Cure research takes centre stage: proof of concept for activating the latent reservoir
Simon Collins, HIV i-Base
For the last two years the major HIV conferences, including CROI and the International AIDS Society (IAS) have included cure research prominently in the main programme. This is new and significant.
At CROI in 2010, Anthony Fauci, head of the US National Institute of Allergy and Infectious Diseases (NIAID) announced that the US government would be launching new funding for cure research. 
Many of the researchers in this field have been working on a cure years, some for decades. But the new drive for this research to receive better funding is clearly an important factor in how quickly future progress will be made.
The new funding may, in part at least, have also been driven by the responsibility that America has assumed as the largest donor for global HIV treatment programmes. Over the last ten years, ARV access in low and middle-income countries has increased from less than 0.5 million people in 2002 to over 6.5 million people in 2012. A long-term alternative to lifelong treatment is therefore likely to be an economic as well as a medical necessity. While current cure research uses specialised and expensive procedures, as with all new developments, including ARVs, high initial costs would hopefully be driven down to become more widely affordable.
The IAS has also been developing a leadership role to coordinate global funding for cure studies and to hopefully focus on a research map that will minimise duplication.  The IAS organised workshops prior to each of it’s last two conferences and another is planned prior to the Washington meeting in July 2012.  Several community workshops, including one before CROI this year have also contributed to broadening awareness of the potential for a cure. [4, 5]
In addition to an oral abstract session this year, CROI included several helpful presentations of the current research in the preconference workshops for young investigators, particularly the overview by John Mellors and the talk on animal models for latency by Vincente Planelles. [6, 7]
The Berlin cure
Whether through mediated immunity (referred to as a functional cure) or eradication (a sterilising cure), the ability to overcome lifelong treatment has always been an ultimate goal, even while the focus for recent years shifted to achieving more effective, tolerable and durable treatments.
The first report of a cure following stem cell transplantation from a donor who was naturally resistant to HIV infection (he was homozygous for the delta-32 deletion in CCR5) was at CROI in 2008  and increasing press coverage since had made this a highly publicised case, and brought optimism to cure research.
The mechanism responsible for curing Timothy Brown (a.k.a. the Berlin Patient) who has been off treatment now with no evidence of HIV for over four years has not been isolated to a single component from a complex and risky set of procedures.
In addition to myeloablative chemotherapy and total body irradiation to kill both HIV infected and uninfected immune cells, he received antithymocyte globulins, cylcosporin, mycophenolate acid (MMF) and gemtuzumab (anti-CD33) that would also have killed HIV-infected and uninfected cells, followed by allogeneic stem cell transplants from a donor homozygous for delta-32 mutation, which should have reseeded an immune system resistant to CCR5 HIV infection, he developed graft vs host disease (GVHD) indicating he had accepted the donor immune system. These procedures have a 25% mortality risk and he underwent each procedure twice as the course was repeated.
An oral presentation at CROI reported on ten patients on suppressed ART who underwent autologous (self-donated) hematopoietic stem cell transplantation for AIDS related lymphoma, which is a less risky procedure than that used by Tim Brown. Unfortunately, persistent HIV viraemia was still detected in 9 of 10 patients post-transplant, with a median viral load of 1.5 copies/mL (range: <0.2 to 26) and median total HIV-1 DNA of 554 copies/million PBMCs (range: <0.4 to 2179). 2-LTR circles were detectable post-transplant in only 2 of 10 patients (range: 1 to 7 copies/million PBMC). The only patient with undetectable plasma viral load had the highest levels of HIV-1 DNA and 2-LTR circles. Additionally, plasma viraemia persisted in a patient with undetectable HIV-1 DNA in PBMC. Although the authors concluded that this showed that the CCR5 delta-32 donor was essential in the Berlin case, patients in their study also did not have total body irradiation, graft vs. host disease, and were reinfused with their own stem cells, which could have included HIV-infected T-cells. 
A further US study is about to open of allogeneic stem cell transplant in HIV positive people with bone marrow failure with the hope that 1 or 2 of the 15 patients may also be able to be matched to a delta-32 donor to see if the Berlin case can be repeated. 
This will involve overcoming the difficulty of finding and matching a delta-32 donor who is also compatible on 8 HLA types. At the community cure meeting prior to CROI, John Zaia from the City of Hope Cancer Centre near Los Angeles described an initiative to develop an inventory
of cord blood stem cell donors as an international resource, and to date, out of 13,000 donors tested, 90 have been identified as being homozygous for the CCR5 delta-32 deletion. 
