The Cognition Maintenance Study of Simufilam

Addendum and revision 11 May ’23 — Cassava announced today that dosing inthe Cognition Maintenance Study (CMS) is complete — https://finance.yahoo.com/news/cassava-sciences-completes-patient-dosing-131500155.html.  All that remains is to analyze and report the data  which will happen in the third quarter of 2023. The link notes that “The CMS dataset remains locked and blinded. After unlocking, the dataset will be analyzed by outside biostatisticians.”   This could be a game changer and lead to early FDA approval as the CMS study is double blinded.  So it’s worth republishing an earlier (5/22) post on the subject to explain how this might occur.

Cassava’s Cognition Maintenance Study may prove Simufilam works

The FDA will approve less than perfect therapies if there is nothing useful for a serious condition.  Consider the following from Proc. Natl. Acad. Sci. vol. 119 e2120512119 ’22

“KRAS is the most frequently mutated oncogene in human cancer, with mutations detected across many lineages, particularly in the pancreas, colon, and lungs. Among the most commonly activating KRAS mutations at codons 12, 13, and 61, G12C occurs in ∼13% of lung and 3% of colorectal carcinomas and at lower frequencies in other tumors.

“In locally advanced or metastatic non–small-cell lung cancer (NSCLC) patients with KRAS^G12C mutations who have received at least one prior systemic therapy”  treatment with sotorasib resulted in the following “objective response  in 37.1% of the patients, with a median duration of response was 11.1 months.”   This is hardly a cure, but nonetheless “This promising anticancer activity has resulted in accelerated approval from the US Food & Drug Administration”

Which brings me to the current CMS study from Cassava Sciences.  I’ll let them speak for themselves. https://finance.yahoo.com/news/cassava-sciences-reports-first-quarter-130000375.html

“Cognition Maintenance Study (CMS) – on-going
In May 2021, we initiated a Cognition Maintenance Study (CMS). This is a double-blind, randomized, placebo-controlled study of simufilam in patients with mild-to-moderate Alzheimer’s disease. Study participants are randomized (1:1) to simufilam or placebo for six months. To enroll in the CMS, patients must have previously completed 12 months or more of open-label treatment with simufilam. The CMS is designed to evaluate simufilam’s effects on cognition and health outcomes in Alzheimer’s patients who continue with drug treatment versus patients who discontinue drug treatment. The target enrollment for the CMS is approximately 100 subjects. Over 75 subjects have been enrolled in the CMS and 35 have completed the study.”

Even though the open label study was not randomized, this one will be.

Only someone who has actually taken care of  patients would know the following.  People who are getting no benefit from a drug will soon stop taking it.  This was particularly true for my experience with Cognex for Alzheimer’s disease.

Which is exactly why the fact that 75 patients who’ve been on Simufilam have decided to continue on in the CMS study.  Presumably they feel they are getting some benefit.

There are two possible hookers to this

l. The patients are being paid to enter CMS

2. The original cohort was 200, not all of whom have finished the 1 year.  So we don’t know how many could have been in CMS but chose not to.

As I discussed in an earlier post, the most impressive thing (to me at least) was that at 9 months 5/50 had significant improvement in their cognition — here’s a link — https://luysii.wordpress.com/2021/08/25/cassava-sciences-9-month-data-is-probably-better-than-they-realize/.

The CMS study should give us an idea of how they fared at 1 year and  at 18 months.

If:

l. gains in cognition were maintained on Simufilam

2. gains in cognition were lost off Simufilam

FDA approval should follow quickly.

Results on the 75 will be available this year.   Also available this year will be 1 year results on all 200 entering the open label study.

There are two other double blind studies in progress which will provide  more definitive answers, but they are far from full and will take much longer to complete.  So stay tuned.

The fact that not everyone responds to Simufilam is irrelevant to its eventual FDA approval

A very intelligent friend does not share my optimism about Simufilam.

