Update Compendium 2024

Bottom-line Up Front (BLUF): A few weeks ago, the FDA published draft guidance for Early Alzheimer's disease. The guidance makes marked improvements for very early stage trials by placing greater emphasis on pathophysiological changes (measurable changes in the brain as a cause of disease), and negates the need for the activities of daily living metric. Overall, the draft is favorable to Anavex Life Sciences and a myriad of other developers. Below you will find a link to the draft guidance (downloadable), as well as a link to Spirit of the Coast's official comment to the FDA in order to improve upon the draft. Our feedback primarily revolves around specificity on desirable physiological changes with emphasis on those that are most likely to extend patient lifespan (brain atrophy).

• Draft Guidance

• Spirit of the Coast Comment

Bottom-line Up Front (BLUF): Within this report we make comment on various points of interest: ADCS-ADL success, the tau biomarker, the arterial spin labeling endpoint, possible gender-dependent effect, SIGMAR1 mRNA levels, odds ratio and effect sizes, placebo effect, and time-saved as result of therapeutic intervention. These are subjects we've researched over the last couple of weeks and wanted to wrap up here as separate-but-related points of interest. We hope to spark interest in readers as we all wait for Anavex's full 2b/3 dataset.

Gate Keeping

To clarify SOTC stance on the ADCS-ADL endpoint, it is our opinion that ADCS-ADL more likely than not met statistical significance as a standalone primary endpoint (p=0.05 or better), but did not meet the threshold needed as a co-primary endpoint (combining ADAS-COG and ADCS-ADL to reach p=0.05 together). Based on our research, it appears that the results are a waterfall. What is meant by this is, for the ADAS-COG to be combined with CDR-SB, the ‘first measure’ must be met. In this case, the first measure being both ADAS-COG and ADCS-ADL were met at the p-value of 0.05 (separately). 

Gate keeping procedures indicate that both ADAS-COG and ADCS-ADL did in fact reach 0.05 separately; however, that’s not good enough to claim success outright because the scores are supposed to be combined to meet 0.05 (i.e. both need to be 0.025 to add up to 0.05).

Thus, as was pre-prescribed in the SAP, Anavex could use the gate keeping measure (because both co primary endpoints were stat sig separately) to try to come to a truly successful trial (0.05 or lower when combining two endpoints). They were able to do this with the CDR-SB and ADAS-COG because the first measure (ADAS-COG and ADCS-ADL meeting at least 0.05 separately) was met. 

Gate Keeping Conclusion: it is more likely than not that ADCS-ADL met statistical significance somewhere between 0.05 and 0.0275. Our confidence is increased by the fact that the company's latest 10-series filings reiterate that the ADAS-COG and ADCS-ADL endpoints were both met. 

There are two fundamental truths:

1. According to the CTAD PR and subsequent 10-series filings, the trial met both co-primary endpoints; ADAS-COG and ADCS-ADL. This implies both endpoints were met with at least 0.05.

2. The second truth is that the trial was a success because of the ADAS-COG and CDR-SB reaching 0.05 combined. This implies that ADAS-COG and ADCS-ADL could not reach 0.05 when combined.

We look forward to the eventual release of the mean-scores for ADCS-ADL in the pooled dosed cohorts as has already been released for ADAS-COG and CDR-SB.

Tau Levels

Plasma (blood) T-tau and CSF P-tau/T-tau were collected as biomarkers in Anavex's 2b/3 Alzheimer's trial. We have yet to receive any information regarding tau and Blarcamesine's therapeutic effect on the proteins. Tau association to brain atrophy has now been elucidated in multiple peer reviewed papers - not just for Alzheimer's, but also in other tauopathies like stroke, corticobasal degeneration, lewy body dementia, frontotemporal dementia, and more. In all of these tauopathies, tau increase is correlated to an increase in brain atrophy. This is unsurprising as tau is largely associated with 'brain injury'. With the knowledge that Blarcamesine dramatically slowed or ceased atrophy in the whole brain, it is likely that the drug had a beneficial effect on tau as well. In preclinical studies, Blarcamesine has already shown to block or decrease tau levels by altering p35 levels, and thereby reducing CDK5 dysfunction. Further information on CDK5 can be found throughout the SOTC site; but as a brief reminder, dysfunctional CDK5 is disastrous to the CNS and its substrates are related to at least 39 key functions such as BACE1 expression, Parkin activity, synaptic plasticity, axon formation, tau, and more. Blarcamesine has been shown to correct or reverse CDK5 levels in Alzheimer's and PDD patients.

