Biomarker findings and mature clinical results from KEYNOTE-052: First-line pembrolizumab (pembro) in cisplatin-ineligible advanced urothelial cancer (UC).

Abstract

4502

Background: Comorbidities and renal impairment preclude many with advanced UC from receiving chemotherapy. Initial results from the phase 2 KEYNOTE-052 (NCT02335424) trial suggested first-line pembro is active and safe in cisplatin-ineligible advanced UC. We present updated efficacy and safety data (all pts have ≥6 mo follow-up) and evaluate biomarkers correlated with outcomes. Methods: Eligibility criteria included cisplatin-ineligible (ECOG PS 2, CrCl ≥30- < 60 mL/min, grade ≥2 neuropathy/hearing loss, NYHA Class 3 heart failure), advanced UC, and no prior systemic chemotherapy. Pts received pembro 200 mg IV Q3W. Imaging was performed at wk 9, then Q6W for the first year, and Q12W thereafter. Primary end point was confirmed ORR (RECIST v1.1, independent review). Efficacy and safety were assessed in the 370 pts with ≥1 pembro dose. The associations of an 18-gene expression profile (GEP) and IHC PD-L1 combined positive score (CPS) with ORR were evaluated. Results: As of the Dec 19, 2016, data cutoff, ORR was 29% (95% CI, 24-34): 25 (7%) and 81 (22%) pts achieved complete and partial responses. Another 69 pts (19%) had stable disease as best response, for a clinical benefit rate of 47%. Median time to response was 2 mo (range, 1-5). At a median follow-up of 8 mo (range, 0.1-20) across all pts, median duration of response was not reached (range, 1+-18+ mo). 74% of responses were ongoing. Any-grade and grade ≥3 drug-related AEs occurred in 239 (65%) and 68 (18%) pts. Immune-mediated AEs occurred in 76 (21%) pts. Evidence supporting a positive association with response was seen in the first 100 pts for both biomarkers (GEP, n = 72, P = 0.007, ROC AUC 0.69; CPS, n = 96, P= 0.111, ROC AUC 0.58); biomarker data for all pts will be presented. ORR in the 110 pts with CPS ≥10% was 47% (95% CI, 38-57). Conclusions: Results confirm that pembro elicits clinically meaningful, durable responses in cisplatin-ineligible advanced UC. Consistent with PD-1 pathway biology, biomarkers (GEP and CPS) showed the expected trends of positive association with response to pembro. Pembro was well tolerated across cisplatin-ineligible pts, including elderly and pts with poor performance status. Clinical trial information: NCT02335424.