MAJOR ARTICLE. Downloaded from by guest on 10 November 2018

Transcription

1 MAJOR ARTICLE Measurement of Varicella-Zoster Virus (VZV) Specific Cell-Mediated Immunity: Comparison between VZV Skin Test and Interferon- Enzyme-Linked Immunospot Assay Kay Sadaoka, 1,a Shigefumi Okamoto, 1,a Yasuyuki Gomi, 2 Takeshi Tanimoto, 2 Toyokazu Ishikawa, 2 Tetsushi Yoshikawa, 3 Yoshizo Asano, 3 Koichi Yamanishi, 1 and Yasuko Mori 1 1 Laboratory of Virology and Vaccinology, Division of Biomedical Research, National Institute of Biomedical Innovation, Osaka, 2 Kanonji Institute, Research Foundation for Microbial Diseases of Osaka University, Kannonji, and 3 Department of Pediatrics, Fijita Health University School of Medicine, Toyoake, Japan Cell-mediated immunity (CMI) is critical for the prevention and control of varicella-zoster virus (VZV) related disease. To assess CMI to VZV, a varicella skin test and interferon- enzyme-linked immunospot (ELISPOT) assay were both performed in healthy volunteers, and the results were compared. A total of 151 subjects were examined: 16 aged years, 26 aged years, 18 aged years, 73 aged years, and 18 aged years. All were seropositive by a glycoprotein antigen based enzyme-linked immunosorbent assay (gpelisa). Skin test reactivity was significantly correlated with the ELISPOT count, and both decreased with increasing age, indicating an age-dependent decline in CMI to VZV. In contrast, the antibody titer obtained by the gpelisa did not correlate with skin test reactivity. The results suggest that the skin test and ELISPOT assay are both reliable for assessing CMI to VZV and can easily be applied to screen individuals susceptible to the development of herpes zoster. Varicella-zoster virus (VZV) causes chickenpox and becomes latent in sensory ganglia. Herpes zoster is caused by the reactivation of latent VZV in sensory trigeminal and dorsal root ganglia. Reactivated VZV replicates in the skin and produces the characteristic herpes zoster rash accompanied by acute pain [1]. The incidence and severity of herpes zoster increase with age because of age-related waning of VZV-specific cell-mediated immunity (CMI); more than half of persons who develop herpes zoster are 60 years old [2]. Live attenuated VZV Received 24 March 2008; accepted 12 May 2008; electronically published 5 September Potential conflicts of interest: Y.G., T.T., and T.I. are employees of Biken, which manufactures skin test antigen. All other authors report no potential conflicts. Presented in part: 33rd International Herpesvirus Workshop, Estoril, Portugal, 27 July 1 August 2008 (poster 5.42). Financial support: Ministry of Health, Labor, and Welfare of Japan (Research Promotion of Emerging and Reemerging Infectious Diseases Grant H18-Shinko-013 to Y.M. and T.Y.). a K.S. and S.O. contributed equally to this work. Reprints or correspondence: Dr. Yasuko Mori, Laboratory of Virology and Vaccinology, Div. of Biomedical Research, National Institute of Biomedical Innovation, 7-6-8, Saito-Asagi, Ibaraki, Osaka , Japan (ymori@nibio.go.jp). The Journal of Infectious Diseases 2008; 198: by the Infectious Diseases Society of America. All rights reserved /2008/ $15.00 DOI: / vaccine can boost a VZV-specific CMI that has decreased as part of the aging process, thereby preventing the reactivation of VZV and reducing the severity of herpes zoster [3 7]. Furthermore, vaccination with inactivated varicella is effective for reducing the risk of herpes zoster in immunocompromised patients [8]. Therefore, screening individuals who are susceptible to herpes zoster by monitoring the state of their immunity to VZV and boosting their immunity by vaccination could an be effective way of decreasing herpes zoster morbidity. The varicella skin test has long been used to assess CMI to VZV [9, 10]. With this test, an individual s susceptibility to VZV can be easily judged by an erythematous change h after intradermal antigen injection. It is a simple and safe way to assess CMI to VZV, because no special skill or laboratory equipment is needed. Recently, an interferon (IFN) enzyme-linked immunospot (ELISPOT) assay was reported as a novel method for assessing CMI to VZV [11, 12]. Because this assay directly measures the number of T cells secreting IFN- after stimulation with VZV antigen, it is extremely sensitive and specific compared with other assays, such as the lymphoproliferative activity assay [13, VZV-Specific Cell-Mediated Immunity Measurement JID 2008:198 (1 November) 1327

