An MRI-defined measure of cerebral lesion severity to assess therapeutic effects in multiple sclerosis

Abstract

Assess the sensitivity of the Magnetic Resonance Disease Severity Scale (MRDSS), based on cerebral lesions and atrophy, for treatment monitoring of glatiramer acetate (GA) in relapsing-remitting multiple sclerosis (MS). This retrospective non-randomized pilot study included patients who started daily GA [n = 23, age (median, range) 41 (26.2, 53.1) years, Expanded Disability Status Scale (EDSS) score 1.0 (0, 3.5)], or received no disease-modifying therapy (noDMT) [n = 21, age 44.8 (28.2, 55.4), EDSS 0 (0, 2.5)] for 2 years. MRDSS was the sum of z-scores (normalized to a reference sample) of T2 hyperintense lesion volume (T2LV), the ratio of T1 hypointense LV to T2LV (T1/T2), and brain parenchymal fraction (BPF) multiplied by negative 1. The two groups were compared by Wilcoxon rank sum tests; within group change was assessed by Wilcoxon signed rank tests. Glatiramer acetate subjects had less progression than noDMT on T1/T2 [(median z-score change (range), 0 (-1.07, 1.20) vs. 0.41 (-0.30, 2.51), p = 0.003)] and MRDSS [0.01 (-1.33, 1.28) vs. 0.46 (-1.57, 2.46), p = 0.01]; however, not on BPF [0.12 (-0.18, 0.58) vs. 0.10 (-1.47,0.50), p = 0.59] and T2LV [-0.03 (-0.90, 0.57) vs. 0.01 (-1.69, 0.34), p = 0.40]. While GA subjects worsened only on BPF [0.12 (-0.18, 0.58), p = 0.001], noDMT worsened on BPF [0.10 (-1.47, 0.50), p = 0.002], T1/T2 [0.41 (-0.30, 2.51), p = 0.0002], and MRDSS [0.46 (-1.57, 2.46), p = 0.0006]. These preliminary findings show the potential of two new cerebral MRI metrics to track MS therapeutic response. The T1/T2, an index of the destructive potential of lesions, may provide particular sensitivity to treatment effects.

Keywords: Brain atrophy; Composite scale; Glatiramer acetate; MRI; Multiple sclerosis.

Fig. 1

On-study brain atrophy in glatiramer acetate vs. untreated patients. Mean (±standard error of the mean) of brain parenchymal fraction (BPF) at baseline and 2-year follow-up. A lower score indicates advancing disease. Both glatiramer acetate (GA) and no disease modifying therapy (noDMT) cohorts showed significant decreases in BPF (i.e. no brain atrophy) from baseline to follow-up. The p values in the figure are from Wilcoxon signed rank tests for the within group change over time. Furthermore, when comparing the change in zBPF between the two groups, no difference was found (Wilcoxon rank sum test, p = 0.59)

Fig. 2

On-study T2 hyperintense lesion volume changes in glatiramer acetate vs. untreated patients. Mean (±standard error of the mean) of total cerebral T2 hyperintense lesion volume (T2LV) at baseline and 2-year follow-up. A higher score indicates advancing disease. Both glatiramer acetate (GA) and no disease modifying therapy (noDMT) cohorts showed no significant decreases from baseline to follow-up. The p values in the figure are from Wilcoxon signed rank tests for the within group change over time. Furthermore, when comparing the change in zT2LV between the two groups, no difference was found (Wilcoxon rank sum test, p = 0.40)

Fig. 3

On-study T1/T2 ratio changes in glatiramer acetate vs. untreated patients. Mean (±standard error of the mean) of the ratio of total cerebral T1 hypointense to T2 hyperintense lesion volume (T1/T2) at baseline and 2-year follow-up. A higher score indicates advancing disease. The no disease modifying therapy (noDMT) group showed significant worsening, but the glatiramer acetate (GA) treated group did not. The p values in the figure are from Wilcoxon signed rank tests for the within group change over time. Furthermore, when comparing the change in zT1/T2 between groups, a difference was found favoring GA treatment (Wilcoxon rank sum test, p = 0.003)

Fig. 4

On-study MRDSS changes in glatiramer acetate vs. untreated patients. Mean (±standard error of the mean) of the magnetic resonance disease severity scale (zMRDSS) at baseline and 2-year follow-up. A higher score indicates advancing disease. The no disease modifying therapy (noDMT) group showed significant worsening, but the glatiramer acetate (GA) treated group did not. The p values in the figure are from Wilcoxon signed rank tests for the within group change over time. Furthermore, when comparing the change in zMRDSS between groups, a difference was found favoring GA treatment (Wilcoxon rank sum test, p = 0.01)