Simufilam

Simufilam

Clinical data
Other names	PTI-125, PTI-910
Routes of administration	Oral administration (tablets)
ATC code	None
Legal status
Legal status	Experimental
Pharmacokinetic data
Elimination half-life	4.5 hrs
Identifiers
IUPAC name 1-benzyl-8-methyl-1,4,8-triazaspiro(4.5)decan-2-one
CAS Number	1224591-33-6
PubChem CID	46195331
UNII	6NV440YIO0
KEGG	D12003
ChEMBL	ChEMBL4650230
Chemical and physical data
Formula	C15H21N3O
Molar mass	259.353 g·mol−1
3D model (JSmol)	Interactive image
SMILES CN1CCC2(CC1)NCC(=O)N2Cc1ccccc1
InChI InChI=1S/C15H21N3O/c1-17-9-7-15(8-10-17)16-11-14(19)18(15)12-13-5-3-2-4-6-13/h2-6,16H,7-12H2,1H3 Key:BSQPTZYKCAULBH-UHFFFAOYSA-N

Simufilam is an Investigational New Drug as an oral tablet for the treatment of Alzheimer's disease in Phase III clinical trials expected to finish in 2023-24. Developed by an American pharmaceutical firm Cassava Sciences (previously Pain Therapeutics), based in Austin, Texas, simufilam was reported to restore impaired function of neurons and improve cognition in mice. Lindsay Burns (Senior Vice President of Cassava Sciences) and Hoau-Yan Wang of the City University of New York (science advisor to the company) are the main scientists who led the research on the discovery and development of the drug.

Promising results from a first-in-patient, non–placebo-controlled study in 13 Alzheimer's disease subjects were reported in 2020. Reductions of biomarkers from cerebrospinal fluid that are associated with Alzheimer's (for the disease itself, as well as neurodegeneration and inflammation) were an indication of the effectiveness. The research on the drug and its funding company were brought to the limelight when Cassava Sciences reported that the first 50 subjects in an open-label extension study to complete 9 months (and later 12 months) of treatment showed improved cognition. Less than one month later, the Food and Drug Administration received a citizen petition in August 2021 to stop the clinical trials primarily based on alleged anomalies in Western blot images from preclinical papers. Submitted by Jordan A. Thomas of the law firm Labaton Sucharow, the petition represented criticisms from a scientist David Bredt and cardiologist Geoffrey Pitt. Upon the publicity of the allegations, other scientists have also questioned these results, citing the small sample size, alleged methodological flaws in an in vitro technique, alleged manipulations of Western blot images and potential conflict of interest. Importantly, the lead petitioner also has multiple conflicts of interest, not limited to his shorting Cassava Sciences stock prior to filing the petition, which resulted in a quick and substantial financial gain.

In early 2022, the FDA rejected the petition. The Journal of Neuroscience and Neurobiology of Aging reexamined the papers they published on simufilam discovery and development, found no apparent data manipulation, but noted several errors, and issued expressions of concern. Neuroscience examined a paper alleged to be fraudulent and found "no evidence of data manipulation." The Quintessential Capital Management hedge fund, also a short-seller of Cassava Sciences stock, had called simufilam "a worthless compound." Others have noted the allegations initiated by the conflicted petitioners and continued by hedge funds were for financial gain.

History

From research funded by Cassava Sciences (then Pain Therapeutics), Lindsay Burns (Senior Vice President at Cassava Sciences) with Hoau-Yan Wang and Maya Frankfurt of the City University of New York, reported in PLOS One the binding of a 300-kDa protein called filamin A (FLNA) with naloxone to prevent opioid tolerance and drug dependence. This report, the authors claimed, was a critical discovery of the binding of certain opioid antagonists (naloxone and naltrexone) to FLNA, a cytoskeletal protein that is critical in maintaining cell shape and division. Burns and Wang reported the pentapeptide binding site on FLNA the next year. In 2010, they identified a new compound from their pharmacophore modeling, in vitro screening and medicinal chemistry efforts, describing it as "a novel analgesic" and named it PTI-609 (PTI for Pain Therapeutics, Inc.). The molecule could also bind to FLNA as did naloxone and naltrexone and, unlike naloxone and naltrexone, also activated the mu opioid receptor. While working on this analgesia program, Dr. Wang separately identified a large protein associating with the alpha 7 nicotinic receptor when Abeta42 bound and signaled through this receptor in Alzheimer's disease models. He identified it as FLNA, and Dr. Wang and Dr. Burns decided to test the FLNA-binding compounds from their medicinal chemistry library to see if any compounds could disrupt this FLNA interaction with the alpha 7 nicotinic receptor and therefore test the hypothesis that it was critical to the toxic signaling of soluble amyloid. In 2012, they identified a new novel compound, named PTI-125, in The Journal of Neuroscience that did disrupt this FLNA linkage with the alpha 7 nicotinic receptor as well as the toxic signaling of Abeta42. Their report demonstrated that FLNA assists in the toxic signaling of amyloid beta via the alpha 7 nicotinic acetylcholine receptor in Alzheimer's disease and presented PTI-125 as a novel therapeutic strategy for Alzheimer's disease. Their report in Neurobiology of Aging in 2017 showed that PTI-125 induced improvements in Alzheimer's disease pathology as it binds, and restores to normal, an altered conformation of FLNA in experimental Alzheimer's disease transgenic mice.