2 Table 1. Characteristic 14] or responder cell frequency method [15]. Although both the varicella skin test and IFN- ELISPOT assay can be used to measure CMI to VZV, a correlation between the results of these 2 tests has not been clearly shown. In this study, we performed both assays to measure CMI to VZV in 151 healthy volunteers and evaluated the validity of these methods. METHODS Characteristics of 151 healthy volunteers. Value Age, median (range), years 49.1 (23 66) Male sex, no. 83 Case history, no. (%) Herpes zoster 12 (7.9) Varicella vaccination 10 (6.6) Herpes zoster and vaccination 5 (3.3) No history of zoster and/or vaccination 124 (82.1) NOTE. Case history data were self-reported on a questionnaire administered at enrollment. Study design. In the same session, a VZV skin test was performed and a blood specimen collected for the IFN- ELISPOT assay and ELISA. Serum samples were collected separately for each participant in the study. This study was approved by the ethical committees of all the involved institutions. Written informed consent was obtained from all the enrolled volunteers. Population. In 151 healthy volunteers, we studied CMI to VZV using both the IFN- ELISPOT assay and the VZV skin test and humoral immunity using glycoprotein antigen based ELISA (gpelisa). The characteristics of the study population are shown in tables 1 and 2. The mean age was 49.1 years (range, years), and 83 subjects were male (55%). Of the 151 subjects, 12 had a history of herpes zoster, 10 had received varicella vaccination, and 5 had both. Sixteen subjects were aged years (10.6%), 26 were aged years (17.2%), 18 were aged years (11.9%), 73 were aged years (48.3%), and 18 were aged years (11.9%). Isolation of peripheral blood mononuclear cells (PBMCs) from whole blood. Whole blood was collected from donors into Venoject heparin-containing tubes (Terumo). The blood was diluted with PBS without calcium and magnesium and layered on top of a Ficoll solution. The tubes were spun at 880 g for 25 min at 20 C, and the buffy layer containing the PBMCs was collected. The cells were washed twice with PBS, resuspended in stock solution consisting of 90% heat-inactivated fetal bovine serum (Gibco BRL) and 10% dimethyl sulfoxide (Wako), and placed into a Bicell cryogenic freezing container (Nihon Freezer). The freezing container was stored at 80 C overnight, and the frozen cell samples were then transferred to liquid nitrogen (vapor phase) for long-term storage. Cell sample preparation from frozen PBMCs. Complete medium, consisting of RPMI 1640 medium supplemented with 5% heat-inactivated fetal bovine serum (Gibco BRL), 1 mmol/l L-glutamine, 20 g/ml gentamicin, and 50 mol/l -mercaptoethanol, was warmed to room temperature and supplemented with Benzonase (Novagen) to a final concentration of 40 U/mL. The frozen PBMCs were thawed at 37 C and suspended in the Benzonase-supplemented complete medium. The cells were washed again, resuspended in complete medium without Benzonase, and counted. After another wash, cells were resuspended in complete medium without Benzonase at a concentration of cells/ml and used for assays. IFN- ELISPOT assay. The wells of a 96-well MultiScreen HA plate (Millipore) were coated with 100 L of an anti human Table 2. Age-associated decline in skin test reactivity and interferon (IFN) enzyme-linked immunospot (ELISPOT) counts. Group No. of subjects Skin test positive, % Skin test result, mm ELISPOT count gpelisa positive, % gpelisa antibody titer All subjects Age years , years years years years Sex Male Female NOTE. Subjects are stratified according to age or sex. For each stratum, average values are shown for the skin test, IFN- ELISPOT assay, and glycoprotein antigen based ELISA (gpelisa). In the IFN- ELISPOT assay, peripheral blood mononuclear cells were used. Positive skin test results were defined as reactions 10 mm (longest diameter), and positive gpelisa results were defined as antibody titers JID 2008:198 (1 November) Sadaoka et al.

