Journal of neurology vol:250 issue:9 pages:1037-1043
BACKGROUND: Interferon- beta (IFNbeta) acts by a variety of mechanisms in relapsing-remitting multiple sclerosis (MS). One of these is a cellular down-regulation of gelatinase B or matrix metalloproteinase-9 (MMP-9), which is known from biochemical, biological and immunohistochemical evidences to play a disease-promoting role in MS. AIMS: a) To investigate the influence of IFNbeta-1a (30 or 60 micro g I. M./week) on serum MMP-9 levels in patients with primary progressive MS (PPMS). b) To correlate serum MMP-9 levels with clinical and magnetic resonance imaging (MRI) findings. METHODS: Serial blood samples were collected every 3 months from 49 patients participating in a phase II trial of IFNbeta-1a in PPMS. Serum MMP-9 was quantified by ELISA and correlations with clinical (EDSS) as well as MRI findings (brain and spinal cord atrophy, ventricular volume, T1 and T2 lesion load) were calculated. RESULTS: No significant differences were found between serial serum MMP-9 levels in IFNbeta-treated versus placebo-treated patients. MMP-9 levels did not differ between patients who progressed or did not progress during the study interval. Although mean absolute serum MMP-9 levels over the study period correlated with an increase in T2 lesion load (relative T2 change: r=0.51, p<0.001; absolute T2 change: r=0.30, p=0.038), absolute increase in brain ventricular volume (r=0.29, p=0.05) and increased brain atrophy (r=0.35, p=0.02), only the correlation with T2 lesion load was sustained throughout the study period. No correlations were found between MMP-9 levels and relative changes in ventricular volume or with relative/absolute changes in T1 lesion load and in spinal cord atrophy. None of the MRI measures correlated with MMP-9 changes between baseline levels and those on treatment. CONCLUSION: Although some evidence suggests a down-regulating effect of IFNbeta on MMP-9, this was not confirmed for a once weekly intramuscular dose of IFNbeta-1a in patients with PPMS. The sustained correlation between serum MMP-9 and changes in T2 volumes, and the lack of correlation with clinical or MRI measures of disease progression may suggest that MMP-9 is more directly related to non-specific central nervous system damage than to axonal loss.