Association of hepatitis B with antirheumatic drugs: a case–control study

Background Though concern of hepatitis B virus (HBV) reactivation by antirheumatic agents has limited therapeutic opportunities in HBV-infected rheumatoid arthritis (RA) patients, the relative risks (RR) among such agents have not been clarified. Objective We compared the reporting of antirheumatic-agent-associated hepatitis B. Patients We assessed 92 hepatitis B cases and 98,069 controls from a population of 98,161 RA patients registered into the US Food and Drug Administration’s (FDA’s) adverse event database between 2004 and 2010. Measurements A reporting odds ratio (ROR), a signal suggesting a risk for hepatitis B among antirheumatic agents, was measured. Results Treatment with corticosteroids [ROR 2.3 (95 % confidence interval 1.3–4.0)], methotrexate [4.9 (3.9–6.0)], rituximab [7.2 (5.3–9.9)], tacrolimus [4.2 (1.5–11.9)], or reporting from Japan [2.2 (1.1–4.2)] were associated with higher signal, whereas adalimumab had a lower ROR [0.2 (0.1–0.4)]. Limitations There are known limitations of spontaneous reporting, such as underreporting, the Weber effect, reporting bias, indication bias, and limited clinical information such as HBV status. Conclusions Adalimumab’s low reporting rate is most likely be due to notoriety. However, the possibility that adalimumab might suppress reactivation of HBV cannot be denied. Until the possibility is clarified in well-designed clinical studies, physicians should use adalimumab cautiously in patients with HBV.