In Pursuit of a Cure

An update on efforts to cure HIV from the 5th International Workshop on HIV Persistence during Therapy

By Richard Jefferys

From December 6-9, 2011, around 250 scientists assembled on the sunny island of St. Maarten for the 5th International Workshop on HIV Persistence during Therapy. The tropical location conjures up notions of escape and fantasy and until recently, the guiding philosophy behind this biannual meeting—that HIV persistence can be addressed and the infection cured—was widely perceived to belong in the realm of fevered dreams.

There are several parallels between cure research and the vaccine field, which for many years had to fend off criticism that there was no clear “proof of concept” to demonstrate that immunization against HIV was possible. That is until the results of the RV144 trial in Thailand finally quelled those arguments. Researchers pursuing a cure for HIV faced an even more vertiginous mountain of skepticism, but this mountain has now been moved; not by a large, randomized clinical trial, but by a single individual named Timothy Ray Brown.

Brown’s case, which has understandably garnered considerable media coverage, first came to light in 2008 in a poster presentation at the 15th Conference on Retroviruses & Opportunistic Infections (CROI). After being on successful antiretroviral therapy (ART) for many years, Brown was diagnosed with acute myeloid leukemia (AML), necessitating a complex series of anti-cancer treatments and, ultimately, a stem cell transplant. The hematologist responsible for Brown’s care, German doctor Gero Hütter, identified a donor homozygous for the CCR5Δ32 allele, which abrogates expression of the HIV co-receptor CCR5 on cells. After two stem cell transplants from this donor, along with a daunting panoply of chemotherapies and immune suppressive treatments, Brown has remained free not only of AML, but any sign of HIV in blood and tissues, despite being off ART for over four years and counting. In a paper published in 2011, Hütter and colleagues felt able to state: “From these results, it is reasonable to conclude that cure of HIV infection has been achieved in this patient” (1).

In much the same way that the RV144 results invigorated the HIV vaccine field in 2009, the 2011 persistence workshop was suffused with a new optimism and sense of purpose as a result of this one-person proof of concept. Signs of the mainstreaming of cure research abounded: two large pharmaceutical companies—Gilead Sciences and Janssen Pharmaceuticals—described programs that aim to identify compounds capable of targeting the HIV reservoirs that persist despite ART. The ratcheting up of US National Institutes of Health (NIH) support was in evidence due to the participation of representatives from the Martin Delaney Collaboratory program, under which multiple groups of investigators have been funded, and the NIH-sponsored AIDS Clinical Trials Group (ACTG) network, which has recently made the search for a cure a top priority.

But for all the sunnier parallels between the vaccine and cure research domains, they are also both clouded by uncertainty. The mechanisms by which protection was achieved in RV144, and how the trial result might be translated into an efficacious vaccine for all populations, remain to be elucidated. Similarly, exactly how Timothy Brown’s cure was obtained is not completely understood, and the challenges associated with attempting to convert this complex individual case into a scalable, accessible, curative therapy are gargantuan.

In broad strokes, current strategies discussed at the workshop focus largely on two main routes for curing HIV: rousting the virus from cells where its DNA has integrated into the cellular genome—latently infected cells—and/or enhancing the ability of the host to restrain viral activity, either via gene therapies or therapeutic vaccines that aim to bolster HIV-specific immunity. Two possible curative scenarios are envisioned: eradication, wherein all HIV is eliminated from the body, or what is termed a “functional cure,” in which the body is able to tolerate the presence of some residual virus without ill effect.

An animal model

Another similarity between HIV vaccine and cure research is the ongoing quest for an optimized animal model. Jeff Lifson, director of the AIDS and Cancer Virus Program at the National Cancer Institute in Frederick, who has extensively studied SIV infection of macaques, in the context of both preventive vaccines and therapeutic interventions, provided an overview of this pursuit at the workshop. He explained that there is not yet an ideal system for studying viral persistence in macaques; rather, there are several published approaches that continue to be refined.

