Talk:Cassava Sciences/Archive 1

Archive 1

Clinical Trial Data

There should be a section added that covers all of their clinical trials and the results including interim results. I could add that as a section if that's okay. It really is the most important part of the cassava story. 72.184.147.25 (talk) 01:04, 17 August 2022 (UTC)

Why? If it's not covered by secondary sources, it's probably WP:UNDUE here. Wikipedia is WP:NOTNEWS and WP:NOT a webhost (for Cassava press releases and primary sources); it follows secondary sources, does not lead. SandyGeorgia (Talk) 01:14, 17 August 2022 (UTC)

PS, you may also be mixing up what might be appropriate for a drug article vs what is appropriate for the company article. At any rate, I'm not even sure that primary source results would be welcome at a drug article. SandyGeorgia (Talk) 03:14, 17 August 2022 (UTC)

Here is a news article that mentions the recent clinical data: https://www.cassavasciences.com/news-releases/news-release-details/cassava-sciences-reports-second-quarter-financial-results-2022 SighSci (talk) 18:21, 17 August 2022 (UTC)

That is not a news article; that is a Cassava press release (not an independent third-party source and certainly not neutral). A friend has had a medical emergency and I won't be able to complete these edits today; if you could come to a better understanding of Wikipedia sourcing, it would save time. We need to look at removing a lot of the press releases already used. They may be used on a limited basis to state Cassava's position, but not to source factual information. Please read WP:V, WP:OR, WP:RS and WP:PSTS. It would save a lot of time, and corrections could be done much quicker, if your requests were short and based on Wikipedia policy. SandyGeorgia (Talk) 19:10, 17 August 2022 (UTC)

Sorry, sent the wrong link (I am not that dense). Here is the news article. https://www.marketwatch.com/story/cassava-sciences-shares-extend-rally-after-board-member-sanford-robertson-increases-stake-271660752662?mod=newsviewer_click

Also, here is another news article from today stating that a different journal (JPAD) also found no evidence of data manipulation in a simufilam paper. Now all three published papers involving the drug have been cleared of allegations of manipulation by the respective journal editors. https://www.marketwatch.com/story/cassava-sciences-academic-journal-found-no-evidence-of-data-manipulation-271660829914

Please update the "journals investigate" section! SighSci (talk) 15:32, 18 August 2022 (UTC)

Best wishes to your friend; hope all is ok. SighSci (talk) 15:34, 18 August 2022 (UTC)

Thanks. But you are still providing dressed up or outright press releases that reflect what Cassava states or claims, rather than independent news reporting. I should be able to start digging in to all of this tomorrow. When marketwatch basically repeats Cassava press releases, that's not independent third-party reliable sourcing. SandyGeorgia (Talk) 04:24, 19 August 2022 (UTC)

PS, I have a google alert set up for Cassava Sciences, so if there is any real reporting from reliable sources, I'm likely to see it; the batch I got tonight from google is all nothing but newswires recycling Cassava Sciences press releases. We don't do it that way on Wikipedia. SandyGeorgia (Talk) 05:06, 19 August 2022 (UTC)

Agree we need to avoid churnalism. Alexbrn (talk) 05:19, 19 August 2022 (UTC)

There is nothing actionable in this section, which like so many others is filling the talk page with repetitive information and impeding progress. It would be helpful to archive close the unuseful sections here so we can focus on the policy-based content improvements. See also #Clinical data updates with press references and #Enrollment in Phase 3 studies. This is the company article; once we get it sorted, perhaps the unregistered and new editors will have a better sense of Wikipedia policies and guidelines and what sort of info might be useful at the drug article, simufilam, and that does not bear repetition here. SandyGeorgia (Talk) 19:03, 19 August 2022 (UTC)

To be recovered

There is a lot of well-sourced content to be recovered from this version of the article, before COI editing. There seem to have been two different, but equally damaging, kinds of COI/POV editing occurring here—the common variety of puffery to inflate along with less common variety attempts to deflate—neither surprising considering the situation, but both requiring attention. Some good text needs to be recovered and re-instated if someone has time. SandyGeorgia (Talk) 03:52, 17 August 2022 (UTC)

Reminder to also revisit this version to see if it has anything usable. SandyGeorgia (Talk) 22:01, 19 August 2022 (UTC)

Quanterix

To be sorted, undefined in the article. SandyGeorgia (Talk) 19:33, 20 August 2022 (UTC)

A tangent, removed; if there is a good reason to reinstate, please provide sources. SandyGeorgia (Talk) 20:08, 20 August 2022 (UTC)

Clinical data updates with press references

Pharmaceutical companies release a lot of data through press release, company filings or presentations. If you look at Biogen they reference these in their updates. https://wiki.riteme.site/wiki/Biogen

That said, with a reasonably quick search of their history, there are articles that cover each of their data updates on their trials. Phase 2b https://www.neurologylive.com/view/sumifilam-improves-multiple-alzheimer-disease-biomarkers-in-phase-2b-study

“The results point to the potential for Sumifilam to become a transformative treatment for patients with Alzheimer disease.”

Jeffrey Cummings, MD, ScD, founding director, Cleveland Clinic Lou Ruvo Center for Brain Health, and Chambers Professor of Brain Science, University of Nevada–Las Vegas, said in a statement.1 “I am pleased to see early evidence of disease-modifying effects in patients with this investigational drug. The data appear to represent a step forward toward urgently needed treatments for Alzheimer disease.”

Open label 6 month update https://www.neurologylive.com/view/simufilam-improves-cognition-behavior-alzheimer-disease

“The Cassava Sciences agent showed 6-month improvements in both ADAS-Cog and NPI scores, and a phase 3 trial is expected to begin in late 2021.”

“The study, funded by the National Institutes of Health (NIH), showed that after 6 months of treatment with the altered filamin A (FLNA) restoration agent that cognition scores measured by the Alzheimer's Disease Assessment Scale-Cognitive Subscale (ADAS-Cog) improved by 1.6 points from baseline scores of 15.5, equivalent to a 10% mean improvement. Additionally, dementia-related behavior—including anxiety, delusions, and agitation—improved by 1.3 points from baseline scores of 4.5 on the Neuropsychiatric Inventory (NPI). This improvement equaled a 29% mean improvement at 6 months.”

https://www.marketwatch.com/story/cassava-sciences-shares-extend-rally-after-board-member-sanford-robertson-increases-stake-271660752662

“The timing of the purchase comes shortly after the company provided an update for simufilam, Cassava's lead drug candidate to treat Alzheimer's disease, a brain disorder that destroys memory and other important mental functions. The Austin, Texas-based company said on Aug. 3 that an interim analysis on 100 patients who completed at least 12 months with simufilam showed that overall ADAS-Cog12, a cognitive scale, scores improved by an average of 1.5 points and that the drug appears safe and well tolerated.”

Mnachtrab (talk) 10:23, 19 August 2022 (UTC)

First a general response; see WP:OTHERSTUFFEXISTS. What any other (C-class) article has is irrelevant to this article. (Perhaps if you referenced a Featured article, we'd have a starting place for discussion.)

1. There is a proper way to use primary sources and press releases on Wikipedia; in this suite of articles we have not seen that. Please see WP:PSTS.
2. "The Austin, Texas-based company said" see the discussion above at #Clinical Trial Data.
3. This article is about the company; to whatever extent any trial data is relevant, that is more likely to be at the simufilam article.

I will look at the neurologylive sources to see if there is anything useful to this article, but the more I have to process through repetitive off-topic posts on this talk page, the less time I am going to have to actually work on improving the article. SandyGeorgia (Talk) 14:49, 19 August 2022 (UTC)

I have now looked at the neurologylive sources and do not find anything useful for this (the company) page; there is a page for simufilam where detail about past trials and their numbers and dates if appropriate can be explored. At any rate, if biomedical claims are made anywhere on Wikipedia, they should be based on WP:MEDRS sources. There is nothing actionable in this section. SandyGeorgia (Talk) 19:21, 19 August 2022 (UTC)

Next, if the editors appearing at this page would like to see the proper use of sources in a similar topic (related to Cassava through the Charles Piller report in Science, as the Cassava investigation led to the University of Minnesota investigation as a side-product), please review Sylvain Lesné, which is a policy-compliant article. You will see there, if you take the time, that there is limited use of primary sources; that is, only when they are backed by secondary references or when they are used for uncontroversial or non-self-serving information about the topic of the article. Please understand what Wikipedia is and is not; it is not a webhost, not promotional, and not for publishing original thought. SandyGeorgia (Talk) 15:28, 19 August 2022 (UTC)

Several people/entities complained to Science about the Piller article. Note the corrections starting to mount at the bottom of the page. The journal JCI did not declined to comment but was apparently not contacted. Watch for a letter to the editor about this irresponsible journalism soon. When discussing Cassava, Piller sought only negative sources (the same negative sources involved in the FDA petition whom the WSJ and NYT reporters also exclusively spoke to). No one random or neutral. The article was about Lesne, yet the first paragraph would have you believe it is all about Cassava. As erroneously described by Piller, the "most influential Cassava-related paper" had nothing to do with Cassava; as evidence, the authors declared no COIs. He states that it was cited over 1500 times, and then goes on to discredit the whole paper by grossly misrepresenting and sensationalizing the research method involving carefully procured, well-matched pairs of postmortem Alzheimer's disease and healthy control brain tissue, collected an average of 5-6 hours after death. Calls it "dead tissue" (the donor was dead, yes), claims it was "partially thawed and chopped" leaving the reader to envision chopped up brain mush severed of axonal connections (it was warmed from -80 to -20 so a smaller piece could be sliced off the larger, original piece for the assay conducted at body temperature), and states it was "purported to transmit nerve impulses" (later changed to "chemical signals" (also misleading) when the assay only measured protein-protein interactions that indicate insulin receptor function -- importantly comparing disease to control. Then he claims that this brain diabetes (in Alzheimer's brain) paper that he just trashed not only misled a bunch of researchers into following something that can't be true but also supported the science behind simufilam -- just because simufilam improves brain insulin receptor function along with a host of other things. Please do not rely on this "reliable source" at least when discussing Cassava. Finally, there are citations he could have cited for other researchers using the accused method, but he failed to find them. SighSci (talk) 01:32, 21 August 2022 (UTC)

Wikipedia can't right great wrongs, nor does it aim to. When reliable secondary sources have something different to say, that can be reflected on Wikipedia. Also, the point of my example was not Piller (Science is a reliable source), but to illustrate how primary vs. secondary sources are used. SandyGeorgia (Talk) 04:20, 21 August 2022 (UTC)

SavaDx

... is mentioned in the lead but nowhere in the article, and there appear to be no high quality sources that have paid any attention to it. Does anyone have an independent third-party reliable source that discusses it? If not, it doesn't belong in the lead (which by the way does not even bother to tell us its a blood test). SandyGeorgia (Talk) 17:04, 19 August 2022 (UTC)

You could omit in the lead as there is nothing published on SavaDx other than meeting abstracts, or simply mention it as a product in early development. Diagnostics are more difficult to patent because you can't patent something that exists in the body, so companies tend to be quiet until they are very advanced. But if you take that out of the lead, you should consider removing the old abandoned products. Every company has a long list of these. One exception for the lead may be Remoxy, an abuse-deterrent oxycodone formulation that would have replaced OxyContin -- and was filed with FDA for approval and rejected 4 times (possibly a record). Remoxy was essentially impossible to abuse for a heroin-like high (as is the case with OxyContin that can be crushed, dissolved etc to get 12-hours of opioid out in 2 minutes). This is a totally separate story and many suspect foul play, but it is definitely incorrect to say (in the products section) "after an FDA advisory meeting raised concerns about its potential risk of abuse." The whole point was -- could it be called abuse-deterrent or are we just going to let OxyContin keep killing everyone because we can't make up our mind about whether Remoxy is abuse-deterrent or not? The ad comm was successfully confused by FDA-generated data shown as a surprise only during the meeting, using undisclosed methods and very few data points (<10) that contrasted with the robust data (nearly 8K data points) produced by a third-party GLP lab contracted by the company, showing clear abuse-deterrence. 2600:1700:BB80:88A0:C999:C3CC:65DA:1E3D (talk) 22:46, 20 August 2022 (UTC)

I can work on this tomorrow (too tired tonight), but my concern is that by deleting history, we will have a lead very weighted towards the current controversies, rather than touching on all aspects of the content, as a WP:LEAD should. SandyGeorgia (Talk) 03:49, 21 August 2022 (UTC)

When discussing Remoxy where appropriate, instead of just describing it as a twice-daily oxycodone, it is important to note that it was a long-acting formulation designed to be abuse-deterrent. Or designed to deter abuse. Otherwise it's a bit confusing. The 2006 successful clinical trial showed it was effective as an analgesic (https://pubmed.ncbi.nlm.nih.gov/30629280/), but the long delay by FDA was all about proving that when messed with in a lab in a zillion different ways, it could not be defeated to dump a large dose of oxycodone for abusers -- they would not approve it unless it could deter abuse, but they had no established guidelines for this. They kept changing conditions hoping that it would fail, and demanded multiple solvents that were non-ingestible, and a human abuse-deterrent study to show it couldn't be snorted for a high (after first asking for and receiving data showing it wasn't absorbed nasally at all in dogs) -- this study was actually published and could be cited: https://pubmed.ncbi.nlm.nih.gov/30629280/ After the successful human clinical trial for abuse-deterrence by the nasal route, the FDA said it was actually a liability because it would then force abusers to inject it (even though it was not dissolvable in injection vehicles). A separate, earlier human clinical trial in abusers showed it was abuse-deterrent by the oral route: https://academic.oup.com/painmedicine/article/12/4/618/1869381 No need for this detailed history, but citing these publications is appropriate. This very frustrating history of this company is now being used by the negative voices to fit their story of negativity. SighSci (talk) 16:19, 21 August 2022 (UTC)

Cleanup

I have cleaned most of the advert-y and non-policy-based content. I've tried to fill in many of the citations, but have NOT verified that all sources are accurately represented (many were not); please do not assume that any citation I have cleaned up the formatting on is accurately represented in this (woeful) article.

