Trial of rgp120 HIV-1 Vaccine among IDUs
• JID 2006:194 (15 December) • 1661
M A J O R A R T I C L E
Randomized, Double-Blind, Placebo-Controlled
Efficacy Trial of a Bivalent Recombinant Glycoprotein
120 HIV-1 Vaccine among Injection Drug Users
in Bangkok, Thailand
Punnee Pitisuttithum,1Peter Gilbert,4Marc Gurwith,5William Heyward,5Michael Martin,3Fritz van Griensven,3
Dale Hu,6Jordan W. Tappero,3and Kachit Choopanya,2for the Bangkok Vaccine Evaluation Groupa
1Department of Clinical Tropical Medicine, Mahidol University, and
of Public Health–US Centers for Disease Control and Prevention Collaboration, Nonthaburi, Thailand;
and Prevention, Fred Hutchinson Cancer Research Center, Seattle, Washington;
Control and Prevention, Atlanta, Georgia
2Bangkok Metropolitan Administration, Bangkok, and
4Statistical Center for HIV/AIDS Research
6US Centers for Disease
5VaxGen, Inc., Brisbane, California;
recombinant glycoprotein 120 human immunodeficiency virus type 1 (HIV-1) vaccines weresuccessfullyconducted
from 1995 to 1998, prompting the first HIV-1 vaccine efficacy trial in Asia.
This randomized, double-blind, placebo-controlled efficacy trial of AIDSVAX B/E (VaxGen), which
included 36-months of follow-up, was conducted among injection drug users (IDUs) in Bangkok, Thailand. The
primary end point was HIV-1 infection; secondary end points included plasma HIV-1 load, CD4 cell count, onset
of acquired immunodeficiency syndrome–defining conditions, and initiation of antiretroviral therapy.
A total of 2546 IDUs were enrolled between March 1999 and August 2000; the median age was 26
years, and 93.4% were men. The overall HIV-1 incidence was 3.4 infections/100 person-years (95% confidence
interval [CI], 3.0–3.9 infections/100 person-years), and the cumulative incidence was 8.4%. There were no dif-
ferences between the vaccine and placebo arms. HIV-1 subtype E (83 vaccine and 81 placebo recipients) accounted
for 77% of infections. Vaccine efficacy was estimated at 0.1% (95% CI, ?30.8% to 23.8%;
No statistically significant effects of the vaccine on secondary end points were observed.
Despite the successful completion of this efficacy trial, the vaccine did not prevent HIV-1infection
or delay HIV-1 disease progression.
In Thailand, phase 1/2 trials of monovalent subtype B and bivalent subtype B/E (CRF01_AE)
, log-rank test).P p .99
The Thai HIV-1 epidemic began in the late 1980s, with
a rapid introduction of HIV-1 subtype B among in-
jection drug users (IDUs) followed by a largerepidemic
of sexually transmitted subtype E (later designated as
CRF01_AE) . In 1988, the prevalence of HIV-1 in-
fection among IDUs in Bangkok increased from !1%
Received 28 February 2006; accepted 28 July 2006; electronically published 3
Potential conflicts of interest: none reported.
Financial support: VaxGen, Inc.
Clinical Trial registration number: 36424.
aStudy group members are listed after the text.
Reprints or correspondence: Prof. Punnee Pitisuttithum, Clinical Infectious
Diseases Research Unit, Dept. of Clinical Tropical Medicine, Faculty of Tropical
Medicine, Mahidol University, 420/6 Rajvithi Road, Bangkok 10400, Thailand
(email@example.com or firstname.lastname@example.org).
The Journal of Infectious Diseases
? 2006 by the Infectious Diseases Society of America. All rights reserved.
to ∼40% . Thailand has been successful in control-
ling the heterosexual spread of HIV-1, with estimated
new infections decreasing from 143,000 in 1991 to
20,000 in 2004 . This reduction reflects Thailand’s
strong commitment to confronting the epidemic and
implementing prevention strategies. Thailand’s first
National Plan for HIV Vaccine Development and Eval-
uation was established in 1993, with a revision in 1997
. The recombinant (r) gp120 vaccine was selected
for evaluation on the basis of safety and immunoge-
nicity profiles in humans [5–7]. A phase 1/2 trial of a
monovalent subtype B rgp120 vaccine among IDUs in
Bangkok was successfully conducted in 1995–1996 ,
which was followed by a similar trial of a bivalent sub-
type B/E rgp120 vaccine in 1998 . These trials dem-
onstrated that rgp120 was safe and immunogenic. In
parallel, 1209 HIV-1–negative IDUs were enrolled in a
vaccine preparatory study, which documented an HIV-
1662 • JID 2006:194 (15 December) • Pitisuttithum et al.
1 incidence of 5.8 infections/100 person-years, with 79% being
subtype E infections and 21% being subtype B infections .
Follow-up rates were 88.2% at 12 months and 75.9% at 24
months. Fifty percent of volunteers reported a definite will-
ingness to participate in HIV-1 vaccine trials . On the basis
of these data, a phase 3 HIV-1 vaccine efficacytrialofAIDSVAX
B/E (VaxGen) was conducted. Inthepresentreport,wedescribe
the trial methodology, conduct, and outcomes.
