Armen Aprikian ^1* Amit Bahl ² Aurelius Omlin ³ Giulia Baciarello ⁴ Abhiroop Chakravarty ⁵ Prashanth Kondaparthi ⁵ Georgia Gourgioti ⁶ Thomas Mclean ⁶ Alexis Serikoff ⁶ Andrew Chilelli ⁶

• ¹ McGill University Health Centre, Montreal, Canada
• ² University Hospitals Bristol NHS Foundation Trust, Bristol, England, United Kingdom
• ³ Onkozentrum Zurich, University of Zurich and Tumorzentrum Hirslanden, Zurich, Switzerland
• ⁴ Azienda Ospedaliera San Camillo-Forlanini, Rome, Italy
• ⁵ Parexel International, Hyderabad, India
• ⁶ Astellas Pharma Europe, Addlestone, United Kingdom

Introduction: Androgen-receptor pathway inhibitors such as abiraterone and enzalutamide have demonstrated clinical benefit in patients with metastatic castration-resistant prostate cancer (mCRPC). The aim of this study was to conduct a meta-analysis of published realworld evidence studies comparing outcomes among patients treated with enzalutamide or abiraterone in the first-line setting. Methods: We conducted a systematic literature review to identify eligible studies. Evaluated outcomes were: overall survival (OS), progression-free survival, prostate-specific antigen (PSA) progression-free survival, PSA response, all-grade adverse events, grade ≥3 adverse events, treatment discontinuation, and dose reduction. Each outcome's suitability for metaanalysis was evaluated by assessing whether there were sufficient data to make comparisons between studies, consistency between outcome definitions, and whether the studies adjusted for baseline patient characteristics. Outcomes deemed suitable for meta-analysis were analyzed using fixed-effect and random-effect models to obtain pooled-effect sizes. Sensitivity analyses were conducted to evaluate the robustness of conclusions. Results: Of 1849 records reviewed, 30 were eligible for inclusion. Most outcomes were deemed unsuitable for meta-analysis due to a lack of adjustment for baseline characteristics, issues with inconsistent outcome definitions, and the small number of studies reporting each outcome. The only outcome deemed suitable for meta-analysis was OS. A total of 17 studies reported hazard ratios (HRs) for OS, 11 of which were adjusted for baseline characteristics. Among the studies reporting adjusted HRs, the pooled-effect estimate favored enzalutamide over abiraterone (reference group) in the fixed-effect model (HR: 0.90 [95% CI: 0.87-0.93]) and the random-effect model (HR: 0.90 [95% CI: 0.86-0.94]). These results were consistent across all sensitivity analyses. Discussion: Across all analyses, enzalutamide demonstrated a statistically significant improvement in OS compared with abiraterone. These findings highlight the value of realworld evidence studies to demonstrate the potential of different therapies under real-world conditions and over long periods of time.