The article over-relies on press releases and primary documents, when secondary sources covering most of the issues are available. SandyGeorgia (Talk) 21:46, 31 July 2022 (UTC)

Thank you for your efforts! ScienceFlyer (talk) 23:36, 31 July 2022 (UTC)

Here's a good primary article for balance of the whole situation: https://www.complianceweek.com/risk-management/the-cassava-sciences-saga-short-sellers-gaming-the-fda-and-the-damaging-ripple-effects/31416.article SighSci (talk) 01:57, 17 August 2022 (UTC)

Continued below at #Compliance week article research should be added. There is nothing actionable in this section, which should be archive closed so we can focus on what does need to be done without the repetition. SandyGeorgia (Talk) 19:05, 19 August 2022 (UTC)

This Compliance Week article should be cited after the statement in the FDA petition section that Cassava claims the FDA petition process was used abusively. Please cite it. This article is not cited anywhere and it is important to keep this page balanced. SighSci (talk) 13:16, 24 August 2022 (UTC)

Enrollment in Phase 3 studies

The last sentence of the Phase II section expressing the opinion that "enrollment has slowed" has been proven wrong by the recent disclosure that over 400 subjects have enrolled in Phase 3. Please remove the sentence and reference and update to cite current enrollment. Additionally, complaints about the small sample size of the first interim analysis of the 12-month open-label study should be tempered by the recent disclosure of similar results in double the sample size (63 of 100 subjects IMPROVED in cognition scores after one year; this is a population known to decline). 104.10.252.226 (talk) 17:04, 5 August 2022 (UTC)

Additionally, in this Phase II section, the focus on Dr. Wang being an advisor to the company (and therefore conflicted) is misleading. It also This passage needs to note, as is disclosed in the same sources, that Dr. Wang was and still is blind to treatment group. Not knowing what samples correspond to simufilam versus placebo makes it impossible to manipulate data, as is the implication with this complaint about conflict of interest. 104.10.252.226 (talk) 17:28, 5 August 2022 (UTC)

Please provide reliable third-party independent sources for your assertions. SandyGeorgia (Talk) 18:20, 5 August 2022 (UTC)

Here is one source (besides company press release) for the above information: https://seekingalpha.com/article/4529663-cassava-sciences-some-learnings-from-q2-2022-earnings

Clinicaltrials.gov for NCT04079803 shows that the labs measuring biomarkers (both CSF and the plasma p-tau181) were blind to treatment, as indicated by the following: "Masking:

Quadruple (Participant, Care Provider, Investigator, Outcomes Assessor)"

Outcomes assessor means both cognitive testing and biomarker testing personnel. Here is the link to the trial registration: https://clinicaltrials.gov/ct2/show/NCT04079803?term=simufilam&draw=2&rank=6 104.10.252.226 (talk) 21:05, 5 August 2022 (UTC)

You've provided one independent source, seekingalpha.com. Since I have never heard of it, and can't even get past all of its ridiculous popups, please visit WP:RS and WP:RSN to establish why you believe it meets Wikipedia guidelines for reliability. (Hint, I have a script installed that redflags non-reliable sources, so I can tell you not to waste your time on it. It's flaming.)

Please also visit the policy page, WP:DUE. And then WP:OR; Wikipedia follows secondary sources--it doesn't analyze primary studies and sources. If real sources don't mention something, it's rarely due weight to include it.

When you want to add something to a Wikipedia article, it is expedient to provide a WP:RS, or for biomedical content, a WP:MEDRS. Please also understand the difference between primary, secondary and tertiary sources, and how to use them; a solid read of the policy and guideline pages I've listed above will provide a guide to everything that was wrong in this article because of what looks like WP:COI and WP:POV editing. And what Wikipedia is WP:NOT.

You haven't provided any useful information or sources upon which text can be based. SandyGeorgia (Talk) 01:40, 6 August 2022 (UTC)

Thanks. We will all just have to wait until CUNY finally states their conclusion so that reliable sources can write a reliable article (in contrast to the hit pieces in WSJ and NYT that only sought sources that fit a preconceived negative narrative and who were directly involved in the failed citizen's petition to halt a Phase 3 program of a promising drug candidate for Alzheimer's disease). All reputable journalists are choosing to wait until CUNY speaks. At that time, this story will substantially change. And, as the company disclosed last fall, these government agency investigations were initiated at the same time as the petition -- likely by the very same people. 104.10.252.226 (talk) 14:58, 6 August 2022 (UTC)

Good reason to go ahead and remove all of the primary and non-reliable sourcing from this article; the editing that went on here was dreadful, and much of that poor editing is still in the article. I do hope you have read all of the policy and guideline pages I've linked above, because, if that editing continues, I'll be requesting administrative attention to this article. SandyGeorgia (Talk) 15:02, 6 August 2022 (UTC)

As of this August 6 version, Ì believe the article is now down to reliable sources or Cassava press releases only when they can be justified as having been mentioned by secondary sources. I could have missed some. If further adjustments are needed, please refer to reliable published independent sources. Also keep in mind that this is the company article, not the drug article. The article was previously going in to excess detail about specific drugs. SandyGeorgia (Talk) 16:08, 6 August 2022 (UTC)

Neuroscience article

If you only include company press releases that have been mentioned by secondary sources, you are failing to report outcomes of all journal investigations and specifically omitting a journal who investigated an allegation and found no evidence of manipulation (Neuroscience). Unlike the other two journals that found no evidence of data manipulation, Neuroscience decided not to maintain an expression of concern on the article while CUNY completes their investigation. This Neuroscience article itself could be cited, even though there is no expression of concern: https://pubmed.ncbi.nlm.nih.gov/16084657/ This section should be balanced, and currently it is not (nor is the entire article). Furthermore, the planned enrollment in the Phase 3 trials is stuffed into "other concerns." Clearly it does not belong there but could be mentioned after the statement that the drug is in Phase 3 trials. The following sentence claiming that patient enrollment has slowed (by Stat) is from April, was wrong then and is very outdated now. It should be removed. SighSci (talk) 19:12, 16 August 2022 (UTC)

Well, thank you for registering an account; now we can begin to address your edits. First, please read and acknowledge the information at WP:COI as necessary. SandyGeorgia (Talk) 19:41, 16 August 2022 (UTC)

Read. I will only make suggestions. Per this policy, short-sellers of Cassava stock, or anyone paid by them or connected to efforts to impede Cassava, should not have been editing. About the Neuroscience article, it turns out there is actually an editorial note stating that no evidence of manipulation was found that can be cited: https://www.sciencedirect.com/science/article/pii/S0306452221005789?via%3Dihub. Again, this should be included if this section is to be complete or balanced. Note that there are only three published papers that involve Cassava Sciences' drug. It may be more relevant to focus on those, rather than the portion of Cassava's academic collaborator's long list of publications that are unrelated to Alzheimer's or Cassava (but also falsely accused and used to smear Cassava). SighSci (talk) 23:48, 16 August 2022 (UTC)

Excellent progress.

Which text about a "long list of publications that are unrelated to Alzheimer's" are you referring to? I will start a separate post to sort the Neuroscience text, so the formatting will be more easily followed. Please recognize that "regular" (i.e. ongoing, or long-standing) Wikipedia editors are not solely immersed in the vagaries of one particular article or topic, rather addressing multiple issues across many types of articles daily, so the more precise you can be, the sooner issues can be addressed. SandyGeorgia (Talk) 01:28, 17 August 2022 (UTC)

Thanks. The long list is that Dr. Wang has published well over 100 papers and "issues" have been raised with over 30 on PubPeer, by the same small group of people. Editors may or may not have investigated by requesting original high-res images, but they are expected to if they see a need. Any honest errors have been found, for sure. SighSci (talk) 01:53, 17 August 2022 (UTC)

SighSci, in trying to format a response to your concerns, I am realizing that I am confused about what you are asking/stating relative to what was in the article when I started to clean up a biased, COI, and unreliably sourced version. I don't know what you want changed at this point.

You are referencing (above) a February 2022 correction about a naloxone paper. The text now in the article references a December 2021 correction about an entirely different paper. And the Michaels/Wall Street Journal piece, used as a source, refers to a November 2021 correction.

What is in the article now never alleges data manipulation; it says an error was found that "does not affect the conclusions of the article". Unless you can be more specific, it appears that you want something corrected that is never incorrect, and I am becoming concerned that we are discussing different Neuroscience papers.

It looks like we had best take this from the top and get on the same page. For expediency here, please reference exact wording now in the article, so I can be sure what you are referring to and get this addressed as soon as feasible. For example, I see no text in the article about Wang and well over 100 papers, or issues with 30 being raised on PubPeer, so I don't know what you want changed vis-a-vis that comment. Try to format your talk page posts as ... here is what the article says now, based on X source, and here is my proposed change, based on Y source. SandyGeorgia (Talk) 02:18, 17 August 2022 (UTC)

OK. The "journals investigate" section mentions three journals - JNS, PLOS and NBA. One of them (PLOS) mentions 5 articles, none of which discuss the drug or even Alzheimer's. If PLOS is to be included, a fourth journal (Neuroscience -- different from Journal of Neuroscience [JNS]) should also be mentioned. That is the editorial note I recently found and linked to above.

First, this section should start, "In response to allegations, " instead of "Following the public controversies." They did not independently reassess but were responding to specific allegations posted on Pubpeer and apparently reached out to the authors. The text now states: The journal published a correction along with the original images in December 2021 remarking that the "error does not affect the conclusions of the article". This sentence omits the main finding: after examining original images that were requested of the author, JNS found "no evidence of data manipulation." An honest error was found and this is the correction. The sentence should state: "After examining original images, the journal found no evidence of data manipulation [might need to cite company press release here because the erratum does not state this] but identified an error and published an erratum, including original images of blots alleged to be manipulated." Please reference only #31, not the hit piece #10 for this sentence. The following sentence now reads: "After further data concerns were brought to the attention of the journal, it issued an expression of concern stating: "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." It should be changed to this: "After receiving additional complaints about the original blot images posted, the editor issued an expression of concern acknowledging an ongoing investigation by the City University of New York (CUNY) and stating JNeurosci "will await the outcome of that investigation before taking further action." No need to extensively quote.

The first part of the next paragraph now reads: "PLOS One re-examined all the Burns and Wang's research papers in March 2020 and found "similarities in background pixels" in the Western blot data. Five of Wang's papers were retracted, two of which were co-authored with Burns that include the original papers on the discovery of FLNA binding as it relates to opioid receptor signaling. Three papers are on prenatal cocaine." It should be changed to this: "PLOS One re-examined all five of Dr. Wang's research papers and retracted all of them based on "similarities in background pixels" in the Western blot data.[cite retraction notices here] None of these papers are about simufilam or Alzheimer's disease. Two illustrate FLNA's role in opioid receptor signaling. The other three have no Cassava coauthors and discuss prenatal cocaine." Note that citations 35-37 are two retraction notices repeated twice. No need to extensively quote the retraction notices. Otherwise, the multitude of Wang papers that are not related to Cassava that have been cleared -- or even republished with data repeated by a different lab with the same result -- should all be discussed here. This all gets too long for this page that is about Cassava, not Wang, who has published with many different companies and academics on many different topics.

The last paragraph also does not need to extensively quote the expression of concern. I suggest this: The journal Neurobiology of Aging examined a 2017 paper on simufilam. [cite paper here, #38] Using AI-based figure proofing software, the editors found "no compelling evidence of data manipulation intended to misrepresent the results."[cite expression of concern here, #39] The journal did identify errors and issued an expression of concern, again acknowledging the ongoing investigation by CUNY.[cite expression of concern here, #39] SighSci (talk) 14:04, 17 August 2022 (UTC)

A new paragraph should be inserted before the PLOS paragraph. I suggest this:

Neuroscience investigated an allegation of image manipulation in a 2005 paper of Wang and Burns and found no evidence of data manipulation.[cite this editorial note: https://www.sciencedirect.com/science/article/pii/S0306452221005789?via%3Dihub] SighSci (talk) 14:08, 17 August 2022 (UTC)

Ack. I see there are two distinct journals (Neuroscience and Journal of Neuroscience). Unless other editors get to it sooner, I will work on this later today after reviewing all secondary sources again, but your suggestion to ignore a high-quality reliable independent third-party source (The Wall Street Journal, citation #10 in the current version) is not in accordance with Wikipedia policies and guidelines. Also, I suggest not mentioning in talk discussions citations by their current number in the article, as Wikipedia is dynamic, and those can change over time. For example, you might mention instead of #10, Michaels WSJ November 2021, or a PMID or DOI for a medical journal article. SandyGeorgia (Talk) 14:40, 17 August 2022 (UTC)

To clarify - I think I might have removed the Neuroscience article (I am not sure, but I remember seeing it and being confused why it was there). The current articles in that section all relate to FLNA (aka filamin A), I thought were relevant because Cassanova's drug simufilam is thought to act via binding FLNA. The exception being a couple of the PLOS One articles which I left in there because they seemed to be all retracted at the same time due to overlapping issues.

On the other hand the Neuroscience article does not relate to FLNA or Alzheimer's (based on the abstract), so it was not clear to me what the relevancy was to this article. I thought it made more sense to only focus on the articles that relate to the FLNA/drug, since otherwise the article is going to have to list every article related those researchers which has been investigated, which seems a very excessive (I was curious so started looking up what that would involve, and so far have found 8 articles which have been retracted and another 2 with an expression of concern).

In my opinion it would be easier to add a sentence to the start of the section stating that it was focusing on FLNA/simufilan papers, rather then list every article, but I will let someone else make that decision (I am very impression at SandyGeorgia efforts improving this article so I don't want to mess up there work) ZICgene (talk) 12:02, 18 August 2022 (UTC)

That was meant to say "I am very impressed at @SandyGeorgia efforts", apologies for the typo ZICgene (talk) 12:42, 18 August 2022 (UTC)

Thx, ZICgene; I need a block of time to get through the mess of stuff on this talk page and sort it out, and then make the edits, and a friend popped up with a medical situation yesterday that took most of my day, and same today. I do hope someone else can pitch in to start on the pieces, but it's been quite hard to decipher not only what is needed, but also what is policy compliant. SandyGeorgia (Talk) 12:58, 18 August 2022 (UTC)

No worries, hope your friend feels better soon! I don't mind helping tracking down some stuff for this section, I have just been reluctant to jump in because I am not familiar with wikipedia's rules are about company's articles (usually I just edit obscure science articles). ZICgene (talk) 14:01, 18 August 2022 (UTC)

Thanks, the mess on this page is going to take quite some time to decipher, as the damage from COI/POV editing is extensive and the disorder on this talk page makes it hard to know where to begin. If anyone is inclined to help, the earlier version listed below at #To be recovered was not entirely policy compliant, but is worth a read. Cassava stock jumped today on what is described in sources as insider trading based on the possibility of a buyout, so it's easy to see the potential for manipulation in this article. SandyGeorgia (Talk) 14:53, 18 August 2022 (UTC)

Thanks ZICgene and SandyGeorgia.

Of the five PLOS papers that were retracted (with many people disagreeing), only two involve FLNA (and not the drug), and for its role in opioid receptor signaling -- not Alzheimer's. The other three are about prenatal cocaine, not AD and not FLNA, and have no Cassava authors. "All the Burns and Wang papers" is inaccurate, because only two include Burns. All are irrelevant, unless you consider any work on FLNA by these two as relevant. True, they were simultaneously retracted for an issue involving pixels in the background of original blots in areas cropped out of the published images. No one has articulated what it is. If the three non-Cassava, non-simufilam papers are to be included (the prenatal cocaine ones), the Neuroscience one should be too -- at least that one includes a Cassava author. Up to you.

Importantly, new news today is that the remaining publication on simufilam (in JPAD) was cleared of allegations. So now ALL THREE simufilam papers have been cleared by editors. Here is the citation of today (not press release): https://www.marketwatch.com/story/cassava-sciences-academic-journal-found-no-evidence-of-data-manipulation-271660829914 SighSci (talk) 22:21, 18 August 2022 (UTC)

I should have some time to catch up tomorrow; some of the problems here are because people were using primary sources, rather than secondary mentions, and it looks like you are saying some of the wrong primary sources worked their way in. I have been iPad editing on the road most of the day, and have to be able to sit at a real computer to sort this ... perhaps tomorrow. I first want to re-read all secondary sources, and make myself a chart of what's what. Thanks for the patience, SandyGeorgia (Talk) 22:36, 18 August 2022 (UTC)

I just spent some time going back over these papers to make sure I understood the connection of these papers to this Simufilam/PTI-125 before I commented again, and I still think there is a link there that makes it valid to include the PLoS ONE papers.