“Is the data really that positive? ADAS-Cog mean scores changed minimally over 1 year in patients with mild-to-moderate Alzheimer’s disease.  47% of patients improved ADAS-Cog over 1 year by 4.7 points. But 23% of patients declined by <5 points. Mild patients responded better than patients with moderate Alzheimer’s.”

Why are these thoughts irrelevant to the eventual approval of Simufilam by the FDA?

First: no drug for anything works for everyone with the condition

Second: The assumption that Alzheimer dementia is a single disease is based on just that: an assumption.

An example: When I was running a muscular dystrophy clinic in MonrN (’71 – ’87), we saw something called limb girdle muscular dystrophy , in which the patients were weak primarily in muscles about the shoulders and hips. Now we know that there are at least 13 different genetic causes of the disorder.

If the clinical picture of Alzheimer’s disease is due to multiple causes, it is unsurprising that Simufilam doesn’t help all of them.

Also it is time for some humility about our knowledge about Alzheimer’s disease.  We have misunderstood what the senile plaque of Alzheimer’s disease really is for 111 years — see the following post written 12/22 — https://luysii.wordpress.com/2022/12/13/111-years-of-study-of-the-alzheimer-plaque-still-got-it-wrong-until-now/

Third (and probably the most relevant for FDA approval):  Less that perfect drugs will be approved if every other treatment is worse.

The example of immune checkpoint blockade therapy for cancer is particularly relevant.

Some absolutely spectacular results for the therapy has led to the approval of 6 different drugs in this class (all of them monoclonal antibodies against proteins involving the immune system).

One example [ Cell vol. 162 pp. 1186 – 1190 ’15 ]:  “20% of metastatic melanoma patients are cured with Ipilimumab, a fully humanized anti-CTLA4 monoclonal antibody.”

Would that results like this were the rule not the exception. Unfortunately — [ Nature vol. 565 pp. 43 – 48 ’19 ] “Most patients with cancer either do not respond to immune checkpoint blockade or develop resistance to it.”

So what.

Immune checkpoint blockade, despite being less than perfect,  is  still being offered to cancer patients, just the way Simufilam with its nearly 50% chance of improvement at 1 year should be offered to Alzheimer patients.  

Why Cassava’s Simufilam results are not a placebo effect

Any open label study without controls is subject to the reasonable criticism that any benefits seen are due to the placebo effect.  Neurologic and Psychiatric disease trials can have large (33%) placebo effects (e.g. migraine, depression).

This is very unlikely to be the case with the 1 year results of the open label trial of Simufilam in 200 patients with Alzheimer’s disease.  “47% of patients improved on ADAS-Cog over 1 year, and this group improved by 4.7 points”

It is based on my clinical experience with a drug for Alzheimer’s disease released 30 years ago — Tacrine (Cognex).  Initially it was touted as helping Alzheimer’s disease (e.g. improving thinking and memory) although later it was held to slow the decline.   The local medical school advertised it aggressively (primarily as a marketing device).

So I put my Alzheimer patients on Cognex.  I wanted them to get better. They wanted to get better, and their families and caregivers certainly did.  Just about all of them thought it might have helped on followup visits in the first month.  I couldn’t see much difference.  By the second month, they weren’t sure, and later in the first year they didn’t think it helped, and most weren’t using the drug after 1 year.

Not only that, but I was in a call group with 4 other neurologists, and they saw exactly the same thing.  I was practicing in an area with a catchment area of over a million people, and every local neurologist I talked to had the same experience. People thought that “Cognex helped” for a month or two and then they didn’t

This a classic example of a placebo effect.  Moreover it occurred in a therapeutic trial for Alzheimer’s disease.  Crucially, the placebo effect was quite transient and  absent at 1 year.

This is why the Simufilam results mentioned above are not a placebo effect.

Those not interested in neuropharmacology can stop at this point.  There were excellent clinical and theoretical reasons for the use of Tacrine.

