Approaches to Improve the Proliferation and Activity

APPROACHES TO IMPROVE THE

PROLIFERATION AND ACTIVITY OF NATURAL

KILLER CELLS FOR ADOPTIVE CELL THERAPY

EVELYN OJO

Submitted in partial fulfillment of the

requirements for the degree of Doctor of

Philosophy

Thesis Advisor: David Wald, MD/PhD

Department of Pathology

CASE WESTERN RESERVE UNIVERSITY

January, 2019

CASE WESTERN RESERVE UNIVERSITY

SCHOOL OF GRADUATE STUDIES

We hereby approve the thesis/dissertation of

Evelyn Ojo

candidate for the PhD degree *.

(Signed) Alan Levine, PhD (committee chair)

David Wald, MD/PhD

John Wang, PhD

Mark Jackson, PhD

Sanford Markowitz, MD/PhD

Clive Hamlin, MS/PhD

(Date) 07/09/18

*We also certify that written approval has been obtained for any proprietary material

contained therein.

Dedication

I dedicate my thesis to the Almighty God who has been and will always be my inspiration to pursue excellence and touch the world with goodness. I am grateful to God for providing all I need to successfully undergo this training to empower me to serve manki nd through community service and pursuit of knowledge.

Table of Contents

List of Figures 6

List of Abbreviations 8

Acknowledgments 9

Abstract 11

Chapter 1. Immunosurveillance and immune evasion 1 3

1.1: Cancer immunosurveillance 14

1.2: Immunoediting by cancer cells promotes immune evasion 17

1.3: Immune suppression induced by cancer cells and the tumor 20 microenvironment

Chapter 2: Natural killer cell biology in health, disease, and cancer therapy 3 7

2.1: NK cell development and subsets 38

2.2: NK cell functions in health and disease 43

2.3: NK cell surface receptors mediate activation and function 52

2.4: Memory - like NK cell 59

2.5: Adoptive NK cell therapy 6 2

Chapter 3: Other immunotherapies 71

3.1: Tumor infiltrating lymphocytes 72

3.2: Lymphokine - activated killer cells 7 6

3.3: Cytokine - induced killer cells 77

3.4: Engineered T Cell Receptor T cells 79

3.5: Chimeric Antigen Receptors T cells 81

3.6: Bi - specific T cell Engagers 8 5

3.7: Cytokines and monoclonal antibodies 87

Chapter 4 : Membrane bound IL - 21 - based NK cell feeder cells drive robust 96 expansion and metabolic activation of NK cells

Abstract 97

Introduction 98

Materials and Methods 10 1

Results 104

Discussions 121

Chapter 5 : Inhibiting TGF - β signaling preserves the function of highly 125 activated, in vitro expanded natural killer cells in AML and colon cancer models

Abstract 126

Introduction 127

Materials and Methods 13 0

Results 135

Discussions 149

Chapter 6: Discussion and Future Directions 151

6.1: Conclusion and discussion 152

6.2: Future directions 155

References 177

L ist of Figures

Chapter 1

Fig 1.1: An illustration of mutations in TGF - β receptors associated with 27 various cancers.

Chapter 2

Fig 2.1: NK cell licensing enhances NK cell cytotoxic function against 42 MHC - I deficient tumor cells

Fig 2.2 : Mechanisms of NK cell functions 48

Chapter 3

Fig 3.1 : The structure of human IgG1 antibody 92

Chapter 4

Fig 4 .1: OCI - AML3 cell line exhibits low HLA - I expression and modest 104 fold expansion

Fig 4.2 : Validation of successful expression of mbIL - 21 on OCI - AML3 (NKF) 105

Fig 4.3: NKF feeder cells enable NK cell proliferation 107

Fig 4.4 : NKF - NK cells exhibit potent cytotoxic activity against both 109 hematologic and solid cancer cells

Fig 4.5 : mbIL - 21 signaling in NKF - NK cells leads to marked changes 111 in cell surface phenotype

Fig 4.6: NKF - NK cells expand beyond 3 weeks and are ex panded with 113 G - Rex flask

Fig 4.7 : mbIL - 21 signaling promotes increased metabolic activity in 116 NKF - NK cells but does not affect cytotoxicity

Fig 4.8 : NKF - NK cells reduce tumor burden in mouse xenografts 119 and improve mouse survival

Chapter 5

Fig 5 .1: Expanded NK cells demonstrate increased cytotoxicity against 135 HCT116 and HT29 cells as compared to fresh, IL - 2 activated cells.

Fig 5 .2: Sustained exposure to pathologic levels of TGF - β impairs the 137 function of highly activated, expa nded NK cells

Fig 5 .3: Inhibiting TGF β signaling using the small molecule kinase inhibitor 141 LY2157299 preserves the cytotoxic function of expanded NK cells, even after sustained exposure to pathologic levels of TGF - β.

Fig 5 .4: TGF - β signaling impairs the production of TNF - alpha, IFN - gamma, 142 perforin and granzyme B by ex vivo expanded NK cells.

Fig 5 .5: TGF - β inhibition enhances activated NK cell function in a colon 145 cancer model of liver metastases

Fig 5 .6: Robust NK cell infiltration into liver tissue observed in mice who 146 received TGF - β inhibition in addition to NK cell infusion

Fig 5.7 : Representative images of colon cancer metastasis xenograft showing 147 CD45 IHC of liver FFPE sections staini ng for human NK cells.

List of abbreviations

ACT - Adoptive cell therapy ADCC - Antibody - dependent cellular cytotoxicity CAR - Chimeric antigen receptor DC - Dendritic cell DNAM - 1 - DNAX accessory molecule γδ - T cells - Gamma - delta T cells GVHD - Graft - versus - host disease HSCT: Hematopoietic stem cell transplant IFN - γ - Interferon - gamma IL - 2 - NK - IL - 2 - overnight activated NK cells iNK cells - Immature NK cells, stage 3 of development KIRs - Killer immunoglobulin - like receptors mbIL - 21 - Membrane - bound in terleukin 21 MDSCs - Myeloid - derived suppressor cells MHC - I: Major histocompatibility complex - I MICA/B - MHC class I chain - related - A/B NCR - Natural cytotoxic receptor NKF - NK - NKF - expanded NK cells NK cells - Natural killer cells OCI - NK - OCI - AML3 - expanded NK c ells TGF - β – Transforming growth factor - β TNF - α - Tumor necrosis factor - α TRAIL - TNF - related apoptosis inducing ligand

Acknowledgments

My utmost gratitude to my family and friends who tirelessly encouraged me and strengthened my willpower to complete my PhD training. In addition, I thank all previous and current mentors, both formal and informal, who have provided me with resources and advice that helped me to navigate the various academic experiences I chose to pursue. I thank all my teachers , professor s, and deans for always believing in me and providing encouragement during the times I fell short of expectations.

I am ever grateful to my P.I. and mentor, Dr. David Wald, for his guidance, encouragement, and posit ivity during my PhD training. Indeed he provided a nurturing learning and work environment in the lab that encouraged me to express myself freely and pursue projects that I found interesting. I thank him for putting together a lab full of some of the nicest, kindest, hardworking, and warmest peo ple on the planet. The Wald lab members were instrumental in my success as a graduate student and will forever remain in my heart.

I thank my committee members, including past ones, for their suggestions and dedication to training me to be an excellent graduate student and researcher. I appreciate their flexibility in adjusting to sudden cancellations and arrangements. I am grateful t o the

MSTP at CWRU , the pathology department, and the Office of Diversity Initi atives and

Community Engagement for the opportunity of a lifetime to spend considerable time learning how to think critically, ask relevant questions that has the potential to i mprove many lives, and design a plan to answer that question. I thank the administration and students for their willingness to help whenever needed, and their critical advice.

Lastly, I thank my church family, Jesus House Cleveland, for welcoming me and s upporting me during my training, providing both spiritual and physical nourishment.

Their continual support helped me to overcome the toughest days.

Approaches to Improve the P roliferation and Activity

of Natural Killer Cells for A doptive C ell Therapy

Abstract

EVELYN OJO

Cancer is the second - leading cause of death in the United States and worldwide leading to 8.8 million deaths in 2015. Despite major advancements in science and health, patients at advanced stages of particular cancer types experience significant mortality. Cancer patients often relapse following treatment with conventional therapeutic modalities such as chemotherapy and radiation therapy. C onventional treatment approaches do not successfully address the heteroge neity present in most solid tumors , result in severe off - target effects resulting in severe side effects, an d fail to kill dormant cancer cells.

Immunotherapy is the use of immune cells as tools to comb at diseases . Natural killer

(NK) cells are lymphocytes that lyse tumor cells and virally - infected cells through diverse mechanisms including release of cytotoxic granules. NK cell therapy has shown muc h promise in preclinical studies however it fails to demonstrate efficacy in various cancer popula tions . This body of work tackles two cru cial barriers to NK cell therapy in

patients with solid tumors.

NK cell numbers in the tumor has been correlated with good prognosis in cancer patients.

Obtaining clinically efficacious numbers of NK cells is limite d by the relatively small proportion of NK cells in peripheral blood. We , therefore, designed and engineered a novel IL - 21 - based NK cell expansion platform consisting of OCI - AML3 cells transduced with membrane - bound IL - 21 (NKF cells) . We demonstrated that NKF cells are effective at expanding NK cells. The NK cells expanded using NKF cells were able to lyse a wide array of tumor types and demonstrated a favorable metabolic signature as compared to non - expanded NK cells. Additionally, NKF - expanded NK cells si gnificantly reduced the tumor burden in sarcoma cell - infected mice.

Cancer cells have developed mechanisms to evade anti - tumor activities of NK cells, dampening the efficacy of NK cell - mediated lysis of cancer cells. Transforming growth factor - β (TGF - β) is an immunosuppressive cytokine abundantly secreted by cancer cells to modify the cancer landscape to support the growth and development of the tumor. We demonstrated that inhibition of TGF - β using the clini cal grade TGF - β inhibitor, galunisertib, resulted in rescue of NK cell cytokine secretion and cytotoxic function.

In conclusion, this work has evaluated the NKF expansion system and demonstrated its ability to robustly expand NK cells. NKF - expanded NK cells have increased glycoly t ic and oxidative phosphorylation rates and continue to expand beyond 5 weeks of expansion. Feeder cell - expanded NK cells demonstrate superior cytotoxic function following inhibition of TGF - β signaling. Infusion of NKF - expanded NK cells coupled with TGF - β i nhibition has the potential to result in decreased tumor burden in cancer patients.

CHAPTER 1: IMMUN OSURVEILLANCE AND IMMUNE EVASION

1.1: CANCER IMMUNOSURVEILLANCE

Understanding the role of the immune system in regulating cancer development and

progression, or cancer immunosurveillance, originated from researchers like Paul Ehrlich

in the early 1900s [1] . His claims were verified in the late 1900s, following extensive

studies o f the immune system, the discovery of tumor - specific antigens, and mechanisms

of graft rejection [2, 3] . Old et al. defined ‘adoptive immunity’ as elicitation of the immune system by injection of immu nologically active cells systemically [2] . The cancer

immunosurveillance theory postulated that the immune system was responsible for

continuous eradication of transformed cells [4] . The absence of detectable difference in

the development of sarcoma and lung adenoma i n “immunocompromised” (athymic nude

mice) versus immune - competent mouse models led to the abandonment of the

immunosurveillance hypothesis [5 - 8] . Retrospective analysis of the experiments

denouncing immunosurveillance, performed by Rygaard and colleagues, have revealed

that the interpretation of t h o se early studies using the “immunocompromised” (nude)

mice were flawed [8] . Although possessing fewer T cells, nude mice maintain

functioning T cells, making it difficult to predict the effects these cells have on tumor

formation compared to wild - type mice [4, 9, 10] . Other reasons for the invalidation of the

early studies using the nude mice includes the likely presence of other thymic -

independent lymphocytes such as natural killer (NK) cells and the increased sensitivity of

the chosen mice strain to the carcinogen used to challenge the mice [11, 12] .

The cancer immunosurveillance theory was revived with the discovery of a subset of

lymphocytes that exhibited natural cytotoxic activity against various tumor types without

prior exposure to these tumors [13] . However, it would take several years for the

14 successful isolation and identification of NK cells [4] . T he discovery of the importance of perforin and interferon - γ (IFN - γ) in countering chemical - induced tumor formation further strengthened the cancer immunosurveillance theory. Perforin is a cytotoxic molecule released by T cells and NK cells following activation, and is important in mediating the killing of tumor cells. C57BL/6 mice lacking perforin were more prone to spontaneou s lymphomas and methylcholanthrene (MCA) - induced tumor formation compared to wild - type mice [14 - 17] . These studies indicated that components of the immune system play a role in mitigating tumor development. The development of mice lacking the recombination activating gene 1 (RAG1) or RAG2, and IFNGR1 receptor made it possible to test the extent of the immunosurveillance in controlling tumor development

[18] . RAG1 - / - or RAG2 - / - mice lack the RAG1 or RAG2 gene s in only lymphoid cells.

Tumor formation in 129/SvEv mice lacking the IFNGR1 receptor or STAT1, RAG2 - / - mice, and mice lacking both RAG2 and STAT1 (RkSk mice) expression were compared to syngeneic wild - type mice [19] . The non - wild - type mice lines developed three times more chemically - induced t umors compared to wild - type mice. In addition , RkSk mice developed spontaneous breast tumors not observed in the RAG2 - / - or wild - type mice.

In addition to murine studies, studies in humans have also supported the cancer immunosurveillance theory. Early st udies showed that immunosuppressed transplant patients had significantly higher risk for developing cancer [12, 20, 21] . Review of the data from the Cincinnati Transplant Tumor Registry (CTTR) in 1996 showed a 2 - fold higher risk of developing melanoma in transplant patients as compared to the general population [22] . Further analysis of registry data from the CTTR revealed increased incidence of lymphoma, kaposi’s sarcoma, carcinoma of the skin and lips, and in situ

carcinoma of the uterine cervix and anogenital area [23] . Lung tumor formation was 25 - fold higher in 608 cardiac transplant p atients than in the general population, with no known viral etiology [24] . Furthermore, there have been consistent correlation between the numbers of tumor - infiltrating lymphocytes (TILs) in the tumor and patient survival.

This correlation has been observed in patients with melanomas [25] , neuroblastoma [26] , and cancers of the breast [27] , ovary [28] , and colon [29, 30] . In aggregate these studies

form strong support for cancer immunosurveillance and its pr omising role in the

development of cancer therapeutics.

1.2: IMMUNOEDITING OF CANCER CELLS PROMOTES IMMUNE EVASION

Studies have shown that while the immune system can be anti - tumorigenic, it can also

promote tumor development and metastasis by selecting for cancer cells that are less

immunogenic [4, 31] . Immunoediting, the process by which immune cells interface with tumor development and progression, is divided into three phases, elimination, equilibrium, and escape.

The elimination phase is characterized by bulk cancer cell death mediated by the cells of the immune system . Immunosurveillance occurs during this phase of robust immune cell response. As the tumor g rows and invades the basement membrane, it disrupts the surrounding tissue and epithelial cells. These cells release stimulatory signals, recruiting cells of the innate immune system such as NK cells, macrophages, and dendritic cells

(DCs) [32] . Proangiogenic proteins released by transformed or stromal cells aid the migration of immune cells to the tumor site [33] . Indeed, angiogenesis has been shown to be an inducer of tumor formation and progr ession and is evident at advanced cancer stages. Innate immune cells recognize stressed proteins on the transformed cells, and produce inflammatory molecules such as IFN - γ [34] . IFN - γ exerts anti - proliferative effects and apoptosis in tumor cells, and secretion of chemokines CXCL10 and CXCL9 by tumor and stromal cells [35 - 38] . CXCL10 induces chemotaxis of T cells and has ant iangiogenic properties, inhibiting the formation of new blood vessels and causing more tumor cell death [39] . NK cells and macrophages lyse tumor cells by interleukin

(IL) - 12 downstream signaling involving the production of perforin, tumor necrosis factor

(TNF) - related apoptosis - inducing ligand, ROS and reactive nitrogen species (NO) [4,

34] . DCs phagocytize debris from dead tumor cells and home to draining lymph nodes.

There, DCs induce Th1 cells which drive the development of tumor - specific CD8+ T cells. These antigen - specific CD8+ T cells home to the tumor site and kill the rest of the antigen - bearing tumor cells, especially following IFN - γ - mediated increase in immunog enicity [4, 34, 40] . The complete obliterat ion of the tumor cells means that the elimination phase encompasses the immunoediting process and does not continue to the equilibrium phase. G enomic instability , however, drives the generation of tumor cells with dampened immunogenicity , including reduced expression of tumor specific antigens , shifting the cancer immunoediting process into the equilibrium phase.

The equilibrium phase is characterized by lymphocytes exerting potent selective pressure on the tumor cells that survived the elimination phase. A s the equilibrium phase continues, many tumor cells are killed but new genetically variable and unstable clones of tumor cell s develop with mutations that provide them with increased resistance to attack from the immune system. This phase is believed to pe rsist for years and may be the longest phase in the immunoediting process [4] .

As the equilibrium phase progresses, tumor ce ll variants that are resistant to immune cell detection and elimination divide rapidly and uncontrollably till the patient develops clinically observable disease. This escape phase is characterized not only by irrepressible expansion of resistant tumor cel ls but als o by waning host defense tumor - directed attacks caused by tumor - derived and non - tumor - derived immunosuppression [4, 31] . Th is immunosuppression is driven by molecular or cellular mechanisms. Some of these inhibitory mechanisms are preexisting systems evolutionarily developed to prevent overt immune stimulation and subsequent tissue destruction, that have been hijacked by the

18 tumorigenic system. These mechanisms will be discussed below in the context of how they aid immune escape of cancer cells.

1.3: IMMUNE SUPPRESSION INDUCED BY CANCER CELLS AND THE

TUMOR MICROENVIRONMENT

Immunosuppressive Cells

Tumor - associated cells, especially in the tumor microenvironment, are a major source of immunosuppression. These suppressive cells are heterogeneous in nature and are derived from both myeloid and lymphoid progenitors. Globally, these cells are referred to as regulatory cells because they limit or ‘regulate’ the anti - tumor activities of immune cells.

Although these regulatory cells inhibit effector immune cell function, they have been found to be higher in number and more immunosuppressive in cancer patient s [41] .

Regulatory cells include regulatory T cells (Tregs), myeloid - derived suppressive cells

(MDSCs), and tumor - associated macrophages (TAMs).

Regulatory T cells

Regulatory immune cells, such as CD4+ CD25+ FoxP3+ Tregs, induce robust suppression of effector immune cells, facilitating escape of tumor cells. Naturally, Tregs play an important role in maintaining homeostasis and preventing overt T cell - mediated autoimmunity [42] . Fo ntenot and colleagues’ study of the role of FoxP3 in the development and function of Tregs revealed the importance of Tregs in maintaining the equilibrium of pr o - and anti - inflammatory forces [43] . Mice carrying a deletion of, or loss - of - function mutations in, the FoxP3 gene, not only lack ed mature and functional

Tregs but they die d by 4 weeks of age [43] . Overt CD4+ T cell function led to an autoimmune phenomenon that could be prevented with injection of FoxP3+ Tregs into

FoxP3 mutant neonatal mice [43] . Cytokines such as TGF - β can increase Treg activity by

upregulating FoxP3 expression. Tregs are recruited to the tumor site via tumor cell -

derived chemokines and prostaglandin E2 (PGE2) which upregulates FoxP3 expr ession

[31, 41, 44] . Also, tumor - derived transforming growth factor - β (TGF - β) supports the conversion of CD4+ T cells into Tregs [44] .

Tregs inhibit immune cell effector activity through cytokine secretion and cel l - to - cell contact. It is believed that Tregs suppress antigen - reactive T cells through IL - 10 production, thereby preventing robust anti - tumor activity of cytotoxic T cells, since most tumor antigens are self - antigens. Also, Tregs induce the expression of B 7 - H4 by APCs.

B7 - H4+ APCs can then induce T cell cycle arrest, limiting T cell proliferation [44] . A minor Treg - mediated mechanism of suppression of T and NK cell function in vitro is competition for IL - 2, which in turn leads to impaired activation and activity of these effector cel ls. Tregs have also been reported to lyse activated effector T cells in a perforin - dependent manner [45, 46] . Therefore, reduction in Treg activity in cancer patients may maintain effector T cell function and enhance tumor clearance. Inhibition of

Treg ac tivity by direct a dministration of anti - CD25 monoclonal an tibody induced anti - tumor immunity [41, 44] . Treg suppressive effects on T cell IL - 2 synthesis and IFN - γ secretion is observed to be reversible following ex vivo separation of Tregs from ef fector

T cells [45] .

Myeloid derived suppressor cells

MDSCs are another type of regulatory cells that dampen anti - tumor immunity. They are described as a heterogeneous group of immature myeloid cells consisting of precursors of

DCs, granulocytes, and macrophages [47] . MDSCs are induced, expanded, and recruited by co - stimulatory molecules such as IL - 1β, PGE2, and vascular endothelial growth factor

(VEGF) [31, 47, 48] . Similar to Tregs, MDSCs are critical immunosuppres sors of effector T cell activity. CD11b+Gr1+ MDSCs downregulate the ζ - chain of the TCR, impairing T cell activation, and T cell proliferation in cancer and in inflammatory states

[31, 47] . Impaired T cell proliferation significantly di minishes the number of cytotoxic effector cells available to mediate an effective adaptive immu ne response . MDSCs exert their immunosuppression mainly through upregulation of nitric oxide (NO) synthase and arginase - 1, and subsequent deletion of L - arginine. L - arginine is required for protein synthesis and increased synthesis of ROS and NO by T cells. NO mediates apoptosis of T cells, induces nitrosylation of TCR on tumor - infiltrating lymphocytes (TILs) and chemokines in the tumor environment. MDSCs can also sequester cysteine and downregulate L - selectin, essential for T cell homing to lymph nodes [47] .

In addition to their role as potent T cell suppressors, MDSCs actively promote tumor growth and metastasis via angiogenesis . MDSCs induce angio genesis by secreting VEGF and basic fibroblast growth factor (bFGF), and may aid in the establishment of a metastatic niche. T he immunosuppressive effects of MDSCs include the present ation of antigens to Tregs, generating tumor - specific Tregs, and the secr etion of immunosuppressive cytokines such as TGF - β and IL - 10 [49] . Mice melanoma models reveal a correlation between MDSC levels and tumor burden, chronic inflammation and immunosuppression, and decreased numbers of mature DCs [47, 48] . Importantly, MDSC

level in cancer patients is associated with low overall survival and tumor recurrence [49,

50] .

Tumo r - associated macrophages

Tumor - associated macrophages (TAMs) promote tumor growth mainly through cytokine secretion. Macrophages are innate immune cells of monocytic lineage that can alter their function to adapt to various environmental stimuli [51] . They are phagocytic in nature

and act as professional antigen - presenting cells to activate the adaptive immune system.

Macrophages undergo polarization and are conventionally categorized into M1 or M2

populations. Initially, reports indicated that macrophages secrete ROS and NO to hamper

tumor growth, and chemokines to attract T cells to the tumor site. macrophages [51, 52] .

However, data countering this notion emerged , especially data from a study that showed that inhibiting macrophage infiltration of tumors in mice nullified tumor progression and metastasis [53] . These conflicting reports led to the grouping of macrophages, based on overall activity, into M1 and M2 macrophages. The M1 macrophages are the classically activated macrophages that are pro - inflammatory, and aid in host de fense and tumor clearance [51] . These macrophages express major histocompatibility complex (MHC) - II,

CD68, and CD80, which are costimulatory molecules that support macrophage - mediated

CD4+ T cell activation. The M2 macrophages secrete ant i - inflammatory cytokines and in general act to resolve inflammation. Due to their immunosuppressive activities that promote tumor growth, M2 macrophages are termed tumor - associated macrophages. M2 macrophages expressing CD11b and F4/80 synthesize high leve ls of VEGF, IL - 10, and

TGF - β that suppress DC differentiation and function [31] . These immature DCs (iDCs)

are functionally - impaired and express low amounts of the costimulatory molecules

CD80, significantly impairing T cell functions. Furthermore, iDCs are known to promote

the expansion of Treg cells in mice, while inhibiting effector T cell proliferation in vitro, leading to further immunosuppression [31, 44, 54] .

As discussed, regulatory cells play a vital role in maintaining a balance between pro - and

ant i - inflammatory tension. These cells secrete potent immunosuppressive factors, most notably TGF - β and IL - 10, to inhibit T cell effector function and prevent autoimmunity.

Cancer cells have been able to manipulate the activity of these cells in to hinder ing

i mmune cells and also promot ing tumor growth. In addition, these regulatory cells can, with the aid of cytokines, prevent DCs from maturing and thereby convert them into regulatory cells that impair effector cell functions. Blockade of regulatory cell activ ity in cancer may significantly aid in maintaining and enhancing the antitumor activities of the immune system. Caution is needed, however, to avoid rampant immune activation and autoimmune pathology.

Immunosuppressive Cytokines

Cytokines are a large gro up of signaling molecules that are critical for immune cell function. Cytokines mediate cell - cell communication and regulate cellular growth and proliferation. Immunosuppressive cytokines are important mediators of immune cell suppression, serving as a cou nterforce to the pro - inflammatory actions of immune cells.

Cancer cells have been able to hijack these molecules, how e ver, to promote their growth and metastases. A prominent and widely suppressive molecule is TGF - β. As mentioned

earlier, high amounts of T GF - β secreted by tumor - associated stromal cells convert

effector T cells to regulatory T cells. This section provides a more detailed overview of

the immunosuppressive activities of TGF - β and IL - 10.

TGF - β

TGF - β is a member of a large group of secreted gro wth factors known to impact all

aspects of development [55] . The isoforms of TGF - β are TGF - β1, TGF - β2, TGF - β3, but

the TGF - β1 is the mos t expressed in the immune system [42] . TGF - β1 (TGF - β) is

secreted by many cell types, but is highly secreted by tumor cells, and stromal cells in

lymphoid organs [56] . Secreted in its latent form bound to latency - associated

polypeptide, proteolytic cleavage of the inactivating component frees active TGF - β to engage its receptors [42] . The receptors for TGF - β are expressed broadly on all immune cells and signal through a number of pathways such as the SMAD - mediated pathway, supporting the pleiotropic effects of TGF - β. TGF - β signaling in a healthy immune system serves to regulate NK and T cell expansion and eff ector functions, facilitate IgA class switching, restrict autoimmunity, and maintain homeostasis [42] . In fact, ablation of

Tgfb1 gene is embryonically lethal in mice, and TGF - β defici ency leads to severe autoimmune disease [57] . Moreover, differentiation of thymic pre - T cells in vitro into the distinct effector, regulatory, and memory lineages is regulated by TGF - β [42, 58 - 60] .

TGF - β signaling impacts the activities of multiple immune cell types such as T cells, NK cells, and DCs. TGF - β has been shown to inhibit the cytolytic activi ty of CD8+ T cells in cancer patients by downregulating the expression of the cytotoxic molecules IFN - γ,

perforin, and granzyme A [42, 45] . In addition, TGF - β signaling leads to downstream

repression of cell cycle genes such as c - myc, impairing the expansion of cytotoxic T

lymphocytes (CTLs) [42] . Likewise, TGF - β impairs the cytotoxic function and cytokine

production of NK cells [55] . TGF - β impairs antigen presentation by DCs and hampers cytokine production [55] . As described earlier, TGF - β has been shown to p romote the expansion of Tregs and upregulation of FoxP3 in vitro [42] . Interestingly, t umor cells

have been observed to be unresponsive to TGF - β or require high doses to be sensitive

[61 , 62] . The mechanism that reverses TGF - β - mediated growth inhibition in cancer cells has not been elucidated. One hypothesis is that th e polymorphism of TGF - β receptors might be driving the unresponsiveness. A common polymorphism of TGF - βRI, TGF - βRI

(6A), has been discovered to transduce growth inhibitory signaling less effectively as

compared to TGF - βR1 [56] . Additionally, mutati ons in either TGF - βRI or TGF - βRII have been discovered in several tumor types including small - cell lung cancer, pancreatic cancer, breast and colorectal cancers, as illustrated in Fig 1 .1 [56] . Also, some tumors

have decreased expression of TGF - β r eceptors, enabling them to avoid the growth inhibitory effect of TGF - β signaling [56, 63, 64] . TGF - β blockade is one of the cancer treatment strategies that are being actively pursued, and may aid in improving patient response to treatment and patient outcome.

Fig ure 1. 1 An illustration of mutations in TGF - β receptors associated with various

cancers. Retrieved from Boris, 2001 [56] .

IL - 10

IL - 10 is another pleiotropic cytokine that broadly impacts key players in the tug - of - war between cancer cells and the immune system. IL - 10 is secreted by almost all the cells of the immune system including all T cell subsets, and by cancer cells, emphasizi ng its role in aiding in tumor escape [45, 65, 66] . IL - 10 suppresses cytokine production by monocytes, Th1 c ells, and macrophages, dampening their puissance to muster an effective immune response against transformed cells [67, 68] . IL - 10 - mediated reduction of IL - 12 production has far - reaching consequences, most notably the inhibition of IFN - γ secretion and subsequent impaired development of Th1 cells [68] . IL - 10 impairs DC

functiona lity by impairing DC maturation and downregulating CD1 expression on DCs,

and downregulates TAP1 and TAP2 thereby impairing cytotoxic T cell - mediated tumor

elimination [65, 69, 70] . In addition to blocking the expression of key components of the antigen processing machinery, IL - 10 also downregulates the expression of MHC class I on tumor cells, reducing their sensitivity to MHC - I - restricted T cell s but increasing sensitivity to NK cells [67] . The effect of IL - 10 on effector cells include suppression of T cell proliferation, downregulation of MHC - II expression on monocytes and DCs, and downr egulation of costimulatory molecules causing both CD4+ and CD8+ T cells to undergo anergy [70, 71] . IL - 10 inhibit s NO production by monocytes and promote s decreased ability to reject tumors [68] . IL - 10 has been shown to impair effector cell function at controlling the growth of immunogenic tumors in vivo [68] . The consistent presence of IL - 10 in tumor biopsies and ascites from cancer patients suggests that it may

play a vital role in maintaining the tumors [67] . IL - 10 may promote metastases of tumor

cells due to its higher levels in metastatic melanoma as compared to primary lesions [67] .

Despite evidence describing an immunosuppressive role for IL - 10, other studies have

painted a more immunostimulatory picture of the role of IL - 10. IL - 10 has been shown to be immunostimulatory by enhanci ng T and B cell responses in a Th1 - dependent mechanism [72] . IL - 10 in the presence of other cytokines such as IL - 2 synergistically stimulates the expansion and cytotoxicity of CD8+ T cells and NK cells [71, 73] .

Emmerich and colleagues demonstrated in mice an IL - 10 - mediated ovarian cancer tumor rejection through the activation of CD8+ T cells [74] . In fact, a small study conducted revealed a higher incidence o f defective IL - 10R pathway in B - cell lymphoma patients compared to the general population, suggesting that IL - 10 pathway may play a role in

preventing lymphomagenesis [75] . IL - 10 has been reported to support the differe ntiation of B cells into antibody secreting cells and rescues T cells from undergoing apoptosis

[71] . While IL - 10 neutralization may be a promising treatment strategy for some cancers, in inflammation - driven cancers such as colon cancer, IL - 10 secreted by Tregs might be

therapeutic by re ducing the synthesis of pro - inflammatory cytokines [72] . Overall,

targeting IL - 10 as a cancer therapy strategy is an interesting pursuit but more work is

needed to elucidate the critical sources of IL - 10 and delineate the precise targets and how the ef fects of IL - 10 differs at various time in cancer development.

Tumor - induced anergy

The idea that tumors cause CD4+ T cell tolerance was initially observed when

hemagglutinin - specific naïve T cells injected into mice with established lymphoma expressing HA exhibited diminished responses to the antigen in vitro [76] . Following

injection in vivo, the naïve CD4+ T cell underwent clonal expansion and phenotypic

changes were observed associated with antigen recognition. However, the cells could not

be primed in vivo following vaccination with a p otent immunogen, suggesting that the

cells were anergic [69, 76] . This T cell tolerance has been observed in both hematologic

and solid tumors [76 - 79] . It was first believed that the solid tumors evade the immune

system due to the extra - lymphatic location of the former , however studies have identified

tumor - infiltrating lymphocytes in which anergy is tumor - mediated [77, 80] . Supporting

this finding is the functionally impaired state of circulating T cells from melano ma

patients [80] . Interestingly, only a subset of antigen - experienced T cells is anergic while

the rest are naïve or ‘ignorant’ of the tumor cells. The culprit of the lack of response of T

cells to the tumor cell are dysfunctional APCs especially dysfunctional DCs [54, 77, 81] .

It has been discovered that antigen capture by iDCs in a non - inflammatory environment

leads to antigen - specific T cell deletion [81] . In an inflammatory environment, DCs are

able to mature and appropriately prime T cells. I n the tumor microenvironment however ,

immunosuppressive molecules such as IL - 10 and TGF - β inhibit t he maturation of DCs,

promoting T cell anergy [69] .

T cells undergo phenotypic and functional cha nges early during cancer development that

lead to unresponsive T cells [76] . CTLA - 4, a key regulator of T cell activation has been

demonstrated to induce anergy in T cells early in cancer growth [82] . Inhibition of its

signaling and its effect on the mainten ance of T cell antitumor activities is discussed

briefly in a later section. Another strategy to overcoming T cell anergy is combining T

ce ll co - stimulation via OX40 with an immunostimulatory factor such as GM - CSF. It has

been shown that OX40 ligation increases CD4+ T cell expansion and function, the

development of T cell memory subsets, and mediates the reversal of anergy in CD4+ T

cells [83] . Combining OX40 co - stim ulation with GM - CSF - secreting vaccine led to expansion and rekindling of the effector functions of CD8+ T cells in a mouse model of established CD8+ T cell tolerance [ 84] . Although tumor - induced anergy is one of the main obstacles to establishing effective anticancer strategies, better understanding of the various components that drive anergy may lead to the development of more efficacious blockade strategies.

Impa ired antigen presentation

Immunoediting comprises of immune cell - mediated lysis of tumor cells and immune -

mediated shaping of the tumor, resulting in impaired antigen presentation and reduced

immunogenicity. Some cancers such as lung and renal carcinoma evade detection by the

immune system by downregulating MHC - I complexes on tumor cells [85] . Genetic

instability of tumor cells enable them to develop mutations, most commonly in the β2 -

microglobulin gene, leading to dysfunctional β2 - microglobulin subunit [86] . In addition,

tumors can evade the immune system by inhibiting the expression of various members of

the antigen processing machinery such as transporter for antigen presentation (TAP) and

tapasin [86, 87] . TAP is an ATP - dependent peptide transporter, and cells defici ent in

TAP cannot transport cytosolic peptides into the ER [85] . This deficiency therefore leads

to a reduction in the supply of peptides available for binding to MHC - I, making the tumor

cells more susceptible to killing by NK cells. Despite the i ncreased sensitivity to NK cell

killing, other immunosuppressive factors impair NK cell - mediated lysis of tumor cells.

Tapasin, another critical protein involved in presentation of antigens, is a chaperone

involved in the assembly of the heavy chain of MHC - I [86] . Specifi cally, tapasin

facilitates the association of TAP to the heavy chain complex. Lack of expression of

tapasin may significantly impair antigen presentation, causing immune cells to ‘ignore’

cancer cells.

Additionally, downregulation of co - stimulatory molecu les results in impaired antigen presentation and is a method by which tumor cells can evade T cells. Melanoma cells have been reported to downregulate co - stimulatory molecules necessary for antigen - specific T cell activation [47, 85] . Absence of costimulatory molecules may cause both

CD4+ and CD8+ T cells to undergo anergy. In addition to impaired antigen processing

machinery, cancers may generate tumor variants with low antigen presentation [69] .

Expansion of antigen loss variants may enable tumor cells to escape lysis by cytotoxic

lymphocytes, enabling the tumor to progress. A f ew reports of specific antigen loss in

tumors from relapsing patients and patients unsuccessfully treated with peptide vaccines

supports this concept of adaptive low antigen presentation [88 - 90] . T cells are dependent on appropriate antigen presentation for activation, therefore strategies to increase the immunogenicity of tumor cells may improve the efficacy of immune therapy.

Stat3 signaling in tumor cells

Oncogenic signaling pathways not only enable cancer cells to evade detection by immune

cells but also modulate t he immunologic microenvironment. The signal transducers and

activators of transcription (Stat3) pathway is a well - studied pathway that is constitutively stimulated by oncoproteins such as the viral Src oncoprotein, and is commonly constitutively activated in diverse cancers [88, 91] . Stat3 promotes tumor cell proliferation by the upregulation of cell cycle regulatory and proangiogenic genes that aid in the t ransformation process. Stat3 signaling also prevents apoptosis by regulating critical cell survival regulators such as bcl - x and c - myc [92, 93] . Stat3 inhibits th e production of pro - inflammatory molecules, thereby disabling the capability of members of both the innate and adaptive system to become activated. Blockade of constitutive

Stat3 signaling in 3T3 cells, in the absence of stimulus, led to the upregulation o f the pro -

inflammatory cytokines TNF - α, RANTES, and IFN - β [94] . Conversely, constitutive Stat3

activation resulted in the inhibition of DC maturation via VEGF and IL - 10 production.

As discussed previously, iDCs promote effector T cell anergy and demonstrate impaired

antigen presentation and cytokine secr etion. Stat3 ablated mice demonstrated higher

tumor infiltration of T cells compared to wild - type counterparts, suggesting that Stat3

signaling inhibition in the tumor microenvironment could be a potent anticancer strategy

[88] . Increased understanding of the effec t of Stat3 signaling in cancer could lead to development of targeted and effective cancer treatment.

Chronic i nflammatory tumor microenvironment

Pro - inflammatory mediators such as cytokines and chemokines are released as cancer

cells grow and disrupt the tissue architecture, and typically aid cells of the immune

system in killing tumor cells. It has been reported, however, that the inflammatory sta te

of the tumor microenvironment can actually promote tumor growth and metastasis [88] .

Chronic inflammatory state, characterized by the release of cytokines, growth factors, chemokines, ROS, NO, and prostaglandins, have been linked to cancer development [31,

47, 95] . For example, hepatocellular ca rcinoma has been linked to longer Hepatitis B and

C infectious states [95, 96] . Potent inflammatory molecules are tumoricidal but ca n also lead to DNA damage and genetic instability in tumor cells, one of the hallmarks of carcinogenesis [88] . Although a correlation between the chronic inflammation and cancer development and aggressiveness has been proposed, and supported, by a number of studies , the exact mechanisms have yet to be elucidated.

