Breaking the Cycle: The Fight Against Relapses in Autoimmune Encephalitis

Abstract

Rituximab Use for Relapse Prevention in Anti-NMDAR Antibody-Mediated Encephalitis. A Multicenter Cohort Study

N. Seery, R. Wesselingh, P. Beech, L. McLaughlin, T. Rushen, A. Halliday, L Ter Horst, S. Griffith, M. Forcadela, T. Tan, C. Kazzi, C. Nesbitt, J. Broadley, K. Buzzard, A. Duncan, W. D'Souza, Y. Tran, A. Van Der Walt, G. Skinner, B. Taylor, A. Swayne, A. Brodtmann, D. Gillis, E. Gerard Butler, T. Kalincik, U. Seneviratne, R. Macdonell, S. Blum, S. Ramanathan, C. Malpas, S. Reddel, T. Hardy, T. O’Brien, P. Sanfilippo, H. Butzkueven, M. Monif for the Australian Autoimmune Encephalitis Consortium.

Background and Objectives: Rituximab is an anti-CD20 monoclonal antibody used in patients with anti-NMDAR antibody (Ab)-mediated encephalitis as both an acute escalation therapy and a longer-term relapse risk–reduction treatment. The potential long-term benefit of a single course administered during the acute disease phase on future relapse risk is uncertain. Moreover, the optimal dosing duration to reduce relapse risk is unknown. The aim of this study was to evaluate the effect of a single course of rituximab on relapse incidence. We also studied the duration of effect of a course of rituximab in adult patients with anti-NMDAR Ab-mediated encephalitis.

Methods: We recruited 67 patients with anti-NMDAR Ab-mediated encephalitis from 10 Australian hospitals. Rituximab exposure was quantified as a time-varying covariate in Cox proportional hazard models.

Results: A single course of rituximab was associated with longer time to first relapse (hazard ratio [HR]: 0.11, 95% CI: 0.02–0.70, p = .02). For patients in whom redosing is considered, rituximab was associated with longer time to first relapse at 6 months after the last infusion, after adjusting for concurrent immunotherapies and the presence of ovarian teratoma at disease onset (HR: 0.05, 95% CI: 0.00–0.48, p = .005). The treatment effect did not persist out to 12 months after a given course (HR: 0.60, 95% CI: 0.15–2.44, p = .47).

Discussion: A single course of rituximab reduces the risk of relapse of anti-NMDAR antibody-mediated encephalitis. In select patients for whom redosing of rituximab is considered, administration at 6 months delays relapses.

Classification of Evidence: This study provides Class IV evidence that rituximab delays relapses in patients with anti-NMDAR antibody-mediated encephalitis.

Acute and Long-Term Immune-Treatment Strategies in Anti-LGI1 Antibody-Mediated Encephalitis: A Multicenter Cohort Study

N. Seery, R. Wesselingh, P. Beech, L. McLaughlin, T. Rushen, A. Halliday, L. Horst, S. Griffith, M. Forcadela, T. Tan, C. Kazzi, C. Nesbitt, J. Broadley, K. Buzzard, A. Duncan, W. D'Souza, Y. Tran, A. Van Der Walt, G. Skinner, B. Taylor, A. Swayne, A. Brodtmann, D. Gillis, E. Butler, T. Kalincik, U. Seneviratne, R. Macdonell, S. Blum, S. Ramanathan, C. Malpas, S. Reddel, T. Hardy, T. O'Brien, P. Sanfilippo, H. Butzkueven, M. Monif for the Australian Autoimmune Encephalitis Consortium.

Background and Objectives: Few studies have evaluated acute immunotherapy and relapse prevention strategies in patients with anti–leucine–rich glioma-inactivated 1 (LGI1) antibody (Ab)-mediated encephalitis. The objective of this study was to analyze the outcomes of acute and long-term immunotherapy strategies in this population.

