;12(4):410-6. 294. Rieger R, Edenhofer F, Lasmézas CI, Weiss S. The human 37-kDa laminin receptor precursor interacts with the prion protein in eukaryotic cells. Nat Med. 1997;3(12):1383-8. 295. Coggin JH, Barsoum AL, Rohrer JW. 37 kiloDalton oncofetal antigen protein and immature laminin receptor protein are identical, universal T-cell inducing immunogens on primary rodent and human cancers. Anticancer Res. 1999;19(6C):5535-42. 296. Rohrer JW, Barsoum AL, Dyess DL, Tucker JA, Coggin JH. Human breast carcinoma patients develop clonable oncofetal antigen-specific effector and regulatory T lymphocytes. J Immunol. 1999;162(11):6880-92. 297. Zelle-Rieser C, Barsoum AL, Sallusto F, Ramoner R, Rohrer JW, Höltl L, et al. Expression and immunogenicity of oncofetal antigen-immature laminin receptor in human renal cell carcinoma. J Urol. 2001;165(5):1705-9. 298. Siegel S, Wagner A, Kabelitz D, Marget M, Coggin J, Barsoum A, et al. Induction of cytotoxic T-cell responses against the oncofetal antigen-immature laminin receptor for the treatment of hematologic malignancies. Blood. 2003;102(13):4416-23. 299. Sahin U, Türeci Ö. Personalized vaccines for cancer immunotherapy. Science.

254

2018;359(6382):1355-60. 300. Lennerz V. The response of autologous T cells to a human melanoma is dominated by mutated neoantigens. Proc Natl Acad Sci Usa. 2005;102:16013-8. 301. Bassani-Sternberg M, Bräunlein E, Klar R, Engleitner T, Sinitcyn P, Audehm S, et al. Direct identification of clinically relevant neoepitopes presented on native human melanoma tissue by mass spectrometry. Nat Commun. 2016;7:13404. 302. Ott PA, Hu Z, Keskin DB, Shukla SA, Sun J, Bozym DJ, et al. An immunogenic personal neoantigen vaccine for patients with melanoma. Nature. 2017;547(7662):217-21. 303. Laidlaw BJ, Craft JE, Kaech SM. The multifaceted role of CD4(+) T cells in CD8(+) T cell memory. Nat Rev Immunol. 2016;16(2):102-11. 304. Kaufman HL, Ruby CE, Hughes T, Slingluff CL. Current status of granulocyte- macrophage colony-stimulating factor in the immunotherapy of melanoma. J Immunother Cancer. 2014;2:11. 305. Gnjatic S, Sawhney NB, Bhardwaj N. Toll-like receptor agonists: are they good adjuvants? Cancer J. 2010;16(4):382-91. 306. Galluzzi L, Buqué A, Kepp O, Zitvogel L, Kroemer G. Immunological Effects of Conventional Chemotherapy and Targeted Anticancer Agents. Cancer Cell. 2015;28(6):690- 714. 307. Peng D, Kryczek I, Nagarsheth N, Zhao L, Wei S, Wang W, et al. Epigenetic silencing of TH1-type chemokines shapes tumour immunity and immunotherapy. Nature. 2015;527(7577):249-53. 308. Barnes PJ, Adcock IM, Ito K. Histone acetylation and deacetylation: importance in inflammatory lung diseases. Eur Respir J. 2005;25(3):552-63. 309. Yang XJ, Seto E. The Rpd3/Hda1 family of lysine deacetylases: from bacteria and yeast to mice and men. Nat Rev Mol Cell Biol. 2008;9(3):206-18. 310. de Ruijter AJ, van Gennip AH, Caron HN, Kemp S, van Kuilenburg AB. Histone deacetylases (HDACs): characterization of the classical HDAC family. Biochem J. 2003;370(Pt 3):737-49. 311. Shakespear MR, Halili MA, Irvine KM, Fairlie DP, Sweet MJ. Histone deacetylases as regulators of inflammation and immunity. Trends Immunol. 2011;32:335-43. 312. Hancock WW, Akimova T, Beier UH, Liu Y, Wang L. HDAC inhibitor therapy in autoimmunity and transplantation. Ann Rheum Dis. 2012;71:i46-i54. 313. Sweet MJ, Shakespear MR, Kamal NA, Fairlie DP. HDAC inhibitors: modulating leukocyte differentiation, survival, proliferation and inflammation. Immunol Cell Biol. 2012;90:14-22. 314. Shakespear MR. Histone deacetylase 7 promotes Toll-like receptor 4-dependent

