Eliezer M. Van Allen - US grants

DESCRIPTION (provided by applicant): For metastatic castration resistant prostate cancer, recent advances have led to the deployment of second- generation ADT (ADT2) therapies, including abiraterone acetate (AA), which targets a component of androgen biosynthesis, and enzalutamide, which targets the androgen receptor directly. Both AA and enzalutamide have demonstrated an overall survival benefit in patients with metastatic CRPC; however, most patients still develop resistance to these agents, which drives prostate cancer-associated morbidity and mortality. Several mechanisms of resistance to ADT2 have recently been identified, although the overall spectrum of resistance mechanisms to ADT2 remains incompletely characterized, as does the biological impact of these events. Moreover, the extent to which such mechanisms might generalize across ADT2 regimens or operate in specific therapeutic contexts remains unknown. Finally, subsequent treatment options for this patient population beyond the use of cytotoxic chemotherapies (e.g. taxanes) are not well defined. The goal of this proposal is to create and apply computational biology algorithms that 1) systematically interrogate genomic resistance effectors to ADT2 in clinically relevant time points, 2) integrate in vitro models of ADT2 resistance with genomic features to define biological modules germane to ADT2 resistance, and 3) model clinical resistance with genomic data to inform subsequent treatment strategies. In doing so, we aim to discover new modules for clinical ADT2 resistance, provide insight into the expansion of rational treatment approaches for ADT2-resistant patients, and create an inferential framework through which clinicians may ultimately predict those treatment strategies most likely to benefit individual patients based on tumor genomic profiles. These efforts will facilitate a focused and comprehensive assessment of ADT2 resistance in the neoadjuvant and metastatic CRPC settings, explain genetic resistance to ADT2 in prostate cancer, and define subsequent therapeutic strategies.

PROJECT SUMMARY The increased accessibility of comprehensive molecular characterization of tumors and germline samples from cancer patients has accelerated translational discoveries and significantly impacted patient care. These approaches ultimately form the basis for precision cancer medicine, whereby ?clinically actionable? molecular data about a patient's tumor and germline genomic profile, specifically defined as diagnostic, prognostic, and predictive markers, are used at the point of care to guide treatment decision-making. While these strategies have been successful in certain use cases, the approaches to understand somatic and germline components of cancer patients are typically considered independently, and systematic characterization of the interaction between the somatic and germline genomes in the context of diagnostic and predictive clinical relevance have not yet been systematically performed across large cohorts of patients. This is in part the result of an absence of computational algorithms that are able to consider these features simultaneously, along with a lack of patient cohorts with both somatic and germline features and clinical annotations of relevant treatment responses to guide these investigations. Our previous studies have demonstrated, through innovative computational oncology approaches, how integrated germline and somatic analysis can determine diagnostic and predictive features that have immediate clinical impact in select clinical contexts. The goal of this proposal is to directly respond to Provocative Question PQ3: Do genetic interactions between germline variations and somatic mutations contribute to differences in tumor evolution or response to therapy? Our overarching hypothesis is that complex interactions between germline and somatic features within and across key DNA repair and immune pathways mediate inherited clinical risk, and selective response to existing chemotherapies and emerging immunotherapies. Specifically, in this proposal, we will leverage existing and emerging cohorts of tumor and germline whole exome/transcriptome data from patients, along with relevant phenotypic data regarding response to chemotherapies and immunotherapies, and develop innovative computational biology algorithms to systematically dissect these cohorts and determine how interactions between germline and somatic events shape clinical actionability. This proposal is unique in that it leverages the extensive and novel resources at both the Dana-Farber Cancer Institute/Harvard Cancer Center and the Broad Institute of MIT and Harvard, along with an international team of collaborators, to address the hypotheses outlined herein. The proposed specific aims are: 1) To determine inherited cancer risk in solid tumors through integrative computational biology, 2) To evaluate the impact of somatic and germline interactions on DNA repair defects and response to platinum-based chemotherapies in solid tumors, and 3) To identify somatic and germline features that coordinate to alter the immune microenvironment and impact selective response to immune checkpoint blockade in solid tumors. These studies will define key relationships between germline and somatic variants that shape tumor biology, with implications for understanding patient risk for cancer development and selective response to chemotherapy and immunotherapy. In addition, this project will establish new computational algorithms to enable widespread integrated consideration of germline and somatic features for broader use in the scientific community. Finally, this project will accelerate the clinical relevance of germline and somatic molecular profiling to enable precision cancer medicine, and serve more broadly as an innovative model for intersecting clinical oncology with computational biology.

