MAJOR ARTICLE

Herpes Simplex Virus (HSV) Suppression with

Valacyclovir Reduces Rectal and Blood Plasma
HIV-1 Levels in HIV-1/HSV-2Seropositive Men:
A Randomized, Double-Blind, Placebo-Controlled
Crossover Trial
Richard A. Zuckerman,1 Aldo Lucchetti,6 William L. H. Whittington,2 Jorge Snchez,6 Robert W. Coombs,2,3
Rosario Zuiga,6 Amalia S. Magaret,3 Anna Wald,2,3,4,5 Lawrence Corey,2,3,5 and Connie Celum2,4
1Section of Infectious Disease and International Health, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire; Departments of
2Medicine, 3Laboratory Medicine, and 4Epidemiology, University of Washington, and 5Program in Infectious Diseases, Fred Hutchinson Cancer

Downloaded from http://jid.oxfordjournals.org/ by guest on June 14, 2016

Research Center, Seattle; 6Asociacin Civil Impacta Salud y Educacin, Lima, Peru

Background. Herpes simplex virus type 2 (HSV-2) infection is common among human immunodeficiency virus
(HIV)infected persons, and HSV reactivation increases plasma and genital HIV-1 levels. We studied HIV-1 levels
during HSV suppression in coinfected persons in a placebo-controlled crossover trial.
Methods. Twenty antiretroviral therapy (ART)naive HIV-1/HSV-2seropositive men who have sex with men in
Lima, Peru, with CD4 cell counts 200 cells/L were randomized to receive either valacyclovir at 500 mg twice daily
or placebo for 8 weeks, after which they underwent a 2-week washout period and then received the alternative regimen
for 8 weeks. Specimens included daily anogenital swabs (for HSV DNA polymerase chain reaction [PCR]), thrice
weekly rectal mucosal secretions (for HIV-1 RNA and HSV DNA PCR) obtained by anoscopy, and weekly plasma (for
HIV-1 RNA PCR). Outcomes were rectal and plasma HIV-1 RNA levels by treatment arm.
Results. HIV-1 was detected in 73% of 844 rectal and 99% of 288 plasma specimens. HSV was detected in 29%
and 4% of mucocutaneous specimens obtained during placebo and valacyclovir administration, respectively
(P .001). Valacyclovir resulted in a 0.16 (95% confidence interval [CI], 0.07 0.25; P .0008; 33% decrease) log10
copies/mL lower mean within-subject rectal HIV-1 level and a 0.33 (95% CI, 0.23 0.42; P .0001; 53% decrease)
log10 copies/mL lower plasma HIV-1 level, compared with values for placebo.
Conclusions. Valacyclovir significantly reduces rectal and plasma HIV-1 levels in HIV-1/HSV-2 coinfected men.
HSV suppression may provide clinical benefits to persons not receiving highly active ART as well as public health benefits.
Trial registration. ClinicalTrials.gov identifier: NCT00378976.

Most HIV-infected persons are also infected with herpes
simplex virus type 2 (HSV-2) [1]. The risk of HIV-1
Received 2 April 2007; accepted 27 April 2007; electronically published 31
October 2007.
Potential conflicts of interest: C.C. has received research grant support from the
National Institutes of Health (NIH), the Bill and Melinda Gates Foundation, and
GlaxoSmithKline (GSK) and has served on an advisory board for GSK. J.S. has
received grant support from the NIH and GSK. A.W. has received grant support
from the NIH, GSK, Antigenics, 3M, Roche, and Vical; she is a consultant for
Novartis, PowderMed, and MediGene and is a speaker for Merck Vaccines. The
University of Washington Virology Division Laboratories have received grant
funding from GSK and Novartis to perform herpes simplex virus serologic assays
and polymerase chain reaction assays for studies funded by these companies. L.C.
directs these laboratories. He receives no salary support from these grants.
The Journal of Infectious Diseases 2007; 196:1500 8
2007 by the Infectious Diseases Society of America. All rights reserved.
0022-1899/2007/19610-0013$15.00
DOI: 10.1086/522523
transmission is higher in HIV-1 serodiscordant couples
when the source partner has reported recent genital ul-
cers [2]. Plasma and genital HIV-1 levels are increased
during both symptomatic and asymptomatic HSV reac-
tivations [3 6]. In vitro studies have demonstrated that
HSV proteins increase HIV-1 expression [79], HSV
coinfection of HIV-infected cells [10], and levels of pro-
inflammatory cytokines during HSV reactivation, which
Presented in part: 44th Annual Meeting of the Infectious Diseases Society of
America, Toronto, 1115 October 2006 (abstract LB-25).
Financial support: GlaxoSmithKline (research grant R103); National Institutes of
Health (Centers for AIDS Research Clinical Research and Laboratory Core Grants
AI-27757 and AI-38858, R37 AI-42528, and HSV Program Project Grant AI-30731).
Reprints or correspondence: Dr. Connie Celum, University of Washington,
Harborview Medical Center, Box 359927, 325 9th Ave., Seattle, WA 98104
(ccelum@u.washington.edu).

