High Constitutive Interleukin 10 Level Interferes With the Immune Response to Varicella-Zoster Virus in Elderly Recipients of Live Attenuated Zoster Vaccine

The Journal of Infectious Diseases MAJOR ARTICLE High Constitutive Interleukin 10 Level Interferes With the Immune Response to Varicella-Zoster Virus in Elderly Recipients of Live Attenuated Zoster Vaccine Anne A. Gershon,1 David Brooks,6 Donald D. Stevenson,3 William K. Chin,5 Michael B. A. Oldstone,4 and Michael D. Gershon2 1Department of Pediatrics and 2Department of Pathology and Cell Biology, Columbia University Vagelos College of Physicians and Surgeons, New York, New York; 3ViralImmunobiology Laboratory, Department of Immunology and Microbiology, Scripps Research Institute, and 4Division of Allergy and Immunology, Scripps Clinic Medical Group, La Jolla, California; 5Allergy, Asthma, Clinical Immunology Clinic, Naval Medical Center Portsmouth, Virginia; and 6Princess Margaret Cancer Center, University of Toronto Medical School, Canada Live attenuated zoster vaccine (Zostavax) is licensed for protection against herpes zoster (HZ) and postherpetic neuralgia. This vaccine is similar to but more potent than the vaccine that prevents primary varicella zoster virus (VZV) infection [1]. A dose of VZV that is 1/14th that in Zostavax is highly effective in protecting susceptible children from varicella (85% after 1 dose and 98% after 2 doses) [2, 3]; however, Zostavax protects only 50%–65% of healthy adults (age, 60–80 years) from HZ [4]. Evidence suggests that immune responses of younger subjects to Zostavax are better than those of elderly adults [5]; nevertheless, it is still unclear why Zostavax protects relatively few recipients from HZ while varicella vaccine achieves nearly universal protection against varicella. Inadequate immune responses to hepatitis and influenza vaccines have been attributed to high preexistent (constitutive) levels Received 27 August 2018; editorial decision 6 November 2018; accepted 12 November 2018.; published online November 16, 2018. Correspondence: A. A. Gershon, MD, Department of Pediatrics, Columbia University Vagelos College of Physicians, New York, NY (). The Journal of Infectious Diseases®  2019;219:1338–46 © The Author(s) 2018. Published by Oxford University Press for the Infectious Diseases Society of America. All rights reserved. For permissions, e-mail: . DOI: 10.1093/infdis/jiy660 1338 • JID 2019:219 (15 April) • Gershon et al of IL-10 [6, 7]. We therefore tested the hypothesis that a high constitutive IL-10 level in elderly subjects impedes the development of a vigorous immune response to VZV. Antibody levels and T-cell responses to VZV were measured after Zostavax vaccination, to evaluate the vigor of the immune response. Antiviral T-cell responses are thought to be responsible for protection against HZ [8, 9], which is due to reactivation of latent VZV. In contrast, humoral immunity protects against the primary VZV infection that causes varicella [10]. Because vaccination stimulates both humoral and cellular immunity, the postvaccination increase in antibody titer has been found to provide a valuable estimate of the vigor of the total immune response [11]. We therefore evaluated the vaccine-induced change in levels of antibody to VZV to test the prediction that the strength of this phenomenon would vary inversely with the constitutive IL-10 level. Release of interferon γ (IFN-γ) from T cells was also assayed in a subset of subjects. Two methods were used to evaluate humoral immunity. One was the fluorescent antibody to membrane antigen (FAMA) assay [12], which yields results that correlate with protection from varicella and is accepted as the gold standard against which other methods to measure protective immunity are compared [13]. The other was the glycoprotein enzymelinked immunosorbent assay (gpELISA), which yields results Introduction. Live attenuated zoster vaccine (Zostavax) was used to test the hypothesis that constitutive level of interleukin 10 (IL-10), which may be high in elderly subjects, impairs vaccine efficacy. If constitutive IL-10 impairs vaccine efficacy, the effectiveness of viral vaccines might be improved by transient inhibition of IL-10 before vaccination. Methods. Zostavax was given to 26 patients (age, 60–80 years). IL-10 and immunity to varicella zoster virus (VZV) were measured at baseline and after vaccination. Fluorescent antibody to membrane antigen (FAMA) assays and glycoprotein enzyme-linked immunosorbent assays (gpELISAs) were used to assess humoral immunity; anti–varicella virus T-cell responses were studied in a subset of subjects. In a prospective animal model, T-cell responses to chimeric vaccines against lymphocytic choriomeningitis virus (LCMV) were assessed in mice that express or lack IL-10. Results. FAMA assays revealed significant boosting (by 4-fold) of humoral immunity, which occurred only in subjects (10 of 26) with a low constitutive IL-10 level (ie, <20 pg/mL); moreover, the Zostavax-induced FAMA and gpELISA responses were inversely related to the constitutive IL-10 level. Significant VZV-specific T-cell responses followed vaccination only in subjects with a low constitutive IL-10 level. Vaccine-induced LCMV-specific T-cell responses in mice lacking IL-10 were greater than in wild-type animals. Conclusions. A high constitutive IL-10 level adversely affects vaccine efficacy. Keywords. Varicella; herpes zoster; vaccination; vaccine failure; cytokine; humoral immunity; FAMA; IL-10; T cells. that do not correlate with protection but is the method most commonly used to assess immunity to VZV [10, 11] . METHODS Constitutive IL-10 levels on the day of vaccination were variable and positively skewed; the overall mean level (±SD) was 384 ± 205 pg/mL, but the median value was 54 pg/mL, and the coefficient of variation was 272%. The IL-10 level on the day after vaccination was significantly increased (mean [±SD], 794 ± 272 pg/mL; P = .01, by the Wilcoxon matched pairs signed rank test); however, the level of IL-10 for each individual on the day after vaccination was highly correlated with that measured in the blood specimen collected on the previous day (coefficient of correlation, 0.98; P < .0001). IL-10 levels in the current set of subjects were not significantly correlated with age (Spearman r = −0.0134; P = .9482). Fourteen individuals had IL-10 levels of ≤20 pg/mL at the time of vaccination, with a mean value (±SD) of 2 ± 1 pg/mL; this group, as in other studies of controls [20, 21], was defined as having a low constitutive IL-10 level. Twelve of 26 individuals had IL-10 levels of >20 pg at the time of vaccination (mean [±SD], 821 ± 418 pg/mL); this group was defined as having a high constitutive IL-10 level. No relationship was found between the use of statins or antihypertensive agents and constitutive levels of IL-10. Vaccination was found to enhance immunity significantly in the overall population of vaccinees, no matter whether the FAMA assay or gpELISA was used for evaluation. For the FAMA assay, the geometric mean ratio (±SD) of the postvaccination titer to the pre (...truncated)