Abstract

BackgroundWe report composite results from the Merck phase I program of near-consensus clade B human immunodeficiency virus (HIV) type 1 gag vaccines

MethodsHealthy HIV-uninfected adults were enrolled in 6 blinded placebo-controlled studies evaluating the immunogenicity of (1) a 4-dose regimen of a DNA vaccine, (2) a 3-dose priming regimen of the DNA vaccine with a booster dose of an adenovirus type 5 (Ad5)-vectored vaccine, or (3) a 3-dose regimen of the Ad5 vaccine. The DNA plasmid was provided with or without an aluminum phosphate or CRL1005 adjuvant. The primary end point was the unfractionated HIV-1 gag-specific interferon γ enzyme-linked immunospot (ELISpot) response 4 weeks after the final dose

ResultsOverall, 254 (83%) of 307 subjects randomized to the vaccine groups were evaluable. Adjuvants did not enhance immunogenicity of the DNA vaccine. Postboost ELISpot responder frequencies were higher for Ad5-containing regimens than for the DNA/DNA regimen (33%) but were similar for DNA/Ad5 (55%) and Ad5/Ad5 (50%). DNA/DNA elicited mainly a CD4 response, whereas Ad5/Ad5 elicited mainly a CD8 response; DNA/Ad5 generated CD4 and CD8 responses comparable to those of DNA/DNA and Ad5/Ad5, respectively

ConclusionsThe DNA vaccine alone or as a priming regimen for the Ad5 vaccine did not increase unfractionated ELISpot responses compared with the Ad5 vaccine alone. Qualitative T cell responses to different vaccine regimens deserve further study

Even with significant advances in therapy, the AIDS pandemic continues to be responsible for devastating morbidity and mortality throughout the world [1–4], especially in regions with limited access to antiretroviral medications. Human immunodeficiency virus (HIV) vaccine strategies to elicit protection by neutralizing antibodies have met with repeated failures [5–7]. Diversity of envelope glycoproteins within and between subtypes allows the virus to adapt to selective pressure and poses a formidable obstacle to developing a broadly effective vaccine based on humoral immunity directed against a single surface epitope [8]. An alternative approach has focused on vaccines that stimulate cell-mediated immune (CMI) responses against several relatively conserved HIV-1 proteins in an attempt to prevent or modulate infection [9–15]. In phase I studies, the Merck adenovirus type 5 (Ad5)–vectored near-consensus clade B HIV-1 vaccines were immunogenic in HIV-seronegative subjects [16–18]. Disappointingly, the STEP proof-of-concept trial failed to demonstrate benefit from a 3-dose regimen of a trivalent Ad5 vaccine, and numerically more HIV infections were acquired by vaccine recipients than by placebo recipients in some subgroups [19, 20]. Despite the unexpected results of the STEP study, it remains imperative to explore all aspects of the Merck HIV vaccine program to inform future vaccine efforts against HIV and other intracellular pathogens [20, 21]

Development of an effective CMI-based vaccine regimen against HIV must address several challenges [20–22]. The STEP trial results underscored the need to better understand the role of immunodominance and antigenic breadth in vaccine development. Both quantitative and qualitative aspects of effective CMI responses to natural infection and vaccines are still not sufficiently understood. Vaccines using viral vectors such as Ad5 can elicit robust immune responses to transgene inserts [16–19, 23, 24]. However, for vectors derived from common human viral pathogens, neutralizing antibodies from natural infection may attenuate or otherwise negatively affect vaccine immunogenicity [25]. Preexistent and/or vaccine-induced immunity to the Ad5 vector substantially dampens the CMI response to Ad5-vectored HIV-1 vaccines [16, 17]. One strategy to overcome high Ad5 titers would be to use a DNA vaccine for the entire series or as a priming series before a booster dose of an Ad5 vaccine [26]. Furthermore, the use of adjuvants may significantly enhance immune responses to some DNA vaccines [27–30]

In a nonhuman primate model, the combination of DNA priming and Ad5 boosting resulted in better short-term control of simian immunodeficiency virus compared with a regimen consisting of Ad5-vaccine alone [26], indicating that a DNA-prime Ad5-boost vaccine regimen could enhance immune responses at least transiently. Nonetheless, single-vaccine regimens have significant logistical advantages with regard to manufacturing, distribution, and implementation of immunization programs. A clear benefit would need to be reasonably anticipated for a prime-boost regimen using 2 different vaccines to be a preferred option

We have conducted 6 independent phase I studies of a HIV-1 gag vaccine administered as follows: (1) priming and boosting with the DNA vaccine (DNA/DNA regimen), (2) priming with DNA vaccine and boosting with the Ad5 vaccine (DNA/Ad5 regimen), and (3) priming and boosting with the Ad5 vaccine (Ad5/Ad5 regimen). The effects of these different regimens on enzyme-linked immunospot (ELISpot) responses to HIV-1 clade B Gag peptide pools measured 4 weeks after the last dose are the major focus of our analysis

