International AIDS Society (IAS) Symposium "Towards an HIV Cure" July 20-21 2012 Wash DC
written for NATAP by David Margolis MD, University of North Carolina
"the French VISCONTI cohort.....examined patients who started ART within 10 weeks of HIV infection (average 40 days), and then controlled viremia after ART interruption, following several years of suppressive therapy.....Bacchus and Saez-Cirion suggested that at least 12 months of ART post-AHI would control viremia for at least 2 years in 5-15% of patients. In my opinion many of these patient appeared to be treated in early, not acute infection...... I think it remains to be demonstrated that this is a new phenomenon, how durable this control really is over longer periods of time, and that ART is increasing the frequency of elite control, rather than a chance observation.: see below
IAC: 2 New Studies at IAC Finding 'Functional Cure' - (08/03/12)
Public health: Towards a cure for HIV - (07/25/12)
Gilead HIV Eradication Program - (06/27/12)
Over two days prior to the opening of the IAS meeting in Washington DC last week, the IAS held a symposium to launching the global scientific strategy "Towards an HIV Cure". The order of the symposium was organized in the same way as the published global strategy, simultaneously published in Nature Immunology Reviews (August 2012). This effort has been spearheaded by Dr. Francois Barre-Sinoussi, incoming president of the IAS and Nobel laureate as co-discoverer of HIV-1. The effort to mobilize resources and attention to develop curative therapies for HIV infection will be a focus of her tenure as IAS president, and caps a planning process that she has led since IAS Vienna in 2010.
Closing Session - Psychosocial benefits of an HIV cure:
Before I review the opening of the meeting, and its contents, let me recount the meeting's closure. Fred Verdult of Volle Maan in the Netherlands performed a survey of HIV+ patients in his country. David Evans of Project Inform has recently performed a related internet survey in the US. Verdult asked HIV+ people to rank the reasons that they desire to have an HIV cure. The answers, I think, would not have been predicted by scientists and clinicians, but to me seemed to come from the burden of the stigma and uncertainty that plague many HIV+ people. Verdult then described different types of cures that might be achieved, from absolute HIV eradication with subsequent total freedom from infectiousness and long-term negative health effects (the most difficult to achieve) to a more cancer-survivor model where one was not fully assured of the duration of "cure" and other downstream benefits (a more likely near-term result). I cannot do the full presentation justice, I think, but the take-home message to me is that there are aspects of the state of being HIV-infected that scientists and clinicians often do not appreciate. What scientists and clinicians are talking about when we talk about cure may not be what patients are hearing, or what they want. Patients may be more willing to undergo research risks in some situations. However, patients may be much less interested in certain clinical goals than is assumed by scientists and clinicians, and so if the achieved "cure" does not bring them what they want or expect this must impact the risk-benefit calculation that makes clinical trials justifable. This is an investigation that deserves more attention and effort, and should inform the science that we plan, and the data that is discussed at the meeting. Perhaps next time, Mr. Verdult should present at the start of the meeting.
The meeting was opened by NIAID director Anthony Fauci, who painted a broad overview of the effort to end the AIDS pandemic via the worldwide implementation of a combination of approaches over the next decades. He outlined scaled-up efforts at diagnosing people with HIV infection by cheaper assays and bringing them in to medical care to receive earlier antiretroviral therapy (ART). Combination approaches to prevent HIV infection by employing interventions such as pre-exposure prophylaxis, circumcision, and improved microbicides in select populations were discussed. Here the key challenges are largely those of effective and affordable implementation of these public health strategies. The remaining scientific challenges, of course, were the development of effective protective vaccines, and of therapy to eradicate infection. In the area of cure, Fauci reviewed approaches employing ART with interleukin 2, initially studied over 10 years ago, and the one of the current strategies know as "shock and kill" or "kick and kill," wherein persistent, latent infection is perturbed with a drug that induces the expression of hidden, latent HIV and the desired effect is the clearance of infected cellular reservoirs. At the end of his talk, Fauci announced a new letter of agreement between the NIH and the Agence Nationale de Recherche sur le SIDA (ARNS) for collaboration on HIV cure research.
The next plenary was given by Bob Siliciano of Hopkins, who reviewed HIV latency, the state of HIV infection when the virus is silently integrated into the genome of rare, resting CD4+ T cells, and of recent results from studies in his laboratory. Siliciano pointed out that latent infection is favored in non-activated CD4 cells, as the silencing of HIV expression is enforced in cells that have low levels of cellular factors required for virus expression, such as NFkB.
