Brigitte Autran¹, Patrice Debré,¹, Bruce Walker² & Christine Katlama³   about the authors

¹ Brigitte Autran and Patrice Debré are at the Laboratoire d'Immunologie Cellulaire, Centre Hospitalier Universitaire Pitié-Salpétrière, 75651 Paris, France.
² Bruce Walker is at the Partners AIDS Research Center, Howard Hughes Medical Institute, Massachusetts General Hospital and Division of AIDS, Harvard Medical School, Boston MA02129, USA.
³ Christine Katlama is at the Service de Maladies Infectieuses, Centre Hospitalier Universitaire Pitié-Salpétrière, 75651 Paris, France.

correspondence to: Brigitte Autran brigitte.autran@psl.ap-hop-paris.fr

The successes of anti-retroviral treatments against HIV are limited today by the cost and toxicity of lifelong administration. An innovative therapeutic strategy has been proposed to boost the immune system of infected patients with HIV vaccines and to help limit the use of anti-retroviral treatments. This perspective article reviews the crucial questions raised by such a strategy and the main international efforts that are already set up to provide rapid answers — in particular, a not-for-profit international network that is dedicated to the development of therapeutic immunization programmes against HIV.

The recent successes in the development of anti-retroviral therapies that contain HIV-1 infection are, at present, challenged by the cost of this lifelong therapy and by its toxicity. As a result of these toxic side effects, treatment delays or interruptions are required. In the absence of a new therapeutic breakthrough, alternative strategies are needed. Immune-based therapeutic strategies that boost the immune response against HIV have been proposed to limit the use of anti-retroviral drugs. Indeed, enhancing immune control of the virus before anti-retroviral treatments are discontinued should help to prolong the period without therapy. However, this new strategy raises several issues that require rapid and efficient evaluation by the coordinated efforts of international and multidisciplinary communities. Apart from national agencies, private sponsorship and not-for-profit organizations that are dedicated to the development of therapeutic immunization against HIV will be instrumental in accelerating the development of such a challenging enterprise — by increasing the flexibility and rapidity of funding, as well as by enhancing the complementarity and creativity of researchers.

Limitations of drug therapy

The course of HIV infection has changed markedly following the introduction, in 1996, of potent new anti-retroviral regimens — known as highly active anti-retroviral therapy (HAART) (Fig. 1). Indeed, HAART regimens reduce the immunological damage that is caused by HIV and restore CD4⁺ T-cell numbers and host defences against pathogens¹. The clinical benefits were rapidly realized, with a marked reduction in the level of AIDS-related mortality and morbidity in countries that have access to HAART².

		Figure 1 |  Aims of therapeutic immunization against HIV.  Four phases can be distinguished during the course of an HIV infection. The pre-HAART (highly active anti-retroviral therapy) phase indicates the natural course after primary infection. During this phase, virus replication occurs, CD4+ T helper (TH) and CD8+ T cells specific for HIV are induced and a steady-state level of virus production is reached. With persistent virus replication, the number of CD4+ TH cells specific for HIV is markedly reduced and the CD4+ T-cell counts fall progressively. The initiation of HAART results in the control of virus replication to below threshold detection levels (less than two log copies per ml of plasma), reconstitution of the CD4+ T-cell compartment, a failure to restore HIV-specific CD4+ TH-cell responses and a reduction in the number of HIV-specific CD8+ T cells. During the combined HAART plus vaccine phase, the numbers of HIV-specific CD4+ and CD8+ T cells are expected to increase following immunization, while CD4+ T-cell counts and virus loads are still controlled with HAART. During the post-HAART phase, the previously increased levels of HIV-specific CD4+ and CD8+ T cells are expected to persist and control the replication of HIV when HAART is discontinued, with the maintenance of CD4+ T-cell counts in the context of a controlled infection.

