Rituximab for severe immune checkpoint inhibitor-related adverse events: a multicenter case series

Abstract

Background

Immune checkpoint inhibitors (ICIs) can induce severe immune-related adverse events (irAEs) that are life-threatening and may be refractory to standard immunosuppressive therapy. Rituximab has emerged as a potential treatment option, but evidence is limited.

Methods

We conducted a retrospective multicenter study including patients who received rituximab for severe ICI-related irAEs between 2018 and 2023. Clinical characteristics, type and severity of irAEs, previous immunosuppressive regimens, outcomes after rituximab, and oncologic status were collected. Overall survival was estimated from rituximab initiation.

Results

Eighteen patients were identified (median age 66 years, 55% male). irAEs were mainly neurological syndromes (n = 7), myositis and myocarditis (n = 6), and interstitial lung disease (n = 3). All events were grade 3–4 according to common terminology criteria for adverse events v5.0 and refractory to corticosteroids, often combined with additional immunosuppressants. Rituximab was administered as second- or third-line therapy. Clinical improvement was observed in 83% of patients, including complete resolution in 39%. After a median follow-up from rituximab initiation of 9.7 months (interquartile range [IQR] 5.1–23.9), seven patients (39%) died, but due to irAE only for three patients (17%).

Conclusions

Rituximab appears to be an effective therapeutic option for selected severe ICI-related irAEs, particularly when conventional immunosuppressive regimens fail. These real-world data highlight the potential role of B-cell-targeted therapy in the management of complex irAEs and support further evaluation in prospective studies.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00262-026-04325-y.

Introduction

Immune checkpoint inhibitor (ICI) therapies targeting programmed death (ligand)-1 (PD-1/PD-L1) and cytotoxic T lymphocyte antigen-4 (CTLA-4) have become the standard of care for many cancers. This novel class of immunotherapy drugs works by blocking intrinsic negative regulators of the immune system, thus increasing immune activity and enhancing antitumor responses. Increased T-cell activation also promotes inflammatory cytokines production and humoral immune responses, which may lead to autoimmune complications known as immune-related adverse events (irAEs) [1]. With the increased use of ICI therapies in cancer treatments, the frequency of irAEs and high-grade irAEs is rising.

Clinicians must be familiar with the clinical presentation and management of irAEs. Accordingly, clinical practice guidelines have been developed by the American Society of Clinical Oncology (ASCO) [2] and the European Society for Medical Oncology (ESMO) [3]. ICI treatment is generally continued with close monitoring for grade 1 toxicities, except for certain neurologic, hematologic, and cardiac events. Grade 2 toxicities may require temporary suspension and corticosteroids. Grade 3 warrants suspension and high-dose corticosteroids, while grade 4 usually leads to permanent discontinuation, except for controlled endocrinopathies.

Some refractory cases may require further immunosuppressive therapy. For such situations, recommendations are less stringent and the treatment is left to the discretion of the physician in accordance with his local guidelines. One alternative is rituximab, a chimeric anti-CD20 monoclonal antibody that targets B lymphocytes, without directly affecting T-cell-mediated antitumor immunity [4].

Rituximab has been increasingly shown to be effective in treating irAEs, such as vasculitis, encephalopathies, myasthenia, myositis, hematological disorders. Currently, data in line with the use of rituximab in the management of irAEs come from scarce case reports. Therefore, we aim to review case reports from our regional center experience, to better understand the use of rituximab in the management of ICI-related toxicity.

Patients and methods

Study design and data collection

We conducted a retrospective, observational, multicenter study across four centers (Montpellier University Hospital, Nîmes University Hospital, Montpellier Cancer Institute, Clémentville Clinic).We included patients ≥ 18 years old, treated with ICI for any cancer type, who received rituximab for a severe irAE between December 2018 and December 2023.

