INTRODUCTION

Antiretroviral therapy (ART) has proven remarkably effective at reducing acquired immunodeficiency syndrome (AIDS)-related morbidity and mortality. However, ART is unable to eradicate the virus and is associated with substantial drug-related toxicities and increasing drug resistance. It would therefore be desirable to develop immune-based therapies that enhance immune control of viral replication and reduce the requirement for ART. The goal of an HIV-1 therapeutic vaccine would be to enhance virus-specific cellular and humoral immune responses, which could in turn lead to reduced HIV-1 RNA levels, increased CD4+ T-lymphocyte counts, and delayed progression to AIDS. Several candidate therapeutic vaccines have already been evaluated for safety and immunogenicity, but no substantial clinical benefits have been demonstrated. This chapter reviews the immunologic rationale for HIV-1 therapeutic vaccines and highlights key preclinical and clinical studies to date.

IMMUNOLOGIC RATIONALE

Increasing evidence has demonstrated that adaptive immune responses are critical for control of viral replication. A therapeutic vaccine aims to enhance HIV-1-specific immunity and to improve this immune control. CD8+ T-lymphocytes exert antiviral cytolytic activity as well as other effector functions, and CD4+ T-lymphocytes provide immunologic ‘help’ to B cells and CD8+ T-lymphocytes.^1–3 The importance of CD8+ T-lymphocytes has been directly demonstrated by experimental depletion of CD8+ T-lymphocytes in SIV-infected rhesus macaques by monoclonal antibody administration.^4,5 Moreover, in humans with acute HIV-1 infection, potent CD8+ T-lymphocyte responses were temporally correlated with initial control of HIV-1 viremia.^6,7 Long-term nonprogressors (LTNPs) have also been shown to have vigorous HIV-1-specific CD8+ and CD4+ T-lymphocyte responses that were associated with preservation of CD4+ T-lymphocyte counts and durable control of viral replication without the use of ART.^1,8,9 T-lymphocytes from LTNPs retain a high proliferative capacity and exhibit high levels of perforinexpression, suggesting the potential importance of these immune functions.¹⁰

Patients who initiate ART during acute HIV-1 infection typically have preserved HIV-1-specific CD41 T-lymphocyte responses as well as detectable CD8+ T-lymphocyte responses.^2,11–13 In an observational study, supervised treatment interruptions (STIs) of HIV-1-infected individuals treated with ART during acute infection resulted in viral rebound, which in turn stimulated expansion of HIV-1-specific CD4+ and CD8+ T-lymphocyte responses and led to transiently improved virologic control during subsequent STIs.¹³ Although the durability of this immune control was limited,¹⁴ this study showed that immune-based interventions can exert detectable effects on viral replication. STI studies in chronically HIV-1-infected individuals, however, have failed to demonstrate improved control of viral replication.¹⁵ In addition, a recent study has highlighted potential safety concerns associated with performing STIs in chronically HIV-1-infected individuals.¹⁶ These data suggest that it may be substantially more challenging to design therapeutic vaccines for the majority of patients who are treated during chronic HIV-1 infection as compared with individuals who begin ART during acute HIV-1 infection.

The immunologic rationale of therapeutic vaccination is illustrated in Figure 31-1. As shown in Figure 31-1A, HIV-1-infected subjects treated with effective ART typically exhibit low levels of HIV-1-specific humoral and cellular immune responses and undetectable viral RNA levels. Following discontinuation of ART, HIV-1 RNA levels predictably rebound within several weeks to pretreatment ‘setpoint’ levels. As shown in Figure 31-1B, therapeutic vaccination in the context of suppressive ART may enhance virus-specific immune responses, which could improve immune control of viral replication following discontinuation of ART.

Figure 31-1 Schematic depiction of rationale for therapeutic vaccination. HIV-1-specific cellular immune responses (black) and viral RNA levels (blue) in HIV-1-infected subjects on ART. (A) Discontinuation of ART results in predictable viral rebound with increased virus-specific cellular immune responses that fail to control the resurgent virus. (B) Therapeutic vaccination while on suppressive ART results in enhancement of HIV-1-specific cellular immune responses that may partially control viral rebound following withdrawal of ART.