One approach is to use rhesus macaques infected with reverse transcriptase (RT)-simian immunodeficiency virus (SIV)/HIV hybrids known as SHIV, that encode the reverse transcriptase gene from HIV in order to render the virus susceptible to a wider array of antiretrovirals (2). This strategy is reminiscent of the creation of SHIVs for vaccine research, but in that case Env from HIV was inserted to facilitate the study of antibody-based vaccines.

Another approach is to study rhesus macaques infected with SIVmac251 and treated with multi-drug antiretroviral regimens, including integrase inhibitors (3). And yet another is to study pigtailed macaques dually infected with SIV 17E-Fr and SIVdeltaB670 treated with the ARVs tenofovir, the integrase inhibitor L-870812, saquinavir, and atazanavir (4).

Lifson invoked the metaphor of “building the boat as we sail” to characterize the current state of play in studying persistence in nonhuman primates (NHPs). But, he argued that the diversity of current models was not necessarily a bad thing, citing the adoption of the SHIV89.6P challenge virus by the vaccine field as a cautionary tale of premature standardization (the virus initially appeared to have a number of advantages but turned out to have essentially no relevance to HIV infection).

Part of the difficulty with studying persistence in NHPs is identifying ARV regimens that can mimic the viral control these drugs exhibit in humans. Lifson outlined work conducted by his group to select a multi-drug regimen capable of suppressing the highly virulent SIVmac251 strain, noting that while suppression of viral load to <30 copies/ml blood could be achieved, it took considerably longer than is seen in HIV infection. It could be, Lifson suggested, that SIV infection of Indian rhesus macaques is too virulent. Chinese rhesus macaques appear to have a slower disease course, and Lifson speculated that this subspecies could be better suited for cure-related studies.

During his talk, Lifson also introduced a relatively new technology that is now being employed in both vaccine and viral persistence research: digital polymerase chain reaction (PCR). Unlike traditional PCR, which exponentially amplifies DNA or RNA from a single sample then attempts to back-calculate how much was originally present, the digital version divides a sample into multiple tiny wells and then performs PCR on each well in order to give a binary readout as to whether the target sequence is present or not. The total amount of DNA or RNA in the sample is then calculated by summing the results from the wells using Poisson distribution, allowing for far more accurate quantification of small amounts of genetic material. The approach is ideally suited to measuring very low levels of viral RNA and DNA in people or animals on suppressive ART.

The vagaries of current macaque methodologies for studying viral persistence were further illustrated by the subsequent presentations. Paul Luciw, a professor in the Department of Pathology and Laboratory Medicine at the University of California at Davis, described an experiment involving an RT-SHIV that, unlike SIV, is susceptible to the non-nucleoside reverse transcriptase inhibitor efavirenz. Animals were treated with a triple drug combination comprising efavirenz (fed in peanut butter sandwiches), tenofovir, and emitricabine for 32-35 weeks, then randomly assigned to either receive two additional drugs targeting the latent viral reservoir (prostratin and valproic acid), or remain on antiretrovirals for an additional eight weeks. Prostratin and valproic acid are among the compounds that have emerged from basic science research into the molecular mechanisms regulating HIV latency. Prostratin is a protein kinase C activator that is reported to stimulate latent virus by activating the cellular enzyme nuclear factor kappa B (NF-κB) (5). Valproic acid belongs to a class of drugs called histone deacetylase (HDAC) inhibitors, which have emerged as leading candidates for coaxing latent HIV into action because they interfere with cellular factors that package genetic material and prevent its active transcription. Luciw showed that receipt of prostratin and valproic acid were associated with significant reductions in viral RNA and DNA in tissues, but when ART was subsequently interrupted there were no differences in viral load rebounds between the groups. Luciw acknowledged that the activity of valproic acid against latent virus has been called into question in the time since his experiment was launched, and said follow-up studies with more potent HDAC inhibitors are planned.