Clinically it was apparent that drugs that blocked the effects of the neurotransmitter acetyl choline on one type of its receptors (muscarinic) profoundly impaired memory. Scopolamine is one such drug.  One of the earliest and most invariable symptoms of Alzheimer’s disease is deficient memory.

That’s the clinical part.  Here’s the theory.  So logically, increasing acetyl choline should help memory.  How to do this?  Well there are enzymes that break acetyl choline down (the acetyl cholinesterases).  So by inhibiting cholinesterases, acetyl choline levels in the brain should increase, and memory should be improved.

Impeccable logic and theory.  Unfortunately, like many such it didn’t work.

Cassava shorts should be worried

Yesterday, 1 November ’22, a blockbuster  article was published in the Journal of Clinical Investigation (JCI) written by its editor Elizabeth McNally — https://www.jci.org/articles/view/166176.

It is just over a year ago since the first of the articles attacking Cassava Sciences appeared.  The first was in the New Yorker which profiled Jordan Thomas as the second coming of Christ for exposing supposed fraudulent data published by Cassava principals —

Radden Keefe P. The Bounty Hunter. The New Yorker. Updated January 17, 2022. Accessed October 11, 2022. https://www.newyorker.com/magazine/2022/01/24/jordan-thomas-army-of-whistle-blowers.

Nowhere in the article was it mentioned that the ‘whistle-blowers’ stood to gain financially because they had shorted the stock.

 

Addendum 2 Nov— see comment by Elizabeth Bik at the end — this is incorrect.  It was mentioned that the whistleblowers had short positions in the stock.

There were similar articles in Science — 2022;377(6604):358–363

and the New York Times https://www.nytimes.com/2022/04/18/health/alzheimers-cassava-simufilam.html.

They relied on the same assertions given to the FDA asking that the clinical trials be stopped because of ‘danger’ to the patients.

In none of these articles was it mentioned that those claiming fraud by the Cassava investigators were short the stock.

Addendum 2 Nov— see comment by Elizabeth Bik at the end — this is incorrect.  It was mentioned that the whistleblowers had short positions in the stock.

It’s worth reading McNally’s article completely.  It isn’t very long.

A few highlights (“the Journal” refers to the JCI)

“Throughout 2022, the Journal has been repeatedly contacted to comment on the 2012 JCI paper. Although we cannot be certain, there now appear to be new “short and distorters.” A recent round of emails was sent simultaneously to multiple journals and editors, identifying 25 articles with potential problems and providing recommendations on how the journals should respond. Importantly, these accusatory emails do not identify any financial conflicts of interest on the part of the whistleblowers. The emails insist that an investigation begin within 24 hours and request that the journals update them on investigative progress. As an editor, I am expressing concern because this represents a new means of manipulating the scientific publishing industry.”

So journal editors are like docs. They talk to each other to find out what’s really going on.  It is likely that McNally called up other journal editors to find out if her experience was common.

Here is why those sending the eMails should not sleep well of a night.

“Last, if the Journal uncovers allegations made for the purposes of stock manipulation, with evidence of misinformation, the JCI may elect to express its concern to the US Securities and Exchange Commission or the Department of Justice.”

It’s about time.

Science, The New Yorker, and the New York Times are guilty of very sloppy reporting.

Whether the ‘whistle-blowers’ are guilty of anything will be determined by the suits (from investors losing money on Cassava, or perhaps Cassava itself) which are almost sure to follow.

As some of you know, I think Cassava’s data is even better than they realize. Be warned the following link is long, detailed and will require your concentration  —

Cassava Sciences 9 month data is probably better than they realize

Why Cassava’s 1 year results should allow compassionate use of Simufilam

Cassava reported results on 100 Alzheimer patients in an open label (e.g. no controls) trial of Simufilam for 1 year — https://finance.yahoo.com/news/cassava-sciences-reports-second-quarter-131500494.html.  The average results were unimpressive (to the uninitiated) with only a minimal average overall improvement of an ADAS-Cog11 score of 1.5 points.  This is probably why the stock (SAVA) dropped a point yesterday after the news.  Since everything turns on ADAS-Cog11 here is a link to a complete description — https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5929311/.  The test takes about 45 minutes placing it out of reach of a busy practicing clinical neurologist.

