Merkel Cell Carcinoma: Immunogenicity and the characterization of CD4 T cell responses to the Merkel cell polyomavirus

Abstract

Merkel cell carcinoma (MCC) is a deadly, virus-associated skin cancer with a 5-year relative mortality rate of 46%. The Merkel cell polyomavirus (MCPyV) is clonally integrated into 80% of MCCs and persistent expression of MCPyV T-antigen oncoproteins is required for tumor survival and growth, potentially providing ideal targets for immune based therapies. In the remaining 20% of MCCs that are virus-negative, remarkably high numbers of UV-induced neoantigens are detected, suggesting that both MCC subsets harbor immunogenic epitopes. Over the last few years, this hypothesis has been strongly supported by extraordinarily high response rates to agents blocking the PD-1 pathway in patients with both virus-positive and virus-negative MCC. However, still roughly half of patients do not benefit from these modalities, indicating an urgent need to identify biomarkers predictive of response and immune evasion mechanisms that underlie PD-1 blockade resistance. While much of the work presented here was initiated and/or completed prior to the use of these novel therapies, these data provide the basis for ongoing efforts to delineate predictors and mechanisms of resistance to PD-1 blockade therapy. Within the opening chapters, we explore pathogen-driven cancers more broadly (Chapter 1) before delving specifically into MCPyV-induced MCC, its rising incidence rate (Chapter 2) and known mechanisms of immune evasion (Chapter 3). Previous studies have indicated that a robust CD8 T cell response is associated with dramatically improved MCC outcomes, therefore, we sought to characterize several mechanisms of CD8 T cell dysfunction. The first is described in Chapter 4 in which we show that the downregulation of the adhesion molecule E-selectin within MCC tumor vasculature is associated with intratumoral T cell exclusion and reduced survival. However, even if CD8 T cells can infiltrate tumors, there is abundant literature to indicate that effective CD8 T cell responses require CD4 help and that this ‘help' is often impaired in the setting of cancer. Consequently, Chapters 5-8 focus upon elucidation of the CD4 helper T cell response against MCC. Specifically, in Chapter 5 we discuss the multitude of CD4 subtypes that have been described and their relevance in the setting of cancer and cancer therapies. In order to elucidate the phenotype and function of MCPyV-specific CD4 T cells in the context of MCC, we needed to first identify CD4 T cell epitopes within MCPyV and develop reagents enabling their isolation. This work is the focus of Chapter 6. In Chapter 7 we examine an especially fascinating, newly identified CD4 epitope ‘WEDLFCDESLSSPEPPSSSE’ locating within the MCPyV Large T-antigen. This epitope is highly immunogenic and has several key features which make it an ideal target for immune-based therapies such as a therapeutic cancer vaccine. Discovery of this epitope also resulted in the generation of HLA class-II tetramers allowing for the first time isolation of MCPyV-specific CD4 T cells directly ex vivo without antigenic stimulation. However, in many patients the frequency of these cells was found to be below the limit of the detection by standard methods. As a result, in Chapter 8 we describe the development of a novel method using a digital scanning microscope to specifically and sensitively identify rare antigen-specific T cells. Finally, in Chapter 9, we shift away from the CD4 T cell describe a unique subset of MCC patients who present without a detectable primary skin lesion and who have a remarkable 50% higher rate of survival as compared to stage-matched patients with primary skin lesions. These patients have several elevated markers of immunity suggesting that clearance of the primary skin lesion is immune-mediate. This past year (2017) historically marked the first FDA approval of an agent for the treatment of advanced MCC. Therefore, as we continue to treat more MCC patients with this agent (avelumab; anti-PD-L1) and other immune checkpoint inhibitors, the findings described in this dissertation will allow us to evaluate potential biomarkers of response and resistance including E-selectin downregulation and evaluation of CD4 T cell phenotype and function. For patients who do not respond to PD-1 blockade, these studies will help inform the use of existing therapies in potentially novel combinations and support the development of new approaches, such as a therapeutic cancer vaccine. Ultimately, we believe that these efforts will translate to improving patient outcomes.