hjalmeida/iStock via Getty Images
Recent results for Annovis’s (ANVS) buntanetap appear to add to the list of treatments that do not help moderate Alzheimer’s disease patients. This list includes anti-amyloid antibody drugs such as Leqembi (Eisai and Biogen) and donanemab (Eli Lilly), and sigma-1 receptor agonists such as blarcamesine (Anavex) (AVXL) and probably simufilam (Cassava Sciences) (SAVA). These treatments can lead to at least some initial seemingly significant improvements in mild Alzheimer’s disease but often not substantially better than the current standard of care. For buntanetap, the mild Alzheimer’s disease group showed a 3.3 point improvement in ADAS-Cog11 scores (Alzheimer’s Disease Assessment Scale-Cognitive Subscale) versus .94 point improvement in the placebo group whereas moderate Alzheimer’s disease patients improved by only .65 points versus a 1.14 decline in those on placebo at twelve weeks (results). One must remember that some patients in both the drug and placebo group are also taking either Aricept for mild Alzheimer’s disease or Namenda for moderate Alzheimer’s disease, so some of the improvement in the drug group are due to these other Alzheimer’s medications whereas most of the decline in the placebo group is due to people on no Alzheimer’s medications. That is why Alzheimer’s drug company executives and statisticians should not be puzzled by the better than expected placebo responses. Moreover, for mild Alzheimer’s disease, buntanetap produced no improvement in activities of daily living and in clinician’s evaluations of disease progression (outcome). For reasons discussed below, I don’t recommend investing in either Annovis or Cassava Sciences while I continue to see a glimmer of hope for Anavex.
There are reasons why so many current drug candidates for Alzheimer’s disease do not help moderate patients, and it is not primarily because the disease is too far advanced to treat at this stage (although that may limit the effectiveness of even the best treatments). The three main explanations is that they are working too far upstream in the process, they do not directly scavenge oxidants, and they do not at least partially reverse oxidative and nitrostative stress.
Many of the currently studied treatments for Alzheimer’s disease work at one point or another in the following pathway.
G protein-coupled receptor or Receptor Tyrosine Kinase
↓
Intracellular Calcium Release and Diacylglycerol
↓
Protein Kinase C Alpha
↓
NMDA Receptor
↓
Hydrogen peroxide (early) and Peroxynitrite
↓
Neuroinflammation
Both the amyloid precursor protein (whose formation is inhibited by buntanetap) and amyloid activate G-protein coupled receptors and via protein kinase C alpha cause oxidative and nitrostative stress and neuroninflammation (which in turn causes more oxidative and nitrostative stress). The amyloid precursor protein and amyloid are two of more than a dozen factors that can overactivate these receptors and thus targeting them has little to no effect on most people (one prominent partial exception is APOE4 carriers who have relatively large amounts of amyloid in their brain).
The following critical observation is for amyloid but likely also applies to the amyloid precursor protein as well:
Malinow’s team found that when mice are missing the PKC alpha gene, neurons functioned normally, even when amyloid was present. Then, when they restored PKC alpha, amyloid beta once again impaired neuronal function. In other words, amyloid beta doesn’t inhibit brain function unless PKC alpha is active (source of quote).
However as Alzheimer’s disease progresses, protein kinase C alpha becomes largely inactivated due to nitration, so treatments that work upstream of protein kinase C alpha become ineffective. The problem then is not overactivation of NMDA receptors by protein kinase C alpha but by excessive amounts of glutamate (which “pool” around these receptors due to the nitration of glutamate transporters).
It is possible that buntanetap and Aricept/donepezil (a sigma-1 receptor agonist that inhibits intracellular calcium release) work at least modestly better in combination than either does separately because together they are inhibiting two steps in the process that via protein Kinase C alpha and NMDA receptors leads to oxidation and nitration rather than just one step. Butanentap leads to a 3.3 point improvement in mild Alzheimer’s disease patients at 12 weeks whereas Aricept leads to a little less than a 2 point improvement.
By contrast, blarcamesine and simufilam share part of the same mechanism as Aricept (i.e. the inhibition of intracellular calcium release). For mild cognitive impairment and mild Alzheimer’s disease, patients on simufilam experienced basically no change in ADAS-Cog11 scores at 52 weeks (graph, p.21), whereas those on blarcamesine declined by .3 points at 57 weeks (table 3). For mild Alzheimer’s disease, patients on Aricept experienced a 1.1 point decline (table 3). It is possible that simufilam and blarcamesine are replacing Aricept in those in the drug group not on Aricept.