3 recombinant IFN- monoclonal antibody (clone 2G1; Endogen) at a concentration of 3 g/ml, overnight at 4 C. The wells were then washed 3 times with sterile PBS and blocked by adding 200 L of complete medium followed by incubation at 37 C with 5% CO 2 for h. The wells were then washed once with 100 L of complete medium. Next, 100 L of PBMC suspension at cells/ml was added to each well, followed by an equal volume of complete medium containing the VZV antigen. UVtreated (5000 J/m 2 ) varicella vaccine (Biken) containing pfu/ml was used as the VZV antigen. Phytohemagglutinin-L 4 (Wako) at 2 g/ml was included as a positive control. The assay plates were incubated for 40 h at 37 C in 5% CO 2 and 95% humidity. The plates were then washed 4 times with PBS containing 0.1% Tween 20 (ELISPOT wash buffer). A biotinylated anti human recombinant IFN- antibody (clone B133.5; Endogen) was diluted to 0.4 g/ml in ELISPOT wash buffer, and 100 L was added to each well. After a 1-h incubation at room temperature, the plates were washed 4 times with ELI- SPOT wash buffer. Next, 100 L of streptavidin-horseradish peroxidase (BD), diluted 1:1000 in ELISPOT wash buffer, was added to each well, and the plates were incubated at room temperature for 45 min. The plates were washed 4 times with ELI- SPOT wash buffer, 100 L of 3,3',5,5'-tetramethylbenzidine H substrate (Moss) was added, and the spots were developed for 3 min at room temperature. The substrate was then removed from the wells, and the wells were rinsed with water to stop the reaction. The plates were allowed to air dry, and the spots were then enumerated using the KS ELISPOT system (Carl Zeiss) for automated plate scanning, imaging, and spot counting. Results of gpelisa. VZV glycoprotein (VZVgp) was purified following the method of Provost et al. [16], using the VZV Oka vaccine strain and MRC-5 cells. A similarly prepared glycoprotein extract of uninfected MRC-5 cells (MRC-5gp) was used as the negative control. The wells of 96-well ELISA plates were coated with purified VZVgp or MRC-5gp for 24 h at 4 C. The plates were washed 4 times with PBS supplemented with 0.05% Tween 20 (PBS-T) and then blocked for 24 h with PBS supplemented with 0.05% human serum albumin. After the plates were washed 4 times with PBS-T, serum standards, negative and positive control serum samples, and test serum samples, all diluted in PBS-T supplemented with 0.25% human serum albumin (dilution buffer), were added to the VZVgp- or MRC-5gp coated wells. The first plates included a set of blank wells containing dilution buffer alone, to assess the background reactivity. The plates were incubated for 60 min at 37 C, followed by 4 washes with PBS-T. Human IgGs bound to the plates were detected with peroxidase-conjugated anti human IgG for 60 min at 37 C, and the color was developed with 2,2 -azinobis(3-ethylbenzothiazoline-6-sulfonic acid). After the reaction was stopped, the optical density was read at 405 nm. The values were calculated by subtracting the optical density of MRC-5gp from that of VZVgp, and the values of test serum Figure 1. Determination of optimum conditions for the enzyme-linked immunospot (ELISPOT) assay. A, Peripheral blood mononuclear cells (PBMCs) collected from 4 healthy volunteers (samples A, B, C, and D) were subjected to the ELISPOT assay. In each sample, to cells were stimulated with live of varicella-zoster virus (VZV) antigen ( pfu/ml), and the spots were automatically enumerated by the KS ELISPOT system (Carl Zeiss). B, Spot count of sample A stimulated with untreated live VZV antigen, heat-inactivated VZV antigen, or UV-inactivated VZV antigen. There was a significant reduction in spot-forming activity using heat-inactivated VZV