Andrea Savarino, program director of HIV Eradication Strategies at the Istituto Superiore di Sanità in Rome, and his graduate student Iart Luca Shytaj were able to report more salutary findings from studies involving Indian rhesus macaques infected with SIVmac251. Savarino’s research group has previously published work suggesting that two existing drugs, buthionine sulfoximine (BSO) and auranofin, have the capacity to deplete latent virus (6; 7). BSO is thought to work by inhibiting the synthesis of the antioxidant glutathione, thereby creating an intracellular environment favorable to HIV transcription. Auranofin is a gold-based rheumatoid arthritis drug that inhibits the proliferation of central memory CD4+ T cells, a major reservoir of latent HIV. At the workshop Savarino presented analyses indicating that combining these drugs with multiple antiretrovirals may have allowed three macaques to maintain SIVmac251 viral loads below the limit of detection (<40 copies/ml) for several months after ART was suspended. Savarino and colleagues believe this finding may be an augury that anti-reservoir approaches can contribute to a functional cure.

Shifting to the mechanisms by which latent HIV reservoirs are maintained, Vicente Planelles, professor in the Department of Pathology, Microbiology, and Immunology at the University of Utah School of Medicine, explained how immunological memory—the biological blessing that allows vaccines to work—can be a curse in the context of viral latency. Specifically, HIV integrates into the DNA of central memory CD4+ T cells, a population endowed with the capacity for long-term persistence and self-renewal by proliferation. Planelles documented that when latently infected central memory CD4+ T cells undergo mitosis, the integrated HIV DNA is copied along with the cellular genome. Some proposed therapies that stimulate central memory CD4+ T-cell proliferation, such as the cytokine interleukin (IL)-7, might therefore increase rather than reduce the size of the viral reservoir. Conversely, drugs that inhibit the division of central memory CD4+ T cells, such as the rheumatoid arthritis treatment auranofin used in Savarino’s experiment, may be able to reduce the amount of latent HIV, but could also potentially risk impairing immunological memory.

The power of ART

For many years now, the most controversial question in the cure research field has been whether ART fully suppresses HIV replication in the majority of individuals. The answer to the question continues to be a point of contention and several talks at the workshop offered differing perspectives. Sarah Palmer, a professor in the Department of Microbiology, Tumor, and Cell Biology at the Karolinska Institute, debuted results from a detailed evaluation of viral genetics prior to ART initiation and after long-term treatment (up to 12 years) in 12 individuals, seven of whom started therapy during acute infection and five during the chronic phase. Palmer was unable to find any evidence of HIV evolution suggestive of ongoing replication in samples from blood and multiple tissues. In a dramatic illustration of the scenario outlined by Planelles, Palmer highlighted a case where a large expansion of HIV DNA was seen that was clearly non-functional (it contained a deletion of the entire protease gene), a finding that could only be explained by the proliferation of the cell containing the defective viral DNA.

Palmer’s conclusion that little or no HIV replication occurs in most people on ART—at least at the sites sampled—was supported by a number of other presentations at the workshop involving intensification of standard ART with additional drugs such as the integrase inhibitor raltegravir and the CCR5 inhibitor maraviroc. Martin Markowitz, professor and clinical director at the Aaron Diamond AIDS Research Center, gave an overview of one such study, comparing ART combinations involving three versus five drugs administered during acute infection. Through follow-up of up to 96 weeks, no significant differences in various measures of the HIV reservoirs in blood and tissues could be documented between the two regimens.

However, a countervailing view was offered by a triple-header of scientists studying ART penetration into lymphoid tissues. The work centered around 12 treatment-naive individuals starting combination ART. Pharmacologist Courtney Fletcher, dean of the College of Pharmacy at the University of Nebraska Medical Center, assessed levels of each component of their ART regimens in blood and lymphoid tissues and reported that, in some cases, suppressive levels were not reached in gut-associated lymphoid tissue (GALT) and lymph nodes. Mario Stevenson, chief of the Division of Infectious Diseases at the University of Miami, then described the virology results, showing that in some of the study participants, levels of HIV DNA forms called 2-LTR circles increased in lymphoid tissue during follow up, indicating ongoing replication was occurring (2-LTR circles are circularized forms of unintegrated HIV DNA produced during viral replication). Stevenson argued that this could occur without evidence of viral evolution if the majority of the events involved just one round of replication. The third co-investigator, Timothy Schacker, professor of medicine and director of the Infectious Disease Clinic at the University of Minnesota, then outlined his ongoing work to correlate the findings with another measure of viral persistence, the trapping of HIV RNA on follicular dendritic cells (FDCs) in lymph nodes, as measured using a technique called in situ hybridization. The data were very preliminary and only derived from a subset of individuals, but he suggested that there was a link between poor drug penetration and the markers of persistent replication. The study is now being expanded, and more data from additional volunteers along with more follow-up time should help shed light on whether this is a broadly applicable phenomenon, as well as the extent to which it might contribute to sustaining HIV reservoirs in the face of ART.

