OBJECTIVE: To explore, in a longitudinal study, the usefulness of optical coherence tomography (OCT) in monitoring people with multiple sclerosis (MS) by testing the association between retinal nerve fiber layer (RNFL) thinning and clinical and brain MRI criteria of no evidence of disease activity (NEDA). METHODS: OCT, visual evoked potentials (VEPs), and disability, using the Expanded Disability Status Scale (EDSS), were tested at baseline and after 2 years in 72 patients, 63 with routine yearly brain MRI. RESULTS: Longitudinal mean binocular RNFL thinning, in absence of optic neuritis during follow-up, was correlated with EDSS worsening, also controlling for baseline EDSS, RNFL, disease duration, and MS subtype (Spearman ρ -0.462, p <0.001; partial correlation coefficient -0.437, p <0.001). At follow-up, patients classified as NEDA (20; 31.7%) had RNFL loss of -0.93 μm ± 1.35 SD, while patients with active disease had -2.83 μm ± 2 SD thinning (t test; p <0.001). At logistic regression, mean RNFL reduction correctly classified 76.2% of patients as NEDA at 2 years (R2 0.355; p = 0.003). A cutoff of -1.25 μm RNFL loss classified NEDA status with specificity 81.4% and sensitivity 80% (receiver operating characteristic curve: area under the curve 0.8; p <0.001). No significant longitudinal correlations were found between changes in RNFL and in VEP latencies or scores. CONCLUSIONS: NEDA is associated with a relatively preserved RNFL over 2 years. A greater neuroretinal loss was detected even in patients with clinical evidence of disease activity independently from changes in brain MRI lesions, prompting further validation of OCT as an additional tool in MS monitoring. © 2017 American Academy of Neurology.