Recommendations of the Advisory Committee on Immunization Practices for Use of Herpes Zoster Vaccines

Kathleen L. Dooling, MD; Angela Guo, MPH; Manisha Patel, MD; Grace M. Lee, MD; Kelly Moore, MD; Edward A. Belongia, MD; Rafael Harpaz, MD


Morbidity and Mortality Weekly Report. 2018;67(3):103-108. 

In This Article

Summary of Findings

As a result of the GRADE process, key outcomes were designated as critical (prevention of herpes zoster and postherpetic neuralgia, serious adverse events following vaccination) or important (duration of protection, reactogenicity). All outcomes were considered for both RZV and ZVL compared with no vaccination. There were no clinical studies that compared the vaccines directly with one another (head-to-head). Supporting evidence for the Work Group's findings is available online ([22]

Recombinant Zoster Vaccine (RZV)

Efficacy of RZV was evaluated in a two-part, phase III multicenter clinical trial which enrolled >30,000 participants, who were randomized 1:1 to receive vaccine or saline placebo.[14,15] The median follow-up time was 3.2 years for Zoster Efficacy Study in Adults 50 Years of Age or Older (ZOE-50),[14] and 3.7 years for Zoster Efficacy Study in Adults 70 Years of Age or Older (ZOE-70).[15] The efficacy for the prevention of herpes zoster was 96.6% (95% confidence interval [CI] = 89.6–99.3) in persons aged 50–59 years and 97.4% (95% CI = 90.1–99.7) in persons aged 60–69 years.[14] Using pooled data from both study arms, vaccine efficacy was 91.3% (95% CI = 86.8–94.5) in participants aged ≥70 years.[15] Vaccine efficacy in the first year after vaccination was 97.6% (95% CI = 90.9–99.8) and was 84.7% (95% CI = 69.0–93.4) or higher for the remaining 3 years of the study in persons aged ≥70 years. Efficacy for prevention of postherpetic neuralgia was 91.2% (95% CI = 75.9–97.7) in adults aged ≥50 years and 88.8% (95% CI = 68.7–97.1) in those aged ≥70 years.[15]

Serious adverse events (an undesirable experience associated with the vaccine that results in death, hospitalization, disability or requires medical or surgical intervention to prevent a serious outcome) were examined in eight studies sponsored by GSK, which included 29,965 subjects (15,264 RZV recipients).[22] Overall, rates of serious adverse events over the study periods were similar in the RZV and placebo groups.

Injection-site and systemic grade 3 solicited adverse events (reactions related to vaccination which were severe enough to prevent normal activities) were actively surveyed in eight studies involving 10,590 subjects.[22] Among the subset of subjects completing the 7-day diary card for reactogenicity in phase III clinical trials (9,936), 16.5% of vaccine recipients reported any grade 3 adverse event compared with 3.1% of placebo recipients.[14,15] Grade 3 injection-site reactions (pain, redness, and swelling) were reported by 9.4% of vaccine recipients, compared with 0.3% of placebo recipients and grade 3 solicited systemic events (myalgia, fatigue, headache, shivering, fever, and gastrointestinal symptoms) were reported by 10.8% of vaccine recipients and 2.4% of placebo recipients.[14,15] Whereas there were no differences in the proportions of local grade 3 reactions between dose 1 and dose 2, systemic grade 3 reactions were reported more frequently after dose 2.[1] Overall, the most common solicited adverse reactions (grade 1–3) were pain (78%), myalgia (45%), and fatigue (45%).[1]

Zoster Vaccine Live (ZVL)

Two randomized clinical trials and seven observational studies were reviewed to evaluate the performance of a single dose of ZVL in preventing herpes zoster.[22] A randomized clinical trial in persons aged 50–59 years found that the efficacy was 70% (95% CI = 54–81) (median follow-up time was 1.3 years).[12] A randomized trial in persons aged ≥60 years found that the efficacy was 64% (95% CI = 56–71) in persons aged 60–69 years and 38% (95% CI = 25–48) in persons aged ≥70 (median follow-up time was 3.1 years).[4] Estimates from observational studies and randomized controlled trials (RCTs) are consistent; observational estimates are within the 95% CI of the RCT estimates.[22] The duration of protection has been studied out to 11 years, including the first 4 years of the RCT and then follow-on, nonblinded studies which used a modeled control group from years 7–11.[4,10,11] Shorter follow-up periods have been evaluated in observational studies using administrative health data.[22] Studies concur that there is a substantial decrease in effectiveness following the first year after receipt of ZVL, and, by 6 years postvaccination, vaccine effectiveness against herpes zoster is <35%.[10,23–25] During years 7–8 postvaccination, observational study estimates of effectiveness ranged from 21%–32%.[23,24] In the longest study of ZVL, estimates of effectiveness were no longer statistically significant 9–11 years postvaccination.[11] In a phase III clinical trial, vaccine efficacy against post herpetic neuralgia was 65.7% (95% CI = 20.4–86.7) in persons aged 60–69 years and 66.8% (95% CI = 43.3–81.3) in participants aged ≥70 years (median follow-up of 3.1 years);[4] these estimates are consistent with estimates from observational studies.[22] Notably, in observational studies, vaccine effectiveness against postherpetic neuralgia was longer-lasting than effectiveness against herpes zoster itself.[23,26]

Serious adverse events related to ZVL were examined in eight high quality RCTs, 13 RCTs with limitations, and an additional seven observational studies.[22] Overall, serious adverse events occurred at similar rates in vaccinated and placebo groups. Whereas injection site reactions were reported in 48% of vaccine recipients and 17% of placebo recipients in phase III clinical trials, post hoc analysis indicates that no more than 0.9% of vaccine recipients reported any given injection site symptom as grade 3.[22] In addition, in rare instances, ZVL vaccine strain has been documented to cause disseminated rash as well as herpes zoster in immunocompetent recipients,[22,27] and life-threatening and fatal complications in immunocompromised recipients.[28,29]

Cost Effectiveness

The CDC analysis was conducted from a societal perspective over a lifetime. It estimated that vaccination with RZV, compared with no vaccination, cost $31,000 per quality adjusted life year (QALY), on average, for immunocompetent adults aged ≥50 years. The numbers of persons needed to be vaccinated with RZV to prevent one case of herpes zoster and one case of postherpetic neuralgia are 11–17 and 70–187, respectively. Estimates of costs per QALY for vaccination with RZV 8 weeks following ZVL (estimated by immediate revaccination in the model) ranged from $15,000 per QALY in persons aged 80–89 years to $117,000 per QALY for persons aged 50–59 years. Under most assumptions, vaccination with RZV prevented more disease at lower overall costs than did vaccination with ZVL. In probabilistic sensitivity analyses, 73.5% 2-dose completion (range = 38.8%–96.3%) coupled with 1-dose initial effectiveness estimates of 90% and 69% were applied, and RZV remained the most cost-effective strategy.[13]

ACIP also reviewed independent cost-effectiveness analyses by an academic group,[18] GSK,[19] and Merck (Merck, unpublished data, 2017). The academic group estimated RZV costs per QALY of $30,000 when vaccination occurred at age 60 years. The GSK model estimated RZV costs per QALY of $12,000, on average, for recipients aged ≥60 years. Although analytic approaches and model inputs differed, both groups found that RZV was more cost effective than ZVL. Merck modeled vaccination at age ≥60 years and estimated $107,000 per QALY for RZV and $83,000 per QALY for ZVL, with ZVL as the most cost-effective vaccine in most scenarios.