OVERLORD-MS
Rituximab versus Ocrelizumab in Newly Diagnosed Relapsing Multiple Sclerosis
Year of Publication: 2026
Authors: Torkildsen Ø, Brustad HK, Høgestøl EA, ..., for the OVERLORD-MS Investigators
Journal: New England Journal of Medicine
Citation: N Engl J Med 2026;395:44-53
PDF: Clinical Question
Is rituximab noninferior to ocrelizumab for suppressing MRI-detected disease activity in newly diagnosed relapsing multiple sclerosis?
Bottom Line
In newly diagnosed relapsing MS, rituximab was noninferior to ocrelizumab in suppressing MRI disease activity from month 6 to 24, with similar serious adverse events, supporting biosimilar rituximab as a cost-effective alternative to ocrelizumab.
Major Points
- First phase 3 randomized head-to-head trial of rituximab vs ocrelizumab in newly diagnosed relapsing MS
- Rituximab met prespecified noninferiority for MRI activity suppression (risk difference -2.6 pp, 95% CI -9.4 to 4.3; noninferiority margin -10 pp)
- Between months 6 and 24, 92.2% (rituximab) vs 94.8% (ocrelizumab) had no new/enlarging T2 lesions
- Clinical outcomes (relapse rates, disability, cognition) were similar between groups
- Infections more common with rituximab (82% vs 69%), but serious adverse events similar (8% vs 7%)
- Findings support use of biosimilar rituximab as a lower-cost alternative to ocrelizumab
Design
Study Type: Phase 3, multicenter, double-blind, noninferiority randomized controlled trial
Randomization: 1
Blinding: Double-blind (participants and trial personnel unaware of assignments; trial medications prepared by unblinded pharmacy personnel with identical infusion setup)
Allocation: 3:2 (rituximab:ocrelizumab); initially 2:1 with implementation error at some sites reversing allocation, subsequently corrected with adjusted 9:1 allocation to achieve overall 3:2
Enrollment Period: November 2020 to November 2022
Follow-up Duration: 24 months, with prespecified blinded follow-up at month 30
Centers: 12
Countries: Norway, Sweden
Sample Size: 218
Analyzed: 216
Analysis: Modified intention-to-treat (all who received ≥1 dose); logistic regression with treatment group as covariate; multiple imputation for missing data; sensitivity analyses in per-protocol and complete-case populations
Power Calculation: Assumed 0.95 proportion without new/enlarging T2 lesions, noninferiority margin 10 pp, two-sided alpha 0.05, 90% power, 20% dropout; target sample size 208
Registration: ClinicalTrials.gov NCT04578639; EudraCT 2020-001205-23; EU Clinical Trials Register 2024-510716-71-00
Inclusion Criteria
- Age 18-60 years
- Diagnosis of relapsing multiple sclerosis within previous 12 months
- Evidence of recent inflammatory disease activity (≥1 clinical relapse OR ≥1 new/enlarging MRI lesion within previous 12 months)
- EDSS score 0 to 4.0
- Diagnosis established per McDonald criteria by treating neurologists
Exclusion Criteria
- Progressive form of multiple sclerosis
- Previous exposure to disease-modifying therapies
- Active infection
- Pregnancy or lactation
- Contraindications to B-cell-depleting therapy or MRI examinations
Baseline Characteristics
| Characteristic | Rituximab (n=132) | Ocrelizumab (n=84) |
|---|---|---|
| Age (yr) | 37.4 ± 9.4 | 36.6 ± 10.1 |
| Female (%) | 72 (95/132) | 68 (57/84) |
| Months since first clinical event | 20.8 ± 42.0 | 24.1 ± 38.3 |
| Months since diagnosis | 0.6 ± 2.0 | 0.4 ± 0.5 |
| EDSS score | 1.6 ± 1.1 | 1.7 ± 1.0 |
| Relapses in past year | 1.2 ± 0.4 | 1.1 ± 0.6 |
| Total relapses | 1.5 ± 0.7 | 1.8 ± 1.0 |
| Days since last relapse | 120 ± 218 | 131 ± 310 |
| Current smoker (%) | 10 (13/132) | 8 (7/84) |
| BMI | 26.3 ± 5.2 | 25.9 ± 4.8 |
| T2 lesions ≥20 (%) | 64 (84/132) | 61 (51/84) |
| Contrast-enhancing lesions ≥1 (%) | 42 (56/132) | 40 (34/84) |
| Oligoclonal bands positive (%) | 83 (109/132) | 92 (77/84) |
Arms
| Field | Rituximab | Control |
|---|---|---|
| N | 132 | 84 |
| Intervention | Intravenous rituximab (Rixathon, Sandoz) 1000 mg at baseline, then 500 mg every 6 months; premedication with IV methylprednisolone 100 mg, oral antihistamine, and antipyretic | Intravenous ocrelizumab (Ocrevus, Roche) 600 mg at baseline and every 6 months; premedication with IV methylprednisolone 100 mg, oral antihistamine, and antipyretic |
