Investment Thesis
Depending on whether you are a bull or a bear, Cassava Sciences (NASDAQ:SAVA) and its lead candidate Simufilam either represents one of the most promising approaches to treating Alzheimer’s Disease in decades – with the data to show for it – or, a rogues gallery of a biotech that has been manipulating its data for nearly two decades in order to pump its stock, before dumping it on a gullible public.
The apparent release of Motion to Dismiss documents (which have been shared on Twitter) prepared by lawyers, in relation to a securities fraud action brought by Cassava shareholders (discussed in Cassava’s Q222 10Q submission), appears to be the most robust and lengthy response to accusations of data manipulation and fraudulent activity yet.
Below, I have summarized three key takeaways from the documents that shed a little more light on the issues at stake, and why I don’t think that it will derail the 2 Phase 3 studies of Simufilam currently underway, as requested by the Citizen’s Petition filed with the FDA that prompted the class action (which the FDA declined to do), or exonerate Cassava completely.
There’s not much middle ground here. To recap on the story so far.
Cassava – Background to Citizen’s Petition and Securities Litigation
Cassava is led by CEO Remi Barbier. Barbier tried and failed on four separate occasions to secure approval for a gel formulation of opioid Oxycodone as CEO of Pain Therapeutics, before changing the company’s name to Cassava Sciences in 2019 and opting to develop an Alzheimer’s drug with a novel mechanism of action (“MoA”) instead.
The drug – now known as Simufilam – targets a scaffolding protein known as Filamin-A (“FLNA”) which is often found to be misfolded in the brains of Alzheimer’s patients. Cassava believes Simufilam can restore the natural function of FLNA, thereby slowing or even reversing the deleterious effects of Alzheimer’s.
The foundational research into this drug was completed by Cassava’s Chief Scientific Officer (and CEO Barbier’s wife) Dr. Lindsay Burns, a Harvard educated neuroscientist and former Olympic rower, and Cassava’s academic advisor, Dr. Hoau-Yan Wang.
The research was first published in 2008, leading to Simufilam’s discovery in 2011, multiple studies in animals, which included findings of “dramatic improvements in brain health,” and eventually, the filing of an Investigational New Drug (“IND”) application with the FDA – which was approved in 2017.
In-Human Studies
The IND enabled Cassava to begin testing Simufilam in humans, and to complete successful Phase 1 and 2a studies with financial support from the National Institutes of Health (“NIH”). Progress was checked, however, when a Phase 2b study failed to meet its primary endpoint of reducing levels of tau protein in cerebrospinal samples and other biomarker assessments. I covered the news for Seeking Alpha back in June 2020.
That could have been the end of the road for Cassava but management decided to have the data from the 64-patient study – initially analyzed by the Lund University of Sweden – re-evaluated by an “outside lab.” That lab happened to be Dr Hoau-Yan Wang’s lab at City University New York – which was able to demonstrate a “significant reduction of indicators of Alzheimer’s in patients who took Simufilam versus placebo.”
Armed with the re-evaluated data, Cassava launched an open label Phase 2 study which began to read out some highly impressive data – after six months of treatment, patients’ cognition scores rose 1.6 points on the ADAS-Cog11 score – a gold standard measure of cognitive decline – at nine months, scores had risen three points higher, and at 12 months, the figure was +1.5 points, with management also finding that:
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).
If Simufilam produces data as compelling in its Phase 3 studies, the logical conclusion is that the drug’s approval – and widespread use – is a near certainty, given its clean safety profile.
The Citizen’s Petition and Lawsuit
During the period between December 2020 and August 2021, Cassava’s share price began to grow and grow, as investors bought into the open label data – arguably some of the most impressive data generated by any Alzheimer’s drug – although it should be noted the trial did not have a placebo arm to compare results with.
Cassava’s share price rose from ~$3, to ~$123 at its August ’21 peak – a gain of >4,000%, which made Cassava the best performing stock across any sector of the stock market. What happened next ultimately drove Cassava’s share price back below $20 by July this year.
