One-Year Clinical Outcomes in an IBD Cohort Who Have Previously Had Anti-TNFa Trough and Antibody Levels Assessed.

Abstract

Loss of response (LOR) is a big concern for anti-TNFa therapies in inflammatory bowel disease. Immunomonitoring may be useful to optimize response rates and overcome secondary LOR. This was an observational retrospective cohort study of a group of patients with inflammatory bowel disease on infliximab (IFX) and adalimumab (ADA) who had anti-TNFa trough and antibody levels measured, during maintenance phase of treatment. Anti-TNFa trough and antibody levels were measured using standard enzyme-linked immunosorbent assay techniques. Baseline patient characteristics were determined and patients were reviewed 1 year later. Clinical assessment took place with partial Mayo scores for ulcerative colitis and Harvey-Bradshaw index for Crohn's disease. C-reactive protein (CRP) and albumin were also measured. Poor outcomes were defined as the following: need for rescue steroids, dose intensification, surgery, or treatment discontinuation. Seventy-four patients were included in the study, 37 (50%) were female, mean age 41 years, 61 (82%) had Crohn's disease, and 42 (57%) ulcerative colitis. Forty-two (57%) patients received IFX and 32 (43%) ADA. Mean IFX trough was 3.6 μg/mL and mean ADA troughs were 3.78 μg/mL. Twenty-seven percent of patients (n = 20) overall had a poor outcome, with a similar proportion in each group 24% (n = 10) IFX and 31% (n = 10) ADA (P value 0.24). Of the cohort, 14.2% (6/42) treated with IFX had subtherapeutic trough levels, 6.2% (2/32) of ADA patients had a trough level <1 μg/mL (P value = 0.273) There was no difference in mean trough according to outcome (4.9 μg/mL poor versus 5.4 μg/mL good, P value 0.14). Low IFX trough levels did correlate with high CRP, low albumin and response rates, mean CRP 6.66 μg/mL (n = 3), mean albumin 37 g/L for patients with low trough levels and poor response versus CRP 2.0 μg/mL (n = 24), mean albumin 43 g/L for patients with high trough levels and good response (P = 0.009, 95% confidence interval, -0.78 to -0.12). LOR is still a big concern with anti-TNFa therapies. Stand-alone anti-TNFa trough and antibody levels are not useful at predicting LOR/disease progression at 1 year, but low trough levels do correlate well with elevated CRP, hypoalbuminaemia, and poor response rates.