Talk of HIV cure goes viral
Published Online: 15 March 2013|
Download the PDF here
Download the PDF here
Download the PDF here
The good news that a baby born with HIV has been functionally cured made international headlines recently, but the wider implications are hard to fathom. David Holmes reports.
On Monday, March 4, the world woke to the welcome news that doctors in the US state of Mississippi had made medical history after presiding over the first documented functional cure of a baby born with HIV.
The extraordinary outcome is the chance result of two decisions, by turns incisive and unorthodox, first by staff at the University of Mississippi Medical Centre (UMMC), then by the baby's mother. Presenting the case at the Conference on Retroviruses and Opportunistic Infections in Atlanta, Georgia, Johns Hopkins' virologist Deborah Persaud detailed how the baby's mother was unaware she was HIV-positive until the results of a routine test came back during the later stages of labour, by which time she was too close to delivery to risk using antiretrovirals. Given the substantial risk of mother-to-child transmission, the medical centre's paediatric HIV specialist Hannah Gay opted for an aggressive course of treatment: a three-drug cocktail of intravenous zidovudine, lamivudine, and nevirapine (replaced with lopinavir plus ritonavir at 7 days) at a therapeutic rather than a prophylactic dose and, crucially, given just 30 h after birth.
Tests on two separate blood samples taken before therapy was started confirmed that the baby was infected with HIV, and plasma viral load tests on samples taken at 7, 12, and 20 days were also all positive before the viral load became undetectable using standard assays at 29 days. The child continued on antiretrovirals until, for reasons not disclosed, the mother withdrew the then 18 month old child from therapy. When the mother and child resurfaced and the child was tested again at 24 months, there was unexpectedly still no sign of HIV.
At this stage clinicians at the UMMC contacted collaborating laboratories at the National Institutes of Allergy and Infectious Diseases and the University of California San Diego, who were able to bring ultra-sensitive tests to bear to find out whether any traces of HIV could be detected.
They found a single copy of HIV RNA in plasma, but no replication-competent HIV in 22 million co-cultured purified resting CD4 T cells. Testing did detect a reservoir of HIV DNA in peripheral blood mononuclear cells, but no evidence that the virus was replicating. So although all traces of HIV had not been eradicated (known as a sterilising cure), researchers concluded that the child could be considered functionally cured.
The only adult known to have been functionally cured of HIV is Timothy Ray Brown, widely known as the Berlin patient. Brown had been infected with HIV for over 10 years when, in 2007, he was given an allogeneic stem-cell transplant to treat leukaemia. The procedure replaced Brown's immune cells with cells from a donor with a rare mutation that rendered their, and subsequently Brown's, CD4 cells immune to HIV. After Brown's leukaemia entered remission and he stopped antiretroviral therapy, his HIV plasma viral load did not rebound, and he remains functionally cured.
Both cases have important implications in the search for a wider cure, according to Oxford University's John Frater. "The Berlin patient proves that with sophisticated, expensive (and even life-threatening) measures it is possible to eradicate HIV from a patient with a chronic infection-but this is proof of principle, and although not relevant to the general population, is vital evidence for researchers. The Mississippi child suggests that there may be something critical in the lifecycle of the virus that means that in the first few hours after transmission the infection hasn't fully taken hold and can be aborted using standard therapy", he says.
Speaking at a press conference Persaud said the timing of the treatment was decisive in establishing the functional cure, noting that "perhaps the initiation of very early antiretroviral therapy prevented the formation of the viral reservoirs in central memory CD4 T cells that are the barriers to cure, and really sets the stage for a paediatric cure agenda going forward". Persaud is now planning trials to test whether the approach can be effective in other high-risk newborns-but many questions need to be answered before broader conclusions can be made says Frater: "It is not clear what really happened. Will the virus come back eventually? Is this child simply an 'elite controller' who would have done well anyway? Would an adult with such acute HIV-1 infection respond in the same way as this neonate?" WHO said in a statement it would keep current recommendations in place advising the "early HIV testing of pregnant women and provision of ARVs to all HIV positive mothers, along with infant prophylaxis, to prevent HIV transmission to the infant".
