Background During randomized controlled trials with tocilizumab, an IL-6 inhibitor, perforations of the lower intestinal tract (LIP) occurred only in the exposed patient population, not in the control group. The relevance of this finding for daily practice is unclear because we know little about the background risk for LIP in rheumatoid arthritis (RA) patients on different treatments.
Objectives To examine the incidence of LIP and their appearance in RA patients treated with different biologics (bDMARDs) or conventional synthetic DMARDs (csDMARDs).
Methods We used data from the prospective longitudinal German biologics register RABBIT with 14,300 RA patients included since May 2001 starting a csDMARD or bDMARD. All serious gastrointestinal adverse events reported until 30th April 2015 which were possibly associated with perforations (including haemorrhages) were filtered (n=137). All events were validated (twice, with additional external validation) based on medical records or specific queries. Reviewers were blinded for treatment exposure. Only events with a definite, non-traumatic and non-iatrogenic perforation of the lower intestinal tract were selected for the analysis. Treatment exposure was defined as treatment given in the last three months before occurrence of the event.
Results In total, 36 LIPs (colon/sigma: 31, appendix: 4, terminal ileum: 1) were observed in 48,102 patient years (PY). In total, 27 of the 36 patients with LIP had concomitant GCs, with a daily dose of ≥7,5mg in 12 patients. In the multivariate analysis higher age, GC use and treatment with TCZ were significantly associated with a higher risk of LIP (hazard ratio (HR) 1.3 per 5mg increase in GC dose [95%CI=1.2, 1.4], and HR 1.5 [1.3, 1.8] per 5 years increase in age). Compared to csDMARDs, treatment with TCZ was associated with a 5 times higher risk for LIP (HR 5.2 [2.4,11.6]) whereas no association was found for TNF inhibitors, abatacept or rituximab.
None of the patients with LIP had a history of diverticulitis known to the treating rheumatologist. In patients treated with TCZ the presentation of LIP events was untypical: acute and localized pain was only reported by some patients, diffuse abdominal pain was more frequent. CRP was completely suppressed or only slightly elevated. In total, 11 (31%) of the patients died after LIP: 2 of 11 csDMARD, 3 of 12 TNFi and 6 of 11 TCZ treated patients.
Conclusions This first comparative analysis of all bDMARDs available for the treatment of RA showed a significant risk of LIPs in patients treated with TCZ. It confirms a signal detected in clinical trials and specified the localization to lower GI tract only. TCZ patients with LIP present untypically. In order to ensure timely intervention, patients should be advised that symptoms of LIP might be mild and that CRP is not a valid diagnostic marker in their case.
Disclosure RABBIT is supported by a joint, unconditional grant from AbbVie, Bristol-Myers Squibb, Celltrion, Hospira, MSD Sharp & Dohme, Pfizer, Roche, and UCB.
Disclosure of Interest None declared