The Journal of Infectious Diseases

BRIEF REPORT

Acyclovir Prophylaxis Reduces the among HIV-uninfected persons: among adults aged ≥65 years in
the United States, the licensed live attenuated zoster vaccine (Zos-
Incidence of Herpes Zoster Among tavax) reduced the incidence of herpes zoster by 48% (95% con-
HIV-Infected Individuals: Results of a fidence interval [CI], 39%–56%) [2]. The zoster vaccine is safe and
Randomized Clinical Trial immunogenic for HIV-infected persons with a CD4+ T-cell count
of >200 cells/µL [3], but it is recommended for use starting at age
Ruanne V. Barnabas,1,2,3,6 Jared M. Baeten,1,2,3 Jairam R. Lingappa,1,2
Katherine K. Thomas,1 James P. Hughes,4,6 Nelly R. Mugo,1 Sinead Delany-Moretlwe,7
60 years and is contraindicated in persons with CD4+ T-cell
Glenda Gray,9 Helen Rees,7 Andrew Mujugira,1,3 Allan Ronald,10 Wendy Stevens,8 counts of <200 cells/µL [4].
Saidi Kapiga,11 Anna Wald,1,2,5,6 and Connie Celum1,2,3; for the Partners in Prevention
Herpes zoster incidence is 12–17-fold higher in HIV-infected
HSV/HIV Transmission Study Team
persons (2.94 cases/100 person-years), compared with HIV-
Departments of 1Global Health, 2Medicine, 3Epidemiology, 4Biostatistics, 5Laboratory Medicine,
University of Washington, and 6Vaccine and Infectious Diseases Division, Fred Hutchinson uninfected individuals (0.20 cases/100 person-years) [5]. Even
Cancer Research Center, Seattle, Washington; 7Wits Reproductive Health and HIV Institute, and with antiretroviral therapy (ART) initiation, the incidence of her-
8
Department of Molecular Medicine and Haematology, University of Witwatersrand,
Johannesburg, 9Perinatal HIV Research Unit, Soweto, South Africa; 10Department of Medicine, pes zoster remains 2–3-fold higher among HIV-infected persons
University of Manitoba, Winnipeg, Canada; and 11Department of Infectious Disease and, notably, is higher immediately after ART initiation [6].
Epidemiology, London School of Hygiene and Tropical Medicine, United Kingdom
High-dose acyclovir (2.4 g daily) reduced zoster recurrence at
12 months by 68% among HIV-infected persons in the pre-
Human immunodeficiency virus (HIV)–infected persons have
higher rates of herpes zoster than HIV-uninfected individuals. ART era [7], but there are limited data and no guidelines for
We assessed whether twice daily treatment with 400 mg of oral standard, lower-dose antiviral prophylaxis dosing for prevention
acyclovir reduces the incidence of herpes zoster in a random- of zoster [4]. We hypothesized that daily acyclovir prophylaxis
ized, double-blind, placebo-controlled trial among 3408 per- would reduce the incidence of herpes zoster among HIV-infected
sons coinfected with HIV and herpes simplex virus type individuals.
2. During 5175 person-years of follow-up, 26 cases of herpes
zoster occurred among those assigned acyclovir, compared METHODS
with 69 cases among those assigned placebo (rates, 1.00 and Study Design and Participants
2.68/100 person-years, respectively), a relative decrease of 62%
Between November 2004 and April 2007, we conducted the
(hazard ratio, 0.38; 95% confidence interval, .24–.67; P < .001).
Partners in Prevention HSV/HIV Transmission Study [8], a
Daily acyclovir prophylaxis significantly reduced herpes zoster
randomized, double-blind, placebo-controlled trial of acyclovir
incidence among HIV-infected persons.
Keywords. herpes zoster; acyclovir; HIV; acyclovir prophy- (400 mg twice daily), among 3408 HIV-infected women and
laxis; shingles; VZV. men from 7 African countries (Botswana, Kenya, Rwanda,
South Africa, Tanzania, Uganda, and Zambia) who were coin-
fected with herpes simplex virus type 2 (HSV-2). The primary
Herpes zoster, or shingles, is a painful, vesicular rash that most
aim of the study was to assess the impact of acyclovir prophy-
often develops in elderly or immunocompromised individuals, in-
laxis on HIV transmission to HIV-uninfected heterosexual
cluding human immunodeficiency virus (HIV)–infected persons
partners; as previously reported, no reduction in HIV transmis-
[1]. This localized cutaneous eruption results from reactivation of
sion was seen, although acyclovir prophylaxis did reduce the
varicella-zoster virus (VZV), which lies dormant in the dorsal
frequency of genital ulcers due to HSV-2 and resulted in a mod-
root ganglia following primary varicella virus infection (chicken-
est reduction in plasma HIV RNA load (0.25 log10 copies/mL)
pox) during childhood. Prompt initiation of therapy with high-
[8]. A predefined secondary outcome of the trial was incident
dose acyclovir, valacyclovir, or famciclovir limits the severity, du-
herpes zoster. VZV serologic assays were not performed. Eligi-
ration, and complications of the outbreak. Vaccination is effective
ble participants reported no current use of ART, and all had
CD4+ T-cell counts of >250 cells/µL and no history of AIDS-
Received 7 April 2015; accepted 27 May 2015; published online 8 June 2015. defining conditions. The University of Washington Human
Presented in part: 20th Conference on Retroviruses and Opportunistic Infections, Atlanta, Subjects Review Committee and ethics review committees at
Georgia, 5–8 March 2012.
Correspondence: R. V. Barnabas, International Clinical Research Center, Department of Global the each of the collaborating organizations approved the study
Health, University of Washington, UW Box 359927, 325 Ninth Ave, Seattle, WA 98104 protocol. All participants provided written informed consent.
([email protected]).
The Journal of Infectious Diseases® 2016;213:551–5 Procedures
© The Author 2015. Published by Oxford University Press for the Infectious Diseases Society
of America. All rights reserved. For permissions, e-mail [email protected].
Throughout the study, all participants received individualized,
DOI: 10.1093/infdis/jiv318 confidential HIV counseling, risk-reduction counseling,

