Replies to post #392670 on Anavex Life Sciences Corp (AVXL)

[12x]

It’s true WS doesn’t buy it. That’s why SP is $8 not $80.

If SDL mean change not significant, there is nothing one can do to change that.

However, I think what the company is trying to do is to point out to the regulators that the drug has sustained improvement on a portion of patients with minimal side effects. We can argue how small that portion is and whether it’s enough to persuade FDA to approve, but it’s the best path forward (use 50mg data is better but that’s a different story for another day).

[boi568]

If the odds ratio is the wrong test for measuring reduction in decline, but it is the right test for exploring absolute improvement in the dosed population (and is part of the SAP, as well), the Anavex explanation makes sense.

Anavex adjusted its co-primary endpoints to match the mab trials -- a reduction in decline, rather than a change from baseline. In that circumstance, only a reduction in decline, the odds ratio should not be the analytic tool you reach for. But if you find, nonetheless, that you have a group of improvers above baseline, you can reach for the SAP-included odds ratio test to explore that outcome. That seems to be what they did. Given earlier AD 2a results, Anavex would have wanted this test in the 2b/3 statistical toolkit.

A couple of points here: First, the absolute improver group is not a large fraction of the entire dosed group. It will probably be almost twice as large a fraction, though, when we see the 50mg arm alone. Second, there remains the possibility that there is another subgroup that more or less maintains cognition and function. My inclination is that different groups of responders have different APOE genetics. This will also result in less-than-classic outcome distribution curves.

We just have to wait for the expanded readout coming soon to get these answers.