Clinical versus radiographic progression and overall survival for patients (pts) with metastatic castration-resistant prostate cancer (mCRPC) from COU-AA-302.

Abstract

193 Background: COU-AA-302 evaluated abiraterone acetate plus prednisone (AA) vs prednisone (P) in chemotherapy-naïve mCRPC pts, with overall survival (OS) and radiographic progression-free survival (rPFS) as co-primary end points. Per study criteria, pts with radiographic progression (RAD) only were allowed to continue treatment, while those with unequivocal clinical progression (UCP) only were not, and were censored for rPFS. We evaluated the clinical significance of survival outcomes for pts with UCP only vs RAD only from the prospective COU-AA-302 trial. Methods: UCP was defined per protocol as ≥ 1 of the following: initiation of chronic opiates, ECOG performance status (PS) decline to ≥ 3, or initiation of chemotherapy, palliative radiation therapy, or surgery. OS was evaluated for each type of progression using Cox proportional hazard models. Results: 500 (92%) pts in the AA arm and 540 (100%) in the P arm discontinued study treatment. Of the 736 pts who discontinued treatment for a protocol-defined reason, 280 (38%) discontinued for UCP only, 332 (45%) for RAD only, and 124 (17%) for both UCP and RAD. Clinical events cited as the reason for discontinuation for UCP (AA vs P arm) included pain requiring opiates (22% vs 25%), ECOG PS ≥ 3 (4% vs 5%), and initiation of chemotherapy (50% vs 53%), radiation therapy (36% vs 27%) and surgery (3% vs 5%). UCP only pts had shorter median OS compared with RAD only pts (Table). Conclusions: UCP is a criterion used as an indicator for a censored event, yet appears to confer inferior survival relative to RAD. The high frequency of UCP implies that it may be an important determinant of clinical outcome. The events that drive UCP should be defined as part of the development of more informative interim trial end points, in line with the PCWG3-proposed “no longer clinically benefitting” outcome measure, which captures pts with UCP. Clinical trial information: NCT00887198. [Table: see text]