Anavex Life Sciences Phase2a For Alzheimer’s Vindicated with 509 Sample Size Phase2b/3. Blarcamesine $AVXL

This Blog Is Only For Educational Purposes. Do Not Trade On This Blog Alone. Do Your Due Diligence or Consult a Professional.

Late Means Late

The presented data has been just released to Anavex and processed to be delivered just in nick of time to Dr. Farlane to be presented at CTAD 2022. Unfortunately, I had not saved the presentation slides as I initially downloaded them and they have been remoed now. If anybody has a copy I would be greatfull to receive a copy.

I have already mentioned that Phase2b/3 is repeat of phase2a with just larger sample size. In general remarks I can say that now we know why in data release in Phase2a the patients were listed by blood concentration. It was explained at the time that the dose and the concentration membership mostly overlaped. The problem is that for some 7.5% of patients there is the incidence of dizziness and confusion after dosage and that it might be so persistent that the members of the 50mg cohort never reach the required 50mg dose as the titration goes on. One way of mitigating this effect is to dose the patients before going to sleep, when the initial blood concentration is the highest after oral dosing. This is a widly practised way of dosing oral CNS drugs. About 48 patients suffered from this limitation, if I am right about it. In the light of that present situation the data merges the two cohorts of 30mg and 50mg into one.

Redoing Phase2a

The first thing which struck me was that the Baseline average ADAS-Cog score when interpreted on the MMSE scale was near or about 21.5 MMSE points. (Later I looked up the slides, and h measurment was 23.5 MMSE) The clinicaltrials.gov webpage for the trial listed the conditions, among others, to be MMSe score 20 to 28 and PET scan positive for Amyloid plaque. From the listed ADAS-Cog score I gather that the latter condition lowered the average score or rather brought the average down to 23.5 MMSE (updated per slides) making this very similar in number to Phase2a but not the same as they must be clustered about the mean or average without much dispersion. No wonder that the enrollment has taken so much time. The other consequence is that versus $SAVA’s Simufilam with average or mean MMSE score of somewere about 24-26 MMSE, Blarcamesine treats a bit worse off patients.

Calculating the Number of Responders

From the email released by Anavex at the time of presentation by Dr. Farlane;

ANAVEX®2-73 demonstrated visible improvement in patients with Alzheimer’s disease. Patients treated with ANAVEX®2-73 were 84% more likely, to have improved cognition by ADAS-Cog score change of -0.50 points or better from baseline to end of treatment than patients on placebo, Odds Ratio = 1.84 (p = 0.015). On average, patients, who improved cognitively with ANAVEX®2-73 treatment, improved by ADAS-Cog cognition score of -4.03 points. ANAVEX®2-73 treatment was 167% more likely to improve function compared with placebo, at a clinically meaningful improvement of ADCS-ADL score change of +3.5 points or better, Odds Ratio = 2.67 (p=0.0255). This reflects a robust improved and clinically meaningful outcome in cognition and function from baseline. 

We have two pieces of information. First is the criteria for branding a patient a responder, the other is the incidence as related to the placebo cohort. Here, we have to dispell some misconceptions. For patients in the Mild cases of Alzheimer’s it is not uncommon to spontaneously improve cognition scores. With time this erratic behavior diminishes and with time disapears completely in a given sample. The best information on this I could find was in a plot of placebo improvement scores in the Donezepil trial that is included in the label for the drug.

The plot with white square is for the placebo arm. The fly in the oinment is that when it comes to my purposes the duration is just 24 week, whereas Blarcamesine trial was of duration of 48 weeks. Let’s make an assumption that we can approximate the passage of time by shifting the entire curve to the right and then read the percentage of those who still improve within the population of the placebo arm. The magnitude of the shift might be somewhere between 2 or 3 ADAS-Cog points. The choice of the shift magnitude was inspired by the 4.11 points deterioration of the placebo Phase2b/3 arm over 48 weeks.

For Delta = 2 there 48% patients improving therefore 184% of that will be 88% of 309 subjects of combined cohorts = 272

For Delta = 3 there 40% patients improving therefore 184% of that will be 74% of 309 subjects of combined cohorts = 229

After accounting for the dropouts of about 6% the average membership in either cohort is 155. The above membership numbers in the cognitive improvement or hold are larger than just either cohort. That might mean that even the 30mg cohort has about at least 50% members who did not detriorate or just improved vs. the 48% or 40% of the same for the placebo arm. This would be consistent with the results from Phase2a where no efficacy has been detected in the 30mg cohort. By default the therapeutic effect or drug efficacy has to be very high in the 50mg cohort to produce such low p-value as reported. Also, the average score for the responders is ADAS-Cog -4.03, so most of this improvement must be within the 50mg cohort with great strength. No wonder that the p-value of .015 is suggesting low probability of false positive or false efficacy of the drug.

User MayoMobile (stocktwits) had conversation with Dr Farlane discussing the existance of the super responders. I believe that they are the cause behind the low p-value of the combine cohorts even in face of possible lack of therapetic effect in the 30mg cohort.

However Cognition Improves the Real McCoy is Daily Living

The odds are here greater over placebo but I have no data at this time assess this as I have no placebo data on ADCS-ADL. Though odds are greater but the p-value which determines the false positive is greater too. They measure two different things. One measures incidence or relative probability of incidence, the other probability of error in establishing efficacy.

In conclusion, Phase2b/3 has validated the Phase2a therapetic signal. My personal conclusion is that there is a group of superresponders who basically are those at the highest dose or concentration in blood that can be turned back from the abyss of cognitive deterioration. They might be present in both cohorts by the virtue of concentration and these are going to be predominately among the 50mg cohort. They are the reason behind low p-value for the trial. The problem with dizziness and confusion can be overcome by the late in the day intake so that the maximum dose can be administred to maximize the benefits.