Why is even the 1.5 point improvement impressive to the initiated (me)?  Over 32 years in clinical neurology, I’d estimate that I saw at least 1 demented patient each week.  Now probably only 300 or so of the 1,664 were followed for a year.  Guess what?  None of them remained stable for a year, and all got worse.  Absolutely none of them  ever got better after a year.  So at least some stabilization of the disease is possible for a year.  The statistics say that Alzheimer patients lose 5 points a year on ADAS-Cog.

But that’s pretty small beer.  Who wants to keep a demented patient around but stable.  Here is the remarkable part of the Cassava results at a year.

63% of the 100 Patients Showed an Improvement in ADAS-Cog11 Scores, and This Group of Patients Improved an Average of 5.6 Points (S.D. ± 3.8). The statistics say that Alzheimer patients lose 5 points a year on ADAS-Cog.

This is unprecedented and is a strong argument for quick approval of Simufilam (or at least compassionate use).

The cynic will say that I’m just looking at the happy part of the Bell curve.  There must have been people who declined to average the improvement in the 63% down to a measly 1.5 points on the ADAS-Cog.

This is where clinical experience comes in.  No drug helps everyone with a given disease.  “Only 20% of cancer patients respond long term to a type of immune checkpoint blockade (of PD-1)” Science vol. 363p. 1377 ’19.  Nonetheless immune checkpoint blockade of several types was approved by the FDA, simply because there was nothing better available.

So if nearly 2/3 of Alzheimer patients will improve at one year on Simufilam, why not  let the FDA offer it to them now under compassionate use.

 

 

Cassava’s Cognition Maintenance Study may prove Simufilam works

The FDA will approve less than perfect therapies if there is nothing useful for a serious condition.  Consider the following from Proc. Natl. Acad. Sci. vol. 119 e2120512119 ’22

“KRAS is the most frequently mutated oncogene in human cancer, with mutations detected across many lineages, particularly in the pancreas, colon, and lungs. Among the most commonly activating KRAS mutations at codons 12, 13, and 61, G12C occurs in ∼13% of lung and 3% of colorectal carcinomas and at lower frequencies in other tumors.

“In locally advanced or metastatic non–small-cell lung cancer (NSCLC) patients with KRAS^G12C mutations who have received at least one prior systemic therapy”  treatment with sotorasib resulted in the following “objective response  in 37.1% of the patients, with a median duration of response was 11.1 months.”   This is hardly a cure, but nonetheless “This promising anticancer activity has resulted in accelerated approval from the US Food & Drug Administration”

Which brings me to the current CMS study from Cassava Sciences.  I’ll let them speak for themselves. https://finance.yahoo.com/news/cassava-sciences-reports-first-quarter-130000375.html

“Cognition Maintenance Study (CMS) – on-going
In May 2021, we initiated a Cognition Maintenance Study (CMS). This is a double-blind, randomized, placebo-controlled study of simufilam in patients with mild-to-moderate Alzheimer’s disease. Study participants are randomized (1:1) to simufilam or placebo for six months. To enroll in the CMS, patients must have previously completed 12 months or more of open-label treatment with simufilam. The CMS is designed to evaluate simufilam’s effects on cognition and health outcomes in Alzheimer’s patients who continue with drug treatment versus patients who discontinue drug treatment. The target enrollment for the CMS is approximately 100 subjects. Over 75 subjects have been enrolled in the CMS and 35 have completed the study.”

Even though the open label study was not randomized, this one will be.

Only someone who has actually taken care of  patients would know the following.  People who are getting no benefit from a drug will soon stop taking it.  This was particularly true for my experience with Cognex for Alzheimer’s disease.

