The study, published as a letter to the editors in the Journal of Neurology, comes from the Neurology Unit and Multiple Sclerosis Center at IRCCS Ospedale San Raffaele in Milan, with co-first authors Veronica Marino and Lucia Moiola and co-senior authors Simone Guerrieri and Massimo Filippi. The team conducted a retrospective, real-world comparison of patients with multiple sclerosis who discontinued natalizumab and were subsequently switched either to ocrelizumab, an intravenous anti-CD20 monoclonal antibody, or to ofatumumab, a subcutaneously injected antibody of the same class. The work was approved by the institutional ethics committee at San Raffaele and conducted in accordance with the Helsinki Declaration, with patients providing dedicated informed consent for inclusion in the Italian Multiple Sclerosis and Related Disorders Register.

To understand why this comparison matters, it helps to look at the biology. Both ocrelizumab and ofatumumab belong to a class of drugs that target the CD20 molecule, a protein found on the surface of most B cells, the immune cells that are central players in the inflammatory attacks of relapsing multiple sclerosis. By binding CD20, these antibodies deplete circulating B cells, quieting the autoimmune storm. Yet the two drugs are not identical twins. Ocrelizumab is delivered by intravenous infusion every six months and is approved for both relapsing and primary progressive forms of the disease. Ofatumumab is self-injected under the skin once a month and binds a slightly different region of the CD20 molecule, a difference in epitope that influences how the antibody interacts with its target and with the immune system more broadly.

Those pharmacological distinctions translate into different clinical profiles. Ofatumumab’s subcutaneous route produces rapid and sustained B-cell depletion with a more continuous exposure, while ocrelizumab’s intermittent infusions create a pulsatile pattern of depletion and repopulation. Previous research, including pharmacokinetic and pharmacodynamic analyses of the pivotal OPERA and ORATORIO trials for ocrelizumab and the phase 2 APLIOS study of subcutaneous ofatumumab, has documented these characteristics. A recent comparative analysis published in Annals of Neurology in 2025 suggested that treatment outcomes can differ between patients receiving the two agents, adding urgency to the question of which drug is the better landing spot for patients stepping down from natalizumab.

The natalizumab discontinuation problem is well known to specialists. Natalizumab works by blocking the trafficking of immune cells across the blood-brain barrier, and when it is stopped, that barrier traffic resumes, sometimes with a vengeance. Systematic reviews and meta-analyses have documented post-natalizumab disease reactivation, including severe rebound phenomena in which patients experience bursts of new relapses and magnetic resonance imaging activity within weeks to months of the last infusion. Risk factors for reactivation after cessation have been identified in large cohort studies, and the timing of the switch is critical. A prolonged gap between stopping natalizumab and starting the next therapy leaves a window in which the disease can flare, while certain successor drugs, particularly those with slower onset, may not close that window quickly enough.

Against that backdrop, anti-CD20 agents have emerged as favored switch options, because they act on a different arm of the immune response and can maintain disease control across the transition. Several earlier studies, including pilot work on switching to ocrelizumab from natalizumab given at extended interval dosing in patients at high risk of progressive multifocal leukoencephalopathy, have reported favorable outcomes over follow-up periods as long as 96 weeks. But head-to-head real-world data comparing ocrelizumab and ofatumumab specifically in the post-natalizumab setting had been lacking, leaving clinicians to extrapolate from trials that enrolled different populations. The Milan study was designed to fill that gap using the kind of messy, unselected patient data that randomized trials rarely capture.

The retrospective design has both strengths and limits that readers should keep in mind. On the strength side, real-world cohorts reflect the full spectrum of patients seen in clinic, including those with comorbidities, atypical disease courses, and prior treatment histories that trial populations often exclude. The San Raffaele center is one of the largest multiple sclerosis referral hubs in Italy, and its registry-based approach allows researchers to track outcomes such as relapse rates, magnetic resonance imaging activity, and safety events under routine clinical conditions. On the limit side, retrospective comparisons cannot fully control for why clinicians chose one drug over the other, and such selection bias can color the results. The authors acknowledge the inherent constraints of this design, and the findings should be read as hypothesis-generating evidence that complements, rather than replaces, randomized trials.

Safety is a central concern in any switch strategy. Anti-CD20 therapies deplete B cells broadly, and while they have accumulated a substantial safety record since ocrelizumab’s approval, questions remain about infection risk, hypogammaglobulinemia over the long term, and the theoretical concern that B-cell depletion could modulate the immune response to the JC virus itself. The risk stratification for progressive multifocal leukoencephalopathy relies on anti-JC virus antibody testing using second-generation assays such as the STRATIFY JCV DxSelect, which supports clinical decisions about when natalizumab must be abandoned. The Milan team evaluated safety alongside efficacy in both switch groups, providing clinicians with a side-by-side view of how patients tolerate each transition in everyday practice.

