Risk of Herpes Zoster in Patients With Rheumatoid Arthritis Treated With Anti-TNF-α Agents

Context The risk of bacterial infection is increased in patients treated with drugs that inhibit tumor necrosis factor alpha ( TNF- alpha). Little is known about the reactivation of latent viral infections during treatment with TNF- alpha inhibitors. Objective To investigate whether TNF- alpha inhibitors together as a class, or separately as either monoclonal anti-TNF-alpha antibodies (adalimumab, infliximab) or a fusion protein ( etanercept), are related to higher rates of herpes zoster in patients with rheumatoid arthritis. Design, Setting, and Patients Patients were enrolled in the German biologics register RABBIT, a prospective cohort, between May 2001 and December 2006 at the initiation of treatment with infliximab, etanercept, adalimumab, or anakinra, or when they changed conventional disease- modifying antirheumatic drug ( DMARD). Treatment, clinical status, and adverse events were assessed by rheumatologists at fixed points during follow- up. Main Outcome Measures Hazard ratio ( HR) of herpes zoster episodes following anti - TNF- alpha treatment. Study aims were to detect a clinically significant difference ( HR, 2.0) between TNF- alpha inhibitors as a class compared with DMARDs and to detect an HR of at least 2.5 for each of 2 types of TNF- alpha inhibitors, the monoclonal antibodies or the fusion protein, compared with conventional DMARDs. Results Among 5040 patients receiving TNF- alpha inhibitors or conventional DMARDs, 86 episodes of herpes zoster occurred in 82 patients. Thirty- nine occurrences could be attributed to treatment with anti - TNF- alpha antibodies, 23 to etanercept, and 24 to conventional DMARDs. The crude incidence rate per 1000 patient- years was 11.1 ( 95% confidence interval [ CI], 7.9- 15.1) for the monoclonal antibodies, 8.9 ( 95% CI, 5.6-13.3) for etanercept, and 5.6 ( 95% CI, 3.6- 8.3) for conventional DMARDs. Adjusted for age, rheumatoid arthritis severity, and glucocorticoid use, a significantly increased risk was observed for treatment with the monoclonal antibodies ( HR, 1.82 [ 95% CI, 1.05- 3.15]), although this risk was lower than the threshold for clinical significance. No significant associations were found for etanercept use ( HR, 1.36 [ 95% CI, 0.73-2.55]) or for anti - TNF- alpha treatment ( HR, 1.63 [ 95% CI, 0.97- 2.74]) as a class. Conclusion Treatment with monoclonal anti - TNF- alpha antibodies may be associated with increased risk of herpes zoster, but this requires further study.