Checkpoint Inhibitor Rheumatic Toxicity in ERIN Registry

Key Takeaways

• Among German rheumatology-referred adults receiving checkpoint inhibitors, rheumatoid arthritis–like polyarthritis was the leading rheumatic adverse-event phenotype.
• Rheumatic events were moderate or severe in 90%, and 38.3% of patients permanently discontinued checkpoint inhibitors.
• Systemic glucocorticoids were commonly used, but 58.3% had glucocorticoid-refractory disease and 53.3% required second-line immunomodulatory therapy.
• At a median follow-up of 10 months (IQR, 4.1–31.1), 33.3% had achieved complete remission and 51.7% partial remission; outcomes may change with longer follow-up.
• Male sex was associated with second-line systemic therapy in univariate analysis, without establishing independent prediction.

Rheumatic symptoms during treatment with immune checkpoint inhibitors (ICIs) raise questions about the inflammatory condition involved, symptom burden, and whether cancer treatment can continue. Patients reaching specialist care may differ from the broader population receiving ICIs. Investigators evaluated these clinical and treatment questions among patients referred for rheumatologic assessment.

In their German ERIN registry analysis, Gailis and colleagues examined 60 adults receiving ICIs who presented with rheumatic immune-related adverse events (rh-irAEs) as of January 1, 2026. This multicenter observational cohort comprised patients referred to rheumatologists, not a denominator of everyone receiving ICIs. Investigators classified presentations using a rheumatology-oriented phenotype framework and graded severity with Common Terminology Criteria for Adverse Events (CTCAE) version 5.0. They assessed associations with second-line systemic therapy in univariate analyses; there was no treatment-assigned comparator.

Rheumatoid arthritis–like polyarthritis accounted for 41.7% of presentations. Polymyalgia rheumatica–like and peripheral spondyloarthritis–like disease were the next most common phenotypes. Axial spondyloarthritis–like manifestations also occurred in younger patients. This distribution describes presentations after rheumatology referral, not the frequency of these conditions among all people receiving ICIs.

Some patients were hospitalized for rheumatic events, a measure distinct from permanently stopping cancer treatment. Glucocorticoid refractoriness was accompanied by frequent use of second-line immunomodulatory therapy, but treatment was not assigned by the registry. Oncologic progression occurred in 15.0% of the cohort and 28.1% of patients requiring second-line therapy. Because the latter were a selected subgroup, these observations cannot establish whether treatment contributed to progression. In univariate analyses, male sex was associated with second-line systemic therapy use.

Referral to rheumatologists selects a subset of patients with rh-irAEs, so these findings cannot establish event incidence or be applied without qualification to all ICI recipients. The cohort was German, and the findings do not establish whether the same patterns occur in the United States. Because treatments and outcomes were observed rather than assigned and compared, escalation beyond glucocorticoids cannot be credited with improving rheumatic outcomes on these data.

Gailis and colleagues documented varied rheumatic presentations, substantial symptom burden, escalation beyond glucocorticoids, and interruption of cancer treatment in this specialist-referred group. They also identified axial spondyloarthritis–like manifestations in younger patients as warranting further study.

Clinician Questions

Does male sex independently predict second-line systemic treatment for checkpoint inhibitor–associated rheumatic adverse events?

Among German rheumatology-referred adults with immune checkpoint inhibitor–associated rheumatic immune-related adverse events, male sex was associated with second-line systemic treatment in univariate analysis. Because that analysis did not establish an effect independent of other patient or disease characteristics, it cannot identify male sex as an independent predictor.

Did second-line immunomodulatory treatment cause cancer progression in patients with checkpoint inhibitor–associated rheumatic toxicity?

In the German rheumatology-referred cohort of adults with checkpoint inhibitor–associated rheumatic immune-related adverse events, oncologic progression was observed among those needing second-line therapy. The observational comparison cannot determine whether immunomodulatory treatment caused progression; patients requiring escalation were not a treatment-assigned comparison group.