Unfortunate this explanation needs to include some biology jargon, but I will try to keep it as simple as possible.

The PLoS ONE papers (“High-Affinity Naloxone Binding to Filamin A Prevents Mu Opioid Receptor–Gs Coupling Underlying Opioid Tolerance and Dependence” and “Naloxone's Pentapeptide Binding Site on Filamin A Blocks Mu Opioid Receptor–Gs Coupling and CREB Activation of Acute Morphine”) establish that FLNA contains a 5 amino acid peptide/region that is a target for drugs. The same papers establish that drugs binding that peptide/region alters the behaviour of FLNA.

The later paper that identifies Simufilam/PTI-125 as a novel compound to treat Alzheimer’s (“Reducing amyloid-related Alzheimer's disease pathogenesis by a small molecule targeting filamin A”) proposes Simufilam/PTI-125 acts by binding FLNA at the same peptide/region. Since this paper is based on the same region as the earlier PLoS ONE papers, those PLoS ONE papers are basically the foundation for the later research. There retraction raises major questions about the mechanism by which Simufilam/PTI-125 acts. That is why it is relevant to include those two PLoS ONE studies in the article. ZICgene (talk) 12:49, 19 August 2022 (UTC)

ZICgene thanks for pitching in. Unless I am misunderstanding (which may be the case), there are two issues we need to separate, which SighSci may be also misunderstanding. At this (company) article, the topic is not only Alzheimer's; it's anything relevant to the company. How far we go in to the relationship between Wang/Burns/Cassava depends on what secondary sources say. At the drug article (simufilam) we stay on topic which is the research leading to the experimental drug; that could include non-Cassava work, again, if secondary sources lend support. If a paper, as SighSci says, has no Cassava authorship, it may still be relevant at the drug article. I will put together a table so we can all get on the same page; I have needed to be home at a regular computer to pull all this together, so we can sort the proper use of primary sources and press releases, which are rife throughout these articles (this one, and Lindsay Burns and simufilam). SandyGeorgia (Talk) 14:27, 19 August 2022 (UTC)

This "Journals investigate" section really needs to be overhauled, especially in light of yesterday's news on the JPAD paper (the third simufilam paper). Now, all three papers on simufilam have been cleared of allegations of data manipulation by the respective editors. Here are two citations for the news of the third clearance: https://www.marketwatch.com/story/cassava-sciences-academic-journal-found-no-evidence-of-data-manipulation-271660829914, https://seekingalpha.com/news/3874626-cassava-sciences-cleared-of-data-manipulation-by-medical-journal-shares-rise-17?source=content_type%3Areact%7Cfirst_level_url%3Amarket-news%7Csection_asset%3Atrending%7Csection%3Atechnology

The current first paragraph about JNS (the first simufilam paper) fails to mention that they found no evidence of data manipulation, communicated to the company who did a press release in Nov. 2021 but also shown in the erratum notice that could also be cited. The expression of concern is a placeholder until the CUNY investigation reports out.

The second simufilam paper (NBA), currently the final paragraph of this section, should be moved up in priority because it involves the drug. This might be a good place to add the third paper's clearance (JPAD), discussed above.

Remember if you talk about all five PLOS retractions, three have nothing to do with the company, the drug or FLNA, but instead are about prenatal cocaine. If these are to be included, you should also mention the 2005 Neuroscience article that at least has a Cassava author. This page is about Cassava, not Dr. Wang. If it were about Dr. Wang, there are many more papers that have been investigated and cleared of allegations that could be listed. This list actually shows the extent of the attacks on him personally that are being used to damage his reputation and that of the company in an effort to impede the ongoing Phase 3 clinical trials (like the failed petition to the FDA). But I don't suggest you should include them.

Somewhere on this page (suggest in the FDA petition section), this article should be cited (it shows the financial COI of the petitioners): https://www.complianceweek.com/risk-management/the-cassava-sciences-saga-short-sellers-gaming-the-fda-and-the-damaging-ripple-effects/31416.article

Thanks for all your work! SighSci (talk) 17:30, 19 August 2022 (UTC)

The foundational paper was the 2012 J Neurosci paper, not the two PLOS papers that identify FLNA as involved in opioid receptor signaling. It was a huge leap to identify FLNA as involved in AD, and quite separate from opioid receptor signaling -- but there was a med chem library of compounds that allowed it to be investigated. What's more is that the other three PLOS papers as I've repeated exhaustively above, have nothing to do with FLNA, AD, simufilam OR Cassava. Right now, this section omits the third of only three simufilam paper's clearance of the false allegations (JPAD), as stated above. It also omits the Neuroscience (different from J Neuroscience) 2005 paper that was cleared and IS a Cassava paper. Can someone please fix all this? SighSci (talk) 13:23, 24 August 2022 (UTC)

Planned enrollment in Phase 3 trials

SighSci wrote: Furthermore, the planned enrollment in the Phase 3 trials is stuffed into "other concerns." Clearly it does not belong there but could be mentioned after the statement that the drug is in Phase 3 trials.

If you can be more specific about what you want addressed, Wikipedians can get to it quicker. The Other concerns section has the text: According to a Cassava Sciences press release, the first trial would enroll about 750 participants, and the second 1,000.[29] Stat stated that enrollment had slowed as of April 2022, as people were deterred from enlisting due to the prevailing controversies.[30] You want this mentioned "after the statement that the drug is in Phase 3 trials"; since the words Phase 3 don't occur anywhere in the article, where is it precisely that you propose this text be moved? (Phase III is mentioned for another drug.)

You also state that: The following sentence claiming that patient enrollment has slowed (by Stat) is from April, was wrong then and is very outdated now. It should be removed. Do you have a reliable, independent, third-party source stating that it "was wrong then" or "is outdated"? SandyGeorgia (Talk) 04:38, 17 August 2022 (UTC)

The details of how many subjects are planned to be included in each of the trials is a non sequitur in the "other concerns" section -- it could naturally follow the first mention that the company is conducting Phase 3 trials of its drug -- in the intro. The Stat article made an unfounded claim that patient enrollment had slowed and added this to the "concerns" section. This opinion was based on enrollment numbers released in April, when the trials had barely initiated. It is now August and the company reported over 400 subjects enrolled. This will have to cite a press release because it is not really "news," but regardless, using old numbers for this type of claim is inaccurate. The latest number is here: https://www.cassavasciences.com/news-releases/news-release-details/cassava-sciences-reports-second-quarter-financial-results-2022 SighSci (talk) 17:44, 17 August 2022 (UTC)

See response below re overuse of Cassava press releases. SandyGeorgia (Talk) 19:07, 17 August 2022 (UTC)

Also, please stop disclaiming WP:RS like The Wall Street Journal and STAT. These are considering reliable sources for Wikipedia, and your disagreement with them should be based on equally reliable sources, not Cassava company statements and press releases. Cassava's positions can be stated if done so neutrally and according to due weight, and not done in a self-serving manner. STAT and WSJ are reliable sources. If you claim them to be "wrong", you need an equally reliable independent third-party neutral source that backs up your claim. SandyGeorgia (Talk) 19:13, 17 August 2022 (UTC)

Regarding the NYT article, it might be worth noting/citing the Letter to the Editor of the NYT that was filed with the SEC by the company as an 8-K: https://www.cassavasciences.com/node/15906/html SighSci (talk) 01:46, 22 August 2022 (UTC)

Intro needs help

The intro should talk about what is current first, not that the company was established under a different name with a different focus 24 years ago. This paragraph should be moved to the end or deleted: "Established in 1998 as Pain Therapeutics, the company moved its focus from opioids to drug development for neurodegenerative diseases and changed its name. Cassava initially worked on three drugs: Oxytrex and Remoxy, which were pain drugs, and PTI-901, which aimed to treat irritable irritable bowel syndrome. None of the drugs have been approved by the Food and Drug Administration (FDA)."

Suggested first paragraph: Cassava Sciences (symbol: SAVA) is a pharmaceutical company based in Austin, Texas, U.S. Its founder, chief executive officer and president is Remi Barbier. Cassava is developing simufilam (previously known as PTI-125 and sumifilam), an oral-tablet drug candidate for the treatment of Alzheimer's, as well as SavaDx, a blood-based diagnostic for Alzheimer's disease. Phase 3 trials of simufilam started in 2021. According to a Cassava Sciences press release, the first trial would enroll about 750 participants, and the second 1,000.[1] As of August 2022, over 400 people had enrolled in the trials. [2]

After this sentence: "A Citizen Petition was filed to the FDA in an attempt to suspend the clinical trials, but it was rejected as the FDA ruled the request fell out of the purview of the Citizen Petitions." Please add the following: "Because the petition was filed by short-sellers, many viewed it as market manipulation for financial gain." Please cite this article for this new sentence: https://www.complianceweek.com/risk-management/the-cassava-sciences-saga-short-sellers-gaming-the-fda-and-the-damaging-ripple-effects/31416.article

At the end of the paragraph discussing the DOJ investigation, please add this sentence and citations below: "Reuters and Bloomberg also reported that the DOJ launched an expansive investigation into "short and distort" campaigns of short-sellers who aim to profit from manipulating stocks."

• https://endpts.com/biotech-short-seller-lands-on-feds-watch-list-amid-sweeping-probe-report/
• https://www.reuters.com/markets/europe/us-doj-launches-expansive-probe-into-short-selling-bloomberg-news-2021-12-10/
• https://www.bnnbloomberg.ca/hedge-funds-ensnared-in-expansive-doj-probe-into-short-selling-1.1694005 SighSci (talk) 18:19, 17 August 2022 (UTC)

Generally (so I can start looking at all of this tomorrow):

1. Is there an independent source supporting start of Phase 3 trials?
2. The number of planned enrollees in a trial is excess detail for a lead; Wikipedia is WP:NOT a webhost, or a recruiting tool. This is the company article; there is a drug article. The number of planned enrollees is probably not apt for the lead over there either.
3. It's unlikely that anything based on the company's press releases is WP:DUE for the lead, particularly when it contradicts WP:RS.
4. I don't believe the FDA matter belongs in the article oopsie, lead at all; it's a dated and resolved issue.
5. The entire sentence proposed beginning with Because the petition was filed by short-sellers, many viewed is POV, and a MOS:WEASEL, contradicted by secondary sources, and not likely to be supported by the source you provide (which I can't read because it's member-only content). Please provide extensive (avoiding WP:CHERRY, that is, both sides of the story) quotes from the article so I can see if anything from it can be unweasled and converted to encyclopedic (WP:NOTNEWS) prose.

   Further, if we are to use complianceweek.com at all, please explain why you think it meets Wikipedia's requirements for reliable sources. Things like: what is its reputation for fact-checking, its journalistic credentials, its journalistic oversight, its policies for retracting and correcting errors, other important reliable sources that use it? What are the staff qualifications and expertise? Etc ... SandyGeorgia (Talk) 05:40, 19 August 2022 (UTC)

6. You've requested that text be added about the DOJ investigating short sellers based on three sources, not one of which ever mentions Cassava; please read WP:SYNTH. I am compelled to again explain that Wikipedia is not a WEBHOST, and text must be based on reliable, independent, third-party sources, that is neutral and avoids original research. We can't take a DOJ investigation unrelated to Cassava and use it to draw connections in this article; that is synthesis of sources.

That leaves me very little, but a bit, left to work with from your suggestion when I start looking more in depth tomorrow. SandyGeorgia (Talk) 04:55, 19 August 2022 (UTC)

In response to above:

1. If you want a date of Phase 3 start, I don't know. It must be somewhere. I don't think it is that important. Omicron paused everything industry wide, so start date is not that relevant.

2. I agree. Number of planned enrollees was stuffed into the "other concerns" earlier, juxtaposed with the article falsely claiming "enrollment has slowed." It is unnecessary.

3. Sounds good. Take it out, along with the planned enrollment numbers.

4. Agree. Old news still being rehashed as FUD. Certainly doesn't belong in the intro.

5. The Compliance Week article shows the financial COI of the petitioners. It specifically covers this kind of thing so should have the expertise needed. Also, here is a Bloomberg article reporting that the short-sellers made $100 million by shorting the stock and then filing the petition, which they knew would tank the stock price (and yes, is illegal market manipulation for financial gain): https://www.bloomberg.com/news/articles/2021-08-31/cassava-short-sellers-reap-100-million-in-august-stock-rout#xj4y7vzkg

6. You can omit, but DOJ investigations are supposed to be confidential so would not state. However, as this stock is one of the most shorted stocks (there are dynamic rankings), it surely is part of that massive investigation. SighSci (talk) 19:46, 19 August 2022 (UTC)

SighSci this 17:18, August 19, 2022 version of the lead is what I have done so far from your post above. Some of your requests are not actionable for the reasons I mentioned, and this is still a work in progress. I've eliminated the need for a source to document the need for a precise date Phase 3 trials started by rewording the sentence; the precise date isn't that relevant to this article anyway. I've also reworked everything so that content in the lead can be cited to secondary sources when they are available, minimizing WP:UNDUE content. I've worked in more general company info that belongs in any company lead (likely more is needed, but we can get to that once the rest is better managed).

We still have to resolve #No products approved by the FDA and #SavaDx; discussion of those can continue below in those sections. This is only a first pass, as I continue to try to sort the mess on this page, and the lead will surely continue to evolve and change.

That said, is there anything else that you see as needing urgent attention in the lead, while we continue to try to sort the rest? SandyGeorgia (Talk) 19:17, 19 August 2022 (UTC)

Anyone in biotech will tell you that no products approved by the FDA (and hence no product revenues) is the norm. This is not really notable. The company has stated that SavaDx is lower priority and earlier stage, so it should not be mentioned before simufilam. SighSci (talk) 19:49, 19 August 2022 (UTC)

Some corrections and updates so far to the lead, here (I believe I had the date on the name change wrong???) My question is whether SavaDx should be mentioned at all (in the lead, that is); are there any secondary mentions of it? SandyGeorgia (Talk) 20:46, 19 August 2022 (UTC)

SciSigh Re point 5 above: we already know from multiple reliable sources (and the obvious) that short sellers have a financial motive. We do not need to use a members-only paywalled source to cover that.

Bloomberg is a very reliable source and usable for this (see WP:RSP); when I get to that part, I can probably use that source (have not yet read it all). Good (neutral) content lets the facts speak for themselves.

I left the FDA in the lead, because I was able to (I hope) work it in contextually. A high-quality financial source like the WSJ should be reliable to cover what is significant in the industry. I suggest that you include in your expectations of where we will end up the matter of scientists who do not have financial skin in the game but have also expressed concern: this article is not going to end at a place of "everyone who raised concerns had a financial motive".