(64K) Eye Of Science/Science Photo Library
Early treatment may prevent latent HIV infection developing in T cells
Researchers have known for some time that starting patients on antiretrovirals soon after infection can, in some instances, enable patients to maintain low viral loads after discontinuing therapy despite having a reservoir of latent disease. This was recently reported to be the case in the VISCONTI (virological and Immunological Studies in controllers after treatment interruption) cohort of 12 patients. But such early treatment is not possible in most cases, and only seems to confer an ability to naturally control the disease in a few patients.
Post-Treatment HIV-1 Controllers with a Long-Term Virological Remission after the Interruption of Early Initiated Antiretroviral Therapy ANRS VISCONTI Study - (03/15/13)
French Study Indicates Some Patients Can Control H.I.V. After Stopping Treatment, 'Functional Cure', Visconti Study - NY Times - (03/19/13)
Although antiretrovirals have revolutionised the control of HIV, problems with their cost, potential toxicity, drug resistance, and treatment fatigue, mean that a cure will probably need to be found in the long term. Most efforts have focused on ways to eliminate the reservoir of latent HIV that is able to evade antiretrovirals by hiding in cells. From a strategic perspective, this can be approached in four ways explains Frater: "exhaust (activate the proviral reservoir in conjunction with antiretroviral therapy), kill (target latently infected cells), silence (silence transcription from the proviral reservoir), or replace (engineer a new population of HIV resistant cells)". Of these, the exhaust and replace approaches have received most attention, and Jerome Zack, Director of the UCLA Center for AIDS Research, and his team are one of many pursuing the exhaust strategy. "The goal is to find agents that induce expression of the latent virus while the patient is on antiretrovirals to inhibit spread of new virus, which should either expose the cell to the immune system or make it targetable by virus-specific reagents such as immunotoxins", he explains.
In a Review in TLID (The Lancet Infectious Disease, see below), Stephen Kent and colleagues report that researchers have already shown that latent virus in patients on antiretroviral therapy can be activated with a single dose of vorinostat, an anticancer drug that inhibits the enzyme histone deacetylase. But the studies are still in the "early developmental stages", says Zack, and there are still fundamental questions about the basic biology of HIV that remain unanswered. "We still do not know all of the potential reservoirs of virus, and how to attack them", he says, "and we are also still learning about how HIV causes disease". For now, at least, a cure remains a hope rather than an expectation.
several NATAP writers covering CROI discussed this case in their reports, there were mixed opinions by various experts at CROI on what happened with the baby, it might have been PEP, post exposure prophylaxis, rather than cure, that perhaps HIV had not yet established itself in the reservoir:
CROI: CROI Report 2013: Observations on the "Cure Agenda" - By David Margolis MD, UNC Chapel Hill and the Collaboratory of AIDS Researchers for Eradication (CARE) (03/18/13)
CROI: Complications of HIV and its treatment, Impressions from the 20th CROI - Pablo Tebas, MD University of Pennsylvania - (03/20/13)
CROI: 20th Conference on Retroviruses and Opportunistic Infections: Eradication/Cure, Prevention, Novel Strategies for administration of PrEP (delivery alternatives) Antiretroviral Therapy/drug resistance/new drugs/Comorbidities - Eric S. Daar, M.D. Chief, Division of HIV Medicine Harbor-UCLA Medical Center Professor of Medicine David Geffen School of Medicine at UCLA - (03/17/13)
CROI: Revealing the Reservoir, and A Second Cure? - Ronald Swanstrom PhD, UNC Chapel Hill - (03/18/13)
Medicine and the Media
Hype and the HIV cure
BMJ 2013 published march 18
Some media reports exaggerated the significance of the recent case of a functional cure of a baby with HIV, but that's a result of publicising unpublished, non-peer reviewed research, says Margaret McCartney
Atul Gawande, the surgeon and author, was one of the first to react. "This is huge, stunning, world changing," he tweeted, linking to a story of a "startling development" in the New York Times titled "In a medical first, a baby with HIV is deemed cured."1 The article explained how a baby in the United States had been treated with antiretrovirals 30 hours after birth, and by 18 months of age the baby had no detectable viral load. The New York Times article explained that this demonstrated "proof of principle . . . if we can replicate this case."