Materials and Methods

Objectives and outcomesThe primary objective of the 6 studies (Merck V520 protocols 002, 007, 008, 011, 012, and 018) was to assess the safety, tolerability, and immunogenicity of various delivery strategies for a near-consensus clade B HIV-1 gag vaccine. Results from protocols 007, 012, and 018 have been reported elsewhere [17, 18]. Protocol 002 evaluated a regimen of repeated vaccinations with gag DNA vaccine in phosphate-buffered saline (PBS) at 1 or 5 mg per dose. Protocol 008 evaluated a priming series of the gag DNA vaccine at the 5-mg dosage with or without aluminum phosphate (AlPO4) adjuvant, followed by a boosting dose with the same DNA vaccine or the Ad5 vaccine containing 1×1010 viral particles (vp). Protocol 011 evaluated a priming series of the gag DNA vaccine at the 5-mg dose, containing either the AlPO4 or CRL1005 adjuvant, followed by a boosting dose with the same DNA vaccine or the Ad5 vaccine containing 1×108 or 1×1010 vp

The aim of the current analysis was to assess the effects on the immune response of the following 3 strategies: (1) a 4-dose regimen of a DNA gag vaccine alone, containing 5 mg per 1-mL dose; (2) a 3-dose priming regimen of the DNA gag vaccine (5 mg/dose), followed by a booster dose of an Ad5 gag vaccine at 1×1010 vp per 1-mL dose; and (3) a 3-dose regimen of an Ad5 gag vaccine (1×1010 vp/dose). Depending on the study, the DNA vaccine was provided in PBS with or without an AlPO4 or CRL1005 adjuvant. Immunogenicity end points were the proportions of subjects with a positive unfractionated gag-specific interferon (IFN) γ ELISpot response to 15-mer Gag peptide pools using peripheral blood mononuclear cells (PBMCs) obtained periodically after vaccination, including 4 weeks after the last priming dose (week 12 for DNA-based regimens and week 8 for the Ad5/Ad5 regimen) and 4 weeks after the final dose (week 30 for all regimens)

Vaccine compositionsThree DNA vaccine formulations were studied. Each of the DNA formulations contained the HIV-1 gag plasmid expressing the gag gene from an HIV-1 strain (CAM-1) that closely resembles the clade B consensus amino acid sequence. The vaccine backbone comprises a cytomegalovirus promoter and intron A, a bovine growth hormone-derived polyadenylation and transcriptional termination sequence, and is derived from a pUC backbone. The plasmid is propagated in Escherichia coli and then isolated using a chromatography-based process. The gag transgene is expressed to high levels in mammalian cells, whereas the plasmid replicates only in E. coli

The HIV-1 gag DNA plasmid was formulated in a sterile solution of PBS (6 mmol/L sodium phosphate, 150 mmol/L sodium chloride; pH 7.2) with or without either AlPO4 adjuvant (0.7 mg/mL) or the nonionic block copolymer adjuvant CRL1005 (7.5 mg/mL) with 0.6 mmol/L benzalkonium chloride. CRL1005 (CytRx) is a linear triblock copolymer with a polyoxypropylene core of molecular weight 12 kDa and 5% polyoxyetheylene [29, 30]. Placebo consisted of an identical PBS solution, with or without 0.7 mg of AlPO4 per milliliter

Two similar adenoviral vectors (Ad5 HIV-1 gag and MRKAd5 HIV-1 gag) were studied, both based on a replication-defective adenovirus type 5 backbone [14, 16–18]. The immunogenicity and safety profiles of these vectors were similar, but the original Ad5 vector exhibited more genetic instability with passage than did MRKAd5

Study designsAll studies in this analysis were multicenter, blinded, randomized, placebo-controlled phase I trials (Figure 1). Healthy subjects between 18 and 55 years of age without evidence of HIV infection were assessed for eligibility. Subjects were excluded if they were considered to be at high risk of acquiring HIV infection during the study or had a history of significant chronic medical conditions. Female subjects with childbearing potential were required to have a negative urine pregnancy test immediately before receiving any injections. Protocols were approved by the institutional review boards at participating centers. Written informed consent was obtained from all subjects. Within each study, enrolled subjects were randomized to receive vaccine or placebo according to allocation schedules generated by computer. Ad5 neutralizing antibody titers were measured at baseline in all studies, and these titers were used to stratify enrollment in some protocols. Investigators, subjects, clinical safety monitors, and laboratory personnel performing the biological assays were blinded to treatment assignments. Clinic visits for safety evaluations were scheduled for 2–4 weeks after each dose. Study personnel monitored injection site, systemic, and laboratory adverse events. All studies except protocol 018 were conducted exclusively in the United States. Although protocol 018 was conducted internationally, only data from subjects enrolled at sites in the United States were included, to avoid confounding due to genetic heterogeneity among subjects and different PBMC-processing protocols across the world regions [18]

