Evaluation of response to enzalutamide (E) consecutively after disease progression on abiraterone/prednisone (AP) and potential predictors of response

294 Background: Efficacy data of E immediately after AP failure in castrate resistant prostate cancer (CRPC) is limited. Although clinical cross-resistant has been observed, the optimal sequence of these agents is not defined. Specific patient pt) characteristics on AP could be useful in selecting appropriate pts for subsequent therapy with E. Methods: We retrospectively reviewed records of 40 pts with CRPC treated at the Cleveland Clinic with E immediately after disease progression on AP. We evaluated clinical / pathological features that could predict subsequent response or lack thereof to E. For this exploratory analysis, best % change in PSA (%Ch-PSA) was used as indicator of treatment response. Results: Median age at E initiation was 69 (58-81), median time from diagnosis to AP initiation was 6.1 years (0.9-16.3). Median PSA was 30.9ng/mL (1.5-1680) at the time of AP initiation and 66.9ng/mL (2.52-3130) at E initiation. 30/40 and 35/40 pts had bone metastasis at AP and E initiation, respectively. Prior treatments to AP included Ketoconazole (45%), Docetaxel (35%). Median time to PSA progression on AP was 9.6 m (0.5-49.6) and 2.3 m (1.8-3.1) on subsequent E. 26/39 pts on AP achieved PSA decline >25%; among these 20 had >50% PSA decline. Similarly, 13 and 8 pts receiving subsequent E achieved >25% and >50% PSA decline, respectively. 10/40 pts did not have PSA response with either agent. 4 pts had PSA decline > 50% on both agents. There was no significant correlation between %Ch-PSA on E and %Ch-PSA on AP, pre-E PSA, time to PSA progression on AP, or time from diagnosis to AP. Longer interval between AP and E initiation was associated with better outcome on E (p=.03). Prior treatment to AP did not predict subsequent response to E. Conclusions: Except for the association between longer interval between AP and E initiation with PSA response on subsequent E, other clinicopathological factors did not predict response to consecutive E in this cohort of pts. Identifying clinical and/or molecular factors predictive of response to AP and E in this setting is of critical importance. Ongoing randomized prospective studies will help determine the optimal treatment sequencing in men with CRPC.