Gene therapy studies show potential for HIV control without drugs

Gene therapy approaches that involve the genetic modification of human haematopoietic stem cells have the potental to engineer HIV control by introducing cells resistant to HIV infection, the 20^th Conference on Retroviruses and Opportunistic Infections (CROI 2013) heard last week.

The proof of concept for this approach to HIV control in humans comes from the case of the 'Berlin patient', who received a bone marrow transplant from a donor with natural resistance to HIV infection. The donor was homozygous for the CCR5 delta 32 mutation, meaning that cells potentially vulnerable to HIV infection would never display the CCR5 receptor necessary for HIV to gain entry to that cell. As a result of complete ablation of the recipient's own stem cells by chemotherapy, his CD4 cells were replaced by cells derived from the donor's CCR5-lacking population. Over three years after the procedure the recipient remains HIV-free without antiretroviral treatment, and has been described as "functionally cured" by physicians.

However, the chance of reproducing this outcome using transfer of donor cells is very low due to the shortage of potential donors who are both HLA-compatible (essential for avoidance of graft-versus-host disease) and CCR5-delta32 homozygous. In any case, the essential elements that contributed to this functional cure are still not fully understood.

The introduction of modified genes into stem cells harvested from a person's bone marrow will be a necessarily individualised treatment, but the costs of gene therapy are likely to come down in the future, and if experimental approaches prove successful in controlling HIV without antiretroviral drugs, gene therapies may deliver a cost-effective form of treatment in the future.

The first human studies of gene therapy have sought to modify the expression of the CCR5 receptor. Sangamo Biosciences is developing a zinc finger nuclease which prevents CCR5 expression; study results have been presented at a number of conferences including ICAAC 2011 and CROI 2013 showing that the procedure is safe and that it results in long-term gains in CD4 cell numbers in people also taking antiretroviral therapy.

Two papers presented at CROI 2013 described the use of modified immune cells that make their own peptides (short chains of amino acids) and act as fusion inhibitors, similar to the drug T-20 (enfuvirtide, Fuzeon). In one experiment these cells, which essentially made their own anti-HIV drug, acted as a treatment or therapeutic vaccine, curbing the viral load of a laboratory virus engineered to be far more virulent than human HIV; in the other, they prevented infection by a similar virus.

In the first study, rhesus monkey stem cells, the progenitors of T-cells and all immune cells, were genetically altered to express a fusion inhibitor peptide called mC46. While not able to prevent infection in monkeys challenged with a highly pathogenic monkey/human SHIV (a laboratory-manufactured virus combining parts of both the human and simian immunodeficiency virus genomes, created for research purposes), it did produce infections characterised by viral loads 2.5 and 3.15 logs lower than in the control group of monkeys.

In another study, human CD4 cells had a fusion-inhibitor peptide called C34 attached to their co-receptors (CCR5 or CXCR4). These were able to resist infection by HIV in the test tube.

Fusion-inhibitor cells raise CD4 count and bring down viral load in monkeys

In the first paper, researchers from the Fred Hutchinson Cancer Research Institute in Seattle took haematopoeic stem cells (HSCs), bone-marrow cells that are the progenitors of all blood cells, from two pigtail macaque monkeys and transformed them genetically by splicing an inserted gene sequence for mC46 into their CCR5 receptor gene.

They then injected the cells back into the monkeys. One monkey had 20%of its HSCs replaced by the C34-producing cells and the second had over 50% replaced.

A week later, they infected them and two control monkeys with a particularly lethal strain of genetically engineered human/monkey SHIV, which destroys CD4 cells fast and usually develops a steady-state viral load in the order of several million copies/ml.

In the control animals, their CD4 counts declined from around 600 cells/mm³ before infection to between 10 and 50 cells/mm³ within two to three weeks. In the monkeys with mC46, the CD4 cell count dipped to about 100 cells/mm³ within two weeks of infection, but then rose slowly back to pre-infection levels over the next six months.

At the time of highest SHIV viral load and fewest CD4 cells, 90% of the CD4 cells in the mC46 monkeys had the fusion-inhibitor-generating insert in them, which is what one would expect, given that SHIV so decimates non-mutated CD4 cells.

What was unexpected to the scientists, though, was that after the period of peak viral load, the non-mutated CD4 cells made a partial recovery – in one monkey to about 60% of all CD4 cells and in the other about 20%. This is promising, as it shows that one would not need to replace all or even most of the CD4 cells in the body with HIV-resistant ones in order to contain an HIV infection, and that increasing numbers of non-resistant cells does not lead to a new burst of virus.

This may also mean that it would not be necessary to take the dangerous step of having to destroy a person’s immune system with whole-body radiation, as happened to the 'Berlin patient' Timothy Ray Brown, in order for the new cells to repopulate the immune system.

As we said, SHIV reproduces furiously and peak viral load in all monkeys ten days after infection was one billion copies/ml. After that, viral load in the control monkeys declined to about half a million in one and about ten million in the other. In the mC46 monkeys, it fell to about 100,000 in the monkey with 20% of its cells replaced by mC46 cells and down to a few hundred in the one with more than 50% of its cells replaced.

Viral load was about 320-fold lower (2.5 logs) in the first monkey and about 1400-fold (3.15 logs) lower in the second. This would transform a typical HIV viral load in an untreated human of 70,000 copies/ml to 50 copies/ml before any ARVs were taken.

Although there was a partial recovery of unmutated CD4 cells in the blood, memory cells in the lymph nodes that form the ‘reservoir’ of proviral HIV DNA remained predominantly the HIV-resistant mC46  cells, which is exactly where one would want them to be.

Fusion-inhibitor human cells resist viral infection in test tube

The second experiment was a test-tube one, using human CD4 cell cultures. Researchers from the University of Pennsylvania physically attached another fusion inhibitor peptide called C34 to the CCR5 or CXCR4 co-receptors. They found that the cells could not be infected with HIV.

Interestingly, although HIV attaches preferentially to one or other of these co-receptors, they found that the action of the fusion inhibitor peptide was non-specific: whichever receptor they attached it to, whether in cells with one or both co-receptors, the cells became resistant to infection, so the technique could work with HIV of R5 or X4 type. Its action was specific to the co-receptors, though: attaching it to the CD4 molecule produced no effect.

The co-receptors on the cells could still do their biological job of attaching to the immune-activation molecules (MIP-1 beta and SDF-1) that they attach to in the body, which indicates less likelihood of toxicity.

The researchers now hope to produce genetically altered cells that produce the fusion inhibitor peptide as part of their co-receptors, as in the previous study, and to move into animal experiments.