BRIEF REPORT • JID 2016:213 (15 February) • 551

couples counseling, free condoms, and treatment of sexually 462 cells/mL, and randomization was balanced with respect to
transmitted infections according to World Health Organization these characteristics. Four percent in each arm reported a histo-
guidelines. HIV-infected participants were seen once per month ry of herpes zoster in the prior year, and 1% in each arm had
for provision of study medication (acyclovir 400 mg twice daily zoster on baseline clinical examination.
or matching placebo). CD4+ T-cell counts were assessed every 6 During 24 months, the study accrued 2605 person-years of
months, and the plasma HIV RNA load was measured at base- follow-up in the acyclovir arm and 2570 person-years of fol-
line. Herpes zoster detected at a monthly visit was documented low-up in the placebo arm. The median follow-up time was
during physical examination; once each quarter, a medical his- 20 months. Acyclovir adherence, assessed by pill count and
tory was obtained to record self-reported zoster not observed at self-report, was high, with an estimated 96% of doses dispensed
a regular visit. During the quarterly medical visits, clinicians consumed and drug consumed on 90% of days overall. A total
asked participants whether they had had any skin rash consis- of 151 participants in the acyclovir arm initiated ART, com-
tent with zoster and recorded reported cases. pared with 180 participants in the placebo arm [9], resulting
At the quarterly visits, participants were asked whether they had in 125 and 158 person-years at risk while receiving ART in
received ART since the last visit, and whether those who initiated the acyclovir and placebo groups, respectively.
ART continued to receive treatment during follow-up. At the time Ninety-five cases of herpes zoster occurred during follow-up,
the study was undertaken, national guidelines generally recom- of which 15 were observed by skin examination only, 62 were
mended ART initiation for patients with CD4+ T-cell counts of self-reported as occurring between study visits, and 18 were
≤200–250 cells/µL or clinical AIDS. Participants who met national documented by both examination and self-report. Of these 95
guidelines for initiation of ART during follow-up, as a result of a herpes zoster events, 26 occurred in 22 people in the acyclovir
decline in CD4+ T-cell count or a change in clinical status, were arm (incidence, 1.00 cases/100 person-years), compared with
referred to local HIV clinics for ART. HIV-infected women who 69 in 64 people in the placebo arm (incidence, 2.68 cases/100
became pregnant during the study were referred to antenatal clin- person-years; Table 1). The cumulative proportion of partici-
ics for services to prevent mother-to-child transmission. pants with herpes zoster during the study was 4.5% (95% CI,
3.5%–5.8%) in the placebo arm and 1.7% (1.1%–2.5%) in the
Statistical Analysis acyclovir arm (P < .001 for the difference in time to first
For the present analysis, the primary outcome was an episode of event; Figure 1). Overall, analysis of all events revealed that acy-
herpes zoster during follow-up, either observed on examination clovir prophylaxis decreased the incidence of herpes zoster by
or reported by the participant. The primary analysis was inten- 62% in the acyclovir arm, compared with the placebo arm (haz-
tion to treat, and survival analysis was used to estimate the effect ard ratio [HR], 0.38; 95% CI, .24–.67; P < .001). The protective
of acyclovir prophylaxis on herpes zoster incidence in the acy- effect of acyclovir on zoster was not modified by sex, age (<30
clovir arm, compared with the placebo arm. Multiple visits with years and ≥30 years), CD4+ T-cell count (≤350 and >350 cells/
an event within the same quarter were rare and were treated as µL; ≤250 and >250 cells/µL), or baseline plasma HIV RNA load
one episode; events in different quarters were treated as separate (<10 000, 10 000–99 999, and ≥100 000 copies/mL).
events. Cox regression with the Anderson–Gill counting meth- There were 6 cases of herpes zoster among 6 participants who
od and robust variance was used to include multiple events per initiated ART after study enrollment; all were in the placebo
participant in the survival analysis. A second survival analysis, arm, and all occurred during the first 3 months after ART ini-
using only time to the first zoster event, was used to create cu- tiation (range, 3–77 days). Incidence in the placebo arm was
mulative probability curves for zoster incidence and to compare 3.80 cases/100 person-years, compared with 0.00 cases/100 per-
curves by arm, using the log-rank test. To explore the impact of son-years in the acyclovir arm (HR, 0.00; 95% CI, .00–1.08).
ART on zoster incidence, and because HIV-infected persons re- ART did not significantly modify the effect of acyclovir on zos-
ceiving ART continue to have an increased risk of zoster as ter (P = .41), but the number of participants receiving ART was
compared to uninfected individuals [6] and because ART initi- small. Acyclovir did not prevent zoster recurrences among per-
ation is associated with a short-term increase in the risk of her- sons reporting prior zoster episodes at baseline (HR, 1.82; 95%
pes zoster [6], data were not censored at ART initiation. Data CI, .33–9.96) but did prevent recurrences among participants
were analyzed using SAS (version 9.3; Cary, North Carolina). who had not experienced zoster prior to enrollment (HR,
0.29; 95% CI, .18–.49; P for interaction = .01).
RESULTS
DISCUSSION
The Partners in Prevention HSV/HIV Transmission Study en-
rolled 3381 participant infected with HIV and HSV-2, with In this randomized, double-blind, placebo-controlled trial of
1693 were randomized to acyclovir and 1688 to placebo. As de- acyclovir 400 mg twice daily among African HIV-infected per-
tailed elsewhere [8], two thirds of participants were women, the sons, acyclovir prophylaxis substantially reduced herpes zoster
median age was 32 years, the median CD4+ T-cell count was events by 62% regardless of sex, age, CD4+ T-cell count, plasma