When the article is improved to a more decent state, we could certainly approach the COI noticeboard to get more eyes on it and other opinions of what to include or not in the lead, but for now it is in a state that few other editors would be thrilled to have to wade through it, and we'd be likely to get no good feedback. With a bit more work, we should be able to get it to a better place for further review. I still have a lot to get through, and all day tomorrow free to continue to work. If you remain dissatisfied about whether to use Compliance Week, the same applies: we could get more eyes on it by approaching the reliable sources noticeboard, but most editors would throw up their hands when seeing the dismal state the article is currently in, and we may not get good feedback just yet. SandyGeorgia (Talk) 21:04, 19 August 2022 (UTC)

PS, before I use Bloomberg, a good source, I need to fully review the Quanterix sources and matter ... I am loathe to work in that source before I have a balanced representation of all of it, and that includes Quanterix. SandyGeorgia (Talk) 21:08, 19 August 2022 (UTC)

Thanks for so much work. The lead is fine. 2600:1700:BB80:88A0:C999:C3CC:65DA:1E3D (talk) 22:26, 20 August 2022 (UTC)

Thanks; we'll get there. SandyGeorgia (Talk) 03:55, 21 August 2022 (UTC)

Intro is much better, but there is one thing that is still quite misleading. The Reuters article did not state that the DOJ investigation started "after" the FDA rejection of the petition. Remember the existence of any DOJ investigation is supposed to be confidential but was leaked at an opportune moment for the short-sellers. The company had disclosed in a press release that "certain government agencies" were investigating in November, because the SEC investigation was also leaked to WSJ in November. No one knows when these initiated but it is a good guess that everything was simultaneous around the time of the petition. SighSci (talk) 16:28, 25 August 2022 (UTC)

Compensation plan lawsuit

https://news.bloomberglaw.com/esg/cassava-investor-sues-over-pay-plan-adopted-before-stock-spike SandyGeorgia (Talk) 16:02, 27 August 2022 (UTC)

Saving should secondary sources warrant: Delaware docket report, Case ID 2022-0737, Jeanne Calamore, et al v Remi Barbier, et al SandyGeorgia (Talk) 16:23, 27 August 2022 (UTC)

Leadership

Thank you. Another thing that stands out in the current version is that the "leadership" section only mentions the CEO and one SVP. No officers besides the CEO, or board members, who should be mentioned if discussing leadership. Suggest you include some of those (e.g. https://www.bloomberg.com/profile/person/1654181 and https://www.globenewswire.com/news-release/2021/01/04/2152667/0/en/Cassava-Sciences-Appoints-Dr-James-Kupiec-as-Chief-Clinical-Development-Officer.html and board members https://www.franciscopartners.com/team/sanford-robertson and https://seekingalpha.com/news/3705973-cassava-sciences-adds-richard-barry-to-its-board-of-directors) or omit the singling out of Burns just because she is married to Barbier and at the crosshairs of the false allegations. She is not really leadership and there are other SVPs at the company. SighSci (talk) 21:20, 22 August 2022 (UTC)

SighSci one should actually not do this (per talk page guidelines), but I hope you don't mind that I split your post in two for ease of response. If you would keep in mind to make posts in a way that they could be more easily addressed, you would increase the likelihood that they would be. I split this out so that I can next ask others for help on this particular portion of your post; by burying it an unrelated topic, you make it harder for others to read and respond. SandyGeorgia (Talk) 17:32, 27 August 2022 (UTC)

Z1720 as you have experience with company articles, and more specific experience with COI on those, I'm hoping you can help out here. SighSci has a declared COI wrt this company.

SighSci (incorrectly) states that SVP Lindsay Burns is the only person besides the CEO mentioned in the leadership section of this article "because she is married to Barbier" (the CEO, president and founder of the company), and again providing churnalism sourcing, while (correctly) noting that she is "at the crosshairs" of what SighSci considers "false allegations". Wang is also mentioned, and is also in those "crosshairs", all (including the marriage aspect) according to high quality reliable sources like The Wall Street Journal, The New York Times, and The New Yorker.

Summarizing, first for SciSigh:

1. Burns is not "singled out" because she's married to Barbier; she's mentioned because high-quality reliable sources specifically name her, because
2. She is quite at the heart of the allegations, as the SVP and co-author (with Wang) of some papers in question, which reliable sources specifically mention (along with Barbier and Wang, also in a leadership role on the advisory board), and
3. Unlike most of the rest of the people in the company's leadership, she decidedly meets Wikipedia notability.

For Z1720, since SciSigh raised this concern several days ago, I have attempted (without success) to find any

1. WikiProject or other guideline describing best practice for what/whom to include in company leadership positions, or
2. featured content to serve as an example.

I can make a case for including the Board of Directors based on some featured content, but it's not a strong case, but I can find no other justification for including all senior management (as WP:NOT laundry list or webhost). Is there something I've missed? Do we include a not-notable list of all of the company's leadership, based only on Cassava's website, or do we only include those mentioned by secondary sources and/or meeting notability?

Feedback needed; Firefangledfeathers or ZICgene might also have ideas. Otherwise, we may need to approach the COI noticeboard (we may be headed that direction anyway if we can't make some headway here in terms of understanding how Wikipedia works so I don't have to keep spinning my wheels reading churnalistic press release sourcing). SandyGeorgia (Talk) 17:53, 27 August 2022 (UTC)

Now, having asked the questions, I'll include my own conclusion. Wikipedia is WP:NOT a webhost, we don't need a list of non-notable senior managment or those not covered by secondary sources, and I can find no guideline or featured content that suggest that best practice would be to include what SciSigh is asking for. SandyGeorgia (Talk) 18:06, 27 August 2022 (UTC)

@SandyGeorgia: Wikiprojects addressing companies and COI are semi-active, at best, so it's been mostly individual editors who judge content to include based on policy and guidelines. There might be a response on Wikipedia:WikiProject Edit requests and WP:COIN if you want additional opinions.

If you are looking for FAs on companies, Cracker Barrel and BAE Systems are the best options; though both are not great, they are FAs that I would not try to delist immediately.

For listing senior management, I usually either list everyone of a certain rank, or none of them. Two things I consider are 1) how many of these people are there and 2) are they notable. In this case, Burns is mentioned as an author of papers promoted/paid for by the company since she is mentioned in other parts of the article, I think it would be beneficial to mention that she is a SVP, but I don't think it is important to mention that she is the wife of Barbier in the leadership section (if it is important in other aspects of the company, such as talking about events in company history, it can be mentioned there). I would probably mention everyone listed as management on the company website here in a paragraph of prose about the company management, and try to find other sources that state when these people entered these positions of responsibility. Z1720 (talk) 19:24, 27 August 2022 (UTC)

Yes, I checked Cracker Barrel and BAE and found them not very helpful. I'll wait to see what other feedback surfaces, and then implement suggestions. Thx, Z! SandyGeorgia (Talk) 19:31, 27 August 2022 (UTC)

This sounds reasonable. Otherwise the section should not be called "leadership" if it only mentions the CEO. I agree that wife of Barbier is an unnecessary personal detail here and not about the company. SighSci (talk) 19:54, 30 August 2022 (UTC)

SighSci I will not be acknowledging your posts until you have responded here at the Noticeboard for Conflict of Interest. SandyGeorgia (Talk) 20:01, 30 August 2022 (UTC)

IIb to III

There's a content gap here that needs to be filled in; reminder. How we got from failed Phase IIb to III. SandyGeorgia (Talk) 04:06, 21 August 2022 (UTC)

It wasn't a failed trial. It was a failed analysis of CSF samples in the first lab that was evident by high variability in the placebo group, which should not happen over the short duration of the trial. Also, there was no correlation between biomarkers in change from baseline in the placebo in the first analysis. (It is implausible that an individual would show dramatic worsening on one biomarker and improvement on another -- this was a sign of the very high variability). Backup CSF samples were sent to the second lab and this analysis showed significant differences vs. placebo and high correlations in change from baseline between biomarkers. Both labs were blind to treatment group. Results are not published yet; it appears to be on hold while the CUNY investigation concludes, but this is mentioned in a comment on Alzforum. So no citation. SighSci (talk) 02:06, 23 August 2022 (UTC)

Done, [3] SandyGeorgia (Talk) 20:36, 31 August 2022 (UTC)

Repetitive and POV text on short selling

Mnachtrab your redundant edits of today are indicating potential POV issues.

This is a summary of your edits so far. Here are the problems:

1. That Bredt and Pitt were shortsellers is already mentioned in the third paragraph of the Citizen petition section.
2. You have added a second mention of them as short sellers to the second paragraph in a sentence that refers to remuneration one can receive for whistleblower tips.
3. And then you've added a third mention of them as shortsellers to the fourth paragraph.

So, in one short section of the article, the article now manages to name the filers of the petition as shortsellers three times. (Do you notice that every time Barbier defends the company, it is not redundantly mentioned that he stands to reap $108 million if he can convince the stock market?). Your edits are not only redundant; by repeating a point three times in one section, they've introduced a clear POV.

I can correct your redundancy with some restructuring to get all the repetitive info (eg the missing name of the attorney and his relationship to all) in one place, but am not going to attempt repair without discussion first. Please discuss your edits. SandyGeorgia (Talk) 14:07, 1 September 2022 (UTC)

Good points on mentioning it too much. In critical areas where introducing new characters, its important to make the link. I like the idea of linking them by name versus saying they are short sellers over and over. For example, for the attorney mentioned by the $18000 payment we could just name Jordan Thomas. I am fine with that as he is clearly representing short sellers. — Preceding unsigned comment added by Mnachtrab (talk • contribs) 14:35, 1 September 2022 (UTC)

Please thread your talk page responses correctly by adding a colon before them, and please sign your talk page edits by adding four tildes ( ~~~~ ) after them.

Allright; once you indicate you are done with your edits, I will restructure the section to (hopefully) add clarity on all the relationships – mentioning them once (not three times). I am not going to do that until I am certain you are done editing, as I don't WP:EDITWAR. Is there something else you intend to add to that section, so I don't waste my time? If you intend to use Compliance Week, I will be requesting quotes from the article to be sure you are fairly representing the entire article (per the discussion at the Reliable Sources Noticeboard). SandyGeorgia (Talk) 14:45, 1 September 2022 (UTC)

Further, POV is suggested with this edit (and its edit summary), where you've introduced undue information without context. The over 400 is a meaningless number, based on a primary source, without any context given from a secondary source, and is also introducing POV, by removing information that did supply context. If you continue editing in a way that suggests POV, it may be more expedient to simply revert the lot and start over. SandyGeorgia (Talk) 14:59, 1 September 2022 (UTC)

COI notice added to the article, as your edits have reintroduced a POV present in previous versions, and discussed at the COI noticeboard. SandyGeorgia (Talk) 15:05, 1 September 2022 (UTC)

As of 16:04 September 1, we are back to a non-COI version. I will repair the attorney name issue once POV editing stops. SandyGeorgia (Talk) 16:43, 1 September 2022 (UTC)

I stopped editing. I'll do suggestions through talk for now. Mnachtrab (talk) 18:09, 1 September 2022 (UTC)

The 2020 Cash Incentive Bonus Plan

I put some more explanation of what the The 2020 Cash Incentive Bonus Plan is all about and how it works. It looks like it was deleted. It seems people are confused by the plan and I think my comments explain it well as they reference the actual plan in the SEC filings. Also its important to know no payments have ever been made as the plan has clear rules on how it gets paid. I put it in the section it was brought up in under leadership. Is there a reason you deleted my edits Sandy? I think I am just stating facts as they are in their filings. Do you want to move this under a different section?

As of the August 4th 2022 filing of the Form 10-Q for the period ending June 30th 2022 , no actual cash payments were authorized or made to participants under the Plan.

The 2020 Cash Incentive Bonus Plan was approved by the Board of Directors of Cassava Sciences on August 26th of 2020 and communicated to the SEC in an 8K filing on September 1st, 2020. Reaching a Target Value Milestone referenced in the 2020 Cash Incentive Bonus Plan does not mean the bonus is released as a Bonus Payment to Participants in the plan. Two events can cause a Bonus Payment or Partial Bonus payment. First, a merger or sale that result in a "change in ownership of a corporation" or "change in ownership of a substantial portion of a corporations assets". Second, the Company will defer actual payment of any Bonus Payments under the Plan until such time as the Company has sufficient cash remaining as determined by the Compensation Committee, after pro forma payment of all such Bonus Payments, to meet the Company’s projected cash needs for the subsequent twenty-four (24) months (“Remaining Sufficient Cash”). In each case, Remaining Sufficient Cash will be based upon the Company’s then-current operating plan and budget previously approved by the Board for the subsequent twenty-four (24) months, net of any direct expenses related to ongoing or projected Phase II/III and Phase III clinical studies. If the Company does not have Remaining Sufficient Cash to pay the Bonus Payments in full, the Company shall make partial Bonus Payments to each Participant to the extent it has Remaining Sufficient Cash to do so. Such partial Bonus Payments shall be paid pro rata to each Participant as an equal percentage of the Bonus Payments which each Participant is entitled to receive. Mnachtrab (talk) 15:32, 1 September 2022 (UTC)

To me i'm not sure its really that relevant overall to even be talked about in their page except that detractors of Cassava bring it up a lot. Mnachtrab (talk) 15:45, 1 September 2022 (UTC)

References

We base content on secondary sources and they have discussed this plan, which we, very briefly summarise. We cannot use primary sources wuch as those you have linked to, to trump what has been written in secondary sources. This is explained in WP:PSTS. Please familiarise yourself with this policy. SmartSE (talk) 16:10, 1 September 2022 (UTC)

I see the edit. Why does it need secondary sources? The quotes are taken directly from SEC filing of the language of the plan. And the comment about the payments made is right from their 10Q. The quotes included in the article are confusing and make it seem like they received cash bonuses. Also calling it an "unusual compensation scheme" is very biased. The plan is not as represented in this quote from the cassava wiki. My edits are much more fact based than what is written in the article:

Patrick Keefe wrote in The New Yorker in January 2022 that Cassava had an "unusual compensation scheme" based on its stock market valuation. The cash bonuses were not based on producing successful drugs or gaining FDA approval, rather are tied to the price of the company's stock and "specific valuations for twenty consecutive days". The WSJ stated that, under this plan, Barbier could reap $108 million.

Honestly, like I said I do not think this bonus plan is very important in the grand scheme of things so I suggest just removing this reference as a possible solution. Or if you want I could simplify it to some of the language from the 10Q like this:

As of the August 4th 2022 filing of the Form 10-Q for the period ending June 30th 2022 No actual cash payments were authorized or made to participants under the Plan.

Participants of the 2020 Cash Incentive Bonus Plan not be paid any cash bonuses unless (1) the Company completes a merger or acquisition transaction that constitutes a sale of ownership of the Company or its assets (a Merger Transaction) or (2) the Compensation Committee of the Board (the Compensation Committee) determines the Company has sufficient cash on hand in the board approved operating budget net of direct expenses related to the projected Phase Two and Three clinical studies. Mnachtrab (talk) 16:19, 1 September 2022 (UTC)

Mnachtrab please read WP:THREAD so I don't have to keep fixing your posts.

Also, please respond at WP:COIN regarding any conflict of interest you may have with respect to Cassava, its research, and Alzheimer's. Until you have done that, I won't be directly acting on your requests here on this talk, and suggest a revert of your contributions so far.