Tumor - mediated cell death

One of the most controversial immune escape concept proposed thus far is that tumor

cells can directly induce immune cell death, named the ‘counterattack hypothesis’.

Initially i t was reported that ligands of death receptors can induce downstream signaling

resulting in membrane blebbing, nuclear condensation, and caspase activation,

phenotypic characteristics of apoptotic cell death, following ligation of their cognate

receptors [97] . Death receptors are a subgroup of TNF receptor family with death

domains, each capable of initiating apoptosis [97] . The most commonly studied death

receptors are Fas (CD95, APO - 1) and TRAILR1/2 (DR4, DR5, respectively). Fas ligand

(FasL) expressi on by melanoma cells conferred immune privilege to tumor cells in vivo,

and blockade of FasL on colon cancer cells suppressed tumor growth in vivo [97, 98] .

FasL is a member of the TNF family whose roles include the maintenance of immune

homeostasis, by limiting inflammation and its tissue - destructive effects [98] . This role is

supported by the increase in Fas expression upon T cell activation and differential

sensitivity of T cell subsets to Fas - induced apoptosis. Death receptor ligands are

expressed by a variety of tumor types, indicating that death ligands mediate immune

privilege in human tumors [98] . Andreola and colleagues reported a novel mechanism of

FasL - mediated apoptosis featuring melanoma cells that secrete microvesicles containing

FasL that can induce apoptosis in lymphoid cells [99] .

Countering the results to the counterattack hypothesis a re results reporting an

immunostimulatory role for Fas. Fas has been shown to provide costimulatory signals for

T cells in vitro, culminating in T cell proliferation and IL - 2 secretion [97, 100] . Also, a

screen of several melanoma cell lines revealed that these cells did not express functional

FasL, questioning a report that FasL is an important mechanism of immune evasion

[101] . Furthermore, enforced FasL overexpression has resulted in neutrophilic infiltration in a number of tissues, leading to acute inflammation and tumor rejection [102] . In the presence of TGF - β, neutrophil response was inhibited resulting in the survival of the tumor in vivo. FasL - induced neutrophil infiltration has not been observed with normal expression of FasL, making it difficult to ascertain the physiological significance of this phenomenon. Ryan et al. addressed the uncertainty of neutrophil infiltration occurring with physiological levels of FasL expression [98] . In their study, they compared lymphocytic infiltration and tumor burden in mice infected with murine colon cancer expressing FasL against mice infected with colon cancer cells expressing very low amounts of FasL. FasL downregulation reduced tumor growth and increased lymphocytic infiltration in immunocompetent mice, but did not affect neutrop hil count. It is possible that FasL has an immunostimulatory role, however activation - induced upregulation of

FasL and Fas on T cells could lead to suicidal or fra tricidal cell death [103 ] . In addition to

FasL, other death receptor ligands such as TRAIL have been indicated in tumor - induced cell death [104] . Similar to FasL, TRAIL has been reported as a mechanism of cytotoxic lymphocytes - mediated tumor cell death, and of metastasis prevention [103] . The differential role of these death receptor ligands could be attributed to dif ferences in the level of these ligands, sensitivity of the immune cell to apoptosis, source of the ligands

(tumor or immune), and presence of regulatory factors in the tumor microenvironment.

Recently, it was reported that cannibalistic tumor cells may act ually ingest antitumor cells, serving as another mechanism by which tumor cells can attack antigen - specific lymphoid cells. Cannibalism is a strategy used by cells to survive in nutrient - depleted situations, and has been observed to correlate with poor pro gnosis in malignant tumors

[105] . Cannibal cells have been described as ‘bird - eye cells’ or ‘signet - ring cells’, and although the phenomenon has been de scribed, the underlying mechanism or driving force for this phenomenon remains unknown [106] . Most of the reports of cannibalism have focused on phagocytized immune cells, although ingested melanoma cells have also been reported [105 - 109] . It is important to determine whether cannibal cells distinguish between normal or neoplastic cells. Apoptotic cells have been observed in these cannibal

cells and Luana et al. described the ingestion of live T cells in vitro [106] .

CHAPTER 2: N ATURAL K ILLER CELL BIOLOGY IN HEALTH, DISEASE, AND CANCER THERAPY

2 .1: NK CELL DEVELOPMENT AND SUBSETS

Natural killer (NK) cells are large lymphocytic, granular cells that arise from the common lymphoid progenitor [110] . Discov er ed in 1975, NK cells comprise 5 - 15% of cells in peripheral blood. NK cells lyse transformed cells without prior stimulation , a unique characteristic that earmarks them [13, 111] . NK cell receptors are germline encoded and do not undergo gene rearrangements , leading to the ir categorization innate immune cells

[11 0] . New information concerning NK cells continues to challenge the notion that NK cells are a homogeneous group. Instead similar to T cells, various subsets of NK cells are being discovered that play varying roles in health and disease. For example, rec ently a subset of NK cells were observed to have memory - like characteristics following specific viral infection or cytokine stimulation [112] . These memory - like NK cells are discussed in depth in the NK cell development section.

NK cells are identified by the expression of CD56 (NCAM) and absence of CD3. CD56 is a neural cell adhesion molecule most commonly associated with NK cells, whose role in the bone marrow is to support niches for stem cells [113] . The role of CD56 on NK cells is still unknown, however its levels correlate with NK cell activation. CD56 is expressed by other immune cells such as dendritic cells (DCs), gamma delta T (γδ - T) cells and monocytes, making it an imprecise marker of NK cells. Furthermore, a small subset of NK cells exists that is CD56 - [114] . This subpopulation of NK cells are CD3 -

CD4 - CD14 - CD19 - CD16+NKp46+, and are most commonly found in ill individuals.

Interestingly, this CD56 - NK cells demonstrate impaired cytolytic and cytokine production functions but preserved chemokin e functions, supporting the notion that

CD56 plays a role in NK cell activity.

Approximately 90% of periphera l blood NK cells are CD16 bright , serving as another NK cell identifier . CD16 or FcγRIII is a receptor notable for mediating antibody - dependent cellular cytotoxicity (ADCC). However, CD16 is not an absolute NK cell defining marker because CD16 is also expressed by macrophages, mast cells, and a subset of monocytes [115] .

There are two main NK cell subsets: CD56 bright CD16 dim and CD56 dim CD16 bright cells

[116] . CD56 bright cells make up 10% of NK cells in blood, and mainly secrete cytokines such as IFN - γ, TNF - α, GM - CSF, and IL - 3 [117] . The CD56 bright NK cells are thought to mature into more cytotoxic CD56 dim CD16 bright cells. NK cells can further differentiate to express high levels of CD57+, a marker of terminal differentiation. CD57+ NK cells are highly cytotoxic but do not proliferate.

NK cell development

NK cells are generated an d mature in the bone marrow prior to homing to secondary

lymphoid tissues to modulate the activities of other immune cells . NK cell differentiation

is sectioned into 5 main stages that involve functional and phenotypic changes , observed

from ex vivo and in vitro work [114, 118] . In the first stage, the NK cell progenitors are

CD34+CD122 - , and are named pro - NK cells. The pro - NK cells have the potential to

differentiate into T cells, B cells, and DCs. The second level of differentiation is the pre -

NK stage. Cells at this stage are CD34+CD 117+CD122 - , and are responsive to soluble

IL - 15 or IL - 2. The third stage describes the iNK cells that have lost their multilymphoid

potential [118, 119] . Essentially, iNK cells are the first committed NK cell precursors,

however they do not perform perforin - mediated lysis or produce IFN - γ. In addition, iNK cells do not express most cell surface receptors found on mature NK cells, differentiating them from CD56 - NK cells. CD56bright NK cells are stage 4 NK cells, and notab ly

express CD94/NKG2A and other NK cell - related receptors such as NKG2D. The last stage contains CD56dimCD16bright NK cells that express killer immunoglobulin receptors (KIRs), a marker of NK cell maturation.

The transcription factor T - bet (Tbx21) is a me mber of the T - box family of transcription factors that contain a conserved DNA binding domain called the T - box [120] . T - bet is the only member of the T - box family to be selectively expressed in the lymphoid system. T cells and DCs from T - bet – deficient (T - bet - / - ) mice cannot produce IFN - γ [121] . The role of T - bet in NK cell development was first established by the observation th at NK cells in

T - bet mice resided mainly in the bone marrow and a reduced proportion were detected in the spleen, liver, and peripheral blood. Examination of NK cells from T - bet - / - mice revealed that they were immature cells bearing CD27 and CD11b [121] .

The role of Eomesodermin (Eomes) is critical for the early stages of NK cell development as Eomes - / - mice die early in embryo. Therefore delineating th e role of

Eomes in NK cell development was possible only by using a compound mutant mice,

Eomes+/ - Tbx21 - / - mice. The loss of one Eomes allele resulted in a drastic downregulation of CD122, the beta - chain of IL - 2R and IL - 15R [120] . IL - 15 signaling pathway is necessary for NK cell development and activity. Using mice harboring floxed alleles of Eomes and expressing Cre recombinase that is restricted to cells of the hematopoietic lineage, deletion of Eomes resulted in differential di stribution of NK cells.

NK cells were located in the liver, lymph node, and bone marrow while the spleen and

blood had a significant reduction in NK cells. Deletion of both T - bet and Eomes resulted

in absence of NK cells in all organs [122] . Therefore, T - bet and Eomes are important for normal NK cel l development .

In add ition to T - box transcription factors , NK cell maturity is also dependent on cell - cell

interactions. NK cell ‘licensing’ involves a ligation of a KIR by a self - MHC I molecule.

NK cells from mice lacking MHC - I proteins are hyporesponsiv e when engaging target cells lacking MHC - I molecules, as illustrated in Fig 1.1 [123] . The inverse of this

phenomenon is observed when the NK cells are engaging target cells expressing self -

MHC - I molecules. This suggests that NK cells’ KIR engagement is critical for optimal

NK cell function [124 - 126] .

Figure 2.1 . NK cell licensing enhances NK cell cytotoxic function against MHC - I deficient tumor cells. Licensed NK cells recognize MHC - I - deficient tumor cells based on the ‘missing self’ ligand absent on the tumor cell (A) and induced cell ligand such as stress molecules present on the tumor cell (B). Unlicensed NK cells cannot recognize

MHC - I null cancer cells (C) but they can recognize induced - self ligands on cancer cells and initiate cytolysis (D) [126] .

2 . 2 : NK CELL FUNCTION S IN HEALTH AND DISEASE

As stated earlier, NK cells have 2 main roles; the elimination of immunogenic cells and secretion of cytokines [127] . NK cells are most notable for their activity against virally - infect ed and transformed cells.

The role of NK cells in clearing viral infections

The role of NK cells in controlling viral infections was elucidated with studies using the mouse model of murine cytomegalovirus (MCMV) . NK cell depletion or defects in NK cell activity resulted in susceptibility to MCMV while adoptive transfer of NK cell led to resistance [128] . The mechanism underlying NK cell control of MCMV infection varies with mouse strain. In C57/BL6 mice, NK cells recognize the cytomegalovirus ( CMV ) protein m157 via its activating receptor Ly49H . In Ma/My mice, Ly49P expressed on NK cells recognize H - 2D k molecules to confer resistance to MCMV infection [129] . Many

MCMV genes promote NK cell evasion through downregulation of NKG2D ligands and expression of decoy ligands, especially MHC class I homologs that inhibit NK cell function [130] . This supports the important role NK cells play in limiting MCMV.

In humans, NK cell has been implicated as host defense mediators against infections with varicella zoster virus (VZV), herpes simp lex virus (HSV), CMV, and hepatitis B virus.

NK cells from immunocompromised children with cutaneous herpes zoster showed reduced NK cell activity within 3 days of onset of infection, as compared to when the children are in the healing phase [131] . Additionally, NK cells from mononucleosis patients demonstrated high cytotoxic activity against EBV - infected cells, and similar results have been reported for human CMV - i nfected cells. G enetic defects resulting in

NK cell deficiencies are consistent with a role for NK cells in defense against human

herpesviral infection [132] . Patients who have a NK - severe combined immunodeficiency

and undergo allogeneic hematopoietic stem cell transplantation or gene therapy fail to

reconstitute NK cells for approximately 30 years post - HSCT. These patients do not

become susceptible to viral infections, suggesting that NK cells are not crucial for

controlling viral infections but might be playing a redundant rol e [133] .

NK cells are being recognized for their role in in the control of immunopathology. This role is observed in NK cell - mediated dampening of type I IFN - dependent immunopathology typically caused by uncontrolled dissemination of the virus [132] . NK cell depletion accelerates the development of myocarditis induced by overt coxsackie virus infection [134] . NK cells also restrict macrophage activation by lysing overstimulated macrophages, as seen in Prf1 - / - mice that develop an immunopathology due to uncontrolled activities of cytotoxic T lymphocytes and macrophages [135] .

The role of NK cells in immunosurveillance and tumor rejection

NK cell targeting of cancer cells has been established in in vitro studies using human and

non - human cells [132] . In vivo data has relied on models using antibody - dependent NK

cell depletion in mice , targeting NK1.1 or the glycolipid - asialo - GM1. Antibodies to

N K1.1 may affect NKT cell popula tions, while the asialo - GM1 antibodies may also target myeloid cells, epithelial cell, and T cell subsets that express asialo - GM1 [132] .

Therefore, caution is required when analyzing the results from studies based on antibody depletion of NK cells. In some murine models, NK cell - mediated elimination of tumor

cells resulted in the development of tumor - specific T cell responses [136] . NK cell -

mediated induction of the adaptive immune system makes adoptive NK cell therapy a

very attractive concept to potentially reactivate the host’s adaptive immune cells a gainst

the cancer cells.

The role of NK cells in immunosurveillance is supported by murine studies reporting

increases susceptibility to tumor development following NK cell depletion. Mice

depleted of NK cells died faster from methylcholanthrene - induced s arcoma than mice

with intact NKG2D pathway [137] . NK cells also play a role in limiting the growth of

spontaneous B cell lymphomas in mice deficient in perforin and β - 2 - micro globulin [138] .

An 11 - yr epidemiologic survey revealed an association of low NK cell activity in peripheral blood with increased risk of developing cancer [139] . NK cell inhibitory KIR

blockade resulted in increased NK cell effector function against syngeneic tumors [140] .

Clinical development of anti - KIR antibodies is an area of strong interest and clinical

trials combining anti - KIR antibodies with other immunotherapies are ongoing.

Roles of NK cells in immunopathology

NK cells can serve as regulatory ce lls in viral infections, in humans and mice, by killing

immature DCs [141] . Also, ho st NK cells regulate donor T cell development by

eliminating donor DCs present in the graft. Conversely, NK cells secret IFN - γ and TNF - α

that promote DC maturation [141] . In addition to DCs, NK cells can also directly regulate

adaptive immune respo nses. IFN - γ from NK cells primes CD4+ T helper type 1 (Th1)

cells in the lymph node. N K cells can also target activated T cells lacking sufficient

expression of MHC class I molecules for lysis , thereby preventing T cell - dependent

autoimmunity [142] . NK cells can suppress, in vitro, autoreactive B cells derived from

Fas - deficient mice, and NK cell depletion leads to severe autoimmunity in the mouse

[143] .

Endothelial cells are primary tar gets of NK cell - mediated attack, especially in organ transplantation, due to MHC - I protein and KIR mismatch [144] . On the other hand,

uterine NK (uNK) cells promote angiogenesis in the pregnant endometrial tissue through

the secretion of VEGF and placental growth factor (PLGF) to support the endometrial

tissue during implantation [145] . The role of uNK cells in infertility or spontaneous

abortion is still being debated [132]

Mechanisms of NK cell function

NK cells mediate cytotoxic functions through a few mechanisms. Ligation of activating

receptors on NK cell s may induce the release of cytotoxic granules containing several

proteases and nucleases as well as perforin and granzyme [131] . Perforin polymerizes to

form pores in the membrane of target cells that permit accumulation of ions in the target

cell. Increase in intracellular osmotic pr essure causes swelling of the target cell and

eventual cell lysis in perforin - dependent cellular cytotoxicity . In addition, perforin

facilitates the entry of granzymes into the target cells. Granzyme s are members of a

family of serine proteases, with granz yme A and B being the most studied , that activate

caspase molecules leading to the induction of apoptosis [103] . Perforin - dependent

cytotoxicity is crucial for NK cell control of several tumors [14, 16] .

NK cells also utilize death receptor ligands to induce target cell apoptosis. NK cells

express ligands of various TNF family of receptors containing the death domain. Upon

binding to its cognate receptor, a death receptor ligand such as FasL induces a

conf ormational change in the receptor and recruitment of adaptor proteins eventually

resulting in apoptosis of the target cell [146] . IL - 18 - treated NK cells have demonstrated

antitumor effects against melanoma cells in a murine study [147] .

Recent reports have established the role of exosome secretion by immune cells to

promote antitumor immunity, and by tumor cells to counter immune cell activities. Both

resting and activated human NK cells sec rete exosomes that contain the NK cell - specific

marker CD56 and other molecules including activating receptors, death receptor ligands,

and perforin molecules [146] . NK cell - derived exosomes have demonstrated cytotoxic activity against various tumor cells including T cell leukemia, Burkitt lymphoma, and metastatic breast cancer cells [148] . Cell surface receptors are potent inducers of NK cell activity. One such receptor is CD16 which binds to the Fc region of antibodies to induce antibody - dependent cell - mediated cytotoxicity (ADCC) of tumor cells. NK cell -

dependent ADCC can be enhanced with monoclonal antibodies such as rituximab against

B cell lymphoma. In addition, bi - specific and tri - specific antibodies can enhance NK cell -

mediated cytotoxicity throug h ADCC. A bi - specific protein, ULBP2 - BB4, that

recognizes NKG2D via ULBP2 and CD138 expressed on plasma cells showed enhanced

NK cell - mediated elimination of primary malignant plasma cells and multiple myeloma

cell lines [149] .

Interactions with ligands and/or receptors on other immune cells impacts the functions of

NK cells and the immune c ells i nvolved, as illustrated in Fig 2 .2 [150 - 153] . NK cells lyse

47 immature dendritic cells and may directly interact with other lymphocytes through

CD40L and OX40 [115, 154, 155] . In addition, cytokine secretion by NK cells not only mediate cytotoxicity against target cells but plays an immunomodulatory role by activating ot her immune cells such as DCs and T cells. Importantly, NK cells exhibit rapid cytolysis of target cells as compared to the slower induction of T cell response

[117] . Therefore, NK cells are important for the recognition and clearance of virally - infected and malignant cells.

B C Cell - cell interactions • DCs CD16 - mediated antibody - • T cells dependent cellular • Monocytes cytotoxicity (ADCC)

NK cells

A Lysis of target cells • Virally - infected cells Cytokine production • Cancer cells • IFN - γ E • TNF - α D Cytotoxic granules Death receptor ligands

Perforin and Granzyme TRAIL FASL

Figure 2 .2: Mechanisms of NK cell functions. NK cell effector mechanisms include cytokine secretion (A), cell - cell interactions (B), lysis via ADCC (C), release of cytotoxic granules containing perforin and granzyme (D), and expression of death ligands (E).

Secretion of cytokines such as IFN - γ, GM - CSF, IL - 10, MIP1, and RANTE S is a pri mary function of NK cells (Fig 2 . 2A). Cytokines such as IFN - γ and TNF - α mediate cytotoxic effects on target cells [115] . IFN - γ also modulates the expression of FasL and TRAIL, activating antitumor immunity [146] . IL - 10 induces NK cell proliferation, cytotoxic function, and IFN - γ production in synergy with IL - 18. In addition, secreted cytokines play an immunomodulatory role by shaping the immune response in the tumor microenvironment. Cytokines and chemokines recr uit cells of the innate and adaptive immune systems to generate an antigen - specific immune response [115] . Also, these soluble factors pr omote the maturation of other cells of the innate immune system such as dendritic cells, that are professional antigen presenting cells (APCs). Chemokines are a special subset of cytokines that facilitate the homing of immune cells [156] . NK cells secrete chemokines including MIP - 1α/β and RANTES [157] .

Chemokines sec reted by other immune cells play an important role in NK cell function by enabling NK cells to migrate to various parts of the body [158] . The broad variety of

chemokine receptors expressed on the surface of NK cells provides a gateway for NK

cells to respond to ch emokines . These receptors are a subcategory of class A G - protein

coupled receptors (GPCR) that undergo a conformational change upon binding of their

cognate ligands [159, 160] . The arrangement of the two N - terminal cysteine residues

creates four families of chemokines: CXC, CC, (X)C, and CX3C. In the CXC family, an

amin o acid separates the N - terminal cysteines while the CC family has adjacent cysteine

residues. The CXC3C family has three amino acids separating the cysteine residues, and

the (X)C family has one missing N - terminal and one C - terminal residues.

NK cell homing is driven by chemokine gradient with the highest chemokine concentration at the site of secretion [156] . As NK cells move towards increased chemokine gradient, they undergo cell polarization and adhesion to the endothelial cellular. Three of the key chemokine receptors found on NK cells are CXCR3, CXCR4, and CXCR6. CXCR3 can recognize CXCL4 and CXCL4L1 which are both platelet - derived cytokines released upon platelet activation. In addition, CXCR3 can recognize

CXCL9, CXCL10, CXCL11, and CXCL13 which are important for recruitment of white blood cells to the inflamed tissue. The antagonistic ligand for CXCR3 is CCL11. CXCR4 recognizes CXCL12 which is important for inducing homeostatic migration and for sequestering NK cells in the bone marrow. Other ligands for CXCR4 includes macrophage mig ration inhibitory factor (MIF) and ubiquitin. CXCR6 recognizes

CXCL16 which is an inflammatory cytokine but can also engage in homeostatic signaling.

Various NK cell subsets have been shown to differentially express a number of cytokine receptors [1 59] . Ligands for CXCR3, CXCL9 and CXCL10 are at low levels of expression during homeostatic conditions. However, they are upregulated in NK cells in the presence of IFN - γ and other related cytokines secreted by tumor - infiltrating immune cells. In MM pat ients, upregulation of CXCR3 ligands CD56high NK cells home to the lymph nodes, express CXCR3 preferentially, and have higher CXCR4 expression compared to CD56low NK cells [159] .

Lima and colleagues established a differential expression of chemokine receptors by

CD56bright versus the mature CD56low NK cells [161] . Analysis of peripheral blood from healthy donors revealed that CD56brightCD16low NK cells CXCR3/CCR5+ while

CD56l owCD16bright cells are CXCR3 - /CCR5+ . The clinical implications of this difference are yet to be elucidated. Understanding of the differential expression of chemokine receptor on various subsets of NK cells have been largely ignored.

2 .3: NK CELL SURFACE RECEPTORS MEDIATE NK CELL ACTIVATION

AND FUNCTION

NK cells express a repertoire of germline - encoded cell surface receptors that enable NK cells to interact with the microenvironment and communicate with other cells [162] . A

balance of the signaling from these receptors orchestrates NK cell cytotoxic activity and

cytokine secretion [163] . Depending on the strength of an activating stimulus, it may or

may not be able to overcome the inhibitory signaling cascade intracellularly.

Activating Receptors

NK cell activation is dependent on strong activating signals generated from the binding

of stress - induced molecules to their cognate receptors on NK cells overpowering inhibitory signaling in NK cells. Most activating receptors signal through their the

imm unoreceptor tyrosine - based activation motifs (ITAM) domains that bind to ITAM domains of adaptor proteins FcεRIγ and DAP12. The signaling chain DAP10 , instead,

contains the tyrosine - based motif YINM [115, 162] . Phosphorylation of adaptor proteins by Src family of tyrosine kinases such as Src and Yes triggers a phosphorylation - dependent signaling cascade eventually resulting in increased intracellular Ca2+ levels.

High calcium levels induces the reorganization of actin cytoskeleton for the release of cytotoxic granules and the transcription of cytokine and chemokine genes [146] . The effector f unction resulting from DAP10 signaling differs from that of DAP12 signaling.

Studies in mice lacking DAP10 or DAP12 have shown that DAP12 signaling induces both cytotoxicity and cytokine secretion while DAP10 signaling mostly activates cytotoxicity [146, 164] .

Natural killer group 2 member D (NKG2D) is a homodimeric C - type lectin - like type II transmembrane receptor (glycoprotein) [165] . It is constitutively expressed on NK cells,

γδT cells, NKT celld, and αβ CD8+ T cel ls. NKG2D expression is upregulated by IL - 15

and downregulated by the immunosuppressive molecule TGF - β [116] . NKG2D ligands

are mainly MHC class I chain - related - A (MICA) and – B (MICB), UL16 - binding

protein/RAE - like transcript 1 (ULBP/RAETI) molecules [166] . The expression of MICA

an d MICB is an indicator of cellular stress due to tumorigenesis or infection. ULBPs are

expressed differentially on various cell lines and tumors. NKG2D intracellular signaling

is mediated by DAP10 which bears the YINM motif, and is not as sensitive to inhi bitory

signals induced as KIRs, as signaling pathways through DAP12 [115] .

The natural cytotoxicity receptors (NCRs) are NKp30 (NCR3, CD 337), NKp44 (NCR2,

CD336), and NKp46 (NCR1, CD335) [115, 162] . Thes e receptors have short cytoplasmic

unit that does not mediate signal transduction. Instead, their transmembrane regions

contain charged amino acids that interact with signal - transducing polypeptides. Thus

unlike inhibitory receptors, activating receptors a re multi - chain complexes and are not single - chain units. NKp30 and NKp46’s ligand is membrane - associated heparin sulfate

proteoglycans (HSPGs). NKp46 is expressed by both resting and activated NK cells, and

is highly involved in the lysis of cancer cell li nes [115, 167, 168 ] . The ligand for NKp46 remains elusive, however hema gglutinin molecules of influenza viruses are its ligands in virally - infected cells. NKp30 is also expressed by both resting and activated NK cells.

Importantly, NKp30 may aid in bridging innate and adaptive immunity through the interaction of NKp30 recepto r on NK cells and NKp30 ligand on dendritic cells (DCs).

Specifically, NKp30 has been linked to the recognition and cytolysis of immature DCs

[115, 169] . NKp44 is the only NCR expressed only on activated NK cells , and also on

γδT cells. The only ligand associated with NKp4 4 is viral hemagglutinin.

DNAX accessory molecule (DNAM - 1, CD226) is an activating receptor constitutively expressed on approximately 50% of NK cells [153, 162, 170] . Ligands for DNAM - 1,

PVR and Nectin - 2, are upregulated on some tumor cells, supporting the importance of this receptor in NK cell effe ctive functions. As an adhesion molecule, DNAM - 1 binds to other receptors such as CD96 and TIGIT to modulate NK cell functions [153] . The role of TIGIT as a checkpoint molecule is discussed in the immunotherapies chapter.

A well - characterized receptor, 2B4 (CD244) , induces IFN - γ production and NK cell - mediated cytotoxicity [170, 171] . CD244 acts as an activating or inhibitory receptor depending on the adaptor molecule recruited at its cytoplasmic tail [146] . Human CD244 neutralizing antibodies effectively blocked the killing of CD48 - expressing target cells, while NK cells from CD244 deficient mice demonstrate enhanced cytotoxicity and cytokine secretion, and rejected a melanoma [172, 173] . CD244 acts as an activating receptor in humans but an inhibitory re ceptor in mice. Interestingly, ligation of 2B4 by its ligand, CD48, followed by ligation of DNAM - 1 leads to greater NK cell - mediated cytotoxicity.

CD16, also known as FCγRIII (IgG) or Leu 11, is a low affinity receptor for IgG [115] .

CD16 is expressed on NK cells to facilitate antibody - dependent cellular cytotoxicity

(ADCC) (Fig 1.2C). CD16 binds to the Fc porti on of an antibody that is bound to a tumor - associated antigen on a cancer cell [170] . This phenomenon is being maximized by combining NK cell therapy with therapeutic antibodies such as rituximab.

CD57, also known as Leu 7 or human natural killer - 1 (HNK - 1), is a marker of senescence in T cells [174] . CD57 is also expressed by NK cells, subpopulations of monocytes, macrophages, mast cells, and neutrophils. CD57 is only expressed by

CD56dim NK cells, and not by the more immature CD56bright NK cells [174] . CD57+

NK cells are highly cytotoxic, more sensitive to CD16 stimulation, and decreased proliferative capacity. CD57+ NK cell subset are considered the most terminally differentiated subsets.

Other activating receptors include the KIRs with short cytoplasmic tails gener ate activating signals via tyrosine phosphorylation of DAP12 and other proteins [175] . The activating KIRs are KIR2DS1, KIR2DS2, KIR2DS3, KIR2DS4, KIR2DS5, KIR3DS1.

CD94/NKG2C and CD94/NKG2E are lectin - like receptors t hat are activated by HLA - E proteins, and complex with DAP12 protein. These proteins have varying affinities for

HLA - E.

Inhibitory Receptors

The inhibitory receptors on NK cells signal through their immunoreceptor tyrosine - based inhibitory motif (ITIM). ITIMs bind to adaptor proteins that propagate the inhibitory signal through the SHP phosphatases. Inhibitory signals are thought to dampen activating signals and prevent inappropriate NK cell activation through vav - 1 inactivation, impairing actin polymeri zation that plays a role in cellular cytotoxicity [115,

176] .

Killer immunoglobulin - like receptors (KIRs, CD158) recognize different allelic groups of

HLA - A, HLA - B, HLA - C [115, 177] . KIRs are encoded in the leukocyte receptor

complex and NK gene complex on human chromosome 19 and 12, respectively. KIRs

with long cytopla smic tails induce inhibitory signals to counter the signals from

activating receptors, while that of KIRs with short cyt oplasmic tails are activating. NK

cells are major histocompatibility complex (MHC) class I - specific and display clonal

inhibitory recept ors on the cell surface. Each NK cell expresses at least one inhibitory

receptor that recognizes self MHC class I molecules. Inhibitory signaling is generated

when an inhibitory KIR is engaged by a self MHC - I molecule, leading to src - kinases -

mediated tyros ine phosphorylation of the downstream ITIM and recruitment of SH2 -

domain - containing tyrosine phosphatase 1 or 2 (SHP1 or SHP2) proteins [115 ] . During

viral infection or tumor transformation, cells often downregulate MHC - I molecules, increasing the probability of NK cell activation and consequently target cell lysis.

CD94/NKG2A is a type II transmembrane receptor and the only inhibitory rec eptor among the c - type lectin NK receptors [170] . Ligands are HLA - E. Although

CD94/NKG2A receptor is structurally different from the inhibitory KIRs, it signals through ITIM phosphorylation and recruitment of SHPs. Ligation of the CD94/NKG2A complex init iates inhibitory signals that dampens NK cell effector functions.

Accessory Receptors

In addition to the main activating and inhibitory receptors, NK cells also express

accessory receptors that play important roles on NK cell effector functions. CD27 is a

negative regulator of NK cell cytotoxic function though the mechanism is not fully

und erstood [170, 178] . CD27+ NK cell s have lower levels of perforin and granzyme B, but produce more IFN - γ and are more proliferative as compared to CD27 - NK cells. The

CD27 ligand CD70 downregulates its receptor, providing a mechanism to block the suppressive effects of CD27 on NK cells. Ly mphocyte function antigen - 1 (LFA - 1) is an adhesion molecule that is critical for the cytolysis activities of cytotoxic T lymphocytes and NK cells [179] . CD54, also knowns as ICAM1, is an adhesion molecule and a ligand for LFA - 1 [180] . In collaboration with LFA - 1, CD54 promotes NK cell effector function by facilitating NK - tumor cell interaction prio r to cell lysis. Studies have linked CD54 expression with increased tumor cell sensitivity to NK cell - mediated lysis. CD62L is an adhesion molecule, highly expressed by CD56bright NK cells, and participates in extravasation of NK cells through the vascular endothelium [158] .

Death Recept or Ligands

The death receptor - mediated cytolysis is a perforin - independent mechanism for cancer

cell lysis by NK cells [ 170] . Death receptors such as Fas and TRAIL - R 1/2 are members

of the TNF superfamily containing death domains. The death receptor ligands ligate the

cognate receptors , leading to recruitment of the adaptor protein Fas - associated death

domain (FADD) via t heir death domains. FADD binds procaspase - 8 triggering its

transactivation and subsequent caspase cleavages ultimately resulting in apoptosis of

target cells [103] . The two main death ligands are FasL and TNF - related apoptosis -

inducing ligand (TRAIL). FasL is only expressed on NK cells and cytotoxic T cells, and

upon engaging its receptor leads to nuclear condensation, membrane blebbing, and

activation of caspases [178, 181] . TRAIL is expre ssed constitutively on some NK cells and induces apoptosis through the activation of caspases. Zamai and colleagues reported

the differential expression of TRAIL is based on the developmental stage of NK cells

[182] . Immature NK cells expressing CD161 rely on TRAIL but not FasL for caspase -

based induction of apoptosis. In addition, CD161+ NK cells cannot mediate cytotoxic

granule release - dependent cytotoxicity. More mature CD56+ NK cells utilize Fa sL - and

granzyme - dependent cytotoxicity but not TRAIL.

Death receptors are capable of inducing other types of cell deaths aside from apoptosis.

In circumstances in which caspase - 8 is inactive, TRAIL - Rs can recruit receptor

interacting protein (RIP)1 and R IP3 to form a necrosome, followed by phosphorylation of mixed lineage kinase domain - like protein. MLKL then mediates necrotic cell death by permeabilizing the plasma membrane. Alternatively, if apoptosis is blocked, ligated

TRAIL - Rs r ecruit RIP and form a complex with TNF receptor - associated factor 2 and

TNF receptor type 1 - associated death domain. RIP1 then promotes the activation of NF -

κB and other transcription factors to promote survival signals [103] .

The role that death receptors play in N K cell - mediated control of tumor metastasis has

been demonstrated in various in vitro and in vivo studies [103] . Administration of

neutralizing antibody against TRAIL resulted in metastases of several TRAIL - sensitive

tumors [183] . In a c hemically - induced murine tumor model, TRAIL - R deficiency

resulted in spontaneous lymph node metastases but had no effect on the primary tumor,

supporting a specific role for TRAIL in controlling metastasis [184] .

2 .4 : MEMORY - LIKE NK CELLS

The discovery of memory - like NK cells has been observed in three circumstances - antigen - specific memory NK cells in the liver, CMV infection, and cytokine induction.

Hepatic memory NK cells mediate hapten - specific contact hypersensitivity response was demonstrated in the absence of T and B cells [146 ] . This response is abrogated in mice deficient for IL - 12, IFN - γ, or I F N - α R signaling [185] . Memory - like NK cells associated

with CMV was first discovered expressing antigen - specific receptors in murine

cytomegalovirus (MCMV) [186] . In addition, a more efficient secondary response was

observed against the same antigen. The current model hypothesizes that MCMV infection

induces the release of proinflammatory cytokines such as IFN - α that non - specifically

induce N K cell activation and proliferation. Ligation of IFN - α and IL - 12 receptors

stimulates downstream STAT1 and STAT4 that induce NK cell IFN - γ production and

cytolytic activity [187] . The key feature in memory formation is the interaction of the

MCMV m1 57 glycoprotein and Ly49H activating receptor on NK cells and signaling though the adaptor proteins DAP10 and DAP12 [186, 188] . Absence of either adaptor protein does not affect memory - like NK cell development, howe ver absence of both results in susceptibility to MCMV infection. In addition to the adaptor proteins, co - stimulatory molecules are also important for memory - like NK cell development. MCMV induces CD155 and/or CD112 expression on monocytes and dendritic cel ls [189] .

Binding of CD155 and/or CD112 to NK cell DNAM - 1 promotes NK cell differe ntiation after the initial challenge and subsequent challenges with MCMV. The memory - like NK cells express high levels of the maturation markers CD11b, Ly6C and KLRG1, and

Ly49H [190] .

Human NK cells mature and differen tiate into memory - like NK cells following infection

with human cytomegalovirus (HCMV). NK - deficient humans are susceptible to infection by multiple herpesviruses, however only HCMV has demonstrated a detectable change in

the NK cell repertoire [191] . This change is reflected in increased frequency of CD57,

CD16, and NKG2C expression on NK cells. In addition, these modified NK cells express

reduced levels of NKG2A and the natural cytotoxicity receptors, while expression of

activating KIRs are increased. Another group described an overlap of these CD57+ cells

with NK cells bearing reduced expression of the transcription factor PLZF, leading to

altered responsiveness to cytokine stimulation [192] . Blocking antibodies against HLA - E

or NKG2C, in an in vitro HCMV infection co - culture system, prevented the expansion of

CD57+NKG2C+ NK cells, suggesting that HLA - E binding to NKG2C is necessary for the expansion of NK memory subsets [193] . NK cells from a co - culture system consisting of NK cells from HCMV - infec ted individuals and HCMV - infected fibroblasts , and lacking exogenous cytokines, are able to proliferate and secrete IFN - γ in a NKG2C - independent manner [194] . Similarly, NK cells from individuals lacking the NKG2C gene demonstrate rapid maturation and express activating KIRs in a manner similar to

NK cells from NKG2C+ individuals , suggesting that NKG2C is not the only stimulus driving HCMV - mediated changes in NK cell phenotype [195] . Assessment of NK cells from donors carrying the NKG2C gene or a homozygous deletion of the gene revealed an important role for CD2 in augmenting ADCC [196] . CD2 co - stimulation may behave similarly t o DNAM - 1in Ly49H - driven responses in mice, serving as a second - signal upon interaction with target cells . The identification of FcεRIγ - deficient NK cells has elucidated the intracellular mechanisms that support NK cell adaptation to HCMV

60 infection. FcεRIγ - deficient NK cells retain CD3ζ expression but lack NKG2C, and surprisingly display enhanced cytokine production and cytotoxic potential. Furthermore,

HCMV - dependent loss of FcεRIγ is associated with decreased expression of PLZF and other PLZF target genes as observed in memory - like NK cells [197] .

