CD8 cell fitness predicts who benefits most from antibody-based cure strategies

Over the last few years a number of studies aiming to cure HIV or at least achieve long-term viral suppression off antiretroviral therapy (ART) have produced tantalising results. These studies have given participants one or both of the long-acting versions of two broadly neutralising antibodies (bnAbs), 3BNC117 (teropavimab) and 10-1074 (zinlirvimab), sometimes in conjunction with immune boosting oral drugs.

These studies have produced interestingly similar results, according to a paper by Dr Zahra Kiani and colleagues at the Ragon Institute at Harvard that analysed four of them. (There are more details of the four studies in the references section).

Normally, if someone with HIV stops taking ART, the virus becomes detectable (‘rebounds’) within two to four weeks. Only 4% of people taken off ART do not rebound or have a delayed rebound – so-called Post-Treatment Controllers (PTCs).

But in these four studies a high proportion, often a majority, of participants given the bnAbs have taken much longer to rebound – in the order of 3-5 months. These were dubbed Post-Intervention Controllers (PICs). In addition, in several studies one or two participants each have experienced, and are still experiencing, much longer periods – sometimes several years – with no viral rebound.

The question now dominating cure researchers’ minds is: what factors contribute to these longer periods of HIV control without drugs (‘remissions’)? And why did a minority in each study not respond to the bnAbs at all? If scientists could answer these questions, they could fine-tune cure therapies to make long-term remission the norm rather than the exception.

Kiani’s study looked at the immune responses to HIV seen in participants in the four previously published studies. They compared responses in people who did, and did not, experience delayed rebounds, especially the cellular immune responses seen in CD8 cells.

CD8 cells include cytotoxic T-lymphocytes (CTLs) – cells that can detect and kill cells infected with viruses, including HIV. In the case of HIV, however, the virus’s infection of the related CD4 (T-helper) cells drives the immune system into a chronic state of over-activation: this causes the exhaustion and death of CD8 cells.

So strategies to induce a more effective cellular immune response to HIV, capable of suppressing HIV production in the absence of ART, walk a tightrope. Boosting the immune response to HIV too much results in there being more CD4 cells for HIV to infect and exhausts the CD8 cells. But the immune response does have to be strong and prompt enough to nip any new signs of viral rebound in the bud. And this implies, or so that the designers of the study theorised, that the response to HIV must be adaptable enough to get round its ability to outwit the body’s usual antiviral responses.

In people who did respond to bnAb treatment – experiencing a delayed rather than immediate rebound (Post-Intervention Controllers or PICs), their CD8 cells had preserved their ability to proliferate and respond afresh to new waves of HIV replication. The researchers called this capacity “stemness”, as the cells somewhat resembled stem cells newly matured in the thymus and lymph nodes. They contrasted them with the cells seen in people who didn’t respond to the bnAbs (Post-Intervention Non-Controllers or PINCs), which appeared exhausted or over-specialised.

Kiani’s research shows that PICs already had CD8 cells with greater proliferative capacity than PINCs. Adding the antibodies increased this capacity by 30% – but in the non-controllers too.

They then however tested the reactivity of subjects’ CD8 cells when tested against fragments of HIV proteins known to trigger immune responses. They found that the CD8 cells of responders were positive for two cell-surface receptor molecules called CD45RA and CD62L – the non-responders lacked them.

CD45RA is a marker for cells that are in a quiescent, rather than fully activated state – but which can proliferate upon stimulation with viral antigens (immunogens). In contrast cells lacking this are already in a fully differentiated state, meaning they have exhausted their ability to respond further to new pulses of viral replication. They’re ‘burnt out’. CD62L, meanwhile, enables CD8 cells to travel to lymph nodes, where they experience further maturation.

A deeper genetic probe of the cells showed that the cells of responders also carried receptors called CCR7 (which does a similar job to CD62L), CD27 (which, importantly, is also carried by B-cells, which make antibodies) and a crucial molecule called TCF-7 (or TCF-1), which is, to quote the researchers, “vital for the self-renewal and generation of memory T-cells”, i.e. ones newly sensitised to viral antigens. Flow cytometry diagrams showed that they had a very different profile to the cells in non-responders.

Crucially, these cell-proliferation markers were largely present before participants were taken off ART and so were not generated by the reappearance of HIV.

Adding a therapeutic vaccine

If people given bnAbs can stay off ART longer if they already have less specialised, more responsive CD8 cells, could a therapeutic vaccine prime the immune system to produce them? Such a vaccine would be designed to enhance the proliferation of these cells around the time ART is stopped. The hypothesis is that the viral suppression, and therefore time off ART, brought about by bnAb administration could be further prolonged if there was already a population of CD8 cells ready to suppress viral rebound.

This is what the second study, by Dr Michel Peluso and colleagues from the University of California San Francisco, set out to discover.

As this was a phase I study, it only had ten participants. They were first given a therapeutic vaccination consisting of four shots of different parts of HIV’s genome over a five-month period, followed by two bnAbs – 10-1074 plus a different second bnAb called VRC07-523LS. Over the following ten weeks they added a weekly dose of the immune boosting drug lefitolimod. They then added a second dose of the bnAbs and stopped ART.

Seven of the participants (70%) experienced prolonged remission from viral rebound. One, who is still undetectable after 18 months off ART, was exceptional in that they started PrEP almost on the day they acquired HIV and began ART shortly after, so had a low viral load from the start. The other six, however, had CD8 responses remarkably similar to those seen in the previous studies, while the three non-responders did not.

The median length of time to viral rebound (defined as two consecutive viral loads over 30) was four months from stopping ART, but participants did not mean resume ART immediately, as the study’s restart criteria stipulated persistently higher viral loads than this. Of the six responders who rebounded, two had a dramatically slower rebound than non-responders, while the other four’s viral loads bounced between 100 and 1000 copies over the next four to six months. This meant that the median time to resuming ART was 8.5 months among the six, and the maximum time 18 months.

The study was complicated by two participants who had large but temporary rises in liver enzymes in response to the antibodies – necessitating pauses in their dosing. Another complication was COVID: the researchers re-introduced ART as a precaution even in some participants who might have maintained a low viral load off it.

Detailed studies of the CD8 cells of responders showed that they started responding to the presence of HIV during ART interruption, even before it became detectable on viral load tests. Many of these changes were seen in non-responders too, though. But two cell-surface molecules were notably higher in responders’ CD8 cells. One, Ki67, is a key marker of cellular proliferation, also seen in studies of cellular immunity against cancers. The other was TCF-1, the T-cell renewal and proliferation guidance molecule also seen among responders in the Kiani study.

In a commentary on the two studies, Harvard University’s Dr Jonathan Li called TCF-1 “a transcription factor that maintains the cells’ capacity to proliferate and differentiate”, but also commented that the Peluso study “hints at the complexity of inducing such responses”.

What the two studies appear to confirm, however, is that a cure for HIV, or at least a therapy that could induce long-term control off ART, may require a synergy between different branches of the immune system all acting in concert to bring about long-term viral suppression in more people.

References