1500 JID 2007:196 (15 November) Zuckerman et al.

increase HIV-1 replication [11]. These observations support the
hypothesis that, by increasing HIV-1 replication, HSV may have
clinical consequences for coinfected persons as well as public
health consequences.
Proof-of-concept studies among HIV/HSV-coinfected per-
sons are needed to assess whether HSV suppression consistently
decreases HIV-1 levels in plasma and genital secretions. Daily
suppressive therapy for HSV infection is highly effective in re-
ducing both clinical and subclinical HSV reactivation in HIV-
infected persons [12, 13]. A randomized trial in Burkina Faso
recently showed that suppressive valacyclovir significantly re-
duced cervical and plasma HIV-1 levels [14]. This observation is
consistent with a pooled analysis of 8 studies conducted in the
1990s of high-dose acyclovir in combination with mono- or
dual-nucleoside antiretroviral therapy (ART), which indicated a
survival benefit among HIV-infected persons who received acy-
clovir [15].
To evaluate the effect of HSV-2 suppression on anogenital
and plasma HIV-1 levels among men, we conducted a random-
ized, double-blind, placebo-controlled crossover trial of daily
valacyclovir among ART-naive HIV-1/HSV-2 coinfected men
who have sex with men (MSM) with CD4 cell counts 200
cells/L in Lima, Peru.
METHODS
Study Characteristics
Study design. A randomized, double-blind, placebo-controlled
crossover trial of valacyclovir for HSV and HIV-1 suppression
was conducted at the Asociacin Civil Impacta Salud y Edu-
cacin, a research organization in Lima. Eligible persons were
MSM who were 18 years old, were seropositive for HIV-1 and
HSV-2, had no history of antiretroviral use, and had a CD4 cell
count 200 cells/L. Exclusion criteria included current or
planned therapy with antiretrovirals or herpes antivirals (acyclo-
vir, famciclovir, or valacyclovir), a history of adverse reactions to
herpes antivirals, a history of seizures, a serum creatinine level
2.0 mg/dL, and hematocrit 30%.
The human experimentation guidelines of the US Depart-
ment of Health and Human Services and the individual institu-
tions were followed in the conduct of the clinical research. The
institutional review boards of the University of Washington and
the Asociacin Civil Impacta Salud y Educacin approved the
protocol. Participants provided written informed consent and
were compensated for travel and related expenses. Men with
sexually transmitted infection (STI) syndromes at screening
were treated with regimens recommended by the Peruvian Min-
istry of Health. At the time of the study, antiretrovirals were
available in Peru in public clinics only for HIV-infected persons
with CD4 cell counts 200 cells/L or with an AIDS-related
condition.
HSV Suppression to Reduce HIV Levels
Study medication. Valacyclovir (500 mg orally twice daily)
and matching placebo were supplied by GlaxoSmithKline. Sub-
jects were randomly assigned 1:1 (valacyclovir to placebo) in
blocks of 10. After 8 weeks of the initial treatment, each partici-
pant crossed over to the alternative treatment for 8 weeks, sepa-
rated by a 2-week washout period with daily placebo. Medica-
tion was dispensed every 2 weeks, with pill counts performed at
each visit. Open-label valacyclovir (1 g orally twice daily for 3
days) was dispensed for symptomatic herpes recurrences.
Study procedures. At enrollment, participants received
counseling about genital herpes and training regarding study
procedures, were administered a brief questionnaire, underwent
a physical exam, and had blood and rectal secretions collected.
Participants came to clinic 3 times a week during each treatment
period. Anoscopy was performed at each visit to collect rectal
secretions, with Snostrips (for HIV-1 polymerase chain reaction
[PCR]) used first followed by a Dacron swab (for HSV DNA
PCR), and blood was drawn weekly. Participants also collected
daily swabs of genital and perianal skin at home for HSV DNA
Downloaded from http://jid.oxfordjournals.org/ by guest on June 14, 2016
PCR [16, 17]. At each clinic visit, participants returned their
collected specimens (stored at room temperature).
Specimen Collection and Laboratory Procedures
STI tests. At screening and as clinically indicated during the
study, rectal specimens were collected for Neisseria gonorrhoeae
culture, and urine was collected for gonococcal and chlamydial
nucleic acid amplification tests (NAAT; Aptima; Gen-Probe).
Peripheral blood was collected for syphilis serological tests
(rapid plasma reagin [RPR] and HSV-2 antibody [HerpeSelect-2
ELISA; Focus Technologies]). Positive HSV-2 ELISA results (by
use of an index value cutoff of 3.5 to improve specificity) [18,
19] were confirmed by Western blot [20]. Gonorrhea cultures,
HerpeSelect-2 ELISAs, and RPR tests were performed in Lima.
Gonococcal and chlamydial NAATs and HSV Western blots
were performed at the University of Washington.
Rectal specimens. Rectal mucosal specimens were obtained
through an anoscope 3 4 cm above the squamocolumnar junc-
tion. For HIV-1 RNA, 3 Snostrips (Chauvin Pharmaceuticals)
were saturated with rectal mucosal secretions and placed into
500 L of guanidinium solution (4 mol/L guanidinium thiocy-
anate, 25 mmol/L sodium citrate [pH 7], 0.5% N-lauroylsarcosine,
and 0.1 mol/L 2-mercaptoethanol); for HSV DNA PCR, the
swab was placed in transport medium. Specimens were frozen at
70C within 8 h of collection. Samples were collected by 1 of 2
clinicians.
Blood specimens. Peripheral blood was collected into tubes
containing EDTA (Becton Dickinson) and separated within 8 h
into plasma and mononuclear cells by ficoll-hypaque gradient
centrifugation. Lymphocyte subsets were determined by flow cy-
tometry methods in Lima. Plasma aliquots were frozen at 70
C and transported to the University of Washington Retrovirol-
ogy Laboratory.
JID 2007:196 (15 November) 1501
 HIV-1 RNA quantitation assays. HIV-1 RNA was quanti-
fied using the TaqMan real-time RNA PCR (rt-PCR) amplifica-
tion assay [21] or the Amplicor HIV Monitor assay (Roche Mo-
lecular Systems). The rt-PCR assay was modified to include a
second gag oligonucleotide probe with a 5'-carboxyfluorescein
(FAM) reporter dye and a 3'-minor groove binder/nonfluorescent
quencher (AR8: 6FAM-CTA TCC CAT TCT GC-3MGBNFQ) to
enhance sensitivity. HIV-1 RNA external standards and positive
and negative controls were included on each 96-well plate. To
ensure that negative results were not due to loss during nucleic
acid extraction or nonspecific inhibition of the PCR assay, 300
copies of an external synthetic sequence control (Gene Am-
plimer pAW 109 RNA [ABI catalogue N808-0037]) were added
to and copurified with participant and control samples. Each
PCR also contained 800 nmol/L each of the forward primer
(GCC TGG GTT CCC TGT TCC) and reverse primer (CGA
CGT ACC CCT GAC ATG G) and 100 nmol/L of the labeled