Memory - like NK cells can also be generated by cytokine stimulation. Adoptive transfer of murine splenic NK cells activated in vitro with IL - 12, IL - 15, and IL - 18 resulted in the proliferation of the cytokine - primed NK cells upon administration to naïve mice. Upon re - exposure to IL - 12, IL - 15, and IL - 18, the primed NK cells demonstrated augmented

IFN - γ secretion detectable for up to 12 weeks and perforin - dependent inhibition of tumor growth [198, 199] . Similar effects were observed with human cells re - stimulated ex vivo with IL - 12, IL - 15, and IL - 18, following prior in vitro stimulation with the same cytokine milieu [112] . Unlike virally - induced memory - like NK cells, cytokine - induced ones lack

CD57 expression but retain NKp46 and NKG2A. Cytokine - induced memory - like NK cells were tested in a phase I clinical trial that resulted in 5 out of 9 AML patients responding clinically to the adoptively transferred cells [200] .

2. 5 : ADOPTIVE NK CELL THERAPY

Current state of cancer treatment

Nearly all mortality due to cancer is due to metastasis to distant sites [201] . Several

strategies to combat the devastating effects of cancer metastases has been developed.

Radiation therapy was found to ease patient pains but could not address cancer stem cells

that remained dormant. Instead radiation therapy enriched for these cells. In addition,

chemotherapy, pharmacological targeti ng of rapidly developing cells has been successful

in clearing primary tumors. However, chemotherapeutic interventions also fail to clear

dormant cells, causing patients to relapse. Targeted therapies aimed at specific antigens

expresse d on tumor cells hav e been successful in only subsets of cancer patients . The

heterogeneity of cancer makes it difficult to identify potent antigens that are expressed

only on cancer cells but not on normal cells. Adding to the complexity is the genetic and

epigenetic evoluti onary changes that cancers undergo as a consequence of external

pressures such as drug treatment [202, 203] . The heterogeneity makes biopsies difficult,

and there is much work towards the development of liquid biopsies for the detection of

exosomes or circulating tumor cells [201, 204] . Even this more advanced and

comprehensive diagnostic tool presents its own set of challenges. Although single cell

sequencing is more sensitive and provides broad view of the tumor landscape, high level

of technical variability and noise makes clinically relevant information dif ficult to

extract. The same challenge exists with analysis of circulating tumor DNA. The next

sections provide detailed descriptions of various forms of therapies, exploiting the

immune system to fight cancer. The focus of this brief review is on NK cell - based adoptive cell therapy.

Despite the discovery of the cytolytic function of NK cells in 1975, their potential as anti -

cancer agents was not fully recognized until the 2000s when their anti - leukemic effects

were observed after hematopoietic stem cell transplants (HSCTs) were performed. Not only can NK cells lyse transformed cells, but they are the first cells to reconstitute after

HSCTs and may also protect against infection [205] .

Sources of NK cells include bulk peripheral blood, apheresis, bone marrow, umbilical cord, human embryonic stem cells, human placenta, induced pluripotent stem cells, and

NK cell lines [177, 206 - 211] . Another source of NK cells could be another institution;

fresh NK cells in a non - frozen state were successfully shipped. It was observed that NK effector functions were only dependent on IL - 2 treatment, and not on incubation temperature or conditions [209] . This study further strengthens the use of NK cells as a promising cancer therapeutic option, especially f or centers that lack the infras tructure to process the cells.

Autologous NK adoptive cell therapy

NK cell therapy holds much promise due to the innate ability of NK cells to lyse tumor

cells and secrete cytotoxic molecules. The earliest NK cell trials utilized bulk CD3 -

depleted PBMCs or purified autologous NK cells. Many of the initial clinical trials of

autologous NK cells failed to yield positive clinical outcomes. In addition, these trials

were combined with high dose IL - 2 that resulted in severe toxicities [177, 212] .

Subsequent trials explored the role of IL - 2 in expanding host lymphocytes in vivo by

treating patients with low dose IL - 2. Low - dose IL - 2 injection led to up to 2500%+ increase in NK cell expansion and enhanced in vitro cytotoxic function against a number of NK cell - resistant cell lines [177, 213, 214] . Despite the suboptimal performance of

LAK cells, intere st in ACT was spurred by positive responses by some end - stage metastatic cancer patients [215, 216] . Thereafter, many studies explored the efficacy of activated and expanded autologous NK cells combined with low - dose IL - 2, necessary for

NK cell persistence in vivo [214, 217 - 219] . One prominent example was a Phase II trial reported by Burns et al. in which 37 patients with lymphoma and metastatic breast cancer were treated with low - dose IL - 2 and IL - 2 - activated NK cells [219] . Despite attempts to maintain NK cell activation in vivo, patient responses remained suboptimal. A critical reason for impaired donor NK cell activity is the expansion of regulatory T cells (Tregs), known to suppress NK cell activity [220] . Tregs have a high affinity IL - 2R, that supports their expansion following IL - 2 infusions, and also may cause them to compete with NK cells for IL - 2 [221] . Moreover, Tregs secrete high amounts of transforming growth factor - β (TGF - β) which directly inhibits NK cell cytotoxic function and cytokine

secretion [222] . TGF - β also promotes Treg differenti ation. Two clinical trials were

established to assess for the effectiveness of depleting Tregs using a cytotoxic fusion

protein that contains the amino acid sequences for diphtheria toxin and truncated IL - 2

(IL2DT) [221] . By day 7, a Treg population was not detected and 53% of refractory

AML patients attained complete remission. Another important finding from these

autologous NK cell infusion trials is the role of self - MHC I proteins on donor NK cell

activity [221] . It was realized that the suboptimal response observed in these trials could

be due to the inhibitory signals generated in NK cells when they encounter self - MHC

molecules, preventing them from becoming activated. Therefore, future trials shifted to the use of allogeneic NK cells, preventing the inhibitory signaling pathways from being

activated by self MHC - I molecules.

Allogeneic NK cell transplantation takes advantage of the ‘missing - self” hypothesis to

increase the probability of NK cell activation upon encountering stress ligands or foreign

peptides [177] .

Allogeneic adoptive NK cell therapy

In AML, allogeneic NK cell therapy was often combined with hematopoietic stem cell

transplantation [177] . In a Phase I study by Ciurea et al., multiple infusions of ex vivo expanded NK cells were administered to 13 patients with myeloid malignancies undergoing hematopoietic stem cell transplantation [223] . As compared to patients who

did not receive NK cell, patients who received NK cell infusions had higher numbers of

cytokine - secreting NK cells a month a fter the first NK infusion, and had reduced CMV

reactivation by 39%.

Allogeneic NK cell therapy, as a standalone therapy was tested in a couple of clinical

trials. The results from a trial testing CD3 - depleted PBMC products against AML, renal cell carcino ma, and melanoma was reported by Miller and colleagues [224] . Five out of

the nineteen AML subjects treated with the final cell product, containing 40%

CD56+CD3 - cells, in addition to low dose IL - 2 following lymphodepletion achieved

complete response. Three patients out of ten melanoma patients, and 1 renal cell

carcinoma patient, had stable disease 6 weeks following NK infusion [224] . Only 1

patient out of 43 demonstrated persistence of NK cells up to 138 days post - infusion. No

GVHD were reported and adverse events were limited to constitutional symptoms, commonly observed with IL - 2 therapy, an d anemia. This clinical trial demonstrated modest efficacy of NK cell therapy against AML but not against the solid tumors tested.

Also, the trial did not effectively demonstrate the effectiveness of NK cell therapy because the final cell product contained approximately 50% monocytes and B cells. In a subsequent trial, the same group depleted CD19+ cells from the CD3 - depleted PBMC product, and finally isolated CD56+ cells [225] . In addition to the pure NK cell infusion, relapsed/refractory AML patients were given a lymphodepleting regimen, and the Treg - depleting drug Ontak. Eight of out fifteen patients achieved remi ssion, and 27% of patients demonstrated in vivo NK cell expansion, as compared to 10% expansion in patients who did not receive Ontak. In addition, the patients treated with Ontak had higher rates of progression - free survival (33%) compared to patients who did not (5%) at

6 - month time point. Patients tolerated NK infusion well but grade 3 - 5 toxicities including alveolar hemorrhage and fevers were observed following Ontak infusion.

Although allogeneic NK cell therapy has been proven to be safe, there are li mitations that need to be overcome in order for it to become a mainstay cancer treatment modality.

Limitations to effective adoptive NK cell therapy include limited in vivo persistence, poor homing to the tumor site, difficulty in generating clinically eff ective NK cell doses, and impaired functionality due to an immunosuppressive tumor microenvironment. NK cells can become exhausted due to downregulation of the transcription factors T - bet and eomesodermin (eomes) [226] . Tbet and eomes are T - box transcription factors that play important roles in NK cell development , maturation, and function. Lower levels of these transcription factors leads to loss of activating receptor expression and IFN - γ

66 production that significantly impairs NK cell effector functions [227] . Downregulation of activating receptors and upregulation of inhibitory receptors of NK cells prevents NK cells from target ing cancer cells for lysis [170, 228, 229] . NKG2D has been reported to be downregulated in tumors that secrete soluble NKG2D ligands [230] . In additi on to impaired recognition of transformed cells, NK cells are also unable to interact with other immune cells to mount an appropriate immune response to the growing malignancy

[163] . In addition to immunosuppressive factors impairing NK cell function, NK cell therapy is also limited by NK cell numbers [231] . NK cells consist of approximately 10% of cells in peripheral blood an d a maximum of 20% in other sources, making it difficult to obtain therapeutically effective cell doses.

NK cell line - based adoptive cell therapy

In order to avoid the challenge associated with generating clinically relevant NK cell number, another approach is to utilize NK cell lines such as NK - 92, NKL, and NKG

[232, 233] . Utilizing NK cell lines is more efficient than primary NK cells because they are easily accessible and homogeneous, which makes quality control and l arge - scale production easier. The most commonly used one is the NK - 92 cell line, derived from a

NK lymphoma patient, and defined as CD56 bright CD16 neg/low NKG2A+KIR - [233] .

Effectiveness of NK - 92 cells against a variety of tumor types inclu ding glioblastoma and

B cell malignancies has been proven preclinically [233] . Phase I clinical studies have demonstrated that NK - 92 cell infusions are safe in cancer patients [208, 233, 234] . NK -

92 cells were thawed and expanded with high - dose IL - 2. Seven refractory/relapsed AML patients were treated with NK - 92 infusions, and no grade 3 - 4 toxicities were observed

[234] . Activity of the transferred cells were transient in 3 out of the 7 patients. There are

currently 4 NK - 92 - trials against AML, B cell malignancies, and MUC1+ solid tumors.

NK - 92 cells have a number of limitation, however, that need to be considered as it is

being used to treat patients. Foremost, NK - 92 cell s must be irradiated prior to infusion

because they have the potential for tumor engraftment and are EBV+. Irradiation

prohibits their expansion and persistence in vivo, two factors that are important for cell

therapy success. In fact, NK - 92 cells engineer ed with a ErbB2/HER2 - CAR was not

detectable within 7 days of adoptive transfer [235] . Moreover, NK - 92 cells are CD16 -

negative and cannot mediate antibody - dependent cellular cytotoxicity (ADCC), an

important method of NK - mediated tumor cell lysis.

Chimeric antigen receptor - modified NK cells (NK - CARs)

The third approach to adoptive NK cell therapy is genetic modification of primary NK

cells or NK cell lines to overexpress cytokines, Fc receptors, and/or CARSs [232] . The

most promising strategy that has shown preclinical efficacy is NK - CARs [17 7] . NK -

CARs have been designed to target a number of antigens including CD19, CS1, and

Erb2/HER2 [233, 235 - 237] . Another approach to NK - CAR development is a non - antigen

specific approach. An example is a NK - CAR engineered with NKG2D linked to DAP10 and CD3ζ, to induce supra - intense NK - CAR cell activation [238] . This NK - CAR demonstrated high cytotoxicity against hematological malignancies and solid tumors in xenogeneic mouse models, but did not target non - transformed cells. Unlike CAR - T cells t hat have been evaluated in a number of clinical trials, the first in - human clinical trial evaluating a NK - CAR, named CB - NK, is ongoing. Patients will receive a

68 lymphodepleting chemotherapy prior to CB - NK infusion. CB - NK cells are sourced from umbilical, an d engineered to target CD19 and produce IL - 15 [177] . IL - 15 secretion by

CB - NK will support its proliferation and persistence in vivo . As a safety feature, CB - NK cells express the inducible caspase 9 as its suicide gene [177] . Patients that are being recruited for this study include patients with B - lymphoid malignancies, acute lymphoid leukemia ( ALL ) , and chronic lymphoid leukemia ( CLL ) .

In general, NK - CAR is a promising therapeutic modality because unlike CAR - T cells, it can be administered allogeneically, and therefore avoids the cumbersome process of processing each patient’s cells individually. Previously, transduction of NK cells with

CARs were difficult with very low efficiency and high variability as compa red to NK cell lines. Optimization of viral transduction and electroporation processes has resulted in up to 73% transduction efficiency [239] . Another potential limitation to NK cell therapy is the sensitivity to the freeze and thaw process, leading to cell death and loss of activity.

Some groups have reported restoring thawed NK cell function with overnight incubation with cytokines such as IL - 2 [233] . Although not yet proven, NK - CARs are expected to behave similarly. NK cells are dependent on cytokine support for persistence, and although this may prevent long - term adverse events, it limits their efficacy. IL - 2 infusions are toxic and IL - 15 is short - lived. Alternative approaches to cytokine infusion is to i ncorporate genes for IL - 2 or IL - 15 in the NK - CAR construct, similar to the CB - NK cells described previously, to provide constant cytokine support to the transferred cells.

Lastly, due to cytokine release syndrome and other adverse events associated with CA R -

T cells, NK - CARs could be modified with suicide genes to more tightly control the in vivo effects of these cells [233] .

Bi - specific and Tri - specific killer cell engagers (BiKEs and TriKEs)

In addition to CARs, bi - specific killer cell engagers (BiKEs) and tri - specific killer cell engagers (TriKEs) are being employed to enhance the efficacy of NK cell therapy [240] .

BiKEs and TriKEs contain two (BiKEs) and three (TriKEs) single chain variable fragments from antibodies with different specificities. Unlike CARs, BiKEs and TriKEs can facilitate the formation of a tumor synapse and induce NK cell activation. For example, an anti - HER2/CD16 bi - specific antibodies demonstrate a 3.4 - fold increase in affinity for CD16, as compared to an anti - HER2 antibody alone, enhancing NK cytotoxic function via ADCC [241] . BiKEs and TriKEs are smaller than conventional bi - specific and tri - specific antibodies, enabling for a more favorable biodistribution and easier penetration into the tumor [241] . An in vitro cytotoxicity assay resulted in significantly increased NK cel l - mediated killing of Raji cells in co - cultures containing anti -

CD16/CD19 BiKE and anti - CD16/CD19/CD22 TriKE, as compared to co - cultures with rituximab or individual parental antibody [240] . Phase I/II clinical trial of an anti -

CD16/IL - 15/CD33 TriKE has been initiated for the treatment of CD33 - expressing hematologic malignancies.

CHAPTER 3: OTHER IMMUNOTHERAPIES

Adoptive Cell Therapies

Adoptive cell therapy (ACT) has been in existence for more three decades ago and has developed to be a promising cancer treatment modality. Compared to other forms of immunotherapy such as vaccination and small molecule inhibitors, ACT has the mo st potential to lead to effective and durable remission [242] . Prior to NK adoptive cell therapy, dendritic cells, lymphokine - activated killer (LAK) cells, tumor - infiltrating lymphocytes, and T cells were being used to cure a number of illnesses [243 - 245] . The effectiveness of these cancer treatment was inconsistent and the presence of immunosuppressive factors in the tumor environment impairing the anti - cancer efforts of these cells [246, 247] . For n early a decade, vigorous efforts were made to understand the mechanism of action of various empirical and ad hoc treatment strategies being employed at various institutions [248] . From these studies, the field of immunotherapy came to be.

3.1: TUMOR - INFILTRATING LYMPHOCYTES

One of the first solely cell - based form of immunotherapy for treating cancer patients are tumor - infiltrating lymphocytes (TILs). TILs, including T cell subsets and NK cells, have been observed to impact prognosis of patients. The higher the TIL numbers, th e better the progression free survival and overall survival of the patient [242] . In the tumor microenvironment of a number of tumors, TILs are often low in numbers and fail to control the tumor. Hence a form of cellular therapy was developed to increase the number of TILs. Advan tages of this form of cell therapy includes the ability to generate large numbers of cells in vitro that can be activated for high reactivity against tumor antigens.

TILs are extracted from resected specimen and expanded in IL - 2 for 2 - 4 weeks resulting in 40 - 50 million total cells. Thereafter, selection for T cells occur with expansion with anti - CD3 antibody and IL - 2 in the final 2 weeks of expansion resulting in 1000 - to 3000 - fold expansion [242] . TILs can be reactivated and expanded, following cryopreservation, under conditions that can overcome inhibitory signaling or factors that TILs are exposed to in vivo. Depletion of regulatory cells prior to infusion of cells into patients provides an amiable environment for the transferred cells [249] . Prior to TIL infusion, patients are giving a lymphodepleting reg iment of cyclophosphamide and fludarabine to create a supportive environment for the TILs to expand and be effective [242] . A secondary

lymphodepletion by total body irradiation was tested at the NIH, and although the

response rate was up to 72%, there were significant toxicities associated with this

treatment.

There is an extensive history of TIL - ACT being tested in metastatic melanoma.

Rosenberg and colleagues in 2011 reported the results from 3 trials in which 93 patients

with metastatic melanoma were treated with TIL - ACT [249] . Ninety - five percent of the se patients had relapsed from a prior systemic treatment and median follow - up time following TIL therapy was 62 months. 19 patients achieved durable complete tumor regression (>3 yrs), and the 3 - yr survival rate for the entire group was 36%. The study also revealed key information that were considered in subsequent trials: response varied with telomere length of the transferred cells, the number of CD8+ cells infused, and the persistence of the transferred cells in the patients. The observation that sustain ed complete remission can be achieved with TILs supports the contest that immunotherapy can control disease and potentially cure cancer patients. TILs have been tested not only in

73 melanoma but also ovarian cancer, renal cell carcinoma (RCC), and cervical c ancer. The first trial to show clinical efficacy of TIL - ACT in ovarian cancer was in a cohort of patients treated previously with primary resection and chemotherapy [250] . As a maintenance therapy, TIL - ACT group had 100% 3 - yr overall survival compared to

67.5% in the control group. There are at least 5 ongoing trials testing TIL - ACT in ovarian cancer patients, while some completed trials have yet to report the results fr om the studies. Cervical cancer is another immunogenic gynecological cancer that has been targeted for TIL - ACT therapy. One cervical cancer trial reported in 2015 that 3 of the 9 patients enrolled in the study achieved objective response [251] . In addition, 2 of the 3 responding patients were disease free up to 46 months after the trial [242] . The effect of

TIL - ACT on RCC is somewhat controversial with some studi es reporting a favorable prognosis with increased CD8+ TILs and others reporting the opposite correlation [252] .

Lymphocy te infiltration was found to correlate with higher tumor grade, and could explain why high numbers of TILs correlated with poor clinical outcomes. Overall, the varied reports on the effects of TIL - ACT on RCC suggests that further testing and optimization i s required to avoid confounding variables such as 41% TIL production failure reported for a phase III trial, that led to the discontinuation of the trial [252] .

Despite promising response rates, the IL - 2 treatment causes significant toxicities including hypotension and hypoalbuminemia, significantly limiting the availability of this treatment option to highly specialized cen ters [242] . Also, IL - 2 may ai d in the reconstitution of Tregs, which is associated with worse clinical outcome. There are ongoing clinical trials exploring lower IL - 2 dose in conjunction with ACT that may determine whether IL - 2 is the best cytokine for adoptive lymphocyte preservation . Other

74 cytokines that are being investigated include IL - 15. Overall, TIL - ACT promises to be an effective cancer therapy modality that is being broadened for use in other cancers including colorectal, pancreatic, uterine, and gastric cancers [252] . Optimization of TIL -

ACT is still needed due to high cost associated with the individualization of the therapy and toxicities ass ociated with it. Also, the response observed seems to vary between trials and tumors, necessitating a better understanding of the mechanisms that best regulate TIL activity once infused. Lastly, the extraction and expansion process for TILs needs to be str eamlined for broad usage, especially by smaller institutions that lack the infrastructure to generate TILs and treat patients undergoing adverse events. The use of TILs is limited to patients with tumors that are immunogenic and easily accessible for resec tion, and patients that can tolerate the lymphodepletion and IL - 2 - mediated adverse effects [253] .

3.2: LYMPHOKINE - ACTIVATED KILLER (LAK) CELLS

LAK cells are peripheral blood leukocytes that are isolated from patients and activated with IL - 2 in vitro to enhance cytotoxic function against autologous cancer cells. LAK cells demonstrated significant cytotoxic activity in vitro against a wide variety of solid tumors including ovarian cancer, breast cancer, and melanoma [254] . Bertelli et al. reported an increase in survival of 16 inoperable hepatocellular carcinoma patients who were treated with autologous LAK cells and high dose IL - 2, compared to IL - 2 alone, though no decrease in tumor mass was detected [255] . This study also demonstrated that

LAK cell infusi ons are safe and feasible. Rosenberg and colleagues published a report in

2011 of three clinical trials testing LAK cells against RCC. Leukaphereses from each patient were activated ex vivo with IL - 2 for >500 RCC patients. The objective response rate was 2 2% but neither clinical response rate nor overall survival were significantly higher in LAK - treated patients compared to IL - 2 alone [215, 252] . Similar to other cell therapies that require IL - 2 infusion, LAK - ACT induces signifi cant IL - 2 - mediated toxicities. Due to unclear responses to LAK cells and the side effects, interests in LAK -

ACT have generally been diverted to other immunotherapies [253] .

3.3: CYTOKINE - INDUCED KILLER CELLS

Cytokine - induced killer (CIK) cells are a mixture of cytotox ic lymphocytes extracted from peripheral blood. CIKs include CD3+/CD56+ cells (NKT), NK cells, and T cells, but the main effector cells reported are NKT cells that make up 20 - 30% of CIK cells

[253, 256] . Treatment of peripheral blood mononuclear cells (PBMCs) with cytokines like IL - 2, OKT3, and IFN - γ depletes unwanted cells such as B cells and expands CIK cells. Expansion of >1000 - fold has been reported over 21 - day in vitro culture [256] . As compared to LAK cells, CIK cells contain CD8+ T cells with diverse TCR specificities, have high proliferation rate, and demonstrate minimal cytotoxicity to normal cells [253] .

Compared to LAK cells and PBMCs, preclinically CIK cells have been shown to exhibit higher anti - tumor cytotoxic ac tivity in vitro and in murine models against hepatocellular carcinoma [257] . CIK cells have been used to treat patients with various cancer cells including lymphoma, non - small cell lung cancer, and gastric cancer. Against hematologic malignancies, CIK cells were reported to be moderately effective in the handful of phase

I clinical trials performed. In a trial of autologous CIK cells to treat 9 patients with

Hodgkin’s and non - Hodgkin’s lymphoma, 2 patients had partial responses and another 2 had stable disease [256] . Allogeneic CIK cells were used in treating 6 leukemia patients that relapsed after hematopoietic stem cell transplantation. Four patients developed

GVHD, one patient had stable disease, and three achieved complete responses [258] .

Jingting and colleagues performed a phase I trial with autologous CIK cel ls on 57 patients with stage IV gastric cancer [259] . The gro up that received both CIK - ACT and chemotherapy had decreased serum tumor markers and prolonged 2 - year survival as

77 compared to the control group that received chemotherapy alone. Co - culture of CIK cells with DCs prior to infusion has also been shown to resu lt in positive outcomes [256] .

Similar outcomes have been reported for CIK - ACT against other cancer types in phase II and III clinical trials [256, 259] . The survival of CIK cells following infusion in patients requires further evaluation since potent effector cells are prone to apoptosis [256] .

Therefore, strategies to prolong the persistence of CIK cells could lead to improvement in patient outcomes.

3.4: ENGINEERED T CELL RECEPTOR T CELLS

T cells can be endowed with tumor - specific T cell receptors (TCR) for TCR adoptive cell

therapy (TCR - ACT). When a tumor antigen presented on the tumor cell MHC is engaged

by TCR, the immunoreceptor tyrosine - based activation motifs (ITAMs) are phosphorylate d inducing intracellular signaling cascade, and resulting in the release of cytotoxic granules and cytokines from T cells [253] . TCRs can be genetically modified using methods such as lentiviral transduction and electroporation. Prior to infusion of the tumor - specific T cells, p atients commonly undergo a lymphodepleting regimen for solid tumors or stem cell transplant for hematologic malignancies. TCR - ACT has gained traction following the clinical success of TIL - based studies and the difficulty in obtaining

TILs from certain rese cted specimen. For example, 20% of T cell therapy protocols registered with the NIH’s Office of Biotechnology Activities are for gene - modified T cells [260] . TCR - ACT against the cancer - testis antigen NY - ESO - 1 in 11 melanoma subjects and 6 synovial cell sarcoma subjects resulted in an overall response rate of 50% by melanoma subjects, with 4 patients achieving complete remission [261] . The subjects with synovial cell sarcoma achieved a 67% objective response rate. Clonal T cells designed with a TCR specific for a particular tumor antigen was tested but it was ineffective against tumor cells with the ability to downregulate MHC I expres sion [262] .

A trial for melanoma patients targeting MART - 1 resulted in 9 of 14 subjects with reductions in tumor size, but the responses were not durable [260] . TCR - ACT certainly has its advantages including the ability to target intracellular proteins while chimeric antigen receptor T cells are restr icted to recognizing cell surface proteins. In addition,

TCR - ACT has been shown to induce remission in a proportion of patients in various

79 studies. Persistence of the adoptively transferred gene - modified T cells has been associated with prolonged remission , and can further enhance TCR - ACT effectiveness

[260] . For example, selection of central memory T cells may be a strategy for encourag ing enduring T cell population in patients. Also, strategies to prevent the immunosuppressive factors, especially Tregs, present in the tumor may also improve the success of TCR - ACT.

3.5: CHIMERIC ANTIGEN RECEPTOR T (CAR - T) CELLS

Chimeric antigen receptors (CARs) are artificial proteins designed to express an antigen

recognition domain, usually an extracellular single chain variable fragment, that is joined

to an intracellular signaling domain(s) by a hinge region linked to a trans membrane

region [263] . The goal of CAR - T cells is to generate tumor - specific T cells in large

quantity with specific and enhanced anti - tumor response in patients. CARs have been designed against a number of cell surface molecules such as CD19 and HER2, and have been evaluated in phase I clinical trials. The only clinically effective CAR - T cell

therapies are anti - CD19 CARs, whose anti - cancer activities include direct tumor killing

and cytokine production [264] . The reasons why only the CD19 - CAR - T cell therapies have been effective are not fully understood. CD19 - CAR - T cell therapy was used to treat patients, following chemotherapy, with low grade B cell malignancies, resulting in significant antitumor responses including six of eight patients achieving objective remissions [265, 266] . B cell acute lymphoblastic leukemia (ALL) patients have very poor prognosis and often relapse. Brentjens et al. reported the results from a trial involving five relapsed B - ALL subjects t reated with autologous CD19 - CAR - T cells expressing the costimulatory molecules CD28 and CD3ζ, following salvage chemotherapy [265] . After CAR - T cell treatment, the subjects no longer had minimal residual disease, a characteristic that is lin ked to poor prognosis in B - ALL patients. Also, a trial involving three CLL patients resulted in 2 achieving complete remission, and one of the patients developing sustained B cell aplasia [267] .

CD19 - CAR - T has been FDA - approved for the treatment of chemotherapy - resistant B cell malignancies, providing patients previo usly with little hope of survival another method

by which they can seek a cure. Patients with B cell acute lymphoblastic leukemia have up

to 90% response rate, a response rate that is unprecedented and incredible [263] . Despite

the high response observed, patients treated with CAR - T cells experience a range of

toxicities associated with the treatment. Toxicities associated with CAR - T therapy

include hypotension, tumor ly sis syndrome and cross - reactivity with normal proteins by

modified CARs leading to life threatening adverse outcomes such as cytokine release

syndrome and multi - organ failure [264, 266] . Prior to CAR - T cell infusion, patients undergo lymphodepleting regiment that creates a supportive environment fo r the CAR - T

cells to grow and persist. However, this immunosuppression also depletes Tregs which

are critical for the maintenance of homeostasis. Another challenge to the long - term

efficacy of CAR - T cells is the development of tumor cells that have “lost” the targeted

antigen and clearance of CAR - T cells by the immune system.

Newer CAR - T cells are being designed that express CARs based on natural receptors or

based on receptor - binding domain of ligands [262] . These CAR - T cells are at different

preclinical and clinical stages. Some of the natural receptors being engineered on the

CAR - T cells are expressed on NK cell receptors such as NKp30 and NKG2D. The

extracellular dom ain of the NK cell receptor is intact while the intracellular domain is

linked to a variety of cytoplasmic costimulatory molecules. Natural receptors are less

immunogenic and are able to recognize multiple stress - induced or overexpressed ligands

on tumor c ells. Addition of multiple costimulatory domains, characteristic of fourth

generation CARs, enhances the activity of these CAR - T s against liquid and solid tumors.

The NKG2D - CAR - T cells express NKG2D fused to the CD3ζ TCR signaling domain to

recognize NKG2D ligands MICA, MICB, ULBP1 - 6, that are often upregulated on cancer

cells. Preclinically, NKG2D - CAR - T cell has shown improved overall survival in murine

lymphoma, multiple myeloma, and ovarian cancer models [262] . Two safety studies have

been performed to understand the potential toxic effects of NKG2D - CAR - T cells in humans. The first study revealed significant morbidity and mortality with the NKG2D -

CAR - T cells, which was exacerbated with pre - treatment with cyclophosphamide [262,

268] . The second study focused on determining the maximum tolerated dose (MTD), and reported the MTD as 10 7 cells per mouse [264] . This dose did not cause major toxicities

in both healthy and tumor - bearing mice. These preclinical stu dies prompted the initiation

of Phase I clinical trials to evaluate NKG2D - CAR - T cells against multiple cancers

including acute myeloid leukemia (AML), multiple myeloma (MM), and five solid

tumors [262] . Although these natural receptor - based CAR - T cells are less immunogenic as compared to the clinically available anti - CD19 CAR - T cells, there is still a concern for off - target effects. Natural receptor - based CAR - T cells do not have the inhibitory signaling that regulates NK cell activation, raising a concern that these CAR - T cells may exhibit broad off - tumor effects leading to overt toxicity. The off - target effects of these natural receptor - CAR - T cells need to be better und erstood to ensure their clinical safety.

More studies are being performed to optimize in vivo safety and efficacy of CAR - T cells.

Preclinically, strategies for improving efficacy include lowering the threshold of

activation of CAR - T cells, engineering CAR - T cells with dual targeting domains,

improving CAR - T intra - tumoral migration, and increasing CAR - T cell resistance to

intrinsic or induced immunosuppression. In addition, efforts are directed towards

improving the fitness of CAR - T cells to persist in pati ents longer by optimizing the

metabolic capacity of CAR - T cells. Strategies for improving safety of CAR - T cells

includes identification of tumor - associated antigens that can be targeted without causing

severe toxicities, and engineering CAR - T cells with inducible suicide system or on - or

off - switches and coupling the CAR - encoding gene to a hypoxia – sensitive subdomain,

increasing CAR expression in the hypoxic tumor microenvironment [260, 263] . An interesting approach to abrogate the toxic side effects of CAR - T cell therapy is inhibitory

CARs (iCARs) [269] . These iCARs are based on the potent checkpoints CTLA - 4 and

PD - 1 and reported to have a temporary initial effect in vitro and in vivo against prostate cancer cell lines. ICARs therefore enable the CAR - T cells to function upon encountering

the tumor antigen at a later time. CTLA - 4 and PD - 1 engagement on T cells inhibits T cell activation, cytokine release, and proliferation [269] . ICARs mimic the natural regulatory system designed to prevent overt T cell responses and could be the solution to prevent the

CAR - T - associated toxicities.

3.6: BI - SPECIFIC T CELL ENGAGERS ( BiTEs )

Bispecific T cell engagers (BiTEs) are bispecific antibodies composed of two single -

chain Fv domains to recognize a tumor - associated antigen and the other to bind CD3 on

T cells [253] . Compared to other bispecific antibodies designed to bind T cells to tumor cells and act ivate T cell activity, BiTEs have overcome the limitations that others had

[270] . BiTEs demonstrated a 100 - 10,000 - fold higher efficacy in tumor cell lysis as compared to other CD3 - based bispecific antibodies. Impor tantly BiTEs can induce target cell lysis by unstimulated T cells without the need of another T cell stimuli and are homogeneous in nature. The mechanism of action of BiTEs includes induction of an immunological lytic synapse by forcing tumor cells and T c ells into close proximity, polyclonal activation of CD4+ and CD8+ T cells, and redirected target cell lysis by

BiTEs. It has been shown that BiTEs do not lyse tumor cells or induce tumor cell lysis in the absence of T cells, but the reason why is unknown [271] . In vitro analyses have not revealed any human tumor cell line, expressing a tumor antigen, that is fully resistant to

BiTE - mediated effects [270] . A CD19/CD3 - BiTE (blinatumomab) has been approved by the FDA for treating acute lymphocytic leukemia (ALL) followi ng clinical trials demonstrating its safety and efficacy [272] . A phase I trial of blinatumomab against non -

hodgkin’s lymphoma resulted in an overall response rate of 82% [272] . Eleven out of

eighteen patients are still responding at nearly 3 years after the therapy ended. Side

effects were mostly flu - like symptoms such as fatigue, headache, and pyrexia. More

severe, but reversible adverse events were CNS - related sym ptoms such as tremor,

aphasia, and convulsions. Phase II trial of blinatumomab against ALL resulted in

complete response in 16 out of 20 subjects [272] . A multi - institutional phase 3 trial

testing blinatumomab against ALL resulted in an overall survival of 7.7 months for

blinatumomab - treated patients as compared to 4 months in chemother apy - treated ones

[273] . Adverse events grade 3 and higher were reported by the blinatumomab - treated patients, suggesting that there is a need for optimization of this drug.

Limitations of BiTEs includes its short half - life of 2 - 3 hours wh ich requires continuous

IV infusion over several weeks, although this limitation also makes it easier to rapidly prevent or ameliorate treatment - mediated adverse effects [272] . Also, CD19/CD3 - BiTE is able to stimulate T cells without any other stimuli creating the possibility of aberrant T cell activation and activity. Reports so far ha ve claimed a tumor - limited activity of BiTEs but it is not clear whether this phenomenon will be true for other target cells and target antigens. Efficacy of BiTEs is dependent on T cell presence in or recruitment to the tumor. Tumors often contain low num bers of tumor - infiltrating lymphocytes or unresponsive tumor - associated T cells, therefore verification of the ability of BiTEs to recruit and/or reactivate T cells is crucial.

3.7: CYTOKINES AND MONOCLONAL ANTIBODIES

Naturally, cytokines can modulate the activity of immune cells and they have been used

in the clinical and preclinical settings to enhance cellular antitumor responses. Cytokines

such as GM - CSF, IL - 12, and IL - 2 have been used in combination with other immu ne therapies to improve patient outcomes. For example, GM - CSF, a potent regulator of differentiation and proliferation also drives the maturation of DCs and monocytes, and enhances antigen presentation to T cells [232] . GM - CSF has been combined with whole tumor cell vaccines and DC vaccines, as an immune adjuvant [274] . As discussed in the

previous chapter, GM - CSF is also a potent activator of MDSCs and promotes MDSC

expansion, therefore GM - CSF use requires caution to prevent the induction of immune

suppress ion [275] . IL - 12 is a strong promoter of IFN - γ release which promotes Th1

polarization and exhibits anti - angiogenic activity. IL - 12, as a monotherapy, did not

demonstrate efficacy against melanoma and renal cell carcinoma [276] . However, it may

be effective as adjuvant, perhaps when co - administered with a vaccine. IL - 12 causes

severe toxicities following systemic administration [275] . Three important cytokines that

have stirred a great deal of interest for cancer therapy are described below.

IL - 2

IL - 2 has been FDA - approved for the treatment of advanced melanoma in adults, renal

carcinoma, and hematological malignancies, following clinical trials that demonstrated

anti - tumor effects of IL - 2 with or without LAK cell co - adm inistration [276] . IL - 2 is a

well - known activator of immune cells such T and NK cells however it has a couple of

disadvantages. High dose IL - 2 has been observed to induce severe toxicities in patients

including gr ade 3 and 4 line sepsis, grade 4 cardiac ischemia, and grade 4

neuropsychiatric toxicity. Commonly, these toxicities were observed to reverse once

treatment was stopped but up to 4% mortality has been rep orted in IL - 2 - treated patients

[277, 278] . In addition to severe side effects, IL - 2 mediates the expansion of Tregs, a

group of T cells that suppress the effector functions of NK cells and CD4+/CD8+ T cells

[275] . This has led to the identification of other cytokines that have similar effects on NK

cells while omitting the side effects.

IL - 15

Other cytokines in the γ - chain family of cytokines that are notable for their

immunostimulatory roles are IL - 15 and IL - 21. In preclinical studies, IL - 15 has been shown to be critical for memory CD8+T cell and NK cell development and homeostasis, and to rev erse T cell anergy [232] . Unlike IL - 2, IL - 15 does not stimulate Treg expansion and has been shown to be safe in patients, although there was a report of a dose - dependent toxicity when IL - 15 was administered as an exogenous bolus [233] . IL - 15 has drawn significant interest in the last few years and has been evaluated in several clinical trials for the treatm ent of advanced cancers. The first in - human trial of rhIL - 15 infusion in patients in cancer resulted in 10 - fold proliferation of NK cells and efflux of NK cells and CD8+ T cells from peripheral blood [279] . Two patients had an objective respons e namely clearance of lung lesions. Toxicities reported include grade 3 hypotension and thrombocytopenia. In combination with NK adoptive cell therapy, IL - 15 was given to 6 pediatric patients with refractory solid tumors in a phase I/II clinical trial, fol lowing stem cell transplantation [280] . Four of the pat ients showed clinical response but all patients

died of disease at a median follow - up of 310 days. Recombinant human IL - 15 has a short

half - life, prompting the development of an alternative. ALT - 803 is a IL - 15/IL - 15Rα complex that is fused to an IgG1 Fc re gion, and has a longer half - life. The IL - 15 has been mutated in the ALT - 803 complex to increase its biological activity and affinity for the IL -

2R and IL - 15βγR [281] . Recently, a phase I trial was completed in which patients with metastati c non - small cell lung cancer were treated with a combination of ALT - 803 and nivolumab (PD - 1 inhibitor). The MTD was not reached and there were no dose - limiting toxicities reported. The most common adverse events were minimal including flu - like symptoms. Tw o out of eighteen patients who were treated experienced fatigue and were lymphocytopenic. One patient had a grade 3 myocardial infarction. Other trials of ALT -

803 in combination with other immunotherapies are ongoing.

