Anavex Life Sciences Corp (AVXL)

[Anshu2]

Most trails have only one primary endpoint (i.e, what is measured and what statistic/measure is used to compare; the latter is mostly the "delta from baseline to trails-end). In P2b/P3 case, CM never disclosed how the two primary co-endpoints Cog and ADL endpoints will be compared, but at CTAD came up with ORs as the way to do it (which he later changed to LSMs recently).

P2b/P3 was unusual -- in that, it had two primary endpoints. That's actually putting unnecessary burden on yourself -- since you need to pass both with p < 0.05 -- so I fail to see the point of having two primary endpoints.

However, it seems statistically reasonable (but I have never seen before -- so, not sure if FDA will agree) to consider the trail a success -- if any one of the two co-primary endpoints passes with p < 0.025. This is what AVXL did recently with delta/LSMs as the comparison measure.

Point: Irrespective of SAPs, the primary endpoint was known. How it should be compared is also pre-specified (and, in almost all cases, is disclosed to the public). That is what defines true success in a pivotal trial. Anything else -- is ad hoc analysis and puts trial success into a question mark (but doesn't necessarily mean an NDA rejection).