Multiple Sclerosis Drug Trial Results Questioned

The likelihood that a participant in a randomized, controlled trial receives laquinimod or placebo to treat relapsing-remitting multiple sclerosis (RRMS) could explain the differences in study results, according to recent research.

Gary Cutter, PhD, of the Department of Biostatistics in the School of Public Health at the University of Alabama at Birmingham, and colleagues conducted the study and the results were published on BMC Neurology on September 17, 2016.

The present study used a propensity score, which is defined as “the probability of treatment assignment conditional on observed baseline characteristics”, to analyze the results of two randomized phase 3 trials. The two studies were "Assessment of Oral Laquinimod in Preventing Progression in Multiple Sclerosis "(ALLEGRO) and the "Benefit-Risk Assessment of Avonex and Laquinimod" (BRAVO).

“Although the study designs of ALLEGRO and BRAVO were similar and were conducted in roughly the same calendar years, the treatment effect of laquinimod on the annualized relapse rate (ARR) and disability progression was found, based upon a predefined analysis plan, to be more favorable in the former than in the latter trial,” the researchers said.

In ALLEGRO, laquinimod was associated with less risk of disability progression and fewer new or enlarging lesions. BRAVO showed only a non-significant reduction in ARR.

“The goal of using a propensity score was to obtain an estimate of the probability of being assigned to one or another of the treatment arms based on characteristics within the trial when the theoretical probability was known to be 0.50,” said the researchers.

There were two main stages to the analysis: first, the researchers calculated a propensity score for each patient, and second, they incorporated the propensity score into an predefined analysis model.

What they found, the authors say, is that “although randomization guarantees unbiased assignment, randomization does not guarantee quality.”

They go on to explain, “In this example, the original preplanned analysis was most likely affected by imbalances in important baseline characteristics, and the propensity-adjusted analyses confirm that the estimate of the treatment effect is quite similar.”

The researchers conclude by noting that propensity score analyses can be useful for identifying studies which are consistent with each other as well as those that are not.

“Specifically, in the MS treatment paradigm of reduction of relapse rates, propensity scores using baseline MRI characteristics can prove helpful to adjust for and better reflect observed differences in treatment effect in randomized controlled trial,” they concluded.