compared with untreated VZV (P.02, Student s t test). samples were referenced against a fitted standard curve determined from six 2-fold serial dilutions of the standard. VZV skin test. One hundred microliters of VZV skin test antigen (Biken) was injected intradermally, and the erythema was measured 24 and 48 h later. The longest diameter is given in the figures and tables, because there was usually no significant difference between the longest and shortest diameters. The criteria for scoring the reaction were defined as follows: 1 (negative), 10 mm; 2 (positive), 10 mm; and 3 (strongly positive), 10 mm with induration at 48 h. Statistics. The effects of inactivation treatments were compared using Student s t test. Spearman s correlation coefficient by rank test was used to analyze the correlation between skin test and ELISPOT assay results or between gpelisa and skin test VZV-Specific Cell-Mediated Immunity Measurement JID 2008:198 (1 November) 1329

4 Figure 2. Age-associated decline in skin test reactivity and enzyme-linked immunospot (ELISPOT) count. Percentages of subjects with negative skin test results (A), ELISPOT counts (B), and glycoprotein antigen based ELISA (gpelisa) antibody titers (C) were compared between subjects aged years and those aged years. There were significant differences in the percentage of skin test negative subjects (Wilcoxon signed rank test) and the ELISPOT count (Student s t test) between age groups. In contrast, there were no significant differences in gpelisa antibody titer between these groups. PBMCs, peripheral blood mononuclear cells. results. For the comparisons of the ELISPOT counts or gpelisa antibody titers with skin test scores, the Mann-Whitney U test was applied. Age-associated decline in skin test reactivity or ELI- SPOT count was assessed by the Wilcoxon signed rank test or Student s t test. RESULTS Determination of optimum conditions for the ELISPOT assay. For the human IFN- ELISPOT assay, we first determined the optimum cell number. PBMCs were collected from 4 healthy volunteers and subjected to the ELISPOT assay. Each sample was plated at ,2 10 5,4 10 5, and cells/well in triplicate and stimulated with live VZV antigen solution or control solution. The spot count was performed automatically using the KS ELISPOT system. Mean ( SDs) spot counts were calculated after subtracting the control count from that of the VZV-stimulated sample wells. As shown in figure 1A, although the spot count increased almost in proportion to the cell number up to cells/well, there was an obvious reduction at cells/well, indicating that too high a cell number causes signal overlapping and hence a reduction in counting accuracy. On the other hand, at cells/well, the spot count decreased to 0 or almost 0 in 2 samples (samples B and D), indicating that too low a cell number can result in an underestimation of the differences between samples. Therefore, we used cells/well in the subsequent ELISPOT assays. For antigen stimulation, among the concentrations tested ( ,1 10 4, and pfu/ml), pfu/ml live varicella vaccine was determined to be effective. Live VZV antigen stimulates CD4 T cells to secrete IFN-, but it might also infect and replicate in T cells, which could also affect IFN- secretion. Therefore, to assess the amount of IFN- secretion caused by antigen stimulation, we examined the effects of inactivated VZV on the spot count (figure 1B). As shown in figure 1B, there was a significant reduction in spot-forming activity using heated VZV compared with untreated (live) VZV (P.02, Student s t test). On the other hand, up to 5000 J/m 2 UV-treated VZV had stimulation activity similar to that of the 1330 JID 2008:198 (1 November) Sadaoka et al.