Drug development

Until very recently, the hunt for compounds that might be capable of depleting HIV reservoirs was confined to academic laboratories. But, at the persistence workshop, Romas Geleziunas, director of clinical virology at Gilead Sciences, and Roger Sutmuller, principal scientist at Janssen Pharmaceuticals, explained how these companies are now conducting these searches on an industrial scale. While these are the only two companies to have publicly discussed their research programs so far, rumors were flying that several others have quietly started similar efforts.

Geleziunas explained that Gilead decided to eschew cell-line based models of HIV latency as a screening tool due to concerns that their artificiality can produce misleading results. Instead, a primary CD4+ T cell assay developed by Planelles and Alberto Bosque, research assistant professor at the University of Utah (8) has been adapted to allow high-throughput screening of drug libraries. It is early days, but several new HDAC inhibitors have been identified, one of which is undergoing preliminary toxicology testing. Geleziunas offered an example of how the approach can be used to identify interesting compounds. A pilot screen of a subset of Gilead’s drug library produced a 1% hit rate of active drugs, one of which belonged to a class called kinase inhibitors. Because this was an unexpected finding, a library of kinase inhibitors was then evaluated, producing a much higher hit rate of 20%. Geleziunas noted that this is an example of how drug screening can feed back into basic science research on mechanisms of HIV latency—the identification of kinase inhibitors suggests a previously unappreciated role they may play in HIV latency that can now be investigated.

Gilead is also interested in immune-based therapies, based on the evidence that simply reactivating latent HIV may not be sufficient to kill the infected cell. Tae-Wook Chun from the National Institute of Allergy and Infectious Diseases (NIAID) discussed this issue, highlighting the fact that in his laboratory, viral reactivation by HDAC inhibitors has not induced a notable amount of cell death. Geleziunas stated that Gilead is looking at an agonist of toll-like receptor (TLR) 7 as a potential candidate, based on data indicating it can enhance immune responses to hepatitis B and an encouraging safety profile in a Phase I trial.

Janssen Pharmaceuticals is taking a slightly different approach, outlined by Sutmuller, which starts with a Jurkat cell-line based assay for high-throughput screening and then uses an in-house primary cell testing system for further evaluation of promising candidates. A humanized mouse model of HIV infection will be employed to assess the best leads that emerge from this process (9). Some 35,000 compounds have gone through the preliminary screen to date, with another 480,000 waiting in the wings.

Human trials

While these drug screens are a long way from clinical testing, the final day of the workshop did feature a number of discussions related to planning for interventional trials in humans. Dan Kuritzkes, professor of medicine at Harvard University, is also co-leader of the ACTG network that now has a committee on HIV Reservoirs and Viral Eradication charged with developing trial protocols, and Kuritzkes outlined multiple issues that will likely arise in the design of clinical trials of eradication or functional cure strategies. Looming large among them is the complex ethical conundrum of early stage studies, which may have little prospect of benefiting an individual participant and considerable risks—the primary benefit would be to advance the science and hopefully contribute to the longer term development of curative strategies. Kuritzkes stressed that in an era when effective ART can extend the lifespan of HIV-infected people close to that of their HIV-uninfected counterparts, it will be particularly important to involve potential participants in discussions relating to the adjudication of risks along with regulators, investigators, funding agencies, and institutional review boards.