Despite these clear successes, numerous shortcomings are emerging. First, these potent treatments do not eradicate HIV, but, at best, result in long-term containment of infection below the level of detection³. At present, it is estimated that quiescent reservoirs of HIV can persist in the host for at least 50 years, indicating the need for lifelong therapy³. Second, despite progress in providing access to HAART in developing countries, the cost of these lifelong therapies means that they are available to less than 2% of the 42 million individuals worldwide that are presently infected with HIV. Although the 'Accelerated Access' initiative by the United Nations Programme on HIV and AIDS (UNAIDS), together with drug companies, has helped to reduce the cost of HAART in developing countries by tenfold, at 1,300–1,500 US dollars per year the cost of anti-retroviral treatment is still far above the average annual income and is available to only one or two per cent of infected individuals in these countries. Even with the reduced cost of generic drugs, which is estimated to be as low as 300 US dollars per year, the cost far exceeds the resources that are allocated to health care in most countries. In Africa, where the average annual income is below 1,000 US dollars, treatment remains unaffordable (Table 1). Third, difficulties with persistent adherence to lifelong therapy limit the efficacy of HAART. Low compliance and unscheduled gaps in treatment allow for bursts of HIV replication and result in the selection of virus mutations that confer drug resistance⁴. Treatment failures owing to drug resistance are now one of the biggest problems for the clinical management of HIV infection.

Table 1 | Annual cost of anti-retroviral therapies

Finally, the long-term administration of HAART is challenged as a result of the side effects that occur in 40–60% of patients^{5, 6}. These side effects include mitochondrial toxicity resulting in neuropathy and myo pathy, severe LIPODYSTROPHIES, and the accumulation of triglycerides and cholesterol that leads to an increased frequency of MYOCARDIAL INFARCTION. In the absence of measures to prevent these complications, the new worldwide therapeutic guidelines⁷ recommend that the initiation of therapy is delayed.

Despite the continuous development of new drugs — including new drug families such as inhibitors of virus entry — there is little hope that a new therapeutic breakthrough that can block the replication of HIV without producing virus resistance or cumulative side effects will emerge soon. So, there is no question that additional therapeutic options are required.

Immune augmentation

Enhancing the host immune response to HIV might be an alternative strategy for controlling HIV. Indeed, adaptive cell-mediated immunity, together with the as yet poorly understood innate immunity, contributes to the control of chronic virus infection, and HIV seems to be no exception. Studies in both animal models of AIDS and HIV-infected humans indicate that HIV-specific CD4⁺ T helper (T_H) cells and CD8⁺ cytotoxic T lymphocytes (CTLs) can contribute to this control^8-11 (Fig. 1). The establishment of a virus SET POINT after primary infection is thought to indicate the speed and potency of the immune response. Indeed, HIV-specific CD4⁺ and CD8⁺ T cells help to reduce the virus load to a quasi-equilibrium level, which is typically 10,000 fold lower than the average 10 million virus particles per ml of plasma present during acute infection. By contrast, HIV-specific neutralizing antibodies seem to occur too late during infection to have a decisive role in this early control of HIV¹², although their contribution has not yet been fully explored. Once established, the quasi-equilibrium in virus production is mainly controlled by T cells, although this is not complete. This partial immune selection pressure in the face of persisting replication of virus allows for the gradual development of immune-escape mutations and 'immune resistance'. After years of this permanent 'predator–prey' relationship, T cells eventually become exhausted and fail to keep the virus at bay. There are exceptions however — less than 5% of HIV-infected individuals are known as long-term non-progressors. The immune responses in these individuals can maintain efficient control of the virus for more than 20 years in the absence of treatment^{13, 14}. The challenge is to induce this type of long-term effective immunity in individuals who do not achieve it on their own.

So, HIV-specific CD4⁺ and CD8⁺ T-cell responses can be almost as efficient as HAART during the primary infection, resulting in a reduction in virus load by several logs, and can control the replication of HIV for years. However, they cannot eradicate the virus and they exert pressure on the virus that drives the selection of variants that can escape immune control.

