The ability to predict disability development in multiple sclerosis (MS) is limited. While abnormalities of evoked potentials (EP) have been associated with disability, the prognosticating utility of EP in MS remains to be fully elucidated. The present study assessed the utility of multimodal EP as a prognostic biomarker of disability in a cohort of clinically heterogeneous MS patients. Median and tibial nerve somatosensory, visual, and brainstem auditory EP were performed at initial assessment on 63 MS patients (53 relapsing-remitting and 10 secondary progressive) who were followed for an average of 2years. A combined EP score (CEPS) was calculated consisting of the total number of abnormal EP tests, and was correlated with the Expanded Disability Status Scale (EDSS) at baseline and follow-up. There was a significant correlation between multimodal EP and baseline and follow-up EDSS. Specifically, tibial nerve P37 latencies correlated with EDSS (RBASELINE=0.49, p<0.01; RFOLLOW-UP=0.47, p<0.01), as did the median nerve N13 (RBASELINE=0.40, p<0.01; RFOLLOW-UP=0.35, p<0.05) and N20 latencies (RBASELINE=0.43, p<0.01; RFOLLOW-UP=0.47, p<0.01), and P100 full-field (RBASELINE=0.50, p<0.001; RFOLLOW-UP=0.45, p<0.001) and central field latencies (RBASELINE=0.60, p<0.001; RFOLLOW-UP=0.50, p<0.001). In addition, there was a significant correlation between the CEPS with baseline (R=0.65, p<0.001) and follow-up (R=0.57, p<0.01) EDSS. In contrast, white matter disease burden, as measured by T2 lesion load, exhibited a weaker correlation with EDSS (RBASELINE=0.28, p<0.05). In conclusion, these findings suggest that abnormalities of EP, as quantified by the novel CEPS, may be a useful biomarker for prognosticating clinical disability in MS, and may aid in the quantification of MS disease severity and in guiding therapeutic decisions.