The Safety and Immunologic Effectiveness of the Live Varicella-Zoster Vaccine in Patients Receiving Tumor Necrosis Factor Inhibitor Therapy : A Randomized Controlled Trial

Ann Intern Med. 2021 Nov;174(11):1510-1518. doi: 10.7326/M20-6928. Epub 2021 Sep 28.


Background: The safety and effectiveness of live virus vaccines, such as the varicella-zoster vaccine, are unknown in patients with inflammatory diseases receiving immunomodulatory therapy such as tumor necrosis factor inhibitors (TNFis).

Objective: To evaluate the safety and immunogenicity of the live attenuated zoster vaccine (ZVL) in patients receiving TNFis.

Design: Randomized, blinded, placebo-controlled trial. ( NCT02538341).

Setting: Academic and community-based rheumatology, gastroenterology, and dermatology practices.

Patients: Adults aged 50 years or older receiving TNFis for any indication.

Intervention: Random assignment to ZVL versus placebo.

Measurements: Glycoprotein enzyme-linked immunosorbent assay (gpELISA) and enzyme-linked immunosorbent spot (ELISpot) from serum and peripheral blood mononuclear cells measured at baseline and 6 weeks after vaccination. Suspected varicella infection or herpes zoster was clinically assessed using digital photographs and polymerase chain reaction on vesicular fluid.

Results: Between March 2015 and December 2018, 617 participants were randomly assigned in a 1:1 ratio to receive ZVL (n = 310) or placebo (n = 307) at 33 centers. Mean age was 62.7 years (SD, 7.5); 66.1% of participants were female, 90% were White, 8.2% were Black, and 5.9% were Hispanic. The most common TNFi indications were rheumatoid arthritis (57.6%) and psoriatic arthritis (24.1%); TNFi medications were adalimumab (32.7%), infliximab (31.3%), etanercept (21.2%), golimumab (9.1%), and certolizumab (5.7%). Concomitant therapies included methotrexate (48.0%) and oral glucocorticoids (10.5%). Through week 6, no cases of confirmed varicella infection were found; cumulative incidence of varicella infection or shingles was 0.0% (95% CI, 0.0% to 1.2%). At 6 weeks, compared with baseline, the mean increases in geometric mean fold rise as measured by gpELISA and ELISpot were 1.33 percentage points (CI, 1.17 to 1.51 percentage points) and 1.39 percentage points (CI, 1.07 to 1.82 percentage points), respectively.

Limitation: Potentially limited generalizability to patients receiving other types of immunomodulators.

Conclusion: This trial informs safety concerns related to use of live virus vaccines in patients receiving biologics.

Primary funding source: The National Institute of Arthritis and Musculoskeletal and Skin Diseases and the American College of Rheumatology.

Publication types

  • Multicenter Study
  • Pragmatic Clinical Trial
  • Research Support, N.I.H., Extramural
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Arthritis, Psoriatic / drug therapy
  • Arthritis, Rheumatoid / drug therapy
  • Chickenpox / epidemiology
  • Chickenpox / prevention & control*
  • Double-Blind Method
  • Enzyme-Linked Immunosorbent Assay
  • Enzyme-Linked Immunospot Assay
  • Female
  • Herpes Zoster / epidemiology
  • Herpes Zoster / prevention & control*
  • Herpes Zoster Vaccine*
  • Herpesvirus 3, Human / immunology
  • Humans
  • Immunogenicity, Vaccine
  • Immunoglobulin G / blood
  • Male
  • Middle Aged
  • Tumor Necrosis Factor Inhibitors / therapeutic use*
  • Vaccines, Attenuated*


  • Herpes Zoster Vaccine
  • Immunoglobulin G
  • Tumor Necrosis Factor Inhibitors
  • Vaccines, Attenuated

Associated data