An animal study (Shytaj) in which standard antiretroviral therapy (ART) was intensified by adding in drugs from other classes has shown signs that it may be possible, using antiretroviral drugs (ARVs) alone, to produce a permanent reduction in the pool of chronically infected ‘reservoir’ cells that are the source of the HIV that reappears when ART is stopped. Reducing the size of this reservoir is seen by many researchers as a crucial component of a possible cure for HIV infection.
The finding, by a team at the Istituto Superiore di Sanità in Rome, was unexpected: they had been testing whether a new monkey immunodeficiency virus called SIVmac251 more accurately mimicked both the pathogenicity and ARV susceptibility of human HIV than previously used laboratory viruses. The investigators hypothesise that the entry-inhibitor drug maraviroc (Celsentri), which blocks ongoing cellular infection by HIV, may be an essential part of this more profound viral suppression.
Their study was part of a series of studies which, in 2011, produced a prominent decline in the number of chronically infected reservoir cells by combining the five-drug ARV therapy with a gold-containing anti-inflammatory drug called auranofin (Lewis).
The current study was designed to look at the viral decay dynamics (rate of viral load decline) in rhesus macaques given an ARV regimen which was intensified by stages. Eleven monkeys were used in a series of experiments.
One of the features of this study is that the monkeys concerned had all been infected with SIVmac251 for 18 months before therapy was initiated, and several monkeys were already showing signs of CD4 cell loss, with counts in the 400 to 500 cells/mm3 range. In many previous monkey studies, therapy was started soon after the animals were infected.
Initially, four monkeys were put on an escalating ART regimen. This started with tenofovir, FTC and raltegravir; a month later, boosted darunavir was added, and then six weeks after that, maraviroc. After this pilot phase, eight more monkeys were added: two on the four-drug regimen with darunavir; four on the full five-drug regimen with maraviroc; and two that were treated with boosted maraviroc alone for two weeks, followed by the other four drugs.
Two monkeys were then further selected for structured-treatment-interruption (STI) experiments. One of these was one of the four animals given escalating ART. This monkey’s regimen was then additionally intensified for three months with auranofin, after which it was taken off all drugs for a month, given two months back on the five-drug regimen without auranofin, and then taken off all therapy permanently. The other had previously been in the auranofin study, which included a similar escalating-ART regimen but with auranofin. It was then taken off ART for a month, given two months on the five-drug regimen, taken off again for four months, given a final two months of therapy, then taken off permanently.
The first observation was that the addition of maraviroc further suppressed viral replication beyond that achievable by the four-drug regimen. SIVmac251 is more difficult to control virologically than HIV is in humans (because monkey immune cells produce about 20 times more viral particles) and the tenofovir/FTC/raltegravir regimen only suppressed viral load in one monkey out of the original four and, with added darunavir/r, in three out of four. All monkeys achieved undetectability on maraviroc. Adding in maraviroc increased viral load suppression from about 1.5 logs (a fifty-fold reduction in viral load) to about 3 logs (an 800-fold reduction).
In one case (a monkey in which treatment persistently failed to suppress HIV), the animal was found to have a higher viral load in its cerebrospinal fluid (CSF) than in its blood. It finally reached undetectability (in both blood and CSF) when the ritonavir-boosting dose was doubled: ritonavir facilitates drug transport over the blood-brain barrier. This finding provides interesting support for the hypothesis that a number of cases of treatment failure are caused by unsuppressed HIV in the central nervous system.
Secondly and unexpectedly, it was found that adding in maraviroc produced a continued decline in viral DNA – a measure of the number of infected cells rather than free virus. Normally, even on fully suppressive ART, there is a residual viral load that stays fairly constant. The investigators found that, under the five-drug regimen, there was a slow but continued decline in viral DNA over 200 days of five-drug therapy, indicating that the proportion of circulating lymphocytes that were HIV-infected declined from 20 per million to four per million.
Thirdly, measurements in three monkeys of the subsets of T-cells that comprise the chronically infected reservoir of cells – the central-memory and effector-memory cells – found that, in two out of three cases, the proportion of these cells relative to the total T-cell population declined over a period of four months, whereas naive T-cells, a subset not thought to be part of the reservoir, did not.
Fourthly, in four monkeys that were taken off therapy, it was found that the viral load ‘set point’ – the average viral load maintained off therapy – was lower than it was before therapy, in three out of four cases by about one log (ten times lower). The only experiments in humans that have managed to produce a permanently lower set point were in people treated in very early infection.
Finally, the two monkeys subjected to treatment interruptions not only maintained much lower viral loads when taken off therapy, but intermittently had undetectable HIV. The monkey in the original study that included auranofin, which was given three treatment interruptions, initially had a viral load of two million copies/ml but, after the second treatment interruption, maintained one of about 800 copies/ml, varying between undetectable and a couple of peaks of about 10,000 copies/ml.
The second, which had escalating ART, then three months on five drugs and auranofin, then two treatment interruptions, had a pre-treatment viral load of about 80,000 copies/ml and a post-treatment viral load that averaged about 250 copies/ml, varying between undetectability and about 4000 copies/ml. Neither showed any signs of breakthrough to higher viral loads within a time period of about 200 days.
Andrea Savarino, one of the investigators, told aidsmap that the team is now optimising the treatment protocol in macaques. When asked if there would, in his opinion, be benefit in adding maraviroc to ART regimens in humans right now, he said: "Yes, and human studies point in the same direction."
Shytaj IL et al. A highly intensified ART regimen induces long-term viral suppression and restriction of the viral reservoir in a simian AIDS model. PLoS Pathogens8(6): e1002774. doi:10.1371/journal.ppat.1002774. 2012. View open-access article here.
Lewis MG et al. Gold drug auranofin restricts the viral reservoir in the monkey AIDS model and induces containment of viral load following ART suspension. AIDS 25(11):1347-56. 2011. (View free abstract here.)