It also reported that other experts would need "convincing that the baby had truly been infected" and that this may have been a case of prevention of transmission rather than cure, as the BMJ also pointed out.2 It described this development in the context of other knowledge, such as the now established prophylactic treatment of babies born to infected mothers, and it called on seven experts in the article for their views, two of whom were coauthors of the original paper. Although the article acknowledged potential for excitement, it was also clear that deep uncertainty existed.
The same couldn't be said for much of the UK media. Paper 48LB from the 20th Conference on Retroviruses and Opportunistic Infections, titled "Functional HIV cure after very early ART of an infected infant," was presented in Atlanta at 10 am on 4 March 2013.3 However, the press release was issued on 3 March,4 the same day the New York Times ran its report, before the abstract had been presented. The press release made it clear that researchers would "continue to follow the case" and that they thought that "further research is needed to understand whether the experience of the child can be replicated in clinical trials involving other HIV-exposed children."
The press release also hinted that the case was unusual because the mother had received no antenatal care or antiretrovirals when she gave birth prematurely; additionally, treatment was stopped when the child was 18 months old for "reasons that are unclear." The press release had the subheading, "Discovery provides clues for potentially eliminating HIV infection in other children." The uncertainties-the unusual nature of this case and that the findings are yet to be replicated-were not spelt out, but a quotation from one author did say that this was a "promising lead."
This important nuance seems to have been eroded in several news reports. The Guardian's headline was "US doctors cure child born with HIV." It continued, "Doctors in the US have made medical history by effectively curing a child born with HIV . . . the child has a normal life expectancy and is highly unlikely to be infectious to others, doctors believe." The researcher was described as "stunned" at this "extraordinary" outcome.5 The story ended with a note that patients should not stop taking antiretrovirals and that preventive treatments for pregnant, HIV positive women were of proved effectiveness. There was no reminder of the unpublished nature of the report and its lack of peer review or the lack of replication.
Sarah Boseley in the same newspaper the next day explained the limitations of the research: "Is this the big one? Have doctors stumbled across the cure for HIV? Unfortunately not. This is progress . . . but the implications for those already infected or even the still significant numbers of babies born with the virus in the developing world are sadly probably slight."6
The Daily Mail, meanwhile, asked, "Have we found a cure for HIV? Child born with virus is now free of infection after 'miraculous' treatment," and commented, "There is no guarantee that the baby will remain disease free, but early signs do look positive."7 The Telegraph hosted a video from a press conference in Atlanta, with one of the researchers, Deborah Persaud, saying that the work "sets the stage for a paediatric cure."8 However, the Telegraph also ran another video featuring the HIV researcher John Frater from Oxford University, who explained the research, its limitations and uncertainties, and the need for patients taking antiretrovirals to continue taking them.9
Jon Snow, on Channel 4 News, was right to ask straight away, "How sure are you? Has it been peer reviewed? Are you sustainably excited?" Another of the researchers answered that the development had not been published or peer reviewed but that "other researchers need to know."
Publicising conference presentations can have problems. The abstract had not been peer reviewed and was presented after the press release. The press release did not spell out the inherent uncertainties in the meaning of this case report. It was left to doctors not involved in the case, including many interviewed by the New York Times, and journalists to unpick the details and ask harder questions.
The risks of hype expand when unchecked enthusiasm seeks coverage before publication. Although research should be published without undue delay, generating press coverage is not always useful for the research community, patients, or citizens. Many comparisons in the press were made with the case of the "Berlin patient," a man with HIV who had a bone marrow transplantation in 2009 and was subsequently found to have been functionally cured of HIV. Yet this has not, so far, been repeated.9
The search for an HIV cure: tackling latent infection
The Lancet Infectious Diseases, Early Online Publication, 5 March 2013
Stephen J Kent, Jeanette C Reece, Janka Petravic, Alexey Martyushev, Marit Kramski, Robert De Rose, David A Cooper, Anthony D Kelleher,
Sean Emery, Paul U Cameron, Sharon R Lewin, Miles P Davenport
Strategies to eliminate infectious HIV that persists despite present treatments and with the potential to cure HIV infection are of great interest. One patient seems to have been cured of HIV infection after receiving a bone marrow transplant with cells resistant to the virus, although this strategy is not viable for large numbers of infected people. Several clinical trials are underway in which drugs are being used to activate cells that harbour latent HIV. In a recent study, investigators showed that activation of latent HIV infection in patients on antiretroviral therapy could be achieved with a single dose of vorinostat, a licensed anticancer drug that inhibits histone deacetylase. Although far from a cure, such studies provide some guidance towards the logical next steps for research. Clinical studies that use a longer duration of drug dosing, alternative agents, combination approaches, gene therapy, and immune-modulation approaches are all underway.
