Immunotherapy for AIDS Virus Infections: Cautious Optimism for Cell-Based Vaccine

Nina Bhardwaj, Bruce D. Walker


Nat Med. 2003;9(1) 


A vaccine that is able to work as a treatment for HIV-infected patients has been an elusive goal. Now experiments on macaques suggest that injections of modified dendritic cells might boost immunity to hold back disease progression (pages 27-32).

Highly active antiretroviral therapy (HAART) limits HIV-1 replication and retards disease progression, but drug toxicities and emergence of drug-resistant viruses are challenges for long-term control in infected persons. Alternative treatment strategies are clearly needed. The demonstration that brief interruptions in HAART during acute infection act to boost immune defenses against HIV-1[1] indicates that immunotherapy may be possible, but the durability of control with this approach is uncertain[2] and it has been less effective in the setting of chronic infection.[3] Lu et al. now provide remarkable evidence in an animal model that augmented immune control of immunodeficiency virus replication can be achieved by therapeutic immunization.[4] The authors immunized monkeys using a combination of potent antigen-presenting cells (APCs) called dendritic cells (DCs) and a chemically inactivated form of a simian AIDS virus. This approach achieved not only a dramatic reduction in levels of AIDS virus in the blood, but also an increase in CD4 cell numbers, all in the absence of HAART.

Intense interest in immunotherapy for HIV-1 infection has been precipitated by emerging data indicating that the immune system has a major, albeit usually insufficient, role in limiting HIV-1 replication. The correlates of protective immunity remain unknown, but there is a growing consensus that T cells, particularly cytolytic CD8+ T cells (CTLs), are the main mediators of viral control, as evidenced by the dramatic increase in viremia that occurs in animal models of AIDS virus infection after experimental removal of CD8+ T cells.[5,6] Virus-specific helper cells, which are critical to maintenance of CTL function, also likely have a role.[7] Viremia is thought to be less influenced by neutralizing antibodies, which are generally low in magnitude in HIV-1 infection and lack the ability to keep up with evolving viral variants in vivo.[8]

Together these data indicate that increasing the strength and breadth of HIV-1-specific cellular immune responses might have a clinical benefit, and have resulted in renewed interest in so-called therapeutic vaccines. However, for AIDS, as for cancer, there has been no proof of principle that therapeutic vaccines can work.

The experiments of Lu et al. suggest that immunotherapy may indeed be a realistic goal. Their approach does not involve treatment interruption or a classic vaccine, but rather the use of antigen-loaded DCs. These cells reside in an immature state in tissues such as the mucosa, where they capture incoming pathogens including HIV-1 (Fig. 1). Antigen-loaded DCs traffic to draining lymph nodes while undergoing 'maturation', a differentiation step that allows them to efficiently activate naive T cells and initiate immunity. In animal models, DCs generated ex vivo and delivered in vivo are highly effective at inducing protective immune responses against a variety of pathogens and tumors.[9] Moreover, in people, removing DCs and expanding them in vitro can easily be done. However, so far no one has demonstrated whether DCs can be used to increase control of immunodeficiency virus in animals or humans.

Human dendritic cell containing HIV-1. Electron micrograph shows a cell two hours after a pulse of inactivated HIV-1. The virus (dark structures, arrows) accumulates within endosomes. A vaccine based on SIV-pulsed dendritic cells shows promise in fighting off infection in macaques.

Human dendritic cell containing HIV-1. Electron micrograph shows a cell two hours after a pulse of inactivated HIV-1. The virus (dark structures, arrows) accumulates within endosomes. A vaccine based on SIV-pulsed dendritic cells shows promise in fighting off infection in macaques.

Using the simian immunodeficiency virus (SIV) model of AIDS virus infection, the authors isolated DCs from recently infected animals and injected them, after in vitroi ncubation, with SIV that had been chemically inactivated using aldrithiol-2 (AT-2). AT-2 covalently modifies zinc-finger cysteines of the viral nucleocapsid without affecting the conformation of the surface envelope glycoproteins.[10] Therefore the virus still binds to and fuses with target cells, but its life cycle is halted before reverse transcription. DCs are able to process and present antigens from non-replicating virus[11] and can activate T cells with relatively few numbers of HLA-peptide complexes.[12] Macaques previously infected with the pathogenic strain SIVmac251 were given five injections of AT-2-inactivated, SIVmac251-pulsed, autologous DCs at two-week intervals.