VOLUNTEERS, MATERIALS, AND METHODS
This randomized, double-blind, placebo-controlled vaccine ef-
ficacy trial was conducted among IDUs attending 17 Bangkok
Metropolitan Administration (BMA) drug-treatment clinics.
These clinics provide methadone detoxification (45 days) and
methadone maintenance (daily) treatment for ∼8000 heroin
addicts per year . Eligibility criteria were age of 20–60years,
drug injection during the past year, being negative for HIV-1
by ELISA at screening and baseline, and provision of written
consent after passing 2 trial comprehension tests. Female vol-
unteers could not be pregnant at baseline, could not be breast-
feeding, and were required to commit to contraceptive use
during the study. A computer-generated block randomization
list, stratified by clinic, was designed to satisfy a 1:1 vaccine:
Risk Counseling and Assessment
At each visit, volunteers were counseled to eliminate HIV risk
behavior, including drug injection, needle sharing, and unpro-
tected sexual intercourse. Male condoms and bleach to clean
injection equipment were provided free of charge.
Questionnaires to assess risk behavior and social harms re-
lated to trial participation were administered every 6 months.
If a study-related social harm was reported, such as denial of
health insurance or discrimination because of HIV infection,
an intervention was made to resolve the harm, and follow-up
was conducted .
Vaccine and Placebo Preparations
AIDSVAX B/E contains 2 rgp120 HIV-1 envelope antigens: 1
from a CXCR4-dependent laboratory-adapted subtype B strain
(MN), and 1 from a CCR5-dependent primary subtype
CRF01_AE isolate (A244), each produced from stable, trans-
fected CHO cell lines [14, 15]. A244 was isolated in northern
Thailand in 1990 [16, 17]. Purified protein (300 mg of MN and
300 mg of A244) was adsorbed onto a total of 600 mg of alum.
Southeast Asian subtype E strains have been determined to be
CRF01_AE with subtype A–like gag, pol, and env gp41 regions,
but the env gp120 has been characterized as belonging entirely
to an HIV-1 subtype E lineage. The subtype B strain MN used
in the vaccine is similar to but distinct from subtype B?strains
circulating in Thailand [18, 19]. The placebo contained only
Vaccine Administration and Outcome Measurements
Vaccine or placebo was injected intramuscularly at months 0,
1, 6, 12, 18, 24, and 36. At each visit, adverse events were
assessed and blood wascollected,todeterminevaccineantibody
response and HIV-1 status by ELISA and immunobloting. The
presence of 2 bands other than gp120 or gp160 was required
for an immunoblot to be considered confirmatory. To estimate
the date of HIV-1 infection, nucleic acid–based amplification
testing (NAT) was performed in volunteers with serologic evi-
dence of incident infection. The date of infection was estimated
as follows: if HIV-1 RNA was undetectable by NAT in the last
seronegative serum specimen, then the date of infection was
estimated as the midpoint of the dates for last negative and
first positive ELISA/immunoblot. Otherwise, the infection date
was estimated as the date for earliest specimen with detectable
HIV-1 infection was determined at the BMA laboratory by
use of the Genetic Systems–Biorad ELISA and Western blot
kits. At a VaxGen contract laboratory, specimens immediately
collected before the first seropositive sample were assayed for
the presence of HIV-1 RNA by NAT (Procleix HIV-1 discrim-
inatory assay; Chiron). Volunteers with incident HIV-1 infec-
tion were followed up at months !1, 1, 2, 4, 8, 12, 16, 20, and
24. At each visit, blood was collected for determination of
plasma HIV-1 RNA load by reverse-transcription polymerase
chain reaction (RT-PCR) (Amplicor HIV-1 Monitor; version
1.5; Roche Diagnostic Systems), and CD4 and CD8 cell counts
were done by 2-color flow cytometry (FACScan; Becton Dick-
inson) in accordance with US Centers for Disease Control and
Prevention (CDC) guidelines . These assays were performed
at the Thailand Ministry of Public Health (MOPH)–CDC Col-
laboration laboratory with EDTA-anticoagulated blood. At the
VaxGen laboratory, 5 assays were used to measure rgp120 an-
tibody responses: an ELISA for antibodythat blocksthebinding
of A244 to the CD4 coreceptor; an ELISA for anti–A244 V2;
an ELISA for anti–A244 V3; an ELISA for anti–gp120 MN/
A244 (mixed) binding antibodies; and an ELISA for MN neu-
tralization [21, 22]. Specimens collected at the last immuni-
zation visit before the first seropositive sample and 2 weeks
after the last immunization visit were assayed for this purpose.
In addition, assays were performed on random samplesofspec-
imens collected at the time of all immunizations and 2 weeks
after immunization from 10% of uninfected vaccine (n p
) and 1% of uninfected placebo ( 115
HIV-1 subtype determination was performed by VaxGen.
