sentiment_stocks,
I think we are both saying the same thing just maybe understanding differently. I agree 100% that L induced pseudoprogression messed up PFS but I also believe late SOC/placebo pseudoprogression played a part in messing up true SOC/placebo numbers because L works best with pseudoprogression already going on (hot tumor). Late SOC/placebo pseudoprogression has always been a legitimate concern because of having a potential active treatment like L, that has a history of working even better with pseudoprogression going on, being given at the optimum time. The impact of late pseudoprogression, significant even if not major, coupled with crossover treatment would have created a situation where there were basically early treatment, late treatment patients (SOC/placebo crossover due to late pseudoprogression) true progression coupled with true crossover SOC/placebo (most of SOC/placebo) and then the rest of SOC/plscebo that never crossed over.
Bottom line is that only treatment arm patients compared to ECA and SOC/placebo true progressors (rGBM) compared to ECA would or could ever be able to accurately measure treatment effect because of the way the trial was designed the the best outcomes possible being offered to those on SOC/placebo as well as treatment. Best wishes.
News 
Market Data 
Discover 