Figure 1

Treatment paradigms. Specimens obtained at vaccination visits were collected before administration of vaccine. Ad5, adenovirus type 5; PBMC, peripheral blood mononuclear cell

Figure 1

Treatment paradigms. Specimens obtained at vaccination visits were collected before administration of vaccine. Ad5, adenovirus type 5; PBMC, peripheral blood mononuclear cell

Laboratory assaysNeutralizing antibodies to Ad5 were measured as described elsewhere; a titer of <18 was considered negative, as defined by the assay specifications [31]. As in earlier analyses, a cutoff titer of 200 was used to categorize high versus low baseline immunity to Ad5 [16–19]

To assess vaccine immunogenicity, unfractionated IFN-γ ELISpot assays using 15-mer clade B HIV-1 Gag peptide pools matched to the vaccine inserts were performed on PBMCs obtained at weeks 4, 8, 12, 26, 30, 34, and 42 at Merck Research Laboratories to detect HIV-1 gag-specific T cell responses after vaccination [32]. A positive ELISpot response was defined as ⩾55 spot-forming cells (SFCs)/106 PBMCs and ⩾4-fold increased response compared with the control medium. The relative contribution of CD4 and CD8 T cells to the IFN-γ response was assessed by intracellular cytokine flow cytometry [33]; a positive response was defined as ⩾400 CD4+ or CD8+ IFN-γ positive events/106 lymphocytes and ⩾3-fold increased response compared with the control medium

Statistical analysesData were combined across protocols. Cell processing and immunological assays for all specimens used in this analysis were conducted, using standardized procedures that ensured the comparability of results across studies. Because the Ad5 and MRKAd5 vaccine vectors had produced observationally similar ELISpot responses, the analyses in this report combined these 2 vectors in the Ad5 vaccine groups [16]. Because a dose response was apparent for both the DNA and Ad5 vaccines, analysis was limited to the 5-mg dose group for the DNA vaccine and to the 1×1010–vp dose group for the Ad5 vaccine

Only subjects receiving all scheduled doses of vaccine and without major protocol violations were included in the analysis of immunogenicity. Results of unfractionated HIV-1 gag-specific IFN-γ ELISpot assays were summarized according to vaccine regimen at each time point. Summary statistics included the proportions of ELISpot responders and the geometric means of the quantitative ELISpot values for each treatment group. Geometric means were calculated for all subjects, regardless of whether their ELISpot result was deemed positive or negative. Summary data are also presented separately for subjects with high (>200) or low (⩽200) baseline Ad5 titers. Between-group differences in the frequencies of ELISpot responders were evaluated using the method of Miettinen and Nurminen [34]. The key pairwise comparisons of interest were the unadjuvanted DNA vaccine versus either the AlPO4- or CRL1005-adjuvanted DNA vaccine and the Ad5/Ad5 regimen versus either the DNA/DNA or the DNA/Ad5 regimen. Dunnett's multiplicity adjustment [35] was applied separately for the 2 sets of comparisons, and differences were considered statistically significant at P<.05 (2-tailed multiplicity-adjusted P values). An analogous testing approach using pairwise t tests was used to compare geometric mean unfractionated ELISpot responses as well as geometric mean CD4+ and CD8+ T cell responses determined by intracellular cytokine flow cytometry

Results

Subject accounting and baseline characteristicsThe demographic characteristics across the vaccine groups were generally comparable (Table 1). Of 307 randomized subjects eligible for this analysis, 254 (83%) completed the studies through week 30 and were included in the primary immunogenicity analysis. Subjects in protocol 002 received a second booster dose of vaccine ⩾6 months after the first booster dose, and many subjects in protocols 007 and 012 were rolled over into another vaccine trial [36]. Accordingly, the current analysis includes only data up to the week 42 visit in all studies

Table 1

Baseline Characteristics by Treatment Group (Pooled from Protocols 002, 007, 008, 011, 012, and 018)

Table 1

Baseline Characteristics by Treatment Group (Pooled from Protocols 002, 007, 008, 011, 012, and 018)

ImmunogenicityTable 2 shows descriptive summaries of the postpriming and postboosting ELISpot results for the 7 vaccine regimens in subjects in the per-protocol population. The response frequency and magnitude differed by regimen

Table 2

Summary of Human Immunodeficiency Virus Type 1 gag-Specific Enzyme-Linked Immunospot (ELISpot) Responses, by Baseline Adenovirus Type 5 (Ad5) Titer (⩽200 vs >200) and Vaccine Regimen

Table 2

Summary of Human Immunodeficiency Virus Type 1 gag-Specific Enzyme-Linked Immunospot (ELISpot) Responses, by Baseline Adenovirus Type 5 (Ad5) Titer (⩽200 vs >200) and Vaccine Regimen