Tau Levels Conclusion: By virtue of known atrophy prevention/slowing and corrected CDK5 levels in earlier patients, it is highly likely that tau levels were improved in the 2b/3 Alzheimer's study.

Arterial Spin Labeling 

Beyond structural MRI, the company assayed cerebral blood flow (CBF) utilizing arterial spin labeling (ASL). ASL is an ingenuitive way to quantitatively and objectively track blood flow in a patient's brain by magnetizing water in the blood and then tracking the blood. The blood amount is then divided against the mass of the brain tissue it was tracked to in order to compute a value. Blarcamesine has demonstrated effect towards hypertension normalization, having corrected systolic blood pressure in the 2a trial participants. Irregular systolic blood pressure damages arteries, causes heart disease, kidney damage, and stroke, and has been implicated in an increased risk of cognitive decline and dementia. 

Arterial Spin Labelling Conclusion: Efficient CBF is needed for proper brain function, ensuring supply of oxygen and energy substrates. In addition, and possibly more importantly, efficient blood flow helps to clean out the accumulation of dead or misfolded proteins in the brain. Accumulation of dead, misfolded, or zombie (senescence) proteins impair energy (ATP) output and cause neuroinflammation. Healthy CBF reduces brain atrophy, and it is likely this measure was successful in the 2b/3 trial - especially in the 'responder' population.

Gender-dependent Effect

We feel we’ve done a good job in previous reports explaining factors influencing Blarcamesine’s efficacy. Most of these factors were discovered in the Alzheimer’s 2a trial such as: S1R WT gene, earlier disease stage, APOE3 genes, drug concentration, and time on drug. It was also indicated that males had a higher tendency to perform better than females; however, this was not statistically significant. Because of this, it’s likely that males do perform better or have higher odds of responding to treatment, but both genders maintain good/excellent compatibility with the drug.

As we know, females have higher odds of all-cause dementia over males, especially as they reach menopause. We also know that S1R mRNA expression is reduced in women, and S1R expression appears to be enhanced by estrogen. As menopause reduces estrogen, S1R also lowers in conjunction.

Newer evidence published on 25 Jan 2024 revealed gender-based differentiation in the basal forebrain cholinergic system (BF / BFCS) during CNS degeneration. The cholinergic system is important to Anavex’s MOA as the muscarinic receptor agonism is dependent on the health/state of cholinergic regions. If the basal forebrain cholinergic system is atrophied, there is less acetylcholine to stimulate muscarinic and nicotinic receptors. Luckily, even with significant atrophy, Blarcamesine would still be partially effective with its S1R role and dampened M1-M4 ability. Within the report it was noted that:

• Normal aging has minimal effect on basal forebrain volume 

• Basal forebrain volume reduction is a biomarker for diagnosing neurodegenerative disease 

• Asymmetric atrophy was observed in subregions of the basal forebrain in female patients as opposed to male patients 

Below are various excerpts from the research, with especially important portions highlighted in red. A brief synopsis will follow.

“Research combining structural MRI (sMRI) and cerebrospinal fluid (CSF) examination has revealed that cognitively normal older adults with abnormal amyloid beta and tau pathology in CSF biomarkers exhibit a reduction in gray matter (GM) volume within the Ch4 subregion of the BF. A longitudinal sMRI study found that in cognitively normal participants destined to develop AD, the BF area exhibited significant atrophy as early as 4.5 years before the onset of clinical symptoms, indicating that atrophy in the BF is a biomarker that predicts the likelihood of asymptomatic elderly subjects developing AD.”

“It has been reported that approximately two–thirds of patients with AD are females, and they tend to experience faster cognitive decline compared to males. This could be due to the brain of female patients with AD undergo more severe pathological damage, leading to more pronounced hippocampal atrophy and cognitive decline. Extensive brain imaging studies have provided evidence that males with MCI and AD tend to experience slower rates of brain atrophy over time compared to females. In addition, it has been observed that males with MCI show less atrophy in multiple brain regions, and once diagnosed with AD, they exhibit less atrophy in various regions as well. It has been proposed that cholinergic basal forebrain cortical projection neurons within the nucleus basalis, which mediate memory, attention, and the degeneration in AD, may exhibit greater vulnerability in elderly females compared to males.”

“Animal experiments have further revealed significant sex differences in estrogen receptors within BF cholinergic neurons.”

“Our findings indicated a significant reduction in BF volume, specifically in the Ch4 region, even before the diagnosis of AD. Most of the cholinergic innervation of the cortex, which is involved in attention and memory, originates in the Ch4 (nucleus basalis of Meynert) and in the horizontal limb of the diagonal band nucleus of the basal prosencephalon. Functional alterations in this system have been implicated in neurocognitive disorders as well as the cognitive changes described in Parkinson’s disease and AD. The atrophy of BF in patients with AD and MCI has been widely reported. A study utilizing a large multicenter dataset found that all of the subregion volumes of the BFCS were significantly reduced in the AD group.”

