KEYNOTE-921

KEYNOTE-921 randomised 1,030 men whose metastatic castration-resistant prostate cancer had progressed after androgen deprivation and one androgen receptor pathway inhibitor to pembrolizumab or placebo with docetaxel and prednisone.

At the first interim analysis (data cutoff 27 September 2021), median radiographic progression-free survival was 8.6 months (95 percent confidence interval 8.3 to 10.2) with pembrolizumab against 8.3 months (8.2 to 8.5) with placebo, hazard ratio 0.85 (0.71 to 1.01, p=0.03), which did not meet the prespecified boundary. At the final analysis (median follow-up 22.7 months), median overall survival was 19.6 months (18.2 to 20.9) against 19.0 months (17.9 to 20.9), hazard ratio 0.92 (0.78 to 1.09, p=0.17).

Grade 3 or higher treatment-related adverse events occurred in 43.2 percent with pembrolizumab and 36.6 percent with placebo. Two participants on pembrolizumab and seven on placebo died of a treatment-related adverse event. Pneumonitis was the commonest immune-mediated event, 7.0 against 3.1 percent.

The separately published patient-reported outcomes found no meaningful difference in time to pain progression, median 21.1 months against not reached, hazard ratio 1.05 (0.77 to 1.43), or in health-related quality of life. Adding the checkpoint inhibitor neither helped nor harmed how men felt; it simply did not work.