Neurochem's novel oral beta-amyloid fibrillogenesis inhibitor Alzhemed (tramiprosate) did not perform statistically better than placebo at modifying the progression of Alzheimer's disease in a Phase III study of North American patients, the company announced Aug. 26.

"Due to significant interference from high between-site variations that complicated the statistical analyses beyond expectations, it is not possible to draw definitive conclusions with regard to the treatment effect of tramiprosate," the Laval, Quebec, company said.

The company met with FDA earlier in August to seek advice on statistical modeling and analyses that might overcome the confounding factors affecting the data. However, the agency "advised that neither the proposed adjusted models nor any further adjustments could be used for this trial to provide results in support of a claim of clinical efficacy," Neurochem said. The drug received fast track status in July ¹ Also see "Neurochem’s Alzhemed On Fast Track For Alzheimer’s Disease" - Pink Sheet, 24 July, 2007.).

Neurochem has commissioned a Special Advisory Board to analyze the data and assist the firm as it considers options for future development of the experimental Alzheimer's treatment. The panel's deliberations should be finished by the end of the year, CEO Francesco Bellini said during an Aug. 27 conference call.

One outcome of the board's analyses could be changes to the cohort size, duration, or analysis models in the ongoing European Phase III trial, which has completed enrollment of 966 patients but can still be modified, the principal investigator on the North American trial, Paul Aisen, Georgetown University, said during the call. "I think adjustments to the European trial are very much on the table," Aisen said.

The execs noted that one confounding factor that affected the trial is the number of concomitant medications taken by patients. All patients in the study were also taking medicines "that have significant effects on the primary outcome measures in these trials," Aisen said.

Another factor that is likely to affect any disease modification trial involving mild-to-moderately affected Alzheimer's patients, particularly those with mild disease, is that they decline slowly, and the decline is difficult to measure, Aisen said. He said that one thing the panel hopes to identify from the data is "what predicts decline in Alzheimer's disease in general. ...Without decline, you can't show slowing of decline and how that can be applied to the design of a clinical trial and the analysis of clinical trial data," he said.

Bellini said the final data from the study will likely be presented at the Conference on Alzheimer's Disease and Related Disorders July 15-20, 2008, in Madrid.

The CEO also said the company would be filing a response to a second "approvable" letter on its experimental Amyloid A amyloidosis therapy Kiacta (eprodisate) that includes follow-up data. Neurochem does not plan to conduct a second confirmatory trial on Kiacta, an option FDA presented in the letter, Bellini said ² Also see "Neurochem/J&J’s Kiacta Action Date Extended To July 16" - Pink Sheet, 11 April, 2007.). The company announced on July 17 that it had received the second letter.

Kiacta, for which Neurochem is partnered with Johnson & Johnson, has had the benefit of substantial regulatory nurturing. In April 2006, it was granted priority review. Before that it received Orphan Drug and Fast Track designation and was accepted for the Continuous Marketing Application Pilot 1 and 2 programs.

-Shirley Haley (s.haley@elsevier.com)

Poor “Site Effect” Profile? Neurochem Reports Inconclusive Alzheimer’s Data For Alzhemed

More from Archive

More from Pink Sheet