To develop agents that might provide the kick needed for "kick and kill," Siliciano described a model system developed in his laboratory. Here, primary CD4 T cells from an HIV-uninfected donor are stably transduced with the human survival gene called Bcl-2. These cells can then grown in the lab for long periods of time. Cells can then be activated to allow infection with an HIV that also carries a marker gene, and then the activation signals withdrawn. Over time, surviving cells lapse into latency, and one is left with a pool of cells that contain latent viral genomes that will express a marker gene if the virus is awakened. This provides a system to perform screening of large numbers of compounds to discover ones able to reactivate latent HIV. Using this system, the lab found that a drug used to treat alcoholism, Antabuse or disulfiram, reactivated latent viral genomes. This drug is now undergoing clinical testing to see if it can reactivate latent virus in the patients for whom viremia is held in check by ART.
However, the kick-kill strategy rests on the hope that the few cells that contain latent, persistent virus will die or be cleared once the virus is re-induced. Siliciano reviewed a very influential paper published by his group (Shan et al., Immunity 2012). The sway of this work was evidenced by the fact that at least 4 other speakers at the meeting would later refer to this paper. Shan and co-workers showed that infected cells studied in the lab could be induced to express HIV in some situations, but that they would not reliably be killed in the process. This finding has reignited interest in approaches to kill infected cells as part of a curative strategy, either by upregulating the anti-HIV immune response with therapeutic vaccines, or devising ways to make infected cells selectively more susceptible to death following viral induction.
Finally, Siliciano discussed new and unpublished work from his laboratory. An important issue in future attempts to cure HIV infection is the measurement of the size of the problem. The HIV genome, HIV DNA, can easily be measured in the cells of almost all HIV-infected patients, even those on ART for many years. However, in such patients, latent HIV is very rare, found in (often much) less than 1 in a million resting CD4+ T cells. Therefore most of these HIV DNA molecules are defective, "junk" DNA that cannot make virus and are not a threat. The problem is that hidden in the pile of junk are potentially deadly, functional genomes. As we move towards eradication therapies, we will need to know how much progress we are making.
So the Siliciano lab performed standard viral outgrowth assay of resting CD4 T cells, growing millions of cells in cultures of 1 million cells per dish. From some culture wells, replication competent HIV was recovered. But the lab looked for HIV DNA in the wells where virus was not recovered, and found viral DNA in some of these. Through exhaustive amplification and sequencing techniques, the lab tried to explain what was wrong with the DNA viral genomes that did not produce virus in culture. Most of these genomes were defective due to multiple and/or extensive mutations. But about 20% of the genomes appeared to be functional, at least on the level of the genetic sequence. When a few of these genomes were reintroduced in to cells in a way that they could produce viral particles, these viruses appeared to grow reasonably well. The take-home message of the work is that while simple measures of HIV DNA in cells wildly overestimate the size of the true latent reservoir, the precise and exhaustive viral culture assays that attempt to measure the latent reservoir may underestimate its size. In my opinion is it not depressing to discover that the size of the latent reservoir may be two or even 5 times bigger than previously estimated, as in absolute numbers of infected cells, it is still pretty small. But if can we develop eradication therapies, we will need better tools to measure the size of the reservoir as we whittle it down. If we are successful, in the end we come to studies of treatment interruption as it is likely that there will always be some sort of HIV genetic material that can be detected in patients, as was recently reported with the only person likely cured so far, the Berlin patient Tim Brown.
Determine cellular and viral mechanisms that maintain HIV persistence:
Warner Greene of the UCSF Gladstone Institute then reviewed mechanisms by which HIV establishes, maintains and emerges from transcriptional latency in the various viral reservoirs within the infected individual. An understanding of these mechanisms is pivotal to the development of strategies aimed at eradication. He highlighted the various regulatory mechanisms, both cellular and viral, that exist within latently infected cells. Primarily, HIV latency occurs on the transcriptional level, that is to say at the level of expression of the integrated HIV DNA genetic code as viral RNA, which is then translated into HIV proteins, or packaged within the viral particle --- the HIV RNA that is detected in the blood and is called "viral load." Many, many molecular mechanisms contribute to HIV latency: a) the site within the human genome at which HIV integrates, b) the activity of genes near the site of integration, multiprotein complexes that bind the HIV promoter and allow expression such as one called PBAF, and countervailing complexes that inhibit expression such as BAF, c) marks on histone proteins - the scaffold of the human DNA genome that contributes to regulation of all genes-that can augment or inhibit HIV expression, and d) effects that reduce the availability of the critical viral Tat activator, either low levels in resting cells, sequestration of Tat in different Tat/HEXIM protein complexes, and recycling of Tat in transcriptional complexes. Finally Greene reviewed a variety of primary cell models of latency - such as the one just described by Siliciano - and noted that different drugs and signals alone and in combination that break latency appear to behave differently in different models. A better understanding of how best to use these models to screen candidate anti-latency therapies is an active area of study.