First activation of latently infected resting T-cell reservoir in vivo
While many aspects of this research are controversial, there is broad consensus on the need for a strategy to overcome the reservoir of long- lived, latently infected, resting CD4 cells that harbour integrated HIV and that are not reached by current ART.
Most notably, an oral presentation at CROI included results from a proof of concept study that viral latency might be overcome. David Margolis from the University of North Carolina presented results in an oral late breaker presentation that the use of a single dose of the histone deacetylase (HDAC) inhibitor vorinostat (suberoylanilide hydroxamic acid, SAHA) is able to activate latently infected resting CD4 cells.  In 2005, Margolis presented results from using another HDAC inhibitor, valproic acid, to stimulate the latent reservoir.
Of the 11 human histone deacetylase, HDACs 1, 2, and 3 are the primary enzymes that limit activation of HIV integrated into cells by producing a barrier that maintains latency. Vorinostat is a selective inhibitor of HDAC 1, 2, and 3 that has been shown to induce HIV expression from latently infected resting cells ex vivo. However, vorinostat, although approved as a cancer treatment also has mutogenic properties.
In this proof of concept study, the change in the latent reservoir was determined by measuring cell associated HIV RNA specifically in the resting cell population. This involves harvesting approximately four billion lymphocytes from each aviraemic patient by leukopheresis that are treated with magnetic antibody beads to leave 200-1000 resting CD4 cells that can be tested by RNA PCR.
Six study participants had baseline measures of activation, that were tested ex vivo after exposure to vorinostat and that demonstrated that a change was measurable in all patients. Each patient also undertook a single 200 mg safety dose and a separate single 400 mg dose of vorinostat for a PK study to decide the timing for the second leukopheresis used to determine efficacy.
Following a second, therapeutic 400 mg dose, all six patients responded with a highly significant mean 4.8 fold increase (range 1.5-10-fold) of RNA expression in resting CD4 cells (p<0.01). The treatment was well tolerated with no reported side effects associated with vorinostat and none greater than grade 1. Of note, and perhaps surprisingly, no increases in HIV plasma RNA were detected using a single copy/mL test.
The study concluded that is the first demonstration of activation of latent resting HIV-infected CD4 cells in vivo. However, these results are still preliminary. While the proof-of-concept is exciting, Margolis suggested that this might be seen as the equivalent of a “ddC moment in relation to HAART”.
Additionally, other molecules may be more effective compounds to activate latency and in vitro data suggesting panobinostat as more active that vorinostat were presented in a poster. 
Earlier in the same conference session Liang Shan reported that latently infected resting CD4 cells treated with vorinostat survived despite viral cytopathic effects, even in the presence of autologous CD8 cells from most patients on ART concluding “that stimulating HIV-1-specific CTL responses prior to reactivating latent HIV-1 may be essential for successful eradication efforts and should be considered in future clinical trials”. 
Treatment during early infection
Theoretically, the easiest targets for cure research might be those patients diagnosed earliest in their infection, who promptly start treatment and who maintain suppressed viraemia for many years.
Although the latent cell reservoir is established within weeks of infection and is likely to be slowly reduced after years on effective ART, in nearly all patients, viraemia rapidly returns within weeks if treatment is interrupted. Even when HIV is reduced to being present in less than 1 in 1.7 billion cells, this is sufficient for systemic infection to quickly be reestablished (within two months) if treatment is stopped. 
While levels this low might question the importance of a treatment to target the viral reservoir, they can so far only be achieved with very early treatment and/or many years of viral suppression. The need to reduce the viral reservoir more quickly will be a concern for everyone else who started ART during chronic infection.
Rapid viral rebound without treatment has been widely reported in numerous treatment interruption studies. However, several small cohorts have also reported viral control in a minority of patients, usually in those who initiated treatment in acute infection and maintained undetectable viral load for several years.
Last year at CROI, the ANRS Visconti study reported small numbers of patients who started treatment in early infection (after serconversion, median viral load >100,000 copies/mL), maintained viral suppression for >3 years on treatment and who have subsequently controlled viraemia off treatment for >6 years.  This year at CROI similar cases were reported in posters by two other groups.
Maria Salgado and colleagues reported a single case of a patient who initiated treatment during seroconversion (viral load >750,000 c/mL, western blot indeterminate) for three years and after stopping ART has since maintained viral load suppressed to <50 copies/mL off-treatment for more than nine years. Initial and current viral isolates are dual CCR5/CXCR4 tropic and fully replication-competent in vitro. Minimal viral evolution has been detected over the 11 years.
