Background and importance Abiraterone and enzalutamide are used for treating castration-resistant prostate cancer (CRPC). The lack of direct comparisons makes the selection and positioning of these drugs difficult.

Aim and objectives To compare abiraterone and enzalutamide use in metastatic CRPC, and to provide real clinical data on effectiveness and safety.

Material and methods Retrospective observational study conducted in a tertiary hospital in patients with metastatic CRPC.

Patients evaluated and treatment initiated between January 2015 and September 2021.

The primary effectiveness variable was progression-free survival (PFS). Overall survival (OS) and survival probabilities were also estimated. Survival parameters were estimated with the Kaplan–Meier test and compared by the log-rank test using R-software (v.4 - 2021).

As safety variables, the percentage of patients with adverse events (AE) and grade according to the Common Terminology Criteria for Adverse Events (CTCAE) were collected.

Results 99 patients were included (abiraterone=70 and enzalutamide=29; disproportionality due to the prospective design). No significant differences were observed in the patients’ baseline characteristics: mean age (75.6±9.1 years vs 75.8±7.5, respectively) and number of metastases at baseline. These were mainly bone (36.34%) and lung (6%). Gleason at baseline was ≥8 in 45.7% of those treated with abiraterone and 31% with enzalutamide. 92.9% in the abiraterone group had Eastern Cooperative Oncology Group (ECOG) 0–1 and the comparable figure was 89.7% for enzalutamide.

62.9% with abiraterone presented ≥1 AE. Most frequent AE were G1-asthenia (22.3%) and G1-hypertension (12.3%). 8.6% were AE≥G2. In the enzalutamide group, 69% presented ≥1 AE(10.3% ≥G2). Common were G1-asthenia (62.1%) and G1-headache (13.8%).

Median PFS for abiraterone was 31 months (95% CI 20 to NA) and for enzalutamide 42 months (95% CI NA to NA); with no significant differences (p=0.5). Median OS was not reached in either group, with no significant differences (p=0.7). For overall survival, at month 13, 92.2% of patients did not reach the event in the abiraterone group and 81.5% in the enzalutamide group.

The power of the study for PFS was 0.038 and for OS 0.042, indicating that the power to detect differences is low.

Numerical disproportion between individuals makes enzalutamide more sensitive to events; however, the number of events remained proportional, with both curves being practically superimposable.

Conclusion and relevance Statistical differences in PFS were not found. Median OS was not reached in either group; AE were mild to moderate for both groups. We cannot affirm that there are differences in effectiveness and safety between these treatments.

Conflict of interest No conflict of interest