FDA cares about statistical endpoints. FDA has 6 efficacy endpoints they accept for approval consideration: Overall Survival, Disease-Free Survival (and Event-Free Survival), Objective Response Rate, Complete Response, Time to Progression and Progression-Free Survival, Time to Treatment Failure. Read the FDA Guidance if you really would like to know: https://www.fda.gov/media/71195/download
Even Dr. Liau does, as she is using PFS as the lead efficacy endpoint in her latest Keytruda/ATL combination trial.
Then there's the post hoc data dredged recurrent OS. The NWBO trial was a naïve GBM trial, but suddenly NWBO and the clinicians needed a new trial endpoint. So they ginned up a post hoc, data dredged recurrent GBM arm our of the air. Sample size only 64 patients. Bet the FDA will have something to say about a ginned up endpoint.