Unadjuvanted versus adjuvanted DNA vaccine regimensThe addition of an adjuvant did not appear to enhance the priming response to the DNA vaccines. Compared with the DNA vaccine without an adjuvant, neither AlPO4 nor CRL1005 significantly augmented the ELISpot responses to the priming regimen at week 12. The priming series was identical for regimens 1 and 4, regimens 2 and 5, and regimens 3 and 6 listed in Table 2, allowing us to combine the DNA regimens that used the same adjuvant for the analysis at week 12. The percentages of responders assessed 4 weeks after completion of the priming series (week 12) were 21.9% (geometric mean ELISpot response, 67) for the PBS diluent without an adjuvant (regimens 1 + 4; n=64), compared with 22.5% (ELISpot response, 79) and 16.0% (ELISpot response, 70) for PBS with AlPO4 (regimens 2 + 5; n=71) or CRL1005 (regimens 3 + 6; n=50), respectively. The postprime response frequency for all the DNA groups combined (regimens 1–6; n=185) was 20.5% (geometric mean ELISpot response, 72). Furthermore, there was no statistical evidence to suggest that the addition of AlPO4 or CRL1005 to the DNA vaccine formulation affected the ELISpot responses when subjects were later given booster doses with either the DNA or Ad5 vaccine. Although the DNA vaccine with the CRL1005 adjuvant yielded the numerically lowest response rates, these results were not significantly inferior to those of the DNA vaccine without an adjuvant. Consequently, results were pooled across the DNA formulations irrespective of adjuvant, in order to compare vaccine regimens

Ad5/Ad5 versus DNA/DNA or DNA/Ad5 regimens Statistically significant differences in the proportion of ELISpot responders and corresponding geometric means were detected between the DNA/DNA and Ad5/Ad5 regimens but not between the DNA/Ad5 and Ad5/Ad5 regimens assessed 4 weeks after the booster dose at week 30 (Table 3). After empirical exclusion of subjects receiving the DNA vaccine with the CRL1005 adjuvant (which consistently yielded the lowest ELISpot responses) and pooling of the 2 other DNA-prime (with PBS ± AlPO4) Ad5-boost arms, a modestly higher postboost response frequency of 62% was achieved, compared with the 50% postboost response after the Ad5/Ad5 regimen

Table 3

Summary of Human Immunodeficiency Virus Type 1 gag-Specific Enzyme-Linked Immunospot (ELISpot) Responses 4 Weeks after Booster Dose (at Week 30)

Table 3

Summary of Human Immunodeficiency Virus Type 1 gag-Specific Enzyme-Linked Immunospot (ELISpot) Responses 4 Weeks after Booster Dose (at Week 30)

Baseline Ad5 titer effectNot unexpectedly, baseline Ad5 titers did not consistently affect DNA vaccine responses. On the other hand, the immunogenicity of the Ad5 vaccine was systematically affected by baseline Ad5 titers. The time course of ELISpot responses for DNA/DNA, DNA/Ad5, and Ad5/Ad5 vaccine regimens are displayed in Figure 2 by Ad5 status. For the Ad5/Ad5 regimen, subjects in the lower Ad5 titer group had notably higher postprime and postboost responses than subjects in the higher Ad5 titer group (Table 2). In the lower Ad5 titer group, statistically significant differences in the proportion of ELISpot responders and corresponding geometric means were detected between the Ad5/Ad5 and DNA/DNA (but not between the Ad5/Ad5 and DNA/Ad5) regimens; however, no significant differences were detected between vaccine regimens in the higher Ad5 titer group (Table 3). Overall, the DNA-prime Ad5-boost strategy did not materially increase the ELISpot responses relative to the Ad5/Ad5 regimen in the group with high baseline Ad5 titers. After subjects receiving the CRL1005-adjuvanted DNA vaccine were empirically excluded from that group, the corresponding week 30 ELISpot response frequencies were 37% with the DNA-prime (with PBS 3 AlPO4) Ad5-boost regimen versus 24% with the Ad5/Ad5 regimen

Figure 2

Human immunodeficiency virus type 1 gag-specific enzyme-linked immunospot (ELISpot) responses over time by regimen. A positive ELISpot response was defined as ⩾55 spot-forming cells (SFCs)/106 peripheral blood mononuclear cells (PBMCs) and ⩾4-fold higher than the control medium. Geometric mean ELISpot responses are reported as SFCs/106 PBMCs, and include both responders and nonresponders. Numbers under graphs (N) represent all subjects with available data who had completed the vaccine series to the specified time point

Figure 2

Human immunodeficiency virus type 1 gag-specific enzyme-linked immunospot (ELISpot) responses over time by regimen. A positive ELISpot response was defined as ⩾55 spot-forming cells (SFCs)/106 peripheral blood mononuclear cells (PBMCs) and ⩾4-fold higher than the control medium. Geometric mean ELISpot responses are reported as SFCs/106 PBMCs, and include both responders and nonresponders. Numbers under graphs (N) represent all subjects with available data who had completed the vaccine series to the specified time point