552 • JID 2016:213 (15 February) • BRIEF REPORT

Table 1. Herpes Zoster Incidence, by Treatment Arm

Acyclovir Arm Placebo Arm

Subjects, Events, PY at Rate, Cases/ Subjects, Events, PY at Rate, Cases/ P

Variable No. No. Risk 100 PY No. No. Risk 100 PY HR (95% CI)a Value

Overall efficacy 1693 26 2605 1.00 1688 69 2570 2.68 0.38 (.24–.67) <.001
Zoster observed on examination 1693 12 2604 0.46 1688 21 2570 0.82 0.57 (.22–1.68) .33
Report of zoster 1693 16 2603 0.61 1688 64 2568 2.49 0.25 (.14–.43) <.001
Subgroup, based on enrollment characteristics
Sex
Male 561 10 875 1.14 536 28 838 3.34 0.34 (.16–.70)
Female 1132 16 1730 0.93 1152 41 1732 2.37 0.41 (.18–.91)
Effect modificationc .75
Age, y
<30 664 13 1011 1.29 665 24 991 2.42 0.55 (.28–1.48)
≥30 1028 13 1593 0.82 1023 45 1579 2.85 0.29 (.16–.54)
Effect modificationc .24
CD4+ T-cell count, cells/µL
<350 424 8 648 1.23 461 34 703 4.84 0.27 (.13–.59)
≥350 1269 18 1957 0.92 1227 35 1867 1.87 0.50 (.23–1.06)
Effect modificationc .22
≤250 4 0 7 0.00 6 0 9 0.00 . . .
>250 1689 26 2598 1.00 1682 69 2561 2.69 0.38 (.22–.67)
Effect modificationc >.99
Viral load, copies/mL
<10 000 778 6 1190 0.50 800 17 1214 1.40 0.37 (.15–.92)
10 000–99 999 645 12 1004 1.20 641 28 986 2.84 0.44 (.16–1.17)
≥100 000 253 8 325 2.08 232 24 348 6.90 0.30 (.13–0.68)
Effect modificationc .85
After enrollment
ART use in period before clinical visit
Yes . . . 0 125 0.00 . . . 6 158 3.80 0.00 (.00–1.08)b
No . . . 26 2480 1.05 . . . 63 2412 2.61 0.38 (.24–.60)
Effect modificationd .41

