How could this trial (pivotal phase 3 DCVax-L for GBM) fail?
1) control arm outlives treatment arm? (very unlikely for GBM according to what have been known from literature and myriad of presentations from treating doctors);
2) No separation between control and treatment, and/or early and late (crossover) vaccinated patients? (very unlikely according to supposed mechanism of action of DCVax-L, and the suggestion of Dr. Linda Liau in her various presentations);
3) PFS messed up due to pseudo-progression? (may be true initially, but corrected after adjudication. More importantly, as the last patient in the trial lasts five years, PFS really doesn't matter anymore because it is just a surrogate for OS);
4) Treatment effects if any cannot be detected in a statistically significant way, ie, the p value being greater than 0.05? (may be likely due to crossover, but the efficacy can be backed up by comparing to similar external data and some immunogenetics data showing distinct immune-reactions before and after vaccination, etc.);
5) Not a pre-specified group or subgroup shows any statistically significant benefits? (very unlikely); and
6) All of the above or a combination of some of the above? (mostly unlikely)
A quick rough recap on what we have known today before unblinding:
1) mOS for all patients in the trial: about 23.1 months from surgery (compared to the standard of care: 14-17 months); and
2) One year mOS: 89.3 %; two year mOS: 46.4 %s; three year mOS: 28.2 % (based on 2018 updated data); and five year mOS: about 15-17% for all patients according to Les from his recent discussion (compared to standard of care which is about 5% at five years).
Conclusion: DCVax-L will be at least approved for treating GBM based on landmark long-term overall survivals (there are other ways for DCVax-L to get approved as well).