I will be happy to address the missing attorney name, once your POV edits are removed. Meanwhile, my attempts to address the gaps in the article has been impeded by interference from COI and paid editing. SandyGeorgia (Talk) 16:30, 1 September 2022 (UTC)

But, wow, just wow; thanks for sharing. "Participants of the 2020 Cash Incentive Bonus Plan not be paid any cash bonuses unless (1) the Company completes a merger or acquisition transaction that constitutes a sale of ownership of the Company or its assets (a Merger Transaction)." SandyGeorgia (Talk) 16:32, 1 September 2022 (UTC)

(edit conflict) I just provided a link explaining why secondary sources are required - did you read it? What you "think" is frankly irrelevant - what matters is that Keefe and the WSJ thought it was important enough to mention in their articles. SmartSE (talk) 16:33, 1 September 2022 (UTC)

Just imagine the original research we would have to engage in if we were to explain how the merger and acquisition wording affects Burns and Barbier; that's why we don't use primary sources, although MNachtrab's post provides some most interesting context. We don't include this until/unless secondary sources mention it-- not the least because the picture it presents is not very flattering. SandyGeorgia (Talk) 16:36, 1 September 2022 (UTC)

I see a little bit on what your saying but that New Yorker article is written like an opinion piece. Calling a board approved compensation plan a "scheme" then quoting that in this company wiki does not seem appropriate. Definition of scheme is "make plans, especially in a devious way or with intent to do something illegal or wrong" The article also falsely states it is pegged "just to the share price". It has to hit a share price in an acquisition or achievement of free cash flow above the 24 month budget after paying for phase 3 trials AND it needs approved by the compensation committee.

For the wall street journal reference, it is a little less biased in its reference because it says ALL MILESTONES, but in the wiki cassava page the word REAP is used vs receive:

Wall Street Journal sentence

"Under the bonus plan, which also applies to Cassava’s officers, Mr. Barbier could receive as much as $108 million if all valuation milestones are met, according to securities filings." Mnachtrab (talk) 16:47, 1 September 2022 (UTC)

And I added a conflict of interest to my profile that I own cassava stock Mnachtrab (talk) 16:48, 1 September 2022 (UTC)

Good, now we've got that clear; if you have any other COI to reveal, for example, any personal relationship relating to Alzheimer's disease or its research, that should also be revealed.

I will be on the road, hotspot editing for the rest of today, and will be happy to add and fix the missing attorney name when I am home, and when your POV edits cease. Your "opinion piece" comments are irrelevant, as the sources are not opinion pieces and are reliable sources. On the other hand, the press releases so frequently added here are most certainly opinion ... Cassava's. SandyGeorgia (Talk) 16:57, 1 September 2022 (UTC)

Wonderful; I see Smartse already got that. I might work on improving the flow of the entire section if I'm ever given time without COI edits impeding progress. SandyGeorgia (Talk) 17:00, 1 September 2022 (UTC)

Re: reap vs. receive, please review WP:PARAPHRASE. SandyGeorgia (Talk) 16:58, 1 September 2022 (UTC)

"Reap" rephrased. SandyGeorgia (Talk) 21:03, 1 September 2022 (UTC)

Suggested Addition to Alzheimer's Section

I suggest deleting the reference to the STAT article as it is opinions and not fact. If that is not deemed appropriate, maybe change the reference to an actual quote like below. It is much more fact based:

STAT stated in April that 60 patients have begun treatment across its two Phase 3 clinical trials. The studies opened in September and November of last year, respectively, so in roughly six months, Cassava has only managed to recruit 3% of the patients it needs to complete the Phase 3 program.

Here is a suggested addition below the stat news reference of an update to the trial on August 4th since nobody has covered it we should be able to use a primary source of their 10Q filing:

As of August 4 2022 in a 10-Q filing, over 400 subjects have enrolled in the phase 3 program or over 23% and patients are being screened in in clinical trial sites in the U.S., Canada, Puerto Rico and Australia.

Or here is another option from a seeking alpha article if its better to quote a 3rd party bio analyst:

As of August 4th Edmund Ingram writes "these 2 trials have enrolled >400 patients, with a roughly equal split between the 2 trials. With the target being ~1,750 patients, we are therefore now at 23% enrolment, which has taken ~9 months.

Edmund Ingram author of Haggerson BioHealth - detailed analysis focused on biotech and healthcare stocks https://seekingalpha.com/article/4529663-cassava-sciences-some-learnings-from-q2-2022-earnings Mnachtrab (talk) 17:17, 1 September 2022 (UTC)

References

You don't seem to be paying much attention - yet again you are suggesting removing a secondary source and replacing it with a primary source. Seekingalpha is self-published and therefore not reliable. See e.g. Wikipedia:Reliable_sources/Noticeboard/Archive_289#Seeking_Alpha. SmartSE (talk) 17:33, 1 September 2022 (UTC)

Where am I doing that? I think it is a little rude to say I am not paying attention. That comment is not neccesary.

I am suggesting adding a new update that has been sent in addition to the stat. It is a drug trial. STAT updated it in april saying only 60 have enrolled. 4 months later the company and the seeking alpha article updated the number thourgh august to 400. The stat update is no out of date so it could be removed. If it is new data that is not available on a secondary source yet, can that be updated since it only has a primary source for that data. Other parts of the article refer to primary sources. Mnachtrab (talk) 20:39, 1 September 2022 (UTC)

Mnachtrab for the third time, please read WP:THREAD. SandyGeorgia (Talk) 20:43, 1 September 2022 (UTC)

Perhaps you have not noticed that nowhere does the current version of the wikipedia article mention any specific current enrollment numbers. Please review WP:NOTNEWS. To mention the current enrollment numbers, we would need to provide context, which comes from secondary sources, not press releases and not non-reliable sources. SandyGeorgia (Talk) 20:45, 1 September 2022 (UTC)

It reports from STAT early in the enrollment and says enrollment is slow as of April. Why would you say its slow as of April but not update more later if it speeds up? Other wiki sites and this cassava site reference company press releases for drug related information. Also its not a press release its referenced in a 10-Q filing with the SEC. Mnachtrab (talk) 20:54, 1 September 2022 (UTC)

Mnachtrab for the fourth time, please read WP:THREAD. SandyGeorgia (Talk) 21:04, 1 September 2022 (UTC)

I am not responding again to your time-wasting. As has already been explained multiple times because STAT is a secondary source, and we don't use primary sources to contradict secondary sources. If you can't understand this, perhaps you need to be disinvited to post on the talk page again. SandyGeorgia (Talk) 21:09, 1 September 2022 (UTC)

I will say it for the second time. I am not contradicting a secondary source. I am saying new information was released updating the enrollment from April to August. The august data update I provided. It is only available as a primary source. Mnachtrab (talk) 22:02, 1 September 2022 (UTC)

Allegations of research fraud

This section currently states: "Two anonymous sources told Reuters that the investigation involved criminal charges over allegations that the company had "manipulated research results"". Criminal charges is incorrect. The investigation involves requests for information only. The company has not been "charged" with anything. Suggest removing "criminal charges over". They are investigating allegations only, and this should be accurate. SighSci (talk) 17:15, 29 August 2022 (UTC)

I removed "criminal charges over". I think the possibility of later criminal charges is covered well enough by the preceding sentence's "criminal investigation". It'll save a bit of volunteer time if you ground your arguments in the sources more. This one was plentifully and immediately supported by the Reuters source, but it's not always that easy. Firefangledfeathers (talk / contribs) 17:33, 29 August 2022 (UTC)

Good change, not because the text was inaccurate, but because it was redundant. Hoping to catch up on the rest here by today or tomorrow. SandyGeorgia (Talk) 17:39, 29 August 2022 (UTC)

Thank you, and thanks for the suggestion. It's not always the sourcing that is an issue but the spin put on the sentence that needs undoing. SighSci (talk) 18:10, 29 August 2022 (UTC)

If you would be less negative, and refrain from using the talk page for unnecessary opinion, you'd probably be more likely to get more editors to respond more quickly. Criminal investigation implies allegations of criminal charges. It's not "spin"; it was a redundancy (committed by me). It is no less "spun" or "unspun" now; it is just less redundant. SandyGeorgia (Talk) 19:23, 29 August 2022 (UTC)

In this same section, in referencing the Biospace article on the rejection of the FDA petition, "request" is not clear. It is referencing the request to FDA to conduct an investigation (because evidence was not provided by the petitioners). Here's the quote from the article: "In its response, the FDA said the petitions “do not purport to set forth all relevant factual information.” The FDA said the petition calls for the FDA to launch a fact-finding investigation, which is not under the purview of the Citizen Petition." Suggested change would be to say "...Feb 2022, as the request that FDA conduct an investigation fell out of the purview of the CP process." This "out of purview" keeps getting spun to imply that FDA acknowledges wrong-doing but just can't halt these trials, which is incorrect. SighSci (talk) 18:27, 29 August 2022 (UTC)

Continued in separate section below. SandyGeorgia (Talk) 19:36, 29 August 2022 (UTC)

FDA Citizen petition "request"

SciSigh let me show you how to make a short neutral request based on sources that anyone can follow; this would save time, stop the talk page clogging, and make it more likely others would help.

Please change the following in this version of the Allegations of research fraud section:

• Current: A citizen petition attempting to suspend the clinical trials had been rejected by the Food and Drug Administration (FDA) in February 2022 as the request that FDA conduct an investigation fell out of the purview of the citizen petition process.
• Suggested: A citizen petition attempting to suspend the clinical trials had been rejected by the Food and Drug Administration (FDA) in February 2022 as the request fell out of the purview of the citizen petition process.
• Sources: Biospace and Reuters

And then you would add a short explanation here. Is that the change you intended? Because to get here, I had to first sort out that you appended something new on to an existing discussion (start a new section for a new request-- then we can start archiving away stuff that is done and not have such an overwhelming talk page that no new editor will want to engage). Then I had to figure out which text you were referencing when you said "request" and included too many quote marks to sort what was what, and then I had to get the source (since you didn't link it on this page either, and then I could add a side by side comparison the change you seek. Do you not see how hard you are making it to try to help here, or understand that, as all Wikipedia editors are busy across many articles, your approach has made it very difficult, and demotivating, to engage here? Editors trying to help here don't have a "spin"; they are trying to reflect sources. Many of your posts on this talk page do have a spin, and it's not helpful. I have plenty of other articles that need my attention, and don't enjoy misspending the little time I have.

Let me know if I've got it right above, and if so, that can be done. SandyGeorgia (Talk) 19:36, 29 August 2022 (UTC)

Full support on SG's advice here. Emphasis on the "short" in "short explanation". On this specific issue, I think the "current" and "suggested" are switched. I'm neutral on the change, and I'm inclined to enact requested edits that I'm neutral on, pending a short waiting period for other voices to chime in. Firefangledfeathers (talk / contribs) 20:05, 29 August 2022 (UTC)

Well that shows how confusing SciSigh's posts are ... I think I got Current and Suggested change correct ... ???? Current is what is in the version I linked. SandyGeorgia (Talk) 01:28, 30 August 2022 (UTC)

I did have the two above backwards; got it now, I think. SandyGeorgia (Talk) 06:52, 2 September 2022 (UTC)

Pausing point

If any (neutral) editor still following can still handle the volume of <stuff> to get through on this talk page, with this version, I believe I am at a point of having recovered and worked in everything lost before the COI/paid editing, and addressed everything that is worthy of inclusion from this talk page. I haven't yet read back through The New Yorker to see if I need to recover anything from there. I'm going to pause for feedback from others, and once we've gotten the body of the article in better shape, the lead may need examination. SandyGeorgia (Talk) 00:55, 2 September 2022 (UTC)

And I've now been through The New Yorker again, worked in a bit more, but that article is so long and goes so many different directions that it puts me to sleep, so I hope someone else will read through it. SandyGeorgia (Talk) 02:19, 2 September 2022 (UTC)

I pushed on and worked in some bits from Compliance Week, but tired, didn't finish; see #CW articles section above. SandyGeorgia (Talk) 08:14, 2 September 2022 (UTC)

Elisabeth Bik interview

https://www.fiercebiotech.com/biotech/data-manipulation-scientist-elizabeth-bik-cassava-athira-alzheimer-s-data-manipulation SandyGeorgia (Talk) 03:18, 3 September 2022 (UTC)

Compliance week article research should be added

There is a great compliance week article that covers the Cassava Sciences story with deep research from 2017 though the FDA denial of the citizens petition.

https://www.complianceweek.com/risk-management/the-cassava-sciences-saga-short-sellers-gaming-the-fda-and-the-damaging-ripple-effects/31416.article

The author is an investigative reporter and Compliance week is a business intelligence service on corporate governance, risk, and compliance that features daily news and analysis, a quarterly print magazine, proprietary databases, industry-leading events, and a variety of interactive features and forums.

She tells the story of early grants from the NIH, positive results from early studies, approval of their phase 3 program under SPAs. She goes on to cover the events surrounding the citizens petition filing that stated:

"In the original CP, dated Aug. 18, Thomas certified that “to the best knowledge and belief of the undersigned, this petition includes all information and views on which the petition relies, and that it includes representative data and information known to the petitioner which are unfavorable to the petition.”

and

"Yet, on Aug. 26, Labaton Sucharow issued a press release revealing the new information that Thomas’ clients “also hold short positions in Cassava stock.” Short selling is the practice of investors betting on a company’s stock price falling—a strategy the Department of Justice is currently investigating.

She talks about some of the other conflicts of interest from the filers of the CP. She talked about how "The other short seller was Geoffrey Pitt, a cardiologist and professor at Weill Cornell Medical College. According to anarticle in The New Yorker published in January 2022, Pitt had never shorted a stock before but decided to do so after Thomas agreed to take their case."

She goes on to discuss how Citizens petition is frequently used to game the FDA and how the FDA denied the CP. Jaclyn did extensive research and it is a great investigative article that loops in many of the events covered on this wiki page. Mnachtrab (talk) 09:13, 19 August 2022 (UTC)

Please see my post just above this one at #Intro needs help, and answer the specific questions I asked about this source. "Jaclyn did extensive research and it is a great investigative article" is an opinion that doesn't address the reliability of a source in ways that Wikipedia requires. "The author is an investigative reporter" gives us ZERO facts to go on, nor does repeating the About info from Compliance Week's website. Anyone can put up a website. I have searched the archives at WP:RSN, and have found nothing there; we are likely to end up there, waiting at least a week for answers as to whether we can use this source, if someone doesn't answer those questions here. Who is their staff, what are their journalistic credentials, what is their reputation for fact-checking, how do they handle retractions, what other reputable sources rely on them ... things like that are needed to establish reliability.

