Replies to post #509518 on Anavex Life Sciences Corp (AVXL)

[Investor2014]

Yes unfortunately reading through your arguments and checking the sources - I'd have to agree this was in plain sight all along. I didn't do those calculations - I guess assuming the amazing Dr. Jin, 58 renowned scientists and peer reviewers would not have allowed it to slip by. The rapporteurs and team likely were not fooled.

[oldmystic]

if they can successfully convince CHMP that the ABCLEAR1/S1RWT data should be used for conditional approval

How would you handicap conditional approval upon re-examination?

[Hosai]

Completion rates weren't a strong part of Blarca trial. Though I would note in regards to the 'survival bias' argument made by some critics in terms of the higher dropouts in main trial in dosed group (though in 30mg group the disparity was equal to Donanemabs ITT disparity), in the OLE the completion rates reversed with higher dropout in original placebo group.

OLE completion rates from start of OLE to finish

from week 48 to 144 (96 weeks)
original Dosed completion = 58.22%
original Placebo completion = 47%

From week 48 to 192 (144 weeks) - only Aus patients recorded this far I believe

original Dosed completion = 34.17
original Placebo completion = 30.3%

Therefore if you believed in the survival bias argument by some critics (that weaker patients drop out skewing the results to the positive for the cohorts with higher drop out) then the original placebo group should have done better in the OLE, however obv this didn't happen the groups moved further away from one another in both primary endpoints and eventually for ADL become stat sig different.

https://e1cd7807-443e-425e-9433-41548681800c.filesusr.com/ugd/191fc6_efd81521f39749a49a61c8440e82589e.pdf

This is different from Lecanemab where after 3 years if anything it looks like the late start group was converging a bit on the early start group (arguably it should have been easier for early start group to stay ahead assuming drug works well as they'd been on in 18 months before other group starting taking it, vs Blarca 48 weeks before late start group)

https://investors.biogen.com/news-releases/news-release-details/new-clinical-data-demonstrates-three-years-continuous-treatment

[georgejjl]

You are completely WRONG regarding the following:

the study was poorly designed by grossly underestimating variance, it would be even further underpowered for the 70%S1R WT subset.

GOD bless,

[catdaddy]

Someone out there disagrees, Doc:

Concise Refutation + Bullet-Point Summary
Key Takeaway: The bear case is overstated. JPAD delivered a large, pre-specified cognition signal in early AD with supportive biomarkers and OLE data — a credible package for conditional approval.
Bullet-Point Summary
• S1R WT subgroup was pre-specified (not post-hoc) ? -2.32 points ADAS-Cog13 (P=0.015) in ~70% of patients
• COL24A1 WT (ABCLEAR2/3, Sep 2025) ? -4.74 points ADAS-Cog13 (P=0.0004) in another ~70% subgroup
• OLE (despite 70% dropout) ? -12.78 points vs ADNI at Year 3 (external control)
• Assumed SD 4.5 was optimistic but defensible for early AD; observed SD ~8 but observed effect 2.0–2.3 points > planned 1.5 ? actual power ~75–85% overall, ~70% in S1R WT
• CDR-SB significant (-0.456, P=0.0175); ADCS-ADL miss is typical in early AD (ceiling effect, same as lecanemab)
• Oral, no ARIA, upstream mechanism ? strong differentiation vs anti-amyloid mAbs
• Ongoing CHMP re-examination + requested FDA meeting (Nov 2025) ? realistic path to conditional approval
Bottom line: Far from “laughable” — the data package is among the stronger oral/small-molecule AD datasets seen to date.