Rituximab was noninferior to ocrelizumab on the absence of new or enlarging T2-weighted magnetic resonance imaging lesions from month 6 to month 24 among patients with newly diagnosed relapsing multiple sclerosis, according to findings from the randomized phase 3 OVERLORD-MS trial.
The trial did not establish superiority or equivalence between the 2 therapies. Secondary clinical outcomes appeared broadly similar, serious adverse events occurred at similar rates, and infections were more common with rituximab; however, the investigators noted that the study was not powered to detect differences in secondary endpoints or rare safety outcomes.
Researchers conducted the randomized, double-blind, multicenter, noninferiority trial at 12 neurology departments in Norway and Sweden. Eligible participants were aged 18 to 60 years, had received a diagnosis of relapsing multiple sclerosis within the previous 12 months, had evidence of recent inflammatory disease activity, and had an Expanded Disability Status Scale score of 0 to 4.0. Evidence of recent inflammatory disease activity was defined as at least 1 clinical relapse or at least 1 new or enlarging MRI lesion within the previous 12 months.
Patients were excluded if they had progressive multiple sclerosis, previous exposure to disease-modifying therapy, active infection, pregnancy or lactation, or contraindications to B-cell–depleting therapy or MRI.
A total of 218 participants underwent randomization in a 3:2 ratio, and 216 received at least 1 infusion and were included in the modified intention-to-treat population: 132 assigned to rituximab and 84 assigned to ocrelizumab. Participants received intravenous rituximab 1000 mg at baseline followed by 500 mg every 6 months, or intravenous ocrelizumab 600 mg at baseline and every 6 months. MRI scans were obtained at baseline and at months 6, 12, and 24, and MRI outcomes were assessed at a blinded central reading center.
The primary endpoint was absence of new or enlarging lesions on T2-weighted MRI between month 6 and month 24. The investigators used month 6 as the reference time point to allow for full treatment effect after initiation of B-cell–depleting therapy and to minimize the influence of preexisting inflammatory activity.
Between months 6 and 24, 117 of 132 participants receiving rituximab, or 89%, and 78 of 84 receiving ocrelizumab, or 93%, had no new or enlarging T2-weighted MRI lesions. The corresponding model-estimated probabilities were 92.2% and 94.8%, respectively. The risk difference for rituximab minus ocrelizumab was −2.6 percentage points, with a 95% CI of −9.4 to 4.3, meeting the prespecified noninferiority criterion of a lower 95% CI boundary no less than −10 percentage points.
The investigators reported that results were consistent in prespecified sensitivity analyses, including a Newcombe hybrid risk-difference analysis in the modified intention-to-treat population and a per-protocol analysis. Seven participants had missing primary-endpoint data, which were handled with multiple imputation as prespecified.
Secondary MRI outcomes also showed low inflammatory activity in both treatment groups. From baseline to month 24, the model-estimated probability of having no new or enlarging T2-weighted MRI lesions was 80% with rituximab and 79% with ocrelizumab. No new contrast-enhancing lesions were observed among participants with available MRI data at months 6, 12, or 24. Brain-volume analyses were descriptive only because of missing data, and no formal between-group statistical comparisons were performed.
Clinical outcomes were infrequent in both groups. The estimated annualized relapse rate over 24 months was 0.09 with rituximab and 0.04 with ocrelizumab, with a between-group difference of 0.05 relapses per year. A total of 92% of participants receiving rituximab and 94% receiving ocrelizumab remained relapse-free through 24 months. Confirmed disability progression at month 24, confirmed at month 30, occurred in 3% and 7%, respectively, whereas confirmed disability improvement occurred in 29% and 24%. Worsening on the Symbol Digit Modalities Test was uncommon in both groups.
The investigators cautioned that secondary endpoints were not adjusted for multiplicity and that point estimates and confidence intervals for these outcomes should not be interpreted as causal.
Adverse events occurred in 86% of participants receiving rituximab and 80% receiving ocrelizumab. Serious adverse events occurred in 8% and 7%, respectively, and no deaths were reported. Infections were more common with rituximab than with ocrelizumab, occurring in 82% vs 69% of participants, although most were mild upper respiratory tract infections. Serious infections occurred in 4 participants in each group. Infusion-related reactions occurred in 23% of participants receiving rituximab and 25% receiving ocrelizumab. One malignant melanoma was reported in the ocrelizumab group; no neoplasms were reported in the rituximab group.
The authors noted several limitations. The trial was powered for the prespecified MRI-based noninferiority comparison, and the low number of MRI and clinical events limited the precision of secondary analyses and precluded meaningful subgroup analyses. The trial also was not powered to detect differences in safety outcomes or rare adverse events. Follow-up was limited to 30 months, and the study population consisted of previously untreated patients with newly diagnosed disease from Norway and Sweden, which may limit generalizability to patients with longer disease duration, previous treatment exposure, or more diverse populations. Approximately half the participants were enrolled at a single high-volume center.
The authors also reported that the trial had a randomization implementation error at some sites that reversed the allocation sequence. After the error was identified, new site-specific allocation lists were generated with an adjusted allocation ratio to preserve the intended overall 3:2 allocation.
“[T]he trial addressed noninferiority only; larger studies are needed to assess superiority,” wrote lead author Øivind Torkildsen, MD, of Neuro-SysMed, Department of Neurology, Haukeland University Hospital, and the Department of Clinical Medicine, University of Bergen, Norway, and colleagues.
The study was funded by the Research Council of Norway, KLINBEFORSK, the Swedish Research Council, Region Stockholm, the Swedish Brain Fund, the Erling Perssons Foundation, and the Horizon Europe Framework Program. The authors reported no commercial involvement in the trial, and disclosure forms are available with the published article.