A law firm – Labaton Sucharow, acting on behalf of anonymous clients – apparently later revealed by Cassava to be David Bredt and associate Geoffrey Pitt, who both held short positions in Cassava stock – then filed a Citizen’s Petition with the FDA, requesting the agency halt the 2 planned Phase 3 studies of Simufilam with a patient population of 1,750 – one lasting 52-weeks, the other 76 weeks.
The Citizen’s Petition contained a range of accusations directed at Cassava broadly divided into three categories.
- Firstly, the “validity of clinical biomarker data” – specifically in relation to the CSF samples analysed by the “outside lab.”
- Secondly, the “integrity of Western blot analyses” – the Citizen’s Petition believes multiple data presentations used to show the efficacy of Simufilam were faked.
- Finally the “integrity of analyses involving human brain tissue” – the Petition contends that since human brain specimens were used over a period of 10-plus years, Cassava’s findings relating to those specimens are “contrary to a basic understanding of neurobiology.
Fallout From The Petition
Although it was clear from the outset that the Citizen’s Petition had been filed by people with an apparent vested financial interest in the collapse of Cassava – i.e. they were short Cassava stock, as opposed to impartial observers – some of the “mud” thrown by it stuck – most notably in relation to the potentially faked Western Blotting, and the re-evaluation of the CSF samples by an “outside lab” that was in fact run by long-time Cassava collaborator Dr Wang.
As mentioned, Cassava stock was discounted from its all-time peak of $123 by 87% by late July this year, despite Cassava’s decisive rejection of the claims made by the petition.
Shortly after the Petition became common knowledge the Securities and Exchange Commission (“SEC”) and Department of Justice asked Cassava to “provide them with corporate information and documents.” Cassava’s disclosure of this fact in November 2021 was apparently enough to convince many observers that the company was subject to a reported criminal investigation.
Many of the journals that had published Cassava’s earlier work discussing Filamin A and Simufilam launched investigations to see whether it was necessary to retract articles, and social media became a popular way for scientists to critique Cassava and suggest the company was guilty of data manipulation.
As Cassava’s share price withered, one bright spot for Cassava bulls was the fact the FDA opted against taking any action in response to the Citizen’s Petition, meaning the Phase 3 trials were allowed to continue. As of Q222, some 400 patients have been enrolled in the studies.
The target enrollment is 1,750, and the primary endpoints will be based on ADAS-Cog12 (a cognitive scale) and ADCS-ADL (a functional scale).
Cassava’s Response To Lawsuit It’s Most Comprehensive Statement Yet
Perhaps inevitably, hot on the heels of the Citizen’s Petition came a lawsuit filed by disgruntled investors claiming that:
Cassava and its personnel defrauded investors for years by making public statements that concealed “rampant data manipulation and significant anomalies” in Cassava’s research.
Yesterday, Cassava’s motion to dismiss the lawsuit was shared via Twitter (and probably other social media channels besides) and it’s by far the most comprehensive and detailed response to its accusers that Cassava has made to date.
Let’s examine three of the most important aspects of Cassava and its law firm Orrick, Herrington & Sutcliffe’s response.
Key Point 1 – The Journal Findings and CUNY Investigation
As mentioned previously, of all the accusations leveled at Cassava it was the data manipulation in peer reviewed publications and the CUNY lab analysis that were the most damaging.
Cassava notes (in its motion to dismiss documents) that both the New York Times and Wall Street Journal ran articles discussing the Citizen’s petition and interviewing several of the most vocal critics of Cassava’s work, stating that:
These articles included no new facts, but simply rehashed the allegations from the Citizen Petitions with new commentary.
It must have been a significant frustration to Cassava – presuming the company believes it is 100% innocent – to have had to read these articles, as well as an article published in Reuters a few months later suggesting the DoJ had “opened a criminal investigation into Cassava.”
Management did issue a statement denying all claims made against it, but in the motion to dismiss documentation there is comprehensive detail around which publications have retracted what data. Cassava first notes that:
While Bik (a vocal social media critic) and the petitioners only reviewed the “as published” Western blot images, Drs. Burns and/or Wang provided the journals with the underlying images.