IL - 21

Interleukin - 21 is a cytokine, similar to IL - 2, that activates and induces proliferation of

NK cells and cytotoxic T lymphocytes. IL - 21 is mainly produced by CD4+ T cells and promotes the expansion of naïve and memory CD8+ T cells in vitro [232] . IL - 21 has been shown to augment NK cell antitumor activities, and inhibit the activation and activity of iDCs [232] . Therefore, IL - 21 has also been an area of interest and there is a trial ongoing of IL - 21 - expanded NK cells for the treatment of AML patients.

Enhancing NK cell ADCC

Antibody - dependent cellular cytotoxicity (ADCC) is a mechanism of cell - mediated

immune response that has been a target for the development of immunotherapies.

Following the success of allogeneic NK cell - m ediated clearance of AML, improving NK

cell function via enhancement of ADCC received heightened efforts attempt to improve

the outcomes of adoptive NK cell therapy. Monoclonal antibodies against oncogenic

proteins have been designed and tested against var ious tumor types and have resulted in improvements in overall survival and time to relapse in a variety of cancers including breast and hematological cancers [282] . Monoclonal antibodies serve as tools for the

promotion of more specific and efficient antigen binding. IgG1 is the primarily employed

immunoglobulin for therapeutic development due to its long half - life in blood (~21 days)

and increased capability of inducing strong ADCC as compared to other heavy chain

isotypes.

Human IgG1 consists of two immunoglobulin light chains and two immunoglobulin

heavy chains covalently and non - covalently associated to form two Fab and one Fc

regions connected by a hinge region (Fig 3.1) . The Fab regions consist of variable and

conserved chains of the heavy and light chains that contain identical antigen - binding site.

The Fc region contains sites for molecules that can induce effector functions, including

the F c receptor CD16 [282] .

Although these therapeutic antibodies have achieved high in vitro cytotoxity, their

efficacy in vivo is low. The reduced effectiveness in vivo could be due to competition for

Fc receptors by endogenous serum IgG. In addition, genetic analyses have indicated that

polymorphism of Fcγ receptors resulted in differential affinities for therapeutic

antibodies. B cell follicular ly mphoma patients that express the high affinity 158 - Val

allotype of FcγRIII responded better to rituximab as compared to patients carrying the

low affinity 131 - His allotype [283] . Study of patients with HER 2+ tumors indicated a

higher ability to mediate ADCC in complete and partial responders to trastuzumab

90 treatment. Surgical samples and biopsies revealed that none of the tumor downmodulated

HER2 and a strong infiltration of leukocytes were observed in all the samples, suggesting that ADCC is an important mechanism of action for trastuzumab in vivo [284] . The importance of ADCC is well recognized and future therapeutics will be developed to enhance ADCC.

The development of monoclonal antibodies against oncogenic targets have expanded tremendou sly over the last two decades. Antibodies have been developed to target CD20 and CD19 for B cell malignancies, EGFR and HER2 for breast malignancies, TNF and

CEACAM5 for colon cancer, and many others. This section will discuss two important monoclonal anti bodies that led to the important contributions of monoclonal antibodies as effective targeted treatments.

Figure 3.1 . The Structure of human IgG1 antibody. IgG1 monoclonal antibodies consist of two immunoglobulin light chains and heavy chains connected by a disulfide bond between CH1 and CL. A pair of the variable regions of the light and heavy chains form the antigen binding site. The CH2 domains of the Fc regions contain covalently - attached oligosaccharide that is critical for generating an effector r esponse following antigen binding. Adapted from [282] .

The anti - CD20 antibody, rituximab, was FDA - approved in 1997 for the treatment of non -

Hod gkin’s lymphoma after demonstrating up to 48% response rate in patients with

relapsed low grade B cell lymphoma [285, 286] . The development of the afucosylated anti - CD20 antibody obinutuzumab changed the landscape of treatment of patients with lymphomas as obi nutumab demonstrated increased ADCC and antitumor activity as compared to other anti - CD20 antibody such as rituximab. A phase III trial in which patients were treated with either obinutuzumab and chemotherapy or rituximab and chemotherapy resulted in an es timated 3 - yr progression - free survival rate of 80% for the obinutuzumab arm as compared to 73.3% in rituximab - treated patients [287] . Response rates and rates of adverse events leading to death were similar for the two arms, while high grade adverse events were observed in 74.6% of the patients who received obinutuzmab as compared to 67.8% of rituximab patients. Combination of obinutuzumab and chemotherapy was FDA - approved in 2017 for the treatment of patients with follicular lymphoma. Despite the effectiven ess of the various immunotherapies, patients with certain B cell malignancies, such as follicular lymphoma eventually relapse. Other

forms of combination therapies are being developed including immunotherapy and

radiotherapy [286] .

Anti - human epidermal growth factor receptor 2 (HER2) therapy has led to a dramatic increase in the survival of patients with HER2+ metastatic breast cancer [288] . HER2 is a proto - oncogene, overexpressed in approximately 30% of breast cancers, and associated with aggressive disease and poor survival in breast cancer patients. Trastuzumab

(Herceptin) is an a nti - HER2 monoclonal antibody that was tested in a phase I trial 2001.

Compared to chemotherapy alone patients, patients treated with both trastuzumab and chemotherapy had a longer time to disease progression and higher objective response rate. Trastuzumab is associated with significant cardiotoxicity in 27% of patients treated with trastuzumab and chemotherapy that resolved with standard medical management

[289] . Trastuzumab resulted in a response rate of 15 - 40% when used as a monotherapy by patients with metastatic HER2+ breast cancer in another trial [282, 284] . When compared directly, trastuzumab combined with chemotherapy was more efficacious to trastuzumab as a monotherapy. The treatment guidelines now require HER2+ breast cancer patients to be treated with chemotherapy and trastuzumab, unless the patient cannot tolerate the regimen at which time the patient should receive trastuzumab alone

[288] . The cardiotoxicity associated w ith combining trastuzumab and anthracyclines led to the abandonment of anthracyclines for taxanes, and cardiac function of patients must be monitored while on trastuzumab. Initially, the mechanism of action of trastuzumab was ascribed to down - modulation of HER2 receptor based on in vitro studies. Treatment of HER2 - overexpressing breast cancer cell lines with trastuzumab resulted in decreased protein expression of HER2 and inhibition of tumor growth [284] . The success of

93 trastuzumab led to the development of other HER2 - targeting agents such as new antibodies, kinase - inhibitors, and antibody - conjugated drugs.

Inhibitory receptor blockade

NK cell activity can be greatly enhanced by blockade of inhibitory signaling. Much work has been directed towards developing neutralizing antibodies to a few of potent NK cell inhibitory receptors. This section will include a discussion of a handful of the se neutralizing antibodies. A group of the most important inhibitory receptors are the KIRs which recognize classical and nonclassical MHC molecules. Lirilumab (IPH2102/BMS -

986015) is a fully human anti - KIR monoclonal antibody designed to block the inhibit ory receptors KIR2DL1, 2, 3. Lirilumab was tested in a phase I trial against AML which resulted in mild and transient adverse events mainly of infusion syndrome [290] . At higher doses, lirilumab resulted in sustained full KIR saturation for more than 2 weeks and transient increases in TNF - α was detected. Phase II trial of lirilumab in smoldering multiple myeloma resulted in only 1 out of 9 patient achieving minimal response defined as <50% decrease in M protein [291] . No grade 3 or 4 toxicities were reported. Lirilumab is currently being used to treat patients with advanced solid tumors in combination with other immunotherapies such a s nivolumab and elotuzumab in various clinical trials.

KIR3DL2 is another target for immunotherapy as it is expressed abundantly on cutaneous

T cell lymphoma (CTCL) cells and expressed on only a small proportion of immune cells. CTCL is a group of heterog eneous diseases that differ in immunophenotypic characteristics, and prognosis and survival of its victims. A humanized antibody that

94 recognizes KIR3DL2, IPH4102, proved to be effective at driving the lysis of primary

Sézary patient cells [292] . IPH4102 is currently in phase I trial to determine its safety in the treatment of CTCL patients.

NKG2A is a NK cell inhibitory receptor that complexes with CD94, and is expressed by not only NK cells but also by cytotoxic T lymphocytes. CD94/NKG2A provide a negative feedback loop to limit TCR activation. The non - classical MHC I molecule

HLA - E is the natural ligand for CD94/NKG2A, and is often overexpressed on cancer cells, serving as a core mechanism of immune suppression [293] . Monalizumab

(IPH2201) is a monoclonal antibody clinically developed to inhibit NKG2A signaling in

NK cells. Specifically, it binds to the NKG2A subunit of the CD94/NKG2A complex.

Blockade of NKG2A on NK cells from CLL patients resulted in restoration of NK cytotoxic activity against HLA - E - expressing targets. Monalizumab was also proven to demonstrate high specificity and affinity for NKG2A in preclinical studies [294] .

Monalizumab is currently in clinical trials in combination with other immune modulatory agents against various solid tumors [146] .

CHAPTER 4 : MEMBRANE - BOUND IL - 21 - BASED NK CELL FEEDER CELLS

DRIVE ROBUST EXPANSION AND METABOLIC ACTIVATION OF NK

CELLS

ABSTRACT

Natural Killer (NK) cells are innate immune lymphocytes that can kill cancer cells in an

HLA - independent fashion. NK adoptive cell therapy is a promising cancer therapeutic approach, but there are still significant challenges that limit its feasibility an d clinical

efficacy. One major difficulty is manufacturing sufficient numbers of highly active and

proliferative NK cells ex vivo, to meet the cell doses necessary to provide clinical benefit.

A novel NK cell feeder cell line termed ‘NKF cells’ was created by overexpressing

membrane bound IL - 21. NKF - expanded NK cells were capable of inducing robust and

sustained proliferation (>10,000 fold expansion at 5 weeks) of cytotoxic NK cells. In

addition to robust proliferation, the feeder cell expanded NK cells exh ibit increased levels

of NK cell activating receptors and adhesion receptors, and better cytotoxic function

against a panel of both blood cancer and solid tumor cancer cells when compared to IL - 2

activated non - expanded NK cells. The NKF - expanded NK cells a lso demonstrate efficacy

in a mouse model of sarcoma leading to a significant reduction in lung metastases and in a

model of T cell leukemia leading to improved mouse survival. Studies to elucidate

mechanisms through which membrane bound IL - 21 promotes NK cell proliferation,

revealed that there is an activation of a STAT3/c - Myc pathway and increased NK cell

metabolism with a shift towards aerobic glycolysis. The NKF feeder cell line is a promising

new platform that enables the large scale proliferation of h ighly active NK cells in support

of large scale third party NK cell clinical studies. These results also provide mechanistic

insights into how membrane - bound IL - 21 regulates NK cell expansion.

INTRODUCTION

Natural killer (NK) cells, comprising 10 - 15% of peripheral blood lymphocytes, play an

important role in immune surveillance due to their innate ability to kill cancer and virally

infected cells without prior sensitization [295] . NK cells are identified by the surface expression of CD56 and absence of the T cell marker CD3. The FcγRIII protein CD16 also expressed on a subs et of NK cells, enhances NK cell cytotoxic function by aiding in antibody - dependent cellular cytotoxicity (ADCC). NK cell function is largely controlled by families of cell surface activating and inhibitory receptors that result in the activation or inhibi tion of the cytotoxic response. Activation signals are promoted by activating receptors such as NKG2D that recognize ligands including the stress - induced protein

MICA [296] . Inhibitory receptors recognize molecule s, like MHC class I, that are nearly

universally expressed on normal cells but are downregulated on cancer cells.

Due to their high cytotoxic activity against cancer cells, the adoptive transfer of NK cells

is a promising therapeutic strategy [177 , 297] . In colorectal and pharyngeal cancer, the

presence and activity of tumor - infiltrating NK cells is correlated with better prognosis and

reduction in metastatic risk [298 - 300] . Several types of cancers including hematologic malignancies, sarcomas, neuroblastoma, and ovarian cancer have been found to be particularly sensitive to NK cell therapy and led to improved outcomes [224, 231, 301] . In addition, roles for adoptive NK cell extends to their ability to facilitate the engraftment of hematopoietic stem cell transplants, provide post - transplant cancer surveillance, and suppress infection [302] .

Despite being a promising strategy for combating cancer, t he widespread clinical success of NK cell therapy has been limited partially by challenges in manufacturing large doses

of NK cells that are likely necessary for clinical efficacy [303] . Generating enough primary

NK cells for adoptive cell therapy (ACT) is a significant challenge because NK cells make

up only 10 - 15% of lymphocytes in peripheral blood. Clinical trials suggest that hi gh doses

of NK cells (>10 9 /kg) are both safe and likely necessary for efficacy [304, 305] . Therefore,

methods to expand large numbers of highly active clinical grade NK cells are essential to

optimally move this strategy forward [221, 306] . Another major advantage of

manufacturing large numbers of NK cells is their use as a universal donor “off - the - shelf”

ther apy. This expansion method supports the use of NK cells from a single donor expansion

to treat multiple patients. Universal donor NK cells are possible since unlike T cells, NK

cells target cells in an MHC - independent manner and are not thought to cause gr aft - versus -

host disease (GVHD).

Several ex vivo expansion platforms have been developed. NK cells have been expanded

with IL - 2 as well as various other cytokine combinations such as IL - 12, IL - 15, IL - 18, and

IL - 21 [307 - 309] . These cytokine - based expansion methods result in highly cytotoxi c NK cells with memory - like features, but limited yields (~4 - fold at day 10 of expansion) have been reported due to NK cell senescence. Expansion methods using irradiated accessory cells as antigen - presenting “feeder” cells tend to lead to more robust yiel ds [310 - 312] . For example, expanding NK cells with irradiated PBMCs and OKT3 can expand NK cells

2300 - fold by day 17 [313] . Another system involv es Epstein - Barr virus - transformed lymphoblastoid feeder cells which result in robust expansion for 2 - 4 weeks before the NK cells become senescent [314] . To combat the issue of senescence, K562 feeder cells were engineered to express membrane - bound IL - 21 (mbIL - 21) with 4 - 1 BB ligand allowing

longer culture. [117, 311, 312, 315, 316] . The mechanism for prevention of senescence by

this approach is poorly understood.

Other approaches to expand NK cells for ACT involve the use of immortalized NK cell

lines such as NK - 92 cells. One major challenge with this approach is that the cells must be

irradiated prior to patient administration which limits the efficacy of this therapeutic

strategy because the cells cannot expand in patients and sustain anti - tumor activity [317,

318] .

Here we report the creation of a novel mbIL - 21 based NK cell feeder cell line that can

support the generation of large doses of highly activated NK cells for clinical studies. In

addition, we chara cterize mechanisms through which mbIL - 21 appear to drive NK cell

growth and activation by activating IL - 21 - dependent signaling leading to changes in metabolism enabling the cells proliferate and kill cancer cells.

100

MATERIALS AND METHODS

Cell l ines

OCI - AML3 cells were obtained from DSMZ and HL - 60, 293T, HCT116, HT - 29, and

MDA - MB - 468 cells were from ATCC. TC106 cell line was previously described in

[319] . All cells were cultured in RPMI 1640 media (Hyclone) supplemented with fetal calf serum (Hyclone), penicillin (100U/mL), streptomycin (100ug/mL). Mycoplasma testing was performed on all cell lines at regular intervals using the Mycoplasma

Detection Kit - Quick Test by bimake.com.

NK cell isolation/purification

Peripheral blood mononuclear cells (PBMC’s) were isolated from the peripheral blood of healt hy donors via ficoll (GE Healthcare) gradient centrifugation. NK cells were isolated from PBMC’s through magnetic bead CD3 depletion followed by CD56 isolation

(Miltenyi biotec). NK cells were cultured as specified with irradiated NKF feeder cells

(90Gy). The work using human primary NK cells was approved by the IRB and performed according to the human subjects guidelines at University Hospitals Cleveland

Medical Center.

Cytotoxicity Assay

NK cell cytotoxic function was assessed by the measuring the number of live cells identified by calcein - AM (CAM) labeling. Target cells and NK cells were labelled with

CAM (BD Pharmingen) and calcein - violet (CV) (eBioscience), respectively. NK cells were co - incubated with target cells at the indicated ratios for 4 hours i n triplicate, and the samples were analyzed by flow cytometry (Attune NXT, Invitrogen) in 96 well plates.

101

The CV - positive NK cells were gated out for analysis. Percent cell lysis was calculated as

follows:

( ) # ( # ) 100 #

Phenotyping Assay

NK cells were washed in 3% F BS/PBS and stained with conjugated antibodies, as

specified, for 15 min at room temperature. The following antibodies were from Biolegend

(NKG2D - APC/Cy7, NKp46 - FITC, NKp30 - PE, CD158 - FITC, CXCR6 - PE, CD54 - FITC),

Novus Biologics (c - myc - BB421), BD Biosciences (NKP44 - BB515, CD57 - BV421,

DNAM - 1 - PE, 2B4 - BV421, Ki67 - BB786, p - STAT3 - Percp - cy5.5) and R&D systems

(NKG2C - PE, NK2A - alexa - 488). The cells were assessed by flow cytometry (BD

Fortessa) and data was analyzed using flowjo (BD).

Mouse models

Nod - SCID - IL - 2Rγ - nu ll mice (NSG, Jackson Laboratory) were injected with 1x10 5

TC106 (sarcoma) cells bilaterally s.c. Ten days following TC106 injection, when tumors were palpable, the mice (n=5 per group) received 1x10 6 NK cells injected intravenously

(IV) once a week or veh icle as well as IL - 2 (75,000U/ml). Tumor volumes were measured twice a week. Mice were sacrificed after they lost 15% of their initial weight.

The leukemia xenograft model was established by injecting 1x10 6 Jurkat cells IV into

NSG mice. Seven days follow ing injection, mice (n=5 per group) received 5x10 6 NK cells or vehicle weekly as well as IL - 2 (75,000U/ml). Mice were sacrificed after becoming moribund or losing 15% of their initial weight in accordance with out

102

institutional guidelines. All animal exper iments were approved by the CWRU

Institutional Animal Care and Use Committee.

Metabolism studies

Metabolic studies were performed using the XFe96 Analyzer (Seahorse Bioscience). NK cells were incubated in 200U/mL IL - 2 overnight (IL - 2 - NK), expanded with NKF feeder cells (NKF - NK), or expanded with OCI - AML3 feeder cells (OCI - NK). Cell energy tests were p erformed in Seahorse XF Base Medium in Minimal DMEM (Agilent) supplemented with 1mM pyruvate, 2 mM glutamine, and 10 mM glucose. Drug concentrations during the assay were 0.25µM FCCP, and 1µM oligomycin. Cells were plated in at least duplicates and OCR and ECAR were measured.

Statistical analysis

Nonparametric 2 - tailed unpaired t tests were performed to determine significance. The

following denotations for significance levels were used: * p<0.05, ** p<0.01 and, ***

p<0.001. 2 - way Anova was calculated, to determine significance among groups. The log -

rank (Mantel - Cox) test was performed to determine significance of mouse xenograft

survival data. All data represent at least 3 independent experiments, unless otherwise

specified.

103

RESULTS

Development and o ptimization of a mbIL - 21 based feeder expansion platform

In order to develop a novel NK cell expansion platform, we aimed to select a suspension cell line to use as a feeder cell that was efficiently lysed by NK cells and exhibits a low level of expression of HLA class I expression. A low level of HLA class I proteins is beneficial as certain epitopes are recognized by NK cell inhibitory killer inhibitory receptors (KIRs) which impair NK cell activation. Initially, the leukemia cell lines HL - 60

and OCI - AML3 were selected since they meet the pro posed desired characteristics (Fig

4 .1A). The two cell lines were irradiated and incubated with freshly isolated NK cells for

1 week to assess their expansion potential. OCI - AML3, a myeloid leukemia cell line, led

to a 4.6 - fold expansion compared to 2.9 - fold e xpansion with the HL - 60 cells (Fig 4 .1B).

A B

Figure 4 .1 . OCI - AML3 cell line exhibits low HLA - I expression and modest fold

expansion. A. HLA - A,B,C expression on PBMC, HL - 60 and OCI - AML3 cells as assessed

by flow cytometry. B. Fold expansion of NK cells using HL - 60 and OCI - AML3 as feeder

cells at a 5:1 ratio after 1 week.

104

Figure 4 .2 . Validation of successful expression of mbIL - 21 on OCI - AML3 (NKF).

OCI - AML3 cells were transduced with mbIL - 21 lentiviral construct. Selection for transduced cells was made using puromycin. Transduction was validated by Rt - PCR assessment of OCI - AML3 and N KF cells.

It has been reported that the presence of membrane - bound IL - 21 (mbIL - 21) can prevent

NK cells from undergoing senescence, markedly improving their ability to expand ex vivo

[315] . A novel NK feeder cell was developed using OCI - AML3 cells tra nsduced with mbIL - 21 (NKF) (Fig 4 .2). To expand peripheral blood isolated NK cells, NK cells (9% of the total PBMC (CI 95%: 6.178 - 15.14)) w ere isolated from PBMCs and co - cultured with irradiated NKF cells which were added weekl y (Fig 4.3 A). The purity of expanded NK cells for 15 donors was assessed by flow cytometry. After 2 weeks of expansion,

CD56+/CD3 - cells were approximately 94% (CI 95%: 92.32 - 96.53) of the expanded cells,

CD3+ cells made up <1% (CI 95%: 0.069 - 1.03), and B cells were virtually undetectable

(Fig 4.3 B - C). Approximately 87% of expanded NK cells were CD56+/CD16+ suggesting a large portion of these cells could mediate ADCC. The low T cell contamination post - expansion is important for avoiding potential graft - versus - host - disease for universal donor

NK cells.

105

The NKF expansion platform was optimized based on NKF - to - NK ratios and IL - 2

concentrations. The goal for feeder cell add ition was to generate robust expansion while

limiting the feeder cell numbers to avoid the unnecessary presence of excess dead feeder

cells in the final product. After 3 weeks, expansions at a 5 - 1 (NKF - to - NK) ratio resulted in an 8.3 - fold higher NK cell yi eld compared to a 1 - 1 ratio (p= 0.017) and a 2.7 - fold higher

NK cell yield compared to a 2 - 1 ratio (p=0.057) (Fig 4.3 D). Expansion yield at IL - 2 concentrations ranging from 10 - 1000 U/mL did not result in a statistically significant difference (Fig 4.3 E). This is consistent with previous reports that high IL - 2 concentrations do not impact the proliferative capacity of feeder cell expansion of NK cells [320 - 323] .

The yield at 200U/mL IL - 2 was the highest, therefore subsequent expansions were performed with a 5 - 1 feeder - to - NK cell ratio and 200U/mL IL - 2.

106

C D NKF:NK 1000 5:1 2:1 * 800 1:1

600

400

200

0 1 2 3 Weeks

E [IL - 2] 1000 1000 200 800 10

600

400

200

0 1 2 3 Weeks

Figure 4 .3 . NKF feeder cells enable NK cell proliferation. A. Schema the NKF based

NK cell expansion platform. B. NKF - expanded NK cells are highly pure. Representative

107

flow plots depicting purity of NK cells in the initial PBMC population and after expansion.

C. Relative fractions of cell types in NKF - expanded product, n=15. D, fold expansion of

NK cells at the indicated NKF:NK ratios, n=4. E. Fold expansion of NK cells at the indicated IL - 2 concentr ations, n=5. *p<0.05.

NKF - NK cells exhibit potent cytotoxic activity against both hematologic and solid cancer cells

Besides being able to generate large numbers of NK cells, another major challenge for adoptive NK cell therapy is insufficient cytotoxic activity against cancer cells. Therefore, the cytotoxic activity of NKF - expanded NK cells (NKF - NK) was assessed in comparison to the traditional source of NK cells used for adoptive cell therapy, IL - 2 overnight activated

NK (IL - 2 - NK) cells using a flow cyt ometry based cytotoxicity assay (see methods). NKF -

NK cells exhibited markedly increased cytotoxic activity against a wide variety of cancer cell lines as co mpared to IL - 2 - NK cells (Fig 4.4 A - B). NKF - NK cells achieved 31%

(p=0.003) and 37% (p=0.009) more ki lling of Jurkat and TC106 cells respectively than IL -

2 - NK cells at a NK - target ratio of 1 - 1. Unlike IL - 2 - NK cells, NKF - NK cells were able to effectively target a broad spectrum of cancer cell lines. To further assess NKF - NK cytotoxic activity, these cells were co - incubated with leukemia, lymphoma and colon cancer cell lines at a variety of NK - target cell ratios. NKF - NK cells demonstrated dose - dependent killing o f all cell types tested (Fig 4.4 C).

108

A B s i s y L

l l e C

0 I I 6 C J T - O A A L R K H R U J

Figure 4 .4 . NKF - NK cells exhibit potent cytotoxic activity against both hematologic and solid cancer cells . The cytotoxic activity of NKF - NK or IL - 2 - NK cells was assessed against hematologic malignancy cell lines (A) or solid tumor cell lines (B) after 4hr of co - culture, n=4. C. NKF - NK cells demonstrate dose - dependent cytotoxicity against cancer

cells. The cytotoxic activity of NKF - NK cells was measured using the indicated

NKF:Target ratios, n=4. *p<0.05, **p<0.01.

109

NK expansion with NKF cells leads to marked changes in cell surface phenotype

NK cell activation and function is coordinated by the engagement of cell surface molecules such as activating and inhibitory receptors. Using flow cytometry, the baseline expression levels of key NK cell surface molecules were measured in both IL - 2 - NK and NKF - NK cells. In order to make direct comparisons, the same donors were utilized for both populations of NK cells (n=4). Expansion with NKF cells led to increased expression of the activating receptors NKG2D, NKp30, and NK p44 as compared to IL - 2NK cel ls (Fig

4.5 A). Activating receptors such as NKG2D and the natural cytotoxicity receptors are crucial to NK cell activation and function [162] . Ligands for NKG2D are expressed in a variety of cancers such as sarcomas, lymphomas, leukemia, melanoma, hepatoma, and prostate cancer [324] . NKp46 plays a role in preventing metastases and NKp44 promotes lysis of a broad spectrum of cancer cell lines and cytokine release [325, 326] . There was also an increas e in the inhibitory receptor, NKG2A, a marker that is reduced in terminally differentiated NK cells, as reported by other groups [327, 328] . NKF - NK cells had a significant decrease in expression of the CD158 family of killer immunoglobulin receptors

(KIRs) that reco gnize inhibitory signals from MHC Class I epitope (Fig 4.5 B). Expansion with NKF cells also led to an increase in adhesion receptors (LFA - 1 and CD54) which are important for NK cell conjugation with tumor targe ts, enhancing cytolysis (Fig 4.5 C).

CD57 expr ession, denoting terminal differentiation of NK cells, was also measured and found to be de creased in NKF - NK cells (Fig 4.5 D). NK cells expressing high levels of

CD57 are limited in their continued proliferative capacity in vivo [174] . Finally, the expression of key chemokine receptors that regulate the in vivo trafficking of NK cells were analyzed. NKF expansion led to a decrease in CXCR4, a receptor reported to

110 sequester NK cells in the bone marrow [156] (Fig 4.5 E). In addition, an increase in CXCR6 suggests t hat NKF - NK cells can traffic to the liver, a common site of cancer metastases.

CXCR6 has also been indicated in the development of memory - like NK cells that persist following hapten or viral exposure [329] .

A 8 ** B *** C 6 *** 4 * 2 ***

-2 6 D C 6 4 0 1 2 2 4 4 3 D G G P P P C K K K K K N N N N N Activating Receptors

D ** ** *

*** **

Figure 4 .5 . mbIL - 21 signaling in NKF - NK cells leads to marked changes in cell surface phenotype . Expression of NK activating receptors (A), inhibitory receptors (B), adhesion receptors (C), NK terminal differentiation receptors (D), and trafficking receptors

111

(E) on IL - 2 - NK and NKF - NK cells were measured by flow cytometry, n=4. *p<0.05,

**p<0.01, ***p<0.001.

mbIL - 21 signaling promotes sustained NK cell expansion and increased metabolic activity

While short term co - culture with NKF cells leads to NK cell proliferation and activation, to support “universal donor” NK cell clinical studies sustained proliferation is necessary.

Unlike IL - 2 treated NK cells that rapidly undergo senescence, NKF cells enable long term

NK cell proliferation. For example, aft er 5 weeks of expansion there was an average of

10,973 - fold expansion (Fig 4.6 A).

In order to support the clinical translation of NKF - expanded NK cells, it is also necessary to utilize high capacity culture devices to accommodate the high cell numbers. G - R EX culture systems allow the culture of high densities of cells such as T cells due to the presence of a gas permeable membrane surface [330] . Therefore, we tested the ability of freshly isolated NK cells to expand in the G - REX device. NK cells were expanded for 2 weeks with NKF cells and the average NK cell yield was 89 - fold ( Fig 4.6 B). This demonstrates that the semi - automated G - REX system is capable of expanding NK cells utilizing NKF feeder cells.

112

A B n o i s n a p x E

d l o F

Weeks

Figure 4.6 . A. Fold expansion of NKF - NK cells at 5:1 feeder - to - NK ratio and 200U/mL

IL - 2 after 5 weeks, n=6. ***p<0.001. B, Expansion of NK cells using NKF cells at 5:1

feeder - to - NK ratio and 200U/mL IL - 2 for 2 weeks, using G - Rex flasks, n=8.

The proliferation of N K cells expanded with NKF cells was compared to that from parental

OCI - AML3 cells (OCI - NK). NKF expanded NK cells an average of 843 - fold NK cell

expansion after 3 weeks as compared to 200 - fold using O CI - AML3 cells (p=0.027) (Fig

4.7 A). Consistent with enha nced NK cell proliferation, the proliferation marker Ki67 was upregulated on NKF - NK cells as compared to OCI - NK cells (p=0.038). As expected, Ki67 levels are higher when either feeder cell line is used as compared to I L - 2 - NK cells

(p=0.0021) (Fig 4.7 A).

N ext activated molecules downstream of IL - 21 Receptor (IL - 21R) signaling were assessed to elucidate mechanisms of IL - 21 - dependent sustained NK cell proliferation. IL - 21R signaling is known to activate STAT3 which in turn can directly induce c - myc expression .

NKF - NK cells exhibited higher p - Stat3 expression than OCI - NK cells (p=0.041) and IL -

2 - NK cells (p=0.00038) (Fig 4.7 B). In addition, NKF - NK cells demonstrated higher

113

expression of c - myc as compared to OCI - NK cells (p=0.043) and IL - 2 - NK cells

(p=0.00085) ( Fig 4.7 B).

As the activity and proliferative status of immune cells is largely governed by their metabolic capacity, the impact of mbIL - 21 signaling on NK cell metabolism was investigated. After activation, the metabolism of immune cells typically shifts from

generating energy mainly through oxidative phosphorylation to aerobic glycolysis. The

oxygen consumption rate (OCR) which measures oxidative phosphorylation (oxphos) rate,

and the extracellular acidification rate (ECAR) which measures glycolysis were measured

in NKF - NK, O CI - NK and IL - 2 - NK cells (Fig 4.7 C). At baseline, NKF - NK cells and OCI -

NK cells had the same oxph os and glycolysis rates (Fig 4.7 D - E). Under stress, the NKF -

NK cells had higher OCR (p=0.043) and ECAR (p=0.0015) values c ompared to OCI - NK cells (Fig 4.7 D - E). Both expanded NK cell populations had higher OCR and E CAR values than IL - 2 - NK (Fig 4.7 D - E). The OCR/ECAR ratio of NKF - NK cells was close to 1, suggesting a balance of the glycoly tic and oxphos pathways (Fig 4.7 F).

As mbIL - 21 leads to enhanced NK cell proliferation and metabolism during feeder cell expansion, we also assessed its impact on NK cell cytotoxic activity. While NKF - NK cells demonstrate a marked increase in cytotoxic activity as compared to non - feeder expanded cells (Fig 4.4 A - B), there was no significant difference in cytotoxic function of NKF - NK

and OCI - NK cells (Fig 4.7 H). This result suggests that mbIL - 21 may not significantly

affect the cytotoxicity of NK cells at least in the presence of feeder cells.

114

A *

K K K N N N - I - - - 2 C F L K I O N

B *** 1000 *

800 ) I s l F l e M

600 c 3

T + 6 A 5

T 400 D S p C ( 200

0 K K K K K K N N N N N N - I - - - I - - - 2 F - 2 C F L C K L K I O N I O N

115

D E

200 ** n Aerobic Energetic

G o IL-2-NK F i t a

** )

r OCI-NK

4 i n

i 150 p

s NKF-NK m / e l R o

3 l m

a 100 R i p r ( A

d C R n E o C / 2

h 50 R O c C o t O i Quiescent Glycolytic 1 M 0 0 50 100 150 200 0 ECAR (mpH/min) IL-2-NK OCI-NK NKF-NK Glycolysis

H I

100 OCI-NK NKF-NK 80

0 3 I 0 L J 9 S 6 - 6 A 2 1 6 L M 9 M 2 A R H U H I - M C O

Figure 4 .7 . mbIL - 21 signaling promotes increased metabolic activity in NKF - NK cells but does not affect cytotoxicity. A. NKF cells lead to enhanced NK cell proliferation as compared to OCI - AML3 cells at a 5:1 feeder - to - NK ratio as measured by cell counts. The proliferation marker, Ki67, expression is higher in NKF - NK cells as compared to IL - 2 - NK and OCI - NK cells as assess ed by flow cytometry, n=4. B. NKF - NK cells exhibit increased pSTAT3 and c - myc levels compared to IL - 2 - NK and OCI - NK as measured by flow

116

cytometry, n= 4. C. OCR and ECAR measurements of IL - 2 - NK, OCI - NK, and NKF - NK

cells at baseline and after the addition of oligo (1 µM) and FCCP (0.25 µM). Average OCR

(D ) and ECAR (E) measurements of IL - 2 - NK, OCI - NK, NKF - NK cells at baseline and stressed conditions. F. OCR/ECAR ratio of IL - 2 - NK, OCI - NK and NKF - NK cells. G.

ECAR vs OCR plot illustrating the energetic state of the indicated NK cells, n=4. H. NKF -

NK cells demonstrate similar cytotoxic activity to OCI - NK cells as were measured by the

4 - hr cytotoxicity assay using hematologic malignancy or solid tumor cell lines, n=4.

*p<0.05, **p<0.01, ***p<0.001.

NKF - NK cells r educe tumor burden in mouse tumor xenografts and improve survival

To assess the therapeutic potential of NKF - NK cells for cancer therapy, mouse models of sarcoma and lymphoid leukemia were used. For the sarcoma model, the Ewing’s sarcoma cell line, TC106, was injected subcutaneously into immunodeficient NSG mice. The sarcoma model was employed because it leads to metastases to the lungs, the most common site of sarcoma metastasis in humans and a known site for NK cell trafficking in vivo [331,

332] . In this model, not only did we observe a reduction in the growth of the primary sarcoma tumor with NKF - NK cell administration, but ther e was a dramatic reduction in tumor metastases to the lung (Fig 4.8 A - C). Ki67 staining of lung specimens from vehicle - treated mice and NKF - NK - treated mice revealed decreases in proliferation of tumor cells in NKF - NK - treated mice as compared to vehic le - trea ted mice (p=0.023) (Fig 4.8 D - E).

In addition to evaluating the NKF - expanded NK cells for their ability to reduce tumor growth, these cells were also evaluated for their ability to prolong survival in a highly aggressive circulating lymphoid leukemia model . NSG mice were injected intravenously

117 with Jurkat cells to establish circulating disease. Mice injected with NKF - NK cells demonstrated an approximately 13 - day median increase in survival over contro l treated mice (p=0.0016) (Fig 4 . 8 F).

A B

Days Post - tumor cell injection C

118

E 0.5 * F a e r a

0.4 l a t o

T 0.3 / a e r a

0.2 e v i t i 0.1 s o P 0.0 Veh NKF-NK

p =0.016

Figure 4.8 . NKF - NK cells reduce tumor burden in mouse xenografts and improve mouse survival. A. NKF - NK cells exhibit efficacy in a mouse sarcoma model. TC106

119 cells were injected subcutaneously into NSG mice. Mice were either treated weekly with vehicle (Veh) or NKF - NK cells (NK). Tumor volumes were measured on indicated days.

5 mice per treatment group. B. H&E staining of resected lung tissue from Veh and NKF -

NK mice at 1X magnification. C. Quantification of tumor in resected lung tissue based on

H&E staining. D. Ki67 staining of resected lung tissue from Veh and NKF - NK mice at 1X magnification. E. Quantification of Ki67 staining in resected lung tissue. F. NKF - NK demonstrate efficacy in a mouse model of leukemia. J urkat cells were injected intravenously into NSG mice. Mice were either treated weekly with vehicle (Veh) or NKF -

NK cells (NK). Survival of the mice was determined. *p<0.05, **p<0.01.

DISCUSSIONS

Adoptive NK cell therapy exhibits promise in the area of cancer therapy, but the development of additional robust methods to expand large numbers of highly activated NK cells is important to continue to advance the field. In addition, further understanding of the mechanisms that enable this expansion is important for the development of optimal strategies. Here we report the development of a novel feeder line, NKF, that can support

“universal donor” NK cell therapy clinical trials and elucidate how mbIL - 21 impacts NK cell expansion and activation.

NKF cells were found to enable both large - scale ex vivo growth as well as activation of freshly isolated NK cells. The expanded NK cells exhibit significant cytotoxic activity against a wide variety of hematolog ic and solid cancer cells in vitro as compared to non - expanded, IL - 2 activated NK cells. These NK cells also exhibit anti - tumor activity in

120

mouse models of human leukemia and sarcoma. The expanded cells exhibit phenotypic

changes such as increased levels of NKG2D and NKp30, consistent with an activated state

despite their continued ability to proliferate.

In order to utilize NKF cells to manufacture clinical grade NK cells, a master cell bank was

created that will support a recently initiated clinical tria l (NCT02890758). While it is

feasible to expand NK cells with feeder cells in traditional flasks or gas permeable G - REX

flasks, further developments in NK cell manufacturing are warranted to support large scale,

universal donor clinical studies. While th e existing methods can support the manufacture

of tens of billions of cells, higher capacity culture systems will be necessary to efficiently

and cost effectively generate hundreds of billions or more NK cells from single donor

expansions. Based upon an av erage of expansion of over 10,000 fold at 5 weeks, it should

be feasible to manufacture greater than 4x10 12 NK cells starting from a single donor

apheresis sample. Methods to characterize the feasibility of feeder cell - based expansions

in large capacity b ioreactors such as the Xuri are ongoing.

