Six of 24 people who received a series of immunomodulatory vaccines were able to keep HIV viral load below 2000 copies/mL 6 months after stopping HIV antiretroviral treatment.
This could be one of the first signals that lowering the set point of viral replication could be a key part of controlling HIV without regular treatment, said Steven Deeks, MD, from the University of California, San Francisco, who was not involved in the study. The finding reminded him of the early days of HIV drug development in the 1990s, he told Medscape Medical News.
“What we’re looking for right now is AZT,” but for a cure, said Deeks, referencing azidothymidine, the first HIV medicine to show clinical activity against HIV. “AZT by itself back in 1991 didn’t do much. But AZT with 3TC [lamivudine] and a protease inhibitor? That was a home run.
“Maybe this is the AZT of some future cure regimen.”
The data, from the AELIX-002 single-center study, was presented at the virtual Conference on Retroviruses and Opportunistic Infections 2021. This series of vaccines isn’t like the SARS-CoV-2 vaccines meant to prevent COVID-19. Instead, what researchers tested was whether a series of shots designed to beef up different CD4 and CD8 T-cell responses to HIV could help participants’ own immune systems slow viral replication in the absence of treatment.
To understand how the trial worked, first you have to understand how the immune system typically responds to the virus. It’s not that the immune systems of people living with HIV don’t respond at all to HIV; they simply don’t innately recognize that the virus is there. By the time they do, the response alone isn’t sufficient to stop viral replication in almost anyone. By then, HIV has usually replicated and stored itself in lymph and other reservoir cells, which is why you can’t cure HIV by simply eliminating new replication of the virus.
This is where the AELIX-002 trial and its immunomodulatory vaccines come in. In this trial, researchers recruited 45 people, mostly 30-something gay and bisexual men living with HIV in Spain. To qualify for the trial, participants had to have started HIV treatment early — in the first 6 months after diagnosis — and have experienced undetectable viral loads for at least a year. They also had to have healthy immune systems, marked by CD4 counts of more than 400 cells/mm3 for 6 months and a nadir CD4 count of at least 200 cells/mm3. In other words, they couldn’t qualify if they had experienced the immune collapse of AIDS. That’s because previous research has found that it’s easier to clear the viral reservoir when that reservoir isn’t terribly deep.
The vaccines, called HIVACAT T-cell Immunogens (HTIs) — named for the HIV vaccine research program HIVACAT — consist of one DNA-vector vaccine and two virus-vector vaccines. In this case, they were DNA.HTI, MVA.HTI, and ChAdOx1.HTI.
Essentially, the vaccines were designed to “reproduce the immune response that we see in individuals who spontaneously control HIV,” said Beatriz Mothe, MD, PhD, from the IrsiCaixa AIDS Research Institute and HIVACAT, who presented the study.
In this trial, participants were randomized, in a two-to-one ratio, to receive either the active combination or placebo injections. It was a rigorous regimen; participants received DNA.HTI shots at 0, 4, and 8 weeks, followed by MVA.HTI shots at 12 and 20 weeks. Then, participants were free from shots for 24 weeks. At the end of those 24 weeks, the clock reset, with people in the active group receiving ChAdOx1.HTI at the new 0 and 12 weeks. Finally, at 24 weeks, participants received the MVA.HTI for the last time.
All three of HIVACAT’s immunogens had been previously studied and shown to be safe, and they were safe in this trial, as well. There were no serious adverse events, just headaches and flu-like symptoms that cleared in days.
When they started taking samples of immune cells during this phase of the trial, they started to see signs that the vaccinations were doing something: After week 12, when treatment participants received their fourth shot and first VMA.HTI shot, researchers saw immune responses shoot up. And those responses “were boosted with subsequent vaccinations, compared to the placebo,” Mothe said.
“Importantly, these responses were highly specific toward HTI,” she added. And the responses were durable.
Other lab results looked good, too. When investigators looked at not just how specific the immune responses were but also at the breadth of those responses, they registered five different immune responses in people receiving the vaccines.
“I want to point out here that we took a conservative approach” when testing for immune activity, Mothe told the audience. “When computing a response, in the cases where two adjacent peptides were reactive, we assumed that they could be reactive to the same epitope.”
In other words, if they’d calculated the same data differently, they could have said that the breadth of activity was higher.
But these were just associations — a “signal” that the vaccines were working. They couldn’t prove that it was the vaccines that were inducing the immune response, Mothe said.
What they didn’t see was a specific pattern in the breadth of responses to the vaccines. This suggested to Mothe that “all regions were equally immunogenic.”