pAW probe (VIC-CCA GGC CAA TGT CTC ACC AAG CTC
TG-MGBNFQ). The laboratory was certified by the National
Institute of Allergy and Infectious Diseasessponsored Virology
Quality Assurance Program to perform both assays.
HIV-1 RNA was consistently not detected in plasma from 3 of
the 20 participants by the rt-PCR assay. The primer and probe
binding regions of gag were sequenced from plasma aliquots
from these 3 subjects. Mutations were detected in the gag-
binding regions for the HXB2 and AR8 probes or the SK431
primer, likely accounting for the negative rt-PCR results. Plasma
and rectal specimens from these 3 participants were subse-
quently tested with the Amplicor HIV Monitor assay, which
yielded HIV-1 RNA; these results were included in the analysis.
For the remaining men, baseline specimens were analyzed by
both assays, and no significant quantitative differences were
found.
For all specimens, vials were warmed, briefly mixed in a vortex
mixer, and microfuged for 5 s at 12,000 g before 200 L of fluid
was withdrawn for testing. Silica extraction to remove inhibitory
factors was performed before PCR analysis of rectal samples. For
the TaqMan assay, the lower limit of detection (LLOD) for
HIV-1 quantitation in blood plasma was 120 (2.1 log10) HIV-1
RNA copies/mL; for the Amplicor HIV Monitor assay, the
LLOD in plasma was 400 (2.6 log10) HIV-1 RNA copies/mL.
Because of the small amount of fluid absorbed by the 3 rectal
Snostrips (25 L) and the dilution with guanadinium, LLODs
were typically 6000 (3.8 log10) and 12,800 (4.1 log10) HIV-1 RNA
copies/mL for the TaqMan and Amplicor HIV Monitor assays,
respectively. For rectal specimens, further dilutional steps were
performed for specimens when repeat testing was required; thus,
the LLOD exceeded these values for 9.9% of specimens.
HSV DNA Assay
DNA was extracted from 200 L of each specimen by use of the
QIAamp 96 DNA Blood Kit (Qiagen) and was eluted into 100 L
1502 JID 2007:196 (15 November) Zuckerman et al.
of AE buffer. A fluorescent probe based rt-PCR (TaqMan; Ap-
plied Biosystems) assay was used to quantitate HSV, using 10 L
of the extracted DNA for each PCR, with primers and probe
sequences and PCR conditions as described eslewhere [22, 23].
To ensure that negative results were not due to nonspecific in-
hibition, each PCR also contained 50,000 copies of EXO DNA,
30 mmol/L of primers EXO186F and EXO315R, and 50 nmol/L
of probe EXO-P, labeled at the 5' end with VIC (Perkin-Elmer
Cetus) and at the 3' end with TAMRA [24]. All negative HSV
PCR results required detection of EXO DNA. One positive con-
trol with 5000 copies of HSV was coprocessed with specimens.
Specimens were processed in parallel with aliquots of 1 PBS.
Wells without DNA also were included in each PCR run.
Statistical Methods
Sample size. The primary study end point was the effect of
valacyclovir on rectal mucosal HIV-1 levels. On the basis of on
our previous studies among MSM [25] and assuming a 10%
dropout rate, we estimated that a sample size of 20 men would
Downloaded from http://jid.oxfordjournals.org/ by guest on June 14, 2016
provide 80% power to detect a change of 0.3 log10 copies/mL in
HIV-1 shedding in the rectum. A crossover design was used be-
cause of lower variability in HIV-1 RNA levels in plasma and
rectal secretions on multiple observations from a person than
between persons [25].
Statistical analysis. HSV shedding rate was computed by
dividing the days with detectable HSV in swabs collected at
home or in the clinic by the total days of swab collection. HSV
shedding was measured on the basis of (1) daily external ano-
genital swabs and (2) clinic-obtained rectal mucosal swabs. Pri-
mary analyses involved overall HSV shedding rates (either exter-
nal anogenital or rectal mucosal). All analyses were done on an
intent-to-treat basis, excluding the first day of study drug ad-
ministration from each arm. Investigators and technologists
were unaware of treatment assignments, and unblinding oc-
curred only after all laboratory assays were completed. For un-
detectable HIV-1 values, the midpoint between zero and the
LLOD was used; HIV-1 values were log-transformed. HSV shed-
ding was examined as a binary variable (detected vs. not de-
tected).
Univariate nonlinear mixed-effects models with a Poisson
link were used to compute the rate ratio comparing HIV-1 de-
tection rates in each treatment arm. The mean quantity of HIV-1
in rectal Snostrips and plasma were compared by treatment arm.
Potential predictors of HIV-1 level were evaluated using linear
mixed-effect models. Because HIV-1 quantity is modeled on the
log10 scale, coefficients () from models are exponentiated (10)
and compared with 1 to compute percent changes in HIV-1
quantity. Multivariate analysis with backward elimination using
mixed-effects models were also performed to compare treat-
ment arms after adjusting for CD4 cell count and HSV shedding,
including interaction terms. Potential sequence effects were
Table 1. Characteristics of 20 men who have sex with men
enrolled in a randomized, double-blind, placebo-controlled cross-
over trial of valacyclovir for the suppression of HIV-1 and herpes
simplex virus type 2 (HSV-2).
Characteristic
Age, median (range), years
History of anal or genital herpes
Symptomatic genital herpes during the trial
Prior herpes antiviral medication use
Baseline laboratory values
CD4 cell count, median (range), cells/L
Hematocrit, median (range), %
Serum creatinine level, median (range),
mg/dL
Sexually transmitted infections
Reactive RPR
Urethral or rectal gonorrheal or
chlamydial infection
Adherence to study procedures
Completed study
869 cells/L) (table 1). Of these 20 HIV-1/HSV-2seropositive
men, 4 (20%) reported prior genital or anal herpes episodes.
Serological evidence of prior Treponema pallidum infection was
detected in 8 men (40%), of whom 2 were treated for early syph-
Value ilis on the basis of titers before enrollment; the other 6 were
considered to have been treated adequately in the past. No rectal
31 (2244)
4 (20)
or urethral gonococcal or chlamydial infections were diagnosed
4 (20) during the study.
2 (10) Nineteen men (95%) completed the 18-week study; 1 partic-
ipant withdrew after 14 weeks. Participants completed a median
406 (232869) of 46 (range, 31 48) of 48 possible clinic visits. On the basis of
42.5 (34.648.0) pill counts, the men took a median of 96.2% (range, 65.8%
100%) of study drug dispensed. Study drug was well tolerated,
1.1 (0.791.2)
with no reports of serious adverse events. Four men were treated
with open-label valacyclovir for herpes recurrences during the
8 (40)a
study, of which 3 occurred during placebo administration.
0 (0) Overall, the analysis database included 2155 days with at least 1
genital and/or perianal swab collected for HSV DNA PCR, in-
19 (95) cluding 904 HSV swabs from anoscopy, 844 Snostrip samples for

