The Risk of Tuberculosis (TB) in Rheumatoid Arthritis Patients Treated with TNF Inhibitors and the Safety of Resuming Biologic Dmards for Patients Who Developed TB after Anti-TNF Treatment

Abstract

Background/Purpose: The association between TNF inhibitor (TNFI) treatment and the development of tuberculosis (TB) has been confirmed through several observational studies. Current guidelines strongly recommend latent tuberculosis infection (LTBI) screening treatment before starting TNFI and they might reduce the risk of development of active TB in TNFI users. However, there is still limited evidence of the risk of TB in RA patients who used TNFI after introduction of LTBI screening and treatment guidelines. Moreover, the safety of resuming biologic DMARDs for RA patients who developed TB after anti-TNF treatment has not been well-known. This study aims to estimate the incidence of TB in RA patients who used TNFI after introduction of LTBI screening and treatment guidelines and to evaluate the safety of resuming biologic DMARDs for patients who developed TB after anti-TNFtreatment.

Methods: A retrospective cohort of RA patients who started TNFI was established using Korean national healthcare claims database between January 2009 and December 2013. There was 1 year wash-out period of biologic DMARDs to determine the incident users of biologic DMARDs. After excluding the patients who started biologics other than TNFI, we made an inception cohort for RA patients who started TNFI as a first biologic DMARD. Patients were followed for the development of TB or the last observational date (December 31, 2013). The development of the TB was defined as the appearance of ICD 10 code of TB plus at least 3 among 4 agents of isoniazid, rifampin, ethambutol, and pyrazinamide. Through this definition, the incidence rate (IR) per 100,000 person-year (PY) and standardized incidence ratio (SIR) of TB for TNFI starters based on total RA patients were calculated. We also classified the patients who developed TB into two groups; resuming biologic DMARD group vs. conventional DMARD group. After comparing the characteristics between two groups, the TB relapse rate among patients who resumed biologic DMARDs was estimated.

Results: We included 4,638 RA patients who had started TNFI as the first biologic DMARD, contributing 8,542 PYs of follow-up. A total of 81 patients had been developed TB infection during follow-up. The IR and the SIR of TNFI users based on total RA patients were 1,100 per 100,000 PY [860-1,340/100,000 PY, 95% confidence interval (CI)] and 2.04 (1.62-2.54, 95% CI). If we restrict the maximal follow-up period in 1 year, the IR and SIR of TNFI starters increased to 1,660/100,000 PY and 3.08, respectively. Among the 81 patients who developed TB during follow-up, 30 patients (37.0%) had continued or resumed biologic DMARDs. Mean interval between TB development and resuming biologic DMARDs was 3.3 months. Two cases of TB were developed in 30 patients with observational period of 45.7 PY.

Conclusion: The risk of TB in RA patients who start TNFI is still higher than that of total RA patients after introduction of LTBI treatment guidelines. Resuming biologic DMARDs after TB development should be undertaken with careful monitoring of TB relapse.