He is reported to currently have low titers of neutralising antibodies to heterologous and autologous HIV-1 isolates, and his CD8+ T cells do not have potent HIV suppressive activity suggesting a mechanism other than CTL-mediated suppression reported in elite controllers. 
Alain Lafeuillade from General Hospital, Toulon (who is also one of the key organisers of the International HIV Persistence Workshop that has been meeting every two years since 2003) reported that 17% (8/45) of a cohort of patients treated at seroconversion for a median of 2.2 years (range 1.8 to 4.0) have remained off treatment for more than 10 years, two of whom remain suppressed to <20 copies/mL (median 2,500 copies/mL for the other six). The 37 people who restarted treatment (due to confirmed CD4 decline to <350 cells/mm3) did this after a
median of 5.0 years (range 3.0-8.0) off-treatment. The study suggested the protective mechanism could be relate to early ART reducing the HIV reservoir but also emphasised that such responses seem to be rare. 
A poster from Joseph Margolick and colleagues reported small differences in viraemia between people diagnosed in early infection (within a year of infection) and randomised to immediate treatment (n=57) year) and those who did not start early treatment (n=24). However, study numbers were very low at the evaluation point (24 months after stopping treatment of 24 months after diagnosis) due to ~20% loss to follow- up and exclusion of people who restarted treatment for other reasons. 
Generally small differences were also reported from early treatment in the larger SPARTAC study that randomised almost 400 people (diagnosed within 6 months of infection) to deferred ART or immediate treatment for either 3 months or 12 months, and who then stopped treatment. 
However, in the context of eradication research, two oral presentations suggested that early treatment, while too late to prevent the establishment of the viral reservoir, might reduce the pool of latently infected cells.
Maria Buzon and colleagues estimated the size of the viral reservoir in patients treated for more than ten years who initiated ART within 3 months of infection (n=9) and compared levels integrated and total HIV DNA levels to people who started treatment during chronic infection (n=26) and to elite controllers (n=37). 
Integrated and total DNA levels were significantly lower in both primary treated (p=0.06 and p=0.001, respectively) and elite controllers (p=0.003 and p<0.0001, respectively) compared to those treated in chronic infection. In addition, the ratio between total and integrated HIV-DNA was significantly lower in early treated and elite controllers (both p=0.04 vs chronic) with no differences between acute and EC groups.
Although patient numbers were small, differences were also reported when comparing how soon treatment had been started with patients treated during Fiebig stage III or IV vs stage V having significantly lower levels of both total and integrated HIV DNA after two years.
An oral presentation by Alan Perelson from the Los Alamos National Laboratory used mathematical modelling to look at the impact of early treatment of 27 people treated during acute infection on the size of the latent reservoir, and the relationship of both to initial viral load and target cell ability. 
This study also reported that earlier ART, including earlier during primary infection, had a measurable impact related to the initial size of the reservoir, with patients who already started with very low levels of resting cell infection (who also had low levels of peak viral load) experiencing less change in the reduction of resting cell decay. The model also suggested that CD4 T cell increases in response to successful ART was not increasing the viral reservoir.
Research into a functional cure
Other groups are focusing on immunological interventions that would support a functional rather than eradicating cure.
Pablo Tebas from the University of Pennsylvania, presented additional safety and efficacy results from the use of zinc finger nuclease (ZFN) modification of CD4 cells (using SB-738) to a CCR5-deleted phenotype (in development by Sangamo BioSciences). 
This process involves harvesting cells by apheresis, treating them with SB-738 to produce 13-35% of cells with CCR5-detetions in vitro. The cells are then expanded, cryopreserved and 5-30 billion cells are reinfused into the donor patient.
Results were combined from three studies: one in ART responders (baseline CD4 >450 cells/mm3) who subsequently interrupted treatment (group 1, n=6) and two in immune non-responders (baseline CD4 <500 cells/mm3) who have not interrupted treatment (group 2, combined n=15). Initial results from these studies were presented at CROI and ICAAC conferences last year.
Most patients were male, white, mean age 48, with a long history of HIV infection (median 12 and 18 years in group 1 and 2 respectively). Mean CD4 count and CD4:CD8 ratio were 921 (+222) cells/mm3 and 1.4 (+0.6) in group 1 and 335 (+89) cells/mm3 and 0.7 (+0.3) in group 2.