PROJECT SUMMARY Approximately 8,000 people in the U.S. are diagnosed with germ cell tumors (GCTs) each year, and the vast majority are young men who develop testicular GCTs. Most patients are cured with conventional chemotherapy, although 30% recur, and half of such patients ultimately succumb to their disease. Given the long life expectancy of these patients, when death from GCT occurs, it accounts for among the greatest number of life years lost of any non-childhood malignancy representing. Our previous studies have demonstrated that GCTs exhibit an extreme burden of reciprocal loss of heterozygosity (RLOH) and high degree of mitochondrial priming for apoptosis. The goal of this proposal is to dissect the molecular features that initiate RLOH in GCTs, determine the relationship between RLOH and defect DNA checkpoints as tumors progress, and evaluate the ability of functional assays to identify highest risk disease prior to chemotherapy initiation. The long-term objective is to enable new mechanisms of patient stratification and identify new therapeutic targets for chemoresistant GCTs, currently an area of unmet medical need with extremely limited therapeutic options under investigation. This proposal is unique in that it leverages the extensive and novel resources at both the Dana-Farber Cancer Institute/Harvard Cancer Center and the Broad Institute of MIT and Harvard, along with an international team of collaborators, to overcome limited preclinical models of this disease and incorporate patient-centered assays focused on human tumor samples to address the hypotheses outlined herein. The proposed specific aims are: 1) To define the genetic defects associated with reciprocal loss of heterozygosity in primary germ cell tumors, 2) To identify the molecular features of tumor evolution leading to chemoresistant germ cell tumors, and 3) To assess the clinical utility of pluripotency markers as prognostic for GCT outcomes. These studies will define the meiotic defects underlying RLOH in GCTs, identify the secondary molecular defects that initiate lethal chemoresistance, and reveal targets for enhanced patient stratification and therapeutic development. In addition, these efforts will accelerate development of new computational algorithms that explore integrative molecular analyses of both the genome and epigenome to address specific hypotheses regarding oncogenic development and progression to chemoresistance that may have broad applicability. Finally, this project will accelerate the clinical and molecular characterization of GCTs, explore the underlying biology driving this rare tumor type, and serve more broadly as an innovative model for studying rare cancers.