5 control. In addition, we confirmed that UV treatment at 5000 J/m 2 completely inactivated the VZV (data not shown). Therefore, the subsequent ELISPOT assays were performed using UVtreated (5000 J/m 2 ) VZV for antigen stimulation. Skin test reactivity, ELISPOT assay, and gpelisa. Of the 151 subjects, 31 had a negative response by the skin test, but all of them had positive serologic results (antibody titer 50) by the gpelisa. The 31 subjects with negative skin test results included 4 vaccinees. As shown in table 2, both the rate of skin test positivity and the erythema diameter showed the tendency of an age-related decline, and the ELISPOT count showed the same pattern as the skin test. As expected, there were significant differences in the rate of skin test positivity and ELISPOT count between subjects in their 20s, 30s, or 40s and those in their 50s or 60s (figure 2A and 2B). The specific antibody titer obtained by the gpelisa remained relatively constant with increasing age (table 2 and figure 2C). There were no significant differences in the values when the results were stratified according to sex (table 2). Of the 151 subjects, 12 had a history of herpes zoster, 10 had received varicella vaccination, and 5 had both. We therefore investigated the correlation between these subjects and the remaining 124 individuals with respect to the ELISPOT and skin test results, but no relationship was seen. Correlation between skin test results and ELISPOT count. All of the values obtained from the VZV skin test, ELISPOT assay, and gpelisa are presented as scatter plots in figure 3. The graphs show the correlation between the VZV skin test and the ELISPOT assay (figure 3A), the gpelisa and the ELISPOT assay (figure 3B), and the gpelisa and the VZV skin test (figure 3C). As shown in figure 3A, the approximation curve indicates a positive linear relationship between the VZV skin test result and the ELISPOT count. In contrast, the slope of the approximation curve is smaller in figure 3B and 3C, indicating only a weak correlation between the gpelisa and the ELISPOT assay or the VZV skin test. Because the VZV skin test involves a rather complex in vivo reaction, minor variations in diameter may occur, depending on the physical condition of the subject. Therefore, the VZV skin test values were scored as follows: 1 (negative), 10 mm; 2 (positive), 10 mm; and 3 (strongly positive), 10 mm with induration. The average ELISPOT counts and gpelisa antibody titers were calculated for all of the subjects (figure 4). The ELISPOT counts increased in proportion with the skin test score (47.8 at score 1, 72.2 at score 2, and at score 3), with P values of.02 and.03 for the comparison between scores 2 and 3 and between scores 1 and 2, respectively (Mann-Whitney U test), indicating a significant correlation between the skin test and ELISPOT results (figure 4A). Furthermore, Spearman s correlation coefficient by rank test yielded a correlation coefficient of 0.43, indicating a moderately positive correlation between the 2 variables. On the other hand, the gpelisa results showed no Figure 3. Correlation between findings of the skin test and enzymelinked immunospot (ELISPOT) assay (A), glycoprotein antigen based ELISA (gpelisa) and ELISPOT assay (B), and gpelisa and skin test (C). The approximation curve is shown in each graph. PBMCs, peripheral blood mononuclear cells. significant differences when compared with the skin test scores (figure 4B) (correlation coefficient, 0.08). In addition, the coefficient was 0.21 when the gpelisa antibody titer was compared with the ELISPOT count, suggesting that the gpelisa antibody titer had only a weak or no linear association with the ELISPOT count or the skin test. Thus, the skin test results correlated well with the ELISPOT count but not with the gpelisa results. VZV-Specific Cell-Mediated Immunity Measurement JID 2008:198 (1 November) 1331