Qualitative T cell responsesThe DNA/DNA vaccine regimen generated a primarily CD4+ T cell response, whereas the Ad5/Ad5 vaccine regimen induced a predominant CD8+ T cell response (Table 4). The DNA/Ad5 regimen produced a CD4+ T cell response comparable to that of the DNA/DNA regimen and a CD8+ T cell response comparable to that of the Ad5/Ad5 regimen

Table 4

Summary of Human Immunodeficiency Virus Type 1 gag-Specific Intracellular Cytokine Staining CD4+ T Cell and CD8+ T Cell Responses 4 Weeks after Booster Dose (at Week 30)

Table 4

Summary of Human Immunodeficiency Virus Type 1 gag-Specific Intracellular Cytokine Staining CD4+ T Cell and CD8+ T Cell Responses 4 Weeks after Booster Dose (at Week 30)

Discussion

In this analysis of near-consensus clade B HIV-1 gag vaccines, we did not demonstrate a clear benefit of a DNA vaccine over an Ad5-only regimen in enhancing CMI responses, as measured by unfractionated IFN-γ ELISpot responses. In particular, DNA priming did not appear to overcome the dampening of immune responses to the Ad5 vaccine in subjects with preexisting Ad5 neutralizing antibodies. This finding in human subjects contrasts with results generated by the corresponding regimens in animal models [23, 26]. However, given the size and nature of the clinical trials, along with the resultant wide confidence intervals around the point estimates for ELISpot response frequencies, we cannot confidently exclude the possibility that meaningful differences among the tested vaccine regimens do indeed exist

Adjuvant effects on immunogenicity may differ based on the specific adjuvant and vaccine being tested, the dosing schedule, and the immune response being measured. The 2 adjuvants used in these studies (AlPO4 and CRL1005) provided no evident immunological benefit over the DNA vaccine without an adjuvant but may have affected the frequency or intensity of injection-site reactions. Contrary to the animal model, the DNA vaccine with CRL1005 generally elicited lower ELISpot responses than the other 2 DNA vaccine formulations in the clinical trials

Pooling study arms when biologically, methodologically, and statistically appropriate to increase sample size may provide enhanced analytical power, but such analyses may still suffer from type 2 error. The groups receiving the DNA vaccine with either of the 2 adjuvants responded similarly to the group receiving the DNA vaccine without an adjuvant. Consequently, the DNA groups were combined regardless of adjuvant for the formal analysis. It nonetheless remains possible that small effects in opposite directions from a common control might cancel each other in the composite data. As a plausible illustration with regard to the DNA vaccine, responses with the AlPO4 adjuvant were the highest and responses with the CRL1005 adjuvant were the lowest. Although the differences were not significantly different, numerically higher ELISpot response frequencies were generated by the DNA/Ad5 regimens when the CRL1005 adjuvant arm was excluded than by the Ad5/Ad5 regimen. Even if these small quantitative differences are real, their clinical relevance can be questioned. It remains possible that DNA priming under some circumstances may provide a marginally beneficial strategy to circumvent preexisting or vaccine-induced immunity to an Ad5 vector

The specific immunogens included in a vaccine and the predominant cell type invoked by the vaccine-adjuvant combination may be critical determinants of the protective efficacy of a CMI-inducing HIV vaccine [37–40]. Our analysis concentrated on unfractionated ELISpot responses to Gag. Inclusion of env in a vaccine could alter its immunogenicity. Potential qualitative differences in the T cell responses elicited by different vaccine regimens may prove more critical than small quantitative differences in ELISpot responses between the DNA/Ad5 and Ad5/Ad5 regimens. The ELISpot assay exhibits the greatest sensitivity in detecting vaccine responses, whereas the less sensitive intracellular cytokine flow cytometry provides supplemental information on the phenotype of the responding T cells [41]. Our data suggest that the cellular response to the DNA/Ad5 regimen may be broader than the response to DNA/DNA or Ad5/Ad5 regimens, but whether a broader response translates into better protection remains an unsettled question in human subjects. The effect of a prophylactic DNA/Ad5 prime-boost vaccine strategy on the HIV set point in uninfected volunteers who acquire infection after vaccination is currently under investigation [42]

The relevance of nonhuman primate models for evaluation of HIV vaccines has been questioned [43]. Our results in humans are at odds with earlier findings in monkeys regarding the relative merits of a DNA/Ad5 versus an Ad5/Ad5 strategy and the advantages of using adjuvants with the DNA vaccine [23, 26]. An Ad5 vaccine regimen could induce a dominant CD8+ T cell response in rhesus monkeys [23]. Complementary experiments had indicated that Ad5 immunity would mitigate but not abolish the response. Boosting with the Ad5 vaccine after priming with a CRL1005-adjuvanted DNA vaccine produced higher ELISpot responses than Ad5/Ad5 vaccination in monkeys. Attenuation of simian immunodeficiency virus SIVmac239 infection by prophylactic immunization with a DNA/Ad5 vaccine regimen was then demonstrated [26]. In contrast, data from the human trials failed to establish any clear-cut quantitative advantage for DNA priming or the use of the CRL1005 adjuvant in overcoming high baseline immunity to Ad5