Which is exactly why the fact that 75 patients who’ve been on Simufilam have decided to continue on in the CMS study.  Presumably they feel they are getting some benefit.

There are two possible hookers to this

l. The patients are being paid to enter CMS

2. The original cohort was 200, not all of whom have finished the 1 year.  So we don’t know how many could have been in CMS but chose not to.

As I discussed in an earlier post, the most impressive thing (to me at least) was that at 9 months 5/50 had significant improvement in their cognition — here’s a link — https://luysii.wordpress.com/2021/08/25/cassava-sciences-9-month-data-is-probably-better-than-they-realize/.

The CMS study should give us an idea of how they fared at 1 year and  at 18 months.

If:

l. gains in cognition were maintained on Simufilam

2. gains in cognition were lost off Simufilam

FDA approval should follow quickly.

Results on the 75 will be available this year.   Also available this year will be 1 year results on all 200 entering the open label study.

There are two other double blind studies in progress which will provide  more definitive answers, but they are far from full and will take much longer to complete.  So stay tuned.

Why most criticisms of Cassava Sciences are irrelevant, and two which are not

Most criticisms of Cassava Sciences are irrelevant (see later), but there are two which could doom the company.  Here they are.

Both criticisms involve the results of the 1 year open label trial of Simufilam. You will recall that patients improved on average.  The study was criticized for cherry picking patients and data.  However a closer reading of the results found that 5/50 patients improved their ADAS-Cog Score by 50% after a year on the drug — for details please see https://luysii.wordpress.com/2021/08/25/cassava-sciences-9-month-data-is-probably-better-than-they-realize/.

These results are spectacular and absolutely nothing like this has been seen in untreated (or even treated) Alzheimer patients.

So what are the criticisms? Forget cherry picking of the data, as Cassava didn’t do it, and even if they did these results are so spectacular that cherry picking is irrelevant.

Doom criticism #1 — the data are false and made up by Cassava.  In Montana this was a practice in the gold mining days, when a few nuggets were sprinkled in an otherwise worthless mine to delude the unwary.  This was called salting the mine.

This is unlikely as the data were reported from sites producing the data, not Cassava.  Further Lindsay Burns told me that the 5 spectacular results came from 5 different sites administering the drug.  I was worried that one site was screwing up and measuring ADAS-Cog incorrectly.

Doom criticism #2 — People are paid to enter these studies.  How much isn’t clear.  Could the 5 have faked the dementia and then gotten better on subsequent testing?  Cassava tests to make sure  subjects are actually taking the drug.   This is a possibility but remote.

Now for the irrelevant criticisms.

#1 The mechanism of action of simufilam makes no sense.  It certainly is quite radical — for details see — https://luysii.wordpress.com/2021/03/25/the-science-behind-cassava-sciences-sava/.

Why is it radical?  The nicotinic cholinergic receptor is an ion channel.  Binding of another protein to it is not postulated to open the channel, but to alter its binding to another protein (filaminA) inside the cell. This is a totally new mechanism for drug action on ion channels — binding to the channel so its binding to something inside the cell changes.  Simufilam is held to change the conformation of filamin A.

Well that’s pretty damning.  Why can it be ignored?

Because the mechanism of the aBeta peptide being the cause of Alzheimer’s has a huge amount of beautiful chemistry behind it — for details see — https://luysii.wordpress.com/2021/10/21/amyloid-structure-at-last-3-the-alzheimer-mutations/

Here’s a quote  from the post — skip this if you don’t have the necessary biochemical background, but to anyone with knowledge of protein chemistry it is beautiful and essentially confirms the amyloid hypothesis of Alzheimer’s disease

“In 2007 there were 7 mutations associated with familial Alzheimer’s disease (10 years later there were 11). Here are 5 of them.

Glutamic Acid at 22 to Glycine (Arctic)

Glutamic Acid at 22 to Glutamine (Dutch)

Glutamic Acid at 22 to Lysine (Italian)

Aspartic Acid at 23 to Asparagine (Iowa)

Alanine at 21 to Glycine (Flemish)

All of them lower the energy of the amyloid fiber.