Abstract-driven presentations followed this plenary. Gray from the Burnet Institute in Melbourne presented studies of unique regulatory mechanisms of brain-derived HIV-1 genomes. Viral promoters, the part of the viral genome called the LTR that acts as the on-off switch for all viral gene expression, from brain viruses had lower activity in both T cells and brain astrocytes, suggesting that CNS-derived viruses have a reduced capacity to initiate viral transcription, and perhaps unique regulatory mechanisms within the CNS, ultimately affecting the fate of viral infection and the development of latency. Li from the NIH demonstrated the potential for infection of brain astrocytes by HIV by directly co-culturing astrocytes with HIV-infected lymphocytes. This novel mode of infection was CXCR4- and CD4-dependent, but DC sign- or α4ß7 integrin independent. This suggests that these cells might be an important reservoir in the brain, and more efficient infection via the cell-to-cell contact transmission might explain the propensity of perivascular infection of astrocytes seen in untreated animal models. Tyagi at George Mason presented work showing repressive chromatin structures at HIV LTR involving the Polycomb Group (PcG) corepressor complex. This work showed again that HIV latency consists of defects both in the initiation of viral gene expression and the elongation of viral transcripts, the process that creates complete, effect viral messages. Wide arrays of modifiers of chromatin structure and transcription complex composition contribute to these restrictions to HIV expression, as in other human genes.
Determine the tissue and cellular sources of persistent HIV in long-term ART-treated individuals:
Janice Clements of Hopkins then presented studies of HIV infection in the SIV model in macaques. Although there are very low levels of HIV infection in blood monocytes, when these cells differentiate into macrophages in tissue they become highly infectable. The level of tissue macrophage infection by HIV in patients on ART is unclear. Similarly, while microglia and astrocytes in brain can be infected, but the persistence in these cells on ART is unknown. This is an important question to be addressed by the SIV model of CNS disease. In the Clements model, animals are co-infected with a macrophage-tropic, neurovirulent SIV and a swarm of CD4-depleting SIV. Perivascular macrophage and microglia infection can be seen in 4 days, as can infection of CD68+ macrophages in spleen. SIV latency on ART has been shown in this macaque model. On ART, residual SIV RNA can be found in spleen and nodes, and to a lesser extent in other tissues. In situ hybridization for SIV RNA can directly detect rare SIV RNA in brain on ART. In a co-culture system of isolated microglia from treated animals with T cell line cells to amplify the virus, virus can be recovered from 1:5000 to 1:50,000 microglial cells, up to 1000-fold less than from untreated animals. Better ART for animal models, and longer treatment periods will be needed to more accurately model the state of treated humans.
Ambrose from Pittsburgh then reported that in the RT-SHIV model, an SIV virus containing HIV-1 reverse transcriptase, treatment for 17-20 weeks with 3 drugs (tenofovir, lamivudine, and efavirenz) or 4 drugs (tenofovir, lamivudine, efavirenz, and an integrase inhibitor) seemed very effective. The majority of virally infected cells were located in lymphoid tissues with variable levels in the gastrointestinal tract. Little viral RNA was evident in tissue after suppressive therapy with either 3 or 4 antiretroviral drugs, plasma viremia did correlate with DNA levels in tissue, and very few animals had PBMCs that were positive for SIV RNA. Kumar from the Lewin lab in Melbourne reviewed their work showing the myeloid dendritic cells facilitated the establishment of latent HIV infection in a cell culture system. Infection of resting CD4+ T-cells in the presence of myeloid (m)DC significantly increased latent infection of non-proliferating CD4+ T-cells compared to infection of T-cells cultured alone (p=0.0005, n=11). When mDC-T-cell contact was prevented using transwells the number of latently infected CD4+ T-cells was reduced (n=3).
