The phase II trial data for AMX0035, an investigational agent being studied to treat amyotrophic lateral sclerosis (ALS), may not be enough for the FDA to decide whether to approve it, the agency said Monday.
The study showed primary endpoint results that were "borderline statistically significant and may not be sufficiently persuasive to allow an effectiveness determination based on a single study," FDA reviewers wrote prepared ahead of Wednesday's .
"However, the FDA has also long stressed the appropriateness of exercising regulatory flexibility" when assessing drugs for serious diseases with unmet medical needs, the reviewers acknowledged.
On Wednesday, AMX0035, a proprietary combination of sodium phenylbutyrate and taurursodiol from Amylyx Pharmaceuticals, goes before the FDA's Peripheral and Central Nervous System Drugs Advisory Committee, which will vote on whether CENTAUR data can support the drug's approval. The vote is nonbinding and will serve as a recommendation to the FDA, which is scheduled to approve or reject AMX0035 on or by June 29.
In CENTAUR, AMX0035 appeared to slow functional decline in people with rapidly advancing ALS compared with placebo over 24 weeks, as measured by ALS Functional Rating Scale-Revised () scores. The trial evaluated 135 ALS patients, including people using two FDA-approved ALS medications, riluzole (Rilutek) or edaravone (Radicava), when they entered the study.
The mean rate of change in ALSFRS-R score was -1.24 points per month with AMX0035 and -1.66 points per month with placebo (difference 0.42 points per month, 95% CI 0.03-0.81; P=0.03) in a modified intention-to-treat analysis. The week-24 mean difference was 2.32 points (P=0.034). Scores on the ALSFRS-R range from 0 to 48, with higher scores indicating better function.
Secondary outcomes were not significantly different between the two groups, and adverse events with AMX0035 mostly were gastrointestinal.
In an that spanned up to 30 months, researchers reported that AMX0035 reduced the risk of death by 44% compared with placebo. Survival analysis of extension study participants suggested that patients randomized to AMX0035 at baseline had a 6.5-month longer median survival than those who started with placebo.
The FDA's concerns come down to uncertainty about relying on a lone study that uses ALSFRS-R to assess efficacy.
"FDA agrees that ALSFRS-R is an acceptable primary endpoint to measure functional change in ALS," said the FDA's Veneeta Tandon, PhD, in a prerecorded of the AMX0035 application.
"However, rate of decline is not generally the appropriate approach to analyzing treatment effect as it assumes that the change in ALSFRS-R is linear over time, which has not been established," Tandon observed. "Additionally, it does not account for loss of data due to death of patients during study. A joint rank analysis is recommended if there are deaths during the study."
"The key issues with this application are first, the single study with evidence in the primary analysis that is not persuasive, with P=0.034 and a corresponding week-24 difference of 2.32 points on the 48-point ALSFRS-R scale," noted FDA reviewer Tristan Massie, PhD, in a prerecorded of the data.
"Second, there are issues with study conduct and analysis assumptions, and many sensitivity analyses provide less persuasive results than primary analysis. In particular, there were issues with randomization implementation and imbalances in use of concomitant ALS medications riluzole and edaravone," Massie pointed out.
"Additionally, there are issues with the handling of deaths or lack thereof and missing data assumptions in the primary analysis," he continued. "Also, the primary analysis assumption of linearity over time in treatment effect is questionable based on the observed data and the prespecified analysis plan."
"Furthermore, secondary endpoint results are not compelling," Massie added. "Finally, with regards to the open-label extension, survival analyses for time to death alone are exploratory and not persuasive."
With no cure for ALS and only a few treatments for symptoms, the FDA is under pressure to approve new drugs for the fatal disease. FDA guidance indicates that one adequate, well-controlled clinical investigation plus confirmatory evidence can be adequate to establish effectiveness. The agency can consider a number of factors when determining whether a single study is appropriate, including the seriousness of the disease and whether there's an unmet medical need.
The phase III trial of AMX0035, which will evaluate approximately 600 people with ALS, is . Amylyx is also studying AMX0035 as an investigational treatment for .