Definitive conclusions about the relative merits of DNA/Ad5 and Ad5/Ad5 vaccine regimens for protection against HIV and other intracellular pathogens cannot be drawn from our analysis, especially for populations with widespread Ad5 immunity. Our data suggest that DNA priming did little to augment unfractionated IFN-γ ELISpot responses to the Ad5 vaccine, even when preimmunization antibody titers to Ad5 were high. A 4-dose immunization series with 2 different vaccines, as used in the DNA/Ad5 strategy studied here, would be logistically more difficult to implement than simpler schedules. If boosting with the last Ad5 vaccine dose were required for an efficacious and durable immune response, the sequential DNA/Ad5 approach could prove counterproductive under circumstances in which the entire vaccine series might not be routinely completed. Other permutations of DNA/Ad5 combination regimens were not studied. Because DNA vaccines induce an immune response that is functionally different from (and perhaps complementary to) that elicited by Ad5-vectored vaccines [37], DNA/Ad5 prime-boost regimens could potentially offer theoretical advantages over either vaccine alone that were not evident in the unfractionated IFN-γ ELISpot responses. Whether the qualitative attributes of a DNA/Ad5 prime-boost regimen will translate into an effective vaccine program has yet to be established

Acknowledgments

We thank the study participants and staff at the study sites for their time and support. The invaluable contributions of Scott Thaler (deceased), Robin Isaacs, Randi Leavitt, Andrew Bett, Timothy Tobery, Deborah Brown, and the entire study teams for all the protocols are immensely appreciated. The technical assistance of Joann DiLullo and Karyn Davis was essential to the preparation of this manuscript

References

1
Beyrer
C
HIV epidemiology update and transmission factors: risks and risk contexts—16th International AIDS Conference epidemiology plenary
Clin Infect Dis
 , 
2007
, vol. 
44
 (pg. 
981
-
7
)
2
Joint United Nations Programme on HIV/AIDS
 