The advantage that blarcamesine may have over other treatments for mild cognitive impairment and mild Alzheimer’s disease is that it also seems to act as a direct antioxidant. As a tetrahydrofuran derivative, it appears to donate hydrogen atoms and electrons which is needed to scavenge hydrogen peroxide and peroxynitrite. This donation also helps to partially reverse oxidation and through the production of water helps to de-nitrate proteins. Two of the consequences of this is higher levels of acetylcholine which is needed for the retrieval of short-term memories and the untangling of tau which improves neurotransmissions.
Anavex ran a small trial (almost three years) in which a few people at the highest concentrations of blarcamesine and with mild cognitive impairment and mild Alzheimer’s disease stayed near baseline in terms of cognition and activities of daily living for 148 weeks (almost three years) (figure two). This is what one can expect from compounds that both inhibit the early formation of oxidants, scavenge oxidants, and reverse part of their damage (figure one).
Not much has gone right for Anavex since this small trial. The company never separated the results of the 30mg from the 50mg dose in its phase 2b/3 trial, perhaps because if it did, the numbers would have been too low to establish statistical significance in the 50mg group. Anavex was probably surprised by how many patients in the 50mg dose had to be down titrated or dropped out of the trial due to confusion and dizziness (this may have been caused by the metabolite formed after the scavenging of oxidants). The company has one more shot at Alzheimer’s disease when it releases data for a 96 week open label extension trial (following the 48 week phase 2b/3 trial). Anavex’s Attention-AD trial is scheduled to end this summer with results possibly available this Fall. I expect that the results will parallel/confirm the earlier small phase 2 trial and if so the company can make the case that it is the first company to produce a drug that nearly stabilized mild cognitive impairment and mild Alzheimer’s disease for almost three years in a relatively large number of individuals. This may be enough to gain approval from the European Medicines Agency and at least accelerated approval from the United States Food and Drug Administration. If the results are less than this, Anavex is in serious trouble despite Anavex 2-73/blarcamesine being tested for other conditions and despite having at least one more drug in its pipeline (Anavex 3-71).
Meanwhile, Cassava Sciences and Annovis have to try to box their way out of the Alzheimer’s disease box since their drug candidates are not direct anti-oxidants. In its phase 2 trials, Cassava Sciences defined mild Alzheimer’s disease as between 21 and 26 whereas Annovis probably more accurately defined mild Alzheimer’s disease as between 21-24. For its phase 3 clinical trials, both companies are including patients with even less cognitive impairment: Cassava Sciences from 26 to 27 (although Cassava Sciences did allow 10 patients with scores of higher than 26 in its phase 2 trials) and Annovis from 24 to 28. The adding of more mild cognitive impaired patients will improve results to some degree at least.
Cassava Sciences and Annovis are also trying to box their way out of the Alzheimer’s box by excluding “non-Alzheimer’s disease” patients based on phosphorylated tau levels. The FDA once relied heavily on amyloid to do this. As a result a substantial portion of potential participants who had Alzheimer’s disease but low levels of amyloid (many non-APOE4 carriers and African Americans) were prevented from participating in trials. On the surface, phosphorylated tau (especially phosphorylated tau217) levels seem like a better measurement at least for determining the risk of developing Alzheimer’s disease in mild cognitive impaired patients. But as cognitive decline increases, the validity of the measure may decrease as well (phosphorylated tau217 appears to depend on intracellular calcium release). The goal of trying to improve results by eliminating “non-Alzheimer’s disease patients” may therefore be largely a fool’s errand.
Buntanetap now has to try to raise even more money for an 18 month phase 3 trial for patients with “early” Alzheimer’s disease and mild Alzheimer’s disease. By contrast, Cassava Sciences may have results for a 12 month phase 3 trial for mild cognitive impaired to moderate Alzheimer’s disease patients by the end of this year and the results for an 18 month phase 3 trial sometime next year. In addition, an extension trial may show that the drug keeps those with mild cognitive impairment near baseline for at least two years, which is modestly better than Aricept. But for mild to moderate Alzheimer’s disease, Cassava Sciences past data jiggering is likely to catch up with it. The company initially suggested a 3.2 improvement in those with mild to moderate Alzheimer’s disease at one year (first fifty patients), but the actual results for all 216 patients were a 4.4 decline in those with moderate Alzheimer’s disease, a .5 decline in those with mild Alzheimer’s disease, and an approximate .7 point improvement in those with mild cognitive impairment (one year results).
Given all this, I would recommend selling Cassava Sciences, consider selling Annovis, and buying at least a small amount of Anavex, especially at current price levels.
Editor’s Note: This article covers one or more microcap stocks. Please be aware of the risks associated with these stocks.
