Flurizan has floundered in an 18-month Phase 3 clinical trial in patients with mild Alzheimer disease. According to a June 30 press release from Myriad Genetics, one of the sponsors of the drug, the study did not achieve statistical significance on either of its primary endpoints—cognition and activities of daily living. The results have prompted the company to discontinue the drug. There is not much more information available on the trial at present, but data are slated to be presented the afternoon of 29 July at the International Conference on Alzheimer’s Disease (ICAD) in Chicago.

Flurizan, the R-enantiomer of the non-steroidal anti-inflammatory, flurbiprofen, is an inhibitor of γ-secretase, the second of two enzymes that cleave amyloid-β (Aβ) from its precursor protein. The hope was that Flurizan would limit production of Aβ and slow or halt disease progression. That strategy may still be valid. “I would attribute the negative Flurizan results to a pharmacodynamic failure: insufficient brain levels to achieve meaningful reduction in Aβ generation,” suggested Paul Aisen, University of California, San Diego, and Director of the Alzheimer’s Disease Cooperative Study (ADCS), in an e-mail to Alzforum. “Though disappointing, the results do not refute the amyloid hypothesis,” he added. Aisen was one of the site investigators on the study when he was at Georgetown University, Washington, DC.—Tom Fagan.


  1. Flurizan Trial: Sic Transit Gloria Mundi
    The news that the Phase 3 trial of flurbiprofen (Flurizan) does not show any beneficial effects is a big disappointment. Coming on the heels of halted active immunization (AN1792), unsatisfactory Phase 2 results of passive vaccination (Bapineuzumab), and failure of a Phase 3 trial of Alzhemed (tramiprosate), should the failure of flurbiprofen prompt us to “re-evaluate” our thinking about the cause of the disease?

    Paul Aisen is justified in asserting that “…the results do not refute the amyloid hypothesis,” and a similar line of reasoning was put forth at each of the previous setbacks. The amyloid hypothesis, which has been extensively scrutinized and has produced the most visible therapeutic target, has morphed from “plaques are bad” to “Aβ42 is bad”’ to “soluble oligomers are bad” to “Aβ dimers are bad,” to name a few manifestations. Yet a dispassionate look at the literature tells us that something is not right with this picture. Up to 30-40 percent of non-demented subjects have ample plaques by PIB scanning, flurbiprofen, which reduces Aβ42 and increases Aβ38, is ineffective, tramiprosate is supposed to act by preventing oligomerization, and the Aβ vaccines have not delivered the expected results.

    There are still a number of “amyloid-directed” clinical trials in the pipeline, and we hope that some of them will turn into an effective therapy. The amyloid hypothesis has been the “engine” behind much of the basic and translational AD research, and these setbacks should not diminish its past contributions to the field. Nonetheless, there is no shame in taking a deep breath and opening up to the possibility that maybe we need to revise our strategy. Wishful thinking and contorting to hold on to a favorite—but perhaps wrong—idea is seldom helpful.

  2. The news about Flurizan is disappointing, especially after the promising Phase 2 report earlier this year in Lancet Neurology. The press release was too brief for any serious comment, but I hope that the analysis took into account the ApoE4 status of the patients, either by minimization at the start or by post-hoc analysis. No drug trial in AD should be done nowadays without taking into account the fact that the ApoE genotype strongly influences the rate of cognitive decline in AD, as shown by Martins et al. (2005).


    . APOE alleles predict the rate of cognitive decline in Alzheimer disease: a nonlinear model. Neurology. 2005 Dec 27;65(12):1888-93. PubMed.

  3. It is surprising and disappointing that the Phase 3 trials of Flurizan and Alzhemed have failed, given that both had very positive Phase 2 studies. It makes one wonder if the same intersite variability that doomed the Alzhemed trial also contributed the failure of the Flurizan trial.

  4. Comment by Rudy J. Castellani, George Perry, Xiongwei Zhu, Hyoung-gon Lee, Mark A. Smith

    1911 to 2008: The Dismal Progress of Knowledge in Alzheimer’s Disease Research
    In a study of senile plaques in brains of the demented and non-demented, Fuller in 1911 noted the presence of plaques in both groups, as well as in a young tabetic without cerebral cortical signs (1). He further summarizes the work of other investigators, in which senile plaques were found in brains of patients with senile dementia, as well as neurologically intact patients dying from other causes. In a review of the issue of clinical-pathological correlation by Huebner (cited in the Fuller paper), performed to help resolve medico-legal issues surrounding brain changes and competency, Huebner notes “the presence of (senile) plaques in the brain is not characteristic of any special (neurological condition)”; that the subject had at least reached the fifth decade is the most that could be medico-legally advocated. In a rather prescient remark, Fuller writes “while many interesting details concerning (senile) plaques of the brain have been established, the last word, perhaps, has not been said. As to the origin and nature of these (senile) plaques opinions differ, some even taking on the character of polemics.”

    Welcome to 1911.

    So here we are now in 2008, surprised to find evidence of amyloid-β with Pittsburgh Compound B in non-demented subjects, and surprised by the continued failure of anti-amyloid-based therapies.

    It bears remembering that the now 25-year-old Brownian-like motion that surrounds the amyloid cascade hypothesis is rooted in a causal association between plaques, or variations thereof (e.g., low n Aβ-mers), and human disease (2,3). Absent such an association (which has long been the case), the cascade represents effect rather than cause (4), and the cascade hypothesis an expensive and even harmful distraction. Therefore, while many interesting details of the amyloid cascade have been established, the last word, perhaps, has not been said. As to the origin and nature of neurodegeneration in AD, opinions differ, but the reticence to accept empirical truths (“Though disappointing, the results do not refute the amyloid hypothesis”) takes on the character of polemics.

    Welcome to 2008.


    . A study of the miliary plaques in the brains of the aged. American Journal of Insanity. 1911;LXVIII:147-219.

    . Neuropathology of Alzheimer disease: pathognomonic but not pathogenic. Acta Neuropathol. 2006 Jun;111(6):503-9. PubMed.

    . Alzheimer disease pathology as a host response. J Neuropathol Exp Neurol. 2008 Jun;67(6):523-31. PubMed.

    . Amyloid-beta and tau serve antioxidant functions in the aging and Alzheimer brain. Free Radic Biol Med. 2002 Nov 1;33(9):1194-9. PubMed.

  5. I think that a most plausible explanation for the failure of flurbiprofen in the clinical trial is that when the disease has been ongoing for years, as it obviously has in people recruited in clinical trials, it is too late to cause improvement by just inhibiting amyloid production, even if amyloid is the initial trigger of the disease.

  6. The failure of Flurizan should not be a surprise. NSAIDs were suspected of protective activity based on epidemiological data. Because of a correlation of "epidemiologically active" compounds with activity in cellular assays measuring Abeta 42/40 ratios, it was suggested that this was the causal mechanism. There has always been a discrepancy between the concentrations required for the biochemical mechanism, and the typical exposure of patients taking NSAIDs. This discrepancy was even directly affirmed in the phase I studies with Flurizan, which failed to present the forecasted pharmacodynamic effect on Abeta 42/40 ratios in the CNS. This discrepancy has simply come home to roost at an aggregated cost of about $200 million.

    The tragedy is that the principal idea of modulation of APP processing deserves to be tested, but, because of how the Flurizan rationale has been communicated, other tests, with perhaps more deserving compounds, may be obstructed.

  7. Outstanding information.

Make a Comment

To make a comment you must login or register.


Therapeutics Citations

  1. Flurizan™

External Citations

  1. press release

Further Reading

No Available Further Reading