I need to once again remind that Wikipedia is not a webhost, and filling this talk page with information that is completely irrelevant to Wikipedia's policies and guidelines is only making it longer and harder to get the content matters settled. There is ZERO in this post I can use. We already know the events surrounding the FDA petition, as we have that from multiple reliable sources, including the Charles Piller podcast, The New Yorker, The Wall Street Journal, The New York Times, etc. We know there was short-selling; it's not a crime. Wikipedia's readers are not served by using a paywalled source (that they can't read) to state what is already available from high-quality reliable sources they can find online or in their local library. SandyGeorgia (Talk) 15:08, 19 August 2022 (UTC)

We raised it up to a discussion on the reliable sources board and www.complianceweek.com is deemed a reliable source and its content can be used and updated on the Cassava Sciences article. For reference here is the link:

Wikipedia:Reliable sources/Noticeboard#Compliance Week Mnachtrab (talk) 00:13, 2 September 2022 (UTC)

And finally, it appears (??) that posters here are wanting to use Compliance Week to infer that the short-sellers of Cassava stock were gaming the FDA. See WP:SYNTH and more importantly, WP:BLP; Wikipedia is not going to be adding any such content until/unless a high quality reliable source does precisely and explicitly that. Not having read the Compliance Weekly source, I nonetheless doubt they have done that either; I would think they wouldn't be eager to open themselves to a lawsuit by making such a claim, so I doubt that article gives us anything we don't already have. SandyGeorgia (Talk) 15:26, 19 August 2022 (UTC)

When noting that the FDA petition that was denied, the negative voices have been stating that it was denied only because it "fell out of the purview of the petition process." If you read the whole response letter, you will see that IF there had been evidence provided by the petitioners, the FDA would have intervened. But the petitioners did not provide evidence; instead they requested that the FDA investigate. Here is the important relevant paragraph in the FDA's letter: "FDA has carefully considered your Petitions and acknowledges the importance of the issues they raise. But as a threshold matter, by their own terms, your Petitions do not purport to set forth all relevant factual information. Rather, you call on FDA to initiate an investigation and factfinding process. We are denying your Petitions to the extent that they request, through the citizen petition process, that FDA initiate an investigation. Under § 10.30 (21 CFR 10.30), citizen petitions can request that FDA issue, amend, or revoke a regulation or an order, or take or refrain from taking an administrative action,2 and are to be resolved based on information in the administrative record.3 An investigation is not an administrative action, and, as your Petitions implicitly acknowledge, investigations necessarily require fact finding beyond what is presented in the current administrative record.

So, while initiating an investigation is outside the purview of the petition process, a simpler view of the FDA's denial of the petition is that the petitioners did not provide evidence that suggested the clinical trials should be halted. Again, this "out of the purview of the FDA" implies that there was indeed wrong-doing but this was the wrong way to go after it. Twisting words to fit the negative narrative, once again. SighSci (talk) 18:21, 21 August 2022 (UTC)

I wrote that that sentence (when it was in different section I think). And I did read the entire FDA letter before writing it. Which is how I know in the last paragraph the FDA petition response says: "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." ZICgene (talk) 20:07, 21 August 2022 (UTC)

SighSci, I am doing my best, and devoting long hours, to getting the mess that was on this page cleaned up, but your failure to get the point about how how to use reliable sources is exhausting my patience and risks becoming a tenditious and abusive use of this talk page should it continue. PLEASE read those pages, and all pages I have asked you to read, and refrain from repeatedly asking that Wikipedia editors engage in original research. I still have much work to do to bring this article to a good state, and this constant diversion into original research is indicating to me that my time might be better spent elsewhere.

When noting that the FDA petition that was denied, the negative voices have been stating that it was denied only because it "fell out of the purview of the petition process." If you read the whole response letter, you will see that ... I have asked you several times to understand that we must follow reliable sources; you continue to characterize those sources pejoratively. That is a waste of time and abuse of this talk page. The FDA petition is a primary source; it is not my job to characterize it or to use it beyond how the secondary sources do, and I am unlikely to do that.

If my time is to be misspent here, perhaps you'd prefer a return to this lovely mess? Please confine your posts on talk to suggestions for improvement based on the appropriate use of sources, and stop filling the page with opinion and suggestions that we use sources inappropriately. I will return tomorrow to see if my work on this article will be a productive use of my time. SandyGeorgia (Talk) 20:33, 21 August 2022 (UTC)

Got it, and thank you for all your efforts. To keep things limited and with appropriate citations, I suggest:

1. Please use add to the sentence describing Remoxy: "Remoxy was twice-daily oxycodone intended to be abuse-deterrent." Please cite these publications: https://pubmed.ncbi.nlm.nih.gov/30629280/ https://pubmed.ncbi.nlm.nih.gov/30629280/. This one could be added to the next sentence about the 2006 Remoxy clinical trial for analgesia being successful: https://academic.oup.com/painmedicine/article/12/4/618/1869381.

2. PTI-901 is not "modified" naltrexone, but is low-dose naltrexone. It is currently being used for a variety of autoimmune and inflammatory disorders, such as Chron's, a type of IBS.

3. The only remaining thing that is really glaring is that the "Journals Investigate" section does not currently mention that the third simufilam publication was also cleared of allegations of research misconduct. The journal did not issue an editorial note on this (instead saying "we take no action"), so the only citation currently would be the company press release (https://www.cassavasciences.com/news-releases/news-release-details/no-evidence-data-manipulation-science-publication-simufilam) unless you want to use MarketWatch: https://www.marketwatch.com/story/cassava-sciences-academic-journal-found-no-evidence-of-data-manipulation-271660829914. Again, 3 of the 5 PLOS retractions are irrelevant (non-Cassava, non-FLNA, non-AD and non-simufilam) and the other 2 are more related to PTI-901 or Oxytrex than to simufilam (FLNA, non-AD, non-simufilam). Finally, please consider adding the clearance by the Neuroscience publication that has a Cassava author. FYI, there are at least seven other (non-Cassava) Wang papers that have also been cleared. I am done for now, and thank you once again. SighSci (talk) 14:49, 22 August 2022 (UTC)

Better. I am iPad editing on the road from iPhone hotspot; will try to start catching up tonight when home. SandyGeorgia (Talk) 15:26, 22 August 2022 (UTC)

One final point (I promise): in the Citizen's petition section, the FDA quote, out of context, does not define "requests." It was the multiple requests to conduct an investigation that were not the appropriate subject of a petition (they in fact filed several petitions all requesting the same thing that were addressed together in this single FDA response letter; hence the plural requests). The request to halt a Phase 3 program is not outside the purview of the FDA -- they do this all the time, usually for safety reasons. They did not act here because they were not presented with evidence but were instead asked to find it. This quote from the letter is clear: "We are denying your Petitions to the extent that they request, through the citizen petition process, that FDA initiate an investigation." SighSci (talk) 21:20, 22 August 2022 (UTC)

CW articles

Continued at RSN. SandyGeorgia (Talk) 14:30, 1 September 2022 (UTC)

Requested at RX. SandyGeorgia (Talk) 04:42, 2 September 2022 (UTC)

1. Jaeger, Jaclyn (November 23, 2021). "Cassava Sciences facing widening probe over short seller claims". Compliance Week. Archived from the original on November 24, 2011. Retrieved September 2, 2022. {{cite web}}: |archive-date= / |archive-url= timestamp mismatch; November 25, 2021 suggested (help)

   Nothing here we don't already know. SandyGeorgia (Talk) 05:11, 2 September 2022 (UTC)

   Agreed. Firefangledfeathers (talk / contribs) 20:13, 2 September 2022 (UTC)

2. Jaeger, Jaclyn (January 11, 2022). "DOJ probe into shadowy underworld of short selling 'long overdue'". Compliance Week. Archived from the original on March 11, 2022. Retrieved September 2, 2022.

   I see nothing more than quotes from Barbier about how he responded; nothing usable as far as I can tell. The business about press release ... after ... appears dated, as revealed in The New Yorker that the short sellers had already revealed themselves to The Wall Street Journal, so don't see how this is useful. SandyGeorgia (Talk) 05:15, 2 September 2022 (UTC)

   Agreed. Firefangledfeathers (talk / contribs) 20:13, 2 September 2022 (UTC)

3. Jaeger, Jaclyn (March 2, 2022). "The Cassava Sciences saga: Short sellers, 'gaming' the FDA, and the damaging ripple effects". Compliance Week. Archived from the original on August 1, 2022. Retrieved September 2, 2022.

Firefangledfeathers here are the articles-- I see they are all older than what we already have-- settling in to read them now. SandyGeorgia (Talk) 05:08, 2 September 2022 (UTC)

Points from number 3 above for discussion

• In 2019, it moved ahead with Phase 2 clinical studies in patients with Alzheimer’s disease. According to Cassava, the Phase 2 study results showed simufilam not only stabilized but improved cognition in Alzheimer’s patients and had not presented any serious side effects. The author left out a big step re initial results, and later reanalysis by an "outside lab", as covered by our other high quality reliable sources. That's an important, and surprising, omission, considering the WSJ had covered that aspect already in November 2021.
• Mentions that Thomas did not reveal the short sellers to the FDA in his certification of the first Citizen's petition; something we also already know and have in other sources and have already stated. If that needs to be spelled out in greater detail, I think we can use the non-paywalled sources to do that. Not a lawyer, but I suspect that is a Jordan A. Thomas issue, not a Bredt/Pitt shortseller one, as Thomas was the attorney specializing in this area, signing the certification, and the specialist who is paid by his clients to know what he's doing. More feedback needed. This info might be more useful at Jordan A. Thomas. Not a lawyer, but I suspect a lot of this goes there, not here, and if we use it here, needs careful wordsmithing.

  Started, SandyGeorgia (Talk) 06:20, 2 September 2022 (UTC)

• Implies an additional potential financial motive for Bredt/MPM Capital competitor. We could probably work that in, but it's only insinuated, never explored, in the article, so needs careful handling/wordsmithing. Do any other sources corroborate that (considering the reporter's big miss on the "outside lab").

  I may be too tired to follow the plot, but MPM announces work on a competing drug (not unusual) in November ... the petition was filed in August (with work towards the petition starting well ahead of August obviously) ... how we phrase anything on this ... it's not passing my DUCK test, so I can't figure out how to word it. SandyGeorgia (Talk) 07:05, 2 September 2022 (UTC)

  One reliable piece: NYT correction of original article. SandyGeorgia (Talk) 12:50, 3 September 2022 (UTC)

  So we might describe Bredt as "who was at the time an executive partner at a firm that raised investment capital for another biotechnology company working on Alzheimer's treatment". Not sure if that is too much synthesis, as it relies on the MPM press release where he was hired in 2021, compared to the NYT announcement that he was no longer there. SandyGeorgia (Talk) 13:02, 3 September 2022 (UTC)

• Says Bredt/Pitt potentially "still made millions"; that is a better, more specific number to use than the $100 million we are now using in the article (which applies to *all* shortsellers of the stock, not just these two), so that looks useful. The way the article reads now, it leaves the impression that Bredt/Pitt could have made $100 million when, in fact, plenty of other investors would have routinely shorted that stock, and the 100 million doesn't apply to them-- we are leaving a faulty impression.

  Added, SandyGeorgia (Talk) 06:04, 2 September 2022 (UTC)

• Long general analysis of the CP process, and then: "The CP against Cassava highlighted how much of a mockery the regulatory process can become when misused" which is followed by examples of useless commentary submitted by the general public. Useful perhaps in another article; don't see how to use that here.
• "Thomas, who has since departed Labaton Sucharow, declined requests for further comment." Not sure what to make of this. If Thomas goofed by (mis)certifying something in his filing, it seems that goes in to his article, not here?

  No longer at that firm, SandyGeorgia (Talk) 06:20, 2 September 2022 (UTC)

That's all I've got, SandyGeorgia (Talk) 05:55, 2 September 2022 (UTC)

Too tired to continue, more feedback needed re filling in the missing pieces, way past my bedtime here. SandyGeorgia (Talk) 06:24, 2 September 2022 (UTC)

The bits we're using source #3 for in the article right now look fine to me. I would advocate for removing attribution for the two lines on what the petition contained when Thomas certified it. Both are fairly straightforward, verifiable factual claims. The "made millions" quoted bit should stay attributed. Firefangledfeathers (talk / contribs) 14:06, 3 September 2022 (UTC)

WP:TICKER

My edit summary went missing somewhere, but that is why I did this. SmartSE (talk) 21:17, 3 September 2022 (UTC)

Smartse new territory for me (med editor), but WP:TICKER seems to say not to put it in the lead, rather to put it in the infobox. What did I miss? SandyGeorgia (Talk) 21:20, 3 September 2022 (UTC)

Never mind ... I now see it is in the infobox ... must be my infobox allergy kicking in! SandyGeorgia (Talk) 21:24, 3 September 2022 (UTC)

Thomas, Schrag and $18,000

@Smartse and Firefangledfeathers: Someone added Thomas's (the attorney) name to the $18,000 payment to Schrag (the image expert), to avoid the passive voice vagueness. (When clients pay an attorney and that attorney pays the expert, it's the same difference to say the clients paid the expert, but that's not my question.) We have to combine sources to connect Thomas's name to the payment, which leaves me uncomfortable vis-a-vis WP:SYNTH. Here is what we now have:

Thomas paid Schrag $18,000 for his hundreds of hours of analysis of the images.

Sources
References

1. The Science podcast says Schrag "was paid $18,000 by the attorney representing the short sellers to do this", but never names Thomas.
2. Piller's Science article (sidebar) says: "He was paid $18,000 by the attorney representing the short sellers to do this", but never names Thomas.
3. Keefe's The New Yorker article names Thomas as the short sellers attorney, but never mentions the $18,000 payment to Schrag.

Not liking this SYNTHiness; Piller names Schrag, an attorney and $18,000, but Keefe names the attorney. Should we go back to passive voice, or is this not synth? SandyGeorgia (Talk) 17:13, 2 September 2022 (UTC)

PS, that had me awake and fretting all night. Just sayin' :) SandyGeorgia (Talk) 17:14, 2 September 2022 (UTC)

And then we have https://www.science.org/content/blog-post/faked-beta-amyloid-data-what-does-it-mean which says:

But some of these later Lesné papers had been flagged on the PubPeer site for potential image doctoring, and the Science report linked to in the first paragraph of this blog details the efforts of neuroscientist Matthew Schrag at Vanderbilt to track these things down. He was originally hired by two other neuroscientists who also sell biopharma stocks short - my kind of people, to be honest - to investigate published research related to Cassava Sciences and their drug Simufilam, and that work led him deeper into the amyloid literature.