Josefsson from the Palmer lab at Karolinska, then made a striking presentation of the genetic relatedness of persistent HIV-1 in a broad spectrum of CD4+ T cells isolated from peripheral blood and gut associated lymphoid tissue from patients on long-term suppressive therapy. Phylogenetic analysis revealed clear evidence against any substantial evolution between the pre-therapy plasma-derived HIV RNA sequences and on-therapy intracellular HIV DNA sequences. Numerous intracellular HIV sequences identified after long-term therapy contained replication-incompetent virus. As presented at CROI, one patient had a predominant intracellular clone in both memory and effector memory T-cells containing a 380 bp deletion after >9 years of therapy. HIV DNA sequences from peripheral blood cells, marrow, gut, and node were all highly related to pre-therapy RNA sequences. In striking slide after slide, she showed how the genetic trees from different cells and tissues precisely overlapped. As in other presentations, she showed that bone marrow progenitors do not have HIV sequences, and that marrow samples that do have HIV are uniformly contaminated with CD4+ cells.
Buzon from the Lichterfeld lab at the MGH/Ragon suggested that HIV may persist in T memory stem cells (Tscm), a recently-discovered subpopulation of T cells that persist for extremely long periods of time and are maintained by stem-cell like properties, such as self-renewal and homeostatic proliferation. It should be said that the properties of this small, newly reported population are somewhat controversial. Buzon found HIV DNA in the Tscm population of some patients, the true frequency and reasons for interpatient differences being unclear. In 8 patients treated for various periods of time with ART, the proportion of Tscm in which HIV DNA was detected was 17-24%. It was unclear if these sequences represented replication competent HIV, and longitudinal studies to prove durability of this potential reservoir are in progress.
Determine the origins of immune activation and inflammation in the presence of ART and their consequences for HIV persistence:
Douek from the NIH Vaccine Center gave an overview discussing the causes and consequences of immune activation in HIV infection and suggest novel therapeutic approaches to suppress the immune activation. CD38+ DR+ T cells that persist despite prolonged ART in some patients mark immune activation. Another marker of innate immune activation, soluble CD14 is a predictor of death, independent of viral load, in patients. Douek pointed to a weak association between cell-based measures of persistence (DNA) and immune activation, but an improved association CD38+/DR+ T cells were measured in GALT tissue.
This plenary was followed by several abstract presentations. Palmer from the Burnet Institute in Melbourne presented a novel hypothesis suggesting that increases in glucose metabolism may be linked to increased T cell activation, and that the effects of HIV infection and replication may trigger this increase. Glut1 is the major glucose transporter and a rate-limiting step in glycolysis in lymphocytes. Glut1 + cells increased in number during HIV infection, and were not fully normalized by ART. This increase in Glut1 expression correlates with CD4 cell loss. Glut1 may therefore be a novel functional marker of CD4 activation and metabolic activity, susceptibility to apoptosis, and an early prognostic marker for disease progression.
Dunham from the Gladstone at UCSF presented an update on the hypothesis that inflammation in the gut, resulting in intestinal damage, microbial translocation, and subsequent systemic inflammation, prevents recovery of peripheral CD4+ T-cells during ART. Systemic levels of interferon were implicated as a driver of inflammation and lack of CD4 recovery on ART. He showed further data on systemic levels of soluble CD154, thought to be shed from monocytes, and the enrichment of the GALT in patients with lack of immune reconstitution on ART with CD14+ inflammatory macrophages and CD123+ plasmacytoid dendritic cells. He suggested that Interventions designed to dampen inflammation and/or to repair the intestinal barrier might allow for repopulation of the CD4 compartment and limit non-AIDS complications in patients with inadequate immune recovery following ART.
Van Luntzen from Hamburg presented an analysis from their lymph node repository of over 120 samples before and during ART. Large deposits of trapped virions on lymphoid tissue decay rapidly during ART, but viral p24 gag protein can be found on cell surfaces for very long periods of time (years). The germinal centers of lymph nodes are the major anatomical source for productive HIV replication and latency during ART. Major perturbations in adaptive and innate immune responses occur within tissue despite successful ART. It must be determined if these antigens are themselves primary drivers of immune dysfunction despite ART, or simply seen in association with this. Whether these persistent antigens represent in some part an unmeasured reservoir of virion particles that can rekindle infection, separate from the latent genomes hidden within resting CD4 cells, is another important unanswered question.
Markowitz from the Aaron Diamond in NY presented related findings from his studies of 40 subjects identified as acutely or early HIV-1 infected and treated with either 3-drug (N=14) or 5-drug (N=26) ART that included both raltegravir and maraviroc. ART was initiated a mean of 52 days after infection (range 19-155 days). In this cohort, the early treated HIV-infected subjects display comparable levels of markers of cellular and systemic immune activation when compared to the healthy HIV-uninfected controls. Markowitz advocated for a large, prolonged study of the effects of early ART.
