We may need to combine many approaches to achieve a cure, delegates hear

It is unlikely that one single approach will achieve a cure for HIV infection, delegates at a community cure workshop held the day before the Conference on Retroviruses and Opportunistic Infections (CROI 2015) in Seattle, USA, heard last week.

Bone marrow transplants

One approach was to seek ways to replace the body’s immune cells with ones resistant to HIV infection. This is what appears to have happened in the case of Timothy Ray Brown, the one person still apparently cured of HIV, with no return of his virus now after seven years. However, he is as yet the only one, with HIV in other cases eventually reappearing, often after a prolonged period, as in the case of the “Mississippi baby”, whose HIV rebounded after not being detectable for over two years.

Javier Martinez-Picado, of the Spanish HIV research institute IrsiCaixa, documented other attempts to repeat Brown’s cure. Brown, who needed a bone marrow transplant due to lymphoma, had his immune system effectively deleted and replaced by one from a donor with cells lacking the receptor CCR5; this is the receptor the majority of HIV, and nearly all HIV that is transmitted, needs to bind to in order to infect cells.

However, people who lack CCR5 are uncommon; it depends on being born to parents both of whom have the right gene and is at best (in northern Europe) found in only 0.1% of the population, being absent in many other parts of the globe.

Martinez-Picado therefore documented attempts to repeat the same kind of transplant but using stem cells – primordial immune cells – taken from cord blood, the foetal blood left in the umbilical cord and placenta after birth, which is a valuable source of stem cells. There is now a European system of blood banks containing 32,000 units of cord blood. However, blood with the CCR5-negative mutation is still still a scarce resource, and bone marrow transplants often have to be supplemented by donations from relatives or others that are genetically matched but not always CCR5-negative.

He documented one case of a patient with lymphoma who had a similar transplant to Brown. By six weeks after the bone marrow transplant, all the T-cells in his system had been replaced with ones lacking CCR5 and his HIV viral load was undetectable by a number of ultrasensitive tests. This patient however died of his lymphoma three months after his transplant and three more out of a list of seven for whom the procedure has been attempted also died – underlining the fact that bone marrow transplants are taxing procedures with a high mortality rate, normally only used when other cancer therapy has failed.

Some HIV also binds to a different cellular receptor, CXCR4 – which has been associated with faster immune decline in some people – and Martinez-Picado documented one patient with virus that was able to use both receptors. After transplant with CCR5-negative T-cells his HIV viral load, which had been undetectable, jumped to 100,000 copies/ml, composed exclusively of virus using the CXCR4 receptor. However, CXCR4-using HIV usually replicates less well and the patient's viral load (off-treatment) was down to undetectable levels again three months later.

Re-introducing CCR5-deleted cells

Attempts to replace healthy HIV-positive people’s CD4 cells with CCR5-negative ones rely on the introduced cells out-competing ones vulnerable to HIV. Although populations of artificially generated CCR5-negative cells have been introduced and successfully persist in some patients, the reservoir of latently infected cells persists and has proven extremely difficult to eradicate.

Sangamo is a company that has pioneered the grafting of autologous T-lymphocytes into trial participants. This means that a participants own T-cells are extracted, treated with zinc finger nuclease enzymes to delete their CCR5 gene, and then re-introduced into the body. A poster at the CROI conference described the enhancement of this technique using cyclophosphamide, an immune-suppressant drug designed to reduce the proportion of unaltered T-cells in the body prior to the engraftment of the modified cells

In one study of five volunteers who received cyclophosphamide and CCR5-deleted CD4 cells, large increases in CD4 count and the maintenance of a stable fraction of CCR5-negative cells were seen. All study participants were then taken off ART and in four cases have remained off ART without significant immune deterioration for 40 weeks.

In one participant who received cyclophosphamide and CCR5-deleted CD4 and CD8 cells and who then stopped taking ART, his viral load became undetectable and remains so, while in another participant, HIV only became detectable after a delay of two months. All three of these participants also remain off ART.

Matt Sharp, one of Sangamo’s original trial participants, was a speaker at the cure symposium and told Aidsmap that his CD4 count remains twice as high as it was before receiving CCR5-deleted cells, five years after the experiment.

PD-1 inhibitors

A number of other approaches were described by presenters. There was the idea of activating the long-lived, latently infected cells in the ‘reservoir’ of HIV with the HDAC inhibitors, cancer drugs like panobinostat, in the hope that the immune system would then ‘see’ them and kill them off. Unfortunately, although the cancer drugs certainly seem to flush some of the HIV-infected T-memory cells out of hiding, these cells are deficient in the signal molecules that alert the immune system to their presence, so are not easily picked off.

One of the most interesting presentations was by Stephen Mason of drug company Bristol-Myers Squibb on drugs designed to block the action of a cellular signalling molecule called PD-1. This is already a target for cancer drugs.

‘PD’ stands for ‘programmed death’: this alarming-sounding name is due to the fact that PD-1 receptor molecules accumulate on the surface of activated T-lymphocytes as they mature and, after a certain point, cells with abundant PD-1 on their surface show ‘anergy’: they no longer respond to antigens (foreign substances) and eventually die. A PD-1 blocker could revive and extend the life of the exhausted HIV-specific CD8 cells that seem to be an important part of the immune response in many ‘controllers’, people who manage to control their viral load without antiretroviral therapy (ART).

