Longitudinal disability curves for predicting the course of relapsing-remitting multiple sclerosis

Abstract

Background: Multiple sclerosis (MS) is a lifelong disease affecting young adults that ultimately can lead to significant neurological disability. Identifying the rate of progression of disability for an individual patient early in the disease process can influence treatment decisions as well as enable monitoring of the disease overtime.

Objective: The aims of the present study were to develop longitudinal disability curves to assess disease progression in patients with relapsing remitting MS. The construction of these disability curves was based on the mean yearly Expanded Disability Status Scale (EDSS) scores, represented as a major percentile group. To determine their predictive ability, validation of the percentile curves was performed.

Methods: Using the Multiple Sclerosis Center computerized database of 1540 patients' records, we identified 1317 subjects with a definite MS and a relapsing-remitting disease course. Longitudinal disability curves were constructed for a subgroup of relapsing-remitting patients (n = 1001) with consecutive (3-6 months) EDSS assessments for a period of up to 10 years since onset. The constructed disability curves were then validated in an additional subgroup of relapsing-remitting MS patients (n = 268) with continuous follow-up visits for a period of 10 years.

Results: Statistical procedures using parametric and nonparametric regression procedures were applied to the data in two stages. In the first stage, selected major percentiles were generated for up to 10 years from disease onset with a variety of parametric procedures including moving averages. In the second stage, the empirical percentiles were smoothed to obtain the final disability progression curves. The log-rank test for equality demonstrated a significant adjustment between the initial percentile assignment and disability progression (P < 0.001). Life table analysis demonstrated that the probability of deviating from the initially assigned percentile to a higher percentile over time, representing more severe disability than expected, is in the range of 6.5% for the 50th percentile to 10.4% for the 75th percentile.

Conclusion: Longitudinal disability curves can be used in MS to assess individual patient disability, can contribute to the overall clinical impression of disease progression and can add to the evaluation of immunomodulating treatment effects overtime.