Viral RNA and DNA were isolated from 0.5–1.0 mL of frozen
) recipients.n p 12
Trial of rgp120 HIV-1 Vaccine among IDUs • JID 2006:194 (15 December) • 1663
Flow of study participants
plasma by use of the ViroSeq Sample Preparation Kit (Applied
Biosystems). Full-length gp120 sequences were amplified from
samples by RT-PCR. The RT-PCRs were performed indepen-
dently by use of commercially availablekits(forRT,FirstStrand
cDNA Synthesis Kit from Amersham Biosciences; for PCR,
Sigma-Aldrich). All resulting PCR products were cloned into
a bacterial plasmid(pCR 2.1-TOPO;Invitrogen)andsequenced
by use of BigDye 3.1 reaction mix and an ABI-3100 automated
DNA sequencer (Applied Biosystems). More than 200 full-
length gp120 sequences were aligned by use of a proprietary
software package (VaxGen) and subjected to phylogeneticanal-
ysis by use of Phylogenetic Analysis Using Parsimony software
(Sinauer Associates). A complete description of the sequence
methods and analyses will be presented elsewhere (Jobes et al.,
manuscript in preparation). All testingfollowedmanufacturers’
Objectives and End Points
HIV-1 infection was the primary end point for vaccine efficacy;
secondary end points were safety and delayed progression of
HIV-1 disease. Disease progression was evaluated on the basis
of clinical (initiation of antiretroviral therapy [ART] and onset
of AIDS-defining conditions) and biological (CD4 cell count
and plasma HIV-1 load) end points. Volunteers with incident
HIV-1 infection received HIV care per Thailandnationalguide-
lines [23, 24]. Before October 2001, HIV-1–infected volunteers
with a CD4 cell count !500 cells/mL were treated with 2 nu-
1664 • JID 2006:194 (15 December) • Pitisuttithum et al.
with all infections included regardless of whether the subtype was determined (A); 1 minus the cumulative incidence of infection with subtype B
(B); and 1 minus the cumulative incidence of infection with subtype E (C).
Kaplan-Meier curves for the time to the estimated date of HIV-1 infection during each 6-month interval for the intention-to-treat cohort,
cleoside reverse-transcriptase inhibitors. In October2001,guide-
lines were revised to include highly active ART for those with
a CD4 cell count !200 cells/mL.
Sample size estimation.
90% power to reject the null hypothesis with a demonstrated
vaccine efficacy of 30% when the true vaccine efficacy was
67.5% (, 2-sided test). Power for the primary efficacyP p .05
analysis of the intention-to-treat (ITT) cohort (having received
at least 1 vaccination) was estimated on the basis of computer
simulations, by use of a discrete failure-time model that as-
sumed 2500 individuals enrolled, a 1:1 vaccine:placebo ratio,
no vaccine effect until the third immunization and “full effect”
thereafter, a placebo-arm infection rate of 4% per year, and
losses to follow-up of 20%, 15%, and 10% in years 1, 2, and
3, respectively. Under these assumptions, 106 incident infec-
Vaccine efficacy was defined as
risk of infection)?100%. The trial design had
tions were expected in the placebo group, and 106, 82, and 50
incident infections wereexpectedinthevaccinegroupifvaccine
efficacy equaled 0%, 30%, and 67.5%, respectively. Statistical
power was estimated by the proportion of 95% confidence
intervals (CI) for vaccine efficacy not covering the null-hy-
pothesis value. For the 1 interim analysis and the final efficacy
analysis, significance levels for vaccine efficacy were P p .027
and , respectively.
P p .0494
Primary end-point analysis.
in the safety analyses. Vaccine efficacy analyses included all
participants in the ITT cohort. Kaplan-Meier curves were used
to estimate the probability of being uninfected as a function
of time from first vaccination. Log-rank tests were used to
compare time-to-infection distributions between study arms.
Cox proportional hazards models were used toestimatevaccine
efficacy, with estimated infection times grouped into six 6-
month periods. A simulation-based procedure was used to es-
timate the 95% CI for vaccine efficacy over time . Adjusted
Trial of rgp120 HIV-1 Vaccine among IDUs • JID 2006:194 (15 December) • 1665
trial, Bangkok, Thailand.
Cumulative HIV-1 incidence among injection drug users (IDUs) participating in the AIDSVAX B/E vaccine
Vaccine (n p 1267)Placebo (n p 1260)All (n p 2527)
Less than primary
9.5 (7.5– 12.0)
aHigher risk was defined as 2 or more of the following criteria being met at the baseline risk assessment: use of injection drugs regularly;
use of injection drugs daily or weekly; use of injection drugs with shared needles; history of incarceration during the past 6 months; partner
was an IDU; or sharing needles with partner. Lower risk was defined as the presence of !2 of these criteria.
vaccine efficacy was estimated by including demographic and
baseline risk-behavior variables as covariates. Similar statistical
methods were used to assess HIV-1 subtype E vaccine efficacy,
with participants with non–subtype E infections censored at
their estimated infection dates.
Secondary end-point analysis.
sessed the time between the date of HIV-1 infection anddisease
progression. Three outcomes were evaluated: treatment initi-
ation; first treatment initiation or viral load 110,000 copies/mL
starting at 1 month after infection, to avoid the acute stage;
and the first AIDS-defining condition. Generalized estimation
equations were used to model pretreatment HIV-1 load and
outcomes between the vaccine and placebo recipients.
Scatter plots were used to descriptively
compare preinfection antibody response levels between HIV-
1–infected and HIV-1–uninfected vaccinerecipients;Wei-John-
son tests were used to evaluate differences betweenthesegroups
at 1 ormoretimepoints.Case-cohortCoxproportionalhazards
models (adjusted for demographic and behavioral variables)
were used to estimate the relative risks of HIV-1 infection for
quartiles of antibody level included as time-dependent covar-
iates . Participants with outlying intervals between im-
munization and sampling (150 days) were excluded. All P val-
ues were 2-sided and were unadjusted for multiplicity.
