Off-label rituximab matched ocrelizumab in newly diagnosed MS trial

An old, cheap antibody used off-label for multiple sclerosis held its own against the expensive, purpose-approved one in the first head-to-head trial of the two, published in the New England Journal of Medicine [s1]. In OVERLORD-MS, the estimated probability of having no new or enlarging brain lesions between months 6 and 24 was 92.2% with rituximab and 94.8% with ocrelizumab — a gap small enough to meet the trial's prespecified definition of noninferiority [s1].

The question the trial settles is one clinicians and payers have argued over for a decade. Ocrelizumab, an anti-CD20 antibody designed and licensed for MS on the strength of trials such as OPERA, which showed it cut relapse rates against interferon beta-1a, is a branded drug carrying a branded price [s2]. Rituximab, an older anti-CD20 antibody built for lymphoma and rheumatoid arthritis and long available as low-cost biosimilars, works by the same mechanism — depleting the B cells that drive MS — and has been used off-label for the disease for years without a randomised comparison to back the practice [s1].

What the trial did

OVERLORD-MS, funded by the Research Council of Norway among others, was a phase 3, multicentre, double-blind noninferiority trial in adults with newly diagnosed relapsing MS and recent disease activity [s1]. Participants were randomly assigned in a 3:2 ratio to rituximab or ocrelizumab, each given every 6 months for 24 months [s1]. Of 218 who underwent randomisation, 216 received treatment: 132 assigned to rituximab and 84 to ocrelizumab [s1].

The primary endpoint was the absence of new or enlarging lesions on T2-weighted MRI from month 6 to month 24 — a sensitive imaging marker of ongoing inflammation that turns over faster than clinical relapses do [s1]. Noninferiority was defined in advance as the lower limit of the 95% confidence interval for the risk difference, rituximab minus ocrelizumab, staying above −10 percentage points [s1]. That margin is the trial's central judgment call: it declares in advance how much MRI activity the field is willing to tolerate in exchange for whatever rituximab offers.

What it found

The estimated probability of no new or enlarging T2 lesions was 92.2% with rituximab and 94.8% with ocrelizumab, a risk difference of −2.6 percentage points (95% confidence interval −9.4 to 4.3) [s1]. Because the lower bound of that interval, −9.4, stayed above the −10-point margin, rituximab was declared noninferior [s1]. Relapse rates, disability outcomes, and cognitive-performance profiles appeared similar in the two groups [s1].

Safety was not identical. Infections were more common with rituximab than with ocrelizumab, reported in 82% versus 69% of participants, though the proportion with serious adverse events was similar — 8% with rituximab and 7% with ocrelizumab [s1].

How to read it

"Noninferior" is a claim bounded by the margin, not a claim of equivalence. The result says rituximab is not worse than ocrelizumab by more than 10 MRI-defined percentage points, and the observed difference was a small 2.6 points in ocrelizumab's favour [s1]. The confidence interval crosses zero, so the trial cannot rule out that the two are genuinely equal, nor that ocrelizumab holds a modest edge — it rules out only a large disadvantage for rituximab.

Two limits are worth naming. This was a comparatively small trial followed for two years, so it is powered for an MRI marker rather than for the long-run disability accrual that patients care about most, and the higher infection rate with rituximab, while it did not translate into more serious events here, is the kind of signal that matters more over years than over months [s1]. The MRI endpoint is a validated stand-in for disease control, but it is a stand-in.

Why it matters

Head-to-head trials of a generic-priced option against a branded one are rare, because no manufacturer has an incentive to run them. OVERLORD-MS is one of the few, and it lands in a disease where the two drugs differ enormously in cost but barely in mechanism [s1][s2]. The economics echo debates covered elsewhere on this site, from why biosimilar and reference biologics perform alike to why US drug prices sit so far above other countries'. A separate class of MS drugs is also advancing, including the BTK inhibitor tolebrutinib now under European review, so anti-CD20 antibodies are not the only option — but they are the one where a cheap alternative already exists.

What to watch

The near-term effect is likely to be felt in guidelines and formularies rather than at the bench: whether MS treatment recommendations begin to name rituximab as an evidence-backed alternative rather than an off-label improvisation, and how payers respond when a randomised trial supports the cheaper drug [s1]. Longer follow-up will test whether the two-year MRI parity holds up on disability and whether the infection difference grows with continued dosing [s1].

This article describes research and is not medical advice. Decisions about MS treatment are for patients and their treating clinicians.

Sources

• Rituximab versus Ocrelizumab in Newly Diagnosed Relapsing Multiple Sclerosis — New England Journal of Medicine, 1 July 2026
• Ocrelizumab versus Interferon Beta-1a in Relapsing Multiple Sclerosis — New England Journal of Medicine, 21 December 2016