Neurochem announced last week that it would halt development of tramiprosate (Alzhemed) in the wake of the failure of its large North American Phase 3 trial (see ARF news story and commentary). The Canadian company said it would immediately discontinue an ongoing European trial and put its resources into the development of a tramiprosate pro-drug that it claims achieves higher levels in the brain. The company will continue an open-label extension of the North American trial.

In a surprising twist to the story, the company said it plans to market tramiprosate, also known as 3-amino-1-propanesulfonic acid or homotaurine, as a nutritional supplement. CEO Francesco Bellini announced that Neurochem would form a separate company to sell this product. He said he expects this move to satisfy demand for the compound from patients and their caregivers, and to produce revenues to fund Neurochem’s continuing research and development activities.

The negative results of a multicenter North American trial came out earlier this year, and in August, Neurochem announced that the FDA would not accept alternative analyses of the data to compensate for problems with the trial. (Phase 2 and 3 trials between the FDA and sponsoring companies are confidential, and the FDA does not comment on them.) Barry Greenberg of Neurochem presented data from the North American trial last Friday, 9 November, at the Marian S. Ware Neurodegenerative Disease Research Retreat at the University of Pennsylvania in Philadelphia. The primary outcome measures of cognitive function showed no significant effect of treatment over placebo, but Greenberg reiterated that the ADAS-Cog and CDR-SB showed some numerical differences in favor of tramiprosate. He said that the company is getting hundreds of letters asking for the drug but cannot claim a treatment effect.

One measure did reach statistical significance in the post-hoc analysis, and that was hippocampal atrophy, Greenberg said. A subset of 312 patients out of the total 790 who completed the study had MRI assessments of hippocampal volume at the beginning and end of the study. Greenberg said that a preliminary analysis of the volume data by external statisticians found a dose-dependent reduction in hippocampal atrophy, after adjustment for confounding factors. Reportedly, a dose of 200 mg/day reduced atrophy by 65 percent, and the 300 mg/day dose halted atrophy over the 18 months.

“Neurochem is committed to analyzing and understanding the data generated in the North American trial,” Greenberg said. The company had initially hoped to use the lessons from the North American trial to improve the design of the European trial, but then decided to scrap it altogether. Instead, Neurochem will prioritize to accelerate development of NRM-8499, a pro-drug of tramiprosate. According to Bellini, that compound is in preclinical development, with results of toxicology tests in several animal species due in early 2008.

In his talk at Penn, Greenberg, as had Paul Aisen in his prior discussion of the trial’s woes (see ARF related news story), warned that problems encountered in the tramiprosate trial may stymie other researchers seeking to carry out large disease modification trials. Length, center variability, diagnostic uncertainly in mild AD, and lack of diagnostic biomarkers or surrogates for therapeutic effects all complicate the effort, Greenberg said. Added to that were limitations of psychometric measurements of cognition, most of which were validated for short-term effects. In addition, it is hard to quantify quality of life. “The clinical effect is ultimately the most important, but the tools we have are not matched to this. We need better tools with higher sensitivity,” Greenberg said. Borrowing a quote from Richard Mohs of Eli Lilly, a noted AD trials expert, Greenberg concluded by saying that trials evolve by dialog among all involved, and need to achieve both scientific rigor and clinical relevance. “It is time to reopen that dialog,” Greenberg concluded.

It should be added that, when put to leaders of AD programs in other pharmaceutical companies and academia, these scientists tended to reply that trials will succeed if the drug is effective. In addition, current trials of newer drugs are increasingly including biomarker measurements to gather the data needed for their validation. Off-the-record rumblings about a paucity of rigorous and relevant research data, either published or presented at conferences, had accompanied tramiprosate’s clinical development through the years.