So, the clients hire the attorney who pays the expert ... how to work all of this, better to use passive voice and leave out names or what ? SandyGeorgia (Talk) 17:22, 2 September 2022 (UTC)

I think the old language was "he received $18,000 from the attorney representing the short sellers who benefit if Cassava Sciences' stock falls" Since a living person is involved, I like that option over something that leans on SYNTH, even probably accurate SYNTH. I'd shorten to "he received $18,000 from the short sellers' attorney.". Firefangledfeathers (talk / contribs) 17:48, 2 September 2022 (UTC)

I, too, am more comfortable with your version, but will wait a bit to allow for further response (Smartse? Others?) Not trying to hide the name, rather avoid SYNTH. SandyGeorgia (Talk) 18:10, 2 September 2022 (UTC)

PS, Firefangledfeathers ... Not hyperbole that I was awake all night, fretting over this, so that I was very tired and never finished up at #CW articles, where I could really use another set of eyes. <hint ... not to distract you from the serious work at Joan of Arc, but hopefully you've now read those CW articles and can offer another opinion re how much and whether to work in some aspects ... > SandyGeorgia (Talk) 18:21, 2 September 2022 (UTC)

Given that I edited it, I obviously didn't think that it violated WP:SYNTH but no worries if you disagree. What I still find slightly clumsy/odd though about the current text, is that sometimes we call them whistleblowers and sometimes we call them short sellers which are both loaded terms. I know that both are correct, but I wonder if there is a way that we can discuss this earlier in the section and then stick with one word consistently - petitioners maybe? SmartSE (talk) 19:26, 3 September 2022 (UTC)

Excellent idea. Firefangledfeathers might be wanting to finish up, and since my editing sessions tend to involve a million little edits over many hours, I'll stay out until I'm sure FFF is finished. Perhaps FFF or you will initiate that, and accomplish it in one edit (to my typical twenty). SandyGeorgia (Talk) 19:29, 3 September 2022 (UTC)

Although there could be (?) some ambiguity in some places, depending on context, since Thomas was a petitioner, but not a shortseller or whistleblower ?? So we'd have to state that he filed on behalf of his clients, the petitioners? SandyGeorgia (Talk) 19:31, 3 September 2022 (UTC)

[4] Done], (although whistleblower comes up repetitively because Schrag was one in addition to Bredt and Pitt. SandyGeorgia (Talk) 20:36, 3 September 2022 (UTC)

Looks good to me. I'm finished with the flurry of edits from earlier, FYI. Please ping me if someone wants to add any BioSpace articles as sources here. Firefangledfeathers (talk / contribs) 01:43, 4 September 2022 (UTC)

Being patient August 6 21 article

Being patient focuses on Alzheimer's news advice and stories and is a secondary source that covers the open label data released later at the end of july 2021.

https://www.beingpatient.com/simufilam-dementia-alzheimers/ Mnachtrab (talk) 10:57, 5 September 2022 (UTC)

Please review churnalism; this has been explained several times on this page. SandyGeorgia (Talk) 15:50, 5 September 2022 (UTC)

No products approved by the FDA

... is mentioned in the lead, but is uncited. I know I've seen that in a reliable source; if anyone knows where to find that, it will save me the time of looking back through them. Else it will have to wait for me to find it again as I re-read all sources. SandyGeorgia (Talk) 17:05, 19 August 2022 (UTC)

I think this is not news, but I think the WSJ (extremely biased piece, who sought only sources who were connected to the petition, no one neutral) mentions this because the reporter wanted a negative piece. Again, most biotechs have no products approved. Once that happens they become pharma companies or are bought by them. 2600:1700:BB80:88A0:C999:C3CC:65DA:1E3D (talk) 22:33, 20 August 2022 (UTC)

In the products section, the Stat citation voicing an opinion that "enrollment has slowed" based on a company-provided number in the early days of the trial should be removed, or at least please add that the company reported over 400 subjects enrolled in early August, which would have to be a press release: https://www.cassavasciences.com/news-releases/news-release-details/cassava-sciences-reports-second-quarter-financial-results-2022. No news picked that up because it is not negative, and you won't find anything positive written by this Stat reporter, who according to the company, called them recently and stated, "Clearly you invented the data" about the early August clinical data release -- because it is good. If there had been something negative he could spin about it, he would have. This guy also tried to trash Moderna -- look it up. SighSci (talk) 01:41, 21 August 2022 (UTC)

Wikipedia content is based on reliable secondary sources; both WSJ and STAT are high quality reliable sources. We can raise questions about topics like this at the COI noticeboard, where surely other editors will have other opinions and plenty of my work will be revised, but we will get better feedback if the article is at least readable (what a mess I encountered a week or so ago, but we are close to having at least a readable article).

@ZICgene, Alexbrn, and ScienceFlyer: shall these two bits be deleted, or should we ask for more feedback at COIN, or do we need an RFC?

• Enrollment had slowed in the Products section, and
• The company has had no approved drugs in the Financial section

SandyGeorgia (Talk) 03:46, 21 August 2022 (UTC)

I cannot see any reason not to follow the secondary sources. One has to be a bit careful with WSJ and science, but in this case the knowledge is not contentious. If secondary sources discuss a topic in a "negative" way that Wikipedia editors don't like, that's not a problem Wikipedia can fix. We are bound to reflect decent sources, not try to arrive at some pre-determined POV by being selective about them. Alexbrn (talk) 07:37, 21 August 2022 (UTC)

I'm not sure why you would not cite the scientific publications demonstrating abuse deterrence of Remoxy (https://pubmed.ncbi.nlm.nih.gov/30629279/) and (https://www.wmpllc.org/ojs/index.php/jom/article/view/2275), or results of the clinical trial for analgesia (https://pubmed.ncbi.nlm.nih.gov/21823550/), and instead cite the NYT article, which is over-relied upon on this page as a source. SighSci (talk) 16:48, 5 September 2022 (UTC)

Because we can't use primary sources to contradict secondary information. The FDA did not approve the drug because (it seems) they disagreed. The NYT is a high-quality source; it is not "over-relied upon" here (it would be if other high-quality sources were saying something different, which they aren't). SandyGeorgia (Talk) 16:53, 5 September 2022 (UTC)

Interesting. I would think that like the game of telephone, the further you get away from the primary source, the less accurate. Plenty of investigative journalists wanted to know who was bribing whom at FDA, but no one could find anything. Shockingly, the author of the New Yorker article also wrote the book Empire of Pain, yet used the Remoxy chapter to further slam Cassava. Will patiently await new news and less biased secondary sources. Thanks again. SighSci (talk) 17:44, 5 September 2022 (UTC)

Re: who was bribing whom at FDA, but no one could find anything. Sigh, sigh, SighSci -- just when you were doing so well at avoiding WP:NOTAFORUM on talk. I would not like to see you disinvited from participating on this talk page, as your help is sometimes valuable. Please Stop Doing This. SandyGeorgia (Talk) 18:05, 5 September 2022 (UTC)

Why do you think Sigh is in my name? Journal articles are peer-reviewed. Not so for these secondary sources. SighSci (talk) 23:12, 5 September 2022 (UTC)

Your username is SighSci ... the extra two sighs are me ... sighing. Re journal peer review, we've seen the issues, and those perhaps help explain one of the many reasons that Wikipedia prefers secondary sources. SandyGeorgia (Talk) 00:10, 6 September 2022 (UTC)

Hatting this as WP:TALKOFFTOPIC.— Shibbolethink (♔ ♕) 17:31, 21 August 2022 (UTC)

I cannot see any reason not to follow the secondary sources. I am aware of the guideline whereby we are requested to assume good faith, so there will be no need to template me about it, however I am going to have to say that I highly doubt that. You inserted material from STAT to the Bret Weinstein article; the STAT author misrepresented what the secondary source reported, and what Weinstein actually said. It was only after much hounding on the Weinstein talk page, and your ability to FINALLY come up with a reliable source for the encounter, that just yesterday you FINALLY removed the incorrect information from the Weinstein article...information YOU ADDED, based on a shitty source, in this case, STAT. I consider such editing every bit as damaging, if not more, than the garden variety vandals I make a specialty of fighting. I am of the opinion the community needs to have a discussion about such behavior, and soon, because it is damaging the reputation of the encyclopedia. Le Marteau (talk) 13:43, 21 August 2022 (UTC) WP:FOLLOWING me around with silly accusations and disrupting this talk page certainly is a problem, yes. Alexbrn (talk) 13:47, 21 August 2022 (UTC) I am not "following" you. My intent is not to inhibit your enjoyment of the encyclopedia. My intent is the protection of the encyclopedia. In my work as such, when I encounter a problem editor, I routinely look at the other edits they have done. Often, they will have made dozens of other problematic edits requiring reversion. Them claiming I am "following" them is a common refrain. I am not going to review your previous work. I will, however, keep you on my radar. And if you want to consider that 'following' I am telling you you are wrong, but that is your prerogative. Le Marteau (talk) 14:56, 21 August 2022 (UTC)

2 Neurology Live Articles written about trial results

Phase 2b https://www.neurologylive.com/view/sumifilam-improves-multiple-alzheimer-disease-biomarkers-in-phase-2b-study

Open label 6 month update https://www.neurologylive.com/view/simufilam-improves-cognition-behavior-alzheimer-disease

These are secondary sources talking about the early trial results. Mnachtrab (talk) 10:53, 5 September 2022 (UTC)

Please review churnalism; this has been explained multiple times on this page. SandyGeorgia (Talk) 15:52, 5 September 2022 (UTC)

Firefangledfeathers and Smartse do you see anything usable from these neurologylive.com sources? It sometimes reports rather than just churning press releases, but these seem to be just churning press releases, and don't seem to me to say anything new. Please check me on that. SandyGeorgia (Talk) 16:08, 5 September 2022 (UTC)

I also thought that they looked like churnalism. More to the point though Mnachtrab needs to suggest content to be added, not just link to sources and leave the hard work to us. Additionally, if the sources are about trial results then WP:MEDRS applies and these are definitely not going to be suitable. SmartSE (talk) 16:17, 5 September 2022 (UTC)

I'd prefer to hold off on reading and evaluating until a concrete proposal is on the table. Firefangledfeathers (talk / contribs) 04:07, 6 September 2022 (UTC)

The other point that you continue to miss, Mnachtrab, is that simufilam exists; we don't need to explore drug background in all its gory detail in this article; I can't see anything in those articles that warrants inclusion here, and it's dubious as to whether this info is even useful at simufilam. SandyGeorgia (Talk) 16:19, 5 September 2022 (UTC)

Additional points from Compliance week (CW) article

The editors of this wiki did a good job of weaving the Compliance weeks assertion that the CP was submitted and did not include the information that the petitioners were short the stock. I’d like the editors to consider adding these two points from the article into the wiki page:

1. A major additional research point of the article as indicated by the title is a combination of a few points that link to the 'gaming' the FDA with this CP. As stated in the article the word 'gaming' was coined by the FDA Acting Commissioner Norman Sharpless and reaffirmed by the FTC citing the desire to work with the FDA on CP abuse and other issues harming competition (CW also linked the petitioner as a competitor.

The cassava wiki article points out many articles and activity that originated from the accusations in the CP, but CW article makes the case that maybe the real motivation of the CP was from a competitor, David Bredt, trying to slow down Cassava by gaming the CP process to also profit from his short seller position.

2. CW also did a significant amount of work on reading all 218 replies and noted 211 of them were from “relatives either currently have, or have died from,Alzheimer’s and pleaded to the FDA in support of the company” I think it's important to show that people who could benefit from the treatment wanted the phase three trial to continue. Mnachtrab (talk) 12:06, 5 September 2022 (UTC)

That kind of information would be more relevant in another article, eg citizen petition or FDA. The CW article does not "make the case"; it alludes to a suggestion (which is an interesting way to report), and then presents issues unrelated to the petitioners, eg, advocates who wrote to the FDA. (Because advocates were successful at aducanumab, in spite of unanimous rejection by the advisory panel, three of the FDA advisory panel members resigned, so it's certainly a problem that the FDA can be swayed. The FDA was not swayed in this case.) We present the sourceable facts and let the reader decide; you appear to want to make an explicit statement that the petitioners were "gaming the system" in spite of three high-quality sources that lay out all of their concerns. To use the aducanumab example, every major outlet in the media reported on that scandal within the FDA and how they succumbed to Alzheimer's advocates; in this case, you want to imply that the petitioners were "gaming the system" based on minimal allusions in one article. See WP:BLP; your point has been registered, and dismissed per important (BLP) policy ... more than once ... and if you continue without WP:DUE sources, your behavior on this talk may appear disruptive and I will seek to have you removed from talk. Please see #"Chemical signals" quote. SandyGeorgia (Talk) 15:48, 5 September 2022 (UTC)

For #1, I think the "gaming" association is too loose to use. For #2, I'm not seeing that 211 analysis at all; if I did, I think it might be worth mentioning briefly at simufilam. Firefangledfeathers (talk / contribs) 04:07, 6 September 2022 (UTC)

Cassava Partnership with Evonik 3/9/21

Cassava made news last year when they partnered with Evonik to produce simufilam. It should be added someone. I will leave it up to you guy to figure out how and where it makes sense, but it could make sense prior to discussion the phase 3 program.

Here is a secondary source that talks about the agreement: https://www.contractpharma.com/contents/view_breaking-news/2021-03-09/cassava-sciences-enters-drug-supply-agreement-with-evonik/

There was news last year that Cassava Sciences Inc., a clinical-stage biotechnology company developing product candidates for Alzheimer’s disease, has entered into a drug supply agreement with Evonik Industries AG for simufilam. Under the agreement, Evonik will supply Cassava Sciences with large-scale, clinical-grade quantities of simufilam, a drug candidate for the treatment of Alzheimer’s disease.

There is a quote from Evonik in the article:

“We are delighted to be collaborating with Cassava and contributing to fight Alzheimer’s together. We are committed to supporting Cassava in their goals to maintain the quality of life for millions of patients around the world and to further advance potential treatment options,” commented Dr. Thomas Riermeier, head of Evonik’s Health Care business line. Mnachtrab (talk) 10:48, 5 September 2022 (UTC)

Please review churnalism; this has been pointed out multiple times on this page. We have at least four high quality sources in use in the article. Evonik has never been mentioned in any of them. I recommend that you spend some digesting WP:DUE and WP:IDHT, and do not persist in filling this page with sources that aren't usable. SandyGeorgia (Talk) 15:55, 5 September 2022 (UTC)

Understood. Good point. I am understanding what you mean now on churnalism. I will spend some time reading those links before suggesting more changes. Mnachtrab (talk) 10:46, 6 September 2022 (UTC)

Correction made to Science.org article in area quoted on Cassava Wiki Article

Pillar made a correction on August 6th that requires a change in the Article from suspected issues to journal clearance:

Clarifications, 9 August, 5:45 p.m.: The sidebar has been updated to more accurately describe the nature of the signals said to be transmitted by dead brain tissue in Cassava-related research. In addition, the story originally said the Journal of Clinical Investigation declined to consider Schrag’s findings; the wording now reflects a clearer chronology provided by the journal.

In the discussion of the Journal of Clinical Investigation the wiki article quotes Pillar "the journal had reviewed high-resolution versions of the [15] images when they were originally submitted and declined to consider [his] findings"

The article now says "Schrag sent the journal’s editor his analysis of more than 15 suspect images in two groups. The editor says the journal analyzed high-resolution versions of the images in the first group. It could not corroborate his findings and therefore did not investigate further."

Not only did they consider his findings, they reviewed the first group, could not corroborate his findings, and chose not to investigate further. This is actually a clearance of the manipulation claims by Schrag. Mnachtrab (talk) 10:33, 7 September 2022 (UTC)

Re the first, did you see #"Chemical signals" quote? Looking at the second now. SandyGeorgia (Talk) 13:26, 7 September 2022 (UTC)

Done; thank you for the tip (which I missed above). Re "This is actually a clearance of the manipulation claims by Schrag", no it's not. That's not only wishful thinking; it's an inaccurate reflection of the source. SandyGeorgia (Talk) 13:40, 7 September 2022 (UTC)

Table of primary sources

@ZICgene and Alexbrn: are either of you able to help fill in Talk:Cassava Sciences/Primary sources ? It should give us a way to sort through the mess at all three articles (here as well as simufilam as well as Lindsay Burns, which is also chock full of primary sources). If so, for the love of all things good and holy and sanity-producing, PLEASE use:

• {{PMID}} to reference articles, erratum, and expression of concern when available
• ISO dates (yyyy-mm-dd) for table sortability
• vancouver-style author format on author names, which you can get by plugging in the PMID to this citation template filler.

