Paradoxes, Pitfalls Persist in HIV Remission Research

AMSTERDAM — HIV DNA levels were similar whether patients recently diagnosed with HIV were treated with a kick-and-kill approach that involved a vaccine plus a latency-reversing agent or treated with a traditional antiretroviral therapy, findings from a new study show.

"The results are definitive but disappointing," said Sarah Fidler, PhD, from Imperial College London, who is chief investigator of the Researchers in Viral Eradication of HIV Reservoirs (RIVER) study.

"I don't think this means that the kick-and-kill approach does not work. But I think we have shown that using the two interventions in this trial — that doesn't work," she said here at the International AIDS Conference 2018.

The 60 men enrolled in the RIVER study were recently diagnosed with HIV and recruited from six sites in the United Kingdom.

All received antiretroviral therapy for about 24 weeks to bring their viral loads to undetectable levels, and then were randomized to one of two treatment groups. Half the patients received agents designed to awaken the HIV viral reservoir (the "kick" in the kick-and-kill approach) and stimulate T-cells to attack the virus (the "kill"). The other half, who served as the control group, received traditional HIV treatment.

The kick agent in the RIVER study was vorinostat (Zolinza, Merck), a chemotherapy designed to target T-cell lymphoma, and the kill agent was a chimeric universal HIV prime–booster vaccine combination.

Counting HIV DNA

The men in the kick-and-kill group received the prime vaccine immediately and oral vorinostat every 72 hours, for a total of 10 doses over 30 days. The booster was administered 8 weeks after the initial vaccine.

The researchers assessed HIV DNA levels at weeks 16 and 18, but did not interrupt treatment to see if the virus was controlled on its own.

This was a strategic choice, Fidler explained. Previous research has shown that immunologic biomarkers can be used to predict time to viral rebound (Nat Commun. 2015;6:8495).

Fidler displayed a scatter plot with HIV DNA levels at enrolment, randomization, and weeks 8, 12, 16, and 18. It showed that individual participants' HIV DNA levels dropped over time, from a mean log10 HIV DNA of 3.84 copies/mL to 3.04 copies/mL.

"Without randomization," she said, "we'd be able to say that we saw a significant effect in terms of change in HIV DNA levels.

She then showed a plot with results broken down by treatment group, which clearly showed no significant difference in HIV DNA levels at any of the time points.

Active but Unsuccessful

This kick-and-kill approach did not work, Fidler announced.

However, when the team conducted subanalyses to assess the ability of vorinostat to inhibit acetylation, which is what it is meant to do, and the ability of CD8 cells to recognize HIV, a different story emerged.

In vitro assays of samples from the study participants showed that the agents did exactly what they were supposed to.

It is not clear why the overall strategy did not work, said Fidler, adding that future substudies of inflammation and other markers could show a path forward.

"It may be that the latency-reversing agent here is not potent enough," she explained. "It may be that the vaccination we used was not recognizing the correct epitopes that rest on reservoir cells."

Whatever it is, she said, she hopes the data spur more innovation and do not permanently close the door on kick-and-kill.

Opting In and Calling Out

The RIVER results mirror findings from the BCN01 study, which was presented at the 2017 Conference on Retroviruses and Opportunistic Infections (abstract 119LB), said Javier Martinez-Picado, PhD, from the AIDS Research Institute irsiCaixa in Barcelona, Spain. It used the same prime–booster vaccine combination but a different latency-reversing agent.

In the 4-year BCN01 study, when treatment was interrupted so that researchers could assess how effective the agents were, viremia was controlled without antiretroviral therapy for 9 months in four of 14 patients.

The RIVER results are "disappointing," said Martinez-Picado, who was one of the BCN01 researchers. Still, he said he is encouraged by the fact that the immune response from CD8 cells differed in the treatment and control groups.

"I think there is a potential space to see some differences in treatment interruption," he added.

He was not the only member of the audience to express this opinion.

Interrupting Research

"We don't really have a good measurement of the reservoir, and we don't really know how to measure the immunologic correlates of control," said Steven Deeks, MD, from the University of California, San Francisco. "So we have no idea whether this worked or not."

The audience laughed, but Deeks followed up: "I think you are obliged to at least strongly consider — in order to learn as much as you can from all these participants who put themselves through this — doing a treatment interruption."

There was a pause and a murmur from the audience. At a preconference session on HIV cure, whether or not to interrupt treatment in cure studies was hotly debated.

"I knew you were going to say that," said Fidler, also to laughter.

It is possible that the RIVER team will use lab results to select specific participants for treatment interruption, she said. Or the team might want to preserve the very low viral reservoirs of participants and wait until a more potent combination of a vaccine and latency-reversing agent can be tested.

"I wonder, on the patient side — and I'd be very interested to hear what people living with HIV in the audience feel — if we want to have another study that's just going to show a bunch of rapidly rebounding individuals," she pointed out.

There were people with HIV in the audience, and one suggested allowing study participants to "opt in" to treatment interruption.

Then Timothy Brown, the so-called "Berlin patient," who was cured of HIV after a stem cell transplant, approached the mic.

"I chose to interrupt my treatment and it worked and I'm cured," he said. "I think it should be the patient's will, and there should be scientific ground behind that to help the patient make the decision."

Merck provided raltegravir and vorinostat for the study. GlaxoSmithKline contributed intellectual property through an academic–industry partnership. Fidler, Martinez-Picado, Deeks, and Brown have disclosed no relevant financial relationships.

International AIDS Conference 2018: Abstract:TUAA0202LB. Presented July 24, 2018.

Follow Medscape on Twitter @Medscape and Heather Boerner @HeatherBoerner

Comments

Commenting is limited to medical professionals. To comment please Log-in.

Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.