As NK cells, unlike T cells, are not thought to elicit cause graft - versus - host disease, there are numerous benefits of developing “universal donor” off - the - shelf NK cells. Logistically, this strategy would dramatic ally lower the cost and increase the accessibility of this therapeutic strategy worldwide. Despite promising approaches for ex vivo expansion, the need of donors for each intended patient and the associated costs (both logistics and financial) that accrue with cell processing per patient still limit the feasibility of adoptive

NK cell therapy. Robust expansion systems such as the NKF feeder cells should enable the ability to manufacture NK cell doses for 100 or more recipients from a single donor. Being abl e to harvest NK cells from one donor would greatly alleviate the costs of this therapeutic

121 approach. Utilizing a donor not matched by HLA to recipients also increases the potential for NK cell alloreactivity to promote ‘graft versus leukemia (GVL) effects (without

GVHD). This has been found to improve disease free survival in certain cancer patient populations [333 - 335] .

Despite the ability to generate large numbers of active NK cells, a major challenge in realizing high levels of clinical efficacy is the maintenance of NK cell activity and proliferation in vivo. Tumor microenvironments exhibit immunosuppressive effects through the production of cytokines such as TGFβ and IL - 10 [336] . In order to improve on current NK cell therapy, preclinical studies suggest the combination of NK cells with immunomodulatory agents such as TGFβ inhibitors offer promise. For example, the combination of expanded NK cells with Galunistertib leads to a marked increase in anti - tumor efficacy in a mouse model of colon cancer metastasis [337] . In addition, the immunomodulatory agent lenalidomide, enhances NK cell cytotoxicity against multiple myeloma cells [338] .

It has been previously demonstrated that feeder cell - based expansion with mbIL - 21 leads to the ability of NK cells to avoid undergoing senescence as evidenced by a reduction in telomere shortening, though little is kno wn about molecular mechanisms through which mbIL - 21 supports continued NK cell proliferation [315] . Utilizing non - expanded NK cells and NK cells expanded with NKF or parental feeder cells lacking mbIL - 21, we further assessed mechanisms through which mbIL - 21 sustains NK cell expansion. Our studies demonstrate that mbIL - 21 leads to the activation of a well characterized IL - 21 dependent pathw ay consisting of STAT3 and cMyc. STAT3 activation is necessary for downstream effects of IL - 21 signaling and is a known inducer of c - Myc. The activation of cMyc is

122 known to regulate various cellular processes, important in NK cell proliferation and activit y including the induction of glycolysis, mitochondrial biogenesis, and cell cycle [339 - 341] .

Interestingly, NK expansion by the parental OCI - AML3 feeder cells that lack mbIL - 21 le d to partial activation of the STAT3/cMyc pathway likely due to the fact that the activation of many other receptor signaling pathways also can lead to minimal STAT3 activation.

Recently it has become well - recognized that metabolic reprogramming of immune cells is essential for their proliferative capacity and immune cell functions, most notably through the activity of c - Myc [339, 342] . In particular, immune cells increase their metabolism, for robust proliferation by ensuring sufficient biosynthetic precursors. This metabolic shift also enables the immune cells to survive in hypoxic environments as often o ccurs in the tumor microenvironment. While most previous studies have focused on T cells, reports exist that this phenomenon occurs in NK cells [343] . For the first time we have shown that feeder cell - expanded NK cells exhibit a marked metabolic shift by increasing both glycolysis and oxidative phosphorylation as compared to both parental OCI feeder cells as well as non - expanded, IL - 2 activated NK cells (IL - 2 - NK).

Overall, NK adoptive cell therapy is a promising therapeutic approach that depends on the development of robust ex vivo expansion pla tforms such as the NKF cells that can support the manufacture of highly active clinical grade NK cells.

123

CHAPTER 5 : INHIBITING TGF - β SIGNALING PRESERVES THE FUNCTION

OF HIGHLY ACTIVATED, IN VITRO EXPANDED NATURAL KILLER CELLS

IN AML AND COLON CANCER MODELS

(Adapted from [337] )

124

Abstract

Natural killer cells harnessed from healthy individuals can be expanded ex vivo using various platforms to produce large doses for adoptive transfer into cancer patients.

During such expansion, NK cells are increasingly activated and more efficient at kil ling

cancer cells. Adoptive transfer however introduces these activated cells into a highly

immunosuppressive tumor microenvironment mediated in part by excessive transforming

growth factor β (TGF - β) from both cancer cells and their surrounding stroma. Thi s

microenvironment ultimately limits the clinical efficacy of NK cell therapy. In this study,

we examined the use of a TGF - β receptor kinase inhibitor, LY2157299, in preserving the

cytotoxic function of ex vivo expanded, highly activated NK cells following sustained

exposure to pathologic levels of TGF - β in vitro and in a liver metastases model of colon

cancer. Using myeloid leukemia and colon cancer cell lines, we show that the TGF - β

driven impairment of NK cell cytotoxicity is mitigated by LY2157299. We d emonstrate

125

this effect using quantitative cytotoxicity assays as well as by showing a preserved

activated phenotype with high NKG2D/CD16 expression and enhanced cytokine

production. In a mouse liver metastases model of colon cancer, we observed significant ly

improved eradication of liver metastases in mice treated with adoptive NK cells

combined with LY2157299 compared with mice receiving NK cells or TGF β inhibition

alone. We propose that the therapeutic efficacy of adoptive NK cell therapy clinically

will be markedly enhanced by complementary approaches targeting TGF - β signaling in vivo.

INTRODUCTION

The clinical development of adoptive immunotherapy with natural killer (NK) cells has been facilitated by various ex vivo expansion platforms that yield cell doses sufficient to achieve some clinical efficacy [207, 211, 221, 312, 330, 344 - 347] . These expansion platforms typically involve co - culture of freshly isolated NK cells with irradiated antigen - presenting cells or feeder cells which are themselves sensitive to NK cell killing

[207, 211, 330, 345 - 347] . In the process of feeder cell killing, NK cells expand robustly and also acquire increasingly activated phenotypes re sulting in large numbers of highly activated NK cells capable of efficient tumor killing at low effector to target ratios.

To ensure the efficacy of these highly activated NK cells in cancer therapy, it is critical that these cells maintain their cytotoxic activity in vivo. A major obstacle in this regard is that the tumor micro - environment is enriched with several immunosuppressive cytokines, one of which is transforming growth factor β 1 (TGF - β) [348 - 353] . TGF - β is produced in

126

excess by tumor cells themselves, as well as by regulatory T cells, myeloid derived

suppressor cells (MDSCs) and other stromal cells in the tumor microenvironment.

Circulating TGF - β levels ranging from 5ng/ml to >20ng/ml have been described in both

hematol ogic malignancies and solid tumor patients [354 - 357] . These levels are higher than seen in healthy volunteers and correspond with impaired cellular immunity [351 -

354, 358 - 360] . Levels below 1ng/ml have been described in the peripheral blood and bone marrow of healthy volunteers while acute myeloid leukemia and myelodysplastic syndrome patients have levels ranging from 6 to 42ng/ml [356, 358] . In a study of 45 colorectal cancer patients, Narai et al reported circulating total TGF - β levels greater than

15ng/ml in patients with metastatic disease [355] . Those with liver metastases had the highest levels, up to 45ng/ml.

Pathologic levels of TGF β have been shown to impair both the innate and adaptive cellular immunity of cancer patients [163, 349, 359 - 361] . Postulated mechanisms by which TGF - β impairs NK cell function include down - regulated expression of activating receptors like NKG2D and CD16 (the FCγR mediating antibody - dependent, cellular cytotoxicity (ADCC)) and cytokine mediators/enzymes. It also coun teracts the NK pro - survival effects of IL - 2 and stimulates further proliferation of regulatory T cells. Small molecule kinase inhibitors and monoclonal antibodies targeting the TGF - β receptor have been explored as a means of enhancing cellular immune respo nse pre - clinically [163,

350, 361, 362] . There is at least one active clinical trial exploring the combination of a

TGF - β receptor inhibitor, LY2157299 (Galunisertib, Eli Lilly) with the PD - 1 monoclonal antibody Nivolumab, with a goal of enhancing the liberated T - cell response.

127

In the p rocess of developing allogeneic adoptive transfer of NK cells as a therapeutic strategy against various malignancies, we have adopted ex vivo expansion of NK cells using antigen - presenting feeder cells. In the process of generating large cell yields during expansion, the resulting NK cells are also significantly more activated and better efficient at killing both liquid and solid tumor targets. Our hypothesis is that these highly activated

NK cells will again have limited clinical efficacy in vivo after bei ng continually exposed to the immunosuppressive, TGF - β rich microenvironment of cancer patients following adoptive transfer. This will limit the clinical efficacy of such therapeutic endeavors. In this study we explored inhibiting TGF - β signaling as a stra tegy to preserve and/or enhance the cytolytic efficacy of ex vivo expanded, highly activated NK cells in the

TGF - β rich milieu of myeloid leukemia and metastatic colon cancer.

128

MATERIALS AND METHODS

NK cell culture and activation

Procurement of peripheral blood samples from healthy volunteers with written informed consent for research use. The Institutional Review Board (IRB) of University Hospitals

Cleveland Medical Center approved the procurement and use of blood samples for this study. Peripheral blood mononuclear cells (MNCs) were separated into buffy coats following density gradient centrifugation of whole blood over Ficoll - Paque Plus (GE

Healthcare Life Sciences). MNCs were subjected to CD3 depletion followed by CD56 enrichmen t using MACS human CD3 depletion and human NK cell enrichment kits respectively according to the manufacturer’s instructions (Miltenyi Biotech). The CD3 - ,

CD56+ NK cells (>98% purity confirmed by flow cytometry) were either incubated overnight in media sup plemented with IL - 2 (GoldBio) for next day assays or were expanded over 14 days in co - culture with irradiated feeder cells (K562 - mbIL21) and IL - 2

(50U/mL) as described by Somanchi et al [347] . NK cells were subsequently maintained

129

in IL - 2 supplemented media either alone or in comb ination with human TGF - β 1 (Cell

Signaling) at 5 - 10ng/ml for up to 96 hours. The base medium was RPMI 1640 +

Glutamine 2.05mM (HyClone), 10% Cosmic Calf Serum (HyClone) and 1% penicillin -

streptomycin.

Immediately prior to all in vitro assays, NK cells were centrifuged out of cell culture and

re - suspended in fresh media to remove continued IL2, TGF β or LY2157299 exposure of

the NK cells and their co - cultured targets.

Cytotoxicity assay

NK cell cytotoxicity against various cancer cell lines was obtained using a quantitative

flow cytometry assay. Cancer cells used were a myeloid leukemia (HL60), and colon

cancer cell lines (HCT116 and HT29) cultured in the base medium described above and

all were obtained from ATCC. Target cells were labeled with Calcein - A M (Calbiochem)

and then co - incubated with NK cells at varying effector (NK cell) to target ratios. 10,000

target cells were used per triplicate well for all experiments. Four hours post - incubation,

the number of live CAM - labeled target cells per 100uL of c ell suspension was quantified by rapid flow cytometry with an Attune NxT flow cytometer (Thermo Fisher Scientific).

Cytotoxicity results are expressed as the proportion of Calcein - AM (CAM) bright cells.

NK Cell Cytotoxicity = (Number of CAM bright cells in a target cell alone well minus

Number of CAM bright cells in NK + target co - culture) divided by Number of CAM bright cells in a target cell alone well.

NK cell phenotyping

130

In cell culture assays NK cells were phenotyped by flow cytometry for CD16 and

NKG 2D at various time points using anti - CD56 - APC (BD - Biosciences), and anti - CD16 -

PE (BD - Biosciences). Antibody staining of single cell - suspensions was performed by

manufacturer’s instructions/protocol. Stained cells were analyzed by flow cytometry.

ELISA ass ays

Quantification of human IFN - γ, TNF - α, Perforin and Granzyme B was performed using commercially available ELISA kits (R&D signaling). For IFN - γ and TNF - α quantification, cell - free supernatants were collected after 4h of co - incubation at 37°C of

40,000 N K cells and 10,000 colon cancer cells (HT29). For Perforin and Granzyme B quantification, cell - free supernatants were collected after 2h of co - incubation of 106 NK cells and 106 HL60 cells. Results shown are the means of triplicate wells ± standard deviati on measurements. Active human TGF - β levels were measured in the serum of human colon cancer murine xenografts following the manufacturer’s protocol (R&D signaling). To prepare mouse sera for the TGF - β assay, cardiac blood collected immediately following mo use euthanasia was transferred to pre - cooled micro - centrifuge tubes containing EDTA and centrifuged for 30 min at 1,500g at 4°C. The supernatant plasma was collected and stored in micro - centrifuge tubes at - 80°C. Samples were thawed at room temperature on the day of TGF β assay.

Colon cancer liver metastases mouse model and human NK cell adoptive transfer

131

100,000 HCT116 cells were surgically implanted into NSG mouse spleens (following hemi - splenectomy); 4 groups, 3 mice/group. The control group received ta il vein injections of 3% FBS/PBS and twice daily oral gavage of inhibitor carrier. A TGF - β inhibitor only group received tail vein injections of 3% FBS/PBS and twice daily oral gavage of LY2157299 (TGF - β inhibitor) at a dose of 75mg/kg twice daily for 2 we eks. A third group (NK alone) received 5 x 106 NK cells each (tail vein), weekly for two weeks, starting 10 days after hemi - splenectomy. The fourth group (NK + TGF - β inhibitor) received 5 x 106 NK cells each (tail vein), weekly for two weeks, starting 10 d ays post - op and the TGF - β inhibitor LY2157299 by oral gavage twice daily at 75mg/kg for two weeks. Mice receiving NK cells also received IL2 (75,000U IP) three times a week

(MWF) for two weeks. The vehicle for LY2157299 was constituted per manufacturer (El i

Lilly) instructions [30]. The vehicle for mouse injections (cells and IL2) was sterile filtered PBS with 3% calf serum.

This study was carried out in strict accordance with recommendations in the Guide for the Care and Use of Laboratory Animals of the Na tional Institutes of Health. Mouse surgeries were performed under ketamine/xylazine anesthesia, and all efforts were made to minimize suffering in the peri - operative period and during oral gavaging including post - operative analgesia per institutional guide lines. All mice were euthanized at a pre - determined end point 1 week after the second NK cell infusion in treated mice.

Euthanasia was performed using carbon dioxide asphyxiation followed by cervical dislocation. All procedures used in mouse care, surgerie s and euthanasia for the purpose of this study were approved by the Institutional Animal Care and Use Committee of Case

Western Reserve University.

132

Statistical analysis

All continuous measurements (NK cell expression of CD16, NKG2D, and change in

cytotoxi city during in vitro studies were compared using the student T - test between two groups (control NK cells were reference sample). Tumor burden measurements computed from imaging in colon cancer liver metastasis xenograft was compared using ANOVA among group s ( ≥ 3). P - values demonstrated are at significance levels of 0.05 for two - tailed hypotheses. Chart indicators for p - values are * if <0.05; ** if ≤0.01, *** if ≤0.001,

**** if ≤0.0001 and ns if ≥0.05

133

RESULTS

Ex vivo expansion of NK cells leads to increased cytotoxic activity against leukemia

and colon cancer

Ex vivo - expanded NK cells exhibit increased cytotoxic activity against colon cancer and

myeloid leukemia cells. We compared the cytotoxic function of freshly isolated NK cells

which we re overnight activated in IL2 with NK cells from the same donor that were expanded for 2 weeks using irradiated K562mIL21 feeder cells. The expanded NK cells were more efficient at killing both colon cancer and myel oid leukemia cells. For

example, at a 1:2 NK cell to target cell ratio, there was 24% killing of HCT116 cells as

compared to 2% killing with fresh NK cells; p = 0.043 (Fig 5.1 A). Overnight - activated

NK cells required an effector to target ratio of 5:1 to achieve 50% killing while expanded cells a chieved this at 2:1 ratio. In addition to HCT116 cells, expanded NK cells also exhibited high cytotoxicity on HT29 cells (Fig 5.1 B). At a 1:1 ratio using 3 different NK cell donors, the expanded NK cells killed 45 – 86% of HT29 cells (Fig 5.1 B). This cytotox ic effect was significantly better than the overnight activated NK cells which

134 killed 4 – 28% of the HT29 cells. Again, expanded NK cells demonstrated superior killing efficiency at 4:1 ratio, with >90% killing across all donors at a 4:1 ratio as compared to

29% - 40% using fresh NK cells (Fig 5.1 C).

Figure 5 .1. Expanded NK cells demonstrate increased cytotoxicity against HCT116

and HT29 cells as compared to fresh, IL - 2 activated cells. The indicated NK cells were assessed for cytotoxic activity against target cancer cells using a calcein AM flow cytometry assay following 4 hours co - incubation. (A) Percentage cell death of HCT116 cells induced by NK cells at the indicated Effector:Target cell ratios. (B and C)

135

Percentage HT29 cell death induced by NK cell s from 3 individual healthy donors at

Effector:Target ratios of 1:1 (B) and 4:1 (C). Bars represent data using IL2 - activated NK

cells (Fresh) and following expansion of NK cells from the same donor (Expanded).

*p<0.05; **p ≤0.01; ***p≤0.001; ****p ≤0.0001; n s p ≥0.05. [337]

Exposure to pathologic TGF - β levels impairs expanded NK cell function in a time -

dependent manner

At TGF - β levels similar to that found in AML patients (5ng/ml), NK cell killing of the

myeloid leukemia cell line HL60 was progressively impaired at 24 hours (11 – 14%

decrease, p = 0.002), 72 hours (33 – 4 1% decrease, p<0.0001) and 96 hours exposure (70 –

78% decrease, p<0.0001) (Fig 5.2 A). This reduced cytotoxic activity correlated with an

NK cell receptor phenotype consisted with less active NK cells. For example, there was a

65 – 68% decline in NKG2D express ion as early as 24 hours in the NK cells (Fig 5.2 B; p

= 0.005). CD16 expression did not change in the first 24 hours of exposure but decreased

by 56% at 96 hours (Fig 5.2 C; p = 0.037). There was no appreciable difference by

increasing the TGF - β ligand conc entration from 5ng/ml to 10ng/ml (Figs 5.2 B and 5.2 C).

136

Figure 5 .2. Sustained exposure to pathologic levels of TGF - β impairs the function of

highly activated, expanded NK cells. After 2 weeks of expansion, NK cells were

maintained in culture supplemented with 200U/ml IL2 with or without TGF - β at 5ng/ml and 10ng/ml. (A) Change in cytotoxicity of expanded NK cells against HL60 cells (4:1 ratio) following exposure to indicated doses of TGF - β after 24h, 60h and 96h. Change presented is in comparison to NK cel ls maintained without TGF - β. (B and C) Change in activated phenotype following 24h and 96h exposure to the indicated doses of TGF - β as compared to control NK cells. Change in proportion of NK cells expressing NKG2D (B) and CD16 (C) are shown. T: TGF - β liga nd. *p<0.05; **p ≤0.01; ****p≤0.0001; ns p ≥0.05. [337]

137

TGF - β i nhibition maintains the function of expanded NK cells despite sustained

exposure to pathologic TGF - β levels

We assessed the phenotype and cytotoxic function of ex vivo expanded NK cells treated

with TGF - β alone or in combination with LY2157299, a clinicall y used oral small

molecule kinase inhibitor of TGF - β receptor 1. Despite continued exposure to TGF - β, the

‘activated’ NK cell phenotype consisting of NKG2D+CD16 bright NK cells was

preserved by the addition of LY2157299 (Fig 5.3 A - 5.3 D). NKG2D expression was decreased by approximately 53% at 24 hours and 61% - 72% by 72 hours with TGFβ alone; p = 0.025 and 0.035 respectively (Fig 5.3 A and 5.3 B). This change was

significantly ameliorated at 24 and 72 hours after treatment with the additio n of

LY2157299 (p = 0.029 and 0.012 respectively). A change in CD16 expression was not

evident at 24 hours (Fig 5.3 C) but was decreased by 11% - 43% by 72 hours (Fig 5.3 D); p

= 0.071. These changes were again prevented by the addition of LY2157299.

We also m easured killing of colon cancer and leukemia cell lines by the ex vivo

expanded NK cells in the presence of TGF - β and/or LY2157299. At a NK to target ratio

of 1:1, TGF - β exposure resulted in a 54% - 73% decrease in killing of HT29 cells (Fig

5.3 E) after 24 h ours and 9 – 77% after 72 hours (Fig 5.3 F) The addition of LY2157299

preserved the cytotoxic function of NK cells in the presence of TGF - β ligand, and a modest gain of function in cells from one donor. Changes in NK cell cytotoxicty after 24 hours exposure t o TGFβ were more modest against HCT116; - 9%, - 53%; and - 13% in individual donors (Fig 5.3 G). After 72 hours TGF - β exposure, a more marked decrease in cytotoxic activity against HCT116 was noted; - 67%, - 13% and - 29% (Fig 5.3 H).

138

Against HL - 60 cells, there wa s a 54% - 64% decrease in NK cell mediated killing after 72 hours with TGF - β exposure at a 1:1 NK cell to target ratio (Fig 5.3 I). Increasing the NK

cell ratio to 4:1 was still associated with roughly a 40% decrease in cytotoxic function.

Again, the addition of the TGF - β inhibitor LY2157299 preserved the high cytotoxic

function of these NK cells.

139

’

140

Figure 5 .3. Inhibiting TGF β signaling using the small molecule kinase inhibitor

LY2157299 preserves the cytotoxic function of expanded NK cells, even after sustained exposure to pathologic levels of TGF - β. After 2 weeks of expansion, NK cells were maintained in culture supplemented with 200U/ml IL2 alone or in combination with TGF - β1 (5ng/ml) and/or LY2157299 (5uM) for the indicated times and tested by flow cytometry and cytotoxic assays. (A, B, C and D) Mean fluorescence intensity (MFI) of NKG2D after 24h (A) an d 72h (B); and for CD16 after 24h (C) and 72h (D). (E, F, G and H) Change in cytotoxicity compared to control NK cells (not treated with TGF - β or

LY2167299) against HT29 cells at 24h (E) and 72h (F); and against HCT116 at 24h (G) and 72h (H). Individual re sults from three different donors using NK to target ratios of

1:1 are presented in each figure. (I) Similar 72 hour results using NK cells obtained from two donors is shown against HL60 cells using NK - to - HL60 ratios of 1:1 and 4:1. T:

TGF - β; G: LY2157299. *p<0.05; **p ≤0.01; ***p≤0.001; ****p≤0.0001; ns p≥0.05.

[337]

The production of TNF - α and IFN - γ by these otherwise highly activated NK cells was also significantly impaired following TGF - β exposure. During co - culture with HT29 cells, NK cells previously incubated with TGF - β released 40% less TNF - α; p = 0.0007

(Fig 5.4 A) and 19% less IFN - γ than control NK cells; p = 0.069 (Fig 5.4 B). Compared with control NK cells, TGF - β treated NK cells maintained in culture with LY2157299 showed a slight trend to increased production of TNF - α (1.25 - fold, p = 0.05) and IFN - γ

(1.06 - f old, p = 0.4). In addition to functional cytotoxicity assays, we measured the release of Perforin and Granzyme B by NK cells co - cultured with a leukemia target cell line

141

(HL60). We again noted a decrease in the levels of both perforin (by 57%; p<0.00001) a nd granzyme B (by 38%; p<0.0001) by NK cells exposed to TGF β ligand (Fig 5.4 C and

5.4 D).

Figure 5 .4. TGF - β signaling impairs the production of TNF - α, IFN - γ, Perforin and

Granzyme B by ex vivo expanded NK cells. Functional assay of expanded NK cells

a fter 72hr - exposure to 5ng/ml TGF - β ligand by measuring release of TNF - α and IFN - γ

into the supernatant at the end of a 4hr cytotoxicity assay. ELISA quantification of TNF -

α (A) and IFN - γ (B) in the supernatant of 40,000 NK cells co - cultured with 10,000 HT2 9

cells/well. ELISA quantification of perforin (C) and granzyme B (D) release after 2 hours

of co - culturing 106 NK cells with 106 HL60 cells are presented. All assays presented are

142

results of triplicates. T: TGF - β; G: LY2157299. ***p<0.001; **** p≤0.0001; ns p ≥0.05.

[337]

TGF - β inhibition results in superior tumor eradication by highly activated NK cells

in a metastatic colon cancer mouse model

We examined the therapeutic efficacy of ex vivo expanded NK cells in an immunodeficient mouse (NSG) model of colon cancer liver metastasis using HCT116 cells. Four weeks aft er splenic implantation of HCT116 cells all mice were euthanized.

At autopsy, control and LY2157299 alone mice had hemorrhagic ascites and grossly enlarged livers, riddled with metastatic deposits without evidence of normal liver tissue

(Fig 5.5 A); panel 1 and 2. Mice treated with NK cells alone also had gross evidence of liver metastases however nodule burden was less and there was some morphologic evidence of normal liver architecture (Fig 5.5 A); panel 3. Mice that received NK cells and LY2157299 had most ly normal liver morphology with rare gross nodules (Fig 5.5 A);

panel 4. H&E stains of liver sections for all 3 mice per group were consistent with gross

anatomy (Fig 5.5 B and 5.5 C). Quantification of the tumor burden showed a 25% decrease

in mice treated w ith NK cells alone (p<0.05) and by approximately 90% in those treated

with NK cel ls and LY2157299; p<0.001 (Fig 5 .5C). Of note, circulating human TGF - β

levels (active) in the mice of all treatment groups was an average of 10.98pg/ml +

1.736pg/ml. These lev els are comparable to serum levels of active TGF - β in colorectal

cancer patients [355] .

143

144

Figure 5 .5. TGF - β inhibition enhances activated NK cell function in a colon cancer model of liver metastases. A liver metastases model of colon cancer xenograft using

HCT116 cells was established. All mice were autopsied at 32 days after cell injection and

H&E sec tions were obtained. Tumor burden on H&E sections was measured for each mouse liver. (A) Autopsy findings of gross liver morphology in one representative mouse from each of the four treatment groups. (B) Representative light microscopy of H&E stained liver sections. (C) Average tumor burden in H&E stained liver sections (N = 3 mice/group) was quantified using the VENTANA digital Image viewing software. * p

<0.05; *** p ≤0.001. [337]

TGF - β inhibition enhances NK cell infiltration into liver tissue in colon cancer metastasis model

In our metastatic colon cancer xenogra ft, we examined FFPE liver sections for NK cell infiltration using immunohistochemistry for human specific CD45 antibody. Mice that received NK cells in addition to LY2157299 had an average of 82 infiltrating NK cells per 10X field compared to an average o f 8 cells in similar fields in mice that did not receive TGF - β inhibition; p = 0.0036 (Fig 5 .6). Of note, mice were necropsied 14 days after the last of two NK cell infusions and 9 days after the last dose of LY2157299.

Pictures of represent ative sections are provided in Fig 5.7 .

145

Figure 5 .6. Robust NK cell infiltration into liver tissue observed in mice who received TGF - β inhibition in addition to NK cell infusion. FFPE sections of liver tissue from mice with metastatic colon cancer were subjected to imm unohistochemistry with human specific CD45. Manual counts of CD45 positive cells (NK cells) was done in four representative 10X fields using the VENTANA digital image viewing software.

Mice that received NK cells alone (NK only) and NK cells with LY2157299 (NK+G).

[337]

146

Figure 5.7 . Representative images of colon cancer metastasis xenograft showing

CD45 IHC of liver FFPE sections staining for human NK cells. 10X view of slide

sections; slide capture using VENTANA digital imaging software. (A) Vehicle only

group; all sections showing tumor infiltration; no CD45+ cells observed by IHC. (B)

LY2157299 (Galunisertib) only group; all sections showing tumor infiltrat ion; no CD45+

cells observed by IHC. (C) NK cell only group showing section of tumor involvement

with infiltrating NK cells (CD45+ brown cells). (D) NK Cells plus LY2157299 group

showing section with rare residual tumor showing surrounding infiltrating NK cells

(CD45+ brown cells) in normal liver tissue. [337]

147

DISCUS SIONS

Despite the promise of NK cell therapy, the approach has been limited due to several

factors including challenges in generating sufficient levels of NK cells for infusion as

well as difficulties in maintaining a high level of activity in vivo of the infused cells. The

development of robust feeder cell lines for ex vivo NK cell expansion has enabled the

production of massive doses of NK cells [345 - 347] . These strategies include the use of antigen presenting cells as stimuli for NK cell activation and subsequent proliferation. As an example, the K562 - mbIL21 cells used in our study are K562 cells transduced to express membrane - bound interleukin 21 and 4 - 1BB ligan d resulting in up to 21,000 - fold proliferation of NK cells after three weeks co - culture [347] . Besides a high level of expansion, an added advantage of ex vivo expansion is that the resulting NK cells are also significantly activated compared to resting NK cells and are more effici ent at killing cancer targets. Consequently, it becomes even more pertinent to ensure that these highly activated killer cells remain protected from the immunosuppressive tumor microenvironment after infusion. Studies have shown that the tumor microenviron ment produces high levels of factors such as IL - 10 and TGFβ leading to NK cell dysfunction

[211, 348 - 351, 360, 363] .

In this report we show that TGF - β inhibition with a clinically used agent, LY2157299, can effectively mitigate TGF - β - mediated NK cell dysfunction using in vitro and in vivo model systems of leukemia and colon cancer. We have demonstrated that the marked impairment in function of highly activated NK cells starting from 24 - hour exposure to a

TGF - β rich milieu correlates with a significant decrease in expression of the activating receptors NKG2D and CD16. Inhibiting TGF - β signaling in this setting maintained the

148 activated phenotype of these NK cells and resulted in more effective NK cell - mediated eradication of colon cancer liver metastases in a mouse model.

LY2157299 and TEW - 7197 are the only two small molecule inhibitors of TGF - β receptor

1 currently in clinical trials for various malignancies, particularly advanced stage, solid tumors like pancr eatic, prostate and hepatocellular carcinoma [364 - 366] . These trials are either ongoing or recently completed with results not yet published. For our current research we chose the compound LY2157299 as it is further along in clinical development and the safety/pharmacokinetic profile has been well characterized in the aforementioned cancer trials as well as in healthy individuals (NCT01965808). There are now clinical trials aimed at enhancing the effector T cell response from checkpoint blockade via TGF - β inhibition by combining LY2157299 with Nivolumab in vario us advanced solid tumors (NCT02423343) and with Durvalumab in pancreatic cancer

(NCT02734160). There are also ongoing trials of adoptively transferred NK cells predominantly for hematologic malignancies without complementary TGF - β inhibition.

From our curr ent study, we propose complementing adoptive transfer of NK cells with such TGF - β inhibiting agents to maintain these cells in their highly activated state and ensure better clinical efficacy. We have focused now on colon cancer and leukemia models; in fut ure studies we will assess the general applicability of this strategy for other tumor types.

149

CHAPTER 6: DISCUSSION AND FUTURE DIRECTIONS

150

6.1: CONCLUSION AND DISCUSSION

NK cell therapy holds great promise in being an effective anti - cancer agent. Although

highly cytotoxic, NK cells do not attack normal cells and have not been reported to cause

GVHD. In addition, it has been shown to efficiently clear tumors in various muri ne

human cancer models. There have been more than 100 NK cell clinical trials, and more than 250 trials ongoing or that have not yet started. There is much interest in NK cells and developing them as potent tools for fighting cancer. Despite the high volum e of

interest in NK cell therapy, two main challenges prevent the translation of successes

observed in preclinical studies to studies in humans. Low NK cell infusions and

immunosuppressive factors in the tumor microenvironment have been formidable

obstacle s to the maximization of NK cell effector activities. In fact, suppressive

molecules such as TGF - β impair the activities of other immune cells, such as T cells and

macrophages, that are critical for effective immune responses [367] . In addition to

soluble factors, suppressive cells such as MDSCs also imp air NK cell function.

Therefore, dampening the effects of these suppressive factors, especially during c ancer

treatment, is critical for the success of not just NK cell therapy but the entire cell therapy

field. Various strategies have been developed towards this aim. For example, inhibition

of TGF - β signaling shows great promise in mitigating TGF - β (Fig 5.5 ) [368] .

We developed a novel NK cell expansion system consisting of myeloid leukemia cell line

expressing membrane - bound IL - 21 (NKF cells) to address the inability to expand NK cells ex vivo for adoptive cell therapy . We proved that the expansion system is able to generate pure and clinically efficacious NK cell doses; up to 20,000 - fold expansion after

5 weeks. Phenotypically, NKF - expanded NK (NKF - NK) cells express higher levels of

151

the activating receptors NKG2D and N Kp30 that are critical for NK cell - mediated

antitumor activities. Importantly, NKF - NK cells mount effective cytotoxicity against a broad spectrum of tumor types including breast cancer and colon cancer cells. We also demonstrate increased glycolytic and ox idative phosphorylation rates in NKF - NK cells as compared to unexpanded or non - IL - 21 expanded NK cells. Increased glycolysis is critical for immune cell activation while also promoting exhaustion. Balance of glycolysis and oxidative phosphorylation may pro mote survival of NKF - NK cells. Adoptive transfer of

NKF - NK cells resulted in marked reduction in sarcoma tumor burden as compared to control mice. In addition, NKF - NK cells demonstrated prolonged survival of mice infected with T cell leukemia cells. This d ata demonstrates that NKF - NK cells, in combination with other immunotherapies such as checkpoint blockers, may lead to enhanced tumor rejection and better prognosis.

We attemp ted to address the challenge that immunosuppressive factors in the tumor microen vironment pose to NK cell therapy by targeting TGF - beta. As discussed in earlier sections, TGF - beta is a major mechanism of tumor - mediated immune suppression; therefore its inhibition could rescue NK cell activity. Utilizing a clinical TGF - betaR inhibitor, galunisertib, we were able to demonstrate that combining NK cell therapy with

TGF - beta signaling inhibition results in near - normal NK cell cytotoxic activity and cytokine secretion . In addition, we showed that rescued NK cell activity may be mediated by u pregulation of NK cell activating receptors, NKG2D and CD16. A liver metastasis murine model provided us with the tool to test the ability of NK cells armed with galunisertib to limit metastasis of colon cancer cells. We demonstrated that indeed inhibiting TGF - beta signaling in adoptively transferred NK cells resulted in reduced

152

primary tumor and metastatic disease. The positive preclinical studies are encouraging, and require evaluation through clinical trials to determine the impact TGF - beta signaling inh ibition will have on NK cell therapy to improve patient outcomes.

The unimpressive clinical results from various trials with monotherapies has indicated that combining various therapies will result in better patient outcomes. Also, it is clear that other types of immunotherapies might be effective in rejecting tumors, and will be the subject of research for years to come. The studies proposed below will not only aid in better understanding the biology of NK cells but also may prove to be effective therapie s against cancer.

153

6.2: FUTURE DIRECTIONS

Galunisertib and TIGIT inhibitor

Receptors for TGF - β are constitutively expressed on NK cells, and serve as a main form

of immunosuppression driven by tumor cell - mediated TGF - β secretion. Hence, inhibition of harmful TGF - β signaling in NK cells and other immune effector cells will signifi cantly

help to re - activate the cytotoxicity of adoptively transferred NK cell. Indeed, we and

others have shown that TGF - β signaling inhibition rescues NK cell cytotoxic function

against a variety of tumors in vitro and in vivo [55, 163, 337, 361, 369, 370] . TGF - β

causes downregulation of NK activating receptor and upregulation of inhibitory

receptors. A recently discovered inhibitory receptor, T cell immunoglobulin and ITIM

(TIGIT), may be one of the targets of TGF - β signaling.

TIGIT is a checkpoint receptor mainly expressed on NK and T cells [371] . The ligand for

TIGIT, poliovirus receptor (PVR), is mainly expressed on APCs, and TIGIT has been

indicated in a variety of diseases including lupus and other immunologic diseases [372] .

NK cells from healthy subjects that ex press low levels of TIGIT have been reported to

demonstrate increased cytokine secretion and degranulation capability [371] .

Interestingly, among healthy individuals there is a broad heterogeneity in TIGIT

expression on NK cells, which might explain the variance in susceptibilit y to various

infectious agents and cancer types. TIGIT levels are low in naïve, resting T cells and

increase with antigenic or inflammatory stimulation [373] . Although well - studied for its

role in T cell biology, very little is known on how TIGIT impairs NK cell activity. It has

been postulated that TIGIT might induce inhibitory signaling in NK cells through its

ITIM do main. NK cells isolated from TIGIT - deficient mice produced more IFN - γ when

154

exposed to PVR - expressing target cells as compared to wild - type cells [373] . This study

aims to understand how TGF - β affects TIGIT expression to aid in the development of

more efficacious immunotherapies.

Foremost we will verify that similar to T cells, TIGIT expression is induced upon NK cell

activation. Using same donor comparisons, TIGIT expression on freshly isolated (naïve ,

resting) and NKF - activated and expanded NK cells will be measured by flow cytometry.

Next, we will assess the regulatory role of TGF - β on TIGIT expression in na ïve and activated NK cells. We will treat freshly isolated (naïve, resting) NK cells with and without TGF - β for 72 hours, and TIGIT expression will be measured via flow cytometry.

Transcriptional regulation of TIGIT by TGF - β signaling will be determined by measuring the changes in mRNA levels of TIGIT in untreated versus TGF - β - treated naïve NK cell s.

We expect TIGIT expression to be higher in NKF - NK cells as compared to naïve NK cells, in correlation with what was reported in T cells [373] . Also, we expect TGF - β to induce upregulation of TIGIT in naïve cells, as observed for other NK cell inhibitory receptors. Parallel experiments will be performed using NKF - NK cells to determine the effect of TGF - β treatment on TIGIT expression. We expect TIGIT expression to significantly increase in NKF - NK ce lls following TGF - β treatment as compared to untreated NK cells. The difference in upregulation may be higher in NKF - NK cells as compared to naïve NK cells because TIGIT expression increases with activation and

TGF - β should further heighten the increase. O nce TGF - β regulation of TIGIT has been established, NK functional assays will be used to correlate TGF - β - mediated upregulation of TIGIT to changes in NK cell activity.