Determine host and immune mechanisms that control infection but allow viral persistence:
Saez-Cirion from Institut Pasteur gave the next plenary, providing an overview of the characteristics of patients who innately control HIV infection. These patients generally display a strong response of IgG2 class antibodies to HIV antigens, lower levels of cell-associated HIV DNA, and a potent CD8 HIV-specific immune response. Such patients are more likely to possess the human cell HLA antigens B27 or B57. As another mark of restricted HIV replication, lower copy numbers of integrated HIV genomes are found (Saez-Cirion, Blood 2011; Chen, JCI 2011). Saez-Cirion again pointed to the findings of Shan (Immunity 2012) as evidence that such patients could more effectively clear residual virus when it was expressed from within resting CD4 T cells. He then foreshadowed findings from the Visconti cohort, which will be discussed in detail further on.
Chakrabarti of Institut Pasteur presented studies of immune response in such HIV controllers in comparison to ART-suppressed patients. Effector cells from controllers tended to secrete more interferon gamma and perforin in response to HIV antigens, and unexpectedly IL17, a cytokine that support CD4 cell maintenance, was low. An interesting but unanswered question was why do anti-HIV responses in controllers persist so well despite low levels of HIV antigen presentation or production? Chakrabarti hypothesized that the immune responding cells from these patients had a rapid proliferative capacity of cells that were highly functional (vi receptor avidity).
French from the University of Western Australia, Perth reinforced these findings in a similar cohort of HIV controllers, focusing on the contribution of antibody responses. 12.5% of controllers but no non-controllers had a positive IgG2 antibody to rp55 (p=0.14). Gag antigens were the dominant Ab response in IgG2. Such antibodies may contribute to control via preferential binding to Fc-gamma receptor IIA (a major receptor for antibodies on plasmacytoid dendritic cells). French proposed that a vaccine strategy might include enhancement of IgG isotype switching, to augment these protective IgG2 class antibodies.
The next two short talks addressed the activity of the innate human ApoBEC proteins, which extert a direct antiviral effect by inducing mutations in the HIV RNAs produced by infected cells. Within infected patients, there is a battle between the viral Vif protein, which degrades human ApoBEC, and ApoBEC, which damages HIV RNA strands. Kikuchi from Tokyo found that the Vif protein of HIV recovered from elite controllers was less active that Vif from patients with progressive infection. However, this finding could not be explained by recognized mutations within Vif, and so requires further study. De Pasquale from Vanderbilt University studied the levels of ApoBEC in resting CD4+ T memory cells, that largest known reservoir of latent HIV that persists despite ART. They found that when tested in the lab, the infectiousness of virions recovered from CD4 cells correlated inversely with A3G protein level in the resting cells. This again reinforced the idea that upregulating ApoBEC or inhibiting Vif could be a novel approach to eliminating HIV, if latent virions that we not be expressed could be induced to escape from latency.
Study, compare, and validate assays to measure persistent infection:
Sarah Palmer opened the next session with a plenary talk that focused on assays to measure rare and persistent HIV in the presence of suppressive ART. She discussed the principal that most persistent infection resides in tissues rather than in circulating cells, but that biopsies of patients reflect only a tiny fraction of the total infected tissue, often cannot acquire enough cells to measure rare events, and unlike blood or plasma samples are subject to the challenge of the spatial distribution of infection, wherein a biopsy may miss the specific place where HIV is present. On the other hand, although plasma HIV RNA measures are relatively inexpensive and easy to acquire, but the PCR amplification needed for the HIV RNA single-copy assay (SCA) is technically difficult, may miss virions trapped in tissues, is subject to stochastic Poisson effects at the one-copy low end of the assay, and cannot measure intracellular RNA expression. The viral outgrowth assay from resting CD4+ cells is the gold-standard test to measure the frequency of latent infection, but is also difficult, slow, costly, and also somewhat imprecise when the frequency of infection is low (less than 1 in 3 million cells) as is often seen in durably treated patients. Measures of total HIV DNA are simple, but as discussed above most of this DNA is irrelevant. Specialized PCR assays to measure only integrated HIV DNA copies are an improvement, but more cells are needed for this assay, improvements in precision are needed at the lower end of the assay, and still at least 80% of the genomes measured are likely to be replication incompetent. Palmer reviewed recent studies presented at meetings over the summer of assays performed on samples donated by Tim Brown, the Berlin patient. In summary, SCA was intermittently positive, but mostly <1 c/ml. HIV RNA and RNA were mostly undetectable in PBMCs, plasma, and CSF by PCR, and viral outgrowth assays of PBMCs were negative. In three labs, PCR assays were positive: in two plasma samples, and 1 GALT biopsy sample. However, at least one assay seems likely to be positive due to contamination. Palmer felt that it was not clear that such studies would ever prove that HIV was eradicated with sufficient certainty to assure the success of an ART interruption.
