Matrix metalloproteinase-9 (MMP-9) is involved in blood-brain barrier (BBB) disruption in active multiple sclerosis (MS), while MMP-2 seems to be associated with the chronic progressive phase of the disease. Recombinant interferon beta-1a (rIFNbeta-1a) is effective in restoring the BBB. We studied the relationships between serum MMP-9, MMP-2, TIMP-1 and TIMP-2 and different magnetic resonance imaging (MRI) measures of disease activity in MS patients during treatment with rIFNbeta-1a. Twenty-one relapsing-remitting (RR) MS patients underwent longitudinally simultaneous blood withdrawals and MRI (before and after standard dose (SD) and triple dose (TD) of gadolinium (Gd)) examinations before and during 48 weeks of rIFNbeta-1a (Rebif 22 mcg three times a week) treatment. Serum MMP-9, MMP-2, TIMP-1 and TIMP-2 were measured, MMP-9 to TIMP-1 and MMP-2 to TIMP-2 ratios were calculated and the numbers of Gd-SD, Gd-TD, new-Gd-SD, new-Gd-TD and new-T2 lesions counted. Serum MMP-9/TIMP-1 ratio (P < 0.0001), as well as the numbers of 'active' lesions (P ranging from 0.0004 to 0.005) decreased during treatment Moreover, serum MMP-9/TIMP-1 ratio proved to be a good positive predictor (estimate = 0.85; P < 0.05) of the numbers of MRI Gd-TD active lesions. These data confirm that serum MMP-9/TMIP-1 ratio may be viewed as a reliable marker and may be predictive of MRI activity in RR MS.