March 26, 2001
LUKE'S MEDICAL CENTER
RUSH MULTIPLE SCLEROSIS CENTER
1725 WEST HARRISON
SUITE 309, CHICAGO, ILLINOIS 60612-3824
The undersigned neurologists and I, who treat patients with multiple sclerosis (MS), are writing to you because we are concerned and distressed by a recently published article on the treatment of MS. The article by Dr. Omar Khan et al, “A Prospective, Open-Label Treatment Trial to Compare the Effect of IFNß-1a (Avonex), IFNß-1b (Betaseron), and Glatiramer Acetate (Copaxone) on the Relapse Rate in Relapsing-Remitting Multiple Sclerosis,” published in the March 2001 issue of the European Journal of Neurology, describes the results of a small, open-label, nonrandomized, unblinded observational trial. It contains, in our opinion, serious flaws in experimental design and methodology, and results such as these should be ignored by clinicians.
The methodological flaws of the Khan trial led to selection and subject biases. Patients were not randomized to treatment, consecutive patients were studied, and patients themselves were allowed to select one of three therapies or no therapy. Another serious design flaw is that both subjects and experimenters were not blinded to study drugs, which may have directly or indirectly influenced the outcomes, thus invalidating the results. Endpoints were not well defined, and data on these endpoints were not collected in a controlled manner. Relapses and disability were at times self-reported by patients and recorded without a visit with the investigating neurologist for verification.
The study was vastly underpowered, with only approximately 40 patients in each active treatment group. Our statistical analyses to calculate the required sample size for a study comparing three active treatments revealed that to detect a 50% difference in EDSS between treatments using Kaplan-Meier analysis, one would need 98 patients in each treatment group, more than twice that surveyed in the Khan trial. However, even this would be unrealistic because none of the surveyed drugs is 50% better than either of the other drugs. To detect a 30% difference in EDSS, 284 patients would be required in each treatment group, a sevenfold increment from Khan’s groups.
The Khan study is a small, open-label, nonrandomized, unblinded observational trial, making the validity of any conclusions regarding efficacy unjustified. Unfortunately, biased results such as these are misleading and confusing to patients and physicians and may adversely affect the way they view and use treatments for MS. We urge you to discuss the above issues with your MS patients and your colleagues whenever the topic arises and reassure your MS patients that the results of this trial should not have any influence on decisions regarding their treatment.
Dusan Stefoski, MD
Director, Rush Multiple Sclerosis Center,
For more information on the how to interpret clinical trials, visit Aboutscience.org .