David Margolis, director of the School of Medicine at the University of North Carolina and a recipient of one of the NIH’s Martin Delaney Collaboratory awards, reported very preliminary results from one of the first major cure-related human studies to be launched, a Phase I trial of the HDAC inhibitor vorinostat (also known as SAHA). The trial is an example of a potentially high-risk endeavor, as the drug is used as a cancer treatment and has a host of potential toxicities, including testing positive on the AMES test for mutagenicity. Margolis acknowledged that it took considerable time and patience to obtain US Food and Drug Administration approval to open the study. So far, four participants (all healthy HIV-infected individuals on stable ART) have received three doses of vorinostat (the first to assess safety, the second for pharmacokinetic analysis, and the third to measure activity). Margolis presented evidence of a mean 4.4-fold increase in HIV RNA expression associated with receipt of the drug. While encouraged by the data, he emphasized that many questions remain, including whether induction of HIV RNA expression leads to clearance of latently infected cells. No serious toxicities were seen in any trial participant.

Data from another clinical trial came from Carl June, professor of pathology and laboratory medicine at the University of Pennsylvania. He provided an update on a gene therapy approach that aims to mimic the knockout of the CCR5 receptor achieved in Timothy Brown, but in a kinder, gentler way. Developed by Sangamo BioSciences, the treatment uses enzymes called zinc finger nucleases to target and disrupt the CCR5 gene in CD4+ T cells extracted from HIV-positive individuals. These modified cells are then expanded in the laboratory and re-infused into the original donor, in hopes of creating a large population of CCR5-negative, HIV-resistant CD4+ T cells. As was reported at the 2011 CROI, participants in two Phase I trials (all receiving ART) have experienced significant CD4+ T cell increases after a single infusion. One of the trials involves a 12-week analytical interruption of ART to gauge whether there is any impact on viral load parameters. So far six individuals have undertaken the interruption, and June noted that the results are extremely interesting.

Perhaps of greatest importance, June has been able to demonstrate a significant correlation between the proportion of modified CD4+ T cells detected and the extent of the diminution in viral load levels documented prior to ART reinitiation. One participant in particular is drawing attention because his viral load declined to undetectable levels (<50 copies/ml) just before the end of the ART interruption. June explained that this individual is heterozygous for the CCR5Δ32 mutation and therefore has the highest proportion of modified CD4+ T cells because the zinc finger nucleases only have to disrupt one CCR5 allele in each CD4+ T cell for expression of the co-receptor to be completely abrogated (as opposed to having to knock out both alleles). Sangamo BioSciences is now investigating methods to increase the proportion of modified cells and attempting to confirm this intriguing observation by recruiting a large cohort of CCR5Δ32 heterozygotes into the trial.

At the end of the three days in St. Maarten, while no earth-shattering breakthroughs had been reported, the sense of a major sea change in the field remained. To close the workshop, the lead organizer Alain Lafeuillade, head of the department of infectious diseases at General Hospital in Toulon, France, introduced Françoise Barré-Sinoussi, director of the Regulation of Retroviral Infections Unit at the Pasteur Institute, Nobel laureate, and incoming President of the International AIDS Society (IAS). Barré-Sinoussi called on attendees to sustain and accelerate the momentum that has gathered behind the pursuit of a cure for HIV and reported that, to this end, IAS is developing a Global Scientific Strategy “Towards an HIV Cure,” and sponsoring a symposium on the subject to take place in Washington, D.C., in July, immediately ahead of the International AIDS Conference.

Richard Jefferys is Coordinator, Michael Palm Basic Science, Vaccines & Prevention Project at the Treatment Action Group.

1. Blood 117, 2791, 2011
2. J. Virol. 84, 2913, 2010
3. Retrovirology 7, 21, 2010
4. Curr. Opin. HIV AIDS 6, 37, 2011
5. J. Biol. Chem. 279, 42008, 2004
6. Retrovirology 6, 52, 2009
7. AIDS 25, 1347, 2011
8. Methods 53, 54, 2011
9. Proc. Natl. Acad. Sci. 103, 15951, 2006