Here’s why

Glutamic Acid at 22 to Glycine (Arctic) — glycine is the smallest amino acid (side chain hydrogen) so this relieves crowding.  It also removes a negatively charged amino acid next to the aspartic acid.  Both lower the energy

Glutamic Acid at 22 to Glutamine (Dutch) — really no change in crowding, but it removes a negative charge next to the negatively charged Aspartic acid

Glutamic Acid at 22 to Lysine (Italian)– no change in crowding, but the lysine is positively charged at physiologic pH, so we have a positive charge next to the negatively charged Aspartic acid, lowering the energy

Aspartic Acid at 23 to Asparagine (Iowa) –really no change in crowding, but it removes a negative charge next to the negatively charged Glutamic acid next door

Alanine at 21 to Glycine (Flemish) — no change in charge, but a reduction in crowding as alanine has a methyl group and glycine a hydrogen.

As a chemist, I find this immensely satisfying.  The structure explains why the mutations in the 42 amino acid aBeta peptide are where they are, and the chemistry explains why the mutations are what they are. ”

Evidence just doesn’t get any better than this.

So we have  beautiful convincing evidence that amyloid from the aBeta protein causes Alzheimer’s disease.  EXCEPT  that— innumerable trials of getting rid of the amyloid in the Alzheimer brain have not helped and often made things worse.

So criticizing Cassava’s theory of why Simufilam appears to do what it does is like a med school classmate (he went to the University of Chicago) who was always saying — “That’s how it works in practice, but how does it work in theory?”

#2 There are problems with the electrophoresis data of years ago, and with the biomarkers of Alzheimer’s disease. Possibly true, but people don’t visit doctors because of abnormal biomarkers.  Such criticisms are irrelevant to the therapeutic results Cassava has reported.

So it’s time to proceed with the studies currently entering patients.  It is important to note that the FDA has approved the study and the way it will be done, Cassava will stand or fall on the placebo controlled study.  They won’t be asked to do a new one.

According to sources familiar with the matter

Cassava Sciences tanked today due to reports the the SEC was investigating.  Here is a direct quote from the Wall Street Journal  — https://www.wsj.com/articles/cassava-sciences-alzheimers-sec-investigation-11637154199?siteid=yhoof2

“The Securities and Exchange Commission is investigating claims that Cassava Sciences Inc., the sixth-best performing U.S. stock this year, manipulated research results of its experimental Alzheimer’s drug, according to people familiar with the matter.”

This has all the appearances of a plant by the shorts.  Why?

Both my late mother in law and father in law were lifelong employees of the SEC.  He was their chief financial analyst, and she authored the pension study which resulted in them being transferrable when people switched jobs (yes it wasn’t always that way !).  I was amazed to hear from them that the SEC burned or shredded all their trash, because the information about what the SEC was investigating was so valuable (to both longs and shorts).

I’d never heard of the SEC leaking things but apparently it has happened in the past (J. P. Morgan — https://archives.cjr.org/the_kicker/sec_investigation_media_leaks_reuters.php).  How often this happens, I don’t know as I don’t follow the market that much, but this is more than passing strange.

Consider how many times you’ve heard the phrase “persons familiar with the matter” or “anonymous sources” in the past 4 years, and think of how often they’ve turned out to plants and far from true.

Update

Source: https://www.sec.gov/enforce/how-investigations-work.html

How Investigations Work

The Enforcement Division assists the Commission in executing its law enforcement function by recommending the commencement of investigations of securities law violations, by recommending that the Commission bring civil actions in federal court or before an administrative law judge, and by prosecuting these cases on behalf of the Commission. As an adjunct to the SEC’s civil enforcement authority, the Division works closely with law enforcement agencies in the U.S. and around the world to bring criminal cases when appropriate.