Time-to-event analysis as-
The protocol of the present study was reviewed and approved
by the ethics review committees of the Thailand MOPH, Mahi-
dol University, and the BMA and by an institutional review
board of the CDC. A data and safetymonitoringboard(DSMB)
events, and deaths) and to conduct the 1 interim efficacy anal-
ysis. At this analysis, the trial could have been stopped if sta-
tistically significant protection from HIV-1 infection was dem-
onstrated among vaccine recipients. No specific futilityanalysis
Screening and enrollment.
risk characteristics of the trial participants have been described
elsewhere [27, 28]. Briefly, between March 1999 and August
2000, 4943 IDUs were screened, and 2546 were enrolled. Their
median age was 26.0 years (range, 20–59 years); 93.4% were
men, and 67.3% had at least a secondary education. During
the 6 months before enrollment, 93.8% reported injecting, and
33.0% reported needle sharing; 61.3% reported havingreceived
methadone detoxification, 20.9% reported having received
methadone maintenance, and 17.9% reported not having re-
ceived drug treatment. Almost all volunteers had injected her-
oin (98.5%), and the remainder had injected stimulants or
benzodiazepines. Daily injection was reported by 39.4%. Of the
446 (17.5%) who reported incarceration during the prior 6
months, 12.0% reported having injected in police holding cells,
and 11.2% reported having injected in prisons.
Of the 2546 IDUs enrolled, 2295 (90.1%)
were followed up for 36 months or until HIV-1 seroconversion
(figure 1). Of the enrolled volunteers, 230 were identified as
The screening, enrollment, and
1666 • JID 2006:194 (15 December) • Pitisuttithum et al.
over time in vaccine recipients (C) and placebo recipients (D), for study participants in the intention-to-treat cohort who became infected with HIV-1.
The solid lines indicate average values, and the dotted lines indicate 95% confidence intervals.
Pretreatment log plasma HIV-1 RNA loads over time in vaccine recipients (A) and placebo recipients (B) and pretreatment CD4 cell counts
being HIV-1 infected. Of these, 79 started ART before October
2001, and 18 started after [23, 24, 29]. Later, immunobloting
or NAT revealed that 19 enrolled participants were HIV-1 in-
fected at screening. Thus, the ITT cohort consisted of 2527
HIV-1–negative volunteers at entry: 1267 in the vaccine group,
and 1260 in the placebo group. The DSMB interim safety and
efficacy analysis in October 2002 recommended that the trial
proceed as planned.
During the trial, self-reportsof druginjectiondecreasedfrom
93.8% to 56.0% (), and self-reports of needle sharingP ! .001
decreased from 33.0% to 16.3% (
urine testing to validate the self-reports was conducted. Thirty-
nine trial-related social harms, including discrimination andloss
of opportunity, were reported by 37 volunteers.Disturbance
in a personal relationship related to the voluntary disclosure of
trial participation was the most common event (33 volunteers).
) . No drug-useP ! .001
Trial of rgp120 HIV-1 Vaccine among IDUs • JID 2006:194 (15 December) • 1667
copies/mL), for study participants in the intention-to-treat cohort who became infected with HIV-1.
Kaplan-Meier curves for the time from infection diagnosis to the composite end point of drug-treatment initiation or viral failure (110,000
Social harms were typicallyresolvedwithcounselingbytrialstaff.
Police harassment or arrests in relation to trialparticipationwere
not among the social harms reported .
Tenderness at the injection site—in 902 (71.0%)
vaccine recipients and 830 (65.7%) placebo recipients—wasthe
most commonly reported adverse event and did not increase
accidental injury was the most common (128 [30.9%]), fol-
lowed by drug overdose (49 [11.8%]), and sepsis (22 [5.3%]).
The most common cause of 102 deaths was drug overdose (38
[37.3%]), sepsis (17 [16.7%]), accidental injury (12 [11.8%]),
and suicide (8 [7.8%]); homicide as a cause of death was not
reported. There were no differences between vaccine and pla-
cebo recipients in these respects.
Rates of infection and vaccine efficacy.
1 incidence rate was 3.4 infections/100 person-years (95% CI,
3.0–3.9 infections/100 person-years). There were 106 HIV-1
infections (8.4%) in the vaccine group and 105 (8.3%) in the
placebo group. Of the 211 HIV-1 infections, 164 (77.7%; 83
in vaccine recipients, and 81 inplaceborecipients)weresubtype
recipients) were subtype B?, 1 was subtype B (in a vaccine recip-
ient), and the remaining 14 (6.6%) were untypeable. The co-
variate-unadjusted estimate of vaccine efficacy was 0.1% (95%
CI, ?30.8 to 23.8;, log-rank test). The estimate of theP p .99
unadjusted vaccine efficacy for subtype CRF01_AE infection
The pooled HIV-
was ?1.4% (95% CI, ?37.7 to 25.4;
Estimated curves for remaining free of HIV-1 infection by sub-
type and study arm are shown in figure 2.
There was no evidence of a calendar-time trend in incidence
by 6-month time period. HIV-1 infection rates by study arm,
sex, age, education, and baseline risk behavior are shown in
table 1. The unadjusted and adjusted estimates of vaccine ef-
ficacy were similar, suggesting no confounding by animbalance
of demographic factors or risk behaviors at baseline.