1 55

NK cells facilitate innate immunity by secreting cytokines and releasing cytotoxic

gran ules that result in target cell lysis. NKF - NK cells will be treated with or without

TGF - β and/or TIGIT inhibitor for 72 hours. In addition, two group of NKF - NK cells will

be treated with or without TGF - β and galunisertib. TIGIT - mediated impairment of NK ce ll activity was demonstrated using B cell lymphoma cell lines [374] . The different

groups of NKF - NK cells will be co - cultured wi th calcein - AM - labeled B cell lymphoma

cell line for 4 hours and cytotoxic activity of the different NKF - NK cell groups will be measured via flow cytometry. We expect TGF - β - treated NKF - NK cells to result in a dramatic decrease in cytotoxic function as compa red to untreated NKF - NK cells. NKF -

NK cells treated with TGF - β and the TIGIT inhibitor may rescue NK cell cytotoxic function to an extent similar to galunisertib - treated NKF - NK cell groups. Supernatants from the co - culture can be assessed for levels of IFN - γ and TNF - α. We expect the cytokine levels to correlate with the cytotoxic activity of the different NKF - NK groups.

Next, we propose experiments to determine the role that the SMAD pathway may play in mediating TGF - β - induced TIGIT expression. PD169316, a potent suppressor of TGF - β signaling, inhibits SMAD phosphorylation to hamper TGF - β - mediated effects. NKF - NK cells will be treated with TGF - β with or without PD169316, an inhibitor of SMAD2/3 phosphorylation for 72 hours. Prior to addition of TGF - β, NKF - N K cells will be pre - treated with PD169316, and the other NKF - NK cell groups will receive an equal volume of DMSO. Post - incubation, TIGIT expression on the treated NKF - NK cells will be assessed via flow cytometry. We expect the NKF - NK cells cultured with TG F - β to demonstrate increased TIGIT expression as compared to untreated NKF - NK cells.

Conversely, the NKF - NK cells treated with both TGF - β and PD169316 will have

156

decreased expression of TIGIT. On the other hand, it is possible for the inverse results to

occ ur. TIGIT expression increases with T cell activation while TGF - β is known to

suppress activation. Therefore, it is possible that TGF - β - induced immunosuppression may also prevent activation - induced TIGIT expression.

Following in vitro experiments to deter mine how TGF - β - induced TIGIT expression affects NK cell activity, we propose in vivo experiments to determine how inhibition of

TIGIT on NK cells can enhance NK cell antitumor efforts. NSG mice will be infected subcutaneously with a B cell lymphoma cell li ne known to express TGF - β and expressing luciferase firefly gene. The mice will be divided into 6 groups and treated weekly with

IL - 2 thus: the first group will be NSG mice gavaged with vehicle (Veh); the second, treated with galunisertib (Gal); the third, injected with NKF - NK cells (NKF - NK); the fourth, treated with adoptively transferred NKF - NK cells and galunisertib (NKF - NK -

Gal); the fifth, treated with TIGIT inhibitor (TIG); the sixth, NKF - NK cells and TIGIT inhibitor (NKF - NK - TIG). The mice will be imag ed weekly for 5 weeks. Mice will be sacrificed once moribund or tumor score is ≥ 7 and the data noted to generate a survival plot. Mice in the Veh, Gal, and TIG groups are expected to develop overt disease because the mice were not given any NKF - NK cells t o mediate tumor rejection. The mice in the

NKF - NK group are expected to develop moderate disease as compared to the rest of the groups. NKF - NK - Gal and NKF - NK - TIG groups are expected to demonstrate minimal disease due to the disinhibition of adoptively tran sferred NK cells. The extent of disease will depend on aggressiveness of the cell line used for the study, and should be considered when choosing a target cell line.

Study Limitations

157

An important consideration to maintain the relevance of this study is to ensure that a

valid TGF - β concentration is used to detect changes in TIGIT expression. This concern

can be addressed by measuring changes in TIGIT expression across a spectrum of TGF - β

concentrations. Secondly, although the 72 - hour time - point chosen for the study is based on the minimal amount of time previously observed to result in profound TGF - β - mediated activities, TGF - β regulation of TIGIT expression might be faster. Therefore, a study investigating TGF - β - mediated TIGIT expression across a number of time periods might be necessary. It is possible for the study to fail to reveal a link between TGF - β

signaling and TIGIT expression, although this is unlikely due to the pleiotropic eff ects of

TGF - β. Importantly, this study does not elucidate the role that non - canonical TGF - β

signaling pathways may play on TIGIT expression, therefore sweeping conclusions are

difficult to reach with results from this study. Regulation of TIGIT expression could be

due to post - transcriptional modifications such as glycosylation, and would require

evaluation using post - transcriptional regulation assays. Donor variability could also affect

the analysis and conclusions that can be made from this study. This is addressed by

performing paired analysis of naïve versus expanded NK cells from the same donor,

serving as a normalization technique that accounts for variance in baseline levels of

TIGIT.

Assessing the effectiveness of combined TGF - β inhibitor and TIGIT inhibitor

The modest effects observed from trials with various immunotherapies suggests that

blockade of multiple sources of immunosuppression may be necessary to obtain positive

clinical outcomes in a wide spectrum of cancer patients. We propose a novel s tudy

158 exploring the effectiveness of the combination of NK adoptive cell therapy with TGF - β and TIGIT inhibition.

The NK cell promoting activity of TGF - β inhibition was demonstrated in a phase 2 trial of galunisertib against advanced hepatocellular carcinom a that resulted in median overall survival of 21.8 months for patients with >20% reduction in TGFbeta1 levels, as compared to 7.91 months for patients with <20% reduction [375] . This data indicates that reduction in TGF - β signaling leads to better patient outcomes. Grade 3 and 4 adverse ev ents were reported, therefore future studies should titrate the galunisertib dosage for most effective dose while limiting the side effects. Other studies have shown that reduction in TGF - β signaling in NK cells does not affect its development or homeostas is, indicating that TGF - β is a safe target for immunotherapy [369] . As discussed in the previous study, TGF - β is a potent immunosuppressor, impairing not only T cell activity but NK cell effector f unctions as well. Although a clinical trial combining OMP - 313M32, a TIGIT inhibitor, and nivolumab is ongoing, there are no studies combining TGF - β inhibition with TIGIT inhibition. We propose a study assessing the efficacy of combining

NK cell therapy wit h galunisertib and OMP - 313M32.

Combination of multiple drugs requires careful titration of both drugs to reduce drug - related toxicity. For example, reports from clinical trials combining nivolumab and ipilimumab has indicated that treating patients with a combination of the standard doses

(3mg/kg) of both drugs led to increased rates of grade 3 and 4 toxicities as compared to treatment at the standard dose for each drug only [376] . Hence, there is a need for flexibility in determining the optimal dose and schedule of immunotherapeutic combinations.

159

Although concentrations of galunisertib and OMP - 313M32 used in various studies have been reported, it is important to verify the safeness of those doses, especially in preclinical studies. This is especially important because rodent studies have been shown to predict human MTDs very well [377] . The most commonly used endpoint for determining toxicity is weight loss, followed by clinical signs. Hence for this study we will use weight loss and tumor score as our endpoints. NSG mi ce strain is free of confounding variables such as other immune cell types, and therefore will be used to test the effectiveness of galunisertib and OMP - 313M32 on NK cell therapy. Importantly, using NSG mice allows us to test the effectiveness of human cel ls, making our study more translational.

NSG mice will be divided into 3 groups: the first group of mice will be treated with galunisertib (Gal), the second, OMP - 313M32 (OMP), and the third, both Gal and OMP

(Gal - OMP). Starting with previously published doses reported safe for administration, that dose will be reduced by 50%, serving as the starting dose. Doses will be escalated incrementally by 50% of the original dose until any mouse reaches one of the primary endpoint of >15% weight loss or tumor score ≥ 7. The MTD is set as the dose prior to the endpoints being met, as described by Aston et al. [377] . Once the MTD has been determined, the mice will be sacrificed and the organs grossly examined for any pathologies associated with the drugs.

After the MTD for each drug and drug combination has been determined, the effect of the drugs on adoptive NK cells will be investigated. NSG mice will be injected subcutaneous ly, at the flank, with luciferase - expressing B cell lymphoma cell line, verified to produce TGF - β. The mice will be treated with vehicle (Veh), Gal and NKF -

160

NK cells (Gal), OMP and NKF - NK cells (OMP), Gal and OMP (Gal - OMP) or Gal,

OMP, and NKF - NK cells (Gal - OMP - NK) weekly along with IL - 2. Gal and OMP will be given twice daily, IL - 2 will be administered every week, and NKF - NK cells will be given weekly. The mice will be imaged weekly using a bioluminescence imaging (BLI) aparatus to measure tumor burden. The study will end and the mice sacrificed once the mice in the Veh group become moribund or have a tumor score ≥ 7. Bone marrow, lung, spleen, liver, and primary tumor will be extracted from the mice for immunohistochemical analysis to detect NKF - NK cells. We expect this study to demonstrate a higher proportion of NKF - NK cells in the tumor tissue in the NK - Gal -

OMP group, as compared to the other groups. From our previous study, we reported a significant reduction in tumor burden in mice that were treated with Gal and NK cells for

3 weeks [337] . It would be interesting to know whether OMP can demonstrate a similar potent disinhibition of NK cells. We expect to observe an insignificant reduction in tumor burden in mice in the Gal - OMP group since they lack human NK cells to effectively control tumor growth.

Study Limitation s

NK cells were not included in the determination of the MTD for the drugs and the combination, creating the possibility that the MTD may be too low or too high. NK cell infusions, both in murine and human studies, have been shown to be safe, hence it was no t included in the MTD experiments. Another limitation of this study is the concern that the MTD calculated for NSG mice may not be relevant in other mice strains. Aston and colleagues demonstrated that there was variability in the MTDs between mice strains for commonly used chemotherapeutics [377] . Thus, before the results from this study are

161 used in designing experiments using other mouse strains, the MTDs for Gal a nd OMP must be determined for the particular strain to avoid wasting mice.

Galunisertib and Treg depletion in Balb/c mice

As discussed above, TGF - β is a prominent immunosuppressive factor. Therefore, inhibition of TGF - β signaling in NK cells will enhance NK cell function. Galunisertib is a TGF - βRI inhibitor that is being clinically developed and tested against a variety of hematologic and non - hematologic malignancies. As a monotherapy, galunisertib has not shown much promise, however its safety profile po sitions it as a good candidate for combination therapy. In addition to TGF - β, regulatory T cells are another source of immunosuppression in NK cells. Tregs are known to potently suppress T and NK cell function. Therefore, TGF - β signaling blockade and Treg depletion might lead to disinhibition of NK cells and consequently tumor rejection. Denileukin diftitox (Ontak) is an antineoplastic agent FDA - approved for the treatment of patients with cutaneous T cell lymphoma. Ontak has also been discovered to deplete Tregs because Ontak targets the

CD25 component of the IL - 2R [378] . Upon binding, Ontak, is internalized by receptor - mediated endocytosis. Ontak is then cleaved, releasing its e nzymatic domain to inhibit protein synthesis.

Foremost, the MTDs for Ontak and galunisertib should be determined for Balb/c mice, as described in the previous study. Balb/c mice will be simultaneously infected with the murine colon carcinoma cell line, C T26, expressing the luciferase protein. Once disease has been established, the mice will be given vehicle (Control) or galunisertib (Gal) by

162

oral gavage daily, Ontak by weekly IV injections (Ontak), or a combination of daily

galunisertib gavage and weekly Ontak injections (Gal - Ontak). The Control mice will also be given weekly IV injections of saline. Tumor burden will be measured via in vivo BLI and compared for the 4 groups of mice to determine how well the combination of galunisertib and ontak is able to control tumor growth. The study will be stopped once the

Control mice begin to become moribund or have a tumor score ≥ 7. The lungs and liver of the mice will be grossly examined to determine the effectiveness of the various treatments on limiting tumor m etastases. The harvested tissue will also be processed and assessed via flow cytometry for tumor - infiltrating NK cells and Tregs. We predict that the Gal - Ontak group will demonstrate the least tumor burden as compared to the other groups and the least meta static burden. Also, the mice in the Gal - Ontak group should have the highest numbers of tumor - infiltrating NK cells per mL and least number of

Tregs.

Study Limitations

The study design relies on the previously published reports on the efficacy of Ontak at depleting Tregs. This can be addressed by first extracting Tregs and NK cells from healthy Balb/c mice and treating them with Ontak in vitro. The viability of the cells following treatment should confirm Treg specific activity of Ontak. Secondly, this stu dy is performed with a murine cancer model and may not be entirely translatable into human studies. Both Ontak and galunisertib have been used in human trials against a variety of malignancies, therefore we have reasons to assume that the heightened effica cy of the combined drug treatment will be translatable in the human setting. Thirdly, the study assumes the mice have an effective immune system and does not establish the cytotoxic

163 activity of the NK cells or other immune effector cells in the mice. This issue can be addressed by performing an in vitro assay in which NK cells isolated from the mice are co - cultured with CT26 cells and assessed for cytotoxic function.

Improving NK cell function and homing with CAR

Chimeric antigen receptors (CARs) have armed immune cells, especially T cells, with the ability to exact potent cytotoxic functions against B cell malignancies. Although not as investigated as CART cells, NK - CARs can rapidly respond to transformed and stre ssed cells and therefore show great promise in tumor rejection. NK - CARs have shown promise against B cell malignancies. A CD19/CD3ζ - based CAR was designed for NK - 92 cells for enhanced activity against B cell leukemia and lymphoma cell lines. NK - 92 - CAR cell s demonstrated enhanced cytotoxicity against Raji cells in vitro and in vivo [379] .

Despite success of NK cell - based therapies in the preclinical phase, clinical phase results have been underwhelming due to immunosuppressive factors inhibiting NK cell activation and function. In addition to immunosuppressive factors in the tumor microenvironment, activated NK cells have been reported to have decreased expression of chemokine recepto rs [380] . Chemokine receptors are critical for facilitating NK cell migration to the tumor site, therefore reversal of chemokine receptor downregulation could increase the nu mber of tumor infiltrating NK cells. CCR7 is a chemokine receptor recently observed to improve NK cell migration in vivo. CCR7 is a chemokine receptor for CCL19 and CCL21, observed to increase NK cell homing to the lymph node by 144% in athymic nude mice [380] . Despite the reported strong induction of NK cells to migrate

164

to lymphoid sites, there is nothing known about the role CCR7 plays in cancer, and

whether it promotes NK c ell migration to tumor sites.

We propose engineering NK cells with a bispecific CAR that targets CD19 and contain

CCR7, and a CD3ζ signaling domain to promote tumor site homing and enhanced

cytotoxicity against B cell malignancies. By lentiviral gene tran sfer, NKF - expanded and activated NK cells can be infected with this CAR. The stably transduced NK cells will be selected for and validated via rtPCR. The bifunctional NK - CAR (NK - BiCAR) can be assessed for ability to lyse various B cell malignancies. Using CD19 - positive targets that express CCL19 and/or CCL21, such as Raji cells, the cytotoxic function and cytokine secretion function of NK - BiCAR will be assessed via an established fluorescence - based cytotoxicity assay and measured via flow cytometry. An empt y construct lacking the

CD19 receptor and CCR7 genes will be used as a negative control (NK - BiNull) to ascertain the potency of NK - BiCAR cells. The specificity of NK - BiCAR for CD19+ cells will be assessed by co - culturing NK - BiCAR cells with a non - B cell ly mphoma cell line such as OCI - AML3, using the previously described cytotoxicity assay. We expect higher killing proportion of B cell lymphoma cell lines by NK - BiCAR cells, as compared to

NK - BiNull cells. In addition, we expect that NK - BiCAR cells will lyse non - B cell lymphoma cells since these cells will express NK activating ligands that can activate NK activating receptors expressed by NK - BiCAR cells. However, we expect that NK - BiCAR cells will more effectively lyse B cell lymphoma cells as compared to non - B cell lymphoma cells. The increased homing efficiency of NK - BiCAR cells as compared to

NK - BiNull cells will be determined using a transwell migration assay that measures the proportion of cells that respond to CCL19 or CCL21. We expect the NK - BiCAR cells to

165

significantly migrate towards the chamber containing CCL19 or CCL21 as compared to

NK - BiNull cells.

Using NSG mice infected with a luciferase - expressing B cell lymphoma subcutaneously,

the in vivo effect of CFSE - labeled NK - BiCAR cells can be assessed a gainst CFSE -

labeled NK - BiNull as measured by an animal imager. The mice will be injected with NK -

BiCAR cells (NK - BiCAR group), NK - BiNull cells (NK - BiNull group), or no cells

(Control group) weekly for approximately 4 weeks, based on previous NK cell based murine experiments. IL - 2 will be administered to the mice to maintain NK - BiCAR and

NK - BiNull cells’ cytotoxic function. The persistence of NK - BiCAR and NK - BiNull cells will be determined by collecting blood from mice every two days and measuring the propor tion of cells that are CFSE - positive. In addition, tumor burden will be measured with a bioluminescence imager weekly to assess for the effect of tumor - infiltrating NK -

CAR cells in promoting tumor rejection. If mice in the NK - BiNull group fail to reduce tu mor burden, then all the mice in the study will be sacrificed once the tumor burden is ≥

1000mm 3 or the tumor score is ≥ 7. The following sites will be extracted from the mice to determine the proportion of NK - BiCAR or NK - BiNull cells that have homed there in as compared to untreated control mice: lymph nodes, spleen, liver, lungs, and bone marrow.

The primary tumor site will be extracted and stained for the presence of CFSE - positive cells to measure the amount of tumor - infiltrating NK - CAR cells. We expect t hat the NK -

BiCAR - treated mice will have significantly less tumor burden and higher tumor -

infiltrating NK - CARs as compared to the NK - BiNull - treated mice and the control mice.

Also, we expect the NK - BiCAR - treated mice to have a higher percentage of cells hom ing to the lymph node as compared to NK - BiNull - treated mice or control mice.

166

Study Limitations

The limitations of this study include the potential safety concern associated with

hyperactivated NK - BiCAR cells. NK - BiCAR cells are primary NK cells, and theref ore

have intact inhibitory signaling system that prevents NK cells from targeting self. Off -

target effects of NK - BiCAR cells will be tested by co - culturing NK - BiCAR cells with

PBMCs, and measuring the killing activity of NK - BiCAR cells. In addition, primar y NK

cells are not very durable and are expected to persist for no longer than 2 weeks in vivo.

Therefore, any harmful effects of NK - BiCAR cells is limited.

Comparing the effectiveness of NK - CAR cells expressing CCR7 versus NK - CAR cells

expressing CXCR3

Similar to CCR7, CXCR3 has also been recognized as an effective inducer of NK cell

migration. CXCR3 mediates murine leukocyte adhesion and trafficking, and has been

reported to rely on CXCL10 production by tumor cells [381] . Also, decreased expression

of CXCR3 on NK cells in a murine model of multiple myeloma resulted in less efficient

homing of adoptively transferred NK cells as compared to healthy mice. This

phenomenon is also observed in multiple myeloma patients [382] . In addition to NK cell

homing, CXCR3 is also associated with increased proliferation and cytokine production

[159] . The known ligands of CXCR3 are CXCL9, CXCL10, and CXCL11. The notable

role of CXCR3 a nd CCR7 prompts the design of experiments to compare the tumor

homing capability of bifunctional NK - CARs that express CXCR3 versus CCR7, to

determine which receptor best drives tumor infiltration of NK cells. As discussed earlier,

167

tumor infiltrating NK cel ls have been associated with good prognosis, therefore

determining the most effective chemokine receptor can significantly improve the clinical

successes of NK - CAR cell therapy.

Foremost, we will determine the differences in organs to which CXCR3 - NK - CAR c ells

traffic to compared to CCR7 - NK - CAR cells. Using the previously described technique,

NKF - NK cells will be engineered with an anti - CD19 CAR linked to CD3ζ signaling

domain and contains CXCR3 (CXCR3 - NK - CAR). NSG mice will be injected with 10 - 15

million C FSE - labeled CXCR3 - NK - CAR cells or CFSE - labeled NK - BiCAR cells expressing CCR7, along with IL - 2. Seven to ten days following cell injection, the mice will be sacrificed and the following organs harvested and processed: spleen, lungs, liver, bone marrow, kid neys, and lymph nodes. Prior to organ extraction, blood will be drawn from each mouse for detecting the NK - CARs in peripheral blood. NK - CAR cells will be detected via flow cytometry. We can hypothesize that compared to other organs, the NK -

CARs are most li kely to migrate to the lymph nodes, as reported in literature. The second - most likely homing site may be the bone marrow.

Once the homing sites for the two NK - CARs have been recognized, then the effectiveness of the NK - CAR cells at inhibiting tumor growth will be investigated.

Secondly, this study should identify which NK - CAR cell best promotes homing of NK cells to the tumor site. NSG mice will be injected intravenously with luciferase - expressing CD19+ lymphoma cell line. The mice will then be injected wi th PBS, NK -

BiNull cells, NK - BiCAR cells, or CXCR3 - NK - CAR cells, along with IL - 2 weekly.

Tumor burden will be measured by in vivo BLI weekly. We predict that the control mice that received only PBS will develop significant disease extending to liver, joints , and

168

lymph nodes. The NK - BiNull cells should develop moderate disease. The mice that

received NK - BiCAR and CXCR3 - NK - CAR cells are expected to have significantly

reduced tumor burden as compared to the control groups. The study will end when the

mice have a tumor score ≥ 7 or appear moribund, and the mice will be sacrificed. The

liver was reported as a site of metastasis, therefore the liver will be harvested to

determine the proportion of the injected cells that homed to the site of disease [379] . We

will also examine the lymph nodes and bone marrow for tumor infiltrating NK - CAR

cells. We expect to observe higher numbers of infiltrating cells in tissues from the mice

treated with NK - BiCAR and CXCR3 - NK - CAR.

Study Limitations

The study only examines the chemotactic effects of two chemokine receptors, CCR7 and

CXCR3. Other chemokines such as CXCR4 have also been reported to play a role in NK cell homing. Therefore, future studies shou ld examine other chemokines that may lead to improved homing of NK cell to the tumor site. Secondly, this study performs a head - to -

head comparison of the two chemokine receptors but does not consider a synergistic

effect that may be possible.

Improving N K cell activity with TGF - βDN CAR

TGF - β is a well - known immnosuppressor, and has been shown to significantly impair

NK cell function by downregulating NK cell activating receptor. Many studies have

attempted to inhibit TGF - β signaling in NK cell and other i mmune cells through the use

of small molecule inhibitors that target different points in the TGF - β signaling cascade.

169

There is an ongoing clinical trial to treat metastatic colorectal cancer patients with

adoptively transferred T cells expressing TCR direc ted against a mutated version of

TGFbetaRII ( NCT03431311) . Mutated TGFbetaRII is expressed on >90% of MSI+ colon cancer cells, making it an attractive target for T cell therapy [383] . Another ongoing trial

on the use of fresolimumab, an antibody that binds to and inhibits all isoforms of TGF - β

(NCT02581787). Although these inhibitors have proven effective in preclinical

experiments, they have yet to demonstrate significant improvement in the clinical

outcomes of cancer patients [384] . For example, a phase III trial performed to determine

the effectiveness of a tumor vaccine consisting of TGF - β2 antisense gene - modified

NSCLC c ell lines resulted in no difference between subjects who received the vaccine

versus those who received placebo regarding overall survival or progression - free survival

[385] . The most successful TGF - βR1 inhibitor, galuniserti b, did not lead to better overall

survival when combined with lomustine in a phase II randomized study in patients with

recurrent glioblastoma as compared to Lomustine and placebo [386] . Due to its relatively

lower adverse event rate, galunisertib is being combined with other inhibitors and

chemotherapeutic drugs in several clinical trials. We propose a more targeted inhibition

of TGF - β signaling directed at adoptively transferre d NK cells.

We propose engineering a CD19 - based NK - CAR also encoding a dominant - negative

TGF - β (DN - TGF - β). This NK - CAR will produce a mutated form of TGF - β that will not

only compete with the wild - type TGF - β1 but also fail to induce TGF - β downstream

signa ling. Using lentiviral gene transfer, primary NK cells will be infected with DN -

TGF - β and stably transduced cells will be selected for using an appropriate antibiotic. As

negative control, the mock CAR vector will also be transduced into primary NK cells

170

( Mock). In an in vitro assay, DN - TGF - β cells’ resistance to TGF - β1 will be tested by

culturing the cells in TGF - β + IL - 2 or IL - 2 only for up to 72 hours. Mock cells will also be treated with TGF - β +IL - 2 or IL - 2 only for up to 72 hours. Using flow cytometry , the cell surface expression changes in DN - TGF - β cells as compared to Mock cells will be measured. We expect Mock cells treated with TGF - β, compared to DN - TGF - β cells, to have decreased expression of NK activating receptors, most especially NKG2D and

NKp4 6, according to published sources. In addition, TGF - β - treated Mock cells should also express increased levels of NK inhibitory receptors, compared to TGF - β - treated DN -

TGF - β cells. Each CD19+ target cell line used in this study will be tested to ensure they secrete pathologic levels of TGF - β. To verify the functional effects of the CAR on abrogating the effects of TGF - β on NK cell cytotoxic function, a calcein - AM - based cytotoxicity assay will be performed comparing the lysis of CD19+ targets by DN - TGF - β cell s and Mock cells. We expect TGF - β - treated DN - TGF - β cells to exhibit the same killing efficiency as untreated DN - TGF - β cells. TGF - β - treated Mock cells should demonstrate significantly decreased killing efficiency as compared to untreated Mock cells. Further more, IFN - γ secretion by TGF - β - treated or untreated DN - TGF - β cells and

Mock cells will be tested via ELISA. We expect TGF - β - treated DN - TGF - β cells to secrete nearly the same IFN - γ levels as untreated cells. TGF - β is known to negatively regulate IFN - γ secre tion therefore TGF - β - treated Mock cells should exhibit reduced IFN -

γ secretion compared to untreated Mock cells.

Using NSG mice, the ability of DN - TGF - β cells to effectively prevent cancer progression in vivo will be tested. CD19+ cell lines will be stabl y transfected with a firefly luciferase expression plasmid and injected subcutaneously into the flanks of 3 groups of mice. The

171

first group will be the control mice receiving only IL - 2 injection (Control); the second

group will be mice receiving the Mock c ells and IL - 2 (Mock group); and the third group will consist of mice receiving DN - TGF - β cells and IL - 2 weekly (DN - TGF - β group). The mice will be imaged weekly using an in vivo BLI scope to measure tumor burden and the tumor score for each mouse will be sco red as required by the Animal Resource Center. At the end of 5 weeks or when the control mice require sacrificing, the experiment will be stopped. The mice will be sacrificed and organs containing lymphoid compartments or known NK homing sites will be extr acted and the proportion of DN - TGF - β cells or Mock cells will be measured via immunohistochemistry. We expect the Mock mice to show less tumor burden as compared to the control mice, as demonstrated by the bioluminescence data obtained from the weekly imag ing. Moreover, the DN - TGF - β group will show hardly any disease, and the rationale being that a study performed using galunisertib and adoptive NK cells against colon cancer resulted in minimal disease.

An additional study will be to determine the survival benefit of the DN - TGF - β cells, since this is a very important marker of clinical success in clinical trials. Using NSG mice and the same disease model described above, DN - TGF - β cells or Mock cells will be injected into the mice weekly, along with IL - 2. A murine group that is only treated with

IL - 2 will be the control group. As a secondary study measure, the mice will be imaged weekly to ensure that any deaths observed is due to cancer progression and not some other cause such as infection. From this study, we will be able to determine whether DN -

TGF - β cells provide a survival advantage. We expect that the DN - TGF - β mice will survive the longest as compared to the other two groups. The last animal study will test the persistence of the NK - CAR cells in mice. T he persistence of NK cells has been

172

associated with patient outcome, therefore 10 - 15 million DN - TGF - β cells or Mock cells will be injected intravenously into NSG mice. Blood will be drawn from the mice every 2 days to measure how long the cells persist in mice via flow cytometry.

Study Limitations

One important limitation for this study is being able to generate enough NK cells to transduce with the CARs described above. With the development of the NKF feeder cell system, we are able to obtain up to 20,000 - fold in 5 weeks, which will generate more than enough cells to treat 3 - 5 patients. Another limitation to consider when translating this NK - CAR into the clinic is the patient - to - patient variability observed not only in expansion rates but also in cytotoxic function. As in other cell therapy trials, NK cell donor cells will be tested beforehand to ensure that they can expand robustly. Lastly, the translation of this study might be difficult due to the toxicity associated with IL - 2 infusion. There are trials ongoing utilizing low - dose IL - 2 but the reduction in toxicity as compared to high dose IL - 2 has not been well established. It is important therefore to be sure that the benefit to the patient, i.e. increased disease - free survival, is worth the adverse even ts related to treatment.

Comparing DN - TGF - β cells with adoptive NK cell therapy administered with

galunisertib

A search in the U.S. National Library of Medicine database of clinical trials for

galunisertib or galunisertib in combination with other therap ies in treating cancer resulted

in 20 entries. Nivolumab resulted in over 600 st udies which can lead to difficulties in

173

determining which of the combinations tested is more efficacious and beneficial for the

various malignancies being targeted . It appears that there may be arbitrary combination of therapies bein g tested on cancer patients and many in the field are encouraging more rational treatment combinations [387] . Head - on comparisons of these combinations in preclinical studies might enforce more rational combinations based on biology of the diseases and the drugs. More direct comparisons may reduce the numbers of clinical trials being conducted and the frequent lack of clinical efficacy reported for these immunotherapies. Therefore, we propose comparing the NK cell - enhancing capability of the dominant negative TGF - β NK - CAR (DN - TGF - β) described earlier and inhibition of

TGF - β signaling using the inhibitor galunisertib. The main question being addressed by this study is whether genetic modification or inhibitor is more efficacious, and may inform which strategy is best to combine with other immunotherapy such as PD - 1 inhibitor.

NSG mice will be infected with luciferase - expressing, TGF - β secreting CD19+ target cell line subcutaneously. The mice will be divided into 5 groups: the first group receives no cells (Control gr oup); the second group receives Mock cells described earlier (Mock); the third group receives the DN - TGF - β cells (DN - TGF - β); the fourth group receives galunisertib (Gal); and the fifth group receives NK cells and galunisertib (NK - Gal). The cells will be in jected weekly along with IL - 2. The tumor burden of the mice will be measured by BLI weekly. We expect that the tumor burden of the mice in the DN - TGF - β and NK - Gal group will be the least as compared to the other groups. At this time, we expect that the DN - TGF - β group will fare better than the NK - Gal group because the dominant negative TGF - β can easily impede the activity of wild - type TGF - β since it is

174

being produced locally. Also, the DN - TGF - β cells are designed to target CD19 specifically and should theref ore lead to more specific lysis. The NK - Gal cells might be

limited by the systemic distribution of galunisertib and might not be as effective as DN -

TGF - β cells. Firm conclusions can be made once the study has been completed. In

addition, blood will be draw n weekly to detect the proportion of cells in the peripheral

blood of the mice to determine which treatment yields NK cells that are better at

persisting. The mice will be sacrificed once the tumor size is ≥1cm 3 or the mice become

moribund. The bone marrow and spleen of the mice from each group will be extracted

and stained for NK cells to determine the proportion of cells that home to those sites.

Study Limitations

IL - 2 is known to cause severe toxicities, therefore IL - 15 could be substituted for IL - 2

whe n translating to human studies. Indeed, IL - 15 has been shown to promote NK cell development and persistence. Unlike widely reported toxicity associated with IL - 2 treatment, only one study has reported toxicity associated with IL - 15 bolus given.

Although t he study attempts to determine the most optimal method of inhibiting TGF - β signaling to promote NK cell activity, it is difficult to make conclusions directly from the result without further analysis. For example, the increased expression of the NK - CAR on the transduced NK cells might not be equal to the galunisertib dose used. Therefore, the study can only provide data to support either the CAR or the inhibitor, at the dose used for the study.

175

REFERE N CES

1. P., E., Über den jetzigen Stand der Chemotherapie. Berichte der deutschen chemischen

Gesellschaft, 1909. 42 (1): p. 17 - 47.

2. L J Old, a. and E.A. Boyse, Immunology of Experimental Tumors. Annual Review of

Medicine, 1964. 15 (1): p. 167 - 186.

3. Burnet, M., Cancer — A Biological Approac h: III. Viruses Associated with Neoplastic

Conditions. IV. Practical Applications. British Medical Journal, 1957. 1 (5023): p. 841 - 847.

4. Dunn, G.P., et al., Cancer immunoediting: from immunosurveillance to tumor escape.

Nature Immunology, 2002. 3 : p. 991.

5. Stutman, O., Tumor Development after 3 - Methylcholanthrene in Immunologically

Deficient Athymic - Nude Mice. Science, 1974. 183 (4124): p. 534 - 536.

6. Stutman, O., Chemical Carcinogenesis in Nude Mice: Comparison Between Nude Mice

From Homozygous Matings a nd Heterozygous Matings and Effect of Age and Carcinogen

Dose2. JNCI: Journal of the National Cancer Institute, 1979. 62 (2): p. 353 - 358.

7. Outzen, H., et al., Spontaneous and induced tumor incidence in germfree 'nude' mice .

Vol. 17. 1975. 1 - 9.

8. Rygaard, J. and C.O. Povlsen, Is immunological surveillance not a cell - mediated immune

function? Transplantation, 1974. 17 (1): p. 135 - 6.

9. Ikehara, S., et al., Functional T cells in athymic nude mice. Proceedings of the National

Academy of Sciences of the United States of America, 1984. 81 (3): p. 886 - 888.

10. Hunig, T., T - cell function and specificity in athymic mice. Immunology Today, 1983. 4 (3):

p. 84 - 87.

11. Colucci, F., M.A. Caligiuri, and J.P. Di Santo, What does it take to make a natural killer?

Nature Revie ws Immunology, 2003. 3 : p. 413.

176

12. Laboratory, C.S.H., et al., Origins of Human Cancer: Incidence of cancer in humans . 1977:

Cold Spring Harbor Laboratory.

13. B., H.R., et al., Natural cytotoxic reactivity of mouse lymphoid cells against syngeneic and

al logeneic tumors. II. Characterization of effector cells. International Journal of Cancer,

1975. 16 (2): p. 230 - 239.

14. Street, S.E.A., E. Cretney, and M.J. Smyth, Perforin and interferon - γ activities

independently control tumor initiation, growth, and meta stasis. Blood, 2001. 97 (1): p.

192 - 197.

15. Russell, J.H. and T.J. Ley, Lymphocyte - Mediated Cytotoxicity. Annual Review of

Immunology, 2002. 20 (1): p. 323 - 370.

16. van den Broek, M.E., et al., Decreased tumor surveillance in perforin - deficient mice. The

Jo urnal of Experimental Medicine, 1996. 184 (5): p. 1781 - 1790.

17. Smyth, M.J., et al., Perforin - Mediated Cytotoxicity Is Critical for Surveillance of

Spontaneous Lymphoma. The Journal of Experimental Medicine, 2000. 192 (5): p. 755 -

760.

18. Shinkai, Y., et al ., RAG - 2 - deficient mice lack mature lymphocytes owing to inability to

initiate V(D)J rearrangement. Cell, 1992. 68 (5): p. 855 - 867.

19. Shankaran, V., et al., IFNγ and lymphocytes prevent primary tumour development and

shape tumour immunogenicity. Nature, 2 001. 410 : p. 1107.

20. Penn, I., C.G. Halgrimson, and T.E. Starzl, De Novo Malignant Tumors in Organ

Transplant Recipients. Transplantation proceedings, 1971. 3 (1): p. 773 - 778.

21. Ross Sheil, A.G., Cancer after transplantation. World Journal of Surgery, 1 986. 10 (3): p.

389 - 396.

177

22. Penn, I., Malignant melanoma in organ allograft recipients. Transplantation, 1996.

61 (2): p. 274 - 8.

23. Penn, I., Posttransplant malignancies. Transplantation Proceedings, 1999. 31 (1): p.

1260 - 1262.

24. Pham, S.M., et al., Solid tumors after heart transplantation: Lethality of lung cancer. The

Annals of Thoracic Surgery, 1995. 60 (6): p. 1623 - 1626.

25. Mihm, M.C., Jr., C.G. Clemente, and N. Cascinelli, Tumor infiltrating lymphocytes in

lymph node melanoma metastases: a histopathologic prognostic indicator and an

expression of local immune response. Lab Invest, 1996. 74 (1): p. 43 - 7.

26. Lucio Palma, Nicola Di Lorenzo, and Beniamino Guidetti, Lymphocytic infiltrates in

primary glioblastomas and recidivous gliomas. Journal of Neurosurgery, 1978. 49 (6): p.

854 - 861.

27. Franco, R., et al., Prognostic significance of her -2/neu expression in breast cancer and its

relationship to other prognostic factors. International Journal of Cancer, 1991. 49 (1): p.

44 - 49.

28. Deligdisch, L., A.J. Jacobs, and C.J. Cohen, Histologic correlates of virulence in ovarian

adenocarcinoma: II. Morphologic correlates of host response. American Journal of

Obstetrics and Gynecology, 1982. 144 (8): p. 885 - 889.

29. L., N., et al., Prognostic significance of histologic host response in cancer of the large

bowel. Cancer, 1981. 47 (5): p. 930 - 936.

30. Naito, Y., et al., CD8⁺ T Cells Infiltrated within Cancer Cell Nests as a

Prognostic Factor in Human Colorectal Cancer. Cancer Research, 1998. 58 (16): p. 3491 -

3494.

178

31. Vinay, D.S., et al., Immune evasion in cancer: Mechanistic basis and therapeutic

strategies. Seminars in Cancer Biology, 2015. 35 : p. S185 - S198.

32. Smyth, M.J., D.I. Godfrey, and J.A. Trapani, A fresh look at tumor immunosu rveillance

and immunotherapy. Nature Immunology, 2001. 2 : p. 293.

33. Hanahan, D. and J. Folkman, Patterns and Emerging Mechanisms of the Angiogenic

Switch during Tumorigenesis. Cell, 1996. 86 (3): p. 353 - 364.

34. Ikeda, H., L.J. Old, and R.D. Schreiber, Th e roles of IFNγ in protection against tumor

development and cancer immunoediting. Cytokine & Growth Factor Reviews, 2002.

13 (2): p. 95 - 109.

35. Bromberg, J.F., et al., Transcriptionally Active Stat1 is Required for the Antiproliferative

Effects of Both Int erferon α and Interferon γ. Proceedings of the National

Academy of Sciences of the United States of America, 1996. 93 (15): p. 7673 - 7678.

