Longitudinal Visual Evoked Potential Changes Show Distinct Associations with Relapse Activity and Disability in Relapsing–Remitting Multiple Sclerosis: A Multicenter Retrospective Cohort Study

Background and Objectives: Visual evoked potentials (VEPs) are established tools for detecting optic pathway involvement in multiple sclerosis (MS); however, the distinct clinical significance of longitudinal changes in latency and amplitude remains unclear. This exploratory study aimed to evaluate the clinical utility of serial VEP assessments in patients with relapsing–remitting MS (RRMS) by examining the associations between temporal changes in latency and amplitude and concurrent markers of disease activity and disability. Materials and Methods: We retrospectively analyzed 83 patients with RRMS from three centers who underwent at least two pattern-reversal VEP assessments ≥ 1 year apart. Patients with recent optic neuritis were excluded. Longitudinal changes in VEP latency and amplitude were evaluated in relation to relapse activity during the same inter-assessment interval, baseline MRI and CSF findings, and disability assessed by the Expanded Disability Status Scale (EDSS). Sensitivity analyses were stratified and adjusted for disease-modifying therapy (DMT) efficacy tier, and progression independent of relapse activity (PIRA) was examined using a formal definition. Results: Over a median follow-up of 2 years, 43.4% of patients developed P100 latency prolongation, which was independently associated with a higher annualized relapse rate during the same interval (odds ratio [OR] 2.91, p = 0.005; area under the curve (AUC) 0.72 for concurrent on-study relapse activity), an association that persisted after adjustment for DMT efficacy tier, but not with disability worsening. In contrast, 56.6% of patients exhibited amplitude reduction, which was associated with greater disability accumulation at the group level, with higher EDSS scores at follow-up despite similar baseline values (p < 0.05); the corresponding inverse correlation between amplitude decline and EDSS change was only nominally significant (ρ = −0.231, p = 0.036), did not survive correction for multiple comparisons, and was attenuated after adjustment for baseline EDSS. Formally defined PIRA events were rare (n = 3), precluding inference. Baseline magnetic resonance imaging (MRI) and cerebrospinal fluid (CSF) findings differed in the latency group, with fewer MRI lesions and a lower immunoglobulin G (IgG) index, whereas no corresponding differences were observed in the amplitude group. Conclusions: In this exploratory cohort, longitudinal prolongation of VEP latency was associated with concurrent relapse activity, whereas amplitude decline showed a weaker, group-level association with disability. These findings are consistent with, but do not demonstrate, distinct inflammatory and neurodegenerative substrates, and warrant prospective validation before serial VEP monitoring can inform individualized management in MS.