| Duration | 24 months | 24 months |
Outcomes
| Outcome | Type | Control | Intervention | HR / OR / RR | P-value |
|---|---|---|---|---|---|
| Absence of new or enlarging lesions on T2-weighted MRI from month 6 to month 24 | Primary | 94.8% (78/84 observed; 93%) | 92.2% (117/132 observed; 89%) | Risk difference -2.6 percentage points (rituximab minus ocrelizumab); met prespecified noninferiority margin of -10 pp | 0.03 (two-sided, for noninferiority) |
| No new/enlarging T2 lesions from baseline to month 6 | Secondary | Rituximab: 74% (98/132); estimated probability 0.78 (95% CI 0.70-0.85) · Ocrelizumab: 77% (65/84); estimated probability 0.79 (95% CI 0.71-0.88) | |||
| No new/enlarging T2 lesions from baseline to month 24 | Secondary | Rituximab: 71% (94/132); estimated probability 0.80 (95% CI 0.66-0.93) · Ocrelizumab: 73% (61/84); estimated probability 0.79 (95% CI 0.66-0.91) | |||
| Change in brain volume, baseline to month 24 (%) | Secondary | Rituximab: -0.70 ± 2.20 · Ocrelizumab: -1.75 ± 2.12 | |||
| Change in brain volume, month 6 to month 24 (%) | Secondary | Rituximab: -0.42 ± 1.77 · Ocrelizumab: -1.47 ± 1.79 | |||
| Estimated annualized relapse rate over 24 months | Secondary | Rituximab: 0.09 (95% CI 0.00-0.18) · Ocrelizumab: 0.04 (95% CI 0.00-0.09) · Mean treatment difference: 0.05 (95% CI -0.02 to 0.12) | |||
| Freedom from relapse over 24 months | Secondary | Rituximab: 92% (121/132); estimated probability 0.92 (95% CI 0.85-0.99) · Ocrelizumab: 94% (79/84); estimated probability 0.95 (95% CI 0.89-1.00) | |||
| Confirmed disability progression at month 24 (sustained ≥6 months, confirmed at month 30) | Secondary | Rituximab: 3% (4/132); estimated probability 0.03 (95% CI 0.00-0.07) · Ocrelizumab: 7% (6/84); estimated probability 0.08 (95% CI 0.02-0.14) · Risk difference: -4.5 pp (95% CI -11.3 to 2.2) | |||
| Confirmed disability improvement at month 24 (sustained ≥6 months) | Secondary | Rituximab: 29% (38/132); estimated probability 0.28 (95% CI 0.16-0.40) · Ocrelizumab: 24% (20/84); estimated probability 0.22 (95% CI 0.11-0.33) | |||
| Worsening of SDMT score from baseline to month 24 | Secondary | Rituximab: 2% (3/132); estimated probability 0.02 (95% CI 0.00-0.06) · Ocrelizumab: 4% (3/84); estimated probability 0.03 (95% CI 0.00-0.09) | |||
| Any infection | Safety | Rituximab: 82% · Ocrelizumab: 69% | |||
| Serious adverse events | Safety | Rituximab: 8% · Ocrelizumab: 7% | |||
| Infections | Adverse | Rituximab 82% vs Ocrelizumab 69% | |||
| Serious adverse events | Adverse | Rituximab 8% vs Ocrelizumab 7% | |||
| Prespecified safety end points | Adverse | Serious adverse events, infusion-related reactions, infections, and cancers | |||
Subgroup Analysis
Sensitivity analyses were consistent with primary result: Newcombe hybrid risk difference -2.6 pp (95% CI -9.3 to 5.6); per-protocol risk difference 2.3 pp (95% CI -6.0 to 10.6); complete-case risk difference -2.5 pp (95% CI -10.6 to 5.6)
Criticisms
- Population limited to Norway and Sweden — generalizability to more diverse populations uncertain
- Implementation error in randomization at some sites (initial 2:1 allocation reversed) required corrective adjusted 9:1 allocation lists
- Unequal group sizes (3:2 allocation) reduces power for ocrelizumab arm
- Brain volume outcomes analyzed only descriptively due to missing data
- No formal multiplicity adjustment for secondary end points — should not be interpreted as causal
- Noninferiority margin of -10 percentage points may be considered generous
- 24-month follow-up limits long-term safety and efficacy conclusions
Funding
Research Council of Norway and others; trial medications supplied and funded by participating hospitals as part of routine clinical procurement; no commercial involvement
Based on: OVERLORD-MS (New England Journal of Medicine, 2026)
Authors: Torkildsen Ø, Brustad HK, Høgestøl EA, ..., for the OVERLORD-MS Investigators
Citation: N Engl J Med 2026;395:44-53
Content summarized and formatted by NeuroTrials.ai.