Combination antiretroviral therapy has had a pronounced effect on the course of HIV infection, turning it from an infection that is typically fatal within a period of years into a manageable chronic disease. However, from a practical, public health, and economic perspective, elimination of HIV-1 infection would be preferable to long-term suppression of viral replication with drugs. Cure will probably require reduction or elimination of the HIV-1 forms that persist in the presence of antiretroviral therapy. Several factors contribute to persistent HIV-1 infection, but a major reason is persistence of a replication-competent pool of virus in resting CD4 T cells.1, 2
The pool of latently infected cells is established very early after the initial infection with HIV-1, usually before diagnosis and antiretroviral therapy in routine practice.3 Very early initiation of antiretroviral therapy has been associated with a reduction in the number of latently infected cells, but virus rebounds when treatment is stopped.4-7 HIV in latently infected cells can remain dormant for the life of the cell and is not substantially affected by intensification of antiretroviral therapy.8-10 Attempts to eliminate these cells are a key theme of eradication strategies,2 with calls to invest heavily in innovative approaches.11 A cure for HIV is usually defined as either a sterilising cure (by which in theory all latent HIV DNA is eliminated) or a functional cure (wherein latent HIV persists but viraemia is very low or absent without the use of antiretroviral therapy). We focus on the potential for a sterilising cure.
Antiretroviral therapy-based strategies
HIV-1 integrates into the chromosomal DNA of infected cells as part of its lifecycle. A small proportion of infected cells remain transcriptionally silent but with fully infectious virus.12 These cells are thought to restart active HIV-1 replication when antiretroviral therapy is withdrawn, even after many years of treatment. How these cells persist for many years is not fully understood. The presence of HIV-1 DNA in very long-lived cells (particularly memory CD4 T cells) seems to be the main reason that this latent pool of proviral HIV-1 does not decay or evolve.13-15 Some of these cells might homoeostatically proliferate to maintain the size of the viral reservoir.16 Alternatively, even very low-level replication could result in infection of new cells and replenishment of the reservoir despite antiretroviral therapy.8 However, the absence of viral evolution or the development of drug resistance argues against continued replication in patients on antiretroviral therapy.
Treatment intensification studies have not shown reduction in the size of the HIV-1 reservoir or prevented the recrudescence of viraemia after withdrawal of treatment.10 One study8 in 2010, however, showed that intensification of successful antiretroviral therapy with the integrase inhibitor raltegravir resulted in a subset of patients having increased unintegrated forms of HIV-1 DNA.8 Although controversial,17 these results imply that the integrase-inhibitor intensification might have slightly reduced replication. Complete blockage of new rounds of infection is likely to be an important starting point for future studies aimed at reducing the size of the latent reservoir.
The Berlin patient and bone marrow transplantation
One person is widely believed to have been cured of HIV-1 infection.18-20 Although anecdotal, the case is instructive and provides hope that other, safer strategies could be developed in the future. The patient, a 40-year-old HIV-positive man living in Berlin, developed acute myeloid leukaemia and received a bone-marrow transplant from a donor with a homozygous mutation (∼32) in the gene encoding C-C chemokine receptor 5, CCR5, which renders the donor cells highly resistant to infection from most HIV-1 strains (figure 1). He stopped antiretroviral therapy and detectable HIV viraemia did not return. Intensive efforts to detect residual HIV-1 from several tissues generally resulted in undetectable or barely detectable HIV-1 DNA or RNA.18
According to a report21 from 2012, two HIV-positive patients with lymphoma who were on antiretroviral therapy received allogeneic bone marrow transplants (but from donors who were not resistant to HIV), and HIV-1 was no longer detectable in either patient. Although these patients have not interrupted their antiretroviral therapy and cure is unconfirmed, these additional cases raise the possibility that factors other than HIV- resistance conferred by the transplanted bone marrow could have cured the Berlin patient. Factors that could contribute to the elimination of HIV latency after bone marrow transplantation include graft-versus-host disease, the use of immunosuppressive drugs, and a possible effect of the hosts being heterozygous for the CCR5 mutation. Although bone marrow transplantation is not a method that could be implemented widely to cure HIV, investigation of the most relevant factors that reduce or eliminate HIV latency in this setting could lead to safer and more suitable alternatives.