Methods: We undertook a multicenter cohort study of 55 patients with anti-LGI1 Ab-mediated encephalitis, either recruited prospectively or identified retrospectively from 10 Australian hospitals as part of the Australian Autoimmune Encephalitis Consortium. Clinical data were collected, including treatment durations of all relevant immunotherapies. Clinical outcomes that we examined included (1) time to first clinical relapse, (2) improvement on modified Rankin Scale (mRS), and (3) favorable binary composite clinical-functional outcome at 12 months. A favorable outcome was defined as fulfilling all three of mRS less than 3, a score of 1 or less in the memory dysfunction component of the Clinical Assessment Scale in Autoimmune Encephalitis, and absence of drug-resistant epilepsy.

Results: Rituximab, adjusted for concomitant use of other immunotherapies, was associated with increased time to first relapse (hazard ratio: 0.10; 95% CI: 0.001–0.85; p = .03). Intravenous pulsed methylprednisolone was associated with an improvement in mRS (OR: 4.48; 95% CI: 1.03–21.3; p = .048) and a favorable composite clinical-functional outcome (OR: 4.96; 95% CI: 1.07–27.2; p = .049) at 12 months.

Discussion: Rituximab may be effective at preventing relapses in patients with anti-LGI1 Ab-mediated encephalitis. Acute methylprednisolone treatment may be associated with favorable outcomes at 12 months.

Classification of Evidence: This study provides Class IV evidence that for patients with anti-LGI1 Ab-mediated encephalitis, rituximab prevents relapses and acute methylprednisolone is associated with favorable outcomes at 12 months.

Commentary

Autoimmune encephalitis (AE) is an immune-mediated inflammation of the cortex or brain deep gray matter, associated with antibodies targeting neuronal receptors or related cell-surface proteins. The AE typically arises in previously healthy individuals and presents as either a paraneoplastic or primary autoimmune disorder. Anti-NMDA receptor (NMDAR) encephalitis, the most prevalent AE syndrome, predominantly affects young individuals and occurs at rates exceeding those of viral encephalitis in this age group. Clinically, it manifests as subacute cognitive decline, psychosis, dyskinesia, and new-onset seizures. Seizures occur in up to 89% of patients, with up to 41% developing status epilepticus (SE) during the acute phase.^1–3 Leucine-rich glioma-inactivated 1 (LGI1) antibody-associated encephalitis is the second most common AE subtype, distinguished by prominent cognitive dysfunction and seizures. Seizures are often the initial and most dominant symptom, reported in 83–100% of patients across various cohorts.^3–5 Focal seizures may occur with or without impaired awareness and can include faciobrachial dystonic seizures (FBDS).^4,5 Status epilepticus is less common than in anti-NMDAR encephalitis, occurring in approximately 22–34% of patients. ³ The chronic phase of AE is characterized by epilepsy and enduring cognitive impairments, which may persist for years following initial recovery and significantly contribute to long-term neurological disability. Disease relapses tend to present with milder symptoms than initial episodes; however, they also contribute to overall morbidity. Recent cohort studies report relapse rates of 7–73% in patients with anti-NMDAR and anti-LGI1 encephalitis, reflecting variability across geographic regions and patient populations.^6,7

The immunopathological findings in anti-NMDAR and anti-LGI1-associated encephalitis reveal the presence of T-cell and B-cell infiltrates and abundant interstitial and perivascular antibody-secreting plasma cells. B-cells play a complex role in the context of AE, contributing to the production of autoantibodies and triggering immune injury via antibody-dependent cell-mediated cytotoxicity. Additionally, B cells play an important role by regulating the release of inflammatory mediators. Relapses of AE may occur as a result of incomplete immunological control of pathogenic B-cells or plasma cell clones due to the delayed or insufficient immunomodulatory therapy, or may be caused by recurrent exposure to antigens. Rituximab, a chimeric anti-CD20 monoclonal antibody, is directed against circulating and lymphoid tissue-associated B cells, is considered the second-line immunotherapy in AE. The induction course of rituximab consists of a single dose or closely timed repeated doses; this can be followed by the maintenance therapy to delay or prevent the disease relapses. The main therapeutic effect of rituximab is derived from rapid and sustained reduction of B cells, leading to reduction of autoantibody levels and attenuation of the pro-inflammatory state. Plasma cells do not express CD20; therefore, the effectiveness of rituximab in an acute and relapsing disease may vary.