255

proinflammatory gene expression in macrophages. J Biol Chem. 2013;288:25362-74. 315. Tao R. Deacetylase inhibition promotes the generation and function of regulatory T cells. Nat Med. 2007;13:1299-307. 316. Sen DR, Kaminski J, Barnitz RA, Kurachi M, Gerdemann U, Yates KB, et al. The epigenetic landscape of T cell exhaustion. Science. 2016;354(6316):1165-9. 317. Zhang F, Zhou X, DiSpirito JR, Wang C, Wang Y, Shen H. Epigenetic manipulation restores functions of defective CD8⁺ T cells from chronic viral infection. Mol Ther. 2014;22(9):1698-706. 318. Glaser KB, Staver MJ, Waring JF, Stender J, Ulrich RG, Davidsen SK. Gene expression profiling of multiple histone deacetylase (HDAC) inhibitors: defining a common gene set produced by HDAC inhibition in T24 and MDA carcinoma cell lines. Mol Cancer Ther. 2003;2(2):151-63. 319. LaBonte MJ, Wilson PM, Fazzone W, Groshen S, Lenz HJ, Ladner RD. DNA microarray profiling of genes differentially regulated by the histone deacetylase inhibitors vorinostat and LBH589 in colon cancer cell lines. BMC Med Genomics. 2009;2:67. 320. Halsall J, Gupta V, O'Neill LP, Turner BM, Nightingale KP. Genes are often sheltered from the global histone hyperacetylation induced by HDAC inhibitors. PLoS One. 2012;7(3):e33453. 321. Dawson MA, Kouzarides T. Cancer epigenetics: from mechanism to therapy. Cell. 2012;150:12-27. 322. Murakami T, Sato A, Chun NA, Hara M, Naito Y, Kobayashi Y, et al. Transcriptional modulation using HDACi depsipeptide promotes immune cell-mediated tumor destruction of murine B16 melanoma. J Invest Dermatol. 2008;128(6):1506-16. 323. Mora-García MeL, Duenas-González A, Hernández-Montes J, De la Cruz- Hernández E, Pérez-Cárdenas E, Weiss-Steider B, et al. Up-regulation of HLA class-I antigen expression and antigen-specific CTL response in cervical cancer cells by the demethylating agent hydralazine and the histone deacetylase inhibitor valproic acid. J Transl Med. 2006;4:55. 324. Armeanu S, Bitzer M, Lauer UM, Venturelli S, Pathil A, Krusch M, et al. Natural killer cell-mediated lysis of hepatoma cells via specific induction of NKG2D ligands by the histone deacetylase inhibitor sodium valproate. Cancer Res. 2005;65(14):6321-9. 325. Bolden JE, Peart MJ, Johnstone RW. Anticancer activities of histone deacetylase inhibitors. Nat Rev Drug Discov. 2006;5(9):769-84. 326. Stagg J, Johnstone RW, Smyth MJ. From cancer immunosurveillance to cancer immunotherapy. Immunol Rev. 2007;220:82-101. 327. Guerriero JL, Sotayo A, Ponichtera HE, Castrillon JA, Pourzia AL, Schad S, et al.

