2023-05-16 14:56:55 ET
Summary
- Blarcamesine, Aricept, and most probably simufilam are sigma-1 receptor agonists.
- Sigma-1 receptor agonists often produce temporary improvements in cognition for those with mild cognitive impairment and can initially stabilize the disease in mild Alzheimer's disease patients, but not thereafter.
- Blarcamesine appears to nearly stabilize mild Alzheimer's disease over longer periods of time than other sigma-1 receptor agonists, perhaps because it is acting as a direct antioxidant.
- Cassava Sciences faces charges of data manipulation, staged the results of its early trial, and probably has an incorrect mechanism of action, and yet the results might be just good enough for FDA approval.
- Based on yet still publicly unreleased biomarker data and dose dependent responses, the FDA could grant accelerated approval to blarcamesine.
Recently, I compared the relative effectiveness of Anavex's ( AVXL ) blarcamesine, Cassava Sciences' (SAVA) simufilam, and Aricept for the treatment of Alzheimer's Disease ( article ). I noted that blarcamesine and Aricept were sigma-1 receptor agonists and that simufilam was likely a sigma-1 receptor agonist. In this article, more conclusive results of the latter will be provided. Given the same mechanism of action the results for simufilam, blarcamesine, and Aricept should be about the same. The difference is that if blarcamesine also acts as a direct antioxidant, it should slow the advance of Alzheimer's disease during its earlier stages for a longer period of time than simufilam and Aricept. If so, Anavex would have a better treatment and represent a better investment than Cassava Sciences.
In addition to a discussion of the science, I will review recent developments for both companies.
Blarcamesine, Simufilam, and Aricept: A Shared Mechanism of Action
Both blarcamesine and simufilam are methyl amines (N-CH3), and methyl amines appear to be sigma-1 receptor agonists ( analysis , alkyl/methyl amines ). Aricept/donepezil has also been identified as a sigma-1 receptor agonist . About 20 percent of the population does not have a fully functioning sigma-1 receptor, so they receive very limited benefit from drugs that act as sigma-1 receptor agonists. Also sigma-1 receptor agonists become less effective as Alzheimer's disease progresses as protein kinase C which initially drives the disease becomes inactivated due to nitration (sigma-1 receptor agonists inhibit intracellular calcium release which in part limits the activation of protein kinase C). By contrast, most people with functioning sigma-1 receptors and with mild cognitive impairment or mild Alzheimer's disease benefit from these agonists because early on they reduce acetylcholinesterase activity , reduce the production and aggregation of amyloid and the phosphorylation of tau ( study ), and most importantly lessen oxidative stress and neuroinflammation ( figure three ).
Significant (some might say remarkable) improvements in individuals taking Aricept , simufilam , and blarcamesine have been reported. In addition, each drug improves cognitive function by a possibly clinically significant 4 points or more on the Alzheimer's Disease Assessment Scale-Cognition (ADAS-Cog) in some Alzheimer's patients ( Aricept , simufilam , blarcamesine ). The percentage of people who improve by this amount vary depending on the length of the study and baseline scores, but especially early on sigma-1 receptor agonists can temporarily modify disease progression (for example, 68% of those with mild cognitive impairment on simufilam improved at one year, but only 47% improved after the additional inclusion of mild to moderate Alzheimer's disease patients). Following is a comparison of the impact of these sigma-1 receptor agonists on Alzheimer's disease.
Simufilam versus Acetylcholinesterase Inhibitors (including Aricept) for mild Alzheimer's disease (mean change in ADAS-Cog11 scores)
Simufilam : 19.1 to 19.6
Acetylcholinesterase Inhibitors : 21 to 22.5
There is about a .7 point increase in decline for each change in ADAS-Cog scores for simufilam (from a 2.4 improvement from a baseline of 15 to a 4.4 point decline from a baseline of 25.7). So a close apples to apples comparison would be the following:
Simufilam: 21 to 22.4
Acetylcholineserase Inhibitors: 21 to 22.5
There is little to distinguish simufilam from acetylcholinesterase inhibitors in the treatment of Alzheimer's disease.
Simufilam versus Blarcamesine
The comparison between simufilam and blarcamesine becomes somewhat murkier. In its phase 2a trial, participants with the highest concentrations of the drug, who had mild cognitive impairment or mild Alzheimer's disease, and who did not have a sigma-1 receptor gene variant experienced a 1.7 point improvement in Mini-Mental State Examination (MMSE) scores at 57 weeks which is roughly the equivalent of a 2.9 point improvement in ADAS-Cog11 scores. If Cassava Sciences had only included the same populations, the results may have been close to the same.
Even after dealing with confounding variables such as different baselines, timelines, and genetic make-ups, the impact of sigma-1 receptor agonists on Alzheimer's disease appear similar: they can lead to temporary improvements in those with mild cognitive impairment, they can stabilize for about a year those with mild Alzheimer's disease, and they have a minimal impact on those with moderate and severe Alzheimer's disease.
Cassava Sciences: Pitfalls and Landmines
Mechanism of Action
Cassava Sciences touted a recent cancer study for having validated its proposed mechanism of action for simufilam. The researchers who conducted the study, instead, discovered a significantly different mechanism of action for simufilam: the drug reduced the phosphorylation of filamin A rather than restored it to its normal form. One might argue this newest finding is actually good news for Cassava Sciences since in theory reducing the phosphorylation/misfolding of proteins should provide a more effective treatment for Alzheimer's disease than unfolding them. The problem, however, is that misfolded proteins such as amyloid and filamin A are at best secondary contributors to Alzheimer's disease early on, so reducing their phosphorylation would have very little impact on the progression of the disease. Moreover at the very earliest stages of the disease a considerable amount of "toxic" phosphorylated/soluble filamin A has already been converted into non-toxic insoluble filamin A, so targeting the production of phosphorylated filamin A may not be effective at all ( study ). In short, the main beneficial mechanism of action of simufilam- likely its only beneficial mechanism of action - is as a sigma-1 receptor agonist.
