Pembrolizumab versus chemotherapy in microsatellite instability-high or mismatch repair-deficient metastatic colorectal cancer: 5-year follow-up from the randomized phase III KEYNOTE-177 study.

Background: Results from the phase III KEYNOTE-177 study established pembrolizumab as a new first-line standard of care for microsatellite instability-high or mismatch repair-deficient (MSI-H/dMMR) metastatic colorectal cancer (mCRC). Previous results from KEYNOTE-177 showed a statistically significant and clinically meaningful improvement in progression-free survival (PFS) with pembrolizumab versus chemotherapy ± bevacizumab/cetuximab in MSI-H/dMMR mCRC. Results after >5 years of follow-up are reported.

Patients and methods: Adults with untreated MSI-H/dMMR mCRC were randomly assigned 1 : 1 to receive pembrolizumab 200 mg intravenously every 3 weeks or chemotherapy. Patients assigned to chemotherapy could cross over to pembrolizumab after centrally confirmed progressive disease. Dual primary endpoints were PFS per RECIST v1.1 and overall survival (OS). Secondary endpoints included duration of response and safety.

Results: At data cut-off (17 July 2023), median follow-up was 73.3 months (range, 64.9-89.2 months). Overall, 307 patients were assigned to receive pembrolizumab (n = 153) or chemotherapy (n = 154). Fifty-seven (37.0%) patients assigned to chemotherapy crossed over to pembrolizumab per protocol; 39 (25.3%) received a programmed cell death protein 1/programmed death-ligand 1 [PD-(L)1] inhibitor off protocol (effective crossover rate, 62%). Median OS was 77.5 months with pembrolizumab versus 36.7 months with chemotherapy (hazard ratio, 0.73; 95% confidence interval 0.53-0.99); 5-year OS rates were 54.8% versus 44.2%. Median PFS was 16.5 months with pembrolizumab and 8.2 months with chemotherapy (hazard ratio, 0.60; 95% confidence interval 0.45-0.79). Median duration of response was 75.4 months (range, 2.3+ to 80.1+ months) with pembrolizumab versus 10.6 months (range, 2.8 to 71.5+ months) with chemotherapy. Compared with chemotherapy, fewer patients in the pembrolizumab arm experienced adverse events (80% versus 99%; grade 3-5, 22% versus 67%).

Conclusions: With >5 years of follow-up, responses to pembrolizumab remained durable. Median OS was more than twice as long in patients treated with pembrolizumab versus chemotherapy in first line despite an effective crossover rate of 62%. Pembrolizumab remains a standard of care for MSI-H/dMMR mCRC.