Markers of HIV-1 disease progression.
participants were followed for a maximum of 36 months
(median, 22 months). No significant differences between HIV-
1–infected vaccine recipients and HIV–1–infected placebo re-
cipients were found with regard to plasma HIV-1 loads or CD4
cell counts (figure 3), onset or clinical course of AIDS-defining
conditions, time to treatment initiation, and time to the first
event of viral failure or treatment initiation (figure 4). This
analysis was practically equivalent to assessing time to viral
failure without “confounding” by treatment, because of the183
events, only 3 were due to treatment initiation before viral
failure. These results were similar when stratified by subtype B
and subtype E infection.
The peak preinfection antibody levels
for gp120, A244 V2, A244 V3, blocking of A244 binding to
CD4, and MN neutralization were not significantly different
between the106 HIV-1–infectedandthe115randomlysampled
, log-rank test).P p .93
1668 • JID 2006:194 (15 December) • Pitisuttithum et al.
lines), for study participants in the intention-to-treat cohort. For vaccine recipients with incident HIV-1 infection, antibody levels were measured in
the last peak sample collected before the estimated date of HIV-1 infection; for uninfected vaccine recipients, antibody levels were measured for all
7 peak time points (at months 0.5, 1.5, 6.5, 12.5, 18.5, 24.5, and 30.5).
Average antibody levels for vaccine recipients with incident HIV-1 infection (solid lines) and for uninfected vaccine recipients (dotted
uninfected vaccine recipients (
ents, with or without adjustment for age, education, and be-
havioral risk, the level of the most recent peak preinfection
immune responses did not correlate with the rate of HIV-1
infection ( ). In the samples assessed for antibody re-P 1 .2
sponse, all vaccine recipients, but none of the 12 placebo re-
cipients, developed antibodies to gp120. Figure 5 shows allpeak
antibody levels for uninfected vaccinerecipientsandforvaccine
recipients with incident HIV-1 infection measured before the
estimated date of HIV-1 infection. The geometric mean peak
1.5, 6.5, 12.5, 18.5, 24.5, and 30.5 were 214, 3972, 5707, 5327,
4482, and 4247, respectively.
). Among vaccine recipi-P 1 .2
In this successfully completed trial of AIDSVAX B/E, the can-
didate vaccine did not prevent HIV-1 infection or delay disease
progression. No evidence was found of vaccine efficacy against
either subtype B or E virus in predefined demographic or risk-
behavior subgroups with the highest levelsof antibodyresponse
to the vaccine.
The lack of vaccine efficacy in this study is similar to that
in a recent trial of a similar vaccine, rgp120 B/B, which was
evaluated among men who have sex with men and women at
high risk in North America and The Netherlands [22, 30].
Unlike in the present trial, in the rgp120 B/B trial there was
an interesting trend toward modest efficacy, although it was
not significant after adjustment for multiplicity in nonwhite
persons and in volunteers with the highest HIV risk behavior.
Possible explanations for any disparity between the 2 trials
include differences in demographics, circulating HIV subtypes,
and route of transmission (sexual vs. parenteral). However,
demographic differences do not seem to play a role, because,
in the rgp120 B/B trial, the few Asian participants did not
contradict the trend in nonwhite persons. Differences in HIV
subtypes is also an unlikely explanation, because, in the Thai
Trial of rgp120 HIV-1 Vaccine among IDUs • JID 2006:194 (15 December) • 1669
trial, the genetic variation between the infecting subtype
CRF01_AE viruses and the vaccine subtype E components was
less than that between the infecting subtype B viruses and the
subtype B vaccine components (D. Jobs, personal communi-
cation). Thus, any differences in outcome between the 2 trials,
if real, are likely the result of dissimilarities between sexual and
parenteral HIV transmission with respect to dynamics, viral
loads, and local and systemic protective mechanisms.
It has been hypothesized that the failure in both trials was
due to the lack of induction of neutralizing antibodies against
genetically diverse primary HIV-1 isolates. However, there are
a number of important findings. The vaccines in both trials
appeared to be safe, and earlier concerns about possible en-
hancement of HIV-1 disease progression [31–33] could not be
confirmed. In the rgp120 B/B trial, higher peak antibody re-
sponses to the vaccine appeared to correlate with a lower risk
of HIV-1 infection . Of note, the AIDSVAX B/E vaccine is
being used as the booster portion of a combination regimen
that uses an attenuated canarypox vector (ALVAC vCP1521;
Aventis Pasture) in the world’s third phase 3 HIV-1 vaccine
efficacy trial, which is currently under way in Thailand .
Despite the lack of efficacy of AIDSVAX B/E as a stand-alone
vaccine, a treasure of useful information has been obtained
from this trial and from the preceding 5 years of epidemio-
logical, biomedical, and sociobehavioral research among IDUs
in Bangkok [27, 28]. Measures of HIV incidence and their
changes over time are perhaps the best documented for any
population group in the world [10, 36–39], and studies of viral
characterization [1, 18, 19, 40, 41] and disease progression[42–
44] have provided crucial information for current and future
HIV vaccine trials. Our study has demonstrated that IDUs can
be enrolled and followed and are compliant andthatcounseling
can quickly, although not completely, lower risk behaviors and
sustain this level over time [27, 28]. The trial alsodemonstrated
the importance of monitoring drug-use trends so that drug-
use counseling can be effectively tailored .
