Dr Schoen on a Real-World Study of Enzalutamide vs Apalutamide in mCSPC

“The important [finding] is that the overall survival between the 2 groups was not different, and [we arrived at that conclusion] using both the propensity score matching method as well as the inverse probability of treatment weighting.”

Martin W. Schoen, MD, MPH, a hematologist/oncologist at St Louis Veterans Affairs Medical Center and an assistant professor of medicine at Saint Louis University, discussed data from a real-world study of enzalutamide (Xtandi) vs apalutamide (Erleada) for the treatment of patients with metastatic castration-sensitive prostate cancer (mCSPC) that he presented in a poster during the 2026 American Urological Association Annual Meeting.

The study drew from a dataset of approximately 4000 patients, with a median age of approximately 73 years across both the enzalutamide and apalutamide groups, Schoen said. Through propensity score matching, 1712 well-balanced patients were identified in each arm, yielding a robust comparative cohort with closely aligned baseline characteristics, he explained. Schoen emphasized that achieving this degree of balance was critical to drawing meaningful conclusions from a real-world dataset, where confounding by indication can substantially limit interpretability.

The primary finding of the analysis was that overall survival (OS) did not differ significantly between the 2 treatment groups (HR, 1.06; 95% CI, 0.91-1.25; P = .452), Schoen said. He highlighted that this result was consistent across both the propensity score matching and inverse probability of treatment weighting methodologies.

Beyond OS, Schoen noted that no significant differences were observed in time to prostate-specific antigen progression (HR, 1.11; 95% CI, 0.90-1.37; P = .336). Additionally, no significant differences between the 2 arms were reported in terms of time to next treatment (HR, 0.95; 95% CI, 0.83-1.08; P = .410) or time to treatment discontinuation (HR, 0.96; 95% CI, 0.88-1.05; P = .384). With a median follow-up of approximately 22 months in both arms, the dataset provided sufficient observational depth to detect meaningful differences had they existed, he concluded.