But PD-1 has a dual role. In regulatory CD8 cells, which are at an earlier stage of the CD8-cell life cycle, it promotes rather than suppresses their activity. What these cells do is to stop the immune system over-reacting to foreign substances. In the case of HIV, it is a crucial step in the development of latency: the phenomenon whereby HIV-infected cells stop actively producing virus and become quiescent, thus establishing the long-lived ‘reservoir’ of HIV-infected cells.

A PD-1 blocker could in theory aid a cure in two ways. Firstly it could prevent the reservoir of latent cells developing in the first place. It could therefore have a role in treatment in early infection or, like HDAC inhibitors, reverse the latency of the long-lasting reservoir cells. Secondly, it could maintain a functional population of activated HIV-specific T-cells that could pick off activated reservoir cells.

Experiments in monkeys not taking ART had demonstrated that PD-1 blockers could produce a temporary dip in viral load and that whereas four out of five monkeys given an inactive antibody died within four months of receiving it, four given a monoclonal antibody against PD-1 were all alive five months after their injections.

Mason showed the results of experiments with a monoclonal-antibody PD-1 blocker, BMS 936559, in eight monkeys that were receiving ART but that were taken off it after receiving the anti-PD-1 drug. Results were interestingly mixed. Two of the eight showed no response at all with their viral load rebounding to pre-ART levels. Two showed an initial viral rebound but then a slow drop in viral load from about 100,000 copies/ml to around a thousand. In two more, viral load stayed at approximately a thousand copies/ml after treatment was interrupted. And in the final two, viral load stayed undetectable for more than two months and then rebounded to levels of 50 to 500 copies/ml.

In all but two of the monkeys given anti-PD-1, but not those given a control antibody, there was no increase in integrated viral DNA within immune cells in the blood or gut, an important sign that the PD-1 inhibitor might have been helping to prevent the ‘re-seeding’ of the viral reservoir. Mason commented that the antibodies used in these experiments were originally adapted from human antibodies and might not be the ideal ones for monkeys: potentially, results in humans may be better.

He also showed experiments in mice that showed that a therapeutic vaccine and a PD-1 inhibitor could complement each other, bringing the viral load in these genetically ‘humanised’ mice down to undetectable levels.

Getting into the T-cell sanctuary

A combination approach will certainly be needed, Mason added, as studies of individual cure ideas tended to show the same pattern: response in some but not in all people, either only in patients already favoured with lower-than-average viral loads, or for no apparent reason. This seems to apply whether the treatment concerned is a broadly neutralising antibody, a latency inhibitor, or a therapeutic vaccine.

Afam Okoye of the Vaccine and Gene Therapy Institute of Oregon Health Sciences University examined why the reservoir of latently infected cells is so resistant to surveillance and destruction by the immune system. It has been discovered that HIV persists in a small subset of T-memory cells called T-follicular memory CD4 cells. These are located solely in the follicles of lymph nodes where B-cells, the cells that produce antibodies, mature. The CD8 cells that would normally destroy virally infected cells are unable to enter lymph node follicles and thus the latently infected reservoir is literally kept physically sequestered from destruction.

Okoye said that one way forward might be the therapeutic use of the CMV-based replicating-vector vaccine that had created considerable interest at the last two HIV vaccine meetings (Okoye works for the same institution as Louis Picker, who designed the vaccine). This drove the immune system to generate such a broad variety of anti-HIV CD8 responses that it is hypothesised that they might be able to overcome the protection offered by lymph nodes to the reservoir cells. Currently experiments are underway in two sets of SIV-infected monkeys – some treated very soon after infection and others treated somewhat later – who have also been given the CMV vaccine to find out if, when ART is stopped this summer, they manage to maintain an undetectable viral load.

The role of innate immunity

During the CROI conference itself, there was an interesting presentation about patients in the VISCONTI cohort. This group of 14 patients all started ART within ten weeks of HIV infection but subsequently stopped it an average of three years later and have not had to resume it. They have now maintained undetectable viral loads off treatment for more than eight years.

Scientists have found that VISCONTI patients have particularly active natural killer cells (NK cells). These cells form part of the most primitive but fastest-acting part of the immune system, the innate immune system, which reacts powerfully but non-specifically to foreign invasion. The advantage of NK cells that react to HIV could be that they themselves are not infected by HIV; therefore the characteristic problem of HIV vaccine and cure research – namely that attempts to increase anti-HIV immunity in the body only produces more cells for HIV to infect – does not apply to them.

HIV elite controllers, who are able to maintain low or undetectable viral loads from the start of HIV infection, have NK cells that are particularly good at killing HIV-infected cells. This was not the case with the VISCONTI patients, but instead their NK cells secreted unusually large amounts of interferon-gamma, a cytokine (immune-signalling protein) that attracts the attention of the large cell-eating cells called macrophages to HIV-infected cells and gets them to destroy them.

NK cells taken from VISCONTI patients were also far more able to control HIV production by the patients’ own HIV-infected CD4 cells when they were extracted and cultured in the test tube.

This is the first time a particular kind of NK cell has been described in so-called “post-treatment controllers”. They are however still only at the test-tube stage and need to be replicated in animals and live subjects to see if the high anti-HIV functionality of their NK cells could be induced by a vaccine.