Strategies to purge latently infected cells
Several interventions in patients who are on stable, long-term antiretroviral therapy have been studied for their effect on the latent viral reservoir. Interleukin 2 activates CD4 T cells harbouring proviral HIV-1 DNA that, in theory, could be cleared with potent combination antiretroviral therapy.22 Interleukin-2 therapy showed promise in reducing the latent HIV reservoir in small studies,22 but in larger clinical trials did not reduce the latent virus reservoir.23 The anti-T-cell antibody muromonab-CD3 (also known as OKT3) also causes direct T-cell activation, but like interleukin 2 did not reduce HIV latency or improve clinical outcomes. Both agents were associated with substantial toxic effects.24, 25 Interleukin 7 can reactivate HIV-1 in vitro,26 although clinical trials of the molecule in patients on antiretroviral therapy have not shown any reduction in the size of the latent HIV-1 reservoir.27 Several groups are studying the ability of other agents (particularly histone deacetylase inhibitors) to activate latent HIV-1 transcription from resting CD4 T cells, and clinical trials are in progress to assess these agents.28
Identification of appropriate strategies for viral reactivation has proceeded in two ways. First, new drug discovery has been aimed at identification of agents that through minimum signalling can activate virus expression from latently infected cells without cellular or immune activation. This approach depends on appropriate models for latency. The gold standard for cellular target remains cells that were latently infected in vivo. However, this approach is not always practical and screening assays that use resting primary CD4 T cells infected in vitro have also been used.29, 30 Latently infected cell lines, although useful for defining the mechanisms that control latency,31 are oligoclonal and poorly mimic the variation in viral integration sites seen in vivo and in primary cell models.32 Until comparisons of in-vivo and in-vitro effects are available through clinical trials, the relative usefulness of drug screening assays for inhibition of HIV-1 latency will remain unclear. Despite these limitations, candidate agents have been identified, including those that act on NF-κB pathways of DNA transcription (eg, prostratin), epigenetic modifiers (eg, histone deacetylase inhibitors), inhibitors of DNA methylation, and agents (eg, disulfiram) that target the protein kinase AKT, which is involved in cell proliferation.33 A large array of small molecules that inhibit HIV-1 latency in at least some in-vitro models is now available.34
The second strategy is to move rapidly to clinical trials with existing, licensed therapeutic drugs that have the ability to reactivate the virus. Patients whose infections are suppressed with antiretroviral therapy are currently being enrolled in early clinical trials for treatment with viral activators. Preliminary results of one such study are promising,35 and further research should be done to define the appropriate doses and durations needed to achieve effective clearance of the latent reservoir. The histone deacetylase inhibitor valproic acid was one of the first drugs used, but it is a weak inhibitor of histone deacetylase at doses achievable in the clinic.36 Valproic acid showed promise in an initial investigation37 in which it was used together with intensified antiretroviral therapy. However, subsequent studies showed that valproic acid had no effect on the reservoir, emphasising that the relation between in-vitro and in-vivo activity of latency-activating agents is as yet unclear (table).35,37-41 Potentially more potent histone deacetylase inhibitors, such as vorinostat, have recently been assessed in clinical trials in the USA with a single dose35 and in Australia with a regimen of 400 mg per day (the licensed daily dose of the drug when used for the treatment of cutaneous T-cell lymphoma) for 14 days (NCT01365065). The histone deacetylase inhibitors panobinostat and romidepsin are both more potent than vorinostat in the activation of latent HIV-1 in vitro, and a clinical trial of panobinostat has started in Denmark (NCT01680094).