256

Class IIa HDAC inhibition reduces breast tumours and metastases through anti-tumour macrophages. Nature. 2017;543(7645):428-32. 328. Cimino-Mathews A, Foote JB, Emens LA. Immune targeting in breast cancer. Oncology (Williston Park). 2015;29(5):375-85. 329. Evans MJ, Kaufman MH. Establishment in culture of pluripotential cells from mouse embryos. Nature. 1981;292:154-6. 330. Thomson JA. Embryonic stem cells derived from human blastocysts. Science. 1998;282:1145-7. 331. Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell. 2006;126:663-76. 332. Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T, Tomoda K, et al. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell. 2007;131(5):861-72. 333. De Los Angeles A, Ferrari F, Xi R, Fujiwara Y, Benvenisty N, Deng H, et al. Hallmarks of pluripotency. Nature. 2015;525(7570):469-78. 334. Hackett JA, Surani MA. Regulatory principles of pluripotency: from the ground state up. Cell Stem Cell. 2014;15(4):416-30. 335. Yu J, Vodyanik MA, Smuga-Otto K, Antosiewicz-Bourget J, Frane JL, Tian S, et al. Induced pluripotent stem cell lines derived from human somatic cells. Science. 2007;318(5858):1917-20. 336. Mikkelsen TS, Hanna J, Zhang X, Ku M, Wernig M, Schorderet P, et al. Dissecting direct reprogramming through integrative genomic analysis. Nature. 2008;454(7200):49-55. 337. Takahashi K. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell. 2007;131:861-72. 338. Soldner F, Hockemeyer D, Beard C, Gao Q, Bell GW, Cook EG, et al. Parkinson's disease patient-derived induced pluripotent stem cells free of viral reprogramming factors. Cell. 2009;136(5):964-77. 339. Aoi T, Yae K, Nakagawa M, Ichisaka T, Okita K, Takahashi K, et al. Generation of pluripotent stem cells from adult mouse liver and stomach cells. Science. 2008;321(5889):699- 702. 340. Sohn YD, Han JW, Yoon YS. Generation of induced pluripotent stem cells from somatic cells. Prog Mol Biol Transl Sci. 2012;111:1-26. 341. Wu X, Li Y, Crise B, Burgess SM. Transcription start regions in the human genome are favored targets for MLV integration. Science. 2003;300(5626):1749-51. 342. Okita K, Ichisaka T, Yamanaka S. Generation of germline-competent induced pluripotent stem cells. Nature. 2007;448(7151):313-7.

257

343. Okita K, Nakagawa M, Hyenjong H, Ichisaka T, Yamanaka S. Generation of mouse induced pluripotent stem cells without viral vectors. Science. 2008;322(5903):949-53. 344. Yusa K, Rad R, Takeda J, Bradley A. Generation of transgene-free induced pluripotent mouse stem cells by the piggyBac transposon. Nat Methods. 2009;6(5):363-9. 345. Kaji K, Norrby K, Paca A, Mileikovsky M, Mohseni P, Woltjen K. Virus-free induction of pluripotency and subsequent excision of reprogramming factors. Nature. 2009;458(7239):771-5. 346. Kim D, Kim CH, Moon JI, Chung YG, Chang MY, Han BS, et al. Generation of human induced pluripotent stem cells by direct delivery of reprogramming proteins. Cell Stem Cell. 2009;4(6):472-6. 347. Zhou H, Wu S, Joo JY, Zhu S, Han DW, Lin T, et al. Generation of induced pluripotent stem cells using recombinant proteins. Cell Stem Cell. 2009;4(5):381-4. 348. Warren L, Manos PD, Ahfeldt T, Loh YH, Li H, Lau F, et al. Highly efficient reprogramming to pluripotency and directed differentiation of human cells with synthetic modified mRNA. Cell Stem Cell. 2010;7(5):618-30. 349. Houbaviy HB, Murray MF, Sharp PA. Embryonic stem cell-specific MicroRNAs. Dev Cell. 2003;5(2):351-8. 350. Miyoshi N, Ishii H, Nagano H, Haraguchi N, Dewi DL, Kano Y, et al. Reprogramming of mouse and human cells to pluripotency using mature microRNAs. Cell Stem Cell. 2011;8(6):633-8. 351. Maherali N, Sridharan R, Xie W, Utikal J, Eminli S, Arnold K, et al. Directly reprogrammed fibroblasts show global epigenetic remodeling and widespread tissue contribution. Cell Stem Cell. 2007;1(1):55-70. 352. Wernig M, Meissner A, Foreman R, Brambrink T, Ku M, Hochedlinger K, et al. In vitro reprogramming of fibroblasts into a pluripotent ES-cell-like state. Nature. 2007;448(7151):318-24. 353. Ghosh Z, Wilson KD, Wu Y, Hu S, Quertermous T, Wu JC. Persistent donor cell gene expression among human induced pluripotent stem cells contributes to differences with human embryonic stem cells. PLoS One. 2010;5(2):e8975. 354. Senju S, Hirata S, Matsuyoshi H, Masuda M, Uemura Y, Araki K, et al. Generation and genetic modification of dendritic cells derived from mouse embryonic stem cells. Blood. 2003;101(9):3501-8. 355. Fairchild PJ, Brook FA, Gardner RL, Graça L, Strong V, Tone Y, et al. Directed differentiation of dendritic cells from mouse embryonic stem cells. Curr Biol. 2000;10(23):1515-8. 356. Senju S, Haruta M, Matsumura K, Matsunaga Y, Fukushima S, Ikeda T, et al.

