Cassava Sciences: Open Label Study Data Unconvincing But Indicative Of Good Things

Cassava Sciences (NASDAQ:SAVA) has gone down about 30% since November, the last time I covered it (this article has not been published). The reason for the decline could have been attributed to the data drop on January 24 when the company reported unconvincing results from a phase 2 study of simufilam; however, that does not hold water because, one, there was no sudden decline on that particular day, and two, there's been a steady decline since the highs it attained on September 28, at $51+.

That rally was led by Biogen and Eisai's positive lecanemab data, which saw a number of Alzheimer's players spiking. SAVA was one of them, and the steady decline in the last 5 months is simply a correction.

This is good in a way because it tells me that the market did not react to the January simufilam data too negatively. Key data from this trial was as follows:

• ADAS-Cog mean scores changed minimally over 1 year in patients with mild-to-moderate Alzheimer's disease treated with open-label simufilam tablets.

• 47% of patients improved on ADAS-Cog over 1 year, and this group improved by 4.7 points. An additional 23% of patients declined less than 5 points on ADAS-Cog over 1 year, and this group declined by 2.5 points.

• Mild patients responded better than patients with moderate Alzheimer's disease.

• Simufilam was safe, well tolerated.

Let us begin by taking the data at its face value. In order to assure a wary market that they can safely do that, Cassava went out of its way to include the following Chain of Custody for the clinical data in its press release:

Investigator sites collected clinical data from study patients. Sites entered their clinical data directly into an electronic data capture system managed by an independent, outside data management vendor. The data management vendor also maintains the clinical database. The data management vendor transmitted the clinical database directly to Pentara Corporation, an independent, outside consulting firm that specializes in complex statistical analysis of clinical trial results. Suzanne Hendrix, PhD, CEO of Pentara, has >150 peer-reviewed publications of clinical trial results and statistical approaches for clinical trials, many focusing on statistical methodology for Alzheimer's disease.

Coming now to the data itself, the company's principal claim from this data seems to be that simufilam is helping some patients prevent cognitive decline. Adas-Cog, or Alzheimer's Disease Assessment Scale-Cognitive Subscale, is designed to assess the level of cognitive dysfunction at the time of the scoring, so two such tests taken over a period of one year can assess the decline in cognitive abilities in these patients. Although it is hard to pinpoint numbers, historical data as provided in an article in JAMA Neurology says:

The expected decline varies with the degree of cognitive impairment of the study population, but it is generally 4 to 6 points per year, while the standard deviation of the change, which increases with the duration of follow-up, is roughly 6 to 9 points.

Other similar studies:

ADAS-Cog 11 scores of subjects with mild AD declined 3.5 points over 12 months in our analysis. - 1

A total of 134 publications (140 trials) were included in this model-based meta-analysis. The typical disease progression rate was 5.82 points per year. - 2

On the other hand, for improvement, here's data from a somewhat aged study (2007):

In 6-month anti-dementia drug trials, a 4-point change in the Alzheimer's Disease Assessment Scale-Cognitive Subscale (ADAS-Cog) is held to be clinically important.

Going back to the data here, contrast this with the 47% of patients with a 4.7 point improvement, and 23% of patients with a 2.5 point decline - both numbers fall well outside the historical range, i.e., they show that the drug works. As Suzanne Hendrix, CEO of Pentara, noted:

The improvement in ADAS-Cog over 1 year in mild patients taking simufilam is well outside the expected range of historic placebo decline rates from numerous other studies.

So far so good, but there are a number of issues with the design and outcome of this study that does not produce conviction. One, this is an open label study, so patients knew what they were taking, drug or placebo. This is severely problematic because Adas-Cog is a very "subjective" test. Patients are given a number of tasks like word recall, orientation (knowing day and month etc), language ability and comprehension, and so on. You cannot measure these things in a test tube. It is widely recognized that if patients know whether they took a drug or a placebo, their responses may vary. Thus, and as the company says:

The open-label design and size of this study may introduce clinical or statistical bias or may generate results that may not fully distinguish between drug effects and random variation

…this open label design is merely, and barely, indicative. In a dementia disease, besides safety and tolerability, I doubt there's much use of a non-randomized study.

The other related issue is there are no p-values provided. In a trial like the one here, the so-called null hypothesis is that the drug has no effect. This hypothesis is tested and aimed to be proven wrong. This is done through analysing observed events in the trial; here, an Adas-Cog of a particular patient is such an event. A p-value tells us whether an event - like a positive (LOW) mean change in adas-cog score at baseline and 12 months - occurred by chance. The lower the p-value, the more probable it is that the said event did not occur by chance, i.e., it was an effect of the drug. Thus, p-value is very important in understanding drug efficacy, and without it, evidence from trial data is just anecdotal. It is possible that the company will provide more granularity in published material, but at least as of now, there is no such clarity.

Another problem with the trial was the large number of study discontinuations - 34 patients out of a total enrollment of 200 patients is 17%, which is pretty high. I do not have statistics handy for rate of discontinuations in Alzheimer's trials, but I would bet that 17% is not small. The company needs some explaining to do. The reasons they gave - withdrawal of informed consent (N=14), adverse events (N=13) and patient non-compliance (N=7) - need more details.

Cassava has two phase 3 trials running - RETHINK-ALZ and REFOCUS-ALZ, with enrollment targets of 750 and 1000 patients respectively, both trials under Special Protocol Assessments from the FDA. An SPA means that the FDA has accepted, in theory, the study design, and if the molecule meets the primary endpoint, then the molecule will be approved. The primary endpoint here, in both trials, is the same - ADAS-Cog12 (a cognitive scale) and ADCS-ADL (a functional scale). 650 patients have been enrolled so far; analysts estimate the two trials will be fully enrolled by next year.

In other news, last year Cassava filed a lawsuit in the US District Court for the Southern District of New York against defendants David Bredt, Geoffrey Pitt, Quintessential Capital Management LLC, Adrian Heilbut, Jesse Brodkin, Enea Milioris and Patrick Markey for an alleged "short and distort" campaign targeting the company and its phase 3 Alzheimer's drug simufilam. The November 2 Cassava lawsuit is docketed here.

Financials

SAVA has a market cap of $1.1bn, a cash balance of $201mn and a short interest of nearly 28%. The company guides towards net cash used for operations for the first half of 2023 to be approximately $45 to $50 million. Research and development expenses were $68 million in the previous quarter, while general and administrative expenses were $12 million. At that rate, the company has funds remaining for 3 more quarters. Expenses have gone up hugely from the November quarter.

Bottomline

SAVA is a highly volatile stock, with considerable retail enthusiasm, a lot of things happening, a lot of news, allegations, lawsuits, disputes, allegedly problematic data, and so on. Everything moves the stock this way or the other. Important catalysts ahead are readouts from the phase 3 trials, but this stock can really move on anything. I do not like to invest in such volatile stocks because, at heart, I am a conservative investor. But people with money to spare will find the SAVA ride a lot of fun.

About the TPT service

Thanks for reading. At the Total Pharma Tracker, we offer the following:-

Our Android app and website features a set of tools for DIY investors, including a work-in-progress software where you can enter any ticker and get extensive curated research material. 

For investors requiring hands-on support, our in-house experts go through our tools and find the best investible stocks, complete with buy/sell strategies and alerts.

Sign up now for our free trial, request access to our tools, and find out, at no cost to you, what we can do for you.