Defining multiple sclerosis disease activity using MRI T-2-weighted difference imaging

Serial brain MRI scanning is widely used for assessing multiple sclerosis d isease activity in the evaluation of new therapies. Traditionally, the net change in T-2-weighted lesion volume between paired scans has been used as a measure of disease progression and as a secondary endpoint in definitive clinical trials. However, as the net change in T-2-weighted lesion volume r eflects only the difference between new and resolved T-2-weighted lesions, this measure significantly under-represents the total T-2- weighted lesion activity. Difference images produced by subtracting labelled T-2-weighted l esion volumes from serial registered T-2-weighted scans allows separate mea surements of individual volumes of new and resolving T-2-weighted lesions, which may reflect underlying disease activity more sensitively. We generate d T-2-weighted differences images to define T-2-weighted lesion changes ove r 1 year for 19 patients with relapsing multiple sclerosis, The mean new T- 2-weighted lesion volume change was three times greater than the mean net T -2-weighted lesion volume change over the study period. New T-2-weighted le sion volumes were more strongly correlated with T-1-weighted gadolinium-enh ancing lesion volumes (I = 0.72, P = 0.001) than were the net T-2-weighted lesion volume changes (r = 0.45, P = 0.01). Baseline T-2-weighted lesion vo lume was more highly correlated with new T-2-weighted lesion volumes (r = 0 .89, P < 0.0001) than with net T-2-weighted lesion change (r = 0.47, P < 0. 001), There was a trend for patients who showed sustained clinical progress ion over the year to have a greater new T-2-weighted lesion volume than tho se who did not. This difference was not seen with net T-2-weighted lesion v olume change. T-2-weighted lesion difference images should provide an addit ional and sensitive tool for monitoring disease activity in multiple sclero sis, Independent definition of new and resolving T-2-weighted lesion volume s also offers the potential for discrimination of the relative effects of e xperimental therapies on new inflammatory activity from the effects on oede ma resolution and lesion repair.