Assessing information in T2-weighted MRI scans from secondary progressive MS patients

Magnetic resonance imaging is used routinely for diagnosing MS and for objective assessment of the extent of disease as a marker of treatment efficacy in MS clinical trials. Nevertheless, in many clinical studies only weak correlations have been reported between MRI findings and clinical outcome measures.

The purpose of this study is to compare clinical outcome measures (neurologic scales) with MRI findings (evaluation of T2-weighted MRI scans using a semiautomated quantitative technique and with an independent assessment by a neurologist) in the context of a randomized clinical trial evaluating the efficacy of cladribine for treatment in secondary progressive MS.

Baseline, 6-month, and 12-month scans from 41 secondary progressive MS patients were examined and ranked in terms of lesion burden from the quantitative assessment and independently in terms of severity by neurologic evaluation. Comparison is made to monthly Expanded Disability Status Scale (EDSS) and Scripps Neurologic Rating Scale (SNRS) determinations in these patients with a nonparametric statistical procedure.

Average rank correlations between any of the MRI assessment procedures and either clinical outcome measure were less than 0.15 in absolute magnitude. The average rank correlation between the two MRI assessment procedures was 0.10. There is only a weak degree of association between the MRI assessment procedures and the clinical parameters, although the study has statistical power in excess of 0.90 to find even a moderate level of association between them.

Disease-related activity in T2-weighted scans of secondary progressive MS patients is a multidimensional construct, and is not summarized adequately solely by quantification of overall lesion burden or by assessment of severity. Neither method of summarizing information from T2-weighted scans is strongly related to measures of the clinical course of disease as assessed by the EDSS or SNRS.