Prospects for immune reconstitution in HIV-1 infection

Immunological studies during the natural course of HIV-1 disease have recently been facilitated by the use of powerful new technologies and animal models. This has enabled us to evaluate and compare cellular immune responses seen at seroconversion, in chronic disease, in late-stage disease, in clinical nonprogression and after treatment with highly active antiretroviral therapy (HAART) or after novel immunotherapy. It has become clear that the maintenance of strong cellular virus-specific immune responses is essential to control HIV-1 viraemia in infected persons, in seropositive recipients of immunotherapy and in seronegative vaccinees [1]. Fully functional CD8+ effector T lymphocytes with cytolytic potential are required to control viral replication in vivo. CD8+ T cells recognize MHC class I molecules loaded with peptides derived from viral proteins [2]. A cytotoxic T lymphocyte (CTL) is able to induce death of a virally infected target cell. CTL kill targets using three mechanisms [3,4]: Secretion of cytotoxic cytokines such as TNFα and IFNγ; Secretion of cytotoxic granules (perforins and granzymes) in a calcium dependent fashion; Interaction of Fas ligand on CTL with Fas (CD95) on target cells. CD8+ T cells are also able to suppress HIV-1 replication in CD4+ T cells in in vitro systems. The mechanism of this viral suppression is distinct from MHC class I restricted killing, as it is noncytotoxic and does not require cell-to-cell contact. An as yet unidentified CD8 antiviral soluble factor termed CAF is responsible for this phenomenon [5]. Furthermore anti-HIV-1 CTL, along with virus-specific CD4 helper T lymphocytes (HTL which are described below), release soluble mediators such as MIP-1α, MIP-1β and RANTES which bind to the HIV-1 coreceptor CCR5 and inhibit entry of viruses which utilize CCR5 [6]. Other soluble factors, which bind to the CXCR4 coreceptor are also produced and these inhibit the entry of CXCR4 utilizing viruses. Type-1 CD4+ regulatory HTL play a pivotal role in the maintenance of CD8+ T cell responses [7]. CD4+ T cell proliferation, activation and release of soluble mediators such as IL-2, require recognition of processed viral antigens in the context of MHC class II molecules [8,9]. The relationship between HTL and CTL is mediated by fully functional antigen presenting cells (APC) such as dendritic cells (DC) or macrophages acting as a ‘bridge’[10]. HTL precursors (HTLp) must recognize processed viral antigen in the context of MHC class II molecules in order to prime the APC for subsequent interactions with the CTLp which is then activated to become a fully functional effector cell (Fig. 1a). Hypothetical diagram of (a) the three-way cell–cell interaction between antigen-specific HTLs and CTLs utilizing APCs and (b) interference with these interactions by HIV-1 during infection. HTL, helper T lymphocyte; APC, antigen-presenting cell; CTL, cytotoxic T lymphocyte. Hypothetical diagram of (a) the three-way cell–cell interaction between antigen-specific HTLs and CTLs utilizing APCs and (b) interference with these interactions by HIV-1 during infection. HTL, helper T lymphocyte; APC, antigen-presenting cell; CTL, cytotoxic T lymphocyte. In most cases HIV-1 specific immune responses, such as those described above, are impaired during the course of chronic HIV-1 disease, and patients lack fully functional CD4+ and CD8+ HIV-1-specific T cells. CD4+ T lymphocytes are targets for HIV-1, even in the early phases of infection [11]. The major reservoir of HIV-1, responsible for long-term persistence of the virus, is the resting memory CD4 T cell, bearing integrated HIV-1 DNA [12]. Productively infected DCs and macrophages have also been observed, and numbers of DC decrease during the course of disease [13]. The extent to which DCs remain phenotypically and functionally intact remains controversial. Defects in overall cellular immune responses may certainly be attributed to infection of such important immune components (Fig. 1b). However infection of other cell types such as epithelial cells in the thymus, may destroy the thymic microenvironment and irrevocably interfere with T cell development in the thymus, thus diminishing numbers of naïve CD4+ T cells in the periphery [14,15]. It is also possible that lack of HIV-1-specific proliferative CD4+ T cell responses may be due to clonal anergy, clonal deletion of activated CD4+ T cells, or activation induced cell death (at time of high viraemia). CD8+ T cell responses decline as disease progresses [16]. In chronic