No journal has concluded that Dr. Burns or Dr. Wang manipulated data or engaged in any misconduct. One online journal (PLOS One) retracted articles because of concerns raised about the published data, however, none of the papers retracted by PLOS One concerned simufilam or Alzheimer’s disease. The majority of the journal inquiries have resulted in exonerations for Drs. Wang and/or Burns.
The document then looks at each publication in turn. Neuroscience and the Journal of Neuroscience found “no evidence” of data manipulation, and neither did the Journal of Prevention of Alzheimer’s Disease.
On the other hand, Molecular Neurodegeneration disclosed that the authors retracted this article because concerns have been raised regarding the data,” the Neurobiology of Aging “did not find compelling evidence” of data manipulation but noted that “errors in the published report were identified.”
PLOS One retracted five papers by Dr Wang, two of which were co-authored by Dr. Burns, and Alzheimer’s Research & Therapy retracted one article, refusing to reinstate after considering fresh evidence provided by the authors.
Meanwhile, an investigation into the CUNY lab run by Dr Wang is said to be:
ongoing and has not resulted in any finding of fault with respect to Dr Wang’s research.
Verdict:
The data published in various journals by Drs. Wang and Burns are important because they were used to support Cassava’s IND application, and it’s possible that, had the errors been found at the time, Cassava would not have been able to secure its IND and begin testing its drug in humans.
It does now seem to be clear that errors were found in some of the data submitted to these journals – some were not deemed consequential enough to retract the article, but in other cases the errors were consequential enough to result in retractions.
The FDA seemed to have put an end to the debate about whether Phase 3 Simufilam trials should go ahead when it rejected the Citizen’s Petition, and acting now would be attempting to close the stable door after the horse has bolted.
As such my feeling is that we will see data from those trials, and should that data prove to be positive compared to placebo, the data manipulation claims will likely be forgotten. Should the 2 pivotal Phase 3 studies fail, however, there will be renewed scrutiny.
Key Point 2 – The Improper Puzzle Pleading Defense
The main goal of Cassava’s motion to dismiss is obviously to ensure it does not have to go to court and defend itself, which could be a major distraction, not to mention a significant drain on finances, and disastrous should the company lose.
Cassava lawyer’s major argument in this regard is related to “Puzzle Pleading.” The company and its lawyer’s contention is that the lawsuit:
forces Defendants and/or the Court to sort out the alleged misstatements and match them with the corresponding, allegedly omitted “true” facts to solve the puzzle of interpreting Plaintiffs’ claims.
In other words, Cassava is arguing that the lawsuit simply throws mud at Cassava via multiple difficult to substantiate accusations, and expects the court to make the guilty case itself, when in fact, the burden to prove wrongdoing legally lies with the plaintiffs. Cassava’s lawyer writes:
The PSLRA (Private Securities Litigation Reform Act), however, requires Plaintiffs to specify with particularity “each statement alleged to have been misleading” and “the reason or reasons why (each) statement is misleading.” Plaintiffs may not, as they do in their Complaint, “recite a list of allegedly false and misleading statements extracted from press releases, analysts’ reports, and public filings and then follow the list” with “a separately located second list” of reasons “all the cited statements were false.”
In other words, the motion to dismiss is suggesting the plaintiffs themselves are not specific enough about the wrongdoing Cassava is supposed to have carried out, contenting themselves instead with a “laundry list” of things Cassava may have done wrong, and hoping the court does the hard work of establishing guilt for them.
Verdict
There’s a lot of focus on the Plaintiffs “Puzzle Pleading” approach in the motion to dismiss documents, while attention also is paid to the plaintiffs’ argument that Cassava is guilty of not confessing to its wrongdoing in a timely manner. But why should Cassava call itself out for wrongdoing when it doesn’t believe it had done anything wrong, as has not been formally accused of doing so by any authority?
The motion to dismiss documents conclude as follows:
In short, the Complaint should be dismissed because it is supported by allegations about other allegations by self-interested parties with no personal knowledge whatsoever of the underlying facts or data. This is the antithesis of the PSLRA’s requirement that the Complaint be supported by particularized allegations of fact.
Key Point 3 – Cassava Is Simply Going About Its Business
The motion to dismiss also makes clear that many of the plaintiffs complaints around Cassava providing false or misleading claims can be dismissed because these statements were, in fact, “undisputedly true and not in any way false or misleading.”
