Injectable cabotegravir makes progress towards human efficacy studies: doubts about injectable rilpivirine

Researchers have determined the dose of an injectable formulation of the integrase inhibitor cabotegravir (formerly GSK744) that will be taken into efficacy trials to see if it can be used for pre-exposure prophylaxis (PrEP). Studies in animals were presented earlier this year suggesting that drug levels stayed high enough in the body for it to be injected quarterly, and now studies of drug concentration in humans have confirmed that an 800mg intramuscular injection will be given once every twelve weeks in efficacy trials.  

Progress on an injectable formulation of another drug, rilpivirine, received a setback, however, when animal studies showed it lost its efficacy against viral challenge after only 18-21 days.

Cabotegravir

Bill Spreen of GlaxoSmithKline reminded the Research for Prevention (R4P)conference in Cape Town, South Africa, that experiments in monkeys had shown that cabotegravir almost completely protected a group of monkeys from infection with a monkey-adapted version of HIV.

Drug-level monitoring studies in human volunteers have found that a dose of 500mg produces levels of the drug that stay within the therapeutic level for as long as 16 weeks and that the drug's ‘half-life’ – its rate of elimination in the body – is 25 times longer with the injectable formulation than with the oral cabotegravir pill that is being developed for HIV treatment. They calculated that an 800mg dose every twelve weeks should result in trough levels of drug that are still over eight times the 90% inhibitory concentration (IC₉₀) of the drug – the level that should reduce viral replication by 90%.

It is important to study the concentration of a drug being used for PrEP in the genital tissues as well as in the blood, as this is where HIV gets into the body in most cases. Here they made some interesting findings. Using a dose of 400mg, or half the dose to be taken into efficacy studies, they found that the concentration in tissues was considerably lower than it was in the blood. The concentration in vaginal tissue was 28%, in the cervix 16% and in the rectum 8% of blood plasma concentrations. This is not unexpected as the same is found when drugs are dosed orally. There was also an expected variation by body weight. Drug elimination rates were 35% higher in the 10% of volunteers with the lowest BMI (body mass index) and 20% in the corresponding proportion of heaviest volunteers. These differences, however, are not sufficient to need different doses for different body weights.

What was more unexpected was that men turned out to eliminate the drug considerably faster than women – from 2.8 times faster in the lowest-BMI subjects to 3.1 times faster in the heaviest ones. This is probably because women have higher levels of fatty tissue, which releases the drug more slowly. Spreen commented that different doses for men and women might be needed if phase II studies suggested this made a clinical difference. These will start next year.

Rilpivirine

What is less certain is whether another drug wil be taken forward as injectable PrEP. This is TMC278LA, which is an injectable formulation of the non-nucleoside reverse transcriptase inhibitor rilpivirine (Edurant, also in Eviplera/Complera).

This drug was recently given to laboratory mice that are genetically adapted to become infected with HIV. In a study where mice were challenged with a single HIV variant a week after receiving TMC278LA, no mice were infected, compared with every mouse in a control group that only received a placebo injection.

In a second experiment, the mice were challenged with three separate strains of HIV a week after receiving the PrEP injection, and were then challenged with a fourth strain two weeks after that (three weeks after PrEP).

One out of eight mice did become infected with one of the three viral strains in the first challenge. However, and disappointingly, another five mice became infected with the fourth strain of virus they received at week three.

Interestingly, signs of infection appeared only after a delay of up to a month and, in one mouse, an infection with one of the first three viral strains only became apparent after a two-week delay, suggesting that the TMC278LA was slowing down viral replication, but the fact that three-quarters of the mice eventually became infected with challenge virus may put a question mark over injectable rilpivirine as PrEP.

Humans are less like mice than monkeys, however, and presenter Olivia Snyder commented that they metabolised drugs faster, so TMC278’s effect could last longer in humans. Further experiments are going ahead to quantify the dosing interval needed, but clearly this drug is not going to be useful for PrEP, though it could have limited use as treatment in some patients.