disease, HIV-1 specific CD8 effector T cells appear to be arrested at a relatively early stage of maturation, failing to differentiate into functional effectors [17]. Differentiation and maturation of virus-specific CTL is of course dependent on signals provided by HIV-1-specific HTL and appropriate APC (Fig. 1). Recent studies indicate that immune responses which control viral replication are often present during acute HIV-1 infection, and that these cellular responses may be preserved by early treatment with HAART [18]. Such responses, including broad virus-specific CD8+ T cell responses, CD4+ T cell responses to a wide range of HIV-1 antigens, and intact APC, are also seen in a small percentage of untreated HIV-1 positive persons more than 10 years after infection – such individuals are termed long-term nonprogressors and show slow (or no) progression to AIDS in the absence of antiretroviral drugs [19]. However in most patients at later stages of disease it appears that there is noticeable lack of HIV-1-specific functionality, which is not reversed by HAART. Treatment of HIV-1 with HAART, resulting in an apparent arrest of viral replication and an increase in CD4+ T cell numbers, is nevertheless unable to eradicate virus from latent viral reservoirs [12,20]. HAART appears to increase APC numbers to an extent, but they are still below normal levels after six months of HAART (S. Patterson personal communication). In no chronically HIV-1 infected person thus far has virus been shown to be completely eradicated. Following HAART, fewer CD4+ T cells become infected with virus and numbers of both memory and naïve CD4 T cells in the periphery may increase. Such increases have been observed to allow some restoration of important immune responses to opportunistic pathogens [21–23]. Thus the clinical outlook for the patient becomes better. However HAART treatment does not appear to allow major reconstitution of impaired HIV-1-specific T cell responses, does not therefore result in long-term nonprogressor status or in viral eradication, and does not allow for long-term discontinuation of drug therapy. In point of fact HIV-1-specific CD8+ T cell responses often diminish after HAART is instigated – possibly due to lack of continual viral replication and consequent low levels of antigenic stimulation [24]. It should also be recognized that HAART very often has unacceptable side-effects, that drug resistant virus may emerge in a majority of patients, and that long-term therapy is anyway too costly for 90% of HIV-1 infected individuals who live in the developing world. Immunotherapy with recombinant soluble molecules such as cytokines and hormones, and therapeutic immunization with novel vaccines or with autologous virus, have the potential to enhance existing antiviral responses, to induce new specific responses and to steer immunity towards a desired phenotype, thus potentiating viral decay and eventual eradication of virus from latent reservoirs in the body. Such strategies, when used in the absence of HAART, have very little effect on viral suppression and do not prolong survival [25]. The limitations of immunotherapy in drug naïve HIV-1 infected individuals are as follows: Immunomodulation using recombinant cytokines to boost T cell activation may boost viral replication in infected cells, or worse may exhaust an already hyperactivated immune system. Treatment with whole-inactivated or recombinant HIV-1 antigens may merely introduce exogenous HIV-1 antigens, which are unlikely to have a beneficial effect on an immune system already flooded with HIV-1 antigenic stimulus. Activated CD4 T cells may become targets of high levels of circulating virus increasing the size of the viral reservoir. Thus it seems apparent that for the successful treatment of HIV-1 infection novel immune-based approaches should be complementary to HAART. In chronic HIV-1 disease most patients lack fully functional CD4+ and CD8+ T cells. However, although numbers are often diminished, cells of the correct phenotype have been observed. These cells appear to be anergic, since, although they appear to produce specific cytokines such as TNFα and IFNγ in response to HIV-1 antigens and peptides, they lack functional proliferative capacity, do not produce IL-2, and have little cytolytic capacity [26–28]. Thus it has been suggested that lack of responsiveness is not altogether due to loss of T cell numbers or to loss of antigen presenting cells – and indeed such defects occur when CD4 T cell and APC numbers are still relatively high. Instead it appears that the cells, although