Within ten days of the first injection, a significant decrease in blood SIV cellular DNA and plasma RNA occurred along with a concomitant rise in cellular and humoral immune responses to SIV. By the third injection, viral load fell from >105 to <103 copies/ml and cellular immune responses peaked at levels six-fold greater than baseline. All immunized animals demonstrated a rise in CD4+ T-cell counts and developed neutralizing anti-bodies. The modulation of pathology was illustrated by the preservation of the lymphoid follicular dendritic cell network that is typically destroyed in progressive HIV infection. Significantly, plasma viremia was controlled for the duration of the study (up to 300 days post vaccination) in seven of ten immunized animals, compared with none of four controls that received DCs alone. The three immunized animals that failed to control viremia had significantly lower levels of cellular immunity and higher levels of lymph-node-associated SIV.

This study provides impressive evidence that improved immune control of an AIDS virus infection is possible, and if confirmed in monkeys and successfully adapted in humans it may represent a major new therapeutic approach to HIV-1. The fact that in vitro enerated DCs pulsed with inactivated AIDS virus were able to induce potent virus-specific immune responses, whereas DCs in vivo in the presence of what should be ample amounts of virus do not, suggests that the central immunologic defect in persons with HIV-1 infection may be in the induction phase of the immune response. The ability of this immunization approach to control viral replication is even more striking considering that induction of augmented immune responses in the setting of untreated infection is likely to occur in a milieu of helper-cell deficiency, ineffective CD8+ T-cell responses and compromised APC function. These studies also provide the first suggestion that therapeutic immunization alone in the absence of HAART may be effective.

The apparent success of this approach is encouraging. On the other hand, there are several features of this non-blinded study that must temper enthusiasm until the results can be confirmed in another cohort. The experiments involved macaques of Chinese origin, whereas most studies so far have involved Indian-origin macaques. The course of SIV infection in Chinese-origin macaques is not well defined, nor is the ability to transfer findings from this model to others. The homogeneity of viral load and CD4 cell counts at the start of the study is not typical of the heterogeneity one usually sees with SIVmac251 infection, even considering the novel exclusion of animals with particularly high or low viral loads, and suggests that there may be features unique to this cohort. The paucity of immune responses in the control infected animals is also unusual, but may relate to the sensitivity of the novel assays used.

The increase and subsequent decline in CTL responses in immunized animals following injections of antigen-loaded DCs were quite rapid, but have been described in other models of DC-based vaccination in humans and mice,[13,14] and may relate to cell trafficking. In addition, the animals were immunized early in their dis-ease course, 56 days after infection. Early immunologic intervention through treatment interruptions, which have shown promise in acute infection in humans, was less successful in chronic infection.[1,3] The extent to which similar control might be achieved in chronic infection following more viral diversification[15] is a critical issue in terms of the potential therapeutic impact of this approach.

There are several additional questions raised by the current study. Although this immunization strategy led to a dramatic drop in viral load and an increase in CD4 count, the durability of immune control is as yet uncertain. Other studies in monkeys indicate that sudden loss of control can occur after nearly a year of control, and can be associated with immune escape from CTL responses.[16] Certainly further longitudinal follow-up will be required, but the duration of control already achieved is quite impressive. Although the data indicate augmentation of CTL function and gradual increase in neutralizing antibodies, the precise factors responsible for the enhanced control are not clear. The cellular immune function studies do not allow an assessment of the relative contributions of CD4 and CD8 responses, or the relative breadth of these responses. Without further investigation it is not possible to say whether these CTLs represent newly activated responses or simply expanded memory responses. It is possible that DCs induced antiviral helper T cells, which in turn supported the generation and maintenance of CTLs and neutralizing antibodies. It will also be important to determine the extent to which the antibodies neutralize the evolving in vivo virus variants and whether they have a significant role in maintaining control of virus replication.

The goal of therapeutic vaccines is to modulate the course of disease by preventing suffering and prolonging life. The approach described by Lu et al., while preliminary, offers new hope of this possibility. This study at the very least revives the concept of inactivated virus as an effective immunogen and emphasizes the advantage of targeting antigens to DCs to induce effective immunity. The striking results of the study are unexpected, and it is imperative to rapidly determine whether similar results can be obtained in other cohorts of primates and whether they can be transferred to humans.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.