Objective: We performed a study of tumor necrosis factor inhibitors (TNFi) compared to non-TNFi biologic therapies in rheumatoid arthritis to test whether body mass index (BMI) modified the effect of each therapy.
Methods: We utilized data from CorEvitas. We studied 3 clinical outcomes based on the Clinical Disease Activity Index (CDAI) at 6 months from therapy initiation: 1) achievement of low disease activity (LDA); 2) a change as large as the minimum clinically important difference (MCID); and 3) the absolute change. We categorized BMI and utilized restricted cubic splines to consider nonlinear associations. We used linear and logistic regression to evaluate associations with response, adjusting for confounders. To determine if comparative effectiveness of therapy varied by BMI, we tested for interactions between BMI and class of therapy.
Results: The sample included 2,891 TNFi and 3,010 non-TNFi initiators. Among all initiators, those with severe obesity experienced lower odds of achieving LDA or MCID and less improvement in CDAI score, although associations were attenuated with adjustment. Low BMI was associated with reduced response rates in adjusted models including lower odds of LDA (odds ratio 0.32 [95% confidence interval (95% CI) 0.15, 0.71], P = 0.005). Analyses stratified by TNFi and non-TNFi therapies demonstrated no differences in clinical response rates for TNFi versus non-TNFi across BMI categories (all P for interaction >0.05). Estimates for non-TNFi biologics fit within the 95% CI for TNFi.
Conclusion: This study observed lower response rates among obese and underweight patients and no evidence of a superior effect of non-TNFi therapy over TNFi therapy in particular BMI categories.
© 2022 American College of Rheumatology. This article has been contributed to by U.S. Government employees and their work is in the public domain in the USA.