36. Kumar, A., et al., Defective TNF - α - Induced Apoptosis in STAT1 - Null Cells Due to Low

Constitutive Levels of Caspases. Science, 1997. 278 (5343): p. 1630 - 1632.

37. Luster, A.D. and J.V. Ravetch, Biochemical characterization of a gamma interferon -

inducible cytokine (IP - 10). The Journal of Experimental Medicine, 1987. 166 (4): p. 1084 -

1097.

38. Liao, F., et al., H uman Mig chemokine: biochemical and functional characterization. The

Journal of Experimental Medicine, 1995. 182 (5): p. 1301 - 1314.

39. Pertl, U., et al., IFN - γ - Inducible Protein - 10 Is Essential for the Generation of a Protective

Tumor - Specific CD8 T Cell R esponse Induced by Single - Chain IL - 12 Gene Therapy. The

Journal of Immunology, 2001. 166 (11): p. 6944 - 6951.

40. Pardoll, D.M., Spinning molecular immunology into successful immunotherapy. Nature

Reviews Immunology, 2002. 2 : p. 227.

179

41. Yokokawa, J., et al. , Enhanced Functionality of

CD4⁺CD25^highFoxP3⁺ Regulatory T Cells in the

Peripheral Blood of Patients with Prostate Cancer. Clinical Cancer Research, 2008. 14 (4):

p. 1032 - 1040.

42. Rubtsov, Y.P. and A.Y. Rudensky, TGFbeta s ignalling in control of T - cell - mediated self -

reactivity. Nat Rev Immunol, 2007. 7 (6): p. 443 - 53.

43. Fontenot, J.D., M.A. Gavin, and A.Y. Rudensky, Foxp3 programs the development and

function of CD4+CD25+ regulatory T cells. Nature Immunology, 2003. 4 : p. 330.

44. Zou, W., Regulatory T cells, tumour immunity and immunotherapy. Nature Reviews

Immunology, 2006. 6 : p. 295.

45. von Boehmer, H., Mechanisms of suppression by suppressor T cells. Nature Immunology,

2005. 6 : p. 338.

46. Grossman, W.J., et al., Human T Regulatory Cells Can Use the Perforin Pathway to Cause

Autologous Target Cell Death. Immunity, 2004. 21 (4): p. 589 - 601.

47. Umansky, V. and A. Sevko, Overcoming immunosuppression in the melanoma

microenvironment induced by chronic inflammation. Cancer I mmunology,

Immunotherapy, 2012. 61 (2): p. 275 - 282.

48. Ostrand - Rosenberg, S., et al., Regulating the suppressors: apoptosis and inflammation

govern the survival of tumor - induced myeloid - derived suppressor cells (MDSC). Cancer

Immunology, Immunotherapy, 2012. 61 (8): p. 1319 - 1325.

49. Liu, Y., et al., Targeting myeloid - derived suppressor cells for cancer immunotherapy.

Cancer Immunology, Immunotherapy, 2018.

180

50. Choi, H.S., et al., The prognostic effects of tumor infiltra ting regulatory T cells and

myeloid derived suppressor cells assessed by multicolor flow cytometry in gastric cancer

patients. Oncotarget, 2016. 7 (7): p. 7940 - 51.

51. Aras, S. and M.R. Zaidi, TAMeless traitors: macrophages in cancer progression and

metasta sis. British Journal Of Cancer, 2017. 117 : p. 1583.

52. Miguel, R.D.V., et al., CTL Induction of Tumoricidal Nitric Oxide Production by

Intratumoral Macrophages Is Critical for Tumor Elimination. The Journal of Immunology,

2010. 185 (11): p. 6706 - 6718.

53. Lin, E.Y., et al., Colony - Stimulating Factor 1 Promotes Progression of Mammary Tumors

to Malignancy. The Journal of Experimental Medicine, 2001. 193 (6): p. 727 - 740.

54. Munn, D.H., et al., Potential Regulatory Function of Human Dendritic Cells Expressing

I ndoleamine 2,3 - Dioxygenase. Science, 2002. 297 (5588): p. 1867 - 1870.

55. Wrzesinski, S.H., Y.Y. Wan, and R.A. Flavell, Transforming growth factor - beta and the

immune response: implications for anticancer therapy. Clin Cancer Res, 2007. 13 (18 Pt

1): p. 5262 - 70.

56. Boris, P., Role of transforming growth factor beta in cancer. Journal of Cellular

Physiology, 2001. 186 (2): p. 153 - 168.

57. Shull, M.M., et al., Targeted disruption of the mouse transforming growth factor - β1

gene results in multifocal inflamma tory disease. Nature, 1992. 359 : p. 693.

58. Suda, T. and A. Zlotnik, In vitro induction of CD8 expression on thymic pre - T cells. II.

Characterization of CD3 - CD4 - CD8 alpha + cells generated in vitro by culturing

CD25+CD3 - CD4 - CD8 - thymocytes with T cell gro wth factor - beta and tumor necrosis

factor - alpha. The Journal of Immunology, 1992. 149 (1): p. 71 - 76.

181

59. Takahama, Y., et al., Early progression of thymocytes along the CD4/CD8 developmental

pathway is regulated by a subset of thymic epithelial cells expres sing transforming

growth factor beta. The Journal of Experimental Medicine, 1994. 179 (5): p. 1495 - 1506.

60. Bendelac, A., P.B. Savage, and L. Teyton, The Biology of NKT Cells. Annual Review of

Immunology, 2007. 25 (1): p. 297 - 336.

61. Fynan, T.M. and M. Rei ss, Resistance to inhibition of cell growth by transforming growth

factor - beta and its role in oncogenesis. Critical reviews in oncogenesis, 1993. 4 (5): p.

493 - 540.

62. Moses, H. and M.H. Barcellos - Hoff TGF - beta biology in mammary development and

breast ca ncer . Cold Spring Harbor perspectives in biology, 2011. 3 , a003277 DOI:

10.1101/cshperspect.a003277.

63. Eisma, R.J., et al., Decreased expression of transforming growth factor beta receptors on

head and neck squamous cell carcinoma tumor cells. The Americ an Journal of Surgery,

1996. 172 (6): p. 641 - 645.

64. Koli, K.M. and C.L. Arteaga, Processing of the Transforming Growth Factor β Type I and II

Receptors: BIOSYNTHESIS AND LIGAND - INDUCED REGULATION. Journal of Biological

Chemistry, 1997. 272 (10): p. 6423 - 64 27.

65. Gerlini, G., et al., Metastatic Melanoma Secreted IL - 10 Down - Regulates CD1 Molecules

on Dendritic Cells in Metastatic Tumor Lesions. The American Journal of Pathology,

2004. 165 (6): p. 1853 - 1863.

66. A., G.G., et al., Interleukin -10 production by h uman carcinoma cell lines and its

relationship to interleukin -6 expression. International Journal of Cancer, 1993. 55 (1): p.

96 - 101.

182

67. Kurte, M., et al., A Synthetic Peptide Homologous to Functional Domain of Human IL - 10

Down - Regulates Expression of MHC Class I and Transporter Associated with Antigen

Processing 1/2 in Human Melanoma Cells. The Journal of Immunology, 2004. 173 (3): p.

1731 - 1737.

68. Hagenbaugh, A., et al., Altered Immune Responses in Interleukin 10 Transgenic Mice. The

Journal of Experiment al Medicine, 1997. 185 (12): p. 2101 - 2110.

69. Rabinovich, G.A., D. Gabrilovich, and E.M. Sotomayor, Immunosuppressive Strategies

that are Mediated by Tumor Cells. Annual Review of Immunology, 2007. 25 (1): p. 267 -

296.

70. Nagpal, G., et al., Computer - aided designing of immunosuppressive peptides based on IL -

10 inducing potential. Scientific Reports, 2017. 7 : p. 42851.

71. Groux, H., et al., Inhibitory and Stimulatory Effects of IL - 10 on Human CD8⁺

T Cells. The Journal of Immunology, 1998. 160 (7): p. 3188 - 3193.

72. Geginat, J., et al., The light and the dark sides of Interleukin - 10 in immune - mediated

diseases and cancer. Cytokine & Growth Factor Reviews, 2016. 30 : p. 87 - 93.

73. Mocellin, S., et al., IL - 10 stimulatory effects on human NK cells explor ed by gene profile

analysis. Genes And Immunity, 2004. 5 : p. 621.

74. Emmerich, J., et al., IL - 10 Directly Activates and Expands Tumor - Resident

CD8⁺ T Cells without De Novo Infiltration from Secondary

Lymphoid Organs. Cancer Research, 2 012. 72 (14): p. 3570 - 3581.

75. Neven, B., et al., A Mendelian predisposition to B - cell lymphoma caused by IL - 10R

deficiency. Blood, 2013. 122 (23): p. 3713 - 3722.

183

76. Staveley - O’Carroll, K., et al., Induction of antigen - specific T cell anergy: An early event in

the course of tumor progression. Proceedings of the National Academy of Sciences,

1998. 95 (3): p. 1178 - 1183.

77. Cuenca, A., et al., Extra - Lymphatic Solid Tumor Growth Is Not Immunologically Ignored

and Results in Early Induction of Antigen - Specific T - Cell Anergy. Dominant Role of Cross -

Tolerance to Tumor Antigens, 2003. 63 (24): p. 9007 - 9015.

78. Overwijk, W.W., et al., Tumor Regression and Autoimmunity after Reversal of a

Functionally Tolerant State of Self - reactive CD8⁺ T Cells. The Journal of

Experimental Medicine, 2003. 198 (4): p. 569 - 580.

79. Willimsky, G. and T. Blankenstein, Sporadic immunogenic tumours avoid destruction by

inducing T - cell tolerance. Nature, 2005. 437 : p. 141.

80. Lee, P.P., et al., Characterization of circulating T cel ls specific for tumor - associated

antigens in melanoma patients. Nature Medicine, 1999. 5 : p. 677.

81. Steinman, R.M., D. Hawiger, and M.C. Nussenzweig, Tolerogenic Dendritic Cells. Annual

Review of Immunology, 2003. 21 (1): p. 685 - 711.

82. Perez, V.L., et al., Induction of Peripheral T Cell Tolerance In Vivo Requires CTLA - 4

Engagement. Immunity, 1997. 6 (4): p. 411 - 417.

83. Gramaglia, I., et al., Ox - 40 Ligand: A Potent Costimulatory Molecule for Sustaining

Primary CD4 T Cell Responses. The Jo urnal of Immunology, 1998. 161 (12): p. 6510 - 6517.

84. Murata, S., et al., OX40 Costimulation Synergizes with GM - CSF Whole - Cell Vaccination to

Overcome Established CD8⁺ T Cell Tolerance to an Endogenous Tumor

Antigen. The Journal of Immunology, 2 006. 176 (2): p. 974 - 983.

184

85. Johnsen, A.K., et al., Deficiency of Transporter for Antigen Presentation (TAP) in Tumor

Cells Allows Evasion of Immune Surveillance and Increases Tumorigenesis. The Journal of

Immunology, 1999. 163 (8): p. 4224 - 4231.

86. Hickli n, D.J., F.M. Marincola, and S. Ferrone, HLA class I antigen downregulation in

human cancers: T - cell immunotherapy revives an old story. Molecular Medicine Today,

1999. 5 (4): p. 178 - 186.

87. Sharma, P., et al., Primary, Adaptive, and Acquired Resistance to Cancer

Immunotherapy. Cell, 2017. 168 (4): p. 707 - 723.

88. Drake, C.G., E. Jaffee, and D.M. Pardoll, Mechanisms of Immune Evasion by Tumors , in

Advances in Immunology . 2006, Academic Press. p. 51 - 81.

89. Yee, C., et al., Adoptive T cell therapy using antig en - specific CD8(+) T cell clones for the

treatment of patients with metastatic melanoma: In vivo persistence, migration, and

antitumor effect of transferred T cells. Proceedings of the National Academy of Sciences

of the United States of America, 2002. 99 ( 25): p. 16168 - 16173.

90. Ohnmacht, G.A., et al., Short - Term Kinetics of Tumor Antigen Expression in Response to

Vaccination. The Journal of Immunology, 2001. 167 (3): p. 1809 - 1820.

91. Bowman, T., et al., Stat3 - mediated Myc expression is required for Src tr ansformation and

PDGF - induced mitogenesis. Proceedings of the National Academy of Sciences, 2001.

98 (13): p. 7319 - 7324.

92. Catlett - Falcone, R., et al., Constitutive Activation of Stat3 Signaling Confers Resistance to

Apoptosis in Human U266 Myeloma Cells. Immunity, 1999. 10 (1): p. 105 - 115.

93. Kiuchi, N., et al., STAT3 Is Required for the gp130 - mediated Full Activation of the c -

myc Gene. The Journal of Experimental Medicine, 1999. 189 (1): p. 63 - 73.

185

94. Wang, T., et al., Regulation of the innate an d adaptive immune responses by Stat - 3

signaling in tumor cells. Nature Medicine, 2003. 10 : p. 48.

95. Grivennikov, S.I., F.R. Greten, and M. Karin, Immunity, Inflammation, and Cancer. Cell,

2010. 140 (6): p. 883 - 899.

96. Karin, M., Nuclear factor - κB in cancer development and progression. Nature, 2006. 441 :

p. 431.

97. Siegel, R.M., et al., The multifaceted role of Fas signaling in immune cell homeostasis

and autoimmunity. Nature Immunology, 2000. 1 : p. 469.

98. Ryan, A.E., et al., Addressing the “Fas Cou nterattack” Controversy: Blocking Fas Ligand

Expression Suppresses Tumor Immune Evasion of Colon Cancer In vivo.

Cancer Research, 2005. 65 (21): p. 9817 - 9823.

99. Andreola, G., et al., Induction of Lymphocyte Apoptosis by Tumor Cell Secretion of Fa sL -

bearing Microvesicles. The Journal of Experimental Medicine, 2002. 195 (10): p. 1303 -

1316.

100. Alam, A., et al., Early Activation of Caspases during T Lymphocyte Stimulation Results in

Selective Substrate Cleavage in Nonapoptotic Cells. The Journal of E xperimental

Medicine, 1999. 190 (12): p. 1879 - 1890.

101. Chappell, D.B., et al., Human Melanoma Cells Do Not Express Fas (Apo - 1/CD95) Ligand.

Cancer Research, 1999. 59 (1): p. 59 - 62.

102. Chen, J. - J., Y. Sun, and G.J. Nabel, Regulation of the Proinflammatory Effects of Fas

Ligand (CD95L). Science, 1998. 282 (5394): p. 1714 - 1717.

103. Martínez - Lostao, L., A. Anel, and J. Pardo, How Do Cytotoxic Lymphocytes Kill Cancer

Cells? Clinical Cancer Research, 2015. 21 (22): p. 5047 - 5056.

186

104. Giovarelli, M., et al., A “S tealth Effect”: Adenocarcinoma Cells Engineered to Express

TRAIL Elude Tumor - Specific and Allogeneic T Cell Reactions. The Journal of Immunology,

1999. 163 (9): p. 4886 - 4893.

105. DeSimone, P.A., R. East, and R.D. Powell, Jr., Phagocytic tumor cell activity in oat cell

carcinoma of the lung. Hum Pathol, 1980. 11 (5 Suppl): p. 535 - 9.

106. Lugini, L., et al., Cannibalism of Live Lymphocytes by Human Metastatic but Not Primary

Melanoma Cells. Cancer Research, 2006. 66 (7): p. 3629 - 3638.

107. Caruso, R.A., et al., Morphological Evidence of Neutrophil - Tumor Cell Phagocytosis

(Cannibalism) in Human Gastric Adenocarcinomas. Ultrastructural Pathology, 2002.

26 (5): p. 315 - 321.

108. Fujii, M., et al., Cytologic diagnosis of male breast cancer with nipple discharge. A cas e

report. Acta Cytol, 1986. 30 (1): p. 21 - 4.

109. Kojima, S., et al., Clinical Significance of “Cannibalism” in Urinary Cytology of Bladder

Cancer. Acta Cytologica, 1998. 42 (6): p. 1365 - 1369.

110. A., K.S., M.J. S., and M.J.D. H., Natural killer cells: a review of manufacturing and clinical

utility. Transfusion, 2013. 53 (2): p. 404 - 410.

111. R., K., K. Eva, and W. H., „Natural ” killer cells in the mouse. I. Cytotoxic cells with

specificity for mouse Moloney leukemia cells. Specificity and distribution acc ording to

genotype. European Journal of Immunology, 1975. 5 (2): p. 112 - 117.

112. Romee, R., J.W. Leong, and T.A. Fehniger, Utilizing Cytokines to Function - Enable Human

NK Cells for the Immunotherapy of Cancer. Scientifica, 2014. 2014 : p. 205796.

113. Van A cker, H.H., et al., CD56 in the Immune System: More Than a Marker for

Cytotoxicity? Frontiers in Immunology, 2017. 8 : p. 892.

187

114. Björkström, N.K., H. - G. Ljunggren, and J.K. Sandberg, CD56 negative NK cells: origin,

function, and role in chronic viral dis ease. Trends in Immunology, 2010. 31 (11): p. 401 -

406.

115. Sigalov, A.B., Multichain immune recognition receptor signaling : from spatiotemporal

organization to human disease . Advances in experimental medicine and biology. 2008,

New York, N.Y.

Austin, Tex. : Springer Science+Business Media ;

Landes Bioscience. xxvii, 357 p.

116. Waldhauer, I. and A. Steinle, NK cells and cancer immunosurveillance. Oncogene, 2008.

27 : p. 5932.

117. Leung, W., Infusions of allogeneic natural killer cells as cancer therapy. Cli n Cancer Res,

2014. 20 (13): p. 3390 - 400.

118. Freud, A.G. and M.A. Caligiuri, Human natural killer cell development. Immunological

Reviews, 2006. 214 (1): p. 56 - 72.

119. Freud, A.G., et al., Evidence for discrete stages of human natural killer cell differentiation

in vivo. The Journal of Experimental Medicine, 2006. 203 (4): p. 1033 - 1043.

120. Simonetta, F., A. Pradier, and E. Roosnek, T - bet and Eomesodermin in NK Cell

Development, Maturation, and Function. Frontiers in Immunology, 2016. 7 : p. 241.

12 1. Townsend, M.J., et al., T - bet Regulates the Terminal Maturation and Homeostasis of NK

and Vα14i NKT Cells. Immunity, 2004. 20 (4): p. 477 - 494.

122. Gordon, Scott M., et al., The Transcription Factors T - bet and Eomes Control Key

Checkpoints of Natural Killer Cell Maturation. Immunity, 2012. 36 (1): p. 55 - 67.

123. Kim, S., et al., Licensing of natural killer cells by host major histocompatibility complex

class I molecules. Nature, 2005. 436 : p. 709.

188

124. Lazetic, S., et al., Human natural kill er cell receptors involved in MHC class I recognition

are disulfide - linked heterodimers of CD94 and NKG2 subunits. The Journal of

Immunology, 1996. 157 (11): p. 4741 - 4745.

125. Borrego, F., et al., Recognition of Human Histocompatibility Leukocyte Antigen ( HLA) - E

Complexed with HLA Class I Signal Sequence – derived Peptides by CD94/NKG2 Confers

Protection from Natural Killer Cell – mediated Lysis. The Journal of Experimental

Medicine, 1998. 187 (5): p. 813 - 818.

126. Tu, M.M., A.B. Mahmoud, and A.P. Makrigiannis, Licensed and Unlicensed NK Cells:

Differential Roles in Cancer and Viral Control. Frontiers in Immunology, 2016. 7 : p. 166.

127. Wing, L., Use of NK cell activity in cure by transplant. British Journal of Haematology,

2011. 155 (1): p. 14 - 29.

128. Lee, S. - H ., T. Miyagi, and C.A. Biron, Keeping NK cells in highly regulated antiviral

warfare. Trends in Immunology, 2007. 28 (6): p. 252 - 259.

129. Desrosiers, M. - P., et al., Epistasis between mouse Klra and major histocompatibility

complex class I loci is associate d with a new mechanism of natural killer cell – mediated

innate resistance to cytomegalovirus infection. Nature genetics, 2005. 37 (6): p. 593 - 599.

130. Vidal, S.M. and L.L. Lanier, NK Cell Recognition of Mouse Cytomegalovirus - Infected Cells ,

in Immunobiology of Natural Killer Cell Receptors , R.W. Compans, et al., Editors. 2006,

Springer Berlin Heidelberg: Berlin, Heidelberg. p. 183 - 206.

131. See, D.M., et al., The Role of Natural Killer Cells in Viral Infections. Scandinavian Journal

of Immunology, 1997. 46 (3 ): p. 217 - 224.

132. Vivier, E., et al., Functions of natural killer cells. Nature Immunology, 2008. 9 : p. 503.

133. Fischer, A., Human Primary Immunodeficiency Diseases. Immunity, 2007. 27 (6): p. 835 -

845.

189

134. Fairweather, D., et al., From Infection to Autoimmunity. Journal of Autoimmunity, 2001.

16 (3): p. 175 - 186.

135. van Dommelen, S.L.H., et al., Perforin and Granzymes Have Distinct Roles in Defensive

Immunity and Immunopathology. Immunity, 2006. 25 (5): p. 835 - 848.

136. Kelly, J.M., et al., Induction of tumor - specific T cell memory by NK cell – mediated tumor

rejection. Nature Immunology, 2001. 3 : p. 83.

137. Smyth, M.J., et al., NKG2D function protects the host from tumor initiation. The Journal

of Experimental Medicine, 2005. 202 (5): p. 583 - 588.

138. S treet, S.E.A., et al., Innate Immune Surveillance of Spontaneous B Cell Lymphomas by

Natural Killer Cells and γδ T Cells. The Journal of Experimental Medicine, 2004. 199 (6): p.

879 - 884.

139. Imai, K., et al., Natural cytotoxic activity of peripheral - blood lymphocytes and cancer

incidence: an 11 - year follow - up study of a general population. The Lancet, 2000.

356 (9244): p. 1795 - 1799.

140. Koh, C.Y., et al., Augmentation of antitumor effects by NK cell inhibitory receptor

blockade in vitro and in vivo. Blood, 2001. 97 (10): p. 3132 - 3137.

141. Hayakawa, Y., et al., NK Cell TRAIL Eliminates Immature Dendritic Cells In Vivo and Limits

Dendritic Cell Vaccination Efficacy. The Journal of Immunology, 2004. 172 (1): p. 123 - 129.

142. Lu, L., et al., Regulation of Activated CD4⁺ T Cells by NK Cells via the Qa -

1–NKG2A Inhibitory Pathway. Immunity, 2007. 26 (5): p. 593 - 604.

143. Takeda, K. and G. Dennert, The development of autoimmunity in C57BL/6 lpr mice

correlates with the disappearance of natural killer type 1 - positive cells: evidence for their

suppressive action on bone marrow stem cell proliferation, B cell immunoglobulin

190

secretion, and autoimmune symptoms. The Journal of Experimental Medicine, 1993.

177 (1): p. 155 - 164.

144. Yoneda, O., et al., F ractalkine - Mediated Endothelial Cell Injury by NK Cells. The Journal of

Immunology, 2000. 164 (8): p. 4055 - 4062.

145. Hanna, J., et al., Decidual NK cells regulate key developmental processes at the human

fetal - maternal interface. Nature Medicine, 2006. 12 : p. 1065.

146. Paul, S. and G. Lal, The Molecular Mechanism of Natural Killer Cells Function and Its

Importance in Cancer Immunotherapy. Frontiers in Immunology, 2017. 8 (1124).

147. Hashimoto, W., et al., Differential Antitumor Effects of Administration of Recombinant

IL - 18 or Recombinant IL - 12 Are Mediated Primarily by Fas - Fas Ligand - and Perforin -

Induced Tumor Apoptosis, Respectively. The Journal of Immunology, 1999. 163 (2): p.

583 - 589.

148. Lugini, L., et al., Immune Surveillance Properties of Human NK C ell - Derived Exosomes.

The Journal of Immunology, 2012. 189 (6): p. 2833 - 2842.

149. von Strandmann, E.P., et al., A novel bispecific protein (ULBP2 - BB4) targeting the NKG2D

receptor on natural killer (NK) cells and CD138 activates NK cells and has potent

ant itumor activity against human multiple myeloma in vitro and in vivo. Blood, 2006.

107 (5): p. 1955 - 1962.

150. Mathan, T., C. Figdor, and S. Buschow, Human Plasmacytoid Dendritic Cells: From

Molecules to Intercellular Communication Network. Frontiers in Immu nology, 2013.

4 (372).

151. Zingoni, A., et al., Cross - Talk between Activated Human NK Cells and CD4⁺ T

Cells via OX40 - OX40 Ligand Interactions. The Journal of Immunology, 2004. 173 (6): p.

3716 - 3724.

191

152. Turaj, A.H., et al., Augmentation of CD13 4 (OX40) - dependent NK anti - tumour activity is

dependent on antibody cross - linking. Scientific Reports, 2018. 8 (1): p. 2278.

153. de Andrade, L.F., M.J. Smyth, and L. Martinet, DNAM - 1 control of natural killer cells

functions through nectin and nectin - like proteins. Immunology And Cell Biology, 2013.

92 : p. 237.

154. Li, S., et al., Rapidly induced, T - cell – independent xenoantibody production is mediated by

marginal zone B cells and requires help from NK cells. Blood, 2007. 110 (12): p. 3926 -

3935.

155. Croft, M., Control of Immunity by the TNFR - Related Molecule OX40 (CD134). Annual

Review of Immunology, 2010. 28 (1): p. 57 - 78.

156. Griffith, J.W., C.L. Sokol, and A.D. Luster, Chemokines and Chemokine Receptors:

Positioning Cells for Host Defense and Immunity. An nual Review of Immunology, 2014.

32 (1): p. 659 - 702.

157. Fauriat, C., et al., Regulation of human NK - cell cytokine and chemokine production by

target cell recognition. Blood, 2010. 115 (11): p. 2167 - 2176.

158. Gras Navarro, A., A.T. Björklund, and M. Cheken ya, Therapeutic Potential and Challenges

of Natural Killer Cells in Treatment of Solid Tumors. Frontiers in Immunology, 2015.

6 (202).

159. Bernardini, G., et al., Dysregulation of Chemokine/Chemokine Receptor Axes and NK Cell

Tissue Localization during Diseases. Frontiers in Immunology, 2016. 7 : p. 402.

160. Zlotnik, A. and O. Yoshie, The Chemokine Superfamily Revisited. Immunity, 2012. 36 (5):

p. 705 - 716.

161. Lima, M., et al., Chemokine Receptor Expression on Normal Blood CD56+ NK - Cells

Elucidates Cell Partners That Comigrate during the Innate and Adaptive Immune

192

Responses and Identifies a Transitional NK - Cell Population. Journal of Immunology

Research, 2015. 2015 : p. 18.

162. Pegram, H.J., et al., Activating and inhibitory receptors of natural killer ce lls. Immunol

Cell Biol, 2011. 89 (2): p. 216 - 24.

163. Chretien, A.S., et al., Cancer - Induced Alterations of NK - Mediated Target Recognition:

Current and Investigational Pharmacological Strategies Aiming at Restoring NK -

Mediated Anti - Tumor Activity. Front Imm unol, 2014. 5 : p. 122.

164. Gilfillan, S., et al., NKG2D recruits two distinct adapters to trigger NK cell activation and

costimulation. Nature Immunology, 2002. 3 : p. 1150.

165. Fikri, Y., et al., Cloning, sequencing, and cell surface expression pattern of bovine

immunoreceptor NKG2D and adaptor molecules DAP10 and DAP12. Immunogenetics,

2007. 59 (8): p. 653 - 9.

166. Baragaño Raneros, A., B. Suarez - Álvarez, and C. López - Larrea, Secretory pathways

generating immunosuppressive NKG2D ligands: New targets for t herapeutic

intervention. Oncoimmunology, 2014. 3 : p. e28497.

167. Maria, S.G., et al., NK cell -mediated lysis of autologous antigen-presenting cells is

triggered by the engagement of the phosphatidylinositol 3 -kinase upon ligation of the

natural cytotoxici ty receptors NKp30 and NKp46. European Journal of Immunology,

2001. 31 (6): p. 1656 - 1665.

168. Simona, S., et al., NKp46 is the major triggering receptor involved in the natural

cytotoxicity of fresh or cultured human NK cells. Correlation between surface d ensity of

NKp46 and natural cytotoxicity against autologous, allogeneic or xenogeneic target cells.

European Journal of Immunology, 1999. 29 (5): p. 1656 - 1666.

193

169. Ferlazzo, G., et al., Human Dendritic Cells Activate Resting Natural Killer (NK) Cells and

A re Recognized via the NKp30 Receptor by Activated NK Cells. The Journal of

Experimental Medicine, 2002. 195 (3): p. 343 - 351.

170. Chester, C., K. Fritsch, and H.E. Kohrt, Natural Killer Cell Immunomodulation: Targeting

Activating, Inhibitory, and Co - stimula tory Receptor Signaling for Cancer Immunotherapy.

Frontiers in Immunology, 2015. 6 : p. 601.

171. Garni - Wagner, B.A., et al., A novel function - associated molecule related to non - MHC -

restricted cytotoxicity mediated by activated natural killer cells and T ce lls. The Journal

of Immunology, 1993. 151 (1): p. 60 - 70.

172. Lee, K. - M., et al., 2B4 Acts As a Non – Major Histocompatibility Complex Binding

Inhibitory Receptor on Mouse Natural Killer Cells. The Journal of Experimental Medicine,

2004. 199 (9): p. 1245 - 1254.

173. Vaidya, S.V., et al., Targeted Disruption of the 2B4 Gene in Mice Reveals an

In Vivo Role of 2B4 (CD244) in the Rejection of B16 Melanoma Cells. The Journal of

Immunology, 2005. 174 (2): p. 800 - 807.

174. Lopez - Vergès, S., et al., CD57 defines a functionally distinct population of mature NK

cells in the human CD56^dimCD16⁺ NK - cell subset. Blood, 2010.

116 (19): p. 3865 - 3874.

175. Pittari, G., et al., Revving up Natural Killer Cells and Cytokine - Induced Killer Cells Against

H ematological Malignancies. Frontiers in Immunology, 2015. 6 : p. 230.

176. Billadeau, D.D., et al., The Vav – Rac1 Pathway in Cytotoxic Lymphocytes Regulates the

Generation of Cell - mediated Killing. The Journal of Experimental Medicine, 1998. 188 (3):

p. 549 - 5 59.

194

177. Mehta, R.S., et al., NK cell therapy for hematologic malignancies. International Journal of

Hematology, 2018. 107 (3): p. 262 - 270.

178. Smyth, M.J., et al., Activation of NK cell cytotoxicity. Molecular Immunology, 2005. 42 (4):

p. 501 - 510.

179. Bar ber, D.F., M. Faure, and E.O. Long, LFA - 1 Contributes an Early Signal for NK Cell

Cytotoxicity. The Journal of Immunology, 2004. 173 (6): p. 3653 - 3659.

180. Jewett, A.n., et al., Cytokine dependent inverse regulation of CD54 (ICAM1) and major

histocompatibi lity complex class I antigens by nuclear factor κB in HEp2 tumor cell line:

effect on the function of natural killer cells. Human Immunology, 2003. 64 (5): p. 505 -

520.

181. Kerr, J.F.R., A.H. Wyllie, and A.R. Currie, Apoptosis: A Basic Biological Phenomenon with

Wide - ranging Implications in Tissue Kinetics. British Journal of Cancer, 1972. 26 (4): p.

239 - 257.

182. Zamai, L., et al., Natural Killer (NK) Cell – mediated Cytotoxicity: Differential Use of TRAIL

and Fas Ligand by Immature and Mature Primary Human N K Cells. The Journal of

Experimental Medicine, 1998. 188 (12): p. 2375 - 2380.

183. Takeda, K., et al., Involvement of tumor necrosis factor - related apoptosis - inducing ligand

in surveillance of tumor metastasis by liver natural killer cells. Nature Medicine, 2001. 7 :

p. 94.

184. Grosse - Wilde, A., et al., TRAIL - R deficiency in mice enhances lymph node metastasis

without affecting primary tumor development. The Journal of Clinical Investigation,

2008. 118 (1): p. 100 - 110.

195

185. Monika, M. - S., et al., Natural kille r cell - mediated contact sensitivity develops rapidly and

depends on interferon - α, interferon - γ and interleukin - 12. Immunology, 2013. 140 (1): p.

98 - 110.

186. A., H.K., C.E. M., and G.M. D., Origins of natural killer cell memory: special creation or

adaptive evolution. Immunology, 2018. 154 (1): p. 38 - 49.

187. Welsh, R.M., C.L. O'Donnell, and L.D. Shultz, Antiviral activity of NK 1.1+ natural killer

cells in C57BL/6 scid mice infected with murine cytomegalovirus. Nat Immun, 1994.

13 (5): p. 239 - 45.

188. Bubić, I., et al., Gain of Virulence Caused by Loss of a Gene in Murine Cytomegalovirus.

Journal of Virology, 2004. 78 (14): p. 7536 - 7544.

189. Pende, D., et al., Expression of the DNAM - 1 ligands, Nectin - 2 (CD112) and poliovirus

receptor (CD155), on dendrit ic cells: relevance for natural killer - dendritic cell interaction.

Blood, 2006. 107 (5): p. 2030 - 2036.

190. Nabekura, T., et al., The costimulatory molecule DNAM - 1 is essential for optimal

differentiation of memory natural killer cells during mouse cytomega lovirus infection.

Immunity, 2014. 40 (2): p. 225 - 234.

191. Biron, C.A., K.S. Byron, and J.L. Sullivan, Severe Herpesvirus Infections in an Adolescent

without Natural Killer Cells. New England Journal of Medicine, 1989. 320 (26): p. 1731 -

1735.

192. Schlums, H., et al., Cytomegalovirus Infection Drives Adaptive Epigenetic Diversification

of NK Cells with Altered Signaling and Effector Function. Immunity, 2015. 42 (3): p. 443 -

456.

193. Rölle, A., et al., IL - 12 – producing monocytes and HLA - E control HCMV - driven NK G2C(+)

NK cell expansion. The Journal of Clinical Investigation, 2014. 124 (12): p. 5305 - 5316.

196

194. Nicholas, N., F. Neva, and G. Michael, NK cells generate memory - type responses to

human cytomegalovirus - infected fibroblasts. European Journal of Immunology, 2017.

47 (6): p. 1032 - 1039.

195. Beziat, V., et al., NK cell responses to cytomegalovirus infection lead to stable imprints in

the human KIR repertoire and involve activating KIRs. Blood, 2013. 121 (14): p. 2678 -

2688.

196. Liu, Lisa L., et al., Critical Rol e of CD2 Co - stimulation in Adaptive Natural Killer Cell

Responses Revealed in NKG2C - Deficient Humans. Cell Reports, 2016. 15 (5): p. 1088 -

1099.

197. Zhang, T., et al., Antibody - dependent memory - like NK cells distinguished by FcRγ -

deficiency. Journal of immu nology (Baltimore, Md. : 1950), 2013. 190 (4): p. 1402 - 1406.

198. Cooper, M.A., et al., Cytokine - induced memory - like natural killer cells. Proceedings of

the National Academy of Sciences, 2009. 106 (6): p. 1915 - 1919.

199. Ni, J., et al., Sustained effector f unction of IL - 12/15/18 – preactivated NK cells against

established tumors. The Journal of Experimental Medicine, 2012. 209 (13): p. 2351 - 2365.

200. Romee, R., et al., Cytokine - induced memory - like natural killer cells exhibit enhanced

responses against myeloid leukemia. Science Translational Medicine, 2016. 8 (357): p.

357ra123 - 357ra123.

201. Castro - Giner, F., et al., Cancer Diagnosis Using a Liquid Biopsy: Challenges and

Expectations. Diagnostics, 2018. 8 (2): p. 31.

202. Gerlinger, M., et al., Intratumor Hetero geneity and Branched Evolution Revealed by

Multiregion Sequencing. New England Journal of Medicine, 2012. 366 (10): p. 883 - 892.

203. Amirouchene - Angelozzi, N., C. Swanton, and A. Bardelli, Tumor Evolution as a

Therapeutic Target. Cancer Discovery, 2017. 7 (8 ): p. 805 - 817.

197

204. Halvaei, S., et al., Exosomes in Cancer Liquid Biopsy: A Focus on Breast Cancer. Molecular

Therapy - Nucleic Acids, 2018. 10 : p. 131 - 141.

205. Mehta, R.S. and K. Rezvani, Immune reconstitution post allogeneic transplant and the

impact o f immune recovery on the risk of infection. Virulence, 2016. 7 (8): p. 901 - 916.

206. Woll, P.S., et al., Human embryonic stem cells differentiate into a homogeneous

population of natural killer cells with potent in vivo antitumor activity. Blood, 2009.

113 ( 24): p. 6094 - 6101.

207. Spanholtz, J., et al., High Log - Scale Expansion of Functional Human Natural Killer Cells

from Umbilical Cord Blood CD34 - Positive Cells for Adoptive Cancer Immunotherapy.

PLOS ONE, 2010. 5 (2): p. e9221.

208. Tonn, T., et al., Treatme nt of patients with advanced cancer with the natural killer cell

line NK - 92. Cytotherapy, 2013. 15 (12): p. 1563 - 1570.

209. A., K.S., et al., Successful “in -flight” activation of natural killer cells during long-distance

shipping. Transfusion, 2013. 53 (2): p. 398 - 403.

210. Hermanson, D.L., et al., Induced Pluripotent Stem Cell - Derived Natural Killer Cells for

Treatment of Ovarian Cancer. Stem Cells, 2016. 34 (1): p. 93 - 101.

211. Kang, L., et al., Characterization and ex vivo Expansion of Human Placenta - Derived

Natural Killer Cells for Cancer Immunotherapy. Frontiers in Immunology, 2013. 4 : p. 101.

212. Blaise, D., et al., The Use of a Sequential High Dose Recombinant Interleukin 2 Regimen

After Autologous Bone Marrow Transplantation Does Not Imp rove the Disease Free

Survival of Patients with Acute Leukemia Transplanted in First Complete Remission.

Leukemia & Lymphoma, 1997. 25 (5 - 6): p. 469 - 478.

198

213. Miller, J.S., et al., Low dose subcutaneous interleukin - 2 after autologous transplantation

generates sustained in vivo natural killer cell activity. Biology of Blood and Marrow

Transplantation, 1997. 3 (1): p. 34 - 44.