Duration of follow-up is now a mean 325 days (range 90 – 738 days).
After infusion, CD4 cells increased by about 1500 cells/mm3 in group 1 (n=6), these then decreased during the treatment interruption but which remained significantly above baseline during follow-up. CD4 responses in group 2 involved an increase of about 500 cells/mm3 which then dropped by about 200-300 which then remained stable out to over a year in the patients who did not interrupt treatment. The expansion of CD4 cells was associated with increases in IL-2, IL-7 and IL-15.
The CD4:CD8 ratio increased significantly in both groups, normalising and remaining at approximately 1.0 throughout follow-up in the group 1 and increasing to approximately 2.5 for the six patients in group 2 decreasing during the treatment interruptions but then remaining stable.
The modified cells continued to be detected through follow-up at 2% of circulating CD4 cells at 48 weeks for most patients. Levels were higher during the treatment interruption for group 2 and then dropping to 2%. Circulation of cells to other tissue sites was confirmed by multiple rectal biopsies where levels of the CCR5-modified cells were comparable to those in blood or higher throughout follow-up.
During the treatment interruption viral load rebounded over the first 8 weeks to around 100,000 copies/mL in a similar way to other interruption studies dropping by one log during the last 4 weeks off-treatment to levels that were generally higher than pre-ART. After three months, when treatment was restarted, viral load become undetectable again in all six patients. One person, later found to be heterozygous for the delta-32 mutation, had a lower rebound (to 10,000 c/mL) and then resuppressed viral load to undetectable by week 8 and remained undetectable off treatment until restarting as per protocol at week 12.
This group used a new method to measure changes in the viral reservoir based on levels of HIV DNA sensitive to low copy numbers (although unable to distinguish between integrated DNA and 2-LTR circles. They reported no detectable change in 4/6 patients with one person have a transient 4-fold increases at week 12 and 20 during the interruption but returning to baseline levels and one person experiencing a 9-fold increase that returned to baseline 16 weeks after restarting treatment.
Side effects were mild and transient, mainly within 24 hours of the infusion (mild chills, fever, headache, fatigue) but included one report of arthritis lasting a few days and abnormal garlic-like body odour.
Next steps include using immunomodulatory drugs such as cylcophosphamide to promote engraftment and increase the percentage of modified cells and studying other patients who are heterozygous for the delta-32 deletion.
Several studies presented studies where pegylated interferon (peg-IFN) was added to ART prior to stopping HIV treatment and continuing peg-IFN. The results suggested that viral rebound was delayed by the peg-IFN via an immune-mediated rather than antiviral mechanism, but these were small studies with short-term follow up (12 and 24 weeks). [23, 24, 25]
The timeline for a cure at this meeting was optimistically referred to as being at least ten years. HIV is a tricky puzzle: the virus is resilient and the range of immune responses is complex. Nevertheless, these advancements in several key and linked areas are crucial advances.
Several networks are encouraging collaborative research in order to be able to compare and evaluate different approaches. [26, 27]
Numerous compounds that are already licensed are already being looked at for their potential to overcome latency. These include prostratin, lonomycin, thapsigargin (a calcium pump inhibitor), PMA, typhostin-A (a non-selective typhosin phosphate inhibitor), CD3/CD8 antibodies for TCR signaling, PLA, toll like receptor 7 (TLR7 including GS-9620) and protein kinase-C (PKC) agonists. Several companies including Gilead and Merck are already screening for and have identified other potential HDAC inhibitors.
The important of informed community participation in partners in this research is particularly important given the ethical considerations for study volunteers. If a mechanism is discovered to cure HIV more widely, it may still only work in some patients.
With current treatment able to nearly normalise life expectancy, a cure has a high bar to overcome. Some of this research will involve asking people on stable treatment to interrupt therapy and some of the interventions will have potentially greater toxicity than their current ART. At least for the foreseeable future, the potential risks in these initial studies are likely to outweigh any personal benefits.
Treatment during primary infection early treatment may put someone at a preferential state to respond to parts of a strategy to cure HIV, and some people diagnosed this early may want to take this decision. Because the latent pool is smaller in such patients, those initiating ART during acute HIV may be the best candidates for pilot studies attempting HIV eradication.
Unless stated otherwise, all references are to the Programme and Abstracts for the 19th Conference of Retroviruses and Opportunistic Infections, 5–8 March 2012, Seattle.