Restoration of the immune response

Successful treatment with HAART leads to the restoration of effective immunity to viral pathogens¹ that is potent enough to allow discontinuation of prophylaxis¹⁵. Can HAART alone help to restore the immune response to HIV? There is no question that immediate treatment with HAART during acute HIV infection leads to increased HIV-specific CD4⁺ T_H-cell responses^16-18, probably by limiting the infection of these activated cells¹⁹. Unfortunately, administration of HAART alone during the chronic phase of infection does not readily restore or even preserve HIV-specific immunity^{20, 21}. Instead, these specific defences usually wane with HAART — a phenomenon that is thought to indicate the antigen-driven homeostasis of pathogen-specific immune responses. Indeed, HAART limits the production of HIV antigens to a threshold below that required to stimulate HIV-specific effector T cells or to stimulate HIV-specific naive cells. Anergy or tolerance are not to blame however, as immune responses to HIV can be restored when the immune system is re-exposed to HIV^{16, 22-24}.

On the basis of these observations, an approach known as structured treatment interruptions (STIs), which is best explained as a type of AUTO-VACCINATION, has been developed. The hope was that brief and regulated exposure to the virus with which an individual is infected in the absence of therapy might lead to immune augmentation and enhanced control of virus. Indeed, marked increases in HIV-specific immunity are detected following STIs in individuals that are treated during the acute phase of infection^{16, 25, 26}. Although transient control of HIV has been achieved in small uncontrolled trials of treatment interruption in these patients, the durability of this control remains questionable^{28, 29}. The results of STIs in individuals with chronic HIV infection have been more disappointing. Rebounds in HIV-specific CD4⁺ and CD8⁺ T-cell numbers, similar to those seen in studies of acute-phase infection, usually follow relapses of virus replication, but they are too weak and transient to ensure control of the virus in the absence of treatment^22-24.

Should such failures be blamed on the immune system itself or the auto-vaccination procedure? Re-exposing a heavily pre-immunized individual to the same virus antigens favours the growth of memory CTL clones²¹, thereby limiting the diversification of the specific repertoire, according to the theory of ANTIGENIC SIN. Exposing newly activated antigen-specific CD4⁺ cells to live virus rapidly impairs their function in helping to induce the differentiation and clonal expansion of naive virus-specific CTLs^{19, 22}.

An alternative strategy, known as therapeutic immunization, aims to boost all components of the immune response to HIV when the virus is still under the pressure of HAART, so that an efficient immune barrier, which involves strong and durable CD4⁺ T-cell help, and strong and diverse CTLs (and perhaps antibodies) arises against the virus before exposure to the live virus, rather than after²⁷ (Fig. 1). Several animal models support the rationale for therapeutic immunization by showing partial control of viraemia after anti-retroviral treatment and therapeutic immunization during primary²⁸ or established²⁹ infection with simian immunodeficiency virus (SIV).

Questions for therapeutic vaccines

The new challenge is, therefore, to augment immunity to HIV while patients are still on therapy, so that even after long periods off therapy, the balance between HIV replication and the host immune response can be maintained. Such a strategy, however, raises several important questions.

First, will a vaccine be as or more effective at augmenting immunity than live pathogenic HIV, and which vaccines are available? A general consensus supports the use of vaccines that can restimulate broadly directed HIV-specific CD4⁺ and CD8⁺ T cells, although, again, little attention has been paid to the role of neutralizing antibodies. Although effective prophylactic vaccines are not yet available, there are several candidates in the pipeline that deserve parallel consideration as therapeutic vaccines.

Recombinant virus vectors alone, or in combination with polynucleotide vaccines³⁰ or peptides, might be the best choice to achieve this goal. Results from the first clinical trials are becoming available. Therapeutic immunization with an HIV–recombinant canarypox virus vector has proven to be immunogenic in acutely infected patients that are treated with HAART³¹. Two recent phase II clinical trials of chronically infected individuals immunized with the HIV–recombinant canarypox vector, either alone³² or in combination with lipopeptides and interleukin-2 (IL-2)³³, were both associated with a marked, although modest, extension of the treatment-discontinuation period. Even more encouraging was the ability of this recombinant vector to broaden the HIV-specific CD8⁺ T-cell repertoire, with cross-recognition of vaccine and autologous virus sequences³⁴, and the marked asso ciation between the generation of CD4⁺ T_H1-cell responses and the delay in restarting therapy³².

