AI
Saturday, December 10, 2022 2:18:41 PM
NWBO stated that 90% of the treatment arm was "crossover" -- that means got a second dose. The 64 in the comparator arm were dosed with DCVax-L in the crossover. As ALL of the patients were blinded, the crossover was the available option to all that progressed -- whether comparator or DCVax-L.
Unless I missed it, it has not been disclosed what the entire trial crossover population was. All that has been disclosed is about 90% received DCVax and 64 of those came from the "rGBM" arm. We do not know how many of the 233 treatment arm "crossed".
This does bring up an interesting point. Their is one bit of unbiased randomized data to look at. The 35(?) who never received DCVax vs those in the treatment arm who elected not to "cross". This will be smallish, but if the results are real it should show a very strong trend.
Yes, it would be post-hoc now. But it would be randomized and bias free,.
Odd we have not heard this presented. Sometime silence tells a lot.,
BTW, for the issue with post progression treatment. In cancer trials OS is often confounded by follow up treatment because most treatments fail and patients will try something else. Does not matter at all if the subsequent treatment is the experimental drug itself (as is common).
The FDA even points this out while saying OS in a randomized setting is still the preferred endpoint. Trials just live with this and accept the power will be reduced because of it. Unless they change to an ECA later of course