If I had gotten on this article before it was clobbered by POV/COI editing, we would have gotten all of this right from the beginning; sorting it out now is one big headache. I'd appreciate any help you can lend as I sort through the secondary sources again. A sample is given: if you use the last row of the table, please be sure to add a new blank row so others can see which column is which via the inline comments. SandyGeorgia (Talk) 18:43, 19 August 2022 (UTC)

I have started working on it, I'll see how far I get tonight. ZICgene (talk) 23:43, 19 August 2022 (UTC)

Awesome, thanks! SandyGeorgia (Talk) 03:36, 20 August 2022 (UTC)

I've been gradually cleaning up the rest of the article (much more to do), but have not yet touched the Journal investigations section, hoping we can get the different papers sorted first. SandyGeorgia (Talk) 21:38, 20 August 2022 (UTC)

I have added the journal articles that are currently mentioned in the Journal investigations to the table, I was not sure whether to go back and add any others that are mentioned in the talk page (or were previously in the section). My internet has been particularly shaky the last few days, so it is slow going ZICgene (talk) 01:36, 21 August 2022 (UTC)

Thanks so much for doing all that tedious work, ZICgene. There may be some leads in the new, first line I added here. I can only get so far on this article before I run out of steam having to sit at a desk to sort all of this. Once we get a handle on what's what, we will be better able to sort what belongs here, vs. at simufilam, vs. at Lindsay Burns. I should be able to dig in again on this part tomorrow. At least I've gotten a (small) handle on the rest of the sections and secondary sources. Much appreciated, SandyGeorgia (Talk) 03:29, 21 August 2022 (UTC)

Please note that the sentence that includes "similarities in background pixels" in the journals section is completely uncited. This is not a quote from any of the sources so should be omitted. SighSci (talk) 21:26, 10 September 2022 (UTC)

Looking now: here's where it came from (in the "just saying" dept). SandyGeorgia (Talk) 23:01, 10 September 2022 (UTC)

Better now? SandyGeorgia (Talk) 23:15, 10 September 2022 (UTC)

"Chemical signals" quote

Please note that the sensationalized quote from the Science article about chopped up (it's not) dead (it's not) brain tissue has been modified by Science. The original version is now quoted, but it has been changed to say "chemical signals" instead of "nerve impulses" by Science. The whole article is very contentious as evidenced by letters to the editor on this article: https://www.science.org/doi/10.1126/science.ade1872 and https://www.science.org/doi/10.1126/science.ade2733 and by another Science reporter who comments on it a week later in Science: https://www.science.org/content/blog-post/faked-beta-amyloid-data-what-does-it-mean SighSci (talk) 15:36, 2 September 2022 (UTC)

SighSci thanks for continuing to help out. Because all of these sources are paywalled, I'm not entirely following where I can find the "new wording" you mention. Are you saying that the original Science article by Piller (July 21, 2022) has been corrected to new wording? If so, I can only access the archive.org version, and cannot see that version, which makes it hard for me to parse or verify your post. If there has in fact been a correction, could you please excerpt the exact full quote here for me, as well as its location? Will next try to locate archive.org versions of your other sources so I can read them. I appreciate your ongoing feedback, but pls work to make it as clear as possible, remembering the paywall effect, and that most Wikipedia editors aren't thoroughly versed in the intricate detail :) SandyGeorgia (Talk) 15:57, 2 September 2022 (UTC)

1. https://web.archive.org/web/20220826072914/https://www.science.org/doi/10.1126/science.ade1872 does not capture the full letter, so I can't read it, but I do see that it is co-written by Cummings, who is a Cassava advisor, and Selkoe, who published with Lesné.
2. I can see the second in its entirety, https://www.science.org/doi/10.1126/science.ade2733
3. The third, https://www.science.org/content/blog-post/faked-beta-amyloid-data-what-does-it-mean pertains almost entirely to Sylvain Lesné (where I hope we got the point all along, and never succumbed to the Daily Kos-style hype surrounding the AB56 findings or the impact on research overall), but about Cassava is not very flattering:

   Now Cassava is a story of their own, and I have frankly been steering clear of it, despite some requests. To me, it’s an excellent example of a biotech stock with a passionate (and often flat-out irrational) fan club. In such cases, if you say bad things about a beloved stock then plenty of helpful strangers will point out that you are an idiot, a shill, an evil agent of the moneyed interests, and much, much more. There’s a detailed sidebar in the Science article on Cassava and on simufilam, which I recommend to anyone who wants to catch up on that aspect. That compound is supposed to restore the function of the protein Filamin A, which is supposed to be beneficial in Alzheimer's, and my own opinion is that neither the published work on this compound nor the conduct of the company inspires my trust. There’s an ongoing investigation into the work at CUNY, and perhaps I’ll return to the subject once it concludes.

So I'm left not knowing how to address the original "chemical signals" matter that you raise. SandyGeorgia (Talk) 16:12, 2 September 2022 (UTC)

SighSci never mind, I've got it ... apparently (because it's a new month?), Science lets me see the full version. SandyGeorgia (Talk) 16:19, 2 September 2022 (UTC)

Better now? SandyGeorgia (Talk) 16:51, 2 September 2022 (UTC)

Sorry for the run-around. Just wanted you to know Science had relented to change just those words, despite it still being an inaccurate description of the method and the paper inaccurately called "Cassava related." These advisors have plenty of work of their own -- so should ALL of their work be Cassava related? What Piller was trying to say is, aha, this other paper also misled an entire field of research, just like Abeta*56, and it also misled simufilam (even though it was published in 2012, the same year as the first simufilam paper). The first letter to editor on the Piller article is split between the two links, oddly. SighSci (talk) 18:48, 2 September 2022 (UTC)

I read the "Cassava related" to mean work that formed the basis for the FLNA/simufilam/AZ connection, along with having been authored by Cassava advisors. At any rate, I toned down the "sensationalized quote". SandyGeorgia (Talk) 19:03, 2 September 2022 (UTC)

One would think that is the rational meaning of Cassava related. But Piller uses it otherwise to tag an entire third of Dr. Wang's publication list, and in particular, in discussing the brain diabetes paper that is not Cassava related and does not "support" the science behind simufilam. Thanks for your edits on this part. Much appreciated. SighSci (talk) 17:39, 5 September 2022 (UTC)

Edits needed to keep this page neutral

Inaccuracies were corrected and detail were previously added to keep this page balanced and neutral. They were immediately reversed and called "suspect" and "problematic." The original page had a negative bias and was part of a negative media and social media campaign on this company. Please keep this page fair and neutral. The "experts" quoted are all negative (and tied to the failed Citizen's Petitions) and so should be called "skeptics" not "experts," as just one example. 2600:1700:5431:1280:C5A:D4CF:987B:A482 (talk) 14:56, 30 July 2022 (UTC)

IP 2600, Please review WP:OR and WP:RS and WP:MEDRS and WP:COI and WP:NPOV for samples of some Wikipedia policies and guidelines. Wikipedia follows secondary reliable sources; it doesn't decide what to write as original research. In this edit, I see multiple problems:

• Use of a primary source for overlabeling results
• Original research and editorializing in the second bit, about benefits not outweighing risks (stick to what sources say).
• Considerable insertion of results from primary studies.
• Marginal sources; plenty of mainstream sources have covered this.
• Text not verified by sources.
• POV editorialzing "suddenly subjected to intense scrutiny by actors paid by the short sellers"
• Original research and editorializing (one sample only, it's everywhere): Only two of these PLOS papers are Cassava papers, and none discuss simufilam or even Alzheimer's disease. An additional retraction was a collaboration with Servier published in Alzheimer's Research & Therapy, for which Dr. Wang was and is still blind to group, making it impossible to manipulate data for a certain outcome.

I'm stopping there, as it's abundantly clear that most of the content reverted was poorly sourced editorializing and original research, suggestive that conflict-of-interest editing is occurring here. I am not reverting, as I'm uninterested in getting involved in a DOJ legal matter but suggest that the bulk of the article should simply be deleted. SandyGeorgia (Talk) 17:35, 30 July 2022 (UTC)

Biomarkers, outside lab

Not sure where to put this, but in the Research Candidates section, the second outside lab to analyze CSF samples for biomarkers did not report cognition data. Biomarkers are not cognition. Please change "improvements in cognition" to "improvements in CSF biomarkers." Also, this Alzheimer's section should be first or at least follow the FLNA hypothesis section. Being split by a discussion of decades old drug candidates does not make sense. PTI-901 is misspelled as PIT-901. SighSci (talk) 16:57, 5 September 2022 (UTC)

Acknowledged but iPad editing now ... will look later today when I am on "real" computer (unless someone else gets to it first ... have to study sources). SandyGeorgia (Talk) 17:46, 5 September 2022 (UTC)

PS, the order (flow) of the sections is because ... (Wang and Burns had earlier published together on FLNA's role in naloxone and opioid receptor signaling.) First explain what FLNA is, or the reader will be lost, then chrono order --> opiod receptor signaling --> Alzheimer's. SandyGeorgia (Talk) 18:07, 5 September 2022 (UTC)

From Keefe2022:

But Barbier said that there must be some mistake, and subsequently announced that the company had commissioned a reanalysis, by an outside lab, and that the data showed a significant improvement in biomarkers for Alzheimer’s compared with a placebo. The stock began to climb. In February, 2021, Cassava touted a round of promising results from another trial—without a placebo—and claimed that, in an unprecedented breakthrough, Simufilam could actually renew cognitive function.

From Mandavilli2022:

The most striking evidence of simufilam’s effectiveness was said to be an improvement in cognition in two-thirds of patients after taking the drug for a year. Some of the 50 patients had no neurological or psychiatric symptoms at all by that time, the company reported.

From Cassava press release: Clearly refers to biomarkers, but then adds:

In addition, Alzheimer’s patients treated with sumifilam showed directional improvements in tests of remembering new information, versus patients on placebo. Improvements in cognition correlated most strongly with ...

In conclusion, I am open to your feedback and suggestions for how to address this, but it appears that Cassava wants to have it both ways-- that is, imply that in addition to the biomarker data analysis, there was an improvement in cognition. And that is what the secondary sources report. It looks to me like we have it right; I'm listening. Please frame responses relative to what sources say; both the secondary sources and the press release seem to be clearly saying that Cassava claimed improved cognition. SandyGeorgia (Talk) 18:21, 5 September 2022 (UTC)

Better now? SandyGeorgia (Talk) 21:42, 5 September 2022 (UTC)

It is perfect the way you have stated the finding of improved cognition, but obviously incorrect in the secondary sources phrasing that Cassava claimed that simufilam could renew cognition. Stating clinical trial results (X happened in Y subjects in this study) and making a claim that the drug improves cognition are two very different things; if Cassava had made such a claim, FDA would be reprimanding. Never does Cassava make such claims that will only come with an FDA approval following Phase 3 trials. Also, it is very confusing to just say "reanalysis" for biomarker data. It was a reanalysis of CSF samples (backup samples, completely new assays), not reanalysis of existing data. To be super clear, my suggestion is to add "of CSF samples" after reanalysis. Thanks. SighSci (talk) 17:53, 12 September 2022 (UTC)

See your talk; I can't decipher what you want done. Cassava's press release does claim that "Alzheimer’s patients treated with sumifilam showed directional improvements in tests of remembering new information", as quoted above. SandyGeorgia (Talk) 20:47, 12 September 2022 (UTC)

Unwatch

Done, please don't ping me back to this article. COI noticeboard. SandyGeorgia (Talk) 19:13, 22 September 2022 (UTC)

NPOV Tag?

Hi everyone. I'm wondering if this page could use an NPOV tag? The page as written emphasizes potential fraud by the company, but doesn't give much space to the fact that some of the critics are short sellers making money off of the stock. I don't think we'll know the truth of it until more clinical results are published, but in the mean time this feels to me like it could use more balance? — Preceding unsigned comment added by Asbruckman (talk • contribs) 13:13, 21 November 2022 (UTC)

See WP:FALSEBALANCE - that's really not how Wikipedia is written. MrOllie (talk) 13:23, 21 November 2022 (UTC)

Thanks for your comment. It's not a "false balance" if the point of view is legit? As I said above, the only thing that will shine real light on this is future clinical results. We'll all have to wait I guess.... Asbruckman (talk) 13:42, 21 November 2022 (UTC)

Asbruckman have you declared a conflict of interest? SandyGeorgia (Talk) 02:30, 18 January 2023 (UTC)

Nature source

@MrOllie, Firefangledfeathers, and ScienceFlyer:, at long last, a reliable source that can be used for updating. Pinging you all as I have no interest in editing this article again, after the personal attacks I endured. SandyGeorgia (Talk) 14:51, 19 January 2023 (UTC)

And correction. SandyGeorgia (Talk) 14:53, 19 January 2023 (UTC)

Noes goes. 🤫 Firefangledfeathers (talk / contribs) 19:00, 19 January 2023 (UTC)

As fun as it was to be flippant, I can get to this by early next week if no one else does. Besides my run-of-the-mill editing, I have a couple of editing commitments ahead of this in the queue. Firefangledfeathers (talk / contribs) 23:58, 20 January 2023 (UTC)

I know ... several from me :) SandyGeorgia (Talk) 00:08, 21 January 2023 (UTC)

Bon courage might you do the honors? SandyGeorgia (Talk) 23:50, 20 January 2023 (UTC)

Stat news

I don't have full access, but Stat says no more effective than placebo.

• https://www.statnews.com/2023/01/24/new-results-show-cassavas-alzheimers-drug-has-placebo-like-efficacy/

SandyGeorgia (Talk) 22:28, 27 January 2023 (UTC)

I now have the full text. SandyGeorgia (Talk) 16:36, 28 January 2023 (UTC)

Lawsuit

I don't have access to the full article, so can't provide update.

• https://news.bloomberglaw.com/securities-law/cassava-loses-bid-to-escape-investor-suit-over-alzheimers-drug

SandyGeorgia (Talk) 18:50, 14 May 2023 (UTC)

Update (this lawsuit is not included in the article, while the Cassava lawsuit is):

• https://www.mondaq.com/unitedstates/securities/1344116/fifth-circuit-securities-litigation-quarterly-q2-2023

That is, bias. SandyGeorgia (Talk) 13:11, 20 July 2023 (UTC)

https://lawstreetmedia.com/news/health/cassava-sued-in-connection-with-simufilam/ SandyGeorgia (Talk) 13:14, 20 July 2023 (UTC)

CUNY investigation

Charles Piller, Science report here, and also covered by The Wall Street Journal. This needs to be worked in here and at Simufilam (and Lindsay Burns? Can't remember if she was co-author or these studies are mentioned in her article). I am on an eclipse-viewing roadtrip and not sure when I can get to this; @Firefangledfeathers, MrOllie, and Smartse:. Cassava's response and SandyGeorgia (Talk) 15:51, 14 October 2023 (UTC)

Yes, "Lindsay Burns, Cassava’s senior vice president for neuroscience and a co-author on several of the papers, bears primary or partial responsibility for some of the possible misconduct or scientific errors." SandyGeorgia (Talk) 15:54, 14 October 2023 (UTC)

The Saga of Cassava

Science blog, SandyGeorgia (Talk) 16:23, 19 October 2023 (UTC)

March 2024 status

https://www.science.org/content/article/damning-fda-inspection-report-undermines-positive-trial-results-possible-alzheimer-s

SandyGeorgia (Talk) 16:12, 12 March 2024 (UTC)