Evaluating prostate cancer phenotype and genotype classification from circulating tumor DNA as biomarkers for predicting treatment outcomes

PROJECT SUMMARY/ABSTRACT Prostate cancer is the second most common cause of cancer mortality among men. The majority of these deaths are due to resistance to androgen deprivation therapy and progression to lethal castration-resistant prostate cancer (CRPC). New generation androgen receptor signaling inhibitors (ARSI) that target the AR signaling axis have been used in the CRPC setting; however, the majority of patients still develop resistance. Recently, prostate-specific membrane antigen (PSMA) has become a promising target for positron-emission tomography imaging (PSMA-PET) and targeted therapies, such as the recently FDA-approved radioligand (PSMA-RL) for CRPC patients who progressed on ARSI treatment. Despite a survival benefit for PSMA-RL therapy, the improved outcome is modest and only half the patients show favorable responses. The emergence of resistance to ARSI and PSMA-RL may arise through changes in tumor phenotype, such as trans-differentiation from prostate adenocarcinoma (ARPC) into treatment-related small-cell neuroendocrine prostate cancer (NEPC) and other phenotypes with loss of AR activity. Current methods require a biopsy to diagnose tumor histology, which can be challenging due to invasive procedures accompanied by morbidity and some tumors are not accessible or have poor sample quality. Furthermore, tumor heterogeneity is a major contributor to therapy resistance and is particularly challenging to identify using a biopsy of a single metastatic site. These challenges exemplify major limitations of current treatment strategies and precision medicine for men with CRPC. Circulating tumor DNA (ctDNA) released from tumor cells into the blood as cell-free DNA (cfDNA) is a non- invasive “liquid biopsy” solution for addressing challenges in tissue accessibility. Current research and clinical efforts have focused on the detection of genetic mutations from ctDNA sequencing as potential biomarkers; however, these do not fully explain why treatments fail. The objective of this proposal is to develop and evaluate innovative methods for classifying aggressive CRPC genotypes and phenotypes from ctDNA, overcoming challenges of tumor heterogeneity. The investigators hypothesize that ctDNA can be used to classify tumor subtypes in CRPC and that this can be used to predict treatment outcomes. In Aim 1, they will study tumor heterogeneity in men who have undergone rapid autopsy to evaluate the ctDNA classifiers for predicting heterogeneous phenotypes from post-mortem plasma. In Aim 2, they will determine the utility of ctDNA for predicting prostate cancer treatment outcomes in a prospective cohort of patients treated with ARSI and a subset of patients screened by PSMA-PET and treated with PSMA-RL therapy. They will evaluate the ctDNA classifiers as biomarker tools to aid in the initial allocation of PSMA-RL therapy and inform early indications of treatment resistance. In Aim 3, they will develop extensions to ctDNA methods that infer gene expression and tumor aggressiveness in prostate cancer phenotypes using preclinical mouse PDX models, including in vivo engineering of phenotype mixtures.