The Division obtains evidence of possible violations of the securities laws from many sources, including market surveillance activities, investor tips and complaints, other Divisions and Offices of the SEC, the self-regulatory organizations and other securities industry sources, and media reports.

All SEC investigations are conducted privately.

The SEC is not part of the Justice Department whose policy it is never to disclose an investigation.  It appears that the SEC’s policy is the same, so whatever came out today was very likely not an SEC announcement (and likely a plant by the shorts).

Addendum 18 November:  The Wall Street Journal article on the SEC probe appears on the front page of today’s paper.  The third paragraph is just one sentence. “An  SEC spokeswoman declined to comment.”  The article goes on to talk about the public petition to the FDA filed in August.  The article mentions that David Bredt is a former neuroscience research chief at Johnson and Johnson and Eli Lilly (both of which have developed and are developing Alzheimer’s drugs — something the article fails to mention).

So clearly the article is a hit piece on Cassava.  Would you like to take a guess who the “people familiar with the matter” in the first paragraph are?

 

Cassava Sciences — the clinical reality underneath the stock gyrations.

The stock of Cassava Sciences (symbol SAVA) has undergone some wild gyrations this year.  On 14 September it traded at 41.70, today just two weeks later it is trading in the upper 60s.

The important thing to keep in mind, is that 1 year out on treatment with SAVA’s drug Simufilam 50 patients with mild Alzheimer disease were (as a group) slightly improved.  This is absolutely unprecedented.  The best that previous therapy could accomplish was a slightly slower rate of decline — see arshttps://science.sciencemag.org/content/sci/373/6555/624.full.pdf — for a recent review of other therapy attempts.  So Cassava’s results are unprecedented.   While Alzheimer (and other dementia) patients fluctuate from day to day (like the tides from minute to minute) at the end of a year they are all worse.

These results have not been attacked, unlike their data on the effect of Simufilam on biomarkers which has been criticized by a person of standing — Elizabeth Bik — https://scienceintegritydigest.com/2021/08/27/cassava-sciences-of-stocks-and-blots/#more-2692.

But that’s irrelevant and guilt by association at best.  As a clinical neurologist, no one was ever brought to see me because of their biomarkers.

They have released part of their 1 year results — https://www.cassavasciences.com/news-releases/news-release-details/cassava-sciences-announces-top-line-results-12-month-interim.  There is a lot more that I’d like to know, but a press release is not a detailed scientific paper.

What follows is a lot of commentary and speculation about the 1 year data which we haven’t seen yet.

The results concern the first 50 patients to complete one year on the drug.  The dropout rate is stated to be under 10%.  Presumably this includes death, in a cohort (presently at around 200) with a significant mortality.  It would be interesting to know how many patients on entry made it to one year.

As a clinical neurologist I was particularly impressed with part of their data at 9 months.  Here’s a link — keep it handy — https://www.cassavasciences.com/static-files/13794384-53b3-452c-ae6c-7a09828ad389.

They measured cognitive changes by something called ADAS-Cog — a full description can be found in the following post — https://luysii.wordpress.com/2021/08/25/cassava-sciences-9-month-data-is-probably-better-than-they-realize/

ADAS-Cog score counts errors, so a perfect score would be 0, and a terrible score would be 70.  The range of deficit on entry was 16 – 26 (but possibly on something else called the MMSE) — this is what the 1 year results used.  The 9 month results used ADAS-Cog.  Perhaps they are actually the same thing — I don’t know.

On the link — https://www.cassavasciences.com/static-files/13794384-53b3-452c-ae6c-7a09828ad389 — look at the diagram titled “Individual Patient Changes in ADAS-Cog (N = 50).

There were 5 patients out of 50 at 9 months with improvements of 11 – 14, which would mean that they were pretty close to normal if their entry score was 16 and 50% improved if their score was 26.  From here out I’m just calling them ‘the 5’.