“The survival and maintenance of BFCS neurons rely on the nerve growth factor (NGF) and its homologous receptors (trkA and p75[NTR]). Compared to HC, p75(NTR) mRNA levels in BF cholinergic neurons decreased by approximately 40% in females with AD, while the p75(NTR) expression in males remained unchanged. Additionally, compared to HC and MCI individuals, males with AD demonstrated a 45% decrease in trkA mRNA levels in BF cholinergic neurons, while females exhibited a 50% decrease, and reduced trkA mRNA levels were associated with poorer global cognitive performance in females. These findings suggested that females may be at a higher risk of cholinergic BF neuron degeneration.”

"An animal experiment showed that estrogen had a protective effect on both male and female BF cholinergic fibers, and the therapeutic potential of estrogen decreased with increasing age."

Gender-dependent Effect Conclusion: There is a complex relationship between estrogen, SIGMAR1 expression, APOE, cholinergic atrophy, and neurodegeneration. So far, monoclonal antibodies have showed reduced efficacy in female patients. Considering Blarcamesine's wide-sweeping MOA and the relationship between estrogen and SIGMAR1 reduction in female patients, we expect the drug to be more efficacious in females than the MABs by boosting endogenous protection and restoration. Multiple studies have now demonstrated that SIGMAR1 confers/interacts with p75 noted in one of the excerpts above, further lending credence to SIGMAR1/estrogen axis of effect. Nonetheless, due to more pronounced BFCS degeneration innate to female patients, it is not likely female and male patients with see an equal response. Perhaps more than anything, we find the sex-dependent effect of Blarcamesine's therapy to be forefront of interest. More information can be found in the "gender and estrogen" portion of the SOTC AAIC 2023 readout here.

SIGMAR1 (S1R) mRNA Levels

A driver of clinical success is attributing drug MOA to clinical efficacy. For the MABs for instance, the goal is to show reduction of amyloid and correlate those results to improvements in CDR-SB (or slowing effect). In the Alzheimer's 2a trial, PDD trial, and Rett suite, increased SIGMAR1 mRNA corresponds to higher clinical output. While a minor note, we expect this trend to continue in the 2b/3 final analysis. This ties in nicely with the next section below.

Odds Ratio + Effect Size:

According to Anavex's 4 Nov 2017 PR, of the 30 patients that completed the primary phase 2a Alzheimer's study to 57 weeks, a strong drug concentration / response relationship was first identified. Drug concentration in the upper tertile (1/3) increased the probability of improved MMSE score 2.1 fold (110%) and ADCS-ADL score 1.6 fold (67%). It was stated in the 25 Jul 2018 PR that 80% of patients were "responders" resulting in improved MMSE and ADCS-ADL.

As shown in the 2a supplementary materials, higher drug dose nearly always (although not exclusively) correlated to higher drug concentration. For these intents and purposes, high dose = high concentration. 

Pulled directly from Anavex's 1 Dec 2022 PR: "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."

As was found in the 2a trial, it is highly likely that the higher dose/concentration group had better scores in both memory (ADAS-COG) and function (ADCS-ADL). 

According to the CTAD 2022 presentation given by Dr. Macfarlane, 287 dosed patients (85.7%) were titrated to 30mg or 50mg. Of those patients, 263 (78.5) did not need to down-titrate from their set 30mg or 50mg dose. The presentation did not breakout the 30mg and 50mg separately; however, we assess it is likely more 50mg patients down-titrated over the 30mg cohort due to higher odds of increased dizziness/headache. It's very difficult to project a meaningful outcome without the exact numbers, but we hypothesize somewhere between 110-120 patients on the 50mg dose maintained the objective dose until end of study. Based on my previous analysis considering patient severity, dose, APOE genes, and S1R WT, there is possibly up to 98 "responders" [+/- 10%] as defined by Anavex's own description (ADAS-COG of -.5 points or better and/or ADCS-ADL +3.5 points of better). Most of these responders would be from the 50mg group with a much lower percentage from the 30mg group. 

Odds Ratio + Effect Size Conclusion: While the pooled data presented so far is positive and met necessary statistical weight for success, the real anticipation comes with broken out subgroups - especially the 30mg vs. 50mg subgroup. As it stands, it is possible to equate an effect size to the pooled odds ratios, as they can be a little bit difficult to understand on their own.