258

Generation of dendritic cells and macrophages from human induced pluripotent stem cells aiming at cell therapy. Gene Ther. 2011;18(9):874-83. 357. Senju S, Matsunaga Y, Fukushima S, Hirata S, Motomura Y, Fukuma D, et al. Immunotherapy with pluripotent stem cell-derived dendritic cells. Semin Immunopathol. 2011;33(6):603-12. 358. Choi KD, Vodyanik MA, Slukvin II. Generation of mature human myelomonocytic cells through expansion and differentiation of pluripotent stem cell-derived lin- CD34+CD43+CD45+ progenitors. J Clin Invest. 2009;119(9):2818-29. 359. Zhang Q, Fujino M, Iwasaki S, Hirano H, Cai S, Kitajima Y, et al. Generation and characterization of regulatory dendritic cells derived from murine induced pluripotent stem cells. Sci Rep. 2014;4:3979. 360. Rami F, Mollainezhad H, Salehi M. Induced Pluripotent Stem Cell as a New Source for Cancer Immunotherapy. Genet Res Int. 2016;2016:3451807. 361. Schmitt TM, de Pooter RF, Gronski MA, Cho SK, Ohashi PS, Zúñiga-Pflücker JC. Induction of T cell development and establishment of T cell competence from embryonic stem cells differentiated in vitro. Nat Immunol. 2004;5(4):410-7. 362. Lei F, Haque R, Weiler L, Vrana KE, Song J. T lineage differentiation from induced pluripotent stem cells. Cell Immunol. 2009;260(1):1-5. 363. Larbi A, Mitjavila-Garcia MT, Flamant S, Valogne Y, Clay D, Usunier B, et al. Generation of multipotent early lymphoid progenitors from human embryonic stem cells. Stem Cells Dev. 2014;23(24):2983-95. 364. Brown ME, Rondon E, Rajesh D, Mack A, Lewis R, Feng X, et al. Derivation of induced pluripotent stem cells from human peripheral blood T lymphocytes. PLoS One. 2010;5(6):e11373. 365. Vizcardo R, Masuda K, Yamada D, Ikawa T, Shimizu K, Fujii S, et al. Regeneration of human tumor antigen-specific T cells from iPSCs derived from mature CD8(+) T cells. Cell Stem Cell. 2013;12(1):31-6. 366. Nishimura T, Kaneko S, Kawana-Tachikawa A, Tajima Y, Goto H, Zhu D, et al. Generation of rejuvenated antigen-specific T cells by reprogramming to pluripotency and redifferentiation. Cell Stem Cell. 2013;12(1):114-26. 367. Crompton JG, Rao M, Restifo NP. Memoirs of a reincarnated T cell. Cell Stem Cell. 2013;12(1):6-8. 368. Themeli M, Kloss CC, Ciriello G, Fedorov VD, Perna F, Gonen M, et al. Generation of tumor-targeted human T lymphocytes from induced pluripotent stem cells for cancer therapy. Nat Biotechnol. 2013;31(10):928-33. 369. Woll PS, Martin CH, Miller JS, Kaufman DS. Human embryonic stem cell-derived