We reviewed the regional database (Occitanie East region’s Multidisciplinary Consultation Meeting), including all patients presented for irAEs, regardless of type and grade. Among patients presenting with immune-related adverse events, cases were identified through an automated extraction of electronic medical records using predefined keyword searches for rituximab exposure (“rituximab,” “RTX,” “RITUX,” or “MABTHERA”).

Data on epidemiological, autoimmunity history, cancer characteristics, ICI therapy, irAE details, management, and outcomes were extracted from electronic medical reports.

Objectives and outcomes

The main objectives are to describe the types of ICI-related toxicities treated with rituximab and assess their outcomes. Complete and partial responses were defined according to previously published criteria (Supplementary material). As secondary objective, we evaluate cancer prognosis after rituximab administration in patients with irAEs. Overall survival (OS) was defined as the time from rituximab initiation to death from any cause.

Statistical analysis

Descriptive statistics are presented as medians (IQR) for quantitative variables and frequencies (number and percentages) for qualitative variables. We used the Kaplan–Meier method to estimate overall survival (OS) from rituximab initiation. Statistical analysis was performed with EasyMedStat (version 3.42; www.easymedstat.com).

Ethics

The study was approved by the local ethic committee (approval number 202100908) prior to any data collection.

Results

During the inclusion period, 794 patients presented for immunotherapy-mediated toxicity, among which 81 patients were identified. Of these, 18 patients actually received rituximab as treatment of their irAEs and were analyzed. The baseline characteristics, main toxicities characteristics, and management are presented in Table 1. Fifty-five percent of patients were male. ICIs were mostly used for treating lung cancer (61%) and skin cancer (22%). All patients received PD-1/PD-L1 inhibitors, either alone (89%) or with a CTLA-4 inhibitor (11%). Among patients with preexisting autoimmune disease, only one patient was receiving baseline immunosuppressive therapy (methotrexate for rheumatoid arthritis).

Table 1.

Patients’ baseline characteristics and main toxicity characteristics and management

	Patients (n = 18)
Sex (n (%))
Male	10 (55.5%)
Age at toxicity onset (median (IQR), years)	64 (58.3–70.5)
History of autoimmune disease (n (%))a	5 (27%)
Cancer organ (n (%))
Lung	11 (61%)
Skin	4 (22%)
Othersb	3 (17%)
Stage of cancer (n (%))
Locally advanced	7 (39%)
Metastatic	11 (61%)
Immune checkpoint inhibitor (n (%))
Monotherapy anti-PD(L)1	16 (89%)
Combi-therapy anti-PD(L)1 + anti-CTLA-4	2 (11%)
Main toxicity (n (%))c
3M syndrome	3 (16.7%)
Myocarditis + Myositis	3 (16.7%)
Polyradiculoneuritis	3 (16.7%)
Interstitial lung disease	3 (16.7%)
Encephalitis	2 (11.1%)
Othersd	6 (33.3%)
Time from ICI initiation (median (IQR), days)	30 (15.75–112.25)
Toxicity management (n (%)c
ICI stopped	17 (94.4%)
Corticosteroids	18 (100%)
IVIG	10 (55.5%)
Plasmapheresis	8 (44.4%)
Other immunosuppressivee	5 (27.7%)
Time to first Rituximab infusion (median (IQR), days)	72 (35.75–212.75)
Main toxicity outcome (n, (%))
Remission	7 (38.9%)
Partial response	8 (44.4%)
Persistent	3 (16.7%)

^aGrave’s disease and psoriasis, Sjögren syndrome, chronic ulcerative colitis, rheumatoid arthritis, immune thrombocytopenia

^bCervical, kidney, oropharynx

^cResults may exceed 100% because patients could have experienced associated toxicities or received multiples treatments

^dBullous pemphigoid, immune thrombocytopenic purpura, myofasciitis, vasculitis, polyarthritis, autoimmune hemolytic anemia (n = 1 each)

^eMethotrexate (n = 1), Mycophenolate mofetil + methotrexate (n = 1), sirolimus (n = 1), abatacept + ruxolitinib (n = 1), Tocilizumab + methotrexate (n = 1)