Inhibitors of DNA methylation such as decitabine are used in cancer chemotherapy and can modify HIV-1 expression,31 but are not yet in clinical trials for HIV-1. Disulfiram, a drug long used to treat alcoholism because of its inhibition of acetaldehyde dehydrogenase, might inhibit DNA methyltransferase and activate latent HIV.42 It is currently being assessed in a clinical trial with a regimen of 500 mg per day for 14 days (NCT01286259). Recent in-vitro work has shown that disulfiram activates latent HIV-1 via activation of the protein kinase AKT.33 Prostratin activates the NF-κB pathway and latent HIV-1 in vitro, but because of its serious toxic effects cannot be assessed in clinical trials. Development of activators of protein kinase C such as bryostatin or bryostatin analogues could be an alternative approach.43 Synergistic effects between histone deacetylase inhibitors, methylation inhibitors, histone methyltransferase inhibitors, activators of NF-κB, and interleukin 7 have all been shown in vitro,30, 44 but such combinations remain unexplored in clinical studies so far.
Gene therapy approaches
The curing of the Berlin patient suggests that clearance of HIV-1 can be achieved through repopulation with CCR5-deficient haemopoietic cells, even in the presence of what seemed to be small amounts of virus using CXCR4 before bone marrow transplantation.18 This finding raises important questions about the relative contribution of the total body irradiation used during transplantation compared with the contribution of low CCR5 expression per se. The 2012 report21 of the clearance of HIV-1 DNA after allogeneic stem cell transplantation for lymphoma (in the absence of CCR5 mutations in the donor) raises the possibility that transplantation, potentially enhanced by graft-versus-host disease, could contribute to clearance of latently infected cells.
Two separate but related approaches for gene therapy are also being explored (figure 2). The first uses zinc-finger nucleases transfected into either autologous CD34-positive stem cells or expanded populations of CD4 T cells that are reinfused back into the patient.45 The zinc finger targets CCR5 and permanently modifies it or truncates it within the genome of the transduced cells, thereby providing a pool of cells permanently resistant to HIV-1 infection (figure 2). CCR5 depletion with zinc-finger nucleases that target CCR5 effectively reduces HIV-1 in mouse models.46 An initial clinical trial showed a possible reduction in viral rebound after infusion of CCR5-deficient cells and structured treatment interruption in six patients.47 Reduction of CCR5 expression was associated with control of viraemia-one patient who had the most substantial CCR5 depletion had undetectable plasma viraemia by week 12. Although an important first step, larger studies with more robust reductions in setpoint viral load in patients who are off antiretroviral therapy are needed. A second approach is to knock down translation of CCR5 by use of a short interfering RNA (siRNA) that is retrovirally transduced into autologous CD34-positive or CD4 T cells (figure 2); this approach has also shown promise in preclinical investigation.48 Phase 1 trials of this method are expected to start in 2013.
An alternative gene therapy-based approach is to enforce latency so that viral rebound does not occur if antiretroviral therapy is stopped. This outcome could be achieved by targeting the HIV-1 promoter with siRNAs or short hairpin RNAs that induce transcriptional gene silencing of the virus via induction of stable epigenetic changes in the integrated viral genome, particularly in the viral promoter or 5' long terminal repeat (figure 2).49 Sustained expression of these constructs in immune cells might allow a functional cure through the induction of long-term latency of the virus, resistant to reactivation by inflammatory, proliferative, and homoeostatic stimuli. Although this approach is encouraging in vitro, it is many years away from reaching the clinic.
Immune modulation and immune effector mechanisms
Immunity is also likely to have an important role in controlling HIV latency. Several studies have examined therapeutic vaccines for their ability to control HIV in the absence of antiretroviral therapy. Most studies into therapeutic vaccines have focused on cytotoxic T-lymphocyte responses by attempting to boost and broaden HIV-1-specific CD4 and CD8 T-cell responses, but with little success to date.50-57 In the context of virus reactivation by drugs such as histone deacetylase inhibitors, antiretroviral therapy blocks the infection of new cells after reactivation. However, little is known about the fate of the reactivated cells. Reactivated latently infected CD4 T cells might die as a result of viral cytopathic effect or elimination by the host immune responses, but without active elimination of the cell, a risk remains that the reactivated virus-expressing cells will return to latency.