259

NK cells acquire functional receptors and cytolytic activity. J Immunol. 2005;175(8):5095-103. 370. Larbi A, Gombert JM, Auvray C, l'Homme B, Magniez A, Féraud O, et al. The HOXB4 homeoprotein promotes the ex vivo enrichment of functional human embryonic stem cell-derived NK cells. PLoS One. 2012;7(6):e39514. 371. Knorr DA, Ni Z, Hermanson D, Hexum MK, Bendzick L, Cooper LJ, et al. Clinical- scale derivation of natural killer cells from human pluripotent stem cells for cancer therapy. Stem Cells Transl Med. 2013;2(4):274-83. 372. Sachamitr P, Hackett S, Fairchild PJ. Induced pluripotent stem cells: challenges and opportunities for cancer immunotherapy. Front Immunol. 2014;5:176. 373. Stanic AK, De Silva AD, Park JJ, Sriram V, Ichikawa S, Hirabyashi Y, et al. Defective presentation of the CD1d1-restricted natural Va14Ja18 NKT lymphocyte antigen caused by beta-D-glucosylceramide synthase deficiency. Proc Natl Acad Sci U S A. 2003;100(4):1849-54. 374. Fujii S, Shimizu K, Okamoto Y, Kunii N, Nakayama T, Motohashi S, et al. NKT cells as an ideal anti-tumor immunotherapeutic. Front Immunol. 2013;4:409. 375. Watarai H, Fujii S, Yamada D, Rybouchkin A, Sakata S, Nagata Y, et al. Murine induced pluripotent stem cells can be derived from and differentiate into natural killer T cells. J Clin Invest. 2010;120(7):2610-8. 376. Yamada D, Iyoda T, Vizcardo R, Shimizu K, Sato Y, Endo TA, et al. Efficient Regeneration of Human Vα24. Stem Cells. 2016;34(12):2852-60. 377. Brewer BG, Mitchell RA, Harandi A, Eaton JW. Embryonic vaccines against cancer: an early history. Exp Mol Pathol. 2009;86(3):192-7. 378. Li Y, Zeng H, Xu RH, Liu B, Li Z. Vaccination with human pluripotent stem cells generates a broad spectrum of immunological and clinical responses against colon cancer. Stem Cells. 2009;27(12):3103-11. 379. Zhang Z, Chen X, Chang X, Ye X, Li Y, Cui H. Vaccination with embryonic stem cells generates effective antitumor immunity against ovarian cancer. Int J Mol Med. 2013;31(1):147-53. 380. Zhang ZJ, Chen XH, Chang XH, Ye X, Li Y, Cui H. Human embryonic stem cells--a potential vaccine for ovarian cancer. Asian Pac J Cancer Prev. 2012;13(9):4295-300. 381. Dong W, Du J, Shen H, Gao D, Li Z, Wang G, et al. Administration of embryonic stem cells generates effective antitumor immunity in mice with minor and heavy tumor load. Cancer Immunol Immunother. 2010;59(11):1697-705. 382. Yaddanapudi K, Mitchell RA, Putty K, Willer S, Sharma RK, Yan J, et al. Vaccination with embryonic stem cells protects against lung cancer: is a broad-spectrum prophylactic vaccine against cancer possible? PLoS One. 2012;7(7):e42289.

