It isn't laughable. It is seriously a concern. The idea that they possibly looked at the results and changing it toward the end is not something that can be ignored. This is seriously a concern for the integrity of the trial. You power trials based on the design and can't change it end flight because you don't get the results you were expecting. You properly design a new one. It is actually worse than data mining because you don't actually let the original design of the trial play out.
All doses given and most patients deceased is not "mid-trial". They knew the trial was screwed when Liau said "Everyone is living longer but it's not helping our trial" because they needed to see clear separation in the arms. They did not. If they were going to change the endpoint, that was the time to do it. That was like 8 years ago. Not 2022. Liau has authored papers on pseudoprogression so claiming ignorance in her own trial is comical.