214. Meropol, N.J., et al., Daily subcutaneous injection of low - dose interleukin 2 expands

natural killer cells in vivo without sig nificant toxicity. Clinical Cancer Research, 1996.

2 (4): p. 669 - 677.

215. Rosenberg, S.A., et al., Observations on the Systemic Administration of Autologous

Lymphokine - Activated Killer Cells and Recombinant Interleukin - 2 to Patients with

Metastatic Cancer. New England Journal of Medicine, 1985. 313 (23): p. 1485 - 1492.

216. Rosenberg, S.A., et al., A Progress Report on the Treatment of 157 Patients with

Advanced Cancer Using Lymphokine - Activated Killer Cells and Interleukin - 2 or High - Dose

Interleukin - 2 Alone. New England Journal of Medicine, 1987. 316 (15): p. 889 - 897.

217. Lister, J., et al., Autologous peripheral blood stem cell transplantation and adoptive

immunotherapy with activated natural killer cells in the immediate posttransplant

period. Clinical Canc er Research, 1995. 1 (6): p. 607 - 614.

218. deMagalhaes - Silverman, M., et al., Posttransplant Adoptive Immunotherapy With

Activated Natural Killer Cells in Patients With Metastatic Breast Cancer. Journal of

Immunotherapy, 2000. 23 (1): p. 154 - 160.

219. Burns, L.J., et al., IL - 2 - based immunotherapy after autologous transplantation for

lymphoma and breast cancer induces immune activation and cytokine release: a phase

I/II trial. Bone Marrow Transplantation, 2003. 32 : p. 177.

220. Sim, G.C., et al., IL - 2 therapy promotes suppressive ICOS+ Treg expansion in melanoma

patients. The Journal of Clinical Investigation, 2014. 124 (1): p. 99 - 110.

199

221. Knorr, D.A., et al., Clinical utility of natural killer cells in cancer therapy and

transplantation. Semin Immunol, 2014. 2 6 (2): p. 161 - 72.

222. Mamessier, E., et al., Human breast cancer cells enhance self tolerance by promoting

evasion from NK cell antitumor immunity. The Journal of Clinical Investigation, 2011.

121 (9): p. 3609 - 3622.

223. Ciurea, S.O., et al., Phase 1 clinic al trial using mbIL21 ex vivo – expanded donor - derived

NK cells after haploidentical transplantation. Blood, 2017. 130 (16): p. 1857 - 1868.

224. Miller, J.S., et al., Successful adoptive transfer and in vivo expansion of human

haploidentical NK cells in patien ts with cancer. Blood, 2005. 105 (8): p. 3051 - 3057.

225. Bachanova, V., et al., Clearance of acute myeloid leukemia by haploidentical natural

killer cells is improved using IL - 2 diphtheria toxin fusion protein. Blood, 2014. 123 (25): p.

3855 - 3863.

226. Gill, S., et al., Rapid development of exhaustion and down - regulation of eomesodermin

limit the antitumor activity of adoptively transferred murine natural killer cells. Blood,

2012. 119 (24): p. 5758 - 5768.

227. Gill, S., et al., Rapid development of exhaustion and down - regulation of eomesodermin

limit the antitumor activity of adoptively transferred murine natural killer cells. Blood,

2012. 119 (24): p. 5758 - 68.

228. Garcia - Iglesias, T., et al., Low NKp30, NKp46 and NKG2D expression and reduced

cytotoxic activity on NK cells in cervical cancer and precursor lesions. BMC Cancer, 2009.

9 : p. 186 - 186.

229. Carlsten, M., et al., Primary Human Tumor Cells Expressing CD155 Impair Tumor

Targeting by Down - Regulating DNAM - 1 on NK Cells. The Journal of Immunology, 2009.

183 (8): p. 4921 - 4930.

200

230. Alejandro, L.S., et al., NKG2D signaling in cancer immunosurveillance. International

Journal of Cancer, 2015. 136 (8): p. 1741 - 1750.

231. Parisi, S., et al., The More, The Better: “Do the Right Thing” For Natural Killer

Immunotherapy in Acute Myeloid Leukemia. Frontiers in Immunology, 2017. 8 : p. 1330.

232. Seledtsov, V.I., A.G. Goncharov, and G.V. Seledtsova, Clinically feasible approaches to

potentiating cancer cell - based immunotherapies. Human Vaccines &

Immunotherapeutics, 2015. 1 1 (4): p. 851 - 869.

233. Mehta, R.S. and K. Rezvani, Chimeric Antigen Receptor Expressing Natural Killer Cells for

the Immunotherapy of Cancer. Frontiers in Immunology, 2018. 9 (283).

234. Boyiadzis, M., et al., Phase 1 clinical trial of adoptive immunotherapy using “off - the -

shelf” activated natural killer cells in patients with refractory and relapsed acute myeloid

leukemia. Cytotherapy, 2017. 19 (10): p. 1225 - 1232.

235. Schönfeld, K., et al., Selective Inhibition of Tumor Growth by Clonal NK Cells Expressing

an ErbB2/HER2 - Specific Chimeric Antigen Receptor. Molecular Therapy, 2015. 23 (2): p.

330 - 338.

236. Shimasaki, N., et al., A clinically adaptable method to enhance the cytotoxicity of natural

killer cells against B - cell malignancies. Cytotherapy , 2012. 14 (7): p. 830 - 840.

237. Chu, J., et al., CS1 - specific chimeric antigen receptor (CAR) - engineered natural killer cells

enhance in vitro and in vivo antitumor activity against human multiple myeloma.

Leukemia, 2013. 28 : p. 917.

238. Chang, Y. - H., et al., A Chimeric Receptor with NKG2D Specificity Enhances Natural Killer

Cell Activation and Killing of Tumor Cells. Cancer Research, 2013. 73 (6): p. 1777 - 1786.

201

239. Boissel, L., et al., Comparison of mRNA and lentiviral based transfection of natural killer

cells with chimeric antigen receptors recognizing lymphoid antigens. Leukemia &

Lymphoma, 2012. 53 (5): p. 958 - 965.

240. Gleason, M.K., et al., Bispecific and Trispecific Killer Cell Engagers Directly Activate

Human NK Cells through CD16 Signaling and Induce Cytotoxicity and Cytokine

Production. Molecular Cancer Therapeutics, 2012. 11 (12): p. 2674 - 2684.

241. Felices, M., et al., Generation of BiKEs and TriKEs to Improve NK Cell - Mediated Targeting

of Tumor Cells , in Natural Killer Cells: Methods and Prot ocols , S.S. Somanchi, Editor.

2016, Springer New York: New York, NY. p. 333 - 346.

242. Mayor, P., K. Starbuck, and E. Zsiros, Adoptive cell transfer using autologous tumor

infiltrating lymphocytes in gynecologic malignancies. Gynecologic Oncology, 2018.

243 . Knuth, A., et al., Cytolytic T - cell clones against an autologous human melanoma:

specificity study and definition of three antigens by immunoselection. Proceedings of the

National Academy of Sciences, 1989. 86 (8): p. 2804 - 2808.

244. Inaba, K., et al., De ndritic Cells as Antigen Presenting Cells in Vivo. International Reviews

of Immunology, 1990. 6 (2 - 3): p. 197 - 206.

245. Davis, I.D., et al., Rational approaches to human cancer immunotherapy. Journal of

Leukocyte Biology, 2003. 73 (1): p. 3 - 29.

246. Jung, N. C., et al., Dendritic Cell - Based Immunotherapy for Solid Tumors. Transl Oncol,

2018. 11 (3): p. 686 - 690.

247. Steinman, R.M. and K. Inaba, Stimulation of the primary mixed leukocyte reaction. Crit

Rev Immunol, 1985. 5 (4): p. 331 - 48.

248. M., S.R., et al., T he Sensitization Phase of T -Cell-Mediated Immunitya. Annals of the New

York Academy of Sciences, 1988. 546 (1): p. 80 - 90.

202

249. Rosenberg, S.A., et al., Durable Complete Responses in Heavily Pretreated Patients with

Metastatic Melanoma Using T - Cell Transfer Immunotherapy. Clinical Cancer Research,

2011. 17 (13): p. 4550 - 4557.

250. Fujita, K., et al., Prolonged disease - free period in patients with advanced epithelial

ovarian cancer after adoptive transfer of tumor - infiltrating lymphocytes. Clinical Cancer

Resea rch, 1995. 1 (5): p. 501 - 507.

251. Stevanović, S., et al., Complete Regression of Metastatic Cervical Cancer After Treatment

With Human Papillomavirus – Targeted Tumor - Infiltrating T Cells. Journal of Clinical

Oncology, 2015. 33 (14): p. 1543 - 1550.

252. Anders en, R., et al., Tumor infiltrating lymphocyte therapy for ovarian cancer and renal

cell carcinoma. Human Vaccines & Immunotherapeutics, 2015. 11 (12): p. 2790 - 2795.

253. Ma, W., et al., T Cell – Associated Immunotherapy for Hepatocellular Carcinoma. Cellular

Physiology and Biochemistry, 2017. 41 (2): p. 609 - 622.

254. Grimm, E.A., et al., Lymphokine - activated killer cell phenomenon. Lysis of natural killer -

resistant fresh solid tumor cells by interleukin 2 - activated autologous human peripheral

blood lymphocytes. The Journal of Experimental Medicine, 1982. 155 (6): p. 1823 - 1841.

255. Bertelli, R., et al., Endolymphatic Immunotherapy in Inoperable Hepatocellular

Carcinoma. Transplantation Proceedings, 2008. 40 (6): p. 1913 - 1915.

256. Jiang, J., C. Wu, and B. Lu, Cyto kine - induced killer cells promote antitumor immunity.

Journal of Translational Medicine, 2013. 11 : p. 83 - 83.

257. Wang, F. - S., et al., Antitumor activities of human autologous cytokine - induced killer (CIK)

cells against hepatocellular carcinoma cells in vitro and in vivo. World Journal of

Gastroenterology, 2002. 8 (3): p. 464 - 468.

203

258. Introna, M., et al., Repeated infusions of donor - derived cytokine - induced killer cells in

patients relapsing after allogeneic stem cell transplantation: a phase I study.

Hae matologica, 2007. 92 (7): p. 952 - 959.

259. JIANG, J., et al., Treatment of Advanced Gastric Cancer by Chemotherapy Combined with

Autologous Cytokine - induced Killer Cells. Anticancer Research, 2006. 26 (3B): p. 2237 -

2242.

260. Corrigan - Curay, J., et al., T - Ce ll Immunotherapy: Looking Forward: T Cell

Immunotherapy: Optimizing Trial Design Bethesda, Maryland 10 - 11 September 2013.

Molecular Therapy, 2014. 22 (9): p. 1564 - 1574.

261. Robbins, P.F., et al., Tumor Regression in Patients With Metastatic Synovial Cell S arcoma

and Melanoma Using Genetically Engineered Lymphocytes Reactive With NY - ESO - 1.

Journal of Clinical Oncology, 2011. 29 (7): p. 917 - 924.

262. Murad, J.M., D.J. Graber, and C.L. Sentman, Advances in the use of natural receptor - or

ligand - based chimeric a ntigen receptors (CARs) in haematologic malignancies. Best

Practice & Research Clinical Haematology, 2018. 31 (2): p. 176 - 183.

263. Halim, L., A. Ajina, and J. Maher, Pre - clinical development of chimeric antigen receptor T -

cell immunotherapy: Implications o f design for efficacy and safety. Best Practice &

Research Clinical Haematology, 2018. 31 (2): p. 117 - 125.

264. Sentman, M. - L., et al., Mechanisms of Acute Toxicity in NKG2D Chimeric Antigen

Receptor T Cell – Treated Mice. The Journal of Immunology, 2016. 197 (12): p. 4674 - 4685.

265. Brentjens, R.J., et al., CD19 - Targeted T Cells Rapidly Induce Molecular Remissions in

Adults with Chemotherapy - Refractory Acute Lymphoblastic Leukemia. Science

Translational Medicine, 2013. 5 (177): p. 177ra38 - 177ra38.

204

266. Kochende rfer, J.N., et al., B - cell depletion and remissions of malignancy along with

cytokine - associated toxicity in a clinical trial of anti - CD19 chimeric - antigen - receptor –

transduced T cells. Blood, 2012. 119 (12): p. 2709 - 2720.

267. Davila, M.L., et al., How do CARs work?: Early insights from recent clinical studies

targeting CD19. Oncoimmunology, 2012. 1 (9): p. 1577 - 1583.

268. VanSeggelen, H., et al., T Cells Engineered With Chimeric Antigen Receptors Targeting

NKG2D Ligands Display Lethal Toxicity in Mice. Mole cular Therapy, 2015. 23 (10): p.

1600 - 1610.

269. Fedorov, V.D., M. Themeli, and M. Sadelain, PD - 1 – and CTLA - 4 – Based Inhibitory

Chimeric Antigen Receptors (iCARs) Divert Off - Target Immunotherapy Responses.

Science Translational Medicine, 2013. 5 (215): p. 215 ra172 - 215ra172.

270. Wolf, E., et al., BiTEs: bispecific antibody constructs with unique anti - tumor activity. Drug

Discovery Today, 2005. 10 (18): p. 1237 - 1244.

271. Baeuerle, P.A., P. Kufer, and R. Bargou, BiTE: Teaching antibodies to engage T - cells for

ca ncer therapy. Curr Opin Mol Ther, 2009. 11 (1): p. 22 - 30.

272. Nagorsen, D. and P.A. Baeuerle, Immunomodulatory therapy of cancer with T cell -

engaging BiTE antibody blinatumomab. Experimental Cell Research, 2011. 317 (9): p.

1255 - 1260.

273. Kantarjian, H., e t al., Blinatumomab versus Chemotherapy for Advanced Acute

Lymphoblastic Leukemia. New England Journal of Medicine, 2017. 376 (9): p. 836 - 847.

274. Chiang, C.L. - L., L.E. Kandalaft, and G. Coukos, Adjuvants for Enhancing the

Immunogenicity of Whole Tumor Cel l Vaccines. International Reviews of Immunology,

2011. 30 (2 - 3): p. 150 - 182.

205

275. Palena, C. and J. Schlom, Vaccines against Human Carcinomas: Strategies to Improve

Antitumor Immune Responses. Journal of Biomedicine and Biotechnology, 2010. 2010 : p.

380697.

276. Baxevanis, C.N., S.A. Perez, and M. Papamichail, Cancer immunotherapy. Critical Reviews

in Clinical Laboratory Sciences, 2009. 46 (4): p. 167 - 189.

277. Fyfe, G., et al., Results of treatment of 255 patients with metastatic renal cell carcinoma

who rec eived high - dose recombinant interleukin - 2 therapy. Journal of Clinical Oncology,

1995. 13 (3): p. 688 - 696.

278. Atkins, M.B., et al., High - Dose Recombinant Interleukin 2 Therapy for Patients With

Metastatic Melanoma: Analysis of 270 Patients Treated Between 1985 and 1993. Journal

of Clinical Oncology, 1999. 17 (7): p. 2105 - 2105.

279. Conlon, K.C., et al., Redistribution, Hyperproliferation, Activation of Natural Killer Cells

and CD8 T Cells, and Cytokine Production During First - in - Human Clinical Trial of

Reco mbinant Human Interleukin - 15 in Patients With Cancer. Journal of Clinical

Oncology, 2015. 33 (1): p. 74 - 82.

280. Pérez - Martínez, A., et al., A phase I/II trial of interleukin - 15 – stimulated natural killer cell

infusion after haplo - identical stem cell transplantation for pediatric refractory solid

tumors. Cytotherapy, 2015. 17 (11): p. 1594 - 1603.

281. Wrangle, J.M., et al., ALT - 803, an IL - 15 superagonist, in combination with nivolumab in

patients with metastatic non - small cell lung cancer: a non - randomis ed, open - label,

phase 1b trial. The Lancet Oncology, 2018. 19 (5): p. 694 - 704.

282. Natsume, A., R. Niwa, and M. Satoh, Improving effector functions of antibodies for

cancer treatment: Enhancing ADCC and CDC. Drug design, development and therapy,

2009. 3 : p . 7 - 16.

206

283. Weng, W. - K. and R. Levy, Two Immunoglobulin G Fragment C Receptor Polymorphisms

Independently Predict Response to Rituximab in Patients With Follicular Lymphoma.

Journal of Clinical Oncology, 2003. 21 (21): p. 3940 - 3947.

284. Gennari, R., et al ., Pilot Study of the Mechanism of Action of Preoperative Trastuzumab

in Patients with Primary Operable Breast Tumors Overexpressing HER2. Clinical Cancer

Research, 2004. 10 (17): p. 5650 - 5655.

285. McLaughlin, P., et al., Rituximab chimeric anti - CD20 monoc lonal antibody therapy for

relapsed indolent lymphoma: half of patients respond to a four - dose treatment program.

Journal of Clinical Oncology, 1998. 16 (8): p. 2825 - 2833.

286. Stenner, F. and C. Renner, Cancer Immunotherapy and the Immune Response in

Folli cular Lymphoma. Frontiers in Oncology, 2018. 8 (219).

287. Marcus, R., et al., Obinutuzumab for the First - Line Treatment of Follicular Lymphoma.

New England Journal of Medicine, 2017. 377 (14): p. 1331 - 1344.

288. Larionov, A.A., Current Therapies for Human E pidermal Growth Factor Receptor 2 -

Positive Metastatic Breast Cancer Patients. Frontiers in Oncology, 2018. 8 (89).

289. Slamon, D.J., et al., Use of Chemotherapy plus a Monoclonal Antibody against HER2 for

Metastatic Breast Cancer That Overexpresses HER2. N ew England Journal of Medicine,

2001. 344 (11): p. 783 - 792.

290. Vey, N., et al., A phase 1 trial of the anti - inhibitory KIR mAb IPH2101 for AML in complete

remission. Blood, 2012. 120 (22): p. 4317 - 4323.

291. Korde, N., et al., A phase II trial of pan - KIR2D blockade with IPH2101 in smoldering

multiple myeloma. Haematologica, 2014. 99 (6): p. e81 - e83.

292. Marie - Cardine, A., et al., IPH4102, a Humanized KIR3DL2 Antibody with Potent Activity

against Cutaneous T - cell Lymphoma. Cancer Research, 2014. 74 (21): p. 6 060 - 6070.

207

293. Veuillen, C., et al., Primary B - CLL Resistance to NK Cell Cytotoxicity can be Overcome In

Vitro and In Vivo by Priming NK Cells and Monoclonal Antibody Therapy. Journal of

Clinical Immunology, 2012. 32 (3): p. 632 - 646.

294. McWilliams, E.M., et al., Therapeutic CD94/NKG2A blockade improves natural killer cell

dysfunction in chronic lymphocytic leukemia. Oncoimmunology, 2016. 5 (10): p.

e1226720.

295. Thornthwaite, J.T., et al., The Natural Killer Cell: A Historical Perspective and the Use of

Supplements to Enhance NKC Activity. Journal of Immune Based Therapies, Vaccines and

Antimicrobials, 2012. Vol.01No.03 : p. 31.

296. Lanier, L.L., NK CELL RECOGNITION. Annual Review of Immunology, 2005. 23 (1): p. 225 -

274.

297. Cantoni, C., e t al., NK Cells, Tumor Cell Transition, and Tumor Progression in Solid

Malignancies: New Hints for NK - Based Immunotherapy? Journal of Immunology

Research, 2016. 2016 : p. 13.

298. Coca, S., et al., The prognostic significance of intratumoral natural killer cells in patients

with colorectal carcinoma. Cancer, 1997. 79 (12): p. 2320 - 2328.

299. Schantz, S.P., B.H. Campbell, and O.M. Guillamondegui, Pharyngeal carcinoma and

natural killer cell activity. The American Journal of Surgery, 1986. 152 (4): p. 467 - 474.

3 00. Donadon, M., et al., Increased Infiltration of Natural Killer and T Cells in Colorectal Liver

Metastases Improves Patient Overall Survival. Journal of Gastrointestinal Surgery, 2017.

21 (8): p. 1226 - 1236.

301. Xie, S., et al., Allogenic natural killer c ell immunotherapy of sizeable ovarian cancer: A

case report. Molecular and Clinical Oncology, 2017. 6 (6): p. 903 - 906.

208

302. Palmer, J.M., et al., Clinical relevance of natural killer cells following hematopoietic stem

cell transplantation. J Cancer, 2013. 4 (1): p. 25 - 35.

303. Wang, W. - T., et al., Elevated absolute NK cell counts in peripheral blood predict good

prognosis in chronic lymphocytic leukemia. Journal of Cancer Research and Clinical

Oncology, 2018. 144 (3): p. 449 - 457.

304. Dahlberg, C.I.M., et al., Natural Killer Cell - Based Therapies Targeting Cancer: Possible

Strategies to Gain and Sustain Anti - Tumor Activity. Frontiers in Immunology, 2015. 6 : p.

605.

305. Kean, L., et al., NK Cells Efficiently Prevent Engraftment of Donor Stem Cells after

Tolerige nic Bone Marrow Transplantation. Blood, 2005. 106 (11): p. 3025 - 3025.

306.

administered to cancer patients a phase I clinical s.pdf>.

307. Koehl, U., et al., Clinical Grade Purification and Expansion of NK Cell Products for an

Optimized Manufacturing Protocol. Frontiers in Oncology, 2013. 3 (118).

308. Wagner, J., et al., A Two - Phase Expansion Protocol Combining Interleukin (IL) - 15 and IL -

21 Improves Natural Killer Cell Proli feration and Cytotoxicity against

Rhabdomyosarcoma. Frontiers in Immunology, 2017. 8 : p. 676.

309. Granzin, M., et al., Shaping of Natural Killer Cell Antitumor Activity by Ex Vivo

Cultivation. Frontiers in Immunology, 2017. 8 (458).

310. Bae, D.S. and J.K. Lee, Development of NK cell expansion methods using feeder cells from

human myelogenous leukemia cell line. Blood Res, 2014. 49 (3): p. 154 - 61.

311. Fujisaki, H., et al., Expansion of highly cytotoxic human natural killer cells for cancer cell

therapy. Can cer Res, 2009. 69 (9): p. 4010 - 7.

209

312. Cheng, M., et al., NK cell - based immunotherapy for malignant diseases. Cell Mol

Immunol, 2013. 10 (3): p. 230 - 52.

313. Ahn, Y. - O., et al., Irradiated and Activated Autologous PBMCs Induce Expansion of Highly

Cytotoxic H uman NK Cells In Vitro. Journal of Immunotherapy, 2013. 36 (7): p. 373 - 381.

314. Granzin, M., et al., Highly efficient IL - 21 and feeder cell - driven ex vivo expansion of

human NK cells with therapeutic activity in a xenograft mouse model of melanoma.

OncoImm unology, 2016. 5 (9): p. e1219007.

315. Denman, C.J., et al., Membrane - bound IL - 21 promotes sustained ex vivo proliferation of

human natural killer cells. PLoS One, 2012. 7 (1): p. e30264.

316. F. Torelli, G., Use of Natural Killer Cells for Acute Leukemia P atients. International

Journal of Hematology Research, 2015. 1 (2): p. 57 - 60.

317. Klingemann, H., L. Boissel, and F. Toneguzzo, Natural Killer Cells for Immunotherapy –

Advantages of the NK - 92 Cell Line over Blood NK Cells. Frontiers in Immunology, 2016. 7 :

p. 91.

318. Zhang, C., et al., Chimeric Antigen Receptor - Engineered NK - 92 Cells: An Off - the - Shelf

Cellular Therapeutic for Targeted Elimination of Cancer Cells and Induction of Protective

Antitumor Immunity. Frontiers in Immunology, 2017. 8 : p. 533.

319. Whang - Peng, J., et al., Cytogenetic characterization of selected small round cell tumors

of childhood. Cancer Genetics and Cytogenetics, 1986. 21 (3): p. 185 - 208.

320. Sharma, R. and A. Das, IL - 2 mediates NK cell proliferation but not hyperactivity.

Immuno logic Research, 2018. 66 (1): p. 151 - 157.

321. London, L., B. Perussia, and G. Trinchieri, Induction of proliferation in vitro of resting

human natural killer cells: expression of surface activation antigens. The Journal of

Immunology, 1985. 134 (2): p. 718 - 727.

210

322. Shimasaki, N., et al., Expanded and armed natural killer cells for cancer treatment.

Cytotherapy, 2016. 18 (11): p. 1422 - 1434.

323. Berg, M., et al., Clinical Grade Ex Vivo - Expanded Human Natural Killer Cells Upregulate

Activating Receptors an d Death Receptor Ligands and Have Enhanced Cytolytic Activity

against Tumor Cells. Cytotherapy, 2009. 11 (3): p. 341 - 355.

324. Sallman, D.A. and J.Y. Djeu, Chapter 5 - Immunological Sculpting: Natural Killer Cell

Receptors and Ligands A2 - Prendergast, Geor ge C , in Cancer Immunotherapy , E.M.

Jaffee, Editor. 2007, Academic Press: Burlington. p. 63 - 81.

325. Glasner, A., et al., Recognition and Prevention of Tumor Metastasis by the NK Receptor

NKp46/NCR1. The Journal of Immunology, 2012. 188 (6): p. 2509 - 2515.

326. Vitale, M., et al., NKp44, a Novel Triggering Surface Molecule Specifically Expressed by

Activated Natural Killer Cells, Is Involved in Non – Major Histocompatibility Complex –

restricted Tumor Cell Lysis. The Journal of Experimental Medicine, 1998. 187 (1 2): p.

2065 - 2072.

327. Béziat, V., et al., NK Cell Terminal Differentiation: Correlated Stepwise Decrease of

NKG2A and Acquisition of KIRs. PLOS ONE, 2010. 5 (8): p. e11966.

328. Voskens, C.J., et al., Ex - vivo expanded human NK cells express activating rece ptors that

mediate cytotoxicity of allogeneic and autologous cancer cell lines by direct recognition

and antibody directed cellular cytotoxicity. Journal of Experimental & Clinical Cancer

Research, 2010. 29 (1): p. 134.

329. Paust, S., et al., Critical role for the chemokine receptor CXCR6 in NK cell – mediated

antigen - specific memory of haptens and viruses. Nature Immunology, 2010. 11 : p. 1127.

330. Lapteva, N., et al., Large - scale ex vivo expansion and characterization of natural killer

cells for clinical ap plications. Cytotherapy, 2012. 14 (9): p. 1131 - 1143.

211

331. Heyking, K., et al., The endochondral bone protein CHM1 sustains an undifferentiated,

invasive phenotype, promoting lung metastasis in Ewing sarcoma. Molecular Oncology,

2017. 11 (9): p. 1288 - 1301.

33 2. Ferlazzo, G. and P. Carrega, Natural killer cell distribution and trafficking in human

tissues. Frontiers in Immunology, 2012. 3 (347).

333. Ruggeri, L., et al., Effectiveness of donor natural killer cell alloreactivity in mismatched

hematopoietic transp lants. Science, 2002. 295 (5562): p. 2097 - 100.

334. Giebel, S., et al., Survival advantage with KIR ligand incompatibility in hematopoietic

stem cell transplantation from unrelated donors. Blood, 2003. 102 (3): p. 814 - 9.

335. Ruggeri, L., et al., Role of nat ural killer cell alloreactivity in HLA - mismatched

hematopoietic stem cell transplantation. Blood, 1999. 94 (1): p. 333 - 9.

336. Chiu, J., D.M. Ernst, and A. Keating, Acquired Natural Killer Cell Dysfunction in the Tumor

Microenvironment of Classic Hodgkin Ly mphoma. Frontiers in Immunology, 2018.

9 (267).

337. Otegbeye, F., et al., Inhibiting TGF - beta signaling preserves the function of highly

activated, in vitro expanded natural killer cells in AML and colon cancer models. PLOS

ONE, 2018. 13 (1): p. e0191358.

3 38. Lagrue, K., et al., Lenalidomide augments actin remodeling and lowers NK - cell activation

thresholds. Blood, 2015. 126 (1): p. 50 - 60.

339. Wang, R., et al., The transcription factor Myc controls metabolic reprogramming upon T

lymphocyte activation. Immun ity, 2011. 35 (6): p. 871 - 882.

340. Spolski, R. and W.J. Leonard, Interleukin - 21: Basic Biology and Implications for Cancer

and Autoimmunity. Annual Review of Immunology, 2008. 26 (1): p. 57 - 79.

212

341. Dang, C.V., MYC, Metabolism, Cell Growth, and Tumorigenesi s. Cold Spring Harbor

Perspectives in Medicine, 2013. 3 (8): p. a014217.

342. Pearce, E.L. and E.J. Pearce, Metabolic pathways in immune cell activation and

quiescence. Immunity, 2013. 38 (4): p. 633 - 43.

343. Marcais, A., et al., The metabolic checkpoint kin ase mTOR is essential for IL - 15 signaling

during the development and activation of NK cells. Nat Immunol, 2014. 15 (8): p. 749 -

757.

344. Geller, M.A., et al., A phase II study of allogeneic natural killer cell therapy to treat

patients with recurrent ovaria n and breast cancer. Cytotherapy, 2011. 13 (1): p. 98 - 107.

345. Duck, C., et al., Expansion and Activation of Natural Killer Cells for Cancer

Immunotherapy. Korean J Lab Med, 2009. 29 (2): p. 89 - 96.

346. Childs, R.W. and M. Berg, Bringing natural killer cell s to the clinic: ex vivo manipulation.

ASH Education Program Book, 2013. 2013 (1): p. 234 - 246.

347. Somanchi, S.S., et al., Expansion, Purification, and Functional Assessment of Human

Peripheral Blood NK Cells. Journal of Visualized Experiments : JoVE, 2011 (48): p. 2540.

348. Bierie, B. and H.L. Moses, TGFβ: the molecular Jekyll and Hyde of cancer. Nature Reviews

Cancer, 2006. 6 : p. 506.

349. Bellone, G., et al., Regulation of NK cell functions by TGF - beta 1. The Journal of

Immunology, 1995. 155 (3): p. 1066 - 1073.

350. Bergmann, L., et al., The inhibition of lymphokine - activated killer cells in acute

myeloblastic leukemia is mediated by transforming growth factor - β1 . Vol. 23. 1996.

1574 - 80.

351. Konjevic, G., et al., Investigation of NK cell function and their modulation in different

malignancies. Immunologic Research, 2012. 52 (1): p. 139 - 156.

213

352. Lee, H.m., K. - S. Kim, and J. Kim, A comparative study of the effects of inhibitory cytokines

on human natural killer cells and the mechanistic features of trans forming growth

factor - beta. Cellular Immunology, 2014. 290 (1): p. 52 - 61.

353. Baginska, J., et al., The Critical Role of the Tumor Microenvironment in Shaping Natural

Killer Cell - Mediated Anti - Tumor Immunity. Frontiers in Immunology, 2013. 4 (490).

354. Lee , J. - C., et al., Elevated TGF - β1 Secretion and Down - Modulation of NKG2D Underlies

Impaired NK Cytotoxicity in Cancer Patients. The Journal of Immunology, 2004. 172 (12):

p. 7335 - 7340.

355. Shin, N., et al., Significance of transforming growth factor β1 as a new tumor marker for

colorectal cancer. International Journal of Cancer, 2002. 97 (4): p. 508 - 511.

356. Akiyama, T., et al., Involvement of transforming growth factor - β and thrombopoietin in

the pathogenesis of myelodysplastic syndrome with myelofibrosis. Leukemia, 2005. 19 :

p. 1558.

357. Dong, M. and G.C. Blobe, Role of transforming growth factor - β in hematologic

malignancies. Blood, 2006. 107 (12): p. 4589 - 4596.

358. Yasuo, H., et al., Thrombopoietin concentrations in peripheral blood correlated with

plate let numbers in two patients with thrombocytopenia by chronic graft -versus-host

disease. American Journal of Hematology, 2003. 73 (4): p. 285 - 289.

359. Trotta, R., et al., TGF - β Utilizes SMAD3 to Inhibit CD16 - Mediated IFN - γ Production and

Antibody - Dependent Cellular Cytotoxicity in Human NK Cells. The Journal of

Immunology, 2008. 181 (6): p. 3784 - 3792.

360. Dasgupta, S., et al., Inhibition of NK Cell Activity through TGF - β1 by Down - Regulation of

NKG2D in a Murine Model of Head and Neck Cancer. The Journal of I mmunology, 2005.

175 (8): p. 5541 - 5550.

214

361. Yang, B., et al., Blocking transforming growth factor - β signaling pathway augments

antitumor effect of adoptive NK - 92 cell therapy. International Immunopharmacology,

2013. 17 (2): p. 198 - 204.

362. Yoon, J.H., et a l., Activin receptor -like kinase5 inhibition suppresses mouse melanoma by

ubiquitin degradation of Smad4, thereby derepressing eomesodermin in cytotoxic T

lymphocytes. EMBO Molecular Medicine, 2013. 5 (11): p. 1720 - 1739.

363. Hong, C. - S., et al., Plasma Exo somes as Markers of Therapeutic Response in Patients with

Acute Myeloid Leukemia. Frontiers in Immunology, 2014. 5 : p. 160.

364. Herbertz, S., et al., Clinical development of galunisertib (LY2157299 monohydrate), a

small molecule inhibitor of transforming growth factor - beta signaling pathway. Drug

Des Devel Ther, 2015. 9 : p. 4479 - 99.

365. Chul -Yong, P., K. Dae-Kee, and S.Y. Yhong, EW -7203, a novel small molecule inhibitor of

transforming growth factor - β (TGF-β) type I receptor/activin receptor-like kinase-5 ,

blocks TGF - β1-mediated epithelial-to-mesenchymal transition in mammary epithelial

cells. Cancer Science, 2011. 102 (10): p. 1889 - 1896.

366. Connolly, E.C., J. Freimuth, and R.J. Akhurst, Complexities of TGF - β Targeted Cancer

Therapy. International Journal of Biological Sciences, 2012. 8 (7): p. 964 - 978.

367.

function implication in growth, expansion and differe.pdf>.

368. Marcoe, J.P., et al., TGF - beta is responsible for NK cell immaturity during ontogeny and

increased susceptibility to infection during mouse infancy. Nat Immunol, 2012. 13 (9): p.

843 - 50.

369. Viel, S., et al., TGF - β inhibits the activation and functions of NK cells by repressing the

mTOR pathway. Science Sign aling, 2016. 9 (415): p. ra19 - ra19.

215

370. Wilson, E.B., et al., Human tumour immune evasion via TGF - beta blocks NK cell

activation but not survival allowing therapeutic restoration of anti - tumour activity. PLoS

One, 2011. 6 (9): p. e22842.

371. Feng, W., et al., TIGIT expression levels on human NK cells correlate with functional

heterogeneity among healthy individuals. European Journal of Immunology, 2015.

45 (10): p. 2886 - 2897.

372. Luo, Q., et al., Elevated expression of TIGIT on CD3+CD4+ T cell s correlates with disease

activity in systemic lupus erythematosus. Allergy, Asthma & Clinical Immunology, 2017.

13 (1): p. 15.

373. Manieri, N.A., E.Y. Chiang, and J.L. Grogan, TIGIT: A Key Inhibitor of the Cancer Immunity

Cycle. Trends in Immunology, 2017 . 38 (1): p. 20 - 28.

374. Stanietsky, N., et al., The interaction of TIGIT with PVR and PVRL2 inhibits human NK cell

cytotoxicity. Proceedings of the National Academy of Sciences, 2009. 106 (42): p. 17858 -

17863.

375. Faivre, S.J., et al., A phase 2 study of galunisertib, a novel transforming growth factor -

beta (TGF - β) receptor I kinase inhibitor, in patients with advanced hepatocellular

carcinoma (HCC) and low serum alpha fetoprotein (AFP). Journal of Clinical Oncology,

2016. 34 (15_suppl): p. 4070 - 4070.

376. Ott, P.A., et al., Combination immunotherapy: a road map. Journal for ImmunoTherapy

of Cancer, 2017. 5 (1): p. 16.

377. Aston, W.J., et al., A systematic investigation of the maximum tolerated dose of cytotoxic

chemotherapy with and without supportive care in mice. BMC Cancer, 2017. 17 : p. 684.

216

378. Karsten, M., et al., Depletion of CD4+CD25+ human regulatory T cells in vivo: Kinetics of

Treg depletion and alterations in immune functions in vivo and in vitro. International

Journal of Cancer, 2007. 120 (12): p . 2723 - 2733.

379. Oelsner, S., et al., Continuously expanding CAR NK - 92 cells display selective cytotoxicity

against B - cell leukemia and lymphoma. Cytotherapy, 2017. 19 (2): p. 235 - 249.

380. Somanchi, S.S., et al., Engineering lymph node homing of ex vivo – e xpanded human

natural killer cells via trogocytosis of the chemokine receptor CCR7. Blood, 2012.

119 (22): p. 5164 - 5172.

381. Murray, S. and A. Lundqvist, Targeting the tumor microenvironment to improve natural

killer cell - based immunotherapies: On being in the right place at the right time, with

resilience. Human Vaccines & Immunotherapeutics, 2016. 12 (3): p. 607 - 611.

382. Ponzetta, A., et al., Multiple Myeloma Impairs Bone Marrow Localization of Effector

Natural Killer Cells by Altering the Chemokine Micro environment. Cancer Research,

2015. 75 (22): p. 4766 - 4777.

383. Swati, B., et al., Mutational inactivation of TGFBR2 in microsatellite unstable colon

cancer arises from the cooperation of genomic instability and the clonal outgrowth of

transforming growth f actor β resistant cells. Genes, Chromosomes and Cancer, 2008.

47 (2): p. 95 - 106.

384. de Gramont, A., S. Faivre, and E. Raymond, Novel TGF - β inhibitors ready for prime time

in onco - immunology. OncoImmunology, 2017. 6 (1): p. e1257453.

385. Giaccone, G., et a l., A phase III study of belagenpumatucel - L, an allogeneic tumour cell

vaccine, as maintenance therapy for non - small cell lung cancer. European Journal of

Cancer, 2015. 51 (16): p. 2321 - 2329.

217

386. Brandes, A.A., et al., A Phase II randomized study of galuni sertib monotherapy or

galunisertib plus lomustine compared with lomustine monotherapy in patients with

recurrent glioblastoma. Neuro - Oncology, 2016. 18 (8): p. 1146 - 1156.

387. Kaufman, H.L., Rational Combination Immunotherapy: Understand the Biology. Cancer

Immunology Research, 2017. 5 (5): p. 355 - 356.

218