Abstract CT008: Pembrolizumab (pembro) in combination with gemcitabine and cisplatin (gem/cis) for advanced biliary tract cancer (BTC): Phase 3 KEYNOTE-966 study

Abstract

Abstract Introduction: BTC has a poor prognosis despite available treatment. KEYNOTE-966 is a randomized, double-blind, phase 3 trial of pembro versus placebo in combination with gem/cis as first-line therapy for unresectable locally advanced or metastatic BTC (NCT04003636). Methods: Eligible patients (pts) with histologically confirmed, RECIST v1.1-measurable disease with no prior systemic therapy in the unresectable or metastatic setting were randomized 1:1 to pembro 200 mg or placebo administered IV on day 1 Q3W for ≤35 cycles added to gemcitabine 1000 mg/m2 administered IV on days 1 and 8 Q3W until disease progression and cisplatin 25 mg/m2 administered IV on days 1 and 8 Q3W for ≤8 cycles. Randomization was stratified by region (Asia vs non-Asia), stage (metastatic vs locally advanced), and tumor origin (gallbladder vs intrahepatic vs extrahepatic). The primary end point was OS. Secondary end points were PFS, ORR, and DOR per RECIST v1.1 by blinded independent central review and safety. Per protocol, the final analysis of PFS and ORR was at the first interim analysis (IA1; data cutoff, December 15, 2021). All other data are from the protocol-specified final analysis (FA; data cutoff, December 15, 2022). One-sided P-value boundaries for significance were 0.0200 for OS, 0.0125 for PFS, and 0.0125 for ORR. Results: 1069 pts were randomized to pembro + gem/cis (n = 533) or placebo + gem/cis (n = 536). Baseline characteristics were generally balanced between arms; 45.5% of pts were from Asia, 88.2% had metastatic disease, and 59.2% had intrahepatic origin. Median study follow-up was 25.6 mo (range 18.3-38.4) at FA. At FA, median (95% CI) OS was 12.7 mo (11.5-13.6) for pembro + gem/cis vs 10.9 mo (9.9-11.6) for placebo + gem/cis (HR 0.83; 95% CI 0.72-0.95; P = 0.0034); 24-mo OS was 24.9% vs 18.1%. OS results were generally consistent across subgroups. At IA1, median (95% CI) PFS was 6.5 mo (5.7-6.9) for pembro + gem/cis vs 5.6 mo (5.1-6.6) for placebo + gem/cis (HR 0.86; 95% CI 0.75-1.00; P = 0.0225); 12-mo PFS was 25.4% vs 19.8%. At IA1, ORR (95% CI) was 28.7% (24.9-32.8) for pembro + gem/cis vs 28.5% (24.8-32.6) for placebo + gem/cis (difference 0.2; 95% CI -5.2 to 5.6; P = 0.4735); median (range) DOR was 9.7 mo (1.2+ to 22.7+) vs 6.9 mo (0.0+ to 19.2+). Grade 3-5 AEs occurred in 85.3% of 529 pts treated in the pembro + gem/cis arm vs 84.1% of 534 treated in the placebo + gem/cis arm (drug related, 71.3% vs 69.3%). Grade 5 AE incidence was 5.9% vs 9.2% (drug related, 1.5% vs 0.6%). Potentially immune-mediated AEs and infusion reactions occurred in 22.1% vs 12.9%. Conclusion: Pembro + gem/cis provided a statistically significant, clinically meaningful OS improvement versus placebo + gem/cis in pts with previously untreated metastatic or unresectable BTC. The safety profile of pembro + gem/cis was as expected and manageable. These data support pembro + gem/cis as a new first-line treatment option in this setting. Citation Format: Robin Kate Kelley, Changhoon Yoo, Richard S. Finn, Junji Furuse, Zhenggang Ren, Thomas Yau, Heinz-Josef Klümpen, Stephen Lam Chan, Arndt Vogel, Masato Ozaka, Chris Verslype, Mohamed Bouattour, Joon Oh Park, Olga Barajas, Uwe Pelzer, Juan W. Valle, Li Yu, Usha Malhotra, Abby B. Siegel, Julien Edeline, Makoto Ueno. Pembrolizumab (pembro) in combination with gemcitabine and cisplatin (gem/cis) for advanced biliary tract cancer (BTC): Phase 3 KEYNOTE-966 study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 2 (Clinical Trials and Late-Breaking Research); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(8_Suppl):Abstract nr CT008.