Clinical Patterns and Follow-Up of Inflammatory Arthritis and Other Immune-Related Adverse Events Induced by Checkpoint Inhibitors. A Multicenter Study

Objectives: To describe different clinical patterns of rheumatic immune-related adverse events (irAEs) induced by immune checkpoint inhibitors (ICI) and their rheumatic and oncologic outcomes.

Methods: We classified clinical syndromes according to five different categories: non-inflammatory arthralgias (NIA), rheumatoid arthritis (RA)-like, psoriatic arthritis (PsA)-like, polymyalgia rheumatica (PMR)-like, and a miscellaneous group of patients with other syndromes. We conducted a baseline visit and then follow-up in order to determine their clinical pattern, treatment response, and outcome.

Results: We included 73 patients (64% male) with a mean age of 66.1 ± 11.6 years. Main underlying diagnosis was lung carcinoma in 29 (39%) patients, melanoma in 20 (27%), and renal-urothelial cancer in 11 (15%). Main ICI included Pembrolizumab in 24 (32%), Nivolumab 17 (23%), and Atezolizumab 7 (9 %). Seventeen out of seventy-three patients had an underlying rheumatic disease before ICI treatment. Fourteen patients developed other irAEs before or simultaneously with rheumatic syndromes. Main rheumatic irAEs included: RA-like in 31 (42.4%), NIA in 19 (26.0%), PMR-like in 10 (13.7%), and PsA-like in 5 (6.8%), among others. Median time from ICI to irAEs was 5 months (IQR 3-9). Those patients who received combined therapy, had a trend for an earlier presentation than those who received monotherapy (4.3 months IQR 1.85-17 vs. 6 months IQR 3-9.25, p = NS). Mean follow-up time was 14.0 ± 10.8 (SD, months). At the last visit, 47 % were taking glucocorticoids and 11% DMARD therapy. At the last visit, 13 (17.8%) patients remained with persistent arthritis, 19 (26%) had intermittent flares, and 39 (53.4%) had a self-limited pattern. Only in 15.1% of patients ICI therapy was discontinued.

Conclusions: We described different patterns according to treatment and irAEs. Combined ICI therapy had an earlier onset of symptoms. Patients who presented as RA-like, had a higher risk of persistent arthritis. After a mean follow-up of more than 1 year, one-fifth of the patients remained with persistent arthritis and 11% required DMARD therapy.