Researchers have introduced genetic changes into human T-cells
that make them virtually immune to HIV infection, and the viral load in
HIV-positive people who are inoculated with these cells returns more slowly
when they are taken off therapy, the recent Conference on Retroviruses and Opportunistic Infections (CROI 2019) heard.

The challenge now is to increase the proportion of cells in
the body that have the immunity gene added. This might possibly result in a functional
cure of HIV, with the virus unable replicate even in the absence of HIV drugs.

The top story at CROI 2019 was that a second
person may have been cured of HIV infection.
But most reports also emphasised
that the bone marrow transplant procedure that achieved both this cure, and the
previous one of Timothy Ray Brown 12 years ago, is dangerous, expensive and
not something that would ever be attempted in a patient who did not have

So can a similar feat be achieved more safely? Promising
results in animal experiments produced monkeys that appear to have most of the
HIV DNA removed from their cells

But a second study, this time in humans,
reproduced the genetic change that cured the two patients more exactly, and by
means of a safe and repeatable technique. The two cured patients had their T-cells replaced with ones
from donors with a genetic mutation called CCR5-delta 32 that means their cells
don’t have the CCR5 receptor molecule on their surface that most strains of HIV
need to attach to before they can infect a cell.

In the experiment presented by Dr Pablo Tebas of the
University of Pennsylvania in Philadelphia, T-cells taken from 15 people with
HIV were cultured in a lab dish with a gene-editing enzyme called a zinc
finger nuclease (ZFN). This modified their CCR5 gene so it changed to the
HIV-resistant variant. The cells were then reintroduced into the patients in
what is called an autologous transplant – i.e. the modified transplant comes
from the patient themselves, so there should be no problem with rejection.

This is not a new technology: ZFN enzymes were used to
modify cells in
experiments originally reported to CROI in 2011
. In
that experiment, a viral vector – the shell of a common-cold virus – was used
to introduce the ZFN enzyme into cells. The reintroduced CCR5-negative cells
initially made up about 22% of the T-cell population, but were replaced
over time by the patients’ own CCR5-positive cells.

The current set of experiments, however, used a different
technique called electroporation to get the ZFNs into cells. The problem with
using viral vectors is that the cells develop immunity to them so they can only
be used once. Electroporation can be used many times to create multiple cycles
of modified cells that can be inoculated. In this technique, the T-cells are
incubated alongside the ZFN enzymes in chambers in the presence of an electric
current. This causes cell walls to become permeable and the ZFNs or other
substances one wishes to introduce will diffuse into the cell.

As well as using a different
and repeatable technique to create autologous CCR5-negative T-cells, the experiments explored a
couple of other possibilities. Firstly, could the proportion of modified
T-cells in the body be increased if a low dose of an immune-suppressant drug (cyclophosphamide)
was used prior to their inoculation, to reduce the number of CCR5-positive
cells in the body? And secondly, while only 1% of people of northern European
ancestry have two copies of the CCR5-delta 32 gene, which means they are
virtually immune to HIV, a larger proportion – up to 20% – have one copy of
this gene (so-called heterozygosity). Would they have a better response?

There were 15 people in this study, in five groups of three.
Three people received no cyclophosphamide. Six people – three heterozygous for
CCR5 and three with no copies of the delta-32 mutation – received a lower dose
of cyclophosphamide, and six people received a higher dose. The doses were not
high enough to create significant toxicity and were about the same used to
treat the autoimmune disease lupus, Dr Tebas said.

The creation of the CCR5-negative T-cells took about ten
weeks, and the cyclophosphamide was given two days before the infusion of the
cells. Eight weeks later the patients, who were all HIV-positive people with
high CD4 counts (median CD4 count 831 copies/mm3), stopped their
antiretroviral therapy (ART). The analytic treatment interruption (ATI) was for
a fixed period of 16 weeks, though if people still had a low viral load
after that time, they had the option of extending the ATI. It was discovered that
one person in the CCR5-hetereozygous group in fact kept taking their ART and
was excluded from the analysis.

There were no significant adverse events related to the
study therapy. The single-dose electroporation technique was at least as
effective as using the adenovirus vector: immediately after infusion, 25% of
participants’ T-cells were CCR5 negative.

There were no cases of viral remission in this study: during the
ATI, the HIV viral load in all subjects reappeared and the proportion of
T-cells that were CCR5-negative slowly declined. There was a tendency for the
CCR5-heterozygous subjects to retain a higher proportion of CCR5-negative
inoculated cells. At four weeks, before the ATI, the proportion of T-cells that
were modified was 4.6% and 4.1% in the two homozygous (both genes
CCR5-positive) groups but 7.4% in the heterozygous group, and at week 24, four
weeks after the ATI, the proportions were 2.2%, 2.6% and 4.6% respectively.

The cyclophosphamide appeared to make no difference to the time to viral rebound.

The heterozygous patients’ viral loads also reappeared more
slowly. Whereas 50% of the homozygous patients already had a detectable viral
load four weeks into their ATI, it took all 12 weeks before three out of
five of the heterozygous patients did.

Three subjects had low viral loads at the end of the ATI and
decided to prolong it. One homozygous patient had an undetectable viral load at
the end of the ATI and delayed restarting until 12 weeks later, at a viral
load of 10,000 copes/ml. Two of the heterozygous patients had viral loads in
the 1000 copies/ml region and did not restart their ART till 20 and 32 weeks
after their ATI respectively, at viral loads of 8000 copies/ml.

So while this experiment produced no patients with prolonged
viral suppression, this is a demonstration of a safer, more repeatable and
non-toxic way of creating a population of HIV-resistant T-cells that can be
infused back into the body and which can to some extent delay HIV viral
rebound, even when they only form less than 10% of the body’s complement of

Asked about future steps, Dr Tebas said that instead of
simply repeating the same techniques to engineer CCR5-negative cells, his team
would investigate using ZFN technology to generate cells lacking the other HIV
co-receptor, C4CR4,  and eventually to
generate CRT T-cells – Chimeric Antigen Receptor cells, a type of artificially
engineered T-cell already being used in anti-cancer therapy which will be able
to seek out and destroy HIV infected T-cells.