The body composition profile is associated with response to anti‐TNF therapy in Crohn's disease and may offer an alternative dosing paradigm

Summary Background Anti‐tumour necrosis factor (TNF)s form a major part of therapy in Crohn's disease and have a primary nonresponse rate of 10%‐30% and a secondary loss of response rate of 5% per year. Myopenia is prevalent in Crohn's disease and is measured using body composition analysis tools. Aim To test the hypothesis that body composition can predict outcomes of anti‐TNF primary nonresponse and secondary loss of response. Methods Between January 2007 and June 2012, 106 anti‐TNF naïve patients underwent anti‐TNF therapy for Crohn's disease with body composition parameters analysed using CT scans to estimate body fat‐free mass. The outcome measures were primary nonresponse and secondary loss of response. COX‐regression analysis was used with 3 year follow‐up data. Results A total of 106 patients were included for analysis with 26 (24.5%) primary nonresponders and 29 (27.4%) with secondary loss of response to anti‐TNF therapy. Sex‐specific cut‐offs for muscle and fat were ascertained by stratification analysis. On univariate analysis, primary nonresponse was associated with low albumin (OR 0.94; 0.88‐0.99, P = .04) and presence of myopenia (OR 4.69; 1.83‐12.01, P = .001) when taking into account patient's medical therapy, severity of disease and body composition. On multivariate analysis, presence of myopenia was associated with primary nonresponse (OR 2.93; 1.28‐6.71, P = .01). Immunomodulator therapy was associated with decreased secondary loss of response (OR 0.48; 0.23‐0.98, P = .04). BMI was poorly correlated with lean body mass (r2 = 0.15, P = .54). Conclusions In this cohort study, body composition profiles did not correlate well with BMI. Myopenia was associated with primary nonresponse with potential implications for dosing and serves as an explanation for pharmacokinetic failure. Linked ContentThis article linked to Murray and Ding et al papers. To view these articles visit https://doi.org/10.1111/apt.14392 and https://doi.org/10.1111/apt.14409.