In 2018, the National Institutes of Health granted the company a research award for early clinical trials of PTI-125 as an Alzheimer's drug. Once fully focused on Alzheimer's disease, the company was renamed Cassava Sciences, Inc. the next year. In August 2020, the United States Adopted Names (USAN) assigned the drug chemical name as simufilam. In 2020, it was announced that the initial biomarker analysis of the Phase II trial had failed due to high variability in CSF biomarker changes seen even in placebo subjects over this 28-day trial. It was later determined that this initial biomarker analysis showed no correlation between biomarkers in change from baseline (mean pearson's r=0.06), which is unexpected because biomarkers generally move together. This lack of correlation was especially evident in individual placebo subjects' data that showed huge improvements in one biomarker coincident with worsening in another. The analysis of backup CSF samples showed good results, replicating results from the first-in-patient study.

The final stage, Phase 3 trials started in February 2021. The first phase would require 750 participants, and the second, 1000. Despite mounting enrollment once a majority of clinical sites were activated, naysayers have alleged that people are deterred from enlisting in the trial. As of May 2022, over 120 people were enrolled in the study, a doubling from one month earlier.

Pharmacology and controversies

The original discovery paper claimed that FLNA is the binding site of certain opioid antagonists in reducing and preventing opioid tolerance and dependence. Later, simufilam was selected from a medicinal chemistry library of compounds screened against this binding site on FLNA. They demonstrated that simufilam binds to FLNA and prevents the tight binding of amyloid beta 42 to the alpha-7 nicotinic receptor. This interaction has been shown to hyperphosphorylate tau, disrupting tau's normal function and leading to degeneration of neurons, a cause of Alzheimer's at the molecular level. Preventing amyloid signaling via the alpha 7 nicotinic receptor has been a key focus in Alzheimer drug development. Burns and Wang showed that the FLNA linkage to the alpha 7 nicotinic receptor is critical to amyloid's toxic signaling through this receptor and that simufilam disrupts FLNA's linkage to this receptor to stop this toxic signaling, potentially providing a novel treatment for Alzheimer's disease.

No other researchers have examined simufilam binding to FLNA or simulifam's restoration of altered filamin A. However, an independent researcher at Yale University has shown that simufilam improved epilepsy in a mouse model. Research papers demonstrating the mechanism of action of simufilam contained an error of units in methods (one instance of milligrams noted as micrograms) and erroneous duplication of images, but neither journal found evidence of data manipulation that was previously alleged. Two papers that reported the discovery of FLNA binding by certain opioid antagonists and FLNA's role in opioid tolerance and dependence were retracted for "similarities in background pixels" in Western blot images without evidence of data manipulation.

The link between FLNA and Alzheimer's disease is not well established, but one other laboratory has noted a role of FLNA in AD and its overexpression has been noted as a result of PS1 mutation in familial AD. It is also overexpressed in some forms of Progressive Supranuclear Palsy (PSP). Lawrence Sterling Honig, professor of neurology at Columbia University Irving Medical Center, had remarked on Burns and Wang's claims: "But in fact, all the evidence seems to be from this [Wang's] lab." The placebo effect, a beneficial non-drug treatment, is an important parameter in clinical trials involving brain diseases as it can have beneficial effects as the drugs, but no study exists showing a placebo effect of cognitive improvement from baseline in mild-to-moderate Alzheimer's disease subjects at 9 or 12 months of treatment. Robert Howard, professor of psychiatry at the University College London, is concerned on the lack of placebo and small sample size and said that the research "at the very least is implausible." Thomas C. Südhof, Nobel laureate neuroscientist at Stanford University, also commented: "The overall conclusions with regard to Alzheimer’s disease make no sense to me whatsoever... [The findings of Burns and Wang] are not in the mainstream of the field, and to me they seem implausible and contrived."

Allegations of research concerns

In August 2021, Jordan A. Thomas of the law firm Labaton Sucharow in New York City, submitted a citizen petition to the FDA to investigate simufilam research. It was only after the stock plummeted that Thomas disclosed that the "anonymous whistleblowers" were in fact short sellers of Cassava Sciences stock. Thomas' petition alleged "grave concerns about the quality and integrity of the laboratory-based studies surrounding this drug candidate and supporting the claims for its efficacy," and made four key points:

1. Claims that filamin A is associated with Alzheimer's disease and that simufilam binds to filamin A are not supported by any independent research;
2. Images of western blots in Burns and Wang's papers were alleged to contain data anomalies and manipulations (later refuted by journal editors);
3. The biomarker data in clinical trials were questioned; and
4. "Remarkable" molecular experiments were said to be suspicious.