6 Figure 4. Significant correlation between the skin test response and enzyme-linked immunospot (ELISPOT) count. The skin test values were scored as follows: 1 (negative [ ]), 10 mm; 2 (positive [ ]), 10 mm; and 3 (strongly positive [ ]), 10 mm with induration. For each score, the average values of the subjects ELISPOT counts and glycoprotein antigen based (gpelisa) antibody titers are shown. Of the 151 volunteers enrolled, 18 for whom induration data were not obtained were excluded from this graph. There were significant differences among scores or between negative and positive test results for the ELISPOT results (A) but not for the gpelisa results (B) (Mann-Whitney U test). PBMCs, peripheral blood mononuclear cells. DISCUSSION Previous studies have indicated that a decline in CMI to VZV is closely correlated with the development of herpes zoster, although the mechanism of VZV reactivation in the neuron is not fully understood [7]. The waning of VZV-specific CMI occurs not only in seriously immunocompromised patients (such as those undergoing treatment for cancer or AIDS and recipients of organ transplants) but also in healthy elderly persons. A significant proportion of older subjects with herpes zoster develop postherpetic neuralgia, a chronic pain syndrome [17]. The pain is often long lasting and difficult to control; therefore, it has a severe impact on patients quality of life. Because the elderly population is rapidly increasing, there is an urgent need for effective ways to reduce the burden of this illness. The zoster vaccine effectively reduces the risk and severity of herpes zoster, especially in the elderly [7]. However, it is still not clear how long the protection is maintained or whether repeated doses should be given and, if so, at what intervals. Therefore, it is important to establish a reliable method for monitoring the state of immunity against VZV in adults that can be easily applied in postvaccination follow-up studies. Here, we performed 2 assays to monitor VZV-specific CMI in healthy individuals: the varicella skin test and the IFN- ELISPOT assay. In addition, humoral immunity was examined by gpelisa. The results were compared among subjects in their 20s, 30s, 40s, 50s, and 60s. All 151 subjects examined had detectable antibody titers, and the values were relatively constant across the subgroups. In contrast, both the varicella skin test and ELISPOT assay results indicated the tendency for CMI to decline with increasing age. These results agree with previous reports that CMI is more closely correlated with an increased risk of herpes zoster development than is humoral antibody titer [18, 19]. Marked increases in already high titers of circulating antibody are unlikely to occur even on reexposure to antigen, and it appears that CMI wanes independently of humoral immunity, which remains high. As such, serology is useless as a correlate for increasing susceptibility to herpes zoster. A correlation between the skin test and the IFN- ELISPOT assay was indicated by a positive linear relationship in the scatter plot, although there were several outliers. The discrepancy between the skin test result and the ELISPOT count represented by the outliers might be due to the different assay methods: compared with the in vitro ELISPOT assay, the in vivo skin test reaction is much more complex and is more likely to be affected by the physical condition of the subject. It is also possible that, in some subjects, the skin test reaction was provoked nonspecifically by non-vzv components in the injected solution or that false-negative results might have been caused by an inadequate injection. Therefore, for higher accuracy, the skin test might require a control solution injection besides the VZV antigen injection to measure the nonspecific reaction. In the ELISPOT assay, we observed that the size of the spots occasionally differed among the subjects examined, even among those showing the same number of spots. Although the spot number was used in the present analyses, the size of the spots also appeared to be an important marker for the strength of the im JID 2008:198 (1 November) Sadaoka et al.

