Anavex Life Sciences : Headlining Questionable Data Positively.

Authors: Ben Aubury, Fedor Kolchin, Nicholas Choupak. Published 9 Jan.

Anavex Life Sciences (NASDAQ:AVXL) is a clinical stage biopharma company with a market cap of $1.07B, currently targeting Mild-to-Moderate Alzheimer’s Disease (AD) with its lead candidate Blarcamesine (ANAVEX 2-73). They published their latest set of clinical data, their Phase IIb/III trial data on 31 Oct 2024. The results headlined as positive news but in my opinion can be interpreted negatively. Their MAA filing was recently accepted by the EMA, starting the clock on a 210 day review period, finishing on 19th July 2025. They also have another catalyst date on 16 Jan, where they are scheduled to present their Phase 2b/3 data at the JP Morgan Healthcare Conference. The positive headlines and the MAA filing acceptance was received very positively by the markets, but do not represent a meaningful increase in the probability of approval and hence value. I believe that their clinical trials do not have the necessary data to make an approvable data set.

Price at publication: $11.43 (close 8 Jan)

Introduction to blarcamesine

Alzheimer’s is the most prevalent form of Dementia. The exact mechanism by which Alzheimer’s develops is opaque, however, there is a growing consensus in the scientific community that Alzheimer’s is caused by build of amyloid-beta protein plaques in the brain due to improper cleavage of myeloid precursor protein by beta and gamma secretases. Blarcamesine is a sigma-1 receptor agonist. There are many known sigma-1 receptor agonists—notably, Donepezil, an approved dementia palliative therapy. Activation of SIGMAR-1 has several downstream effects, the relevant effects here are autophagosome formation and maturation. Autophagy is necessary for cellular recycling and, notably, reduction of neuroinflammation and the survival of neurons in a neurodegenerative disease such as Alzheimer’s. Furthermore, it has been observed that there is a reduced density of sigma-1 receptors in Alzheimer’s disease. Hence, Blarcamesine is hypothesised to combat this reduced density by providing more agonist action to the remaining receptors. The two most prevalent AD drugs are Donepezil and Memantine. Both have been around for a while and are current available in generic. Namzaric is another popular treatment which contains a combined dose of memantine and donepezil. Recently, two new AD drugs have been approved by the FDA: Donanamab (Eli Lilly) and Leqembi (Biogen/Eisei) in Jul. 2024 and Jan. 2023 respectively.

In this presentation, Anavex reports that:

• ITT population had a (barely) “clinically meaningful outcome (ADAS-Cog delta is >2.0)

• Greater significant clinical benefit experienced by Common SIGMAR1 WT gene carriers Results from the ANAVEX2-73-AD-004 Phase IIb/III trial]

• In the Phase IIb/III clinical trial, blarcamesine also demonstrated good comparative safety profile (no ARIA) Blarcamesine slowed clinical decline of ADAS-COG in ITT population by 36.3% The relevant slides are given in Fig. 2 and Fig. 3.

This presentation raises a few questions.

In an earlier slide, Anavex highlighted an emphasis on analysing the difference of the drug on variant/WT genes; “Patients who carry the Common SIGMAR1 wild type (WT)** gene are expected to have stronger response to blarcamesine than patients with the mutation (variant)”. Despite this, variant data is not separated out in this presentation, which removes the ability to compare between the two populations. SA Author C.C. Abbott's Nov 05 piece on Anavex discusses whether it is appropriate for Anavex to make any comparative statements between the WT and variant populations.

Very high dropout rates

The dropout rates for blarcamesine look very high in this presentation. The trial protocol specifies that the 50mg cohort (n=168), 30mg (n=170) and placebo (n=170) contains 508 patients. 164/170 patients in the placebo cohort actually begin the trial on Week 0. More concerningly, 296/338 of blarcamesine patients are enrolled in week 0. Even more concerningly, these numbers drop fast; 57 (19.13%) dropout in the first 12 weeks, 90 (30.2%) in the first 24 weeks and 111 (37.25%) by 36 weeks. Very strangely, the number of participants increased from 36 weeks to 48 weeks - there are some “undropouts”. This inconsistency should reflect poorly on the quality of the dataset which Anavex is presenting in their application. The final dropout rate for blarcamesine is 35.91%, according to the Oct 31 presentation. Disregarding whatever is going on with the undropouts, these are very alarming dropout rates, both higher than other AD drugs and the placebo group. Recently approved drugs like Donanemab and Leqembi (which could be a valid point of comparison for the EMA) have dropout rates of 13% and 10% respectively, at much longer trial durations of 76 weeks and 18 months respectively. More analogue treatments like Donepezil and Memantine are also significantly lower than blarcamesine at 28.9% and 17% at durations of 144 weeks and 28 weeks respectively. Alduhem (which was not approved by the EMA) had a dropout rate of 47% for an 18 month trial (the ENGAGE trial). Whilst this is nominally higher than blarcamesine, it looks a little different once it is adjusted for duration. Aduhelm saw 0.65% of participants drop out per week whilst blarcamesine saw 0.75% drop participants drop out per week on the trial. This could indicate safety concerns which have not been picked up by investors or the media.