- Fauci A. A Fauci. The HIV/AIDS research agenda: a view from NIAID. 17th CROI 2010, 16-19 February 2010, San Francisco. Plenary session 5. February 16, 2010.
- International AIDS Society (IAS). Towards a cure: global scientific strategy. February 2011.
- International AIDS Society (IAS). Towards a cure: global scientific strategy. Washington meeting 2012.
- Treatment Action Group. Report: HIV cure-related clinical research workshop, October 2011.
- ATAC Cure Research Workshop. 4 March 2012.
- Mellors J. HIV Reservoirs and Cure Research. Workshop: Program Committee Workshop for New Investigators and Trainees (Part 2) Monday 5 March 2012 11:00 AM, Seattle.
- Planelles V. Models for the Study of Latency. Workshop: Frontiers in Laboratory Science Workshop. Monday 5 March 2012 2:30 PM, Seattle.
- Hutter G et al. Treatment of HIV-1 infection by allogeneic CCR5-D32/D32 stem cell transplantation: a promising approach. Poster abstract 719.
- Cillo A et al. Plasma viremia and cellular HIV-1 DNA persist despite autologous hematopoietic stem cell transplantation for AIDS-related lymphoma. Oral abstract 154.
- Allogeneic Transplant in HIV Patients (BMT CTN 0903)
- Margolis D et al. Administration of vorinostat disrupts HIV-1 latency in patients on ART. Oral late breaker 157LB.
- Rasmussen T et al. The histone deacetylase inhibitor (HDACi) panobinostat (LBH589) stimulates HIV-1 expression more potently than other HDACi in clinical use and disrupts HIV latency at clinically achievable concentrations. Poster abstract 370.
- Shan et al. Elimination of the latent reservoir for HIV-1 requires induction of cytolytic T lymphocyte responses. Oral abstract 153.
- Chun T-W et al. Rebound of plasma viremia following cessation of antiretroviral therapy despite profoundly low levels of HIV reservoir: implications for eradication. AIDS 2010: 24 (18): p 2803–2808 (27 November).
- Saez-Cirion A et al. Long-term HIV-1 Control after interruption of a treatment initiated at the time of primary infection is associated to low cell-associated HIV DNA levels: ANRS VISCONTI Study. Poster abstract 515.
- 16.Salgado M et al. 357 Title: Prolonged Control of Replication-competent Dual-tropic HIV-1 following Cessation of HAART. Poster abstract 357.
- Lafeuillade A et al. Long-term control of HIV reservoir after a 2-year ART course at acute infection. Poster abstract 358.
- Margolick J et al. 356 Effect of randomized HAART on viral suppression off therapy in patients with acute/early HIV infection. Poster abstract 356.
- Fidler S et al. The effect of short-course antiretroviral therapy in primary HIV infection: final results from an international randomised controlled trial; SPARTAC. 6th IAS Conference on HIV pathogenesis, treatment and prevention. 17-20 July 2011. Rome, Italy. Oral abstract WELBX06. See:
- Buzon M et al. Treatment of early HIV infection reduces viral reservoir to levels found in elite controllers. Oral abstract 151.
- Perelson et al. Immediate antiviral therapy restricts resting CD4+ infection but does not accelerate the decay of latent infection. Oral abstract 152.
- Tebas et al. Induction of acquired CCR5 deficiency with zinc finger nuclease-modified autologous CD4 T cells (SB-728-T) correlates with increases in CD4 count and effects on viral load in HIV-infected subjects. Oral abstract 155.
- Mexas A et al. Concurrent Measurements of Total and Integrated HIV DNA Provide Insight into the Mechanism of Reduced Reservoir Size in an Interferon-alpha followed by Structured Treatment Interruption Trial. Poster abstract 374.
- Papasavvas E et al. Immune Correlates of Sustained IFN-alpha-mediated Suppression of HIV Replication: Association with IFN-alpha-mediated Signaling and Increased NK Cell Responses. Oral abstract 93.
- Azzon L et al. Pegylated interferon-alpha 2A monotherapy induces durable suppression of HIV-1 replication and decreased HIV DNA integration following ART interruption. Poster abstract 631.
- Collaboratory of AIDS Researchers for Eradication (CARE) and the Delaney AIDS Research Enterprise (DARE).
- Defeat HIV – The Delaney Cell and Genome Engineering Initiative, Fred Hutchinson Cancer Research Center, University of Washington, Seattle.