Why would a company change the endpoints 14 years into a trial?

Let's say hypothetically that you are conducting a trial that has PFS as the primary endpoint and OS as the secondary endpoint. The trial is now 14 years old and along the way your trial manager has conducted efficacy IAs and reports that all is well. At the same time you've been monitoring shipping records and you note a good number of patients seem to still be alive. In other words, while remaining technically blinded you have every reason to believe the trial is going very well indeed. If this were the case, then why would you change endpoints? If you have every reason to believe that things are going well, would you change the endpoints to make OS the primary just because OS is now considered the "gold standard"? No, of course you wouldn't. If everything is going well, changing the endpoints would be an incredibly stupid risk, and an unnecessary one when OS is already defined as a secondary endpoint.

We all know why NWBO changed endpoints, and so does Hoffman on iHub and all of his delusional ilk. They also know that the decision to change endpoints was made with post-hoc knowledge. They all know the truth, and they all know that their own "gold standard" explanation makes no sense. But they have a narrative to maintain, so they lie, and they lie, and they lie some more.