Exploration of Viral and Host Gene Expression Dynamics in Hematologic Malignancies

Abstract

Cancer encompasses a diverse set of diseases, characterized by genomic damage, with varied causes and heterogeneous disease courses. Despite advancements in treatment, outcomes remain suboptimal for many cancers. This thesis focuses on viral drivers of cancer and leverages expression data to interrogate blood cancer biology.Microbial drivers can play a significant role in cancer, particularly blood cancers, where numerous pathogen associations have been established. Next-generation sequencing (NGS) offers several advantages over standard clinical methods for pathogen detection, including high-throughput capabilities, integration with host genomics, and the ability to quantify both RNA and DNA. Here, an NGS-based classifier for Epstein-Barr Virus (EBV) is shown to be highly concordant with clinical detection methods. It is also demonstrated that NGS-based capture of RNA for select EBV genes, known to play crucial roles in tumorigenesis, can be quantified to elucidate EBV expression patterns in tumor biopsy samples.While DNA mutation status provides discrete information, RNA captures the aggregated effects of the mutational burden in a particular tumor. The information-rich nature of RNA is only useful if it is interpretable in the context of the disease or question being explored. This is often achieved through pathway analysis, a collection of statistical methods used to link gene sets to biological processes, enabling researchers to interpret high-throughput sequencing data more effectively. This work demonstrates how expression data and a framework for pathway analysis can be used to understand and predict patient response to therapy in a set of 111 relapsed diffuse large B-cell lymphoma (DLBCL) patients treated with chimeric antigen receptor T-cell (CAR-T) therapy.Finally, EBV and host expression profiling are integrated with pathway analysis methods to study extranodal NK/T-cell lymphoma (ENKTCL), a rare non-Hodgkin lymphoma with a survival rate of 30-40%. A diverse patient cohort, spanning a broader range of backgrounds than previous studies, reveals that high European ancestry is associated with significantly better survival. In addition to ancestry, prognostic differences in the tumor microenvironment, specifically cytotoxic T cell infiltration, are identified. These survival differences are linked to variations in mutational landscape, host expression, tumor microenvironment, and EBV expression, offering new insights into the biology of ENKTCL.