260

383. Kooreman NG, Kim Y, de Almeida PE, Termglinchan V, Diecke S, Shao NY, et al. Autologous iPSC-Based Vaccines Elicit Anti-tumor Responses In Vivo. Cell Stem Cell. 2018;22(4):501-13.e7. 384. Kim J, Jung J, Lee SJ, Lee JS, Park MJ. Cancer stem-like cells persist in established cell lines through autocrine activation of EGFR signaling. Oncol Lett. 2012;3(3):607-12. 385. Cao L, Zhou Y, Zhai B, Liao J, Xu W, Zhang R, et al. Sphere-forming cell subpopulations with cancer stem cell properties in human hepatoma cell lines. BMC Gastroenterol. 2011;11:71. 386. Liao MJ, Zhang CC, Zhou B, Zimonjic DB, Mani SA, Kaba M, et al. Enrichment of a population of mammary gland cells that form mammospheres and have in vivo repopulating activity. Cancer Res. 2007;67(17):8131-8. 387. Min SO, Lee SW, Bak SY, Kim KS. Ideal sphere-forming culture conditions to maintain pluripotency in a hepatocellular carcinoma cell lines. Cancer Cell Int. 2015;15:95. 388. Balbous A, Cortes U, Guilloteau K, Villalva C, Flamant S, Gaillard A, et al. A mesenchymal glioma stem cell profile is related to clinical outcome. Oncogenesis. 2014;3:e91. 389. Jeter CR, Badeaux M, Choy G, Chandra D, Patrawala L, Liu C, et al. Functional evidence that the self-renewal gene NANOG regulates human tumor development. Stem Cells. 2009;27(5):993-1005. 390. Chen Y, Shi L, Zhang L, Li R, Liang J, Yu W, et al. The molecular mechanism governing the oncogenic potential of SOX2 in breast cancer. J Biol Chem. 2008;283(26):17969- 78. 391. Wang D, Lu P, Zhang H, Luo M, Zhang X, Wei X, et al. Oct-4 and Nanog promote the epithelial-mesenchymal transition of breast cancer stem cells and are associated with poor prognosis in breast cancer patients. Oncotarget. 2014;5(21):10803-15. 392. Huang ZJ, You J, Luo WY, Chen BS, Feng QZ, Wu BL, et al. Reduced tumorigenicity and drug resistance through the downregulation of octamer-binding protein 4 and Nanog transcriptional factor expression in human breast stem cells. Mol Med Rep. 2015;11(3):1647- 54. 393. Takamizawa J, Konishi H, Yanagisawa K, Tomida S, Osada H, Endoh H, et al. Reduced expression of the let-7 microRNAs in human lung cancers in association with shortened postoperative survival. Cancer Res. 2004;64(11):3753-6. 394. Imbert AM, Garulli C, Choquet E, Koubi M, Aurrand-Lions M, Chabannon C. CD146 expression in human breast cancer cell lines induces phenotypic and functional changes observed in Epithelial to Mesenchymal Transition. PLoS One. 2012;7(8):e43752. 395. Asiedu MK, Ingle JN, Behrens MD, Radisky DC, Knutson KL. TGFbeta/TNF(alpha)-mediated epithelial-mesenchymal transition generates breast cancer

261

stem cells with a claudin-low phenotype. Cancer Res. 2011;71(13):4707-19. 396. Gyamfi J, Eom M, Koo JS, Choi J. Multifaceted Roles of Interleukin-6 in Adipocyte- Breast Cancer Cell Interaction. Transl Oncol. 2018;11(2):275-85. 397. Fantozzi A, Christofori G. Mouse models of breast cancer metastasis. Breast Cancer Res. 2006;8. 398. Lelekakis M. A novel orthotopic model of breast cancer metastasis to bone. Clin Exp Metastasis. 1999;17. 399. Aslakson CJ, Miller FR. Selective events in the metastatic process defined by analysis of the sequential dissemination of subpopulations of a mouse mammary tumour. Cancer Res. 1992;52. 400. Bolin C, Sutherland C, Tawara K, Moselhy J, Jorcyk CL. Novel mouse mammary cell lines for in vivo bioluminescence imaging (BLI) of bone metastasis. Biol Proced Online. 2012;14. 401. Tao K, Fang M, Alroy J, Sahagian GG. Imagable 4T1 model for the study of late stage breast cancer. BMC Cancer. 2008;8. 402. Kim JB. Non-invasive detection of a small number of bioluminescent cancer cells in vivo. PLoS One. 2010;5. 403. Shi Y, Inoue H, Wu JC, Yamanaka S. Induced pluripotent stem cell technology: a decade of progress. Nat Rev Drug Discov. 2017;16(2):115-30. 404. Sullivan S, Stacey GN, Akazawa C, Aoyama N, Baptista R, Bedford P, et al. Quality control guidelines for clinical-grade human induced pluripotent stem cell lines. Regen Med. 2018;13(7):859-66. 405. Lutz MB, Kukutsch N, Ogilvie AL, Rössner S, Koch F, Romani N, et al. An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. J Immunol Methods. 1999;223(1):77-92. 406. Senju S, Haruta M, Matsunaga Y, Fukushima S, Ikeda T, Takahashi K, et al. Characterization of dendritic cells and macrophages generated by directed differentiation from mouse induced pluripotent stem cells. Stem Cells. 2009;27(5):1021-31. 407. Fairchild PJ, Nolan KF, Cartland S, Graça L, Waldmann H. Stable lines of genetically modified dendritic cells from mouse embryonic stem cells. Transplantation. 2003;76(3):606-8.