Legal Troubles
A Federal judge in Texas recently allowed a lawsuit against Cassava Sciences to go forward. The most serious allegation in the lawsuit is that Cassava Sciences' management/team manipulated phase 2b data (specifically removing a tau outlier from the drug group) and then set up large cash bonuses for themselves tied to rises in the stock's value ( trial record, pp. 5-6 ). The plantiffs while acknowledging that Barbier and other Cassava executives did not collect on these bonuses, argued that the company still benefitted from the subsequent rise in stock prices ( trial record, pp. 24-25 ). The exact legal jeopardy that Cassava's management team is in will not be known until the end of the trial.
Staging of the Disease
In its open label trial, Cassava Sciences staged the disease such that the first fifty patients with mild cognitive impairment/amnestic cognitive impairment/prodromal Alzheimer's disease were processed first whereas the results for the last group which had late mild to early moderate Alzheimer's disease was not announced until 16 months later. The first fifty patients improved by a mean of 3.2 points at one year (2.4 points after further adjustment) whereas the last fifty patients declined by 4.4 points ( press release ). In between, Cassava Sciences averaged the scores for those with mild cognitive impairment with those who had mild Alzheimer's disease, possibly to try to obscure the small decline in the latter group. The company, thus, left the impression for months that individual with mild to moderate Alzheimer's disease had improved by 3.2 points at one year which would indeed have been remarkable achievement if it had been true. As it turned, individuals with less advanced mild Alzheimer's disease declined by .5 points and those with late mild to early moderate Alzheimer's disease declined by 4.4 points. This again is very similar to what one would expect with Aricept ( article ).
Projections
If Cassava Sciences has not truly randomized its Cognition Maintenance Study but instead continued those with the least cognitive impairment on the drug and switched those with the most cognitive impairment onto the placebo, then the results will be meaningless. If a baseline ADAS-Cog score is not provided for both groups that would be a strong indication that results are not reflective of a randomized trial. Simufilam will almost certainly perform better than the placebo groups in its two phase 3 clinical trials but the question is whether it performs the same or better than the current standard of care (acetylcholinesterase inhibitors) for early stage Alzheimer's disease from one year to 18 months.
Given the added uncertainty of how the legal case against Cassava Sciences will be decided, it is hard to project the company's future stock value. High expectations of full exoneration and great future results by a very loyal investor base and further purchases by insiders may keep the stock in decent shape until the phase 3 clinical trial results are released. Those expectations may be partially dashed. However, despite all its baggage, Cassava Sciences may have a drug that has equal efficacy but perhaps less side effects than acetylcholinesterase inhibitors, and thus gain FDA approval. Hence the hold rating.
Anavex: Omissions and Potential
Omissions
Anavex should have provided a mean change in activities of daily living score at its Clinical Trials on Alzheimer's Disease (CTAD) presentation in December 2022. The company could also have provided the data for those maintained on the 30mg dose and the 50mg dose. Anavex did not provide details on rare side effects caused by the drug nor has it specified the severity and duration of the two main side effects: dizziness and confusion. Some or all of this information may be forthcoming later this year.
Mechanism of Action: Antioxidants Can Make a Difference
Anavex notes that Anavex 2-73/blarcamesine reduces oxidative stress and inflammation in Alzheimer's disease (during its early stages). This is consistent with the action of sigma-1 receptor agonists. The company has never stated that blarcamesine is a direct antioxidant.
Blarcamesine is a tetrahydrofuran derivative whose antioxidant capacity is uncertain. The best case that can be made that tetrahydrofurans are strong antioxidants is that they are ready hydrogen and oxygen donors which is precisely what is needed to scavenge the nitro-oxidant peroxynitrite, and to partially reverse oxidative and nitrostative damage.
At one year for mild cognitive impairment and mild Alzheimer's disease adding powerful antioxidants to an acetylcholinesterase inhibitor appears to make little difference (maybe sigma-1 receptor agonists are enough to sufficiently reduce oxidative stress as this point). But after that, the combination of acetylcholinesterase inhibitors and strong antioxidants helps to largely stabilize mild Alzheimer's disease ( retrospective study ). This may be what blarcamesine is doing: its methyl amine group is acting as a sigma-1 receptor agonist and its tetrahydrofuran group is acting as an antioxidant. Blarcamesine nearly stabilized most participants with mild cognitive impairment or mild Alzheimer's disease for 148 weeks, which sets it apart from other treatments for Alzheimer's disease ( figure two ).
The Future
Anavex's CEO Christopher Missling has recently promised to release a clear, detailed analysis of the effects of blarcamesine based on dose. He also said that the phase 2b/3 results are consistent with and complement the phase 2a results. If that is the case those on the 50mg dose of Anavex should perform similar to those in the high concentration group in the phase 2a trial who had mild cognitive impairment or mild Alzheimer's disease and who did not have a "negative" sigma-1 gene variant (an approximate 2.9 point improvement in ADAS-Cog11 scores). That along with good biomarker data should be enough for the FDA to grant accelerated approval for the drug for early stage Alzheimer's disease at least in those who have a fully functioning sigma-1 receptor. Anavex could then sell blarcamesine while it runs a confirmatory trial. And if this is indeed what happens, Anavex is in good shape.
For further details see:
Cassava Sciences' Simufilam, Anavex's Blarcamesine, And Aricept: Shared Mechanisms Of Action In The Treatment Of Alzheimer's Disease