We recently published results from the final prespecified analysis of overall survival (OS) in the ARCHES trial, a phase III, double-blind study of enzalutamide (160 mg once daily) plus androgen deprivation therapy (ADT) versus placebo plus ADT. 1 In that analysis, enzalutamide plus ADT demonstrated a clinically meaningful and statistically significant increase in OS compared with placebo plus ADT (hazard ratio [HR], 0.66; 95% CI, 0.53 to 0.81; log-rank P , .001).Sun and Wei 2 have raised concerns about the use and interpretation of the HR for the OS end point and a lack of median Kaplan-Meier (KM) estimates from the data. They note that the validity of the HR depends on a proportional hazards assumption and suggest that this assumption is likely invalid. In our analysis, we considered the proportional hazards assumption to be appropriate, as observed by a visual examination of the log-log plot and the Kolmogorov-type supremum test (P 5 .4440). Furthermore, the HR is not the sole descriptor of treatment effect presented; we also presented the KM curve and provided KM estimates at 24, 36, and 48 months. 1 The HR is appropriate for these data and supported by the summaries of the KM curves and by the log-rank test; the KM curves and logrank test do not require proportional hazards assumptions and are independent of the Cox HR. These statistics are clinically meaningful to clinicians and well-understood by the scientific community.