Inactivated viruses and proteins that induce antibodies specific for HIV and the CD4⁺ T_H1 cells that are required for the generation of CTL responses might be of interest. An inactivated HIV vaccine was, indeed, able to augment virus-specific T_H-cell responses, but whether it confers protection has not been tested³⁵. Combinations of recombinant virus vectors and these inactivated viruses might provide even more potent regimens, and they are presently being examined in an on-going clinical trial in acutely infected patients that are treated with HAART³⁶. An alternative strategy, which has been recently illustrated using an SIV model³⁷, involves the use of autologous dendritic cells that are pulsed with chemically inactivated HIV. Converting this expensive strategy to allow for the treatment of millions of infected individuals is not yet feasible. Nevertheless, important proofs of concept can be established from these studies, and dissection of the effective components of immunity is likely to promote research into the identification of alternative immune stimulators, other than dendritic cells, and more cost-effective strategies that might be amenable for global use³⁸. Finally, other adjuvants or cytokines have still to be evaluated. Several clinical trials are presently being carried out that test the use of recombinant virus vectors and IL-2 (Ref. 33) as a therapeutic vaccine strategy.

Second, when an efficient vaccine is found, new questions will arise, as schedules of immunization and safety concerns might differ for therapeutic and prophylactic vaccine approaches. Indeed, high-dose tolerance that is induced by hyperimmunization of individuals pre-immunized by HIV has to be avoided, and a therapeutic approach might not require as many immunizations as the prophylactic vaccination of seronegative naive individuals. The safety concerns raised by the use of a live attenuated virus vector or a growth factor might also differ when the risk of immune deficiency exists. The use of whole virus as an immunogen is also likely to be approved for infected persons long before it will be tested as a prophylactic strategy.

Third, the goal and end points to be reached by therapeutic vaccination have to be clearly defined. Is our aim to delay the initiation of therapy? Or to prolong the time spent off therapy? Or to reduce the number of anti-retroviral drugs that are required to control virus replication? Even when an ideal, safe and immunogenic candidate vaccine is found, we will still need to establish optimal designs and appropriate end points for clinical trials. Given the rapidity and severity of the CD4⁺ T-cell depletion that is usually observed during STIs in chronic infection, are there ethical issues in proposing the use of a placebo in these trials? End points also vary with time and with the changing guidelines for anti-retroviral therapy. Indeed, the clinical trials for therapeutic vaccines that were designed in 1999 had aimed to maintain the virus load below 10,000–30,000 copies per ml of plasma as a primary end point, but these criteria have changed in the years since then¹⁰. New designs for clinical trials will consider the time it takes to reach a CD4⁺ T-cell count of 250–300 per mm³ after discontinuation of therapy, which is the lower limit for the initiation of anti-retroviral therapy according to most of the present guidelines⁷ (Fig. 2).

		Figure 2 |  Typical design for a clinical trial of therapeutic immunization against HIV.  Patients included in these trials are typically treated for at least a year, either during primary or chronic infection, and have normal or subnormal CD4+ T-cell counts (above 400 cells per mm3) and undetectable plasma virus load. Three phases are considered. During the immunization phase, vaccines are administered when patients are still on highly active anti-retroviral therapy (HAART). During the interruption phase, the treatment is discontinued, usually a month after the last vaccine injection, and careful monitoring of the plasma virus load and the CD4+ T-cell counts is carried out. Anti-retroviral therapy is resumed if consensus criteria for therapy are reached. According to the 2002 guidelines, treatment is re-administered if CD4+ T-cell counts fall below 300 cells per mm3 or virus load rises above 100,000 HIV RNA copies per ml. The primary efficacy end point in these trials is typically the duration of the period off therapy.

Finally, important questions that are fundamental to the concept of therapeutic immunization have to be thoroughly evaluated. Does restoring cellular or humoral immune responses to which the virus has already escaped make any sense? Is durable immune augmentation possible during the chronic phase in HIV-infected individuals who have a history of severe immune deficiency, even if it can be improved with HAART? Rapid answers to these questions are crucial given the hopes raised by this exciting new field of investigation and its consequences for the clinical management of the disease.