The broader significance of the study lies in the evolution of multiple sclerosis treatment. The field has moved decisively toward high-efficacy therapies earlier in the disease course, and natalizumab, once a flagship of that movement, is increasingly managed with extended interval dosing to reduce infection risk before ultimately being withdrawn in patients who accumulate risk factors. As the population of patients needing a post-natalizumab home grows, the choice between intravenous ocrelizumab and subcutaneous ofatumumab becomes a practical fork in the road, influenced by efficacy signals, safety profiles, dosing convenience, patient preference, and health system logistics. Real-world comparisons like this one give neurologists evidence grounded in clinical reality rather than idealized trial conditions.

For patients and clinicians alike, the message from Milan is one of cautious reassurance paired with continued vigilance. Switching from natalizumab to either anti-CD20 agent appears to be a viable strategy for maintaining disease control, but the details of timing, monitoring, and individual risk profiling remain decisive. The study’s data are held at IRCCS Ospedale San Raffaele and are available upon reasonable request to the corresponding author, an approach that supports transparency and further analysis by the research community. As longer follow-up accumulates and more centers contribute their own switch cohorts, the evidence base will mature, helping to convert one of neurology’s trickiest transitions from an act of clinical judgment into an evidence-based protocol.

Subject of Research: Comparison of ocrelizumab and ofatumumab as post-natalizumab switch therapies in multiple sclerosis

Article Title: Ocrelizumab versus ofatumumab efficacy and safety after natalizumab discontinuation in multiple sclerosis: a retrospective real-world comparison

Article References: Marino, V., Moiola, L., Zanetta, C., Rubin, M., Rossi, L., Margoni, M., Nozzolillo, A., Gattuso, I., Viti, V., Petza, E., Rocca, M. A., Guerrieri, S., & Filippi, M. (2026). Ocrelizumab versus ofatumumab efficacy and safety after natalizumab discontinuation in multiple sclerosis: a retrospective real-world comparison. Journal of Neurology, 273(10), Article 577. https://doi.org/10.1007/s00415-026-14124-1

Image Credits: AI Generated

DOI: 10.1007/s00415-026-14124-1

Keywords: multiple sclerosis, natalizumab, ocrelizumab, ofatumumab, anti-CD20 therapy, B-cell depletion, progressive multifocal leukoencephalopathy, JC virus, drug switching, real-world evidence, disease reactivation, neuroimmunology

The research team, led by Antonio Esposito and Marcello Moccia of Federico II University of Naples, followed 34 adults with multiple sclerosis who started ofatumumab and remained on treatment for at least two years, with paired immune assessments at baseline and follow-up. Ofatumumab is administered as 20 milligram subcutaneous injections at weeks 0, 1, and 2, and monthly thereafter. Unlike intravenous anti-CD20 agents such as ocrelizumab and rituximab, ofatumumab acts primarily in the peripheral lymphatic compartment, and its distinct epitope binding and complement-dependent cytotoxicity have fueled speculation that its broader immunological footprint might differ from its intravenous cousins.

To map that footprint, the investigators used a standardized whole-blood flow-cytometry workflow. Fifty-microliter blood samples were incubated with fluorochrome-conjugated antibody mixtures, red blood cells were lysed, and the remaining nucleated cells were acquired on a Beckman Coulter Navios cytometer equipped with 405-, 488-, and 638-nanometer lasers. Sequential gating identified the major lymphocyte lineages: CD3-positive T cells and their CD4 and CD8 subsets, CD19- and CD20-positive B cells, CD56-positive natural killer cells, CD27-positive populations, memory B cells carrying both CD27 and CD19, and the small CD3-positive CD20-positive T-cell subset that anti-CD20 therapy is known to eliminate. Absolute counts were obtained using a double-platform approach under routine quality-control standards, including instrument calibration and participation in the UK-NEQAS external quality assessment scheme.

The longitudinal analysis relied on linear mixed-effects regression, with fixed effects for age, sex, follow-up duration, comorbidities, body mass index, and previous disease-modifying therapy, and a random intercept for each participant. The results were striking. Over roughly two years of treatment, total lymphocyte counts rose by an average of about 251 cells per microliter, while CD3-positive T cells increased by roughly 369, CD4-positive T cells by about 240, CD8-positive T cells by about 132, CD27-positive cells by about 275, and CD27-positive CD3-positive T cells by about 452. All of these increases were statistically significant, with confidence intervals that excluded the null. In parallel, the expected depletion was unmistakable: CD19-positive and CD20-positive B cells fell by about 182 cells per microliter, and CD27-positive CD19-positive memory B cells dropped sharply as well.

Not everything moved, however. The ratio of CD4 to CD8 T cells remained stable, as did CD56-positive natural killer cells and the CD3-positive CD20-positive T-cell subset. That stability of the CD4-to-CD8 ratio is important, because it suggests that no major rebalancing occurred between the two broad arms of the T-cell compartment, even as absolute T-cell numbers climbed. The authors are careful to note that the rise in CD27-positive cells cannot be taken as evidence of naive T-cell expansion. CD27 is expressed across both naive and central memory T-cell compartments and also appears on subsets of natural killer and natural killer T cells. Without additional differentiation markers such as CD45RA, CD45RO, and CCR7, the specific CD27-expressing population driving the increase cannot be resolved with the panel used.

