BACKGROUND: Cognitive impairment is a well-recognized symptom of multiple sclerosis (MS) that can manifest early in the disease course. Deficits in cognitive function can have a major impact on daily life. However, cognitive decline is often under-examined in clinical trials and clinical practice due to lack of adequate data. The objective of this study was to examine the longitudinal effect of ocrelizumab vs interferon beta (IFNβ)-1a on cognitive impairment in 2 phase 3 studies in relapsing MS (RMS). METHODS: The pooled population of participants with RMS (n = 1656) from the OPERA I/II clinical trials received subcutaneous IFNβ-1a (44 μg
n = 829) 3 times weekly or intravenous ocrelizumab (600 mg
n = 827) every 24 weeks. Cognition was assessed with a Symbol Digit Modalities Test (SDMT), administered in written or oral form according to each site investigator's choice, that primarily measured cognitive processing speed at baseline and every 12 weeks until the end of the double-blind treatment (96 weeks). Treatment effects were investigated based on longitudinal linear models for the change from baseline in SDMT and Cox regression for the time to 12- or 24-week confirmed decline of ≥4 points. RESULTS: Among the participants with an SDMT assessment at baseline and ≥1 postbaseline time point (IFNβ-1a, n = 749
ocrelizumab, n = 766), ocrelizumab treatment was associated with a greater mean SDMT improvement over 96 weeks than IFNβ-1a treatment (5.4 [95 % CI, 4.4-6.5] vs 4.0 [95 % CI, 3.0-5.1]
adjusted mean difference, 1.4 [95 % CI, 0.05-2.72]
P = 0.042). The risk of a clinically meaningful SDMT decline (≥4 points) was lower for those treated with ocrelizumab for both ≥12 weeks (IFNβ-1a, 18.4 %
ocrelizumab, 12.7 %
hazard ratio, 0.63 [95 % CI, 0.47-0.85]
P = 0.003) and ≥24 weeks (IFNβ-1a, 12.9 %
ocrelizumab, 7.9 %
HR, 0.57 [95 % CI, 0.39-0.82]
P = 0.003). CONCLUSION: Ocrelizumab treatment resulted in better cognitive outcomes as measured by SDMT in participants with RMS compared with IFNβ-1a treatment. However, methodological limitations need to be considered when interpreting these data. CLINICALTRIALS: gov: NCT01247324, NCT01412333.