PROJECT SUMMARY Cancer genomes typically harbor a substantial number of somatic mutations. Relatively few driver mutations actually alter the function of proteins in tumor cells, whereas most mutations are considered to be functionally neutral passenger mutations. Over the past decade, the search for cancer driver mutations has focused on coding regions and several mutational significance algorithms have been developed for coding mutations. The contribution of mutations in noncoding regulatory regions to tumor formation largely remains unknown and current mutational significance algorithms are not designed to detect driver mutations in noncoding regions, due to biological differences between coding and noncoding mutations. The emerging availability of large whole- genome sequencing datasets (e.g. PCAWG and HMF datasets) creates an ample opportunity to develop new mutational significance algorithms that are particularly designed for the interpretation of noncoding regions. Recently, we have developed a new statistical approach that identifies driver mutations in coding regions based on the nucleotide context. Critically, consideration of the nucleotide context around mutations does not require prior knowledge for functional consequences associated with these mutations. Hence, we hypothesize that generalizing our nucleotide context model to noncoding regions will uncover novel noncoding driver mutations that cannot be detected using the mutational significance approaches currently available. For this purpose, we will develop a statistical framework that incorporates the biological differences between coding and noncoding mutations and that is specifically designed to detect driver mutations in noncoding regions. Specifically, we will consider the context-dependent distribution of passenger mutations, modeling of the background mutation rate, accurately partition the background mutation rate, model the sequence composition of the reference genome, and account for coverage fluctuation. We will then combine these statistical components by computing an independent product of their underlying probabilities. We will derive a significance p-value using a Monte-Carlo simulation approach, and use FDR for multiple hypothesis test correction. This strategy will allow us to accurately estimate the significance of somatic mutations in noncoding genomic regions. We will next apply this statistical framework to whole-genome sequencing data of 5,523 tumor patients, thereby deriving a comprehensive list of candidate driver mutations in noncoding regions. Finally, we will investigate whether noncoding mutations are overrepresented in transcription factor binding sites, regulate gene expression levels, induce alternative splicing, or affect epigenomic states. Upon the completion of this project, we will have developed and applied a statistical framework for discovery of significant somatic mutations in noncoding regions, and defined the mutational landscape of the non-coding cancer genome. All aspects of the methods developed and applied in this project will be made open source and developed in an online platform.

PROJECT SUMMARY Despite recent advances in treatment, metastatic castration resistant prostate cancer (mCRPC) remains incurable, and approximately 30,000 men die of this disease yearly. Advances in immunotherapy with drugs targeting immune checkpoints have raised hopes that these agents will improve outcomes for mCRPC patients. While initial studies of immune checkpoint blockade have been unsuccessful, emerging evidence suggests a subset of prostate cancer (PCa) patients can respond, although the mechanisms of PCa immunotherapy response and resistance are incompletely characterized. Work in other immunotherapy responsive malignancies has found several predictive immune and tumor-intrinsic properties that contribute to response, but the extent to which these (or other) features are operant in PCa is largely unknown. For example, we recently identified mutations in a chromatin remodeling complex that mediates immunotherapy response through T cell interactions in solid tumors, and in parallel discovered a previously unknown PCa genomic subclass defined by mutations in these same chromatin remodelers. These findings indicate that tumor-intrinsic epigenetic dysregulation may also interact with the immune system to modulate PCa immunotherapy responsiveness. The overarching hypothesis of this project is that multiple immune and tumor- intrinsic properties mediate PCa interactions with the immune system, and these interactions can be modified through selective targeting in combination with checkpoint blockade to expand the therapeutic potential of immunotherapy in PCa. We will leverage our team's deep experience in clinically grounded molecular characterization and preclinical models that can test immunotherapy combinations in PCa to define the processes that govern the immunotherapy landscape in PCa. The proposed specific aims are: 1) Define the systemic and infiltrating immune states in PCa associated with clinical response to checkpoint blockade; 2) Establish the immunologic impact of chromatin dysregulation and inhibition in PCa; and 3) Determine the impact of existing DNA damaging agents for sensitizing PCa to PD-1 blockade. This proposal leverages the extensive, novel, and complementary resources at both Dana-Farber/Broad Institute and University of California, San Francisco, led by highly collaborative investigators and an international scientific team, to address the hypotheses outlined herein. Through a combination of functional, molecular, and clinical approaches inherent in these studies, our team is poised to identify mCRPC cohorts that may benefit from this treatment paradigm, determine strategies to augment the use of checkpoint inhibitors in this disease, and mechanistically define the immune and tumor-intrinsic defects that drive immunoresistance in PCa. Broadly, this project will provide a unique approach for the Immuno-Oncology Translation Network (IOTN) community and enable discovery of anti-cancer immunotherapies strategies for PCa that may have larger relevance across the IOTN network and collaborating members of the Cancer Immunotherapy Consortium.