Key points are not available for this paper at this time.
5064 Background: Pembro + ola did not significantly improve rPFS or OS vs NHA in unselected patients (pts) with previously treated mCRPC in the phase 3 KEYLYNK-010 study (NCT03834519). Here, we present the results of prespecified biomarker analyses with respect to rPFS or OS. Methods: Pts were randomly assigned 2:1 to pembro + ola or NHA. PD-L1 combined positive score (CPS) was measured by IHC (22C3 pharmDx); HRRm, BRCAm, TP53, PTEN, SPOP, ATM, CDK12, TMB, and MSI were analyzed in tissue samples using FoundationOne CDx or ctDNA using FoundationOne Liquid CDx; AR-V7 was analyzed using Epic Sciences’ CTC assay. Associations (continuous scale) with outcomes were evaluated using adjusted Cox proportional hazards regression, and 1- (pembro + ola) and 2-sided (NHA) P values were calculated; significance was prespecified at α 0.05). rPFS and OS HRs for pembro + ola vs NHA by biomarker status are shown (Table), with notable results observed for AR-V7 status. Conclusions: For pembro + ola, PD-L1 CPS had a weak positive association with rPFS, while HRRm and BRCAm had strong positive associations with rPFS; these biomarkers did not correlate with NHA outcomes. We observed a potential benefit with pembro + ola over NHA (for rPFS and OS) in AR-V7–positive mCRPC pts, which requires validation. Clinical trial information: NCT03834519 . Table: see text
Yu et al. (Sat,) studied this question.