Abstract CT222: Pembrolizumab (pembro) for previously treated advanced hepatocellular carcinoma (aHCC): Meta-analysis of the phase 3 KEYNOTE-240 and KEYNOTE-394 studies

Abstract Background: In patients (pts) with previously treated aHCC, pembro demonstrated comparable efficacy and safety vs placebo (pbo) in the phase 3 KEYNOTE-240 (NCT02702401) and KEYNOTE-394 (NCT03062358) studies in global and Asian populations, respectively. The hazard ratio ([HR]; 95% confidence interval [CI]) for OS was 0.781 (0.611-0.998) in KEYNOTE-240 and 0.79 (0.63-0.99) in KEYNOTE-394; HR (95% CI) for PFS was 0.78 (0.61-0.99) and 0.74 (0.60-0.92), respectively. ORR differences with pembro vs pbo were similar in KEYNOTE-240 (13.8% [95% CI, 7.7-19.5]) and KEYNOTE-394 (11.4% [95% CI, 6.7-16.0]). We performed a prespecified meta-analysis of KEYNOTE-240 and KEYNOTE-394 to obtain a more precise estimate of the pembro treatment effect. Methods: In KEYNOTE-240 and KEYNOTE-394, pts with confirmed aHCC and progression or intolerance to sorafenib or oxaliplatin-based chemotherapy (KEYNOTE-394 only) were randomized 2:1 to pembro (200 mg IV Q3W) or pbo for ≤35 cycles, both with best supportive care. Inclusion/exclusion criteria were similar. Meta-analysis of pt data pooled from the ITT population of the pembro and pbo arms of each study was completed. OS, PFS (blinded independent central review [BICR] per RECIST 1.1), and ORR (BICR per RECIST 1.1) were assessed. Results: In total, 578 pts who received pembro and 288 who received pbo were included. The HR for OS and PFS and the difference in ORR for pembro vs pbo in all pts was 0.79 (95% CI, 0.67-0.93), 0.76 (0.64-0.89), and 12.5 (8.8-16.2), respectively (Table). Results were consistent across subgroups, including viral etiology, BCLC stage, and age. Conclusions: This meta-analysis of two studies with similar design, inclusion/exclusion criteria, and endpoints, showed improvement in OS, PFS, and ORR with pembro vs pbo across studies. These data also show consistent outcomes between the studies, providing further evidence for the benefit of second-line pembro for aHCC globally. All Patients Patients With Sorafenib-Treated aHCC Pembrolizumab(n = 578) Placebo(n = 288) Pembrolizumab(n = 550) Placebo (n = 274) OS, median (95% CI), mo 14.2 (12.8-16.2) 12.5 (10.2-13.6) 14.2 (12.8-16.0) 12.5 (10.4-13.6) HR (95% CI)a,b 0.79 (0.67-0.93) 0.78 (0.66-0.92) PFS, median (95% CI), mo 2.8 (2.7-2.9) 2.7 (1.6-2.8) 2.8 (2.7-2.9) 2.7 (1.6-2.8) HR (95% CI)a,b 0.76 (0.64-0.89) 0.76 (0.64-0.90) ORR, % (95% CI) 15.4 (12.6-18.6) 2.8 (1.2-5.4) 15.6 (12.7-18.9) 2.9 (1.3-5.7) Estimated treatment difference, (95% CI)b,c 12.5 (8.8-16.2) 12.6 (8.8-16.4) aHCC, advanced hepatocellular carcinoma; CI, confidence interval; HR, hazard ratio.aStratified Cox proportional hazard model with treatment as a single covariate and Efron’s method of tie handling was used to estimate the OS and PFS HR and its 95% CI.bStratification was performed per protocol and within each protocol, by strata used in the analysis of the respective protocol.cMiettinen & Nurminen method was used to estimate the difference in ORR and its 95% CI. Citation Format: Richard S. Finn, Kangsheng Gu, Xi Chen, Philippe Merle, Kyung-Hun Lee, Mohamed Bouattour, Peiguo Cao, Wei Wang, Ann-Lii Cheng, Liangjun Zhu, Ho Yeong Lim, Masatoshi Kudo, Yueyin Pan, Ting-Tsung Chang, Julien Edeline, Wei Li, Ping Yang, Chen Li, Jianfeng Li, Abby B. Siegel, Shukui Qin. Pembrolizumab (pembro) for previously treated advanced hepatocellular carcinoma (aHCC): Meta-analysis of the phase 3 KEYNOTE-240 and KEYNOTE-394 studies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr CT222.