No clean injection equipment was made available to partic-
ipants at study clinics, and some have identified this asanethics
problem. However, Thailand’s narcotics law prohibits the dis-
tribution of clean injection equipment, and its HIV-prevention
policy favors cessation of heroin use through methadone treat-
needles and syringes can be bought at drug and convenience
stories without prescription for the modest price of 4–10 Thai
bahts (US $0.10–$0.20). Indeed, when asked at every 6-month
study visit (as part of ourrisk assessment), 195%ofparticipants
said they could obtain new and unused needles without any
problem. Moreover, the inclusion of needle exchange as a mea-
sure to prevent HIV infection, which would not have been
feasible outside the clinical trial setting, would have seriously
affected the external validity of our phase 3 efficacy study.
Toward the end of the trial, in February 2003, the Thai
Government initiated its “war on drugs,” to reverse the in-
creasing trend in methamphetamine use that had begunduring
the mid 1990s [46, 47]. Heroin users in drug treatment, the
population from which we recruited our participants, were not
the campaign target . Consequently, police harassment and
arrest in relationto trial participationwerenotamongthesocial
harms reported . Several human-rights organizations have
expressed concern  over the reported increase in drug-use-
related homicides during the war on drugs; however, homicide
was not among the causes of death reported in our trial.
An effective vaccine remains the best hope to control the
global HIV-1 epidemic, especially in developing countries. It is
disappointing that the first 2 HIV-1 candidate vaccines did not
prevent infection. Nonetheless, the tremendous amount of sci-
entific information gained from the years of work leading up
to, as well as during, the conduct of these studies will be in-
valuable in preparing for future large-scale HIV-1 vaccine tri-
als—as well as other biomedical intervention trials in high-risk
populations around the world .
BANGKOK VACCINE EVALUATION GROUP
Bangkok Metropolitan Administration.
(principal investigator of the BVEG), Boonrawd Prasittipol,
turas, and Suphak Vanichseni.
Valai Bussaratid, Jaranit Kaewkung-
wal, Dwip Kitayaporn, Sricharoen Migasena, Benjaluck Phon-
rat, Punnee Pitisuttithum, Sawangjai Pungpak, and Pravan
Thailand Ministry of Public Health–US Centers for Disease
Control and Prevention Collaboration.
wanachan, Thitima Cherdtrakulkiat, Wanitchaya Kittikraisak,
Wanna Leelawiwat, Michael Martin, Janet McNicholl, Supa-
wadee Na-Pompet, Chalintorn Sinthuwattanawibul, Jordan W.
Tappero, Frits van Griensven, Punneeporn Wasinrapee, and
US Centers for Disease Control and Prevention.
and Timothy D. Mastro.
Phillip Berman, Lisa Brooks, Marlene Cher-
now, Don Francis, Carolyn Gee, Marc Gurwith, William Hey-
ward, Tina Ippolito, David Jobes, Tina Kalanon, Elizabeth Li,
Aimee Luck, Karin Orelind, Patti Orozco-Cronin, Michael L.
Peterson, and Faruk Sinangil.
Statistical Center for HIV/AIDS Research and Prevention,
Fred Hutchinson Cancer Research Center.
Maggie Wang, and Michael Hudgens.
bert (Statistical Center for HIV/AIDS Research and Prevention,
1670 • JID 2006:194 (15 December) • Pitisuttithum et al.
Marc Gurwith, William Heyward, Michael Martin, Frits van
Griensven, Dale Hu, and Jordan W. Tappero.
We thank Piyamaith Yodnane, Suwanee Raktham, Krit Hiranras, Soon-
thorn Srichan, Veerawat Hantavichai, Paichit Pawabutr, Prayura Kunasol,
Sornchai Looareesuwan, Pornchai Matangkasomabut,PraphanKitisin,Na-
than Winslow, Mark McLaughlin, Marc Drucker, Jim Young, Kate Mac-
Queen, Alan Greenberg, and Charles Vitek.
1. Kitayaporn D, Vanichseni S, Mastro TD, et al. Infection with HIV-1
subtypes B and E in injecting drug users screened for enrollment into
a prospective cohort in Bangkok, Thailand. J Acquir Immune Defic
Syndr Hum Retrovirol 1998;19:289–95.
2. Wright NH, Vanichseni S, Akarasewi P, Wasi C, Choopanya K. Was
the 1988 HIV epidemic among Bangkok’s injecting drug users a com-
mon source outbreak? AIDS 1994;8:529–32.
3. The Thai Working Group on HIV/AIDS Projection, March 2001. Pro-
jections for HIV/AIDS in Thailand: 2000–2020. Nonthaburi, Thailand:
Division of AIDS, Department of Communicable Disease Control,
Ministry of Public Health, 2001.
4. Phoolcharoen W, Ungchusak K, Sittitrai W, Brown T. Thailand:lessons
from a strong national response to HIV/AIDS. AIDS 1998;12 (Suppl
5. Belshe RB, Graham BS, Keefer MC, et al. Neutralizing antibodies to
HIV-1 in seronegative volunteers immunized with recombinant gp120
from the MN strain of HIV-1. JAMA 1994;272:488–9.
