At first glance, PFS seems like an acceptable endpoint. But this trial highlights how other variables like pseudo progression, etc., can introduce additional unexpected complications/risks into the trial data.
OS seems like a much more solid endpoint to work because it it less susceptible to unforeseen issues. Patients are either alive or deceased and the survival benefit is there or it isn't. There are no "interpretation" issues with OS.
Of course you can still have other supporting measures as well (PFS, immune response, QOL, etc.), but the primary measure for many reasons should be OS.
Do you think IMUC will see a nice bounce if they once again become the leader in the race for GBM approval?