Comparative cardiotoxicity of the novel hormonal agents abiraterone and enzalutamide in metastatic castration-resistant prostate cancer using real-world data.

62 Background: Novel hormonal agents (NHAs) such abiraterone (ABI) and enzalutamide (ENZA) have demonstrated similar survival benefits against placebo groups in their respective clinical trials leading to their regulatory approval in both the pre- and post-chemotherapy settings in metastatic castration-resistant prostate cancer (mCRPC). Despite the overall tolerable risk profile, certain signals of cardiovascular toxicity were reported for these agents in clinical trials but little is known about their incidence in clinical practice. The objective was to assess the comparative cardiovascular safety of ABI and ENZA in patients with mCRPC in the real-world. Methods: A retrospective population-based cohort was extracted from Quebec public healthcare administrative databases. Patients were selected on the basis of having received androgen deprivation therapy prior to initiating a novel hormonal agent (ABI or ENZA) between 2012 and 2016. The primary outcome of interest was cardiovascular-related hospitalization (composite outcome that included acute coronary disease, cerebrovascular disease, heart failure, arrhythmia and other cardiovascular causes). Inverse probability of treatment weighting (IPTW) with the propensity score was used to adjust for measured baseline confounders including pre-existing cardiovascular disease. Results: The cohort comprises 2,183 patients, with 1,773 (81.2%) in the ABI group and 410 (18.8%) in the ENZA group. Before IPTW, mean age of the ENZA group was higher than the ABI group (78 vs 76). There were more ENZA patients with pre-existing arrythmia (ABI: 10.7%, ENZA: 15.1%) and diabetes (ABI: 21.5%, ENZA: 25.1%). Crude incidence rates of cardiovascular-related hospitalization were of 10 events per 100 person-years (PYs) and of 7 events per 100 PYs for the ABI and ENZA groups, respectively. After applying IPTW, all baseline variables were well balanced across both groups with standardized differences < 0.05. The ABI group was at greater risk of cardiovascular-related hospitalization compared to the ENZA group (IPTW-hazard ratio (HR): 1.79, 95% confidence interval (95%CI): 1.04-3.09). The risk of hospitalization for heart failure was greater in ABI (IPTW-HR: 3.02, 95%CI: 1.17-7.78). Conclusions: In our study population, there was a greater risk of cardiovascular-related hospitalizations for ABI users relative to ENZA users, in particular for hospitalization for heart failure. Given the lack of evidence from randomized head-to-head comparisons of both agents, these results provide clinicians with additional insight on the cardiovascular risks of mCRPC patients treated with NHAs in the real-world and further large studies are required to corroborate these findings.