Injectable rilpivirine shows promise in phase I trials – but may work better for anal than vaginal sex

A phase one dose-finding and safety study in humans of TMC278 LA, a long-acting, injectable formation of the antiretroviral (ARV) drug rilpivirine, found that a single 1200-mg dose could produce sustained drug levels in rectal tissues that could offer protection against HIV for three months, and did in fact suppress viral replication in so-called explants (biopsies of rectal mucosal cells) for that length of time.

However – and to the surprise of researchers presenting this study at the HIV Research for Prevention conference in Cape Town – drug levels seen in vaginal and cervical cells were only about half of those seen in rectal cells, and viral replication was not suppressed in vaginal and cervical biopsies taken from women given TMC278 LA.

Detailed results – drug levels

Ian McGowan of the University of Pittsburgh presented results from a study that gave 24 women and 12 men a dose of either 600mg or 1200mg TMC278 LA. This was given as either one injection in the buttock (600mg) or two (1200mg) with one hour between shots.

Importantly, McGowan was only presenting here the tissue-concentration results in women – including the rectal ones. As Tuesday’s presentation on cabotegravir showed, absorption rates of these injectable drugs may differ between women and men.

In terms of the whole group of 36 volunteers, eight of the participants were black and three Hispanic, with the others largely white, and their average age was 29.

There were adverse events: in fact mild adverse events were nearly universal, but these consisted of mild or moderate pain at the injection site for three days or so, which might involve a bit of discomfort sitting down, but nothing worse, and the two serious adverse events seen were not related to the drug.

The 600mg dose produced sustained levels of rilpivirine in the blood of more than twice the IC₉₀ (the amount needed to suppress viral replication by 90%) for over 28 days while the 1200mg dose produced levels of more than four times the IC­­­₉₀. Levels equivalent to the IC­₉₀ were reached after 56 days in the 600mg dose and after 100 days, over three months, in the 1200mg dose.

In vaginal, cervical and rectal fluids, vaginal concentrations were about twice as high as rectal ones. Only the 1200mg dose produced levels above the IC₉₀ in rectal fluids.

In vaginal and rectal tissues, however, the situation was reversed, with roughly double the concentration in rectal than vaginal or cervical tissues. In rectal tissues, levels above the IC₉₀ persisted for 64 days in the 600mg dose and over three times the IC₉₀ in the 1200mg dose.

Detailed results – protection from HIV infection

This study uses what is called an explant model to determine if the drug is inhibiting HIV. Obviously one cannot inject people with HIV to find out if the drug works, but one can do the next best thing – take biopsies of rectal, vaginal and cervical cells, keep them alive in laboratory dishes, and try and infect the cells with HIV.

Rectal cells taken from women given TMC278 LA four weeks after they received their injection were completely resistant to HIV infection in the lab dish. Cells biopsied 56 days after injection with 1200mg of drug were still resistant to HIV infection, and only slightly susceptible to infection with the 600mg dose. By day 64, cells taken from people given the 600mg dose were susceptible to HIV infection but were still resistant to infection with the 1200mg dose, eventually becoming susceptible after 84 days.

In contrast, and to the researchers’ surprise, vaginal and cervical cells taken from women given TMC278 LA showed no signs of benefiting from the drug – they were just as susceptible to infection after the injection as before.

This is surprising because one would have expected some effect, even if a weakened one. McGowan and principal investigator Linda-Gail Bekker reminded the audience that the explant-cell technique is only an assay, a surrogate measure of protection, and may not be reflected in the real world. MGowan added that the next step was to perform multiple-dose phase I studies in which women and men were given three doses of the drug.

It is possible that women may need more than one dose for concentrations to build up to protective levels in vaginal and cervical tissue. A modelling study presented at the same conference found that levels of the nucleoside drugs tenofovir and emtricitabine – the drugs in Truvada – reached lower levels in vaginal rather than rectal tissue, and this study may indicate that the same is true on the non-nucleoside drugs like rilpivirine. This does not necessarily mean that injectable rilpivirine may not work for women but Bekker was asked whether it was ethical to start phase II trials early next year without evidence of efficacy in women. She said: “We certainly intend to counsel women volunteering for the study that there is no evidence yet that it is at all effective.” The phase II study, HPTN 076, will take place in lower-risk women in Zimbabwe and South Africa, and men who have sex with men in South Africa and the United States next year.

After a presentation on Tuesday that found that TMC278 LA offered reasonable but rather short-term protection against HIV infection in mice, this study adds another layer of complexity to the data we have on injectable rilpivirine.

A case of drug resistance caused by injectable PrEP

One further case report presented to the conference added to the issues that may need to be taken into account if these injectable, long-acting formulations are to be used for HIV prevention. Kerri Penrose of the University of Pittsburgh presented a case report of a person who had developed HIV drug resistance to rilpivirine because she acquired HIV during the time the drug levels in her body were slowly declining.

The study participant’s infection was a surprise, because the trial for which she volunteered was intended for women at low risk of HIV. Nonetheless, she tested positive for HIV 84 days after being given a 300mg dose of TMC278LA – half the minimum amount in the studies discussed above.

By this time, seven weeks after her injection, drug levels of rilpivirine in her body were below the level needed to prevent HIV infection, so this cannot be seen as a ‘PrEP failure’ – no one would have expected this relatively small dose to work by this time. At the first drug-level test after her diagnosis, she had 7.5 nanograms per millilitre (ng/ml) of rilpivirine in her blood; the IC­₉₀ for rilpivirine is about twice this, 12 ng/ml.

Having enough drug in your body to slow down viral replication but not enough to stop it is how drug resistance arises, because only drug-resistant mutations survive. Although she was infected with ‘wild type’, non-drug-resistant virus, she developed drug resistance and, a month afterwards, 19% of the HIV in her blood had resistance mutations to rilpivirine and other NNRTI drugs. However, the presence of resistant virus was relatively fleeting; as rilpivirine levels fell further, and also because she started HIV treatment, the wild-type virus came to predominate and five months after her diagnosis NNRTI resistance had disappeared.

This study shows that a major issue in the long-acting injectables may be how to stop taking them if one is still at risk of HIV, as drug may persist in the body for months. Clearly the best idea is to cover the ‘long tail’ with oral PrEP but this may not always be obtainable in the low-income settings in which the use of injectable PrEP is being contemplated.