7 mune response. Therefore, further study is required to analyze the relationship between the spot size in this assay and the strength of the VZV-specific immune response. The significant correlation between the results of the skin test and the ELISPOT assay was also indicated by the proportional change in the skin test score and the ELISPOT count (figure 4). Taken together, the results suggest that the skin test, like the ELISPOT assay, is valid for measuring CMI to VZV. Further study is needed to establish the immunoprotective threshold level of the skin test reaction for predicting an individual s susceptibility to herpes zoster. Nevertheless, our findings indicate that the skin test, which is clearly the easiest method developed thus far for measuring CMI to VZV, can be broadly applied to screen people at high risk for herpes zoster and that the results may predict those at greatest risk for reactivation, providing informative criteria for varicella vaccination. Furthermore, the skin test should be useful in longitudinal studies to estimate the durability of the protection provided by vaccination. Acknowledgments We thank Hironori Yoshii, Masaaki Matsuura, and Hiroko Oyaizu (National Institute of Biomedical Innovation) for technical assistance. References 1. Arvin AM. Varicella-zoster virus. Clin Microbiol Rev 1996; 9: Thomas SL, Hall AJ. What does epidemiology tell us about risk factors for herpes zoster? Lancet Infect Dis 2004; 4: Levin MJ, Barber D, Goldblatt E, et al. Use of a live attenuated varicella vaccine to boost varicella-specific immune responses in seropositive people 55 years of age and older: duration of booster effect. J Infect Dis 1998; 178(Suppl 1):S Berger R, Trannoy E, Hollander G, Bailleux F, Rudin C, Creusvaux H. A dose-response study of a live attenuated varicella-zoster virus (Oka strain) vaccine administered to adults 55 years of age and older. J Infect Dis 1998; 178(Suppl 1):S Levin MJ, Smith JG, Kaufhold RM, et al. Decline in varicella-zoster virus (VZV)-specific cell-mediated immunity with increasing age and boosting with a high-dose VZV vaccine. J Infect Dis 2003; 188: Takahashi M, Okada S, Miyagawa H, et al. Enhancement of immunity against VZV by giving live varicella vaccine to the elderly assessed by VZV skin test and IAHA, gpelisa antibody assay. Vaccine 2003; 21: Oxman MN, Levin MJ, Johnson GR, et al. A vaccine to prevent herpes zoster and postherpetic neuralgia in older adults. N Engl J Med 2005; 352: Hata A, Asanuma H, Rinki M, et al. Use of an inactivated varicella vaccine in recipients of hematopoietic-cell transplants. N Engl J Med 2002; 347: Kamiya H, Ihara T, Hattori A, et al. Diagnostic skin test reactions with varicella virus antigen and clinical application of the test. J Infect Dis 1977; 136: Baba K, Yabuuchi H, Okuni H, Takahashi M. Studies with live varicella vaccine and inactivated skin test antigen: protective effect of the vaccine and clinical application of the skin test. Pediatrics 1978; 61: Smith JG, Liu X, Kaufhold RM, Clair J, Caulfield MJ. Development and validation of a gamma interferon ELISPOT assay for quantization of cellular immune responses to varicella-zoster virus. Clin Diagn Lab Immunol 2001; 8: Smith JG, Levin M, Vessey R, et al. Measurement of cell-mediated immunity with a varicella-zoster virus-specific interferon-gamma ELI- SPOT assay: responses in an elderly population receiving a booster immunization. J Med Virol 2003; 70(Suppl 1):S Chilmonczyk BA, Levin MJ, McDuffy R, Hayward AR. Characterization of the human newborn response to herpesvirus antigen. J Immunol 1985; 134: Hayward AR, Herberger M. Lymphocyte responses to varicella zoster virus in the elderly. J Clin Immunol 1987; 7: Hayward AR, Zerbe GO, Levin MJ. Clinical application of responder cell frequency estimates with four years of follow up. J Immunol Methods 1994; 170: Provost PJ, Krah DL, Kuter BJ, et al. Antibody assays suitable for assessing immune responses to live varicella vaccine. Vaccine 1991; 9: Hornberger J, Robertus K. Cost-effectiveness of a vaccine to prevent herpes zoster and postherpetic neuralgia in older adults. Ann Intern Med 2006; 145: Gershon AA, Steinberg SP. Antibody responses to varicella-zoster virus and the role of antibody in host defense. Am J Med Sci 1981; 282: Nagasawa K, Yamauchi Y, Tada Y, Kusaba T, Niho Y, Yoshikawa H. High incidence of herpes zoster in patients with systemic lupus erythematosus: an immunological analysis. Ann Rheum Dis 1990; 49: VZV-Specific Cell-Mediated Immunity Measurement JID 2008:198 (1 November) 1333