Although there is currently no evidence that HIV-1 therapeutic vaccines will afford any clinical benefit, a recent study has demonstrated the clinical utility of a therapeutic vaccine for a different virus. A live, attenuated varicella-zoster virus (VZV) vaccine has been shown to reduce the incidence and severity of illness from clinical herpes zoster as well as the incidence of postherpetic neuralgia, demonstrating that therapeutic vaccination for this persistent viral infection can result in clinical benefits.¹⁷ Although the mechanism of action of the VZV vaccine remains to be established, it presumably involves enhancement of VZV-specific immunity. However, VZV causes latent infection without active viral replication, whereas HIV-1 causes chronic infection with persistent viral replication, and thus it will likely prove substantially more challenging to produce an HIV-1 therapeutic vaccine than a VZV therapeutic vaccine.

NONHUMAN PRIMATE STUDIES

Nonhuman primates have proven instrumental in the development of both prophylactic and therapeutic HIV-1 vaccine candidates. Although rhesus macaques cannot be infected with HIV-1, they can be infected with simian immunodeficiency virus (SIV), which leads to loss of CD4+ T-lymphocytes and clinical AIDS-like syndromes. Although significant biologic differences exist between SIV-infected rhesus macaques and HIV-1-infected humans, this animal model has proven useful for evaluating the immunogenicity and protective efficacy of candidate AIDS vaccines.

There have been several studies to date evaluating the immunogenicity and potential efficacy of therapeutic vaccines using this animal model. Early studies focused on recombinant protein subunit vaccines and whole inactivated virus vaccines. These vaccine modalities, however, have yielded disappointing results, presumably because they failed to elicit broadly reactive neutralizing antibodies and did not induce CD8+ T-lymphocyte responses. Current vaccine strategies are therefore largely focused on gene delivery technologies to augment virus-specific cellular immune responses, including recombinant virus vector-based vaccines, plasmid DNA vaccines, and dendritic cell (DC)-based vaccines.

A proof-of-concept study in the macaque model utilized a recombinant poxvirus vaccine expressing SIV structural genes (NYVAC-SIV). In rhesus macaques treated with ART following primary SIV infection, vaccination with NYVAC-SIV augmented SIV-specific CD4+ and CD8+ T-lymphocyte responses. Following discontinuation of ART, all animals exhibited viral rebound, but SIV replication was spontaneously controlled in six of eight vaccinated animals and in four of seven unvaccinated controls.¹⁸ Although the observed virologic control following treatment discontinuation likely reflected the fact that ART was initiated during acute infection, this study nevertheless demonstrated that virus-specific immune responses could be enhanced by therapeutic vaccination. In a second study from this same group, NYVAC-SIV vaccines expressing gag-pol-env and rev-tat-nef were administered with interleukin-2 (IL-2) as an adjuvant to chronically SIV-infected macaques on ART.¹⁹ Following treatment discontinuation, viral setpoint was significantly lower in vaccinated macaques than in control animals, and this difference persisted for at least 4 months. Viral containment in this study was also inversely correlated with vaccine-induced virus-specific CD41 and CD8+ T-cell responses.

Therapeutic vaccine strategies utilizing plasmid DNA vaccines have focused on novel delivery approaches, given the relatively low immunogenicity of DNA vaccines in nonhuman primates and humans as compared with small animal models. One strategy involves transfecting DCs with plasmid DNA vaccines and then administering these DCs to macaques.²⁰ A topical DNA vaccine has also been developed and was shown to enhance SIV-specific cellular immune responses in chronically SIV-infected macaques on ART.²¹ Following withdrawal of ART, vaccinated animals exhibited improved control of viral rebound as compared with unvaccinated controls. Another study demonstrated that autologous DCs pulsed with inactivated SIV enhanced virus-specific cellular immune responses and resulted in a substantial reduction in viral RNA levels in vaccinated animals as compared with unvaccinated controls.²²

Studies in nonhuman primates have shown that several vaccine candidates can enhance virus-specific immune responses and can partially control viral rebound following discontinuation of ART. Although encouraging, translating these preclinical findings into clinical trials remains a major challenge. Studies in nonhuman primates are often designed to utilize vaccines that are based on sequences identical to those of the infecting virus. Moreover, many experimental vaccines have proven less immunogenic in humans as compared with nonhuman primates. Clinical trials are therefore critical for evaluating the potential utility of therapeutic HIV-1vaccines.