• ADAS-COG ODDS RATIO: 1.84 = a small+ effect size

• ADCS-ADL ODDS RATIO: 2.67 = a medium+ effect size

Keep in mind that Anavex's responder threshold was improvement over baseline, which is unique to Anavex, with predecessor competitors always opting for 'slower decline' as their responder threshold. With this knowledge, the odds ratio and respective effect sizes are more impressive than what meets the eye. A 50mg only analysis would also certainly garner higher odds ratios and effect sizes. 

Time Saved and Placebo Improvement [Donanemab]:

In MCI and early-Alzheimer's stage patients, placebo improvement is a common phenomenon. In the Donanemab trial, the following was noted:

• Low/Medium Tau Group [ideal patients]: 47% of dosed patients saw no progression on CDR-SB, 29% of placebo patients saw no progression

• Pooled Groups [all dosed vs. placebo]: 36% of dosed patients saw no progression on CDR-SB, 23% of placebo patients saw no progression

We do not currently have data on placebo response in Anavex's 2b/3 Alzheimer's trial, but there was surely placebo response as patients are susceptible to placebo effect and earlier-stage patients can still see benefit from diet, exercise, and lifestyle improvements. CDR-SB is scored on a 0.5-point scale, so it's not the best tool to use for progression analysis, but there is no ADAS-COG or ADCS-ADL progression analysis publicly available for progression. In addition to the progression metric, time-saved will be another key output we hope to be included in the final data set. Time-saved was all-the-rage at AAIC 2023, and according to analysis presented there featuring unique calculation methods the following was elucidated:

• Donanemab: saved 5.5 months in 18 mo dosing cycle

• Lecanemab: saved 5.1 months in the same

• Aducanumab: saved 2.6 months in the same

Time Saved and Placebo Improvement [Donanemab] Conclusion: There is still a lot of useful data expected from the 2b/3 trial. At AAIC 2023, Dr. Jeffrey Cummings mentioned how time-saved may be the most useful metric for approval going forward, and that odds ratio analysis will be instrumental in determining which patients are right for each drug. SOTC previously attempted a time-saved analysis using brain atrophy, that analysis can be found in graphic 5, here. We thought it necessary to share Donanemab's placebo response in order to set expectation for the final Anavex data release and put forth the understanding that placebo response, while robust, can be overcame with drug efficacy - even when threshold for efficacy was set as high as Anavex's.

Bottom-line Up Front (BLUF): Two years ago on 6 Feb 22, ForestFoxes asked me a question related to doubts of another poster to Blarcamesine's long-term efficacy. The other poster stated "....the gist of it was that it [S1R agonist] will work and do so quite dramatically-at the beginning, but over time it will become less effective as the person's S1R levels continue to drop (or the patient will require increasing doses with declining improvement over multiple years) and it's therefore not going to prove a viable long-term therapeutic strategy..."

Yesterday, 6 Feb 24, ForestFoxes requested an update to my initial repsonse. Below you will find my response two years ago, as well as my response yesterday. At the least, I believe some followers will find the conversation interesting.

Reply on 6 Feb 22:

First, I don’t believe his point to have merit. The AD 2a trial extended through 5 years with at least half of the patients still taking the drug. (These people were quite old, it is likely that some actually died - not Blarcamesine related).

Secondly, I found in my analysis that most AD drugs dropped efficacy dramatically somewhere around the 35-45 week range (need to go back and find the exact number). Anavex did not see a drop here, in fact, it increased further. With this being the case, Anavex already beats SOC in multiple arenas. 

Of course many drugs lose efficacy after bodies attenuate to the response. But we haven’t seen meaningful evidence towards this at the 5 year mark. And what’s more, it is my hypothesis that the preventative CNS trial Anavex is planning will begin patients at 30mg. If people take 30mg to stave off CNS disorders but begin losing efficacy, the dose can be titrated up to 50 (and in some cases 60mg) to account for that loss.

Reiteration on 6 Feb 24:

Indeed, based on slope change for ADAS-COG and CDR-SB in the 2b/3 trial, Blarcamesine efficacy appears to increase with time, especially the longer the trial went on (week 36-48). 

I was going through the 2a supplementary material the other day and found some interesting tidbits. Primarily, it was found that TIME on drug was the lead indicator of therapeutic value in both cognition, and activities of daily living. 

This was especially true in patients with APOE4 genes (the majority of AD patients). Patients with high blood concentrations of the drug found the most therapeutic benefit. While not exclusively high dose patients, MOST high concentration was found in high dose cohort. High concentration and time on drug were especially important for patients with S1R variant. 