International efforts

There is no question that therapeutic immunization deserves intense investigation as a potential option for the treatment of HIV infection. The questions outlined above will not be readily solved without an intense and collective international effort to explore simultaneously the different strategies that have been proposed.

Rapid evaluation of many vaccine approaches in phase I and II clinical trials requires interdisciplinary teams of experienced immunologists, vaccinologists, virologists, clinical-trial experts and clinicians. Indeed, to reach the goal of effective immuno therapy for HIV-1 infection will require the expertise of many disciplines. The traditional boundaries between basic science and clinical medicine have begun to break down with the introduction of HAART, and many investigative possibilities have opened up. In the era of therapeutic vaccines, vaccinologists will have to become increasingly familiar with the questions that are raised by the clinical management of AIDS. A network of top-quality scientists and clinicians will bring the complementarity that is required for the development of these new strategies. In addition, such a network provides a way to exploit rapidly the main research investments in various vaccine approaches. Indeed, interaction with experienced clinical sites should allow larger clinical trials and a greater number of them to be carried out, thereby providing better opportunities for the evaluation of vaccine efficacy. Collaboration between immunology laboratories with experience in high-tech immunology research tools will facilitate the definition of standards and augment the evaluation of various vaccine strategies. Periodic 'think-tank' meetings are also required to help this community answer the questions raised by vaccine initiatives.

How can such networks be established and sponsored? A marked effort has already been made by several national research agencies to build interdisciplinary networks that can evaluate therapeutic vaccination programmes: the ANRS (National Agency for AIDS research) in France, ACTG (AIDS Clinical Trials Group) in the USA, and several other countries have already set up networks for phase I and II trials. However, these efforts are generally limited to certain immunogens. With the emergence of potential new vaccines, intensified efforts are required. Moreover, the cost and technical expertise that is required to monitor the efficacy of these therapeutic strategies is an important obstacle. Although these national agencies are unique in providing the investment for infrastructures that are relevant to clinical trials, a global approach will require an international effort and the participation of other networks.

The international community has already set up, with the World Health Organization (WHO), united efforts to help organize phase III trials for various vaccine strategies. Therapeutic vaccination against HIV might require smaller scale, high-tech interdisciplinary teams. The European Union (EU) provides unique government facilities to build supra-national interdisciplinary networks between academics and industries. As an example, Theravac is an EU-funded programme that will test the safety, immunogenicity and efficacy of new, highly attenuated, recombinant pox-viruses with French, Dutch, German and Swiss teams of scientists and clinicians.

Combined efforts from the public and private sectors are working to accomplish this goal, such as CANVAC (Canadian Network for Vaccines and Immunotherapeutics) in Canada, which is developing vaccines and clinical trials, or the International Forum for Collaborative HIV Research, which is organizing international think-tank meetings that incorporate various national regulatory agencies, academics, government departments and industries to help solve the questions of vaccine safety, trial design and end points. Large pharmaceutical companies that have already invested in HIV vaccines are increasing their investments in this direction. The QUEST study³⁶ is an example of an exceptional effort, in which one large pharmaceutical company is sponsoring an international team composed of clinicians, virologists and immunologists to evaluate a large, multicentre, international phase II trial of a therapeutic-immunization strategy that combines the recombinant canarypox virus and an inactivated HIV vaccine provided by two distinct companies.

Although highly encouraging, present efforts are too limited to allow for the systematic exploration of all the approaches that deserve testing. Private sponsorship should also provide additional support for these efforts. The main private sponsor IAVI (International AIDS Vaccine Initiative) has already proven to be highly efficient at funding the development of new prophylactic vaccines, and clinical and laboratory facilities, as well as clinical trials³⁹. Similar international efforts have to be incorporated in the field of therapeutic vaccines, with new financing strategies to help accelerate this research.