Downloaded from http://jid.oxfordjournals.org/ by guest on June 14, 2016

Study drug adherence, median (range), mucosal HIV-1 detection, and 288 plasma samples. Adherence
% of pills not returned 96.2 (65.8100)
with self-collection of anogenital swabs for HSV-2 DNA PCR
Proportion of expected days supplying
self-collected genital and anal swabs was high, with a median 98.6% (range, 88.4%100%) of ex-
for HSV-2, median (range), % 98.6 (88.4100) pected swabs collected.
NOTE. Data are no. (%) with characteristic, unless otherwise indicated. HSV detection in genital and rectal swabs. HSV DNA was
a
Two participants with screening rapid plasma reagin (RPR) titers indicative detected in any anogenital specimen at least once for all study
of early syphilis were treated before enrollment. participants. Overall, HSV was detected in 29% of swabs during
placebo administration, compared with 4% of swabs during val-
evaluated. Statistical analysis was performed using SAS for Win- acyclovir administration (P .001) (table 2). By participant,
dows (version 9.1). HSV detection ranged from 11% to 68% of all swabs obtained
during placebo administration and from 0% to 26% of all swabs
RESULTS obtained during valacyclovir administration. Among rectal
swabs obtained via anoscopy, HSV was detected in 25% of swabs
Study population and protocol adherence. Of 63 men obtained during placebo administration and in 3% of swabs ob-
screened, 20 met the study entry criteria and were enrolled. The tained during valacyclovir administration (P .001); the
median age of participants was 31 years (range, 22 44 years), ranges of rectal swabs positive for HSV by participant were 0%
and the median CD4 cell count was 406 cells/L (range, 232 96% for placebo and 0%17% for valacyclovir. The majority of

Table 2. Rates of herpes simplex virus (HSV) and HIV-1 detection and mean log10 HIV-1 levels for 20 HIV-1/HSV-2 coinfected men who
have sex with men.

Arm

Category Both arms Placebo Valacyclovir

HSV detection rate, external anogenital or rectal mucosal sample 356/2155 (17) 309/1071 (29) 47/1084 (4)a
HSV detection rate, rectal mucosal sample only 124/904 (14) 112/446 (25) 12/458 (3)
Rectal mucosal HIV-1 detection rate 620/844 (73) 333/427 (78) 287/417 (69)b
Rectal mucosal HIV-1 level, mean SD, log10 copies/mL 4.90 1.04 5.00 1.04 4.80 1.04a
Plasma HIV-1 detection rate 284/288 (99) 143/145 (99) 141/143 (99)
Plasma HIV-1 level, mean SD, log10 copies/mL 4.32 0.72 4.50 0.71 4.14 0.69a

NOTE. Data are no. with detectable HIV-1 RNA or HSV DNA by polymerase chain reaction/no. of samples obtained, unless otherwise indicated. Observations
begin on the second day for each study arm. Undetectable HIV-1 levels have been imputed to the midpoint between zero and the lower limit of detection.
a
P .001, compared with placebo.
b
P .02, compared with placebo.

HSV Suppression to Reduce HIV Levels JID 2007:196 (15 November) 1503
Table 3. Potential predictors of HIV-1 level in rectal mucosa and plasma, in univariate and multivariate models.

