Pembrolizumab versus chemotherapy for microsatellite instability-high or mismatch repair-deficient metastatic colorectal cancer (KEYNOTE-177): final analysis of a randomised, open-label, phase 3 study

Background: Pembrolizumab has shown improved progression-free survival versus chemotherapy in patients with newly diagnosed microsatellite instability-high or mismatch repair-deficient metastatic colorectal cancer. However, the treatment's effect on overall survival in this cohort of patients was unknown. Here, we present the final overall survival analysis of the KEYNOTE-177 study. Methods: This randomised, open-label, phase 3 study was done in 193 academic medical centres and hospitals in 23 countries. We recruited patients aged at least 18 years, with an Eastern Cooperative Oncology Group performance status of 0 or 1, and who had previously untreated microsatellite instability-high or mismatch repair-deficient metastatic colorectal cancer. Patients were randomly assigned (1:1) in blocks of four using an interactive voice response system or integrated web response system to intravenous pembrolizumab 200 mg every 3 weeks or to the investigator's choice of intravenous mFOLFOX6 (oxaliplatin 85 mg/m² on day 1, leucovorin 400 mg/m² on day 1, and fluorouracil 400 mg/m² bolus on day 1 followed by a continuous infusion of 1200 mg/m² per day for 2 days on days 1–2) or intravenous FOLFIRI (irinotecan 180 mg/m² on day 1, leucovorin 400 mg/m² on day 1, and fluorouracil 400 mg/m² bolus on day 1 followed by a continuous infusion of 1200 mg/m² per day for 2 days on days 1–2), every 2 weeks with or without intravenous bevacizumab 5 mg/kg every 2 weeks or intravenous weekly cetuximab (first dose 400 mg/m², then 250 mg/m² for every subsequent dose). Patients receiving chemotherapy could cross over to pembrolizumab for up to 35 treatment cycles after progression. The co-primary endpoints were overall survival and progression-free survival in the intention-to-treat population. KEYNOTE-177 is registered at ClinicalTrials.gov, NCT02563002, and is no longer enrolling patients. Findings: Between Feb 11, 2016, and Feb 19, 2018, 852 patients were screened, of whom 307 (36%) were randomly assigned to pembrolizumab (n=153) or chemotherapy (n=154). 93 (60%) patients crossed over from chemotherapy to anti-PD-1 or anti-PD-L1 therapy (56 patients to on-study pembrolizumab and 37 patients to off-study therapy). At final analysis (median follow-up of 44·5 months [IQR 39·7–49·8]), median overall survival was not reached (NR; 95% CI 49·2–NR) with pembrolizumab vs 36·7 months (27·6–NR) with chemotherapy (hazard ratio [HR] 0·74; 95% CI 0·53–1·03; p=0·036). Superiority of pembrolizumab versus chemotherapy for overall survival was not demonstrated because the prespecified α of 0·025 needed for statistical significance was not achieved. At this updated analysis, median progression-free survival was 16·5 months (95% CI 5·4–38·1) with pembrolizumab versus 8·2 months (6·1–10·2) with chemotherapy (HR 0·59, 95% CI 0·45–0·79). Treatment-related adverse events of grade 3 or worse occurred in 33 (22%) of 153 patients in the pembrolizumab group versus 95 (66%) of 143 patients in the chemotherapy group. Common adverse events of grade 3 or worse that were attributed to pembrolizumab were increased alanine aminotransferase, colitis, diarrhoea, and fatigue in three (2%) patients each, and those attributed to chemotherapy were decreased neutrophil count (in 24 [17%] patients), neutropenia (22 [15%]), diarrhoea (14 [10%]), and fatigue (13 [9%]). Serious adverse events attributed to study treatment occurred in 25 (16%) patients in the pembrolizumab group and in 41 (29%) patients in the chemotherapy group. No deaths attributed to pembrolizumab occurred; one death due to intestinal perforation was attributed to chemotherapy. Interpretation: In this updated analysis, although pembrolizumab continued to show durable antitumour activity and fewer treatment-related adverse events compared with chemotherapy, there was no significant difference in overall survival between the two treatment groups. These findings support pembrolizumab as an efficacious first-line therapy in patients with microsatellite instability-high or mismatch repair-deficient metastatic colorectal cancer. Funding: MSD.