Imagine seeing a patient every eight weeks for a few quick injections into the buttocks that could prevent HIV between appointments.
That's the promise of the ÉCLAIR trial of the investigational integrase strand transfer inhibitor cabotegravir, which had its phase-2 results unveiled at CROI 2016 in February. ÉCLAIR was one of two trials of long-acting injectable HIV-prevention programs presented at the conference.
"Cabotegravir in oral and long-acting injections were well tolerated," said Martin Markowitz, M.D., the trial's primary investigator, in his presentation at CROI. "We feel this does permit continued development of cabotegravir for PrEP [pre-exposure prophylaxis]."
Designing Long-Acting Agents
This study was a phase-2 trial -- that is, it wasn't designed to figure out how effective the drug was in preventing HIV acquisition in the mostly gay and bisexual men who participated in the trial. It had more fundamental things to figure out first: whether it was safe for human use, what dose would maintain effectiveness between doctors' visits and whether men liked the approach enough to take it.
On all counts, the answer appears to be, "Yes."
The trial randomized the 127 men into two groups. One received a placebo; the other received cabotegravir. Because there would be no way to remove the drug from a participant who had an adverse reaction after receiving a shot, all participants were first dosed via daily pills. That way, if they had adverse reactions or their lab results changed in a concerning way, they could be safely removed from the trial. Seven participants were dropped at this point, due to neutropenia and increased blood creatine phosphokinase. At CROI, Markowitz attributed concern about elevated levels in one Asian-American and two African-American members of the cohort to excessive diligence.
"There was a lot of data not known about giving cabotegravir long acting, so we had a low bar to discontinue patients with changes in white counts, etc.," Markowitz said during his presentation. "Those turned out to be a red herring. Knowing what I know now, I would not have discontinued them. I would have had those patients get dosed [with the long-acting shot]."
Ouch, But OK: Safety
By safety standards, cabotegravir performed well. While reports of adverse events were high -- 98% among participants receiving intramuscular injections of 800mg of cabotegravir every 12 weeks and 90% among patients receiving placebo saline injections -- the majority were minor. They included pain, swelling and itching at the injection site. Seven men in the cabotegravir group experienced fever too. And when participants did experience pain, it was longer lasting in the cabotegravir group, at about 5.4 days of sustained pain after injection.
"It's really interesting to note, and maybe even a little surprising, that the actual act of injecting saline elicited complaints of injection-site pain in 27% of subjects that lasted for about two days," said Markowitz at CROI. "That suggests that there's some background here, that there's some injection pain caused by the act of injecting."
However, he noted that half of participants reported their maximum level of pain as mild. Only about 10% reported their pain levels as being severe.
Four men discontinued the study, citing injection intolerability.
Men, Not Macaques: Surprise Absorption Rates
At CROI, Markowitz presented the audience with a graph. On it were two lines. One showed the plasma cabotegravir levels that afforded 97% protection in earlier studies of rhesus macaques. The second showed the level that afforded 100% protection.
Then he loaded a model. In it, the plasma level of cabotegravir longer-acting rose immediately after injection, slowly declined until the next injection and then spiked again. The intent was to dose participants with enough cabotegravir that effectiveness would never drop below 97%.
But those weren't the results. At CROI, Markowitz layered a third spike-and-trough line on top of the model. This one spiked far higher after dosing and dropped off much more precipitously after dosing, often dropping below effective levels. The findings? About 70% of participants during the injection phase had trough or nadir drug levels below full efficacy. And 15% to 31% of participants at some point during the injection phase had trough levels that wouldn't afford protection from HIV.
So perhaps it should be no surprise that one of the two seroconversions that happened during the trial happened in the cabotegravir arm during the follow-up phase of the trial. This happened around the time that the participant's drug levels were undetectable, said Markowitz.
What this means, essentially, is that men are not macaques -- and that cabotegravir metabolizes differently in human bodies than in our simian counterparts.
"You can see quite clearly, first, that the peaks are higher at each injection visit, consistent with more rapid absorption of drug from the depot," Markowitz said, pointing at the slide. "And the troughs are lower after each injection, consistent with more rapid release of drug from the depot."
For his part, Andrew Owen, Ph.D., wasn't surprised that the absorption rates were so different in men than they were in pre-clinical studies of macaques. Owen, a professor of molecular and clinical pharmacology at the University of Liverpool, has been studying HIV drug development for years, and recently received a grant from the National Institutes of Health to develop long-acting antiretroviral formulations for HIV treatment. So he went to the literature to see what's already known about how long-acting injectables work for other conditions, such as schizophrenia and contraception.
What he found turned into a literature review on long-acting injectables published in the journal Advanced Drug Delivery Reviews in February.
"I was surprised by how little is actually known about how the drug makes it into the blood stream after administration of these formulations," he said. "There are a lot of unknowns around drug release from the depot. One would expect clearance of drug to be predictable from data with the oral form and early long-acting trials. However, higher maximum concentrations and lower minimum concentrations than predicted were observed in the ÉCLAIR study, which confirms we have much more to learn about drug release from the depot."
This difference in absorption and clearance may be part of the reason why cabotegravir levels shot higher than models and then dropped far below efficacious levels in participants receiving the shots. Markowitz said the team is now considering a dosing regimen that would happen every eight weeks instead of every 12 to account for the quicker drug absorption in humans.
Coming Back for More: Acceptability
To Mike Cohen, primary investigator of the landmark HPTN 052 study on treatment-as-prevention and co-principle investigator of the HIV Prevention Trials Network, of which ÉCLAIR is a member, the study wasn't just good: It was essential.
"I think it accomplished its goals and more, in the sense that it demonstrated safety of the injectable in HIV-negative men and it demonstrated their willingness to participate and return," said Cohen.
This last part wasn't entirely predictable to Cohen. But the news seems to continue to be good about injectables. At the end of the phase-2 trial for ÉCLAIR, the men not only returned for follow-up shots, 74% of them reported a preference for the injection over daily cabotegravir pills. And the same appears to be true in two trials that are building on ÉCLAIR's results, he said. HPTN 076 and HPTN 077, both phase-2 studies of other long-acting injectables, saw high rates of retention, too.
"Injections look great," said Cohen. "[ÉCLAIR] demonstrated that men in the study returned repeatedly for injections, plus they were able to collect PK [pharmacokinetic] data to inform the right dose."