1057: Melanoma and Combination Immune Checkpoint Inhibitors Are More Clearly Associated with the Development of Chronic Inflammatory Arthritis Than Pre-existing Autoimmune Disease: A Cross-Sectional Study of Administrative Health Data

Alexandra Ladouceur¹, Marie Hudson², Hassan Behlouli³, Jeffrey Curtis⁴, Louise Pilote³ and Sasha Bernatsky³, ¹Department of Rheumatology of McGill University and CHU-Bordeaux, Montréal, QC, Canada, ²McGill University, Montréal, QC, Canada, ³Research Institute of the McGill University Health Centre, Montreal, QC, Canada, ⁴University of Alabama at Birmingham, Birmingham, AL

Background/Purpose: Immune checkpoint inhibitors (ICIs) improve survival in many malignancies¹, by augmenting the immune system's anti-tumor response. However, ICI can result in immune-related adverse events (irAEs), including inflammatory arthritis (IA)², affecting 3-7% of ICI users³. Since half of ICI-IA fails to resolve and becomes chronic, we aimed to evaluate whether new ICI-IA occurring in patients with a pre-existing (non-IA) autoimmune disease was associated with a greater likelihood of chronicity compared to ICI-IA occurring in patients without any pre-existing autoimmune disease.

Methods: In the MarketScan Commercial Claims, Medicare Supplemental, and Multi-State Medicaid administrative health databases, we identified all adults (18+) with at least one filled prescription for an ICI (ipilimumab, nivolumab, etc.) from Jan. 1^st 2011-Dec. 31^st, 2021. We excluded those without continuous health insurance coverage and those with any International Classification of Diseases (ICD) billing/hospitalization code for any IA or for DMARDs and/or biologics in the year prior to ICI initiation. The primary exposure of interest was a diagnosis of any non-IA autoimmune condition (e.g. multiple sclerosis, hypothyroidism, etc.) prior to ICI initiation. We evaluated the data between 0 and 90 days and at 180 days post-ICI-IA. Chronic ICI-IA was defined as current use of systemic glucocorticoids (GC), DMARDs and/or biologics 180 days after ICI-IA diagnosis.

Results: During the study period, 29,598 individuals were newly exposed to ICI and 395 (1.3%) of these had one or more ICD code for IA after ICI initiation. One hundred forty-nine (N=149, 38%) individuals with ICI-IA had a pre-existing autoimmune disease. The most frequent autoimmune diseases were gastrointestinal (N=56, 38%), ophthalmologic (N=28, 19%), other rheumatic diseases (N=23, 15%) and endocrine (N=17, 11%). One hundred fifty-seven (N=157, 40%) of those with ICI-IA received treatment with GC, DMARDs and/or biologics within the first 90 days after ICI-IA onset. Sixty-eight (N=68, 17% of the 395) were still under one of these treatments at 180 days post ICI-IA diagnosis. Table 1 demonstrates that men were more likely than women to receive GC, DMARDs and/or biologics treatment within the first 90 days after ICI-IA. Table 1 shows that melanoma and treatment with combination ICI (as well as a strong trend for anti-CTLA-4) were associated with chronic ICI-IA, but no pre-existing autoimmune disease was clearly associated with this outcome.

Conclusion: In univariate cross-sectional analyses of patients followed from 90 days post ICI-IA to 180 days, pre-existing autoimmune disease was not clearly associated with increased chronicity according to our definition. Melanoma, combination ICI, and possibly anti-CTLA-4 was associated with chronic ICI-IA. Multivariate hazard regression analyses are under way.

References: 1. Sharma P. et al., Cancer Discov 2021 2. Roberts J. et al., Autoimmun Rev 2020 3. Suarez-Almazor ME. et al., Support Care Cancer 2020




A. Ladouceur: None; M. Hudson: AstraZeneca, 6, Boehringer-Ingelheim, 1, 5, 6, Bristol-Myers Squibb(BMS), 5, Merck, 6, UCB, 5; H. Behlouli: None; J. Curtis: AbbVie, 2, 5, Amgen, 2, 5, Bristol-Myers Squibb, 2, 5, CorEvitas, 2, 5, Eli Lilly and Company, 2, 5, Janssen, 2, 5, Myriad, 2, 5, Novartis, 2, 5, Pfizer, 2, 5, Sanofi, 2, 5, UCB, 2, 5; L. Pilote: None; S. Bernatsky: None.