ADDRESSING TREATMENT RESISTANCE IN MODELS OF LETHAL PROSTATE CANCER BY IDENTIFYING NOVEL TARGETS FOR DRUG DISCOVERY

Background: The majority of prostate cancer deaths are due to late-stage disease that acquired drug resistance, termed castration-resistant prostate cancer (CRPC). About one-fourth of such cases present features of neuroendocrine (NE) differentiation, the mechanisms of which are still poorly understood. Collaborators of Dr. Califano, including the Abate-Shen laboratory, generated in vivo models of NE prostate cancer (NEPC) and castration-resistant neuroendocrine prostate cancer (CRPC-NE) using a murine transposition-based forward mutagenesis screening system (called Sleeping Beauty [SB]). Patterns of genomic lesions generated by the SB system, termed Common Integration Site (CIS), are expected to be causally involved in the lethal phenotype characterized by NE differentiation. Indeed, cross-species analysis of preliminary data from these in vivo models identified several CIS-associated genes, the expression of which completely segregated human patients affected by metastatic lethal prostate cancer from patients diagnosed with localized disease.Objective: The main objective is to identify drugs that target critical tumor dependencies in treatment-resistant lethal prostate cancer. We plan to systematically elucidate mechanisms of treatment-related NE emergence in lethal prostate cancer by using a systems biology approach that integrates CIS and Master Regulator (MR) analyses.Specific Aims:Aim 1. Characterization of MR dependencies in lethal prostate cancerAim 2. Prioritization of candidate drugs targeting identified druggable MRs dependenciesStudy Design: (Aim 1) CIS analysis, using DNA-Seq assay from the SB system, will be performed on both intact and castrated mouse models to identify genes causally involved in lethal variants of prostate cancer. Integrative network analysis using RNA-Seq assay from the SB system, MR analysis, and pathway enrichment analysis from several databanks will be used to identify MR dependencies that are exploitable as drug targets. (Aim 2) Integrative analysis will identify preclinical models that are representative of lethal prostate cancer and suitable for drug screening. Using the selected in vitro model, analysis from pre- and post-treatment highly scalable RNA-Seq screenings on FDA-approved and experimental drugs will enable us to prioritize compounds that revert previously identified lethal prostate cancer MR signatures. We expect to find 2 to 5 drugs that kill in vitro lethal prostate cancer by impinging on driving mechanisms of treatment resistance.Personnel: (Principal Investigator) During my research career, I have worked in three countries (Italy, Germany, and the United States) on large-scale modeling of Gene Regulatory Networks (GRNs), graph-based CIS analysis for Gene Therapy safety profiling, and elucidation of prostate cancer MR proteins along their drug-induced activity. My plan is to use this award to continue the investigation of lethal prostate cancer treatment-related resistance mechanisms and, ultimately, to improve therapeutic strategies for prostate cancer patients. (Mentor) Dr. Andrea Califano is a world leader in systems biology research. He pioneered the first genome-wide regulatory model of human cells and a variety of methods for the identification of MRs responsible for tumorigenesis, neurodegeneration, and cellular differentiation. He also pioneered network-based methods for the elucidation of mechanisms of action of drugs, drug synergy, and drug sensitivity. An ongoing Phase II clinical trial is testing a drug for metastatic gastroenteropancreatic neuroendocrine tumors that was identified by research from the Califano laboratory. Globally, more than 40 doctoral and post-doctoral scholars have benefited from his mentorship. (Co-Mentor) Dr. Michael Shen is a world-leading expert on prostate cancer. He characterized the role of NKX3.1 in prostate cancer and established several clinically relevant models, such as genetically engineered mice (GEMMs) and 3D organoids. His laboratory developed mouse models of CRPC-NE. He has mentored more than 20 postdocs. Training Plan: The Columbia University Medical Center multidisciplinary research environment and the mentorship of two world-renowned scientists will allow me to learn several computational analysis techniques while working on high-impact clinically relevant open challenges in the prostate cancer research field.Impact: We anticipate finding several novel candidate biomarkers and drivers of lethal prostate cancer that will enable patient stratification, outcome prediction, and therapeutic strategy evaluation. Also, this study directly addresses the identification of compounds that target drivers responsible for acquired resistant phenotypes associated with treatment failures. Men affected by lethal prostate cancer and left with no treatment options will benefit from this study. Thus, this study addresses two of the Overarching Challenges established for the FY18 PCRP: (1) to define the biology of lethal prostate cancer to reduce death and (2) to develop treatments that improve outcomes for men with lethal prostate cancer.