Ab1479 to Each Their Own- Classification of Arthritis Phenotype Among Inflammatory Arthritis Patients With Cancer Treated With Immune Checkpoint Inhibitors

Abstract

BackgroundThe use of immune checkpoint inhibitors (ICIs) for the treatment of cancer can be associated with adverse events including inflammatory arthritis (ICI-IA). ICI-IA can encompass a variety of different phenotypes including: polymyalgia rheumatica (PMR)-like, spondyloarthritis-like, rheumatoid arthritis (RA)-like, and others[1]. Due to the heterogeneous clinical presentation of ICI-IA, methods are needed to find important variables and associations to describe and distinguish between clinical phenotypes.ObjectivesUse supervised machine learning techniques to find important variables associated with classification of arthritis phenotypes among patients with ICI-IA.MethodsThis study is ancillary to a retrospective observational study of patients with ICI-IA at 6 U.S sites who were treated with DMARDS (TNFi, IL6R, or methotrexate). Arthritis phenotypes was the classification target of interest. Demographics, patterns of swollen joints, medications, lab values, and concomitant irAEs were all obtained from the medical record. Descriptive statistics were computed. Classification and Regression Trees (CART) were created using the rpart and partykit packages and Random Forest (RdF) using the randomForest R package. One thousand trees were created for the forest. Variable importance scores (CART) and mean decrease in Gini scores (RdF) were created to assess the variable importance of included variables.Results147 patients were included, mean age 62 years, 43% with melanoma. Arthritis phenotypes were PMR-like in 8.3%, spondylarthritis-like 19%, RA-like 48%, polyarthritis 17%, and other 8%. Two versions of the CART and RdF were performed (with and without study site) (Table 1). Including study site, the following were identified by both CART and RdF as important: study site was the most important variable associated with the phenotypic label, followed by cancer type, hand arthritis, age, total joint count (TJC). The CART flow diagram with study site is shown in theFigure 1. Excluding study site, CART and RdF identified hand arthritis, cancer type, age, TJC.ConclusionWe found that classification of arthritis phenotype was study site-specific in our statistical models. After excluding site, age, cancer type, TJC, maximum steroid dose, and hand arthritis were associated with phenotypic classification. The influence of study site illustrates current heterogeneity in ICI-IA phenotyping among experts and the need for formalized classification for enrollment in clinical trials. These results identify variables that could be explored by the ACR/EULAR ICI-IA classification criteria working group.Reference[1]Ghosh N., Bass A. Rheumatic Complications of Immune Checkpoint Inhibitors. Med. Clin. North Am. 2021 Mar; 105(2):227-245.Table 1.Variable Importance (VI) from Classification and Regression Tree (CART) and mean decrease in Gini score (MDG) from Random forest (RdF) associated with Immune Checkpoint Inhibitor Inflammatory Arthritis phenotype assignment.CharacteristicCART VI1RdF MDG2CART VI1without study siteRdF MDG2without study siteStudy site2510.8Not includedAge1210.61312.2Cancer type166.7207.7Maximum steroid dose of prednisone (mg)7.38.7Total Joint Count45.966.2Hand arthritis143.7254.0Factors with MDG <=3cancer stage, rash, Number of IRAEs, sex, Rheumatoid factor positive(RF+), RF+ or cyclic citrullinated peptide factor positive, arthritis affecting: shoulder, knee, ankle, hip, wrist, and knee1CART Variable importance, the higher the number the more influential the variable is in the CART2Mean Decrease in Gini score from RdF is a measure to rank variables; the higher the number the more important the variable.Figure 1.Classification and Regression Tree for Arthritis Phenotype with Study Site included. The bar graphs at the bottom show which arthritis phenotype is the majority for each terminal node and the y-axis shows the error rates for each group from the model.Acknowledgements:NIL.Disclosure of InterestsDeanna Jannat-Khah Shareholder of: Dr. Jannat-Khah owns shares of Walgreens/Boots Alliance, AstraZeneca, and Cytodyn (non Rheumatologic pharmaceutical company), Grant/research support from: Dr. Jannat-Khah has a grant from Hospital for Special Surgery for research., Nilasha Ghosh Grant/research support from: Dr. Ghosh has a grant from Hospital for Special Surgery for research., Laura Cappelli Grant/research support from: Dr. Cappelli has research grants from the NIH (NIAMS K23AR075872) and from Bristol-Myers Squibb., Pankti Reid Consultant of: Dr. Reid was a consultant for Level Ex, Grant/research support from: Dr. Reid has grant support from the following: COVID-19 Funds to Retain Clinical Scientists by the Supporting Early Career University Researchers to Excel through Disruptions Steering Committee and The University of Chicago Institute of Translational Medicine Clinical and Translational Science Award K12/KL2 Grant 5KL2TR002387-05, Jeffrey Sparks Consultant of: Dr. Sparks was a consultant for AbbVie, Amgen, Boehringer Ingelheim, Bristol Myers Squibb, Gilead, Inova Diagnostics, Janssen, Optum, and Pfizer, Grant/research support from: Dr. Sparks has received grants from the following entities: Bristol Myers Squibb, National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIH), Rheumatology Research Foundation, R.Bruce and Joan M. Mickey Research Scholar Fund, Llura Gund Award for Rheumatoid Arthritis Care and Research., Noha Abdel-Wahab Speakers bureau: Dr. Abdel-Wahab was a speaker for ChemoCentryx., Consultant of: Dr. Abdel-Wahab was a consultant for ChemoCentryx., Grant/research support from: Dr Abdel-Wahab has grant funding from the NIAD (K01AI163412) and from the University of Texas MD Anderson, cassandra calabrese Paid instructor for: Dr. Calabrese received an honorarium for a lecture by Sanofi, Consultant of: Dr. Calabrese was a consultant for Lilly, and AstraZeneca., Carlos Aude: None declared, Karmela Kim Chan: None declared, Anne Bass Grant/research support from: Dr Bass has grants from the Hospital for Special Surgery, Memorial Sloane Kettering Cancer Center, and the Rheumatology Research Foundation.