Thomas expressed that "The volume of problematic material uncovered in publicly available sources indicates a thorough audit would likely unveil significant additional scientific misconduct and data manipulation," and that "the methodology allegedly used in these experiments defies logic." Cassava Sciences publicly defended against each claim asserting that "the claims made in this post [petition] regarding scientific integrity are false and misleading. The Company stands behind its science, its scientists and its scientific collaborators, and is responding to ensure the facts are known and respected." Barbier maintained that the research data were genuine, and said, "As a science company, we champion facts that can be evaluated and verified."

Quanterix Corporation did confirm simufilam treatment effects in a biomarker, plasma p-tau181. Cassava Sciences stated: "Cassava Sciences' plasma p-tau data from Alzheimer's patients was generated by Quanterix Corp., an independent company, and presented at the recent Alzheimer’s Association International Conference." Quanterix immediately made a public disclaimer that although Quanterix analyzed the samples blind to the treatment group, it was not involved in analyzing the data, stating: "Quanterix or its employees did not interpret the test results or prepare the data charts presented by Cassava."

In November 2021, David Bredt, a former neuroscientist at Johnson & Johnson, who mysteriously left venture capital firm MPM after a 4-month stint the month the petition was filed, was revealed as the lead whistleblower in The Wall Street Journal. Early in 2021, Bredt became aware of the drug as the shares of Cassava Sciences reached a notable height. However, Bredt had first learned of the program from multiple confidential presentations to J&J by Cassava Sciences (then Pain Therapeutics). Examining the clinical trials, he was concerned with the lack of placebos. Further scrutiny of all the research papers made him conclude that, "They were making statements that were incompatible with biology and with pharmacology," and such important discoveries deserved to "win five Nobel Prizes." Teaming up with a childhood friend Geoffrey Pitt, a cardiologist and a professor at Weill Cornell Medical College, he directed Thomas for the petition, which was filed without disclosing their short positions in Cassava Sciences stock, which some considered "Gaming the FDA."

The petition was denied in February 2022 by Patrizia Cavazzoni, Director of the Center for Drug Evaluation and Research for "lack of evidence." The letter concluded:

We take the issues you raise seriously. Please note that your petitions are being denied solely on the grounds that your requests are not the appropriate subject of a citizen petition. This response does not represent a decision by the Agency to take or refrain from taking any action relating to the subject matter of your petitions.

Reactions from the journals

The 2012 paper in The Journal of Neuroscience was the one that reported that simufilam could bind to filamin A, the key claim for the potential use of the drug in Alzheimer's disease. The study relies on Western blotting, immunohistochemistry and binding affinity of FITC-labeled Abeta 42 for the alpha7 nicotinic receptor. After reassessment, Marina R. Picciotto, the editor-in-chief, reported in November 2021 that although data manipulation was not detected, there was one duplication of an image. The journal published an erratum correcting the duplication and also publishing original Western blot images that had been accused of manipulation in December 2021. The journal declared that the "error does not affect the conclusions of the article." However, the journal received further complaints about the original Western blot images published as clarification of no data manipulation, and it issued an expression of concernon 19 January 2022:

The editors have been made aware of concerns about Western blots in this study, including those published with the article's erratum (Wang et al., 2021). These and other concerns are currently under investigation by the academic authorities at the City University of New York (CUNY). JNeurosci will await the outcome of that investigation before taking further action.

Neuroscience also found no evidence of data manipulation for a 2005 paper that had been questioned.

PLOS One re-examined all the Burns and Wang's research papers and came to conclusion in March 2020 that there were "unresolved issues" relating to "similarities in background pixels" in the uncropped western blot images. It retracted five of Wang's papers, two of which were co-authored with Burns, and discuss the role of filamin A in opioid receptor signaling. The other three papers have no Cassava Sciences authors and are about prenatal cocaine. None of these papers involve simufilam or Alzheimer's disease, and no clear evidence of data manipulation was found. The retraction note on the paper showing naloxone binding to filamin A states:

The data and comments provided did not resolve the concerns about the integrity and reliability of data presented in this article. In light of these issues, the PLOS ONE Editors retract this article. HYW [Hoau-Yan Wang] and LB [Lindsay Burns] did not agree with the retraction. MF [Maya Frankfurt] either did not respond directly or could not be reached. HYW stands by the article’s findings.

Neurobiology of Aging, which published the 2017 paper reporting the ability of simufilam to restore altered filamin A in mice, also reassessed the paper, making a notification that there was "no compelling evidence of data manipulation intended to misrepresent the results," but there was an error in units (e.g. microgram should have been milligram in one instance) in the methods. It issued an expression of concern:

The authors have requested a corrigendum to correct these issues. However, Neurobiology of Aging is aware of an ongoing inquiry of these and other concerns by the sponsoring institution, the City University of New York (CUNY), and will make a final decision as to appropriate corrective action once that inquiry has been concluded.