2008 Report on the global AIDS epidemic: executive summary. July 2008. Available at http://data.unaids.org/pub/GlobalReport/2008/JC1511_GR08_ExecutiveSummary_en.pdf. Accessed 24 February 2009
3
Gray
RH
Li
X
Wawer
MJ
, et al.  . 
Stochastic simulation of the impact of antiretroviral therapy and HIV vaccines on HIV transmission: Rakai, Uganda
AIDS
 , 
2003
, vol. 
17
 (pg. 
1941
-
51
)
4
John
R
Arango-Jaramillo
S
Self
S
Schwartz
DH
Modeling partially effective HIV vaccines in vitro
J Infect Dis
 , 
2004
, vol. 
189
 (pg. 
616
-
23
)
5
Flynn
NM
Forthal
DN
Harro
CD
Judson
FN
Mayer
KH
Placebo-controlled phase 3 trial of a recombinant glycoprotein 120 vaccine to prevent HIV-1 infection
J Infect Dis
 , 
2005
, vol. 
191
 (pg. 
654
-
65
)
6
Pitisuttithum
P
Gilbert
P
Gurwith
M
, et al.  . 
Randomized, double-blind, placebo-controlled efficacy trial of a bivalent recombinant glycoprotein 120 HIV-1 vaccine among injection drug users in Bangkok, Thailand
J Infect Dis
 , 
2006
, vol. 
194
 (pg. 
1661
-
71
)
7
Kwong
PD
Doyle
ML
Casper
DJ
, et al.  . 
HIV-1 evades antibody-mediated neutralization through conformational masking of receptor-binding sites
Nature
 , 
2002
, vol. 
420
 (pg. 
678
-
82
)
8
Lynch
RM
Shen
T
Gnanakaran
S
Derdeyn
CA
Appreciating HIV type 1 diversity: subtype differences in Env
AIDS Res Hum Retroviruses
 , 
2009
, vol. 
25
 (pg. 
237
-
48
)
9
Fu
TM
Dubey
SA
Mehrotra
DV
, et al.  . 
Evaluation of cellular immune responses in subjects chronically infected with HIV type 1
AIDS Res Hum Retroviruses
 , 
2007
, vol. 
23
 (pg. 
67
-
76
)
10
Harrer
T
Harrer
E
Kalams
SA
, et al.  . 
Strong cytotoxic T cell and weak neutralizing antibody responses in a subset of persons with stable nonprogressing HIV type 1 infection
AIDS Res Hum Retroviruses
 , 
1996
, vol. 
12
 (pg. 
585
-
92
)
11
Jin
X
Bauer
DE
Tuttleton
SE
, et al.  . 
Dramatic rise in plasma viremia after CD8+ T cell depletion in simian immunodeficiency virus-infected macaques
J Exp Med
 , 
1999
, vol. 
189
 (pg. 
991
-
8
)
12
Matano
T
Shibata
R
Siemon
C
Connors
M
Lane
HC
Administration of an anti-CD8 monoclonal antibody interferes with the clearance of chimeric simian/human immunodeficiency virus during primary infections of rhesus macaques
J Virol
 , 
1998
, vol. 
72
 (pg. 
164
-
9
)
13
Schmitz
JE
Kuroda
MJ
Santra
S
, et al.  . 
Control of viremia in simian immunodeficiency virus infection by CD8+ lymphocytes
Science
 , 
1999
, vol. 
283
 (pg. 
857
-
60
)
14
Shiver
JW
Fu
TM
Chen
L
, et al.  . 
Replication-incompetent adenoviral vaccine vector elicits effective anti-immunodeficiency-virus immunity
Nature
 , 
2002
, vol. 
415
 (pg. 
331
-
5
)
15
Harvey
BG
Maroni
J
O'Donoghue
KA
, et al.  . 
Safety of local delivery of low- and intermediate-dose adenovirus gene transfer vectors to individuals with a spectrum of morbid conditions
Hum Gene Ther
 , 
2002
, vol. 
13
 (pg. 
15
-
63
)
16
Priddy
FH
Brown
D
Kublin
J
, et al.  . 
Safety and immunogenicity of a replication-incompetent adenovirus type 5 HIV-1 clade B gag/pol/nef vaccine in healthy adults Merck V520–016 study group
Clin Infect Dis
 , 
2008
, vol. 
46
 (pg. 
1769
-
81
)
17
Harro
CD
Robertson
MN
Lally
MA
, et al.  . 
Safety and immunogenicity of adenovirus-vectored near-consensus HIV type 1 clade B gag vaccines in healthy adults
AIDS Res Hum Retroviruses
 , 
2009
, vol. 
25
 (pg. 
103
-
14
)
18
Hammer
S
Miller
M
Pape
J
, et al.  . 
Safety and immunogenicity of the MRKAd5 gag HIV-1 vaccine in a worldwide phase I study of healthy adults (Merck V520–018/HVTN 050) [abstract OA08-02]
Program and abstracts of AIDS Vaccine (Seattle)
 , 
2007
19
Buchbinder
SP
Mehrotra
DV
Duerr
A
, et al.  . 
Efficacy assessment of a cell-mediated immunity HIV-1 vaccine (the Step Study): a double-blind, randomised, placebo-controlled, test-of-concept trial
Lancet
 , 
2008
, vol. 
372
 (pg. 
1881
-
93
)
20
McElrath
MJ
De Rosa
SC
Moodie
Z
, et al.  . 
HIV-1 vaccine-induced immunity in the test-of-concept Step Study: a case-cohort analysis
Lancet
 , 
2008
, vol. 
372
 (pg. 
1894
-
905
)
21
Sekaly
RP
The failed HIV Merck vaccine study: a step back for a launching point for future vaccine development
J Exp Med
 , 
2008
, vol. 
205
 (pg. 
7
-
12
)
22
Masopust
D
Developing an HIV cytotoxic T-lymphocyte vaccine: issues of CD8 T-cell quantity, quality and location
J Intern Med
 , 
2009
, vol. 
265
 (pg. 
125
-
37
)
23
Casimiro
DR
Chen
L
Fu T-M
, et al.  . 
Comparative immunogenicity of rhesus monkeys of DNA plasmid, recombinant vaccinia virus, and replication-defective adenovirus vectors expressing a human immunodeficiency virus type 1 gag gene
J Virol
 , 
2003
, vol. 
77
 (pg. 
6305
-
13
)
24
Catanzaro
AT
Koup
RA
Roederer
M
, et al.  . 
Phase 1 safety and immunogenicity evaluation of a multiclade HIV-1 candidate vaccine delivered by a replication-defective recombinant adenovirus vector
