NorthWest Biotherapeutics Inc (NWBO)

[exwannabe]

1. The first error was the reviewers' claim that the mOS of the entire trial of 331 patients was almost 4 months longer than the mOS of the 232 patients in the treatment group when in fact the mOS of the entire trial was only about 0.8 months longer than the mOS of the treatment group; This error was committed because the reviewers were apparently unaware of the fact that the mOS of all 331 patients in the trial was measured from surgery whereas the mOS of the 232 treatment patients was measured from randomization.

That was an error, true. They probably made the rational assumption that OS was reported from randomization as is expected. But regardless, they were wrong.

2. The second error is described in the post I am replying to (your post) and that error stated that 90% of the 99 patient placebo group crossed over after disease progression when in fact only 64 of those 99 did so.

They used LL's exact statement to JAMA on the reason for not comparing OS within the trial being that "eventually almost 90% crossed over". That is LL's error, not theirs.

3. The third error found by Senti deals with the review's claim that no information concerning the IDH1 mutant status was presented in the trial results when in fact that data was presented. It was reported that very few (about 7?) GBM patients did carry that mutation.

Wrong. The article said IDH mutation status was not reported in the first paper. That is true, it was not reported then. BTW, they never reported exactly how they "found a way" to salvage the IDH data years later when the tumor had already been used to manufacture -L.