Prostate Cancer Vulnerabilities to BH3 Mimetic Drugs

The anti-apoptotic BCL2 family proteins (including BCL2, BCLXL, and MCL1) act by neutralizing BAX and BAK, and by inhibiting the BH3-only pro-apoptotic proteins that can activate BAX/BAK. BH3-mimetics are drugs that enhance apoptosis by binding to and inhibiting BCL2, BCLXL, or MCL1. Navitoclax inhibits BCL2 and BCLXL and has single-agent activity in hematological malignancies, but causes thrombocytopenia due to BCLXL inhibition. A BCL2-specific BH3-mimetic that spares platelets, venetoclax, is similarly active and FDA approved for several hematological malignancies. Unfortunately, these currently clinically available BH3-mimetics have limited single agent activity in most solid tumors, which appears to substantially reflect a greater role for MCL1. BH3 mimetic drugs that inhibit MCL1 have more recently been developed and are in early trials, but preclinical studies from us and others suggest that it will in most cases be necessary to inhibit both BCLXL and MCL1 to achieve robust apoptotic responses, and it is likely that toxicity will limit the ability to combine BCL2/BCLXL and MCL1 inhibitors in patients. Although BH3 mimetics have limited efficacy as single agents in PC, they may have activity in subsets of PC with genomic alterations affecting apoptotic pathways. We recently identified the mitochondrial ubiquitin ligase MARCH5 as the primary mediator MCL1 degradation in response to cellular stress, and found that MARCH5 gene deletion that occurs in up to ~5% of CRPC can sensitize to BH3 mimetic drugs. Other alterations that increase MCL1 expression (including MCL1 gene amplification) are also frequent in PC and can confer increased MCL1 dependence. Conversely, our preliminary data indicate that PC with BRCA2/RB1 loss may be highly dependent on BCLXL. Based on these data, Aim 1 is to identify and characterize genomic alterations that may be used as robust biomarkers for clinical trials of single agent BH3 mimetic therapy. While a subset of PC may be responsive to single agent BH3 mimetic drugs, fully exploiting these drugs will likely require the identification of synergistic combination therapies. Indeed, we have previously identified a number of available drugs that can markedly enhance MCL1 degradation and sensitize to navitoclax. Therefore, Aim 2 is to identify combination therapies that exploit BH3 mimetic agents to drive apoptotic responses in CRPC. Finally, we hypothesize that castration-sensitive prostate cancer (CSPC) cells exposed to intensive androgen signaling inhibition may have a reduced apoptotic threshold and be vulnerable, at least transiently, to the addition of a BH3 mimetic drug. Therefore, Aim 2 will also determine whether BH3 mimetics can be used to exploit vulnerabilities generated acutely by intensive ASI in CSPC. Overall, we hypothesize that BH3 mimetic drugs will be highly effective in a subset of genetically defined CRPCs, and more broadly in combination with other targeted agents. The Specific Aims are 1) Identify genomic alterations in prostate cancer that sensitize to BH3 mimetic drugs and 2) Identify BH3 mimetic combination therapies that are effective in prostate cancer