To derive early demographic, clinical, radiographic, and laboratory factors that predict the 10-year outcome of patients with multiple sclerosis (MS).

The course of MS is variable: some endure many attacks and accrue irreversible disability; others sustain milder disease. Certain prognostic factors confer greater risk, but to an unknown extent, and high-efficacy treatment (HET) reduces disease activity but has side effect risks. Thus, early prediction of disease course may allow for personalized MS treatment.

We reviewed patients enrolled in the Comprehensive Longitudinal Investigation of MS at Brigham and Women’s Hospital (CLIMB) with 10-year follow-up. Thirty-five predictor variables were collected: baseline demographics and comorbidities, and early, defined as within 3 years from symptom onset, clinical, radiographic, and serum biomarker data, including neurofilament light-chain levels (NFL). Outcome variables, defined from year 3 to 10 post-symptom onset, included escalation to HET, new attacks or MRI activity, expanded disability status scale (EDSS), and brain atrophy. Multiple logistic and linear regression analysis was performed.

Complete data was available for 122 patients. Fifty-two (43%) escalated to HET and 86 (70.5%) had new attacks, both associated with younger age (p=0.001 and p=0.009, respectively). MRI activity in 99 (81%) was associated with higher BMI (p=0.006), vascular comorbidities (p=0.006), NFL (p=0.027), and early MRI activity (p=0.015). Forty patients (33%) had EDSS worsening, associated with injectables as first treatment (p=0.018) and less likely in patients with monofocal sensory attacks (p=0.026). Brain atrophy was associated with neurologic comorbidities (p=0.005) and NFL (p=0.008).

Different clinical, radiographic, and serum biomarkers are associated with 10-year outcome measures in MS. A prognostic score to stratify patients for treatment decisions must be encompassing and prospectively validated.