Smoking is not linked to the development of anti-peptidylarginine deiminase 4 autoantibodies in rheumatoid arthritis

Laura C. Cappelli, Maximilian F. Konig, Allan C. Gelber, Clifton O. Bingham, Erika Darrah

Defining environmental factors responsible for development of autoimmunity in rheumatoid arthritis (RA) is critical for understanding mechanisms of disease initiation and propagation. Notably, a history of cigarette smoking has been implicated in the genesis of RA and is associated with worse disease outcomes. Antibodies to peptidylarginine deiminase 4 (PAD4) are also associated with more severe RA. A subset of patients who have PAD4 autoantibodies that cross-react with PAD3 (anti-PAD3/4) are at the highest risk for interstitial lung disease, and this risk is augmented by a history of cigarette smoking. It is unclear, however, if smoking is etiologically linked to the development of anti-PAD4 antibodies. Patients were included in this study if they had physician-diagnosed RA as well as DNA, serum, and a date-matched clinical assessment (n = 274). Anti-PAD4 and anti-CCP antibodies were measured by immunoprecipitation and ELISA, respectively; shared epitope (SE) status was determined by HLA-DRβ1 genotyping. Logistic regression analysis was used to evaluate associations of smoking with PAD4 antibodies, with adjustment for relevant demographic and clinical features. Stratified analyses by disease duration and shared epitope status were also performed. Anti-PAD4 antibodies were present in 25% of RA patients, with 50% of these individuals having anti-PAD3/4 cross-reactive antibodies. Anti-PAD4 antibodies were significantly associated with a longer disease duration, SE alleles, and anti-CCP antibodies. Importantly, there were no significant differences in smoking history between anti-PAD4 positive and negative groups in univariate analyses, stratified analyses, or multivariable models. However, an inverse relationship between smoking and anti-PAD4 antibodies was suggested by a lower prevalence of current smokers among patients with anti-PAD3/4 antibodies compared to antibody negative individuals (p = 0.04). Further, the lowest levels of anti-PAD4 antibodies were observed in current smokers (p = 0.14), and a significant association of SE and anti-PAD4 antibodies was only present among never smokers (p = 0.01). Smoking history was not associated with anti-PAD4 antibodies in patients with RA. The finding that anti-PAD4 antibodies were not associated with smoking suggests that other environmental factors may contribute to the development of autoimmunity to PAD4 in these patients.