present, are unable to respond properly to HIV-1 antigens, and thus fail to control viraemia. Successful immunotherapy must therefore aim to: Reverse the anergic state of immune cells; boost HIV-1 specific immune responses; reduce virus to minimal levels and possibly allow total eradication of virus from reservoirs in tissues and cells of the body; delay disease progression; lower transmission rates from person to person; enable discontinuation of HAART; yield results which may inform the design of successful prophylactic vaccines. The advent of HAART has enabled novel immumodulatory strategies to be envisaged and these are discussed below. Studies in nonhuman primates and in man have shown that structured treatment interruption (STI) after early intervention with HAART may lead to the preservation and induction of useful antiviral responses [29,30]. Early suppression of virus appears to protect the activated CD4+ T cells from infection and deletion. Such early treatment may be sufficient in acute phases of infection, when anti-HIV-1 responses are preserved, and autoimmunization during STI may boost these responses. Preliminary results from STI in drug treated chronically HIV-1 infected individuals appears less successful, and does not allow long-term reconstitution of potentially useful anti-HIV-1 responses [31,32]. Immunization with suitable therapeutic vaccines may modulate the host immune system to effect more appropriate and efficient responses to HIV-1 with the aim of maintaining an indefinite asymptomatic phase and immune control of viral replication. The addition of cytokines such as IL-2 may serve to boost T cell numbers and function. The most extensively tested therapeutic immunogens are briefly described in Table 1. HIV-1 immunotherapeutic vaccines HIV-1 immunotherapeutic vaccines The inactivated gp120 depleted HIV-1 immunogen, RemuneTM. Phase I studies with Remune, administered in the absence of HAART, showed increased HIV-1-specific delayed-type hypersensitivity CD4 responses and antibody titres [33,34], and increases in type I cytokines and chemokines that were accompanied with increased HIV-1-specific in vitro proliferative responses [35]. A larger study in asymptomatic patients in Thailand, where HIV-1 is predominantly clade E, showed significant increases in CD4 T-cell counts, associated with increased HIV-1-specific immunogenicity [36]. In Phase II studies, Remune in conjunction with HAART has induced marked HTL and CTL responses [37,38]. A recent controversial paper reported that Remune in conjunction with HAART had little effect on clinical outcome in chronically infected individuals, although HIV-1-specific cellular responses were clearly induced [39]. In contrast, recent data from the Spanish Remune phase II study indicate that therapeutic immunization with Remune provides a better clinical outcome in terms of virologic suppression in patients on HAART compared to the placebo group [40]. Data from a phase III clinical endpoint study, clearly show beneficial effects of Remune on viral load, CD4+ T cell counts and HIV-1-specific HTL proliferation in a background of multiple drug therapies [41]. We are at present conducting a trial utilizing both Remune and IL-2 in conjunction with HAART, and preliminary data indicates that such immunotherapy can result in the transient induction of potentially protective immune responses in chronic HIV-1 disease (Hardy and Imami, personal communication). Effects on clinical outcome remain to be elucidated. Recently, Moss et al.[42] examined the adjuvant effect of synthetic CpG oligonucleotide immunostimulatory sequences of DNA in conjunction with Remune in animal models, and showed that simultaneous vaccination generated more potent responses. The HIV-1 p17-p24 virus-like particles (VLP), p24-VPL. Phase I and II studies with p24-VLP (a yeast organism vector expressing gag p17 and p24), have been only carried out in the absence of drug therapy or with monotherapy in chronically infected patients. This immunogen had no statistically significant effect on either CD4 count, viral load, or progression to disease, although in a minority of patients HIV-1 specific CTL responses were seen to increase [43–48]. Phase II studies of p24-VLP at two low doses in comparison to the adjuvant (alum) alone, showed transient but insignificant improvements in CD4 T cell counts [45]. Higher doses of this product induced transient proliferative responses to either p17 or p24 and significant CTL responses to gag proteins, however, there was no difference in HIV-1 RNA load between the placebo and