Resting CD4 T cells latently infected with HIV-1 in vitro do not die after virus reactivation by the histone deacetylase inhibitor vorinostat,58, 59 which suggests that reactivation alone will not purge the viral latent reservoir. However, Shan and colleagues58 have shown that when HIV-specific cytotoxic T lymphocytes were first stimulated in vitro, they efficiently killed latently infected cells reactivated by the histone deacetylase inhibitor. Although data are limited to one in-vitro study, this finding adds credence to the idea of using therapeutic vaccines to activate cytotoxic T lymphocyte responses together with agents that reactivate latently infected cells. The induction of potent cytotoxic T lymphocyte responses in HIV-infected patients might, however, be difficult with present HIV vaccine strategies. The improvement of therapeutic vaccines for the induction of such responses is one possible approach that could be pursued.60, 61 For example, dendritic-cell-based vaccines have shown promise in this respect, although these can only be given in highly specialised centres.62 Conjugation of vaccines to anti-dendritic-cell antibodies to target the induction of potent cytotoxic T-lymphocyte responses is a simpler approach that showed promise in preclinical studies63 and is being investigated in a clinical trial (NCT01127464). The efficacy of live cytomegalovirus vector vaccines for the control of simian immunodeficiency virus (SIV) infection in macaques suggests a possible role for similar cytomegalovirus vectors as therapeutic vaccines for HIV-1 in future studies.64 Unfortunately, some of the most effective HIV-specific cytotoxic T lymphocytes are restricted by fairly uncommon HLA class I alleles (eg, HLA-B*27, HLA-B*57) and many existing responses in patients will have already forced viral escape early in the infection.65
An alternative, non-MHC-restricted immune response that might recognise and eliminate reactivated latently infected cells is antibody-dependent cellular cytotoxicity (ADCC). HIV-specific ADCC antibodies mediate the killing of infected cells by binding to viral antigens on the surface of infected cells and engaging innate immune cells such as natural killer cells or monocytes through their Fc receptors.66 However, little is known about the effect of long-term antiretroviral therapy on HIV-specific ADCC. HIV-specific ADCC antibody responses are often high in patients with very slowly progressing HIV infection.67, 68 Natural killer cell effectors are, however, decreased in number and can become dysfunctional in patients with progressive HIV disease, although results of recent studies suggest that function of natural killer cells is largely preserved, particularly in patients on effective antiretroviral therapy.69-77 We speculate that HIV-specific ADCC antibodies could potentially be induced to help to clear reactivated latently infected cells by use of protein therapeutic vaccines combined with potent adjuvants.
Another approach is to target cell-surface molecules that are highly expressed on cells that harbour latent HIV. The programmed cell death receptor PD-1 (also known as PDCD1) is highly expressed on so-called exhausted CD4 T cells that are resistant to activation and harbour latent HIV infection.78-80 The use of antibodies to inhibit PD-1 and its ligand has recently shown substantial activity in cancer trials81, 82 and some promise for the reduction of immune activation in macaques infected with SIV.83 Although such agents will have toxic effects, they could prove useful adjuncts to latency-eradication strategies.
Immune activation and latent HIV
Active HIV-1 infection is associated with increased immune activation. Antiretroviral therapy reduces immune activation, but not to normal. The residual immune activation is associated with residual long-term morbidity, despite antiretroviral therapy.84 Immune activation despite antiretroviral therapy might have a role in the promotion of low-level HIV replication and reseeding of the latent HIV-1 reservoir, although this idea remains controversial. The investigators of the study of raltegravir intensification of antiretroviral therapy noted that the increase in non-integrated HIV-1 forms after the intervention (which implies continued low-level replication) was largely confined to patients with increased immune activation despite antiretroviral therapy.8 This finding lends support to the idea that immune activation might drive continued viral replication in patients on antiretroviral therapy.
Drug treatments to reduce chronic inflammation are of interest for several diseases,85 and such treatments are being investigated for HIV-induced immune activation in several small studies. These studies have two goals: to improve long-term morbidity and to reduce the continuing low-level HIV-1 replication that occurs despite antiretroviral therapy, which would allow more efficient decay or clearance of latently infected cells. The gut-associated lymphoid tissues are damaged during HIV-1 infection and probably contribute to persistent immune activation. Drugs being investigated in clinical trials to reduce gut-damage associated immune activation and so potentially to reduce latent HIV-1 reservoirs include probiotics (NCT01439841),86 anti-lipopolysaccharide antibodies,87 the bowel anti-inflammatory agent mesalazine (NCT01090102), and the antibiotic rifaximin (NCT01466595). Research into treatment with agents such at valganciclovir to reduce herpesvirus-induced immune activation common in patients with HIV-1 is at the early proof-of-concept stage.88 Another approach being investigated in a clinical trial is the reduction of inflammation and thereby the HIV reservoir by use of the angiotensin-converting-enzyme inhibitor lisinopril (NCT01535235). Whether approaches to reduce immune activation in HIV-1 infection will reduce the latent reservoir of HIV DNA is unknown.
