DOP030 Anti-TNF re-induction is as effective, simpler and cheaper compared with dose interval shortening following secondary loss of response in Crohn's disease: a dual centre “real world” study

Abstract

Background: Anti-TNF re-induction (RI) refers to re-dosing infliximab (IFX, 5mg/kg) at weeks 0, 2, 6 or adalimumab (ADA, 160mg, 80mg) at weeks 0, 2, prior to resuming standard IFX (8-weekly) or ADA (2-weekly) dosing. RI presents a simple and cost effective alternative to ongoing dose interval shortening (DIS) to recapture response following secondary loss of response (2°LOR) to maintenance anti-TNF therapy. This study aimed to compare the “real world” cost and efficacy of anti-TNF RI and DIS following 2°LOR in Crohn's Disease (CD) and identify factors associated with durability of response. Methods: CD patients attending two Inflammatory Bowel Disease (IBD) centres in Melbourne, Australia between 1/1/2006–30/8/2016, who developed 2°LOR to maintenance anti-TNF and subsequently underwent RI or DIS, were retrospectively identified from existing IBD databases. DIS was standardised to 6-weekly IFX or weekly ADA. Treatment failure was defined as either CD-related bowel resection, discontinuation of, or switching to another biologic within 12 months of RI or DIS. “Real world” drug costs of RI and DIS were based on Australian Pharmaceutical Benefit Scheme pricing. Non-parametric statistics plus bi- and multivariate linear regression depicting variables associated with time to failure were calculated. Results: Of 423 patients on anti-TNF therapy for CD, 88 (21%) developed 2°LOR to IFX (51, 58%) and ADA (37, 42%) respectively. Median age and disease duration at 2°LOR was 36.6y (17.4–78.0) and 9.4y (0.7–35.2) respectively, with a median duration from anti-TNF initiation to 2°LOR of 2.1y (0.2–8.1). Baseline characteristics were similar between groups, except the RI group had more smokers (48% vs 18%, p<0.01). Subsequent RI or DIS was undertaken in 33 (38%) patients and 55 (62%) patients respectively. Treatment failure rates at 12 months were similar (27% vs 15%, p=0.17). Factors favourably associated with a durable response to either RI or DIS on multivariate linear regression included subtherapeutic serum anti-TNF levels (B 5.37 [4.82, 5.92]), initial RI (B 2.56 [2.24, 2.89]), use of IFX (B 1.52 [1.26, 1.78]), and concurrent, optimised thiopurines (B 0.68 [0.46, 0.91]), p<0.01). RI demonstrated lower median “real world” costs due to fewer anti-TNF doses during the follow-up period (median: AUD $4,839 vs $13,170, doses 2.0 vs 7.4, p<0.01). Figure 1. “Real world” comparison of additional anti-TNF drug costs associated with RI vs DIS. Conclusions: This “real world” study suggests that first-line RI following 2°LOR to anti-TNF therapy in CD was as effective, simpler and cheaper than first-line DIS. Larger prospective evaluation is now planned.