Real-world analysis of prostate-specific antigen (PSA) outcomes among patients with metastatic castration-resistant prostate cancer (mCRPC) treated with enzalutamide (ENZA).

Abstract

33 Background: ENZA showed efficacy in chemotherapy-naïve men with mCRPC in a clinical trial setting (PREVAIL). We report real-world outcomes with ENZA for this population in a urology practice setting. Methods: This retrospective cohort study included men with prostate cancer newly initiating ENZA in the IntrinsiQ Specialty Solutions™ urology electronic medical records database between September 1, 2014 and February 28, 2018. Due to the approved indication in the study time frame, prescription of ENZA (first claim date = index date) was used as a proxy for mCRPC. Patients with evidence of prior chemotherapy and/or abiraterone were excluded. PSA value closest to index date (±30 days) was used as baseline. Men were followed until the earliest of: discontinuing ENZA, leaving practice, death, or study end. Best PSA response (largest decline or smallest increase in absence of decline from baseline), PSA declines of ≥50% and ≥90%, undetectable PSA, and time to PSA progression were analyzed. Results: We identified 931 eligible men. Most (>95%) were ≥60 years old; hypertension (54.6%) and diabetes (17.0%) were the most common comorbidities. Median (interquartile range [IQR]) baseline PSA was 9.0 (2, 37) ng/dL. Median (IQR) follow-up time was 12.5 (7.6, 19.4) months, during which a median (IQR) of 4 (3, 6) PSA tests were observed. Median time between two adjacent PSA tests was 2.0 months. A ≥50% and ≥90% PSA decline was observed in 55.0% and 23.8%, respectively. Best PSA response was a median (IQR) PSA decline of 58% (-89%, 1%), with 14.2% reaching an undetectable PSA value. Median time to PSA progression was 18.5 months (95% confidence interval 15.6, 23.7). Conclusions: This real-world study supports the effectiveness of ENZA in patients with mCRPC. The median PSA at treatment was much lower than PREVAIL, potentially explaining the longer time to PSA progression vs. PREVAIL. However, a lower proportion had PSA declines of ≥50% and ≥90% vs. PREVAIL, which may be attributed to more frequent PSA monitoring within a clinical trial setting and thus more opportunity to capture the true best PSA response.