Charline Bacchus then presented findings from the French VISCONTI cohort. The Virological and Immunological Studies in CONtrollers after Treatment Interruption (VISCONTI) examined patients who started ART within 10 weeks of HIV infection (average 40 days), and then controlled viremia after ART interruption, following several years of suppressive therapy. Bacchus reported studies on a 12 patient subset of this group, who had controlled HIV with RNA <50 c/ml after ART interruption for a median of 76 [IQR: 67.5-84.5] months after 3 [IQR: 1.7-5.9]. There was a lot of interest and excitement around the Visconti cohort. I personally feel that this is due to an unrealistic hope that achieving a functional cure might be simpler than it seems. The reason that these patients were able to control viremia after ART interruption is unclear. In fact, the cohort confounded all expectations. It was stated that elite controllers have a low peak viremia in AHI, but the Visconti cohort post-ART controllers did not. HLA B35 was over-represented in the Visconti cohort, not the expected protective allelles HLA B27 or B57. Prior controllers cohorts have high interferon gamma elispot responses to HIV antigens, but the Visconti cohort did not. Cell activation as measured by CD38+DR+ has been high in controllers but was lower in Visconti. Bacchus and Saez-Cirion suggested that at least 12 months of ART post-AHI would control viremia for at least 2 years in 5-15% of patients. In my opinion many of these patient appeared to be treated in early, not acute infection. And in our experience in Chapel Hill, it is rare to observe elite control as observed in Visconti post ART in AHI, but perhaps 5% is rare. I understand that the Visconti cohort was founded by patients with symptoms of acute HIV infection (a minority of all those with AHI) who brought themselves to treaters in France requesting ART, and were enrolled. About 750 patient were enrolled, and the controller cohort numbers only 14. I think it remains to be demonstrated that this is a new phenomenon, how durable this control really is over longer periods of time, and that ART is increasing the frequency of elite control, rather than a chance observation.
Joe Wong from UCSF discussed extensive work done by his group studying persistent HIV in the gut-associated lymphoid tissue (GALT). As with blood cells, preferential retention of HIV DNA+ cells in GALT appears to be greatest in central memory T-cell and, to a lesser extent, effector memory T-cell subsets. In circulating cells there is more HIV DNA found in naïve and central memory CD4+ T cells. In the Ileum and rectum more HIV DNA is found in transitional and effector subsets. Quantitatively, ileal and rectal cells had more HIV DNA than PBMCs, ileal cells statistically more.
Chomont from the Vaccine and Gene Therapy Institute in Florida presented data from an alternative approach to study latency. CD4 T-cells are isolated from PBMCs of virally suppressed subjects by negative selection. These cells, of which a rare number within the population are naturally latently infected cells, were cultured for 9 days in the presence of antiretroviral drugs to spreading of virus within the culture and new rounds of infection. Then several compounds that disrupt mechanisms of HIV persistence were tested for their ability to induce viral production, as measured by PCR for viral RNA released into the culture media.
Detectable HIV RNA in tissue culture media, presumably reflecting release of viral particles, was measured in about 15% of the cultures of cells from patients without any drug exposure. Upon potent T cell activation via the T cell receptor using anti-CD3/anti-CD28 co-stimulation, about 85% of the pools of cells produced HIV RNA. Well-described anti-latency compounds (Prostratin, SAHA, VPA), and other compounds selected for their capacity to revert viral latency in a cell line model displayed differing abilities to induce RNA expression, in some cases less activity than expected. This assay is similar to that used by our group to study SAHA in patients, with some important technical differences. It will be a key area of study to develop and validate these techniques, as the ability to reliably predict which molecules will have anti-latency activity in vivo is sorely needed.