Log10 rectal HIV-1 level Log10 plasma HIV-1 level

Univariate Multivariate Univariate Multivariate

Predictor Estimate P Estimate P Estimate P Estimate P

Valacyclovir vs. placebo 0.16 .0008 0.16 .0008 0.33 .0001 0.33 .0001
CD4 cell counta 0.19 .11 NS NS 0.12 .18 0.072 .42
HSV detection, anyb 0.04 .34 ND ND 0.19 .0001 NS NS
HSV detection, rectalb 0.39 .0001 NS NS 0.27 .037 ND ND
Interaction of valacyclovir arm with
CD4 cell countc NA NA NS NS NA NA 0.082 .018

NOTE. Estimates represent the effect of the predictors on HIV-1 plasma or rectum levels (log10 copies/mL) in either univariate or multivariate (backward-
elimination) models. HSV, herpes simplex virus; NA, not applicable (interaction terms were not tested univariately); ND, not done (indicated for terms that were
not included in multivariate analysis; only 1 predictor of HSV detection was used in multivariate analysisrectal HSV detection was used as a potential predictor
in rectal HIV detection, and any HSV detection [home or clinic] was used to predict HIV detection in plasma); NS, not statistically significant when included in the
multivariate model (therefore, the term was removed).
a
Each 100-cell/L increase in CD4 cell count.
b
For HSV detection, any indicates detectable HSV-2 from any swab acquired in the study (home or clinic), and rectal indicates detectable HSV-2 DNA in
a rectal swab obtained by anoscopy.
c
Interaction term included to determine whether the treatment effect differed by CD4 cell count (each 100-cell/L increase).

Downloaded from http://jid.oxfordjournals.org/ by guest on June 14, 2016

HSV detection was asymptomatic, with only 4 symptomatic
HSV recurrences.
Rectal HIV-1 levels. Overall, HIV-1 was detected in 620
(73%) of 844 rectal mucosal samples obtained from the 20 par-
ticipants: 78% (333/427) of samples collected during placebo
administration and in 69% (287/417) of samples collected dur-
ing valacyclovir administration (P .02) (table 2). The range
by participant was 0%100% in both the placebo and valacyclo-
vir arms, with 2 participants having 3 rectal specimens with
detectable HIV, regardless of treatment arm. The mean HIV-1
levels in rectal secretions during placebo and valacyclovir ad-
ministration were 5.00 (SD, 1.04) and 4.80 (SD, 1.04) log10 cop-
ies/mL, respectively (P .001) (table 2).
In univariate analysis, the mean within-subject decrease in
rectal HIV-1 levels during valacyclovir administration versus
placebo administration was 0.16 log10 copies/mL (95% confi-
dence interval [CI], 0.07 0.25 log10 copies/mL; P .0008), a
31% decrease (table 3). Participant-level means and SDs for rec-
tal HIV-1 levels for the treatment arms are displayed in figure
1A. No significant sequence effects between treatment arms were
found (data not shown). No significant associations were noted
for rectal HIV-1 shedding and CD4 cell counts (P .11) or for
detection of HSV shedding by either external anogenital or rectal
mucosal swabs (P .34). However, in analyzing only HSV
swabs obtained by anoscope, rectal HSV detection was associ-
ated with an increase of 0.39 log10 copies/mL in the rectal Snos-
trip HIV-1 level, a 150% increase (P .0001); this association
did not remain significant in a multivariate model that included
treatment arm.
Plasma HIV-1 levels. HIV-1 RNA was detected in 284
(98.6%) of 288 plasma samples. Mean HIV-1 levels in plasma
during placebo and valacyclovir administration were 4.50 (SD,
0.71) and 4.14 (SD, 0.71) log10 copies/mL, respectively
(P .001) (table 2). In univariate analysis, the mean within-
subject decrease in plasma HIV-1 levels during valacyclovir ad-
ministration versus placebo administration was 0.33 log10 cop-
ies/mL (95% CI, 0.23 0.42 log10 copies/mL; P .0001),
corresponding to a 53% decrease in the quantity of plasma
HIV-1. Participant-specific means and SDs for plasma HIV-1
levels for the treatment arms are displayed in figure 1B. No sig-
nificant sequence effects between treatment arms were found
(data not shown). No univariate association between plasma
HIV-1 level and CD4 cell counts was observed (P .18). Uni-
variately, the detection of HSV in rectal swabs was associated
with an increase of 0.27 log10 copies/mL in HIV-1 level
(P .037); this association did not remain significant in a mul-
tivariate model that included treatment arm.
In multivariate analysis, the mean within-subject decrease in
plasma HIV-1 levels during daily suppressive valacyclovir was 0.33
log10 copies/mL (P .001). When CD4 cell count and treatment
arm were included as an interaction term, there was a greater
decrease in plasma HIV-1 levels for each 100-cell/mL increase in
CD4 cell count for the valacyclovir arm (P .018) (figure 2).
DISCUSSION
Our findings show that antiviral drugs that effectively suppress
HSV reactivation significantly reduce plasma and mucosal
HIV-1 RNA levels among HIV-1/HSV-2 coinfected MSM. In
this proof-of-concept crossover trial with multiple observations
per participant and intensive mucosal sampling, suppressive
therapy with valacyclovir (500 mg twice daily) was associated
with mean reductions of 31% in rectal and 53% in plasma HIV-1
levels. Importantly, we documented that the significant reduc-
tion in HSV detection coincided with the reduction in HIV-1
level, indicating that this effect on systemic and mucosal HIV-1

1504 JID 2007:196 (15 November) Zuckerman et al.

 Downloaded from http://jid.oxfordjournals.org/ by guest on June 14, 2016
Figure 1. Comparison of rectal (A) and plasma (B) HIV-1 levels between treatment arms, by individual participant. Boxes represent the mean HIV-1
level and brackets denote 1 SD for observations after the first day of each treatment. Undetectable HIV-1 levels were imputed to the midpoint between
zero and the lower limit of detection (LLOD). For participants with mean values near the LLOD, the lower SD limit may span below the LLOD because
of the effect of other values on the SD calculation. In panel A, for participants without detectable values in a given treatment arm, the marker is placed
at the LLOD for that participants values.