And, surprisingly, the results were better when they excluded people with biomarkers that are associated with better spontaneous viral control — those with HLA class 1 alleles. It was a finding that they couldn’t yet explain, she said.
Finally, researchers checked the viral reservoirs by testing both total proviral DNA and just the fragments that were replication-capable at screening and after participants received the last shots. In all participants, the reservoir dropped slightly. They dropped more in the vaccine group, but not significantly.
Then, 8 weeks after the last vaccination, it was time to test whether the vaccines would show any activity in controlling viral load. Participants stopped taking daily ART.
Analytical treatment interruption (ATI), as it’s called, is the only definitive way to see if the treatments have an effect, said Deeks. Three years ago, the idea of stopping treatment that might just “show a bunch of rapidly rebounding individuals” led to spirited disagreements between researchers.
But now, AELIX-002 participants were taking the plunge.
Researchers took CD4 and viral load tests weekly and returned participants to effective ART as soon as they showed any of four signs of rebounding infection: signs of acute retroviral syndrome; a single plasma viral load of 100,000 or more; 8 consecutive weeks of viral loads above 10,000 copies/mL; or two consecutive weeks of CD4 counts less than 350 cells/mm3.
In the first 2 to 3 weeks, all the participants’ viruses responded the same, showing no vaccine effect. HIV came roaring back, with plasma viral loads as high as 107 copies, or 10 million copies/mL. There was no difference in time to the first detectable viral load or the level of peak viremia between the groups.
But then the groups started to diverge. By 12 weeks, just four of those in the placebo group had low enough viral loads to continue off treatment, compared with 10 in the treatment group. Of those 10, eight didn’t have the protective alleles. By the end of the 6-month interruption, five of those had viral loads below 2000 copies/mL at the end of the trial.
This was still a high viral load, said Deeks — most studies only allow people to stay off treatment if they have viral loads between 200 and 1000 copies/mL — but it was a concrete metric researchers could use to see if they were making a difference in viral activity with their agents.
“It might have actually moved the dial,” he told Medscape Medical News. The data “provide tentatively, with nuances, proof of concept that with this immunogenic vaccine, you can safely affect viral load set point.”
Then, in the middle of the treatment interruption, the COVID-19 pandemic hit Spain, and the government implemented one of the more severe lockdown orders. The researchers scrambled to complete the study, said Mothe.
“We had to implement, very fast, a lot of measures to keep the trial going,” she said. That, she told the audience during a live Q&A, seemed to be more stressful for the participants than the actual treatment interruption.
Indeed, two of the eight people who’d received vaccines and still had manageable viral loads chose to reinitiate HIV treatment at 22 and 23 weeks, respectively. It wasn’t because of viral failure, though. It was, Mothe said, because of logistical complications related to the COVID-19 pandemic.
By the end of the study, the difference between the groups was not statistically significant (P = .1834), but it was, perhaps, an “efficacy signal” that researchers should pursue with other potential kick and kill agents.
“This is still very far away from changing the current clinical practice in the management of people living with HIV,” Mothe told Medscape Medical News. “But really, the data that we present show that the vaccine probably is a good design and probably is going to be, or could be, proposed as a T-cell backbone for any combination of a strategy that could combine vaccines with our immunomodulatory agents.”
“A Lot to Learn”
The team is now following five of the remaining vaccine recipients and one spontaneous controller — none of whom have the HLA allele — in an extension trial. More than a year later, all continue to have viral loads below 2000 copies/mL.
Meanwhile, researchers have begun recruiting for AELIX-003, a study that expands the patient group to include people who didn’t start treatment quite so early. They are testing HTI vaccines in combination with a TLR7 agonist to wake up the HIV reservoirs and, potentially, with broadly neutralizing antibodies, which may provide the kill part for the kick and kill.
The data piqued the interest of the researchers present. Katharine Bar, MD, assistant professor of medicine at the Perelman School of Medicine at the University of Pennsylvania in Philadelphia, called the study a “logistically impressive feat” and a model for how studies can continue in the midst of COVID, and how to conduct an ATI.
“The correlates you saw in your study with intact reservoirs being lower, and those who were able to maintain off antiretroviral treatment are quite interesting, as well as the association between immune responses,” she said during the discussion. “I think we could learn a lot from your study.”
Conference on Retroviruses and Opportunistic Infections (CROI) 2021: Abstract 161. Presented March 10, 2021.
Heather Boerner is a science and medical reporter based in Pittsburgh, and can be found on Twitter at @HeatherBoerner. Her book, Positively Negative: Love, Sex, and Science’s Surprising Victory Over HIV, came out in 2014.