Finally, Del Prete and Lifson from SAIC-Frederick described approaches to assess the in vivo residual virus pool, and eradication strategies in SIV-infected rhesus macaques. A novel five drug ART regimen, comprising a triple formulation of two nucleos(t)ide reverse transcriptase inhibitors (FTC, tenofovir) with an integrase inhibitor (dolutegravir) dosed once-daily by subcutaneous injection along with a twice-daily orally-dosed boosted protease inhibitor (ritonavir-boosted darunavir), was administered to six SIVmac239-infected indian rhesus macaques starting at 4 weeks post-infection for at least 32 weeks. An ex vivo limiting dilution SIV induction assay was developed using CD4+ T cells from the ART-suppressed animals. Cells were exposed to the HDAC inhibitors vorinostat (SAHA) and romidepsin with monitoring of cellular histone acetylation and viral RNA levels in culture supernatants. This super-ART was able to drop pretreatment plasma viral loads from 10,000 to 10 million RNA copies/ml (illustrating the breadth in the model, as in people) to undetectable levels in a sensitive assay (< 30 RNA copies/ml plasma) in 4 of 6 macaques within 10 weeks of ART, and all 6 animals had undetectable VLs by 20 weeks. Ex vivo treatment of CD4+ T cells with Vorinostat or Romidepsin induced measurable increases in cellular acetylated histone levels in the laboratory with associated production of SIV RNA, allowing estimation of the frequency of cells harboring inducible virus. These were encouraging results, demonstrating that techniques piloted in the human vorinostat study could be replicated in monkeys, suggesting that future therapies might be piloted in this animal model. Still to be overcome are the challenges of super-ART to achieve full suppression, the need to obtain enough target cells to accurately measure rare events. On the up side, if eradication strategies are successful, it should be easy to detect lack of viral rebound.
Develop and test therapeutic agents or immunological strategies to safely eliminate latent infection in individuals on ART:
The author opened the next session. I reviewed the mechanisms of latency that could become targets for anti-latency therapy, and the assays we used to study vorinostat (SAHA) in man. The final results of this study were published in Nature the following week during IAS (http://www.nature.com/nature/journal/v487/n7408/full/nature11286.html). I then reviewed techniques and assays currently in use to discover and develop other reagents that could break latency, either alone or in combination with HDAC inhibitors like vorinostat, in cell and animal model systems. Finally, I discussed the emerging concept that eradication therapies are likely to encompass a combination approach, and that efforts to improve the immune system's ability to recognize infected cells and clear them (therapeutic vaccines) are likely to play a role in this effort.
Link to Nature article:
AIDS Cure Quest Advances as Cancer Drug Finds Hidden HIV - (07/27/12)
Sutmiller from the Janssen discovery laboratories (formerly Tibotec) described the screening of 35,000 compounds using a Jurkat HIV-GFP system seeking latent HIV reactivating compounds. In a first effort, we have identified hundreds of small molecules with HIV reactivating potential. Positive hits were further tested in a latent HIV-primary CD4+ T cell based assay to identify those capable of reactivating the virus in primary T cells. In addition, to characterize the many compounds with unknown mode of action, they were further assessed using assays for T cell activation, LTR-mutants, specific kinase inhibition and microarrays. The use of primary T cells harboring silenced HIV in the selection and prioritization of compounds seemed critical. Hit compounds were found that acted via protein kinase C (PKC) pathways, as does prostratin, and via HDAC inhibition, like vorinostat.
Link to Janssen slide presentation:
Janssen's HIV Cure Effort: High throughput screening for HIV reactivating compounds - (08/01/12)
Henrich from the Kuritzkes group at Harvard reported an interesting and important case of two patients with long-term reduction in peripheral blood HIV-1 reservoirs following reduced-intensity conditioning allogeneic stem cell transplantation. Of note, patients maintained ART throughout their treatment, and did not receive a transplant that was HIV-resistant, as did the Berlin patient. Analysis was performed on banked samples (1 pre- and 3 post-RICAlloSCT) for both patients. SCA for HIV RNA were stably <1 copy/ml following engraftment. No HIV-1 DNA was detected 8 to 17 months after alloSCT in PBMC from both patients despite presence of modest levels of total PBMC-associated HIV-1 DNA prior to and 2-3 months after SCT (87-271 copies/million PBMCs) at a detection level of ca. 10 copies per million cells. 2-LTR circles were not detected at any time-point despite excellent recovery of episomal mitochondrial DNA. Both patients were heterozygous for ccr5∼32 mutation prior to transplant. Both patients remained virologically suppressed on ART, but were either started on prednisone or continued on tacrolimus/sirolimus immunosuppressive therapy for chronic graft-versus-host disease (GVHD) near the time of loss of HIV-1 reservoir detection. These observations demonstrate the ability of allogeneic transplant, and/or graft-vs-host disease, to clear HIV infected cells as measured by HIV DNA and the inability to recover virus from cultures PBMCs. It demonstrates the ability of ART to protect the newly engrafted immune system from HIV. If the patients choose to interrupt ART and no rebound of viremia is seen, that would provide a confirmation for the experience of the Berlin patient, in the absence of the transplant of HIV-resistant cells.