levels was mediated through HSV suppression. Our detection of
HSV in samples from the distal rectum, where mucosal secre-
tions were sampled to quantify rectal HIV-1 shedding, further
suggests a direct association between anogenital HSV reactiva-
tion and HIV-1 replication. Thus, these findings support a sub-
stantial body of research in which HSV-2 enhanced HIV-1 rep-
lication in vitro and HSV-2 reactivation increased plasma and
anogenital HIV-1 load in vivo, indicating that HSV-2 may en-
hance HIV-1 infectiousness in coinfected persons [2, 26 29].
The mean decrease in plasma HIV-1 level was 0.33 log10 cop-
ies/mL over the 2 months of HSV suppression, with a more pro-
nounced effect among men with higher CD4 cell counts. Our
findings are consistent with those of a recent trial in Burkina
Faso in women, which demonstrated a mean reduction of 0.53
log10 copies/mL in plasma HIV level with valacyclovir suppres-
sion [14]. Natural history studies indicate that such a reduction
in plasma HIV-1 levels may result in clinical benefits if associated
with increased CD4 cell counts and delayed HIV disease progres-

HSV Suppression to Reduce HIV Levels JID 2007:196 (15 November) 1505

Figure 2. Difference in mean plasma HIV level between the valacyclovir and placebo arms in 20 men who have sex with men enrolled in a
randomized crossover trial, by CD4 cell count at screening. The figure shows a greater reduction in HIV levels at higher CD4 cell counts (P .018)
sion [30, 31]. Comparable reductions in plasma HIV-1 levels
(0.5 log10 copies/mL) with zidovudine monotherapy have
translated into clinical benefit [32], although the benefit was at-
tenuated by HIV rebound due to resistance to zidovudine. How-
ever, this would not be expected with antiherpes therapy, be-
cause it acts via suppression of a cofactor in HIV-1 up-regulation
rather than directly on HIV-1 transcription. Interestingly, the
Burkina Faso study, during which the study drug was adminis-
tered for 3 months, noted an increased effect on HIV-1 over
time, suggesting that longer duration of HSV-2 suppression may
achieve greater reductions in HIV-1 levels. In comparison, the
present crossover trial involved 2 months of administration for
each arm, and no temporal effect in HIV-1 levels was observed
(data not shown). Both trials showed an increased effect with a
higher CD4 cell count.
HSV replication occurs in the anogenital mucosa, and the
mechanisms by which HSV reactivation increases HIV-1 levels
in plasma are not well understood. Increased levels of proinflam-
matory genital cytokines and chemokines were observed in a
cross-sectional study of HIV-1/HSV-2 coinfected African
women who were shedding HSV-2, compared with those in
women who were not shedding HSV-2 [33]. Among HIV/HSV-
2 coinfected MSM, anorectal HSV reactivation could directly
increase HIV replication in the gut lymphoid tissue, which con-
tains large numbers of CD4 lymphocytes, dendrocytes, and mac-
rophages. The greater reduction in plasma HIV-1 than rectal
HIV-1 levels during HSV suppression is likely a reflection of
1506 JID 2007:196 (15 November) Zuckerman et al.
Downloaded from http://jid.oxfordjournals.org/ by guest on June 14, 2016
biological factors and greater variability in rectal HIV-1 mea-
surement, given the small volumes collected by Snostrips and the
dilution for sufficient volume for the HIV-1 RNA assays. Com-
plex factors likely influence HIV replication in the gut-
associated lymphoid tissue, which contains more CD4 T cells
than lymph nodes or the peripheral blood [34]. Further studies,
including immunological and virological analyses of rectal bi-
opsy samples, will be required to understand the pathogenesis of
anorectal HSV reactivation and HIV-1 replication.
Most HSV reactivations in this cohort were asymptomatic, so
HSV interventions directed at decreasing HIV-1 levels will re-
quire the use of suppressive rather than episodic treatment.
Given the small proportion who shed HSV-2 during valacyclovir
administration in this trial, additional studies are needed to as-
sess the effect of greater HSV-2 suppression on HIV replication.
Valacyclovir, the hydrochloride salt of the L-valyl ester of acy-
clovir, is rapidly and completely metabolized to acyclovir,
achieves higher plasma levels than do similar doses of acyclovir,
and is very safe. Studies with comparable doses of acyclovir,
which is available generically at lower cost, have demonstrated
comparable efficacy to valacyclovir in suppressing HSV recur-
rences and shedding [35, 36]. Although acyclovir-resistant
strains of HSV are observed among HIV-1 and HSV-2infected
persons [3739], the prevalence remains low (5%). Genital
herpes lesions that are clinically refractory to acyclovir therapy
because of acyclovir resistance are much less common [39, 40],
and transmission of acyclovir-resistant strains is rare [41].
 The limitations of the present study include our inability to
precisely quantify lower levels of rectal HIV-1 and to detect rec-
tal mucosal abnormalities not visualized on anoscopy. Adher-
ence was assessed by pill count and self-report, which may not