6. Gorse GJ, Patel GB, Newman FK, et al. Antibody to native human
immunodeficiency virus type 1 envelope glycoproteins induces by IIIB
7. Fultz PN, Nara P, Barre-Sinoussi F, et al. Vaccine protection of chim-
panzees against challenge with HIV-1-infected peripheral bloodmono-
nuclear cells. Science 1992;256:1687–90.
8. Migasena S, Suntharasamai P, Pitisuttithum P, et al. AIDSVAX? (MN)
in Bangkok injecting drug users: a report on safety and immunogeni-
city, including macrophage-tropic virus neutralization. AIDS Res Hum
9. Pitisuttithum P, Nitayaphan S, Thongcharoen P, et al. Safety and im-
munogenicity of combinations of recombinant subtypeEandBhuman
immunodeficiency virus type 1 envelope glycoprotein 120 vaccines in
healthy Thai adults. J Infect Dis 2003;188:219–27.
10. Vanichseni S, Kitayaporn D, Mastro TD, et al. Continued high HIV-
1 incidence in a vaccine trial preparatory cohort of injection drugusers
in Bangkok, Thailand. AIDS 2001;15:397–405.
11. MacQueen KM, Vanichseni S, Kitayaporn D, et al. Willingness of in-
jection drug users to participate in an HIV vaccine efficacy trial in
Bangkok, Thailand. J Acquir Immune Defic Syndr 1999;21:243–51.
12. Mastro TD, Kitayaporn D, Weninger BG, et al. Estimating the number
of HIV-infected injection drug users in Bangkok: a capture-recapture
method. Am J Public Health 1994;84:1094–9.
13. Pitisuttithum P, Choopanya K, Vanichseni S, et al. Social harms in
injecting drug users participating in the first phase III HIV vaccine
trial in a non-Western country. J Acquir Immune Defic Syndr (in
14. Berman PW. Development of bivalent rgp120 vaccines to prevent HIV
type 1 infection. AIDS Res HumRetroviruses1998;14(Suppl3):S277–89.
15. Francis DP, Gregory T, McElrath MJ, et al. Advancing AIDSVAX? to
phase 3: safety, immunogenicity and plans for phase 3. AIDS Res Hum
Retroviruses 1998;14(Suppl 3):S325–31.
16. McCutchan FE, Hegerich PA, Brennan TP, et al. Genetic variants of
HIV-1 in Thailand. AIDS Res Hum Retroviruses 1992;8:1887–95.
17. Berman PW, Huang W, Riddle L, et al. Development of bivalent (B/
E) vaccines able to neutralize CCR5-dependent virusesfromtheUnited
States and Thailand. Virology 1999;265:1–9.
18. Subbarao S, Vanichseni S, Hu DJ, et al. Genetic characterization of
HIV-1 subtype E and B strains from a prospective cohort of injecting
drug users in Bangkok, Thailand. AIDS Res Hum Retroviruses 2000;
19. Phan K-O, Callahan ME, Vanichseni S, et al. A comparison of full-
length gp120 from incident HIV-1 subtype E and B infections in Bang-
kok injecting drug users to prototype E and B strains that are com-
ponents of a candidate vaccine. AIDS Res Hum Retroviruses 2000;16:
20. Centers for Disease Control and Prevention. 1997 revised guidelines
for performing CD4+ T-cell determinations in persons infected with
human immunodeficiency virus (HIV). MMWR Recomm Rep 1997;
21. Pitisuttithum P, Berman PW, Phonrat B, et al. Phase I/II study of a
candidate vaccine designed against the B and E subtypes of HIV-1. J
Acquir Immune Defic Syndr 2004;37:1160–5.
22. The rgp120 HIV Vaccine Study Group. Placebo-controlledphase3trial
of a recombinant glycoprotein 120 vaccine to prevent HIV-1 infection.
J Infect Dis 2005;191:654–65.
23. Ministry of Public Health. National guidelines for the clinical man-
agement of HIV infection in children and adults. 6th ed. Nonthaburi,
Thailand: Division of AIDS, Department of Communicable Disease
Control, Ministry of Public Health, 2000.
24. Ministry of Public Health. Guidelines for the care and treatment of
HIV/AIDS in children and adults in Thailand. 7th ed. Nonthaburi,
Thailand: Division of AIDS, Department of Communicable Disease
Control, Ministry of Public Health, 2002.
25. Gilbert PB, Wei LJ, Kosorok MR, Clemens JD. Simultaneousinferences
on the contrast of two hazard functions with censored observations.
26. Borgan O, Langholz B, Samuelsen SO, Goldstein L, PogodaJ. Exposure
stratified case-cohort designs. Lifetime Data Anal 2000;6:39–58.
27. van Griensven F, Keawkungwal J, Vanichseni S, et al. Lack of increased
HIV risk behavior among injection drug users participating in the
AIDSVAX? B/E HIV vaccine efficacy trial in Bangkok, Thailand.AIDS
28. Vanichseni S, Tappero JW, PitisuttithumP, etal. Recruitment,screening
and characteristics of injection drug users participating in the AIDS-
VAX? B/E HIV vaccine trial, Bangkok,Thailand.AIDS2004;18:311–6.