Greenberg did not mention in his talk that Neurochem had announced the day before that they would pursue marketing tramiprosate as a nutritional supplement. When questioned about this, he declined comment, saying he was not involved in the decision or in any future work on those plans. In a webcast on 8 November, Bellini said the move was a reaction to the results of the trial, coupled to the “demands from physicians and family members for the drug.” Because it is a natural product, tramiprosate can be sold as a nutritional supplement without the need to establish safety or effectiveness, but it cannot carry therapeutic claims. Bellini claimed that tramiprosate could be on the market as early as 2008, but the planned subsidiary company has not been formed yet.

Could the marketing of a tramiprosate dietary supplement affect future trials of related compounds? This question will only be answered if the product hits the market and catches on with elderly people. Bellini placed the market for nutritional supplements geared toward cognition at a cool US$1 billion per year. No one knows what slice of this pie tramiprosate might eventually capture, and CEOs of struggling public companies have been known to make optimistic projections before. Neurochem’s share price tumbled from around $25 last December to about $2 following the FDA decision last August, and closed at $3.85 last Friday.—Pat McCaffrey and Gabrielle Strobel.


  1. Unacknowledged in the public dialog surrounding Alzhemed's failure is the virtual certainty that trials of other anti-amyloid strategies that remain hopeful will exclude Alzhemed users.

    The desperation of the disease can drive families and carers to make irrational choices. But, Alzhemed "nutraceutical" users, Caveat Emptor: You may unintentionally deny yourselves access to trials of immunotherapies and/or secretase modulators that, unlike Alzhemed/tramiprosate, continue to hold promise for slowing cognitive decline.

  2. The decision to market tramiprosate as a supplement is surprising news. It does not speak well of nutraceuticals as a refuge for drugs that fail in clinical trials. From my perspective, it would be okay to fail in a clinical trial if they had compelling biomarker data from humans showing that the drug can really move a target, and established the dosing required to do it. The hippocampal volume data looks promising, but needs its own supplement, perhaps with a CSF biomarker cocktail that includes tau given the report from Avila (see ARF related news story) that the compound can promote tau aggregation. One problem with people self-dosing a nutraceutical is the typical lack of information about the dose required to move an endpoint. And at least they have a dose response on the hippocampal volume, and apparently at least some patients clamoring for more because they thought that it helped them.

    The great value with nutraceuticals would be for prevention. There, the clinical endpoint is hard to come by, especially in diseases with long and very expensive trials. Also, insurance may not pay without a diagnosis. So I think you could argue that tramiprosate might work for prevention but not with established AD. In the absence of a convincing clinical trial, tramiprosate needs solid data to show that it moves a panel of biomarkers relevant to pathogenesis. I have been unable to understand how tramiprosate could get into the brain as a charged compound, and not understanding makes me suspicious.

    My Alzheimer Center colleagues pointed out that tramiprosate is a modification of taurine. Taurine is a major component of the "energy drink" Red Bull, which is also loaded with caffeine. At SfN, Gary Arendash talked at a news conference about how caffeine is treating his transgenics and is blocking both β- and γ-secretase. So it is beginning to sound like we'll be seeing something like a cross between Red Bull and Rock Star (another energy drink containing ginkgo extract and caffeine) for seniors, Perhaps we should have a naming contest? Combining the last too initials in the cross would tempting, but so far I like Senior Bull.

  3. Reader advisory: These comments may appeal only to readers of a certain age, who remember black-and-white TV and Richard Nixon.

    Sean Silcoff, a reporter for Neurochem’s hometown paper, The Toronto Financial Post, describes the Alzhemed story as playing out like the Monty Python dead parrot sketch, a discussion between a pet store owner and a customer over whether a dead parrot is dead or just resting or stunned.

    Neurochem had been going around and around about the vital status of its North American trial, and landed its most recent stunner on November 8, 2007, the essence of which is that since there is no evidence for tramiprosate’s efficacy by any conventional scientific standard, Neurochem will market homotaurine—the amino acid formerly known as the drug tramiprosate—as a food supplement or nutraceutical so that people with Alzheimer disease can buy it over-the-counter and not be deprived of the drug’s perceived potential benefits that the company touted in prior news releases and presentations.