Lastly, COMT gene variant also needed more time, which correlated to higher activities of daily living scores but not necessarily cognition.

As we know, the most ideal patient type for blarcamesine are patients with S1R WT, early in disease stage, with APOE3 genes and high drug concentration. These patients, ESPECIALLY over a long trial period, will do very well in the trial and represent most likely ‘responders’ or ‘super responders’. I have done analysis prior to determine the odds of being one of these such patients. It was 48% of all MCI and early stage Alzheimer’s patients. When combining the ‘second best’ patient group, blarcamesine can optimally treat 61% of ALL MCI and early AD patients. Which is significant. Nearly all other patients (remaining 39%) respond well to treatment, but simply slow degradation instead of halt/improve.

One of my key takeaways is that ADCS-ADL may not have seen enough separation from placebo during the trial. Which isn’t necessarily surprising as activities of daily living are hardly affected in very early stages. More so than cognition, it was found in the 2a supplementary data that ADCS-ADL was improved most by patients on the trial longest. This makes sense, as cognition and brain degradation are the two drivers of reduced patient function. By improving brain function and halting atrophy (or slowing in both cases) activity scores would have a trailing effect that would take longer to record. 

I’m looking forward to 72-week OLE data, as this is congruent with both Lecanemab and Donanemab trials, and is likely more than enough time to garner ADCS-ADL separation.

I’m also looking forward to a 50mg-only chart, a “responder-only” chart, and oppositely, a total non-responder chart of the worst performing patients to better define what small percentage of patients respond worst to the therapy. Perhaps it is a specific co-morbidity, pathology, or gene keeping those patients behind. 

We have still only seen pooled data for the trial, with responder analysis in “odds ratio” and mean change data for ADAS-COG and CDR-SB. And only for a short 48 weeks. I emphasize “pooled” because in theory, most patients could likely titrate up to 35-45mg. This is why I’m most interested in high dose or high concentration-only charts. Even if the drugs quadruple muscarinic modulation causes transient dizziness or headaches, I believe it is a risk most patients would take if foretold that the drug would provide MEANINGFUL benefit. We already have evidence of this as the vast majority of patients continued to OLE and continued even further to the extended-OLE.

Phase 2 Trial for Anavex 3-71 in Schizophrenia is Imminent 

Bottom-line Up Front (BLUF): On 7 Feb, the trial design for Anavex's 3-71 Schizophrenia trial became available on clinicaltrials.gov. The trial is set to assess approximately 40 patients over 10 or 28 days with two doses (30mg & 60mg), in the United States. The trial features a vast array of endpoints (38) and is supported by the COGNISION consortium, the Cognitive Research Corporation, and Hassman Research Institute. While the trial is exploratory in nature, the inclusion of fluid biomarkers is exciting, and positive outcome there would propel the indication through to a well-deigned phase 2b/3 or phase 3.

Trial Design Notes: 

After analyzing other phase 2 Schizophrenia trials, we found that both the size and length of Anavex's trial to be within established norms (15-125 patients, 2-6 weeks). Looking at the trial endpoints, we most look forward to the PANSS, BACS, CGI-SCH, and biomarkers YKL-40, GFAP, amino metabolites, and of course the Anavex-standard DNA/RNA & whole genome analysis. The biomarkers and genome analysis will objectively correlate PANSS, BACS, and CGI-SCH efficacy to molecular improvement via gene up-regulation. 

Anavex 3-71: 

Having explained differences between Blarcamesine and Anavex 3-71 in the past, below is a brief refresher on why Anavex 3-71 was chosen over Blarcamesine for Schizophrenia. 

• Pros for 3-71:

  • • Massively increase S1R and M1 agonism. Of all the muscarinic receptors, M1 is known to be highly concentrated in the CNS with ties to motor, learning, and memory.

    • Eliminating M3-M4 reduces safety concerns as quadruple modulation of muscarinic receptors has historically been tied to dizziness, headaches, and in some rare cases psychiatric problems.

    • Access to a slew of 5-HT serotonin secondary receptors for psychiatric benefit.

    • Retains good M2 agonism for motor function.

• Con Differentiators from Blarcamesine: 

  • • Lacks SCN2A which is probably the biggest component for autism & seizures.

    • Lacks M3 agonism which is probably one of two greatest factors influencing blood pressure/endothelial aid.

    • Lacks NMDAR, probably the second biggest component for autism.

Overall, as mentioned in previous publications, Blarcamesine is an excellent 'whole body' therapy and has special considerations for autism disorders and epilepsy. Meanwhile, Anavex 3-71 is an ultra-focused cognitive enhancer and psychiatric aid, making it more ideal for Schizophrenia.