Among irAEs requiring rituximab treatment, neurological toxicities, either isolated or associated with myocarditis and/or myositis, accounted for most cases (7/18 patients). These included three cases of myasthenia as part of “3M syndrome” (i.e., the association of myocarditis, myositis, and myasthenia). Three cases of polyradiculoneuritis were observed, including one associated with myocarditis and myositis, one with encephalitis, and one isolated. One additional case of encephalitis was reported. Myositis was reported in 6/18 patients; all in combination with myocarditis, including three cases of “3M syndrome” syndrome. The irAEs were mostly categorized as grade 4 (life-threatening) according to CTCAE v5.0 (13/18, 72.2%). In all cases, the ICI was stopped, except for one patient, for whom the toxicity appeared several months after the cessation of immunotherapy. All patients were treated with corticosteroids before receiving rituximab. Ten and eight patients received intravenous immunoglobulins (IVIG) and plasmapheresis respectively. All concomitant immunosuppressive treatments, except corticosteroids, were discontinued prior to rituximab initiation. All patients except one received two rituximab infusions administered 15 days apart.

Overall, a good clinical response was observed in 83.3% of patients after rituximab initiation: 7/18 (38.9%) remission and 8/18 (44.4%) partial improvement. The three patients with persistent irAE experienced a 3M syndrome for one, encephalitis for the second, and persistent immune thrombocytopenic purpura for the third patient. No significant differences in baseline characteristics were observed between complete responders and partial or non-responders. The additional Table 1 summarizes each patient’s characteristics.

Median duration of follow-up after rituximab initiation is 9.7 months (IQR 5.1–23.9). Seven patients died during follow-up, due to irAE for three patients, and oncologic progression for four patients. At 12 months, the OS was 61.1% (95% confidence interval 32.8–80.4). Figure 1 illustrates Kaplan–Meier analysis for overall survival.

Fig. 1.

Kaplan–Meier curve for overall survival from Rituximab initiation for severe irAE

Discussion

In our study, immunotoxicities requiring rituximab treatment were predominantly neurological, with patients experiencing central or peripheral neurological events. These results are consistent with a previous systematic literature review [5], which identified nine case reports among 1905 publications—all neurological presentation (five myasthenia gravis, one Miller Fisher syndrome, one relapsed multiple sclerosis, one sensory neuronopathy, and one neuro-Sjögren syndrome). The second most frequently involved organ systems were the cardiac and muscular systems. Indeed, 6/18 patients (33.3%) experienced myocarditis associated with myositis, among them are 3 cases of “3M syndrome.” Concerning myositis, sparse case reports are available in literature [6, 7]. This observation may be explained by the fact that many neurological and muscular autoimmune disorders are primarily B-cell mediated. This pathophysiological rationale supports the use of rituximab in refractory cases. Moreover, in refractory cases, physicians frequently turn to treatment strategies used for idiopathic autoimmune diseases, in which rituximab is commonly used as a second line therapy. In contrast, the use of rituximab for ICI-induced myocarditis has not been previously reported in the literature. Our study reported them for the first time. For myocarditis cases, the decision to manage them with rituximab could be driven by their association with neurological or muscular irAEs.

In our study, remission or partial improvement of irAEs was observed in 83.3% of patients, reflecting an effective response to rituximab treatment. Several case reports have shown rituximab’s efficacy in neurological, dermatological, renal, muscular, hematological irAEs [1, 4, 5, 8–10].

