Herpes Zoster in the Age of Focused Immunosuppressive Therapy

Drugs that inhibit tumor necrosis factor α (TNF-α) are now widely used for management of a variety of inflammatory processes (eg, rheumatologic disorders, inflammatory bowel diseases) proven refractory to conventional therapy. As with many medical advances, benefit comes with a price, and an important consequence of anti–TNF-α therapy is increased risk of infection, especially tuberculosis.¹ Other studies have linked anti–TNF-α therapy with increased risk of serious bacterial² and fungal³ infections. Associations between TNF-α antagonists and viral infections have not been as clearly defined, though case reports of severe episodes of herpes zoster in patients receiving these drugs have appeared in the literature.⁴ Prior analyses of large databases have yielded conflicting results regarding a causal association between treatment with biologic agents and herpes zoster.^{5 - 6}

In this issue of JAMA, Strangfeld and colleagues⁷ further explore possible associations between anti-TNF-α therapy and viral diseases. The authors probed a prospectively collected database from the German biologics register of patients with rheumatoid arthritis to determine if the frequency of herpes zoster was increased and, if so, whether the increased risk was linked to therapy.⁷ In a population of 5040 patients with rheumatoid arthritis, the investigators identified 86 cases of herpes zoster in 82 individuals. Thirty-nine of the cases were temporally linked to treatment with anti–TNF-α monoclonal antibodies (adalimumab or infliximab), 23 to the fusion protein etanercept, and 24 to conventional disease-modifying antirheumatic drugs (DMARDs). The herpes zoster incidence rate was 11.1 per 1000 patient-years in the group treated with the monoclonal antibodies, 8.9 per 1000 patient-years for etanercept, and 5.6 per 1000 patient-years for conventional DMARDs. Older age and prednisone use were confirmed to be risk factors for development of herpes zoster, as previously described in other populations.

After adjustment for age, rheumatoid arthritis severity, and prednisone use, a significantly increased risk for herpes zoster was observed for treatment with the monoclonal antibodies (hazard ratio [HR], 1.82 [95% confidence interval {CI}, 1.05-3.15]) but not with etanercept (HR,1.36 [95% CI, 0.73-2.55]). Risk increased when patients switched from DMARDs to therapy with adalimumab or infliximab. The authors concluded that there appears to be an increased risk of herpes zoster in patients receiving anti–TNF-α monoclonal antibody treatment for rheumatoid arthritis.⁷

This report presents some compelling data but also raises interesting questions. The increased incidence of herpes zoster observed in the group treated with anti–TNF-α monoclonal antibodies (11 cases per 1000 patient-years) is not nearly as high as rates seen in other immunocompromised groups such as persons seropositive for human immunodeficiency virus (30-50 cases per 1000 patient-years)^{8 - 10} or solid organ transplant recipients (25-40 cases per 1000 patient-years).^{11 - 14} This moderate increase in risk presumably reflects the more narrowly focused immunosuppressive effects of anti–TNF-α therapy. For reference, the incidence of herpes zoster in the US population is approximately 3.5 cases per 1000 person-years, ranging from 1 to 2 per 1000 person-years in individuals aged 20 to 29 years to 10 to 11 per 1000 person-years for those older than 80 years.^{15 - 16} Put another way, therapy with anti–TNF-α monoclonal antibodies appears to induce a risk similar to that observed in populations older than 80 years.

However, cases of unusually severe herpes zoster seem to be overrepresented in this population of patients with rheumatoid arthritis. Of note, 18 of the 86 herpes zoster episodes (20%) were judged “severe,” and 12 (13%) required hospitalization. In addition, 15 patients (18%) had multidermatomal herpes zoster (although it is not specified whether these were contiguous or noncontiguous dermatomes), 1 had visceral involvement (esophagitis and pneumonitis), and 5 (6%) had recurrent episodes. The highest incidence of multidermatomal herpes zoster was noted among patients receiving therapy with monoclonal antibodies. This unexpectedly large number of atypical cases may be attributable to a reporting bias or possibly to enhanced disease severity resulting from the immunosuppressive effects of treatment for rheumatoid arthritis.

The authors report that only 2 patients (2.4%) experienced postherpetic neuralgia. While the incidence of postherpetic neuralgia varies with the age of the study population (higher rates are seen in older populations) and the definition of postherpetic neuralgia used, the frequency observed in this study is unexpectedly low. Whether the incidence is higher in immunocompromised compared with immunocompetent patients has been debated, but it seems clear that older age is by far the strongest risk factor. Among adults with herpes zoster, approximately 20% to 30% report persistent pain at 90 days; for patients older than 70 years, the rate is approximately 30% to 40%.^{15 ,17}

Although the mean age of the population studied by Strangfeld et al is relatively young (approximately 54 years), it is somewhat surprising that more cases of postherpetic neuralgia were not documented. The low incidence in this study is possibly explained by underreporting. However, if the data are indeed correct, it forces the important question of how these medications may influence the development, perception, or amelioration of neuropathic pain. Investigators have previously attempted, unsuccessfully, to blunt the development of postherpetic neuralgia using the anti-inflammatory effects of corticosteroids.^{18 - 19} If the observations made in this study are confirmed, they may provide new insights into the role of immune-mediated mechanisms in the pathogenesis of postherpetic neuralgia.

Strangfeld et al⁷ comment on the possible role of vaccination for preventing herpes zoster in the rheumatoid arthritis population. The herpes zoster vaccine has clearly been shown to reduce the risk and severity of shingles in immunocompetent adults 60 years or older.²⁰ However, because it is a live-virus vaccine, vaccination of immunocompromised patients is contraindicated, although clinically significant infection caused by VZV_oka vaccine virus following herpes zoster vaccination has not been described.²¹ Until sufficient safety data are available, administration of the herpes zoster vaccine to patients receiving anti–TNF-α therapy is not recommended. However, administration of the vaccine to an adult for whom anti–TNF-α therapy is planned (but not yet initiated) seems reasonable and appropriate to consider, although this approach has not been validated in a clinical trial.²²

At this juncture, an association of TNF-α inhibitor therapy (specifically the monoclonal antibodies) with infection caused by Mycobacterium tuberculosis is definite, with other bacteria and fungi is probable, and with some herpes viruses is likely. The study by Strangfeld et al⁷ provides the best evidence to date for a positive association with reactivation of latent varicellazoster virus. Although data are limited, anecdotal case reports suggest that disease caused by herpes simplex virus reactivation may also occur with increased frequency or severity in patients receiving anti–TNF-α treatment.^{23 - 24} In addition, 3 cases of polymerase chain reaction–proven herpes simplex virus encephalitis have recently been described among patients actively treated with anti–TNF-α monoclonal antibodies for inflammatory arthritis (R.D. Bradford and A.C. Pettit, written communication, June 30, 2008).

The TNF-α inhibitors provide tremendous benefit to a broad spectrum of patients with systemic inflammatory diseases. As with any therapy, time is required for all of the safety concerns related to these potent medications to become apparent. TNF-α inhibitors have revolutionized the management of a number of difficult diseases, especially inflammatory arthritis, but clinicians must continue to remain aware of the potential for serious infectious complications, which now include herpes zoster.