Replies to post #444407 on Anavex Life Sciences Corp (AVXL)

[catdaddy]

Are we still at a 1% chance, Doc?

[Steady_T]

The Lecanemab study had only one primary endpoint so it is not quite comparable to the AD study.
The Trofinetide study did have two endpoints however it did not specify any alternative methods of statistical testing. It required both endpoints to meet p<0.05 to proceed to testing the secondary endpoints.

Thank you for linking those two documents.

[Joseph_K]

Thanks for posting these. I might not agree with your forecast for blarcamesine approvals, but I sure as heck appreciate your generosity in helping provide an education!

There's been conversation on the board about whether Anavex provided in its P2b/3 (AD) SAP for a hierarchy of alternate statistical tests for the results on ADCS-ADL. Anavex reported the results of ADCS-ADL in terms of Odds Ratio. A biostatistician friend of mine is certain that Odds Ratio must have been not-first-listed in a hierarchy of statistical tests (or models) for accessing the ADL data, and that the first-listed would have been difference of means. For comparison, I just looked at the two PDFs you posted, and I couldn't find any hierarchy of alternative statistical tests for any single endpoint in the lecanemab SAP. I'm a novice at this, and I'm wondering: Is there no such hierarchy in that SAP or am I just not seeing it? Is the following, from the trofinetide SAP, that type of hierarchy?

For CGI-I: ...
In the event that the model fails to converge using the unstructured covariance matrix, the
following covariance structures will be modeled in the order given (i.e., from least
parsimonious to most parsimonious): heterogeneous Toeplitz, heterogeneous compound
symmetry, heterogeneous autoregressive(1), Toeplitz, compound symmetry,
autoregressive(1), variance components. The first covariance structure that allows for
convergence will be selected for the final model.

Thanks in advance.