Predictor of primary response to antitumor necrosis factor-α therapy for inflammatory bowel disease: a single-center observational study.

Abstract

It is necessary to find reliable and appropriate predictors of primary response to anti-TNFα therapy (infliximab and adalimumab) in inflammatory bowel disease (IBD) so as to avoid treatment failure and select optimal treatment. The aim of this study is to reveal useful predictors of the response to anti-TNFα treatment from baseline to 2 months after initial administration of anti-TNFα for individual IBD patients using our pharmacokinetic and pharmacodynamic (PK/PD) model at the time of second administration. We retrospectively analyzed 26 IBD patients who received anti-TNFα. In the PK/PD model, inflammation was assumed to be suppressed based on the action of anti-TNFα at the rate constant of Kanti-TNFα (day-1). Kanti-TNFα0 (day-1) is Kanti-TNFα in the absence of anti-TNFα. We expressed inflammation caused by factors not affected by the action of anti-TNFα using the rate constant Kelse (day-1). Using univariate and multivariate linear regressions, we statistically analyzed factors related to the improvement of disease activity index. The significant correlation between Kanti-TNFα0/Kelse and the improvement of disease activity index was shown in Crohn's disease patients (univariate: estimated value 2.4; P = 0.003; and multivariate: 1.8; P = 0.012) and ulcerative colitis patients (univariate: 0.12; P = 0.011), and no other factors were significant. This is the first study to present a useful predictor of primary response to anti-TNFα of individual IBD patients at second administration. The Kanti-TNFα0/Kelse ratio may help to select the optimal therapeutic drug and avoid the improper continuous administration of anti-TNFα in the induction phase.