ORVACS and therapeutic vaccines

ORVACS (Objectif Recherche Vaccins SIDA) is an example of this kind of international collaborative effort. It is a not-for-profit organization, which aims to accelerate research on therapeutic vaccines and immune-based therapeutic strategies against HIV (Box 1). ORVACS was created in 2001 with the support of Liliane Bettencourt — the main share-holder and daughter of the founder of l'Oréal, the giant cosmetics company — and the Bettencourt–Schueller Foundation in Paris. ORVACS has assembled an international network of leading researchers with extensive experience in immunology, vaccinology, anti-viral drug development and clinical trials in HIV infection. So, ORVACS brings to the field of therapeutic vaccines the research complementarity that is required to develop clinical trials for the most promising vaccine approaches seen in the pre-clinical arena.

Interdisciplinary teams have been brought together from three immunology laboratories with a long history of research in T-cell immunity to HIV and vaccines, from France (B.A. and P.D.), the United Kingdom (A. McMichael) and the United States (B.W.), and seven clinical centres with experience in anti-retroviral therapeutic trials, from France (C.K.), the United Kingdom (M. Youle), Germany (S. Staszewsky), Spain (J. Gatell and B. Clotet) and the United States (R. Murphy), and include the largest and most respected clinical centres that concentrate on AIDS research. By creating a critical mass of top-quality researchers and clinicians, establishing a network of laboratories that are able to carry out standardized, quality controlled immunology and providing funding for clinical trials of immune-based therapies, ORVACS is able to integrate experts in HIV research with large vaccine companies or small biotech companies with innovative HIV vaccine candidates. By offering a unique platform of large and experienced clinical centres and laboratories, ORVACS will ensure that the quality of clinical trials required by regulatory and registration agencies is met. By directly funding the clinical trials that are proposed by this scientific network, ORVACS provides a unique opportunity to accelerate this research and bring candidate vaccines to the clinical programmes. Its independent status allows ORVACS to study different vaccine candidates that might belong to different pharmaceutical groups or biotech companies. Autonomy in financial resources allows the ORVACS network to set up new therapeutic projects rapidly that usually take months to be funded by conventional means. The ORVACS network is developing a scientific programme that aims to translate new vaccine concepts from pre-clinical to clinical studies.

So, the purpose of ORVACS is to complement existing national and international efforts in the field to increase research on new immunological concepts or products for therapeutic intervention against HIV. The private sponsorship that funds ORVACS (representing half the budget of the Bettencourt–Scueller Foundation) gives this relatively small network the rapidity and flexibility that is required to investigate the most promising vaccine approaches.

Conclusions

At a time when general access to an expensive lifelong treatment for HIV infection is not yet a reality, when virus eradication does not seem to be obtainable and when lifelong HAART is becoming too toxic, it is time to rethink the clinical management of HIV infection. There is no doubt that maximal virus suppression is the only way to restore immune functions. However, the ultimate goal of therapeutic strategies is now to restore an immune response to HIV that would help to control virus progression without anti-retroviral treatment, prolong the time 'off' therapy and decrease the toxicity and costs of anti-retroviral therapies for HIV-infected individuals throughout the world. Promising vaccine studies indicate that immune augmentation might be a realistic goal. Given the magnitude of the present AIDS problem and the need for durable solutions, all promising avenues must be pursued with the utmost haste. International networks that combine all research forces and sponsors are required to help solve these new issues. The ORVACS initiative provides an incentive to expand present efforts in therapeutic immunization, and to enhance the international partnerships that are required to achieve, as quickly as possible, effective new immune-based interventions for the durable control of HIV-related immune defects.

Boxes

		Box 1 | The aims of ORVACS
		Conception, testing and funding of therapeutic immunization and immune-based strategies against HIV. Definition of standardized methods to provide a top-quality international policy for the immunological evaluation of clinical trials of therapeutic immunization. Design and development of clinical trials exploring new strategies for therapeutic immunization, such as proof-of-concept phase I and/or II trials to evaluate new vaccines.

References

back to top	NATURE REVIEWS | IMMUNOLOGY
© 2003 Nature Publishing Group