29. Martin M, Mansakul W, Tappero JW, et al. Care and treatment of HIV-
infected injecting drug users during an HIV vaccineefficacytrial,Bang-
kok, Thailand [abstract MoPe3350]. The XV International AIDS Con-
ference (Bangkok) 2004. Available at: http://www.iasociety.org/ejias
-search/search.asp. Accessed 11 July 2006.
30. Gilbert PB, Ackers ML, Berman PW, et al. HIV-1 virologic and im-
munologic progression and initiation of antiretroviral therapy among
HIV-1-infected subjects in a trial of the efficacy of recombinant gly-
coprotein 120 vaccine. J Infect Dis 2005;192:974–83.
31. Robinson WE, Montefiori DC, Mitchell WM. Antibody-dependenten-
hancement of human immunodeficiency virus type 1 infection. Lancet
32. Mascola JR, Mathison BJ, Zack PM, Walker MC, Halstead SB, Burke
DS. Summary report: workshop on the potential risks of antibody-
dependent enhancement in human HIV vaccine trials. AIDS Res Hum
33. Morens DM. Antibody-dependent enhancement of infection and the
pathogenesis of viral disease. Clin Infect Dis 1994;19:500–12.
34. Gilbert PB, Peterson ML, Follmann D, et al. Correlation between im-
munologic responses to a recombinant glycoprotein 120 vaccine and
incidence of HIV-1 infection in a phase 3 HIV-1 preventive vaccine
trial. J Infect Dis 2005;191:666–77.
Trial of rgp120 HIV-1 Vaccine among IDUs • JID 2006:194 (15 December) • 1671
35. Cohen J. Public health. AIDS vaccine still alive as booster after second
failure in Thailand. Science 2003;302:1309–10.
36. Hu DJ, Subbarao S, Vanichseni S, et al. Higher viral loads and other
risk factors associated with HIV-1 seroconversion during a period of
high HIV-1 incidence among injection drug users in Bangkok. JAcquir
Immune Defic Syndr 2002;30:240–7.
37. Nguyen L, Hu DJ, Choopanya K, at al. Genetic analysis of incident
HIV-1 strains among injection drug users in Bangkok: evidence for
multiple transmission clusters during a period of high incidence. J
Acquir Immune Defic Syndr 2002;30:248–56.
38. Hu DJ, Vanichseni S, Mock PA, et al. HIV-1 incidence estimates by
detection of recent infection from a cross-sectional sampling of injec-
tion drug users in Bangkok: use of an IgG capture BED enzyme im-
munoassay. AIDS Res Hum Retroviruses 2003;19:727–30.
39. Hu DJ, Subbarao S, Vanichseni S, Mock PA, et al. Frequency of HIV-
1 dual subtype infections, including intersubtype superinfections,among
injection drug users in Bangkok, Thailand. AIDS 2005;19:303–8.
40. Wasi C, Herring B, Raktham S, et al. Determination of HIV-1subtypes
in injecting drug users in Bangkok, Thailand, using peptide binding
enzyme immunoassay and heteroduplex mobility assay: evidence on
increasing infection with HIV-1 subtype E. AIDS 1995;9:843–9.
41. Kalish ML, Baldwin A, Raktham S, et al. The evolving molecular ep-
idemiology of HIV-1 envelope subtypesin injectingdrugusersinBang-
kok, Thailand: implications for HIV vaccine trials.AIDS1995;9:851–7.
42. Hu DJ, Vanichseni S, Mastro TD, et al. Viral load differences in early
infection with two HIV-1 subtypes. AIDS 2001;15:683–91.
43. Buchacz K, Hu DJ, Vanichseni S, Mock P, et al. Early markers of HIV-
1 disease progression in a prospective cohort of seroconvertersinBang-
kok, Thailand: implications for vaccine trials. J Acquir Immune Defic
44. Nguyen L, Li M, Chaowanachan T, Vanichseni S, et al. CCR5 promoter
human haplogroups associated with HIV-1 disease progression in Thai
injection drug users. AIDS 2004;18:1327–33.
45. van Griensven F, Pitisuttithum P, Vanichseni S, et al. Trends in the
injection of midazolam and other drugs and needle sharing among
injection drug users enrolled in the AIDSVAX B/E HIV-1 vaccine trial
in Bangkok, Thailand. Int J Drug Policy 2005;16:171–5.
46. Vongchak T, Kawichai S, Sherman S, et al. The influence of Thailand’s
2003 ‘war on drugs’ policy on self-reported drug use among injection
drug users in Chiang Mai, Thailand. Int J Drug Policy 2005;16:115–21.
47. Bureau of Health Service System Development. Number of IDU in
treatment in Bangkok, 1991–2004. Nonthaburi, Thailand: Department
of Health Services, Ministry of Public Health, 2005.
48. Thailand’s “war on drugs.” The Wire: AmnestyInternational’smonthly
magazine, 2003. Available at: http://web.amnesty.org/web/wire.nsf/
May2003/Thailand. Accessed 15 May 2006.
49. Centers for Disease Control and Prevention (CDC). CDC trialsof daily
oral tenofovir for preventing HIV infection: phase II and III clinical
trails in Botswana, Thailand, and the United States. CDC, 2004. Avail-
able at: http://www.cdc.gov/hiv/pubs/TenofovirFactSheet.pdf.Accessed
13 December 2005.