Lifson then presented the initial data from a trial of vorinostat (SAHA) in the SIVmac239-infected rhesus macaque. As discussed 6 Indian rhesus macaques were treated with 6 drugs (PMPA/FTC/DRV-RTV/L-870812/L-870564) at 4 weeks post-infection. After 22 weeks of ART, SAHA was added in 21-day courses of daily treatment, 45 mg/kg of 2 cycles, and 57 mg/kg for 2 cycles. Viremia was mostly <30 c/ml after ART induction, but was intermittently detectable > 30 c/ml in several animals. When tested ex vivo, all 6 animal had cells that expressed HIVRNA when treated in the lab. However, SAHA was rapidly metabolized, and often undetectable in the plasma 4 hours after dosing. Some oral doses may not have been well absorbed by the animals. At some time points cellular acetylation, a marker of SAHA activity, was seen, but inconsistently. SIV RNA induction in cells taken directly from the animals was seen, but inconsistently and not always when acetylation was measured. Overall, the study did not confirm the findings reported in humans, but there still appear to be technical hurdles to the study of ART and SAHA in this model.
Romerio from the Institute of Human Virology, Baltimore, reported a complete transcriptome analysis of latently infected CD4+ T cells. He has developed an in vitro model suitable to investigate HIV-1 latency in CD4+ T cells. In this system CD4+ T cells are activated with dendritic cells and antigen, infected in vitro with HIV-1, and then brought back to quiescence through a resting phase in the presence of interleukin-7. During the resting phase, the latently infected cells generated lack expression of activation markers; do not undergo cellular proliferation and do not sustain viral replication. However, in these cells HIV p24 gag antigen persists intracellularly, an unusual finding that nevertheless allows the identification and separation of latently infected vs. uninfected cells from the same cell culture at the end of the resting phase, thus allowing specific transcriptome profiling of infected cells at day 56 post-infection. Romerio found that CD2 was upregulated in latently infected cells, along with 33 other unnamed transcripts involved in an array of cellular functions. It would be very exciting if these findings can be validated, and a surface marker that identifies cells that are, or at least are more likely to be, latently infected can be identified.
Develop and test strategies to enhance the capacity of the host response to control active viral replication:
Giuseppe Pantaleo of the University of Lausanne opened the final session with an overview of the many studies of HIV pathogenesis and therapeutic vaccines over the last decades. Although there are many potential approaches that may be developed, there have been only 29 clinical trials in last 15 years, and in only 13 were modest measures of therapeutic efficacy reported. Pantaleo urged a broad and deep re-evaluation and re-engagement in these approaches.
Apetrei of the University of Pittsburgh proposed that the infection of rhesus macaques with the SIVagm strain could serve as a model of functional cure. In this natural host for this SIV strain, initial high level viremia (108 copies/ml) and massive mucosal CD4+ T cell depletion, reminiscent of HIV infection in man, are followed by durable control of SIVagm replication, complete recovery of CD4+ T cells, normalization of T-cell activation, and loss of SIV antibodies. This state can be temporary reversed by experimental CD8-cell depletion, demonstrating that is is driven by an immune response.
Debre from INSERM Paris proposed that a therapeutic vaccine could neutralize HIV, protect effector T cells, and reduce viral reservoirs. In another vein, Takahara from the University of Tokyo showed that a therapeutic vaccine could induce HIV-specific cytotoxic responses, and lower viremia. Lara Vojnov from the University of Wisconsin-Madison raised a new concern, as she showed that at least in the SIV model, SIV-specific CD8+ T cells that could control viral replication T cells were ineffective in lysing SIV-infected macrophages.
The meeting then closed with Verdult's presentation, a young investigator award, and closing words by Dr. Barre-Sinoussi. The meeting clearly demonstrated the energy and excitement in the field, and marked what is hoped to be a serious and durable effort to develop and implement scalable eradication therapies. Dr. Barre-Sinoussi is to be applauded for focusing the resources of the IAS on this goal.