CLINICAL STUDIES

As shown in Table 31-1, studies of therapeutic HIV-1 vaccines in humans have explored a variety of different vaccine modalities, including recombinant protein subunits, inactivated virus particles, virus-like particles, plasmid DNA vaccines, recombinant poxvirus vector-based vaccines and DC-based vaccines.^23–41 These studies have demonstrated the safety and immunogenicity of several vaccine approaches, but none of these studies have convincingly demonstrated clinically meaningful endpoints. In HIV-1-infected individuals receiving ART, vaccination with the attenuated poxvirus vectors ALVAC vCP1452 and modified vaccinia Ankara (MVA) proved safe and immunogenic.^34–3638 The magnitude of vaccine-enhanced immune responses elicited by both of these vectors, however, was modest. Similarly, an inactivated, envelope-depleted HIV-1 preparation (Remune) was shown to be safe and immunogenic²⁸, but a large phase III trial failed to show any significant effect on HIV-1 viral RNA levels or progression-free survival.²⁷ The lack of a beneficial effect in this study may have reflected incomplete suppression of viral replication with ART prior to vaccination as well as the inability of this vaccine to elicit potent virus-specific CD8+ T-lymphocyte responses.

Table 31-1 Results of Recent HIV-1 Therapeutic Vaccine Clinical Studies

A randomized, double-blinded, placebo-controlled study has also recently evaluated the impact of therapeutic vaccination with ALVAC vCP1452 and/or Remune in 79 HIV-1-infected subjects treated with ART during acute infection.³⁹ These vaccines led to detectable increases of HIV-1-specific CD4+ and CD8+ T-cell responses, but they failed to afford any improvement in virologic control 24 weeks after discontinuation of ART. This was a disappointing result, since subjects treated during acute HIV-1 infection theoretically should be the most amenable to immune-based interventional strategies as a result of their relatively preserved immune systems. There was also no significant correlation between gag-specific CD4+ and CD8+ T-lymphocyte responses and viral RNA levels following discontinuation of ART in this study, suggesting that the correlates of immunity are more complex than simply the magnitude of these immune responses.

Another study has recently investigated therapeutic vaccination with ALVAC vCP1433 and HIV-1 lipopeptides (Lipo6T) together with IL-2 as an adjuvant in chronically HIV-1-infected subjects on ART. In this study, a higher percentage of vaccinees developed significant HIV-1 p24 proliferative responses and CD4+ T-lymphocyte responses as compared with controls. The breadth and magnitude of HIV-1-specific IFN-γ-producing CD8+ T-lymphocyte responses also improved in the vaccine group as compared with the control group. After treatment discontinuation, 24% of vaccinees exhibited a statistically significant reduced viral setpoint as compared with only 5% of controls, and virologic control correlated with vaccine-elicited cellular immune responses.⁴⁰

Another novel therapeutic vaccine approach involves the use of antigen-loaded autologous DCs. In one study, autologous DCs were pulsed with virus obtained during an initial interruption of ART and were then utilized as a therapeutic vaccine after ART was re-initiated. Following a second interruption of ART, there was a significant lengthening in mean doubling time during the viral rebound.⁴¹ In a second study, 18 chronically HIV-1-infected individuals who were not receiving ART were treated with monocyte-derived DCs loaded with autologous inactivated HIV-1. HIV-1 RNA levels in vaccinated individuals decreased by 80% over the first 112 days, and eight subjects had virologic suppression of over 90% for at least 1 year.⁴² Moreover, virologic suppression appeared to correlate with HIV-1-specific CD4+ and CD8+ T-lymphocyte responses. These results suggest that certain therapeutic vaccines can indeed exert a degree of virologic control, but these studies involved only a limited number of subjects and the clinical relevance of these observations remains unclear.