262

Titre: Stratégie d'immunothérapie des cancers: Modèle de Cancer du Sein Triple Négatif

Mots clés: Cancer du sein triple négatif, Cellules souches cancéreuses, Pluripotence, Immunothérapie

Résumé: Les cellules souches cancéreuses (CSCs) sont à l’origine de la progression tumorale, des métastases et des rechutes tardives. Elles ont été identifiées dans de nombreux cancers, comme le cancer du sein triple négatif (TNBC) et les cancers de grade III-IV. Elles sont résistantes aux chimiothérapies et radiothérapie et résident dans une niche immuno-répressive. Cette étude vise à évaluer une stratégie d’immunothérapie qui cible sélectivement les CSC dans le modèle murin 4T1-GFP-Luc mimant le TNBC. Le phénotype/ génotype des mamosphères a été initialement caractérisé. Basée sur l’analyse génomique des CSC, nous avons développé une immunothérapie active associée à des agents immuno-modulateurs. Nous avons mesuré la taille des tumeurs et suivi l’apparition des métastases par bioluminescence. Une étude immunologique et analyse génomique de la tumeur a été réalisée. La combinaison thérapeutique provoque le recrutement dans la tumeur de lymphocytes T (CD4+, CD8+) et lymphocytes B par augmentation de chimiokines, une réduction des lymphocytes T reg et cellules myéloïdes suppressives. Cette induction de réponse immunitaire provoque la diminution de la taille de la tumeur et des métastases. Cette nouvelle immunothérapie active de type vaccinale pourra être utilisée en association avec les traitements actuels pour des mesures prophylactiques et curatives dans une grande variété de cancers.

Title: Strategies of cancer immunotherapy: Model of Triple Negative Breast cancer

Keywords: Triple negative breast cancer, Cancer stem cell, Pluripotency, Immunotherapy

Abstract: Cancer stem cells (CSCs) are responsible for tumor progression, metastases, and late relapses. They have been identified in many cancers, such as triple negative breast cancer (TNBC) and in grade III to IV cancers. They are resistant to chemotherapy and radiotherapy and reside in an immuno-repressive niche. This study aims to evaluate an immunotherapy strategy that selectively targets CSCs in the mouse model 4T1-GFP-Luc mimicking TNBC. The phenotype / genotype of mammosphere was initially characterized. Based on genomic analysis of CSC, we have developed an active immunotherapy associated with immunomodulatory agents. We measured the size of tumors and monitored the appearance of metastases by bioluminescence. We performed an immunological study and genomic tumor analysis. The therapeutic combination causes the recruitment of CD4+ and CD8+ T lymphocytes and B lymphocytes with increase of chemokines, the reduction of Tregs and suppressive myeloid cells in the tumor. This induction of intra-tumor immune response leads to a decrease in tumor size and metastases. This new active immunotherapy can be used in combination with current treatments for prophylactic and curative measures in a wide variety of cancers.

Université Paris-Saclay Espace Technologique / Immeuble Discovery Route de l’Orme aux Merisiers RD 128 / 91190 Saint-Aubin, France