Well I sort of agree, my post did say that ABCLEAR3 wasn't pre-specified (though p value very low and the gene interactions very plausible). However, it comes from ITT data that on its own is better than the MABS ITT data in terms of ADAS COG and CDR SB.. The pre-specified sigma WT data is a bit stronger still. Potentially conditional approval could be given for the pre-specified sigma wt group with a phase 4 on ABCLEAR 3 in meantime. Arguably though perhaps shouldn't need to be a sub-group when beating the MABS on ADAS COG and CDR SB the ones regulators care most about. Plus showing the effect growing into the OLE delayed start analysis.
"In contrast, the mAbs had also modest but clear ITT clinical efficacy and robust A/T/N signals across large, replicated trials."
As I said in a previous post I wouldn't necc say "replicated" as the MABS phase 2 results had quite a lot of failures - Lecanumab primary endpoint not stat sig. Secondary CDR SB also not stat sig. ADAS cog successful at one dose not another.
Donanemab P2 primary endpoint slim pass (0.04), failed most of its seconndary endpoints including adas cog and cdr sb - didn't give p values.
Both trials also showed extra brain volume loss and non stat sig Nf-L data so arguably not great on the 'n' part of A/T/N (this part was repeated in their phase 3s).
Obv I would agree their phase 3 trials were a fair amount bigger than Blarcamesine 2b/3 then again it is harder to get low p values in a smaller trial.
Market Data 
Markets 
AI
