P003 Immune checkpoint inhibitor-induced rheumatic adverse events - a single centre retrospective cohort study

Over the last decade, immune checkpoint inhibitor (ICI) use in cancer treatment has rapidly increased, leading to dramatically improved cancer progression-free and overall survival rates. The development of immune-related adverse events (irAEs), including autoimmune rheumatic disease is common. Reporting of rheumatic irAEs has been variable and to our knowledge a UK cohort has not yet been described from the rheumatology perspective.

We undertook a retrospective cohort analysis of all adult patients commencing ICI therapy at the Royal United Hospitals Bath between August 2016 and December 2022. Patients with pre-existing rheumatic disease were excluded. Routinely collected clinical data were reviewed until December 2023, 6 months after discontinuation of ICI, or patient death, whichever occurred first.

636 patients were identified as having received ICI treatment at the RUH during the study period. 78 patients were excluded.

Of the patients included 233/558 were female. 451/558 patients identified as White British. The most common cancer diagnoses were lung, melanoma and renal. Pembrolizumab monotherapy and nivolumab monotherapy were the most common ICI treatments given.

Rheumatic irAEs were reported in 43/558 patients. Of those, 25/43 developed symptoms within 6 months and 35/43 within 12 months of initiating ICI therapy. 6/43 discontinued ICI therapy due to a rheumatic irAE.

Arthralgia was the most reported symptom (32/43). 22/43 developed objective features of inflammatory arthritis, of which 14 had polyarthritis and 8 oligoarthritis. 18/43 were referred to Rheumatology.

Treatment: 32/43 patients received systemic steroids. The average total cumulative dose was equivalent to prednisolone 2367 mg with an average daily dose of 11.2 mg/day. Conventional synthetic disease modifying anti-rheumatic drugs (csDMARDs) were given to 3 patients, of which 1 was due to proceed to biologic therapy (bDMARD) but did not start due to cancer progression. The average cumulative and daily steroid doses for these 3 patients was equivalent to prednisolone 6980 mg and 12.3 mg/day. A further patient was due to initiate bDMARD as first line therapy but did not do so due to patient concerns about the risk of disease progression.

In our cohort, 8% of patients developed a rheumatic irAE, of which 53% developed objective features of inflammatory arthritis. The majority occurred within 12 months of commencing ICI treatment. 74% of patients received steroids though only 51% had documented objective evidence of inflammatory arthritis. The overall steroid burden in these patients is high and carries associated potential risks. The use of DMARDs was low during the study period though post hoc analysis shows this to be increasing. This likely reflects increasing clinician confidence in the use of DMARDs and recognition of the required long-term use of steroids for these patients.

Increasing collaboration between Oncologists, Rheumatologists and the patient will be crucial to optimising care.

M. Ho: None. E. Gates: None. F. Cheese: None. G. Gullick: None. T. Tillet: None. L. Dumas: None. A. Allard: Honoraria; A.A. has received honoraria from Abbvie, Bristol Myers Squibb, Eli Lilly, Novartis, Pfizer and Roche.