Development of an Algorithm Incorporating Pharmacokinetics of Adalimumab in Inflammatory Bowel Diseases

Abstract

OBJECTIVES: Several decision algorithms based on the measurement of infliximab (IFX) trough levels and antibodies to IFX have been proposed. Whether such algorithms can be extrapolated to the pharmacokinetics of adalimumab (ADA) has yet to be determined. METHODS: A prospective study included all consecutive patients with inflammatory bowel disease (IBD) having a disease flare while being on ADA 40 mg every 2 weeks were included. All patients were primary responders to ADA therapy and were anti-tumor necrosis factor (TNF) naive. ADA trough levels and antibodies against ADA (AAA) were measured blinded to clinical data (Elisa LISA-Tracker, Theradiag). All patients were optimized with ADA 40 mg weekly. Four months later, in the absence of clinical remission (CR; Crohn's disease activity index <150 for Crohn's disease (CD), and Mayo score <2 for ulcerative colitis), patients were treated with IFX therapy. Patients were divided into three groups based on ADA trough levels and based on previous studies: group A, ADA>4.9 μg/ml; group B, ADA<4.9 μg/ml and undetectable levels of AAA (<10 ng/ml); and group C, ADA<4.9 μg/ml and AAA >10 ng/ml. RESULTS: A total of 82 patients were included (55% CD; mean age=43 years, disease duration=7.4 years, duration of ADA therapy=17 months). After optimization of ADA treatment, 29.2% of patients achieved CR in group A (N=41), 67% in group B (N=24), and 12% in group C (N=17; P<0.01 between groups A/B and B/C). C-reactive protein level at the time of relapse, disease duration, duration of ADA therapy, and IBD type was not predictive of CR after ADA optimization by univariate analysis. The response to ADA optimization was significantly more durable in group B (15 months) than in groups A and C (4 and 5 months, respectively). Fifty-two patients who failed following ADA optimization (63%) were treated with IFX, and 30.6% of them achieved CR. CR rates following IFX initiation were 6.9%, 25%, and 80% in groups A, B, and C, respectively (P<0.01 between groups C/A and between groups C/B). Duration of response to IFX was significantly higher in group C than in groups A and B (14 vs. 3 and 5 months, respectively, P<0.01). CONCLUSIONS: The presence of low ADA trough levels without AAA is strongly predictive of clinical response in 67% of cases after ADA optimization. Conversely, low ADA levels with detectable AAA are associated with ADA failure, and switching to IFX should be considered. ADA trough levels >4.9 μg/ml are associated with failure of two anti-TNF agents (ADA and IFX) in 90% of cases, and switching to another drug class should be considered.