What, then, explains the T-cell rise? The authors propose that the pattern may reflect broad reorganization of the T-cell compartment following sustained B-cell depletion, a phenomenon that has been hinted at in other anti-CD20 studies. Previous short-term real-world investigations of ofatumumab generally found CD3, CD4, and CD8 counts to be broadly stable, which makes the two-year increases reported here particularly noteworthy. The discrepancy may be a matter of timing: early immune kinetics differ from the intermediate and long-term reorganization that unfolds as the immune system adapts to life without circulating CD20-positive B cells. Other variables complicate the picture, including prior exposure to therapies such as natalizumab or fingolimod, which reshape peripheral immune pools and trafficking patterns, and differences in measurement approaches, such as whether studies use surface-epitope or intracellular CD20 staining, the latter avoiding steric hindrance by therapeutic antibodies still circulating in the blood.

The study also looked for clinical meaning behind these immune shifts. Using multivariable logistic regression, the team tested whether the annualized percentage change in each laboratory measure predicted relapses, magnetic resonance imaging activity, progression on the Expanded Disability Status Scale, improvement on that scale, or the composite outcome of no evidence of disease activity, known as NEDA-3. The answer, in every case, was no. No significant associations emerged between changes in any lymphocyte population and any clinical or radiological outcome. Several models could not be reliably estimated at all because clinical and MRI events were so rare, a consequence of treating patients with a highly effective therapy, and because near-complete B-cell depletion left little variability to model.

The authors are refreshingly candid about what these negative findings do and do not mean. With only 34 participants and a low frequency of disease-activity events, the study was never powered to detect subtle immunological-clinical relationships, and the absence of statistical significance should not be read as evidence that such relationships do not exist. The analyses were explicitly designed to be hypothesis-generating rather than to establish mechanistic pathways or predictive biomarkers. Most participants had prior exposure to other disease-modifying therapies, and although previous treatment was included as a covariate, residual carry-over effects on immune composition cannot be excluded. There was also no parallel comparator group receiving ocrelizumab or rituximab, which limits direct inference about how ofatumumab’s subcutaneous route and pharmacology might differ from intravenous anti-CD20 agents, and no standardized intermediate sampling at six and twelve months to reconstruct the early-to-late kinetics of immune change.

Those limitations point directly toward the next generation of experiments. The immunophenotyping panel used here captured broad lymphocyte populations but could not resolve the smaller, functionally critical subsets that other work has implicated in anti-CD20 responses, including Th17 and Th17.1 cells, T follicular helper cells, and their regulatory counterparts. Cross-sectional research during ofatumumab treatment has documented improved Th17.1-to-regulatory-T-cell and T follicular helper-to-follicular regulatory T-cell ratios, reduced pro-inflammatory cytokine production, and impaired migration of residual CD20-positive T cells across an inflamed model of the blood-brain barrier. A multicenter comparison of ocrelizumab and ofatumumab found more pronounced reductions of B and CD8-positive cells with ocrelizumab, while total lymphocytes, CD4 cells, and natural killer cells remained similar between the two drugs. Functional readouts such as cytokine profiling, migration assays, standardized immunoglobulin tracking, and assessments of vaccine responses and infection risk would round out the safety picture.

For now, the study delivers a clear and reassuring headline finding: two years of ofatumumab produces sustained, near-complete depletion of circulating CD19- and CD20-positive B cells, exactly as its mechanism of action predicts, alongside a measurable and statistically robust expansion of broad T-cell compartments that remains clinically silent, at least as far as conventional outcome measures can detect. Whether that T-cell expansion represents a benign consequence of B-cell depletion, a compensatory homeostatic response, or a subtle repositioning of the immune system toward a more anti-inflammatory profile remains an open question. Answering it will require larger prospective cohorts, repeated longitudinal sampling, deeper immunophenotyping panels, and the functional assays that can translate cell counts into biological meaning. Until then, this two-year window into the immune life of ofatumumab-treated patients stands as a valuable benchmark for what long-term B-cell depletion actually looks like in the blood of people living with multiple sclerosis.

Subject of Research: Longitudinal immune profiling of lymphocyte subsets in multiple sclerosis patients treated with the anti-CD20 antibody ofatumumab

Article Title: Ofatumumab in multiple sclerosis: 2-year immune profiling and clinical correlates

Article References: Esposito, A., Corsaro, L., Cave, I. D., Capasso, F., Corsini, G., Matrone, G., Cerbone, V., Petracca, M., Carotenuto, A., Lanzillo, R., Scalia, G., Castaldo, G., Morra, V. B., & Moccia, M. (2026). Ofatumumab in multiple sclerosis: 2-year immune profiling and clinical correlates. Journal of Neurology, 273(10), Article 579. https://doi.org/10.1007/s00415-026-14088-2

Image Credits: AI Generated

DOI: 10.1007/s00415-026-14088-2

Keywords: multiple sclerosis, ofatumumab, anti-CD20, B-cell depletion, T cells, flow cytometry, immune profiling, lymphocytes, NEDA-3, clinical outcomes, immunotherapy, Journal of Neurology