perfectly measure adherence, although the high rate of HSV sup-
pression in the valacyclovir arm indicates very high adherence to
study drug. The 8-week duration of treatment may have been
too short to detect the maximal effect of HSV suppression on
HIV levels. A longer trial is necessary to assess whether the re-
duction in HIV-1 levels translates to delayed HIV disease pro-
gression and clinical benefits from delayed time to highly active
ART (HAART) initiation.
In summary, this randomized, double-blind, placebo-controlled
crossover trial of valacyclovir in HIV-1/HSV-2 coinfected MSM
with CD4 cell counts 200 cells/L has demonstrated significant
reductions in both mucosal and systemic HIV-1 levels during daily
suppressive HSV therapy. Ongoing randomized trials will deter-
mine whether HSV suppression can reduce HIV-1 transmission
and will address the potential for HSV suppression to delay HAART
initiation. While awaiting those results regarding public health and
clinical benefits of HSV suppression in HIV-1/HSV-2 coinfected
persons, this study reinforces the need for HSV-2 serological testing
among HIV-infected persons and the benefits of suppressive HSV
therapy in those who have CD4 cell counts 200 cells/L and are
not receiving ART.
Acknowledgments
We extend our grateful appreciation to the study participants. Addition-
ally, we thank Shyla Snchez and Julio Chamochumbi for study coordination
and scheduling in Lima, Peru; Carmen Snchez, Sofia Snchez, and Dr. Jorge
Vergara for clinical support and procedures for study participants; Drs. Jef-
frey Ferris and Esmellin Prez for pharmacy support; Jerry Galea for techni-
cal support; Dr. Tuofu Zhu, Joan Dragavon, and the University of Washing-
ton Retrovirology Laboratory staff; and Stacy Selke, Dr. Meei-Li Huang, Dr.
Rhoda Ashley-Morrow, and the University of Washington Virology Re-
search Laboratories for support.
References
1. Corey L, Wald A, Celum CL, Quinn TC. The effects of herpes simplex
virus-2 on HIV-1 acquisition and transmission: a review of two overlap-
ping epidemics. J Acquir Immune Defic Syndr 2004; 35:435 45.
2. Gray RH, Wawer MJ, Brookmeyer R, et al. Probability of HIV-1 trans-
mission per coital act in monogamous, heterosexual, HIV-1-discordant
couples in Rakai, Uganda. Lancet 2001; 357:1149 53.
3. Kreiss JK, Coombs R, Plummer F, et al. Isolation of human immunode-
ficiency virus from genital ulcers in Nairobi prostitutes. J Infect Dis
1989; 160:380 4.
4. Mbopi Keou FX, Gresenguet G, Mayaud P, et al. Genital herpes simplex
virus type 2 shedding is increased in HIV-infected women in Africa.
AIDS 1999; 13:536 7.
5. Schacker TW, Ryncarz AJ, Goddard J, Diem K, Shaughnessy M, Corey L.
Frequent recovery of HIV-1 from genital herpes simplex virus lesions in
HIV-1infected men. JAMA 1998; 280:61 6.
6. McClelland RS, Wang CC, Overbaugh J, et al. Association between
cervical shedding of herpes simplex virus and HIV-1. AIDS
2002; 16:242530.
7. Mosca J, Bednarik D, Raj N, et al. Herpes simplex virus type 1 can reac-
HSV Suppression to Reduce HIV Levels
tivate transcription of latent human immunodeficiency virus. Nature
1987; 325:6770.
8. Margolis D, Ostrove J, Straus S. HSV-1 activation of HIV-1 transcrip-
tion is augmented by a cellular protein that binds near the initiator
element. Virology 1993; 192:370 4.
9. Kucera L, Leake E, Iyer N, Raben D, Myrvik Q. Human immunodefi-
ciency virus type 1 (HIV-1) and herpes simplex virus type 2 (HSV-2) can
coinfect and simultaneously replicate in the same human CD4 cell:
effect of coinfection on infectious HSV-2 and HIV-1 replication. AIDS
Res Hum Retroviruses 1990; 6:6417.
10. Heng M, Heng S, Allen S. Coinfection and synergy of human immuno-
deficiency virus-1 and herpes simplex viurs-1. Lancet 1994; 343:255 8.
11. Moriuchi M, Moriuchi H, Williams R, Straus SE. Herpes simplex virus
infection induces replication of human immunodeficiency virus type 1.
Virology 2000; 278:534 40.
12. Schacker T, Hu HL, Koelle DM, et al. Famciclovir for the suppression of
symptomatic and asymptomatic herpes simplex virus reactivation in
HIV-infected persons: a double-blind, placebo-controlled trial. Ann In-
tern Med 1998; 128:21 8.
13. DeJesus E, Wald A, Warren T, et al. Valacyclovir for the suppression of
recurrent genital herpes in human immunodeficiency virusinfected
subjects. J Infect Dis 2003; 188:1009 16.
14. Nagot N, Ouedraogo A, Foulongne V, et al. Reduction of HIV-1 RNA
levels with therapy to suppress herpes simplex virus. N Engl J Med 2007;
Downloaded from http://jid.oxfordjournals.org/ by guest on June 14, 2016
356:790 9.
15. Ioannidis JPA, Collier AC, Cooper DA, et al. Clinical efficacy of high-
dose acyclovir in patients with human immunodeficiency virus infec-
tion: a meta-analysis of randomized individual patient data. J Infect Dis
1998; 178:349 59.
16. Wald A, Zeh JE, Barnum G, Davis LG, Corey L. Suppression of subclin-
ical shedding of herpes simplex virus type 2 with acyclovir. Ann Intern
Med 1996; 124:8 15.
17. Wald A, Corey L, Cone R, Hobson A, Davis G, Zeh J. Frequent genital
HSV-2 shedding in immunocompetent women. J Clin Invest 1997; 99:
10927.
18. Prince HE, Ernst CE, Hogrefe WR. Evaluation of an enzyme immuno-