For example, when Cassava mentions that an “outside lab” conducted the re-evaluation of the Phase 2b CSF samples, the motion to dismiss contends that this was in fact true, because:
CUNY is a different institution than Cassava, and its labs are not Cassava labs.Thus, the statement that Plaintiffs allege to be false is entirely true.
Similarly, in response to accusations of fraudulent work carried out at the CUNY lab, Cassava had noted that some of the Phase 2b data had been generated by Quanterix, a life sciences firm. Quanterix responded immediately that it had not “not interpreted the test results or prepared the data charts”.
This much is true, and was seized upon by plaintiffs as evidence of Cassava’s duplicity, but it doesn’t actually make Cassava’s statement that it had worked with Quanterix false – the firm had been engaged to perform blinded sample testing – in fact, Quanterix’ clarification about what work it did was unnecessary.
The same defencs applies to Cassava’s statement that “government agencies have asked us to provide them with corporate information and documents.” Plaintiffs argue Cassava alsoshould have mentioned the prospect of a possible criminal investigation.
Once again however, Cassava notes a statement made at the time that “it cannot predict the outcome or impact of any (of) these ongoing matters, including whether a government agency may pursue an enforcement action against (Cassava) or others.” Once again, Cassava appears to have – just about – covered its bases.
Verdict
Throughout the motion to dismiss document the main objection raised to the plaintiffs case is that they fail to adequately explain which misdemeanors they expect to be compensated for, while Cassava has always been straightforward and honest in its communications with investors.
Cassava has generally kept its public comments to a minimum and attempted to steer clear of controversy, although it’s undeniable that there are grey areas that ideally require further investigation or clarification. The key question is whether they’re concerning enough for a full investigation to be launched. It’s probably fair to say that nobody – as yet – has been able to catch Cassava out in a bald-faced lie.
Most of the evidence against the company is based on a refusal to believe Cassava could have developed a successful Alzheimer’s drug with a completely unique mechanism of action – something no triple-digit-billion valuation pharmaceutical has been able to do in 30 years of trying.
“If it sounds too good to be true, then it probably is” is not, however, sufficient grounds for a criminal investigation, or a pivotal trial halt, or a shareholder lawsuit. The Citizen’s petition may have come very close to finding a “smoking gun” wielded by Cassava employees or associates, but very close is likely not close enough for a class action to succeed.
Conclusion – Cassava’s Robust Response To Lawsuit Keeps The Dream Alive
In this post I have not speculated about whether or not Simufilam can be an effective and safe Alzheimer’s drug – ultimately, the entire Cassava Bull Bear debate rests on the Phase 3 studies, in my view, rather than the outside noise, which has risen to a crescendo.
Based on the above – plus the statistic that, of 239 class action cases resolved in 2021, 153 were dismissed and 86 resolved through a settlement, I would say that on balance of probability Cassava’s motion to dismiss is likely to be upheld.
There’s mounting evidence to suggest that Cassava has sailed mightily close to the wind on its Simufilam journey, but equally, there’s no incontrovertible evidence of wrongdoing at the present time.
As the motion to dismiss notes, “all significant stockholdersinCassava did not sell a single share of Cassava stock throughout the class period.” On that basis the evidence seems to suggest that Cassava management are as desperate to see the Phase 3 results as everybody else.
In terms of Cassava’s share price and where it may head next, I think the only sensible conclusion to be drawn is that the company’s current valuation of $1.43bn is, almost certainly, wildly inaccurate.
If Simufilam succeeds in its Phase 3 studies it’s likely to be approved as an effective and safe form of therapy for Alzheimer’s patients, and will almost certainly go on to achieve blockbuster (>$1bn p.a.) sales, meaning the true valuation of the company is likely in excess of $10bn. On the other hand, if Simufilam fails, Cassava is essentially a worthless company.
The risk reward profile is likely too high for any neutral investor, while its unlikely that bullish arguments are going to dissuade bears from shorting the stock, or that bears can persuade bulls that there are too many inconsistencies in the Simufilam story to date. All roads still point to the Phase 3 data, which we won’t see for another 12 months or more.