    This may be great news to some, “but wait, there’s more!” as Ron Popeil, the consummate TV purveyor of Veg-O-Matics, might say. According to the Financial Post, Neurochem expects to “scare up US $1 billion-plus sales” selling this nutraceutical and will use the proceeds to develop an effective AD drug, presumably one minimally effective enough that the FDA could actually approve it. In the meantime, people with or without AD can soon take tramiprosate thinking that it will both help them directly and also help fund Neurochem to develop effective future drugs.

    One such Neurochem gleam-in-the-eye is the previously unheard-of NRM-8499, said to be a pro-drug of the very “food supplement” they will be selling—inevitably—on their TV infomercials. Neurochem states that NRM-8499 is five times more potent than their “Alzhemed” brand of homotaurine. (This of course begs the question why consumers shouldn’t just buy five times more of the nutraceutical than they would have bought of the prescription drug had it been demonstrated effective). The sales message—to be voiced probably by Hollywood and medical celebrities—could be, “Help Neurochem fight AD as you help yourself.” A Beverly Hills celebrity statistician could make an appearance here, too, because statistical modeling is so much a part of the Alzhemed story.

    While they are at it, Neurochem could go further and tithe a portion of their proceeds to the Alzheimer’s Association, gaining a hefty tax credit and stating on each bottle of the Alzhemed brand of homotaurine, “all proceeds to Alzheimer’s charity.”

    Even as the company announces that tramiprosate is not worth pursuing as an ethical pharmaceutical product, it continues to imply that there is efficacy to be found somewhere deep within the North American trial data. Seven months after they knew their results, they still speak of “promising results from preliminary post-hoc analysis,” “descriptive data” showing “numerical differences in favor of tramiprosate on the primary clinical endpoints,” “descriptive data also shows…differences between groups on the primary disease modification endpoint of change in the volume of the hippocampus,” or, to further obfuscate, “preliminary post-hoc analysis…that allowed adjustment for potential confounding factors showed a dose-dependent reduction in hippocampal atrophy in patients treated with tramiprosate.” Never mind inferential statistics; facts about efficacy matter less now than they had before for the marketers of a food supplement.

    And now for something completely different…(as I am trying to “always look on the bright side....”):
    Having acknowledged—not by its press spin but by its business actions—that the North American trial did not show efficacy and that the company could not withstand the risks of its European trial, Neurochem might now just put their bird to rest. It should post the North American trials database, statistical analyses, data files, and clinical study report on their website for others to review and analyze. They cannot get FDA approval and by law cannot make specific AD health claims for a food supplement. Any manufacturer can make homotaurine or conceivably extract it from one or another species of algae, so they no longer have a proprietary need to hide the data. Posting the data would provide the best and most transparent support for their press releases. Consumers will be able to examine directly the efficacy evidence and judge for themselves.

    Open access to the database and statistical models would allow clarification of the alleged methodological problems that Neurochem enumerates but doesn’t document (e.g., site variability, diagnostic uncertainty, confounding, and metrical issues). It may even allow other methodologists—using other Veg-O-Matics—to find things that may have eluded the Neurochem experts. Academics and advisors to companies and the NIH would learn more about AD clinical trials methods, avoid the repetition of mistakes that might be dooming otherwise effective drugs, and improve future trials methods. This, in turn, might help us to find effective drugs for AD sooner by improving the designs of trials and serve the common good. Of course, other companies and the NIA ADCS should post their clinical trials data, as well, and may do so if Neurochem sets the example. Finally, posting these data and reports would not undermine the publication of a primary paper in any major journal.

    I leave it to others to explain why such a modest proposal for companies to share their data from negative trials so that others can benefit would be dead on arrival.

    Disclosure: See 14 September 2007 comment.

Make a Comment

To make a comment you must login or register.


Therapeutics Citations

  1. Alzhemed™

News Citations

  1. FDA Deems U.S. Alzhemed Trial Results Inconclusive
  2. Washington: Site-to-Site Differences Delay Result of Alzhemed Trial

Further Reading