Background: Contrast enhancing lesions (CEL) is a common endpoint in multiple sclerosis (MS) clinical trials. To minimize sample size or placebo exposure, a crossover design without a concurrent control group is attractive. Natural regression may confound this strategy. We assessed the degree of regression in monthly new gadolinium activity in relapsing-remitting (RR) placebo patients.
Methods: A post hoc analysis was performed on 65 RRMS placebo patients in the Prevention of Relapses and disability by Interferon beta-1a Subcutaneously in Multiple Sclerosis (PRISMS) trial. Patients were originally selected for relapses but not preselected for MRI activity. Eleven MRI scans were taken at screening, baseline, and months 1 through 9. Monthly new CEL rates were examined using a random effects Poisson model. Patients were analyzed as a single group and by screening CEL count level subgroups: no, low, and high (0, 1 to 3, >3 CEL).
Results: A total of 32, 19, and 14 patients had no, low, and high CEL counts at screening. The monthly new CEL rates (95% CI) of all patients at baseline, months 1 to 3, 4 to 6, and 7 to 9 were 2.0 (1.3, 2.9), 1.8 (1.3, 2.5), 1.4 (1.0, 2.0), and 1.2 (0.8, 1.7). Compared to baseline, the rate decreased by 10%, 27%, and 39%. The monthly rate of the no subgroup remained stable. The rates for both the low and high subgroups decreased by 4%, 29%, and 48% at months 1 to 3, 4 to 6, and 7 to 9 compared to baseline.
Conclusions: Placebo relapsing-remitting multiple sclerosis patients experience a decline of new gadolinium activity over 9 months. A crossover design without a concurrent comparison group may overestimate the treatment effect.