The 9 month report doesn’t discuss this, and only a clinician would know, but this is the way neurologic patients respond to treatment.  Some do extremely well while others have no effect.  Why?  It’s probably because not really understanding causation, we classify patients clinically (it’s all docs have after all).

I ran a Muscular Dystrophy Clinic for 15 years back in the day.  The Muscular Dystrophy Association was founded by parents of weak kids.  They didn’t know that some weakness was due to the muscle itself (what we’re now calling muscular dystrophy), some was due to disease affected the nerves from the spinal cord to the muscle (what we call a neuropathy now) and others were due to disease of the cells in the spinal cord giving rise to the nerves to the muscle (motor neuron disease).  That all came later.

It is quite presumptuous to say that Alzheimer’s disease is just one thing.  Perhaps the 5 patients doing so very well had it from a different (as yet unknown) cause than the other 45.  Even so such a treatment would be worth having.

So here are a few questions for the folks at Cassava about their data

l. Some 16 different sites were involved in the open label study.  Were all of ‘the 5’  from the same site (doubtful — but if true, perhaps they tested ADAS-Cog differently, casting doubt on these results).

2. What were the ADAS-Cog scores initially on ‘the 5’.

3. What happened to ‘the 5’ in the past 3 months (did they maintain improvement, slide back, or improve further?)

4. We must have lots more people passing the 3, 6, 9 month markers.  Have their results paralleled that of the first 50 reaching the mileposts?   It would be very useful to know if there are now more than 5 with improvements over 10 in ADAS-Cog at 9 months.

The slightly slowing of improvement at 1 year relative to 9 months is typical of neurologic disease.  When L-DOPA was first available in the USA in 1970, some patients because so normal that you couldn’t tell they had Parkinson’s disease, and for a few years, neurologists (myself included) thought we were actually curing the disease.  Of course we weren’t and the underlying pathology of Parkinsonism (death of neurons using dopamine) continued unabated.  The L-DOPA just helped the surviving neurons function more efficiently.  Something similar may be going on with Simufilam and Alzheimer’s.

Now for some blue sky about Simufilam. Just as the gray hair on the head of an 80 year old looks the same under the microscope as one from a prematurely gray 30 year old, the brain changes of Alzheimer’s disease (the senile plaque)  are the same regardless of the age of onset.  Assuming that the senile plaque is in someway related to dementia (despite the lack of effect of therapies trying to remove it) and given that we all accumulate a few as we age, could Simufilam improve cognition in the elderly?   Would it then be intellectual viagra and the blockbuster drug of all blockbuster drugs.

 

What controls should Cassava Sciences use for their open label trial?

MDs gradually woke up to the fallacy of using historical rather than concurrent controls particularly in studies of therapies to prevent heart attack and stroke, as the rates of both dropped significantly in the past 50 years, and survival from individual heart attacks and strokes also improved.

An open label trial is just that, no placebo, no controls.  Such trials are done in the exploratory phase of drug development to look for side effects and (hopefully) therapeutic effects.

Cassava Sciences has been attacked because their open label study of Simufilam had no controls.  Duh !

Here is a suggestion for the concurrent controls for the Cassava study:  the Biogen study leading to approval of aducanumab (Aduhelm).  It’s a little hard to find out exactly when it was done, but it certainly was within the past 10 years. Here is a link to an article on Alzheimer therapy from Science — https://science.sciencemag.org/content/sci/373/6555/624.full.pdf

Cassava’s work is nowhere to be found.  The article contains the following

“Although the marked decrease in amyloid deposits can be viewed as biological evidence of disease modification, this was accompanied by a decidedly mixed outcome on cognitive testing, with one aducanumab trial (EMERGE, NCT02484547) meeting its prespecified primary and secondary endpoints at the highest dose, whereas the other (ENGAGE, NCT02477800) did not achieve them.”

So use Biogen’s data on aducanumab as the placebo control (which I and the FDA advisory committee think it is).  There is a reason the entire committee resigned after the FDA approved the drug.