J Infect Dis
 , 
2006
, vol. 
194
 (pg. 
1638
-
49
)
25
Sprangers
MC
Lakhai
W
Koudstaal
W
, et al.  . 
Quantifying adenovirus-neutralizing antibodies by luciferase transgene detection: addressing pre-existing immunity to vaccine and gene therapy vectors
J Clin Microbiol
 , 
2003
, vol. 
41
 (pg. 
5046
-
52
)
26
Casimiro
DR
Wang
F
Schleif
WA
, et al.  . 
Attenuation of simian immunodeficiency virus SIVmac239 infection by prophylactic immunization with DNA and recombinant adenoviral vaccine vectors expressing gag
J Virol
 , 
2005
, vol. 
79
 (pg. 
15547
-
55
)
27
Ulmer
JB
DeWitt
CM
Chastain
M
, et al.  . 
Enhancement of DNA vaccine potency using conventional aluminum adjuvants
Vaccine
 , 
1999
, vol. 
18
 (pg. 
18
-
28
)
28
Wang
S
Liu
X
Fisher
K
, et al.  . 
Enhanced type I immune response to a hepatitis B DNA vaccine by formulation with calcium- or aluminum phosphate
Vaccine
 , 
2000
, vol. 
18
 (pg. 
1227
-
35
)
29
Newman
MJ
Actor
JK
Balusubramanian
M
Jagannath
C
Use of nonionic block copolymers in vaccines and therapeutics
Crit Rev Ther Drug Carrier Syst
 , 
1998
, vol. 
15
 (pg. 
89
-
142
)
30
Todd
CW
Lee
E
Balusubramanian
M
, et al.  . 
Systematic development of a block copolymer adjuvant for trivalent influenza virus vaccine
Dev Biol Stand
 , 
1998
, vol. 
92
 (pg. 
341
-
51
)
31
Aste-Amezaga
M
Bett
AJ
Wang
F
, et al.  . 
Quantitative adenovirus neutralization assays based on the secreted alkaline phosphatase reporter gene: application in epidemiologic studies and in the design of adenovector vaccines
Human Gene Therapy
 , 
2004
, vol. 
15
 (pg. 
293
-
304
)
32
Dubey
S
Clair
J
Fu Tong-Ming
, et al.  . 
Detection of HIV vaccine-induced cell-mediated immunity in HIV seronegative clinical trial participants using an optimized and validated ELISPOT assay
J Acquir Immune Defic Syndr
 , 
2007
, vol. 
45
 (pg. 
20
-
7
)
33
Trigona
WL
Clair
JH
Persaud
N
, et al.  . 
Intracellular staining for HIV-specific IFN-gamma production: statistical analyses establish reproducibility and criteria for distinguishing positive responses
J Interferon Cytokine Res
 , 
2003
, vol. 
23
 (pg. 
369
-
77
)
34
Miettinen
O
Nurminen
M
Comparative analysis of two rates
Stat Med
 , 
1985
, vol. 
4
 (pg. 
213
-
26
)
35
Dunnett
CW
New tables for multiple comparisons with a control
Biometrics
 , 
1964
, vol. 
20
 (pg. 
482
-
91
)
36
Goepfert
P
Priddy
F
Spearman
P
, et al.  . 
Boosting adenovirus serotype 5 (Ad5) HIV-gag primed subjects with ALVAC HIV (vCP205) [abstract 505]
Program and abstracts of the 12th Conference on Retroviruses and Opportunistic Infections (Boston)
 , 
2005
37
Cox
KS
Clair
JH
Prokop
MT
, et al.  . 
DNAgag/Ad5gag and Ad5gag/Ad5gag vaccines induce multifunctional and distinct T-cell response profiles
J Virol
 , 
2008
, vol. 
82
 (pg. 
8161
-
71
)
38
Roberts
AD
Ely
KH
Woodland
DL
Differential contributions of central and defector memory T-cells to recall responses
J Exp Med
 , 
2005
, vol. 
202
 (pg. 
123
-
33
)
39
Staprans
SI
Barry
AP
Silvestri
G
, et al.  . 
Enhanced SIV replication and accelerated progression to AIDS in macaques primed to mount a CD4 T cell response to the SIV envelope protein
Proc Natl Acad Sci USA
 , 
2004
, vol. 
101
 (pg. 
13026
-
31
)
40
Barouch
DH
Kunstman
J
Glowczwskie
J
, et al.  . 
Viral escape from dominant simian immunodeficiency virus epitope-specific cytotoxic T lymphocytes in DNA-vaccinated rhesus monkeys
J Virol
 , 
2003
, vol. 
77
 (pg. 
7367
-
75
)
41
Tobery
TW
Dubey
SA
Anderson
K
, et al.  . 
A comparison of standard immunogenicity assays for monitoring HIV type 1 gag-specific T cell responses in Ad5 HIV type 1 gag vaccinated human subjects
AIDS Res Human Retrovir
 , 
2006
, vol. 
22
 (pg. 
1081
-
90
)
42
National Institute of Allergy and Infectious Diseases
 
NIAID will not move forward with the PAVE 100 HIV vaccine trial. 17 July 2008. Available at http://www3.niaid.nih.gov/news/newsreleases/2008/pave100.htm. Accessed 7 April 2009
43
Watkins
DI
Burton
ER
Kallas
EG
Moore
JP
Koff
WC
Non-human primate models and failure of the Merck HIV-1 vaccine in humans
Nat Med
 , 
2008
, vol. 
14
 (pg. 
617
-
21
)
Potential conflicts of interest: all non-Merck authors have served as investigators for Merck vaccine studies. C.D.R. has received grant support from Merck and other companies involved in developing human immunodeficiency virus (HIV) vaccines. C.H. has received contract support from Merck related to its influenza, HIV-1, and Staphylococcus aureus vaccine programs and has previously received fees for service on safety advisory committees for HIV-1 vaccine trials and review of clinical safety reports for S. aureus vaccine trials. J.G.K. was formerly employed by Merck and has been a paid consultant for the company. Current employees of Merck may own stock and/or stock options in the company
Financial support: Merck