Brief Report: Risk of Gastrointestinal Perforation Among Rheumatoid Arthritis Patients Receiving Tofacitinib, Tocilizumab, or Other Biologic Treatments. Issue 11 (27th October 2016)

• Record Type: Journal Article
• Authors: Xie, Fenglong; Yun, Huifeng; Bernatsky, Sasha; Curtis, Jeffrey R.
• Is Part Of: Arthritis & rheumatology. Volume 68:Issue 11(2016)
• Page Start: 2612
• Page End: 2617
• Publication Date: 2016-10-27
• Subjects: Arthritis -- Periodicals; Rheumatism -- Periodicals; 616.72
• DOI: 10.1002/art.39761
• Languages: English
• ISSNs: 2326-5191

Abstract

Abstract : Objective: To evaluate gastrointestinal (GI) perforation in rheumatoid arthritis (RA) patients receiving tofacitinib, tocilizumab, or other biologic agents. Methods: Using health plan data from 2006 through 2014, RA patients without prior GI perforation were identified. Those in whom treatment with tofacitinib or a biologic agent was being initiated were followed up for incident GI perforation with hospitalization. Crude incidence rates were calculated by exposure. Adjusted Cox proportional hazards models were used to evaluate the association between GI perforation and exposures. Hazard ratios (HRs) with 95% confidence intervals (95% CIs) were calculated. Results: A cohort of 167, 113 RA patients was analyzed. Among them, 4, 755 began treatment with tofacitinib, 11, 705 with tocilizumab, 115, 047 with a tumor necrosis factor inhibitor (TNFi), 31, 214 with abatacept, and 4, 392 with rituximab. Compared to TNFi recipients, abatacept recipients were older, tofacitinib and rituximab recipients were younger, and tocilizumab recipients were similar in age. Patients beginning treatment with a non‐TNFi agent were more likely to have previously received biologic agents than patients beginning treatment with a TNFi. The incidence of GI perforation per 1, 000 patient‐years was 0.86 (tofacitinib), 1.55 (tocilizumab), 1.07 (abatacept), 0.73 (rituximab), and 0.83 (TNFi). Most perforations occurred in the lower GI tract: the incidence of lower GI tract perforation per 1, 000 patient‐years was 0.86 (tofacitinib), 1.26 (tocilizumab), 0.76 (abatacept), 0.48 (rituximab), and 0.46 (TNFi). Lower GI tract perforation risk was significantly elevated with tocilizumab treatment, and numerically elevated with tofacitinib treatment, versus treatment with TNFi. Adjusted HRs were 2.51 (95% CI 1.31–4.80) for tocilizumab and 1.94 (95% CI 0.49–7.65) for tofacitinib. Older age (HR 1.16 per 5 years [95% CI 1.10–1.22]), diverticulitis/other GI conditions (HR 3.25 [95% CI 1.62–6.50]), and prednisone use at >7.5 mg/day (HR 2.29 [95% CI 1.39–3.78]) were associated with lower GI tract perforation. The incidence of upper GI tract perforation was similar among all drug exposures. Conclusion: The risk of lower GI tract perforation associated with tocilizumab treatment, and possibly tofacitinib treatment, is elevated compared to that associated with TNF blockade.