Abbreviations: ART, antiretroviral therapy; CI, confidence interval; HR, hazard ratio; PY, person-years.
a
Unless otherwise indicated, data were determined using Cox regression with the Anderson–Gill counting method, to allow multiple outcomes per participant.
b
Data denote the incidence rate ratio and exact 95% CI.
c
Evaluated using the likelihood ratio test.
d
Evaluated using exact logistic regression, to model discrete survival time.

HIV RNA load, or ART use. This reduction in herpes zoster The limitations of our study were the small numbers of par-
events was consistent with a study from 1999 that found a ticipants with low CD4+ T-cell counts, because such persons
68% reduction in herpes zoster with high-dose acyclovir [7]. were excluded from enrollment. Previous studies have found
The incidence of herpes zoster (2.68 cases/100 person-years an increased incidence of herpes zoster with declining CD4+
in the placebo arm) in this population was comparable to pre- T-cell count [11, 12], which we did not observe, perhaps because
vious estimates among HIV-infected individuals (2.90 cases/ of limited power. Some studies have shown an increase in the
100 person-years), which is 12–17-fold higher than their age- risk of herpes zoster immediately after ART initiation [6],
matched HIV-uninfected controls [5]. which is thought to be due to immune reconstitution inflamma-
Among immunocompromised individuals, including HIV- tory syndrome. We did not observe ART to alter the incidence
infected persons, herpes zoster may have a prolonged course of herpes zoster events, although the number of individuals and
and disease recurrence is more common [4]. Acyclovir prophy- duration of ART in our study were small.
laxis did not prevent zoster recurrence among persons who re- Current Advisory Committee on Immunization Practices
ported prior episodes in our study, which might indicate a guidelines recommend zoster immunization for immunocom-
defect in VZV-specific T-cell immunity [10]; HIV-infected in- petent persons beginning at age 60 years. Because of the risk
dividuals with prior zoster may require a higher acyclovir dose of vaccine-associated zoster, the zoster vaccine is contraindicat-
for prophylaxis. ed in immunocompromised persons, including HIV-infected

BRIEF REPORT • JID 2016:213 (15 February) • 553

 Figure 1. Cumulative probability of herpes zoster disease, by time to the first zoster event. Recurrent episodes of zoster are not captured in this figure.

