Predicting treatment failure and switching effective treatment immediately in patients with multiple sclerosis (MS) is important. We aimed to evaluate the usefulness of Modified Rio score (MRS) in predicting treatment failure in MS patients. This is a retrospective study, which was conducted in two University Hospital. 129 MS patients treated with İnterferon or glatiramer-acetate from 2 clinical sites, were retrospectively selected. MRS was calculated after the first year of therapy. Treatment failure was defined as the presence of a 1 point increase in EDSS, 2 clinical attacks, 1 clinical attack and progression, 1 clinical attack and new lesion on MRI except associated with an attack, or new lesion in 2 different MRI taken at least 3 months apart. The sensitivity, specificity, positive and negative predictive values of the MRS in predicting treatment failure were determined. 71 (55%) patients with score ‘0’, 41 (31.8%) patients with score ‘1’, 11 (8.5%) patients with score ‘2’, 6 (4.7%) patients with score ‘3’ were detected. 14 patients needed treatment switching during the first three years of the treatment. Sensitivity was 57%, specificity was 92%, positive predictive value was 95%, negative predictive value was 47% and accuracy was 89%. Modified Rio score (MRS) was found to be effective in determining the treatment failure as mentioned before. This study will be useful for clinicians who evaluate the treatment failure like us, and this study revealed that the MRS may also help predict treatment failure.