Timing of interventions to reduce HIV latency
The timing of when antiretroviral therapy is started after HIV-1 infection might affect the extent of latency during treatment. Patients with acute HIV-1 infection treated with antiretroviral drugs have fewer latently infected cells, and the extent of latent infection seems to be associated with the size of the viral load and duration of viraemia before antiretroviral therapy.7 Additionally, a small subset of patients given antiretroviral therapy during acute infection can control HIV well when treatment is stopped.4, 7 Although such individuals are difficult to identify for recruitment to clinical trials, they could represent an opportunity to explore more effective interventions against latency.
A second question is when to give interventions to reduce latently infected cells relative to the timing of antiretroviral therapy. Studies are currently done in patients who are on long-term antiretroviral therapy. This approach has the advantage of minimising the risk that any HIV-1 activation could reseed the latent reservoir. On the other hand, this approach might mean that the latent pool is targeted at a time when it is more resistant to activation. Indeed, the notion of a stable population of latently infected CD4 T cells has historically been defined in the context of long-term antiretroviral therapy. However, some researchers have questioned whether such a stable reservoir of HIV-1 DNA is also present during active infection,89 because of the high immune activation. If the latent HIV-1 DNA pool is much more labile during active infection, then this possibility presents a potential avenue for intervention.
We recently studied the turnover of viral DNA in resting CD4 T cells in active SIV infection of macaques.90 When viral loads were low, the turnover of the putative latent reservoir was very slow (many years), which is consistent with the low turnover of virus in patients with HIV-1 on treatment. However, in animals with high viral loads, the turnover of SIV DNA within resting CD4 T cells was fast (days), which suggests that high immune activation during active infection might prevent the establishment of true latency. This idea is supported by findings from a study in patients after the initiation of antiretroviral therapy that suggested rapid exchange between productively and latently infected cell pools during both chronic and acute infection.91 Additionally, studies in HIV-infected patients show a delay between the appearance of mutations in plasma and their appearance in the proviral DNA, both for immune-adaptive mutations92 and drug-resistance mutations.93-95
These findings suggest a new strategy to address the reduction or elimination of the viral reservoir. Since many agents currently under investigation aim to purge the latent reservoir through activation of latent cells, targeting the reservoir when it is already turning over rapidly during active infection might be easier to achieve than when it is stable under long-term antiretroviral therapy. The possible effects of giving a latency-activating drug at different times relative to antiretroviral therapy can be investigated by use of a simple model of HIV infection (figure 3).96, 97 The model predicts that the short-term treatment with latency activators would have the strongest effect on latent HIV-1 DNA if given immediately before commencement of antiretroviral therapy, when it enhances the already existing high activation rate of latent cells caused by the high viral load. Of course, the precise mechanisms of action of a putative latency-activating drug are uncertain, as is whether these would work synergistically with immune activation (as is assumed here). Moreover, whether the reduction in latent HIV would be sufficient to reliably yield a functional cure in the absence of antiretroviral therapy remains unknown, but studies aimed at the incremental reduction of the reservoir size should ultimately answer such questions.
The timing-of-cure approaches with respect to initiation of antiretroviral therapy could hold some promise, but they are not without their own risks. A very real potential exists for toxic effects from attempts to cure at the time of starting antiretroviral therapy, since cellular activation, the loss of a key subset of immune cells, and potentially further increases of HIV-1 replication could be transiently induced. Paradoxically, the results of our previous study90 and our model (figure 3) suggest that the patients who could benefit the most from such an intervention would be those with high levels of viral replication-ie, those who are likely to have a high turnover of HIV-1 DNA within resting CD4 T cells, making this reservoir more vulnerable.
Innovative strategies to reduce or eliminate latent HIV-1 infection are being pursued to reduce the global burden of HIV and costs and adverse effects associated with lifelong antiretroviral therapy. Several different strategies are in early stage clinical trials. Approaches that attempt to reduce latent HIV at the time that antiretroviral therapy is started should also be investigated.