Pathophysiological mechanisms underlying irAEs are primarily driven by T-cell activation, leading to increased production of inflammatory cytokines. These observations support the use of corticosteroids and immunosuppressive agents targeting both T- and B-cell activity, which are the therapeutic class recommended in ASCO and ESMO Guidelines as first-line therapy for management of irAEs [2, 3]. B and T lymphocytes interact in a regulatory loop essential for a complete and effective immune response. Disruption of this cross talk has profound effects on the immune system. The efficacy of rituximab might be due to the suppression of abnormal humoral immunity and T-cell activity through the depletion of B cells [11]. The literature review of Marino et al. on rituximab in autoimmune diseases, particularly myasthenia gravis, underlined the broad effects on B-cell function, including antigen presentation, cytokine production, and T-cell activation. Furthermore, since a small subset of T lymphocytes expresses CD20, its direct targeting by rituximab may contribute to the therapeutic effect [12]. This hypothesis was demonstrated in multiple sclerosis patients, where rituximab effectively depletes CD3 + CD20dim T cells [13]. These findings and pathophysiological hypotheses are consistent with updated ASCO guidelines on irAE management, which recommend rituximab for dermatologic, neurologic, and hematologic irAEs if IVIG or plasmapheresis is ineffective [2].

Patients in our study had a median follow-up of 9.7 months, with a favorable OS of 61% at one year, suggesting that rituximab may not jeopardize the efficacy of ICI. Nonetheless, in the absence of a control group, overall survival at 12 months after severe irAEs should be interpreted with caution. In the study by Maillet et al. [14], patients experiencing irAE (all grades) had an estimated 12-month OS of 60%, compared with around 25–30% in patients without irAE. In contrast, outcomes appear markedly poorer in life-threatening, organ-specific irAEs. Salem et al. reported a 6-month OS of 70% in patients with severe immune checkpoint inhibitor-associated myocarditis and diaphragmatic involvement treated with abatacept and ruxolitinib [15].

The role of B cells in the antitumor activity of ICI is unknown. Damsky et al. studied the B cells population in 40 melanomas from patients treated with pembrolizumab or nivolumab. They demonstrated that the extent of tumor-infiltrating B cells is not associated with the response to anti-PD-1 inhibitors in melanoma. PD-1 inhibitors induce tumor shrinkage in murine cancer models, even in the absence of B cells or after their depletion. This suggests that PD-1 inhibitors remain effective in patients with impaired B-cell function [16]. Wulfken et al. reported a case of a sustained remission due to PD-1 inhibition in a patient with metastatic melanoma, in a context of a complete depletion of B cells, due to preexisting granulomatosis with polyangiitis treated with rituximab. These findings indicate that B cells were not essential for ICI efficacy in this case. Additionally, it has been suggested that regulatory B-cell depletion might have contributed to enhanced ICI efficacy [17].

We acknowledge some limitations to our study. Firstly, this is a retrospective study with the known biases related to this type of study. Secondly, the number of patients analyzed is low. Finally, rituximab was not the only treatment administered. Thus, it is difficult to attribute the whole efficacy to this single treatment. Despite these limitations, our study is the largest case series published to date. Further randomized studies should prospectively validate these findings.

Conclusion

While first-line therapies effectively treat most irAEs, further options are warranted in refractory cases. Our case series suggests a high efficacy (83.3%) of rituximab on several severe irAEs, notably neurological toxicities and myocarditis. In this context, rituximab appears to be a relevant therapeutic option, particularly as our results also suggest that ICI efficacy on the oncologic disease is maintained. Prospective larger studies are needed to further evaluate the efficacy of rituximab in these conditions.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization was done by S.H., ATJM., and C.C.; methodology, investigation, and writing—original draft were performed by S.H. and C.C.; formal analysis was performed by B.J. and C.C.; resources were done by S.M., A.W., P.W.D.V, E.C., G.P., A.L., Q.D.T, and X.Q.; writing—review and editing was performed by all authors; supervision was done by C.C.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Data availability

De-identified data are available from the corresponding author on reasonable request.

Declarations

Conflict of interest

The authors declare no competing interests.

Ethics approval and consent to participate

This study was conducted in accordance with the Declaration of Helsinki and was approved by the Montpellier University Hospital Ethics Committee (approval number 202100908). The requirement for informed consent was waived by the committee.