Background: Radiologically isolated syndrome (RIS) represents a subclinical period of multiple sclerosis (MS).

Objective: We aimed to characterize and identify risk factors for developing MS in an RIS cohort and to assess various proposed RIS definitions for their predictive value in MS development.

Methods: This cohort study included all patients with at least one typical inflammatory-demyelinating lesion suggestive of MS on brain and/or spinal cord magnetic resonance imaging (MRI). The development of MS symptoms and new T2 lesions were the primary and secondary outcomes, respectively. Cox regression was used to identify risk factors, and diagnostic performance was assessed.

Results: Eighty-eight patients were included, with 25.0% developing MS symptoms over a mean 55.1-month follow-up. Younger age and spinal cord and cortico-/juxtacortical lesions were associated with worse outcomes. The 2017 McDonald dissemination in space (DIS) criteria and 2023 Lebrun RIS definition, which correspond to either the 2005 McDonald DIS criteria or one brain inflammatory-demyelinating lesion associated with two among oligoclonal bands, a spinal cord lesion and dissemination in time during radiological follow-up, showed high sensitivity (0.82 and 0.94, respectively). However, the sensitivity decreased (0.79) when only the baseline characteristics of the 2023 Lebrun RIS definition were considered. Combining the 2017 McDonald DIS criteria with positivity for oligoclonal bands or contrast-enhancing lesions on baseline MRI, which are the current McDonald MS criteria, improved the specificity (to 0.64 and 0.90, respectively).

Conclusions: Our findings support the utility of the current DIS component of the 2017 McDonald MS criteria for RIS patients.