persons with CD4+ T-cell count of <200 cells/µL [4, 13]. Immu- approved the study protocol. All participants provided written informed
consent.
nosuppressed persons, including transplant recipients, who re-
Disclaimer. The work is solely the responsibility of the authors and
ceive antiviral prophylaxis against cytomegalovirus (CMV) or does not necessarily represent the official views of the National Institutes
HSV with valganciclovir, ganciclovir, valacyclovir, or acyclovir of Health (NIH).
have a lower risk of herpes zoster [14]. Our data contribute to Financial support. This work was supported by the Bill and Melinda
Gates Foundation (grant 26469) and the National Institute of Allergy and In-
this evidence by demonstrating prevention of herpes zoster in fectious Diseases, National Cancer Institute, National Institute of Mental
the setting of acyclovir suppressive therapy for HSV-2. Health, National Institute on Drug Abuse, Eunice Kennedy Shriver National
In summary, in this large placebo-controlled trial of daily Institute of Child Health and Human Development, National Heart, Lung,
and Blood Institute, National Institute on Aging, National Institute of General
acyclovir (400 mg twice daily) for HSV-2 suppression, we
Medical Sciences, and National Institute of Diabetes & Digestive & Kidney Dis-
found a 68% reduction in zoster events on the acyclovir arm. eases of the NIH (grant P30AI027757).
Acyclovir is well tolerated, safe, and affordable as a generic Potential conflicts of interest. All authors: No reported conflicts. All
authors have submitted the ICMJE Form for Disclosure of Potential Con-
product, and acyclovir prophylaxis has benefits that include
flicts of Interest. Conflicts that the editors consider relevant to the content
a 73% reduction in HSV-2 genital ulcer disease [8] and a mod- of the manuscript have been disclosed
est (16%) decrease in HIV disease progression [9]. For immu-
nosuppressed patients on antiviral prophylaxis for CMV or References
HSV, prevention of herpes zoster is an additional benefit. Fu- 1. Gnann JW Jr, Whitley RJ. Clinical practice. Herpes zoster. N Engl J Med 2002;
ture research should include a focus on the impact of acyclovir 347:340–6.
2. Langan SM, Smeeth L, Margolis DJ, Thomas SL. Herpes zoster vaccine effective-
(or related antiviral) prophylaxis on herpes zoster incidence ness against incident herpes zoster and post-herpetic neuralgia in an older US
among HIV-infected persons with a low CD4+ T-cell count population: a cohort study. PLoS Med 2013; 10:e1001420.
3. Benson C, Hua L, Anderson JW, et al. Zostavax is generally safe and immunogenic
and among HIV-infected individuals initiating ART, which in HIV+ adults virologically suppressed on ART: results of a phase 2, randomized,
are the factors associated with greatest frequency of zoster double-blind, placebo-controlled trial. Presented at: 19th Conference on Retrovi-
ruses and Opportunistic Infections, Seattle, Washington, 2012.
reactivation. 4. Panel on Opportunistic Infections in HIV-Infected Adults and Adolescents.
Guidelines for the prevention and treatment of opportunistic infections in HIV-
Notes infected adults and adolescents: recommendations from the Centers for Disease
Acknowledgments. R. V. B. oversaw the analysis and wrote the first Control and Prevention, the National Institutes of Health, and the HIV Medicine
Association of the Infectious Diseases Society of America. http://aidsinfo.nih.gov/
draft of the paper, which was revised by all authors. All authors contributed
contentfiles/lvguidelines/adult_oi.pdf. Accessed 16 May 2014.
to data analysis, as well as to the interpretation of findings. K. K. T. did the
5. Buchbinder SP, Katz MH, Hessol NA, et al. Herpes zoster and human immuno-
statistical analysis, with input from R. V. B., J. M. B., J. P. H., J. R. L., and deficiency virus infection. J Infect Dis 1992; 166:1153–6.
C. C. All authors approved the final version of the paper for submission. 6. Grabar S, Tattevin P, Selinger-Leneman H, et al. Incidence of herpes zoster in
The University of Washington human subjects review committee and HIV-infected adults in the combined antiretroviral therapy era: results from the
ethics review committees at the each of the collaborating organizations FHDH-ANRS CO4 cohort. Clin Infect Dis 2015; 60:1269–77.

554 • JID 2016:213 (15 February) • BRIEF REPORT

 7. Leautez S, Bani-Sadr F, Billaud E, Raffi F. [Secondary prophylaxis for herpes zoster 11. Shearer K, Maskew M, Ajayi T, et al. Incidence and predictors of herpes zoster
wi oral acyclovir in HIV patients]. Pathol Biol (Paris) 1999; 47:570–2. among antiretroviral therapy-naive patients initiating HIV treatment in Johannes-
8. Celum C, Wald A, Lingappa JR, et al. Acyclovir and transmission of HIV-1 burg, South Africa. Int J Infect Dis 2014; 23:56–62.
from persons infected with HIV-1 and HSV-2. N Engl J Med 2010; 362: 12. Gebo KA, Kalyani R, Moore RD, Polydefkis MJ. The incidence of, risk factors for,
427–39. and sequelae of herpes zoster among HIV patients in the highly active antiretro-
9. Lingappa JR, Baeten JM, Wald A, et al. Daily acyclovir for HIV-1 disease progres- viral therapy era. J Acquir Immune Defic Syndr 2005; 40:169–74.
sion in people dually infected with HIV-1 and herpes simplex virus type 2: a rand- 13. Bhalla P, Forrest GN, Gershon M, et al. Disseminated, persistent, and fatal infec-
omised placebo-controlled trial. Lancet 2010; 375:824–33. tion due to the vaccine strain of varicella-zoster virus in an adult following stem
10. Levy O, Orange JS, Hibberd P, et al. Disseminated varicella infection due to the cell transplantation. Clin Infect Dis 2015; 60:1068–74.
vaccine strain of varicella-zoster virus, in a patient with a novel deficiency in nat- 14. Pergam SA, Limaye AP, AST Infectious Diseases Community of Practice. Varicella
ural killer T cells. J Infect Dis 2003; 188:948–53. zoster virus in solid organ transplantation. Am J Transplant 2013; 13(suppl 4):138–46.

BRIEF REPORT • JID 2016:213 (15 February) • 555