assay system for measuring herpes simplex virus (HSV) type 1specific
and HSV type 2specific IgG antibodies. J Clin Lab Anal 2000; 14:13 6.
19. Golden MR, Ashley-Morrow R, Swenson P, Hogrefe WR, Handsfield
HH, Wald A. Herpes simplex virus type 2 (HSV-2) Western blot confir-
matory testing among men testing positive for HSV-2 using the focus
enzyme-linked immunosorbent assay in a sexually transmitted disease
clinic. Sex Transm Dis 2005; 32:7717.
20. Ashley RL, Militoni J, Lee F, Nahmias A, Corey L. Comparison of West-
ern blot (Immunoblot) and glycoprotein G-specific immunodot en-
zyme assay for detecting antibodies to herpes simplex virus types 1 and 2
in human sera. J Clin Microbiol 1988; 26:6627.
21. Li CC, Seidel KD, Coombs RW, Frenkel LM. Detection and quantifica-
tion of human immunodeficiency virus type 1 p24 antigen in dried
whole blood and plasma on filter paper stored under various conditions.
J Clin Microbiol 2005; 43:39015.
22. Jerome KR, Huang ML, Wald A, Selke S, Corey L. Quantitative stability
of DNA after extended storage of clinical specimens as determined by
real-time PCR. J Clin Microbiol 2002; 40:2609 11.
23. Wald A, Huang ML, Carrell D, Selke S, Corey L. Polymerase chain reac-
tion for detection of herpes simplex virus (HSV) DNA on mucosal sur-
faces: comparison with HSV isolation in cell culture. J Infect Dis 2003;
188:134551.
24. Limaye AP, Huang ML, Leisenring W, Stensland L, Corey L, Boeckh M.
Cytomegalovirus (CMV) DNA load in plasma for the diagnosis of CMV
disease before engraftment in hematopoietic stem-cell transplant recip-
ients. J Infect Dis 2001; 183:377 82.
25. Zuckerman RA, Whittington WL, Celum CL, et al. Higher concentra-
tion of HIV RNA in rectal mucosa secretions than in blood and seminal
plasma, among men who have sex with men, independent of antiretro-
viral therapy. J Infect Dis 2004; 190:156 61.
JID 2007:196 (15 November) 1507
26. Mole L, Ripich S, Margolis D, Holodniy M. The impact of active herpes
simplex virus infection on human immunodeficiency virus load. J Infect
Dis 1997; 176:766 70.
27. Schacker T, Zeh J, Hu H, Shaughnessy M, Corey L. Changes in plasma
human immunodeficiency virus type 1 RNA associated with herpes sim-
plex virus reactivation and suppression. J Infect Dis 2002; 186:1718 25.
28. Freeman EE, Glynn JR. Factors affecting HIV concordancy in married
couples in four African cities. AIDS 2004; 18:171521.
29. Ouedraogo A, Nagot N, Vergne L, et al. Impact of suppressive herpes
therapy on genital HIV-1 RNA among women taking antiretroviral
therapy: a randomized controlled trial. AIDS 2006; 20:230513.
30. Mellors JW, Munoz A, Giorgi JV, et al. Plasma viral load and CD4
lymphocytes as prognostic markers of HIV-1 infection. Ann Intern Med
1997; 126:946 54.
31. Rodriguez B, Sethi AK, Cheruvu VK, et al. Predictive value of plasma
HIV RNA level on rate of CD4 T-cell decline in untreated HIV infection.
JAMA 2006; 296:1498 506.
32. Katzenstein DA, Hammer SM, Hughes MD, et al. The relation of viro-
logic and immunologic markers to clinical outcomes after nucleoside
therapy in HIV-infected adults with 200 to 500 CD4 cells per cubic
millimeter. AIDS Clinical Trials Group Study 175 Virology Study Team.
N Engl J Med 1996; 335:1091 8.
33. Rebbapragada A, Wachihi C, Pettengell C, et al. Negative mucosal syn-
ergy between herpes simplex type 2 and HIV in the female genital tract.
AIDS 2007; 21:589 98.
1508 JID 2007:196 (15 November) Zuckerman et al.
34. Mattapallil JJ, Douek DC, Hill B, Nishimura Y, Martin M, Roederer M.
Massive infection and loss of memory CD4 T cells in multiple tissues
during acute SIV infection. Nature 2005; 434:10937.
35. Conant MA, Schacker TW, Murphy RL, Gold J, Crutchfield LT, Crooks
RJ. Valaciclovir versus aciclovir for herpes simplex virus infection in
HIV-infected individuals: two randomized trials. Int J STD AIDS 2002;
13:1221.
36. Gupta R, Wald A, Krantz E, et al. Valacyclovir and acyclovir for suppres-
sion of shedding of herpes simplex virus in the genital tract. J Infect Dis
2004; 190:1374 81.
37. Reyes M, Shaik NS, Graber JM, et al. Acyclovir-resistant genital herpes
among persons attending sexually transmitted disease and human im-
munodeficiency virus clinics. Arch Intern Med 2003; 163:76 80.
38. Sarisky RT, Bacon TH, Boon RJ, et al. Profiling penciclovir susceptibility
and prevalence of resistance of herpes simplex virus isolates across
eleven clinical trials. Arch Virol 2003; 148:1757 69.
39. Levin MJ, Bacon TH, Leary JJ. Resistance of herpes simplex virus infec-
tions to nucleoside analogues in HIV-infected patients. Clin Infect Dis
2004; 39(Suppl 5):S248 57.
40. Safrin S, Elbeik T, Phan L, et al. Correlation between response to acyclo-
vir and foscarnet therapy and in vitro susceptibility result for isolates of
herpes simplex virus from human immunodeficiency virus-infected pa-
tients. Antimicrob Agents Chemother 1994; 38:1246 50.
41. Kost R, Hill E, Tigges M, Straus S. Brief report: recurrent acyclovir-
Downloaded from http://jid.oxfordjournals.org/ by guest on June 14, 2016
resistant genital herpes in an immunocompetent patient. N Engl J Med
1993; 329:1777 82.