Unpooled data:Considering the 30mg and 50mg formulations individually.

The company also previously released a more detailed presentation on Jul. 28 at the Alzheimer’s Association Conference, separating the 30mg and 50mg cohorts. This presentation includes a few additional slides breaking down other endpoints and, crucially, separating 30mg and 50mg cohorts. These results can be seen in Fig.5

Fig.5 confirms that the 30mg cohort does not meet clinically significant thresholds whilst the 50mg cohort does (delta of 1.93 vs 2.15). Moreover, the 30mg value is not statistically significant with a p value of 0.026 (the Bonferroni adjusted p value for the two cohorts would be 0.025) to make the conclusion that ADAS-COG13 is meaningfully affected by blarcamesine. Additionally, the dropout rates become very alarming here. The 30mg cohort has a dropout rate of 29.87% whilst the 50mg cohort has a dropout rate of 42.36%. Comparatively, placebo has a dropout rate of 23.17%. When viewed in as a pooled group, blarcamesine may appear to investors as relatively safe and effective. When seperated, we can see that the 30mg formulation is not effective (p value > 0.026 and delta < 2.0), whilst the 50mg formulation does not appear to be safe. Additionally, this presentation contains disclosure that blarcamesine has failed the ADCS-ADL coprimary endpoint. With a failure of one coprimary endpoint, the question of Bonferroni correction and statistical significance arises again. Anavex did not include the ADCS-ADL data in the Oct. 31 presentation, which may have shifted investors' attention away from this point of failure.

Comparable studies

We have already compared the drop out rates for blarcamesine to the drop out rates of other drugs. Anavex has a weaker data set than the previously accepted AD drugs, not only in terms of drop out rates but also in terms of number of participants and the duration.

Comparing Anavex to the clinical trials of donanemab (kisunla), lecanemab (leqembi) and aducanumab (aduhelm):

1. Trial participants:

Blarcamesine: 462

Donanemab: 1736

Lecanemab: 1795

Aducanumab: 1653

Blarcamesine has significantly fewer participants than previously approved AD candidates (NB Aduhelm was not approved by the EMA).

2. Study duration

Blarcamesine: 48 weeks

Donanemab: 76 weeks

Lecanemab: 78 weeks

Aducanumab: 78 weeks

Analyst criticisms and allegations of fraud

Anavex has been criticised by other analysts, saying that they have a habit of changing clinical trial endpoints ("shifting the goalposts") on their Rett's syndrome trials for blarcamesine. Analysts, such as Adam Feuerstein, have questioned these misrepresentations in the past. Another unnamed analyst is cited in multiple publications as saying "there are several key factors that point us to believe the data is provocative, but not yet compelling, given the choice of statistical analyses and other trial design/conduct ‘complexifiers.’". Shareholders have filed several litigation claims against Anavex as well; Levi & Korinsky LLP, Bragar Eagel & Squire and several others have filed class action claims alleging securities fraud against Anavex on behalf of shareholders that have been affected by the misrepresentation of clinical trial results by Anavex. These allegations create a risk that shouldn't be ignored by investors.

Financials

Anavex currently has in excess of 4 years of cash runway. As of their last financial statements, they held $138M in cash and equivalents. Net CFFO for nine months ended June 30 was -$24.15M. Given the rate of cash burn is unchanged, they have around 4 years of runway left. This is a very comfortable level of cash. In addition, Anavex has a right to sell to Lincoln park, who has an obligation to buy $110.8M dollars of stock through to 2026. I would say that runway is not a large concern for Anavex. They currently hold a good cash position especially given that they have already filed with the EMA.

Conclusion

In my opinion, Anavex's clinical trial results present that their two formulations, 30mg and 50mg are ineffective and dangerous, respectively. This is supported by the p-values and the dropout rates. I think that the market has overreacted to minor positive news and has based decisions mainly on positive headlines which are actually unsupported by data. I think that Anavex's data set is quite weak in terms of participants and duration and will not meet the EMA's standard. If you agree, disagree or have further information that can inform this article, please send me an email.