Impact of Race on Loss of Response to Infusion Biologics in Inflammatory Bowel Disease Patients in a Tertiary Care Referral Center

Abstract

Introduction: Inflammatory bowel disease (IBD) is a chronic condition becoming increasingly prevalent in the African-American (AA) population. There is limited data on treatment outcomes between the two populations. We aimed to compare treatment outcomes and disparities in Caucasian (C) and AA patients receiving infliximab and vedolizumab. Methods: A retrospective chart view was performed on IBD patients newly started on infliximab and vedolizumab at Emory University between 2012-2016 after IRB approval. Clinical and demographic variables were analyzed. Continuous variable were compared using student t test, and categorical variables were compared using Chi square test. Binary logistic regression modeling was utilized to determine likelihood predictors of loss of response (LOR). Results: A total of 150 patients were selected including 95 C and 47 AA patients. Table 1 highlights the demographic and clinical comparisons of the groups. A trend towards CD C patients being older than AA (p=0.08) was found as well as more thiopurine use at induction (p=0.07) in C CD patients. Body mass index (BMI) did not differ between the two groups. However, in UC patients, a trend towards more AA patients being obese than C patients pre and post 1-year treatment (p=0.057) was identified. No differences in LOR, time to LOR, or changes in management following LOR was seen between AA and Caucasians patients with IBD. On binary logistic regression race, gender, and age were not predictive of LOR for all comers, p=0.53. In our CD cohort, there was a trend towards LOR to infliximab (OR 4.69, p=0.07) and vedolizumab (OR 0.21, p=0.07) in AA compared to Caucasians when the logistic regression was adjusted for age and gender.Table: Clinical and demographic characteristics by race. Please see the JPEG file for Table 1.Conclusion: Our study reveals similarities and subtle differences in outcomes of AA and Caucasian IBD patients treated with vedolizumab and infliximab. While it is known that AA patients are 1.5x more likely to be obese compared to C patients, our CD cohort did not show a difference between the two populations in terms of percent obesity. However, there was a trend towards more AA UC patients to be obese than C patients, similar to the general population. While race was not a predictor for LOR for all comers, within the CD cohort, we did find a trend towards LOR to infliximab and vedolizumab in AA compared to C, which may be due to findings of possible less thiopurine use in AA CD patients. Such subtle disparities warrant further study in larger cohorts.