immunization groups [46]. A double-blind, randomised, placebo-controlled, phase II study of p24-VLP with and without zidovudine monotherapy showed no significant differences between the treated groups in terms of CD4 and CD8 T cell counts, antibody responses to p24, lymphocyte proliferative responses to p24 or p17, or the T-cell activation marker CD38 [47]. Re-visiting this approach in the context of HAART could be worth while. The HIV-1 envelope-based immunogens. Therapeutic immunizations with envelope-based vaccines have been shown to induce both humoral and cellular immune responses. In initial studies chronically infected recipients of such immunogens did not receive HAART [49–52]. Both rgp120 and rgp160 based immunogens have been shown to induce some anti-HIV-1 CD4+ T cell responses [53], which did not translate into improved clinical outcome [54]. A well studied recombinant gp160 preparation, VaxSyn, which is a noninfectious subunit vaccine derived from the envelope protein of HIV-1IIIB, has been shown to be both safe and immunogenic [49,55,56]. In a three-year, randomised, placebo controlled, double blind study, where asymptomatic patients received either alum alone or VaxSyn, there were no significant differences between the treatment and placebo groups for all parameters tested [57]. In a different two-year study, patients were randomised to receive either VaxSyn in alum plus placebo drug, VaxSyn in alum plus ZDV or alum plus ZDV [58]. Although persistent lymphocyte proliferative responses to HIV-1 gp160 were detected in patients after immunization, VaxSyn conferred no advantage over ZDV monotherapy. A further five-year double blind placebo controlled trial of this immunogen demonstrated the induction of potent cellular immune responses directed towards gp120. No clinical utility of these responses was seen with no effects on CD4 T cell counts, HIV-1 viral load, time of initiation of antiviral therapy, incidence of opportunistic infection or death [59]. A larger trial, where > 800 patients received either the immunogen or placebo, plus antiretroviral drug therapy if required, showed modest effects on immune parameters, and on short-term clinical end-points, but these changes remained insignificant at the end of the study [60]. Canarypox vaccines. The recombinant Canarypox virus vector vCP1452 (ALVAC1452) expressing recombinant env, gag, pol and nef together with recombinant gp160, has been administered to recently infected individuals who were treated with HAART within 90 days of infection [61]. All subjects developed good immune responses to the immunogen. Six out of 14 patients had increased transient T helper cell responses and >50% showed some CD8 T cell responses. DNA vaccines. DNA constructs encoding env and rev, or nef, rev or tat, and regulatory HIV-1 have been used as therapeutic immunogens in small numbers of asymptomatic drug naïve patients and have been shown to induce some transient antigen-specific CTL responses and antibody responses No significant changes in CD4 T cell counts or viral load were observed. of DNA constructs encoding nef, rev or with HAART, induced HTL and CTL responses that but did by reduce viral load are new approaches HIV-1 DNA vaccines used in with recombinant vector vaccines for on the virus or The use of HAART with cytokines alone may enhance control of viral replication and T cell function. therapies are and phase III are in Table therapies phase III trial All immunotherapy doses were administered be when antigen-presenting cells; a natural a three a therapies phase III trial All immunotherapy doses were administered be when antigen-presenting cells; a natural a three a therapy. IL-2 is the most as and have been shown to be during the course of disease Although the mechanisms of in remain IL-2 T and cells and cells, a of other cytokines such as and chemokines which on and HTL, CTL, cells and studies with therapies were by however, of and of has lead to successful clinical increases in CD4 T cell numbers In addition to increasing both naïve and memory T cell numbers and immunotherapy together with HAART is associated with a in viral load than that seen with antiretroviral therapy alone in patients with CD4 T cell counts of randomised, controlled show no significant towards improved clinical outcome Recent studies using IL-2 therapy in conjunction with HAART show that in addition to an at increase in CD4 T cell numbers and a of viral small of viraemia occur which may induce and induction of HIV-1-specific T cell responses The use of HAART with IL-2 alone, described as successful in HIV-1 