Predicting the On-Study Relapse Rate for Multiple Sclerosis Patients in Clinical Trials

Held, U.; Heigenhauser, L.; Shang, C.; Kappos, L. and Polman, C. (2005): Predicting the On-Study Relapse Rate for Multiple Sclerosis Patients in Clinical Trials. Collaborative Research Center 386, Discussion Paper 430

Abstract

Background: The annual relapse rate has been commonly used as a primary efficacy endpoint in phase III multiple sclerosis (MS) clinical trials. The aim of this study was to determine the relative contribution of different possible prognostic factors available at baseline to the on-study relapse rate in MS. Methods: A total of 821 patients from the placebo arms of the Sylvia Lawry Centre for Multiple Sclerosis Research (SLCMSR) database were available for this analysis. The univariate relationships between on-study relapse rate and the baseline demographic, clinical, and MRI-based predictors were assessed. The multiple relationships were then examined using a Poisson regression model. Two predictor subsets were selected. Subset 1 included age at disease onset, disease duration, gender, Expanded Disability Status Scale (EDSS) at baseline, number of relapses in the last 24 months prior to baseline, and the disease course (RR and SP). Subset 2 consisted of Subset 1 plus gadolinium enhancement status in MRI. The number of patients for developing the models with no missing values was 727 for Subset 1 and 306 for Subset 2. Results:The univariate relationships show that the on-study relapse rate was higher for younger and for female patients, for RR patients than for SP patients, and for patients with positive enhancement status at entry (Wilcoxon test, p<0.05). A higher on-study relapse rate was associated with a shorter disease duration, lower entry EDSS, more pre-study relapses and more enhancing lesions in T1 at entry. The fitted Poisson model shows that disease duration (estimate=-0.02) and previous relapse number (estimate=0.59 for 1, 0.91 for 2 and 1.45 for 3 or more relapses vs 0 relapse) remain. We were able to confirm these findings in a second, independent dataset. Conclusions: The relapse number prior to entry into clinical trials together with disease duration are the best predictors for the on-study relapse rate. Disease course and gadolinium enhancement status, given the other covariates, have no significant influence on the on-study relapse rate.

• Item Type: Paper (Research Paper)
• Subjects: Mathematics, Computer Science and Statistics; Mathematics, Computer Science and Statistics > Statistics; Mathematics, Computer Science and Statistics > Statistics > Collaborative Research Center 386
• Dewey Classification: 600 Natural sciences and mathematics; 600 Natural sciences and mathematics > 510 Mathematics