Impact of Sex, Serostatus, and Smoking on Risk for Rheumatoid Arthritis–Associated Interstitial Lung Disease Subtypes
Gregory C McDermott, Keigo Hayashi, Pierre‐Antoine Juge, Ritu R. Gill, Suzanne Byrne, Staci Gagne, Xiaosong Wang, Misti L. Paudel, Matthew Moll, Michael H. Cho, Kathleen M.M. Vanni, Emily Kowalski, Grace Qian, Katarina J Bade, Alene Saavedra, Yumeko Kawano, Michael DiIorio, Taylor Wolfgang, Edy Y. Kim, Paul F. Dellaripa, Michael E. Weinblatt, Nancy A. Shadick, Tracy J. Doyle, Jeffrey A. Sparks
Abstract
OBJECTIVE: Rheumatoid arthritis-associated interstitial lung disease (RA-ILD) includes multiple subtypes with varying histopathology, prognosis, and potential treatments. Limited research has investigated risk factors for different RA-ILD subtypes. Therefore, we examined demographic, serologic, and lifestyle associations with RA-ILD subtypes. METHODS: We systematically identified RA-ILD cases and RA controls without ILD (RA-noILD) in the Brigham RA Sequential Study and Mass General Brigham Biobank RA cohort. We determined RA-ILD subtype (usual interstitial pneumonia [UIP], nonspecific interstitial pneumonia [NSIP], and other/indeterminate) through chest high-resolution computed tomography imaging pattern. We investigated associations of demographic, lifestyle, and serologic factors with major RA-ILD subtypes using multivariable logistic regression. RESULTS: Among 3,328 patients with RA, we identified 208 RA-ILD cases and 547 RA-noILD controls. RA-UIP was associated with older age (odds ratio [OR] 1.03 per year, 95% confidence interval [95% CI] 1.01-1.05), male sex (OR 2.15, 95% CI 1.33-3.48), and seropositivity (OR 2.08, 95% CI 1.24-3.48), whereas RA-NSIP was significantly associated only with seropositive status (OR 3.21, 95% CI 1.36-7.56). Nonfibrotic ILDs were significantly associated with smoking (OR 2.81, 95% CI 1.52-5.21). Having three RA-ILD risk factors (male, seropositive, smoking) had an OR of 6.89 (95% CI 2.41-19.7) for RA-UIP compared with having no RA-ILD risk factors. CONCLUSION: Older age, seropositivity, and male sex were strongly associated with RA-UIP, whereas RA-related autoantibodies were associated with RA-NSIP. These findings suggest RA-ILD sex differences may be driven by RA-UIP and emphasize the importance of further studies to clarify RA-ILD heterogeneity and optimize screening and treatment approaches.