positive individuals with CD4 T cell counts has been useful at all stages of disease IL-2 in addition to HAART T cell numbers and or even viral replication as well as viral reservoirs T-cell activation in the absence of antigenic stimulation yield results remains to be seen strategies used in where were with DNA vaccine with or without show that control of viral replication after was only seen in that received both IL-2 and DNA vaccine phase III studies IL-2 plus HAART with HAART alone in patients in different disease should indicate changes which are induced be of clinical therapy. a increases the and of both and precursors by increasing the numbers and of both and such as and lymphocytes Furthermore maturation and of most MHC class II on APC macrophages and APC Phase have shown that of and to individuals with infection increases CD4 T cell counts and suppress viral thus potentially the of resistant of HIV-1 in such patients In addition the beneficial effect on APC numbers and appears to reduce the incidence of opportunistic Recent studies the effect of and HAART in individuals show that treated group decrease in viral load and a increase in CD4 T cell counts than HAART alone IL-2 therapy. In stage disease patients HAART, where IL-2 alone had little or no the addition of to IL-2 appears to be at both HIV-1-specific and proliferative HTL responses, and fully functional CD8+ effector T cells are also Such data the positive role such therapy could have in the treatment of patients with HIV-1 infection presenting with persistent opportunistic towards therapy. of HAART with IL-2 and most both T cells and APC resulting in useful anti-HIV-1 responses. therapy. is a produced by the components of the and which are essential for T cell and appears to play a major role in T-cell by survival of developing T lymphocytes but may also proliferation and of the Although no specific to HIV-1 are carried in animal it has been shown in addition to thymic increases T-cell therapy. is produced by activated macrophages and a important for HTL and CTL and cell function. during HIV-1 infection. of exogenous remain although it may have potential as a vaccine adjuvant Recent studies and a phase I trial that further of as an immune adjuvant with therapeutic immunization is therapy. has been as an with potential to be used as a therapy that could the immune system and are the immune and it both the and immune system. It increases of cells into both and and thymic by side-effects, and in loss and are seen in HAART treated HIV-1 positive patients. is for treatment of HIV-1 and of the and In addition to increased rates of T-cell development with gp120 in a increases antigen-specific and HTL proliferation, increases of CTL and T-cell towards a We are at present conducting a study of in conjunction with HAART, and preliminary data indicate that such therapeutic intervention results in maturation and of HIV-1-specific CD8 T cells. The of immunotherapy on APC numbers and are also This could be by the use of HAART with suitable IL-2 and which boost T cells and boost by autologous DC immunization may also be autologous in vitro DCs with HIV-1 induce transient cellular immune responses The extent to which can be phenotypically and functionally may on the of and on the of CD4 T cells to prime and immunotherapy with autologous and CTL or have been both with and without HAART, but were by in for the potential of such CTL is derived from in that in addition to viral load these HIV-1-specific T cells to the and become associated with CD4+ HTL immunotherapeutic where of soluble mediators in the context of HAART may these cells and target towards viral reservoirs are envisaged of such cells, utilizing antiviral constructs and Thus by or T cell some of the effects of the virus on T cell and APC and numbers, and by the virus load and potent may have generated a new for the use of therapeutic immunizations with or without other data from and some of which is described above, indicates that it may be possible at all stages of disease to steer the immune response towards the phenotype of the long-term and to enable the host to immune control of the virus to discontinuation of HAART. It is clear that HIV-1-specific CD8+ T cells with cytolytic potential to control on CD4 of these virus-specific CD8+ T cells with viral be due to the decline in antigenic but may also result from CD4 T cell It is essential to such anergy, in chronic treated HIV-1 disease, where (STI) does not The of antigenic using therapeutic immunization and numbers and of T cells and APCs may result in a more immune