PARP inhibitors elicit distinct transcriptional programs in homologous recombination competent castration-resistant prostate cancer

Abstract

Prostate cancer (PCa) is the second most lethal cancer in men in the United States. African American (AA) men have twice the incidence and death rate from the disease than European American (EA) men. Early-stage PCa is treated with hormone deprivation therapy, although patients frequently experience relapse. Advanced stage PCa is associated with increased expression and activity of the DNA damage/repair pathway enzyme, poly (ADP-ribose) polymerase 1 (PARP1). Furthermore, PARP1 inhibitors are FDA-approved for the treatment of advanced PCa tumors that carry mutations in components of a specific DNA damage/repair pathway termed homologous recombination repair (HRR). However, PARPi also provide benefit in model systems without HRR incompetencies. A number of different PARPi have now been developed, tested and approved for use in PCa. These inhibitors utilize multiple biochemical mechanisms of action and exhibit distinct potencies and toxicity profiles. While there is emerging evidence of differences in DNA damage/repair pathway enzyme expression between EA and AA men, PARP1 itself has not been fully explored in the context of race. This study hypothesized that 1) AA and EA PCa may respond differently to PARPi and 2) different PARPi may differentially impact the transcriptome, irrespective of HRR status. To test these hypotheses, PCa patient samples from a racially diverse cohort were examined to define race-based differences in PARP activity/expression. Additionally, biologically relevant doses of five clinically relevant PARPi were established across multiple PCa lines carrying different genetic backgrounds, HRR status, and hormone therapy sensitivities. Collectively, these findings demonstrate a link between racial background and PARP1 expression/activity and define a core transcriptional response that lies downstream of all five PARPi, while simultaneously defining transcriptional programs unique to each inhibitor. These findings broaden our understanding of the effector pathways downstream of individual PARPi and provide a compelling rationale for a broader exploration of the impact of race on the response to PARPi. They may also help refine personalized recommendations for use of specific PARPi.