Although the benefits of early immunotherapy are well recognized, the definitive role of rituximab in post-acute management of AE has not been established. Seery and colleagues have made important advancement in this field. ⁸ The authors conducted a multicenter cohort study of adults with confirmed anti-NMDAR encephalitis, evaluating seizure burden and other acute symptoms during hospitalization and assessing relapse rates longitudinally following immunotherapy. Their study particularly stands out for its use of rituximab as a time-varying covariate in regression models, allowing the analysis of relapse latency relative to treatment timing. Their findings were encouraging: over 70% of patients received rituximab acutely, and more than 70% remained relapse-free over a median follow-up of nearly four years, a duration longer than in prior studies. Importantly, a single treatment course of rituximab was associated with an approximately 90% reduction in relapse risk as it delayed recurrence by about six months independently of other immunotherapies. However, this protective effect diminished after 12 months, suggesting that sustained protection may require ongoing therapy or timed redosing with this agent. These results support the therapeutic value of early rituximab and underline six months as a potential optimal interval for redosing to maintain disease control in anti-NMDAR encephalitis. Interestingly, although seizures occurred in over 60% of patients during the acute phase, they represented the initial manifestation of relapse in only one case, suggesting that rituximab may have disease-modifying effects in AE-associated seizures.

In a related study, Seery and colleagues applied a comparable patient selection strategy and examined the outcomes of acute and long-term immunotherapy in anti-LGI1-associated encephalitis. ⁹ They highlighted the key clinical features of anti-LGI1 encephalitis, including the predominance of FBDS and its frequent occurrence in older adults. Interestingly, only about half the patients received second-line immunotherapy, consistent with previous reports showing the responsiveness of this syndrome to first-line agents. The most compelling finding was that rituximab exposure correlated with a tenfold reduction in relapse risk. Overall, over two-thirds of patients with encephalitis remained relapse-free at the last follow-up, while among those who relapsed, 88% presented with seizures. The observation that some patients relapsed more than 12 months after a rituximab course suggested that repeated dosing and close monitoring of B-cell depletion maybe required for sustained relapse prevention. The authors also highlighted the higher efficacy of pulsed methylprednisolone over intravenous immunoglobulin (IVIG) for the long-term functional outcomes, as measured by the modified Rankin Scale (mRS). Notably, steroids promoted meaningful functional and clinical improvement when they were started within three months of symptom onset. These findings suggest that beyond alleviating seizures, corticosteroid treatment may confer lasting neurocognitive benefits. Although 80% of patients achieved a favorable mRS score at one year, this proportion declined to 65% when more sensitive measures of cognitive impairment and drug-resistant seizures were applied. This gap reflects a growing recognition that mRS and other conventional disability scales may not reflect the true morbidity burden that is compounded by cognitive decline, fatigue, psychiatric symptoms, and epilepsy.

These studies collectively provide class IV evidence supporting the efficacy of rituximab in reducing relapse risk in anti-NMDAR and anti-LGI1-associated encephalitis. This reinforces findings from previous smaller studies and provides practical guidance on incorporating rituximab into immunotherapy regimens. These studies applied an elegant methodological approach to assessing relapse prevention. Further, they enrolled moderately large patient cohorts with long-term follow-up and systematically recorded the use of immunotherapeutic agents. The limitations include reliance on existing AE registries and more extensive post-acute surveillance in relapsing patients, which may introduce selection bias. Further, the lack of continuous EEG monitoring may have led to underreporting of nonconvulsive SE or missed seizure relapses, particularly in anti-NMDAR cohort. Overall, these findings provide valuable guidance for clinical decision-making, where practitioners must weigh effective relapse prevention against the risks of prolonged immunosuppression. The findings also emphasize the need for comprehensive clinical and neurophysiological monitoring to fully capture disease burden and optimize long-term management.