Interview with Professor Kuritzkes: Long-acting formulations for treatment and prevention
What does the term ‘long-acting formulations’ encompass in terms of HIV?
From a therapeutic perspective, long-acting formulations probably range from something that is only slightly better than daily therapy, such as weekly administration, to ideally something that could be administered every 6 months or even once a year. At the very extreme would be vector delivery systems, with a genetically engineered virus that would continuously produce a monoclonal antibody or something similar, requiring administration only once or twice for life; but that’s still at the aspirational extreme. More practically, I think we’re looking at either monthly administration with injectables or potentially twice a year with implantables.
In terms of prevention, there are a range of formulations ranging from injectables and implantables to rings and foils. Injectables and implantables have potential significant advantages over other modalities such as rings or foils, particularly in terms of adherence. For example, although the ASPIRE study certainly showed evidence of protection with the dapivirine ring, there was clearly an association between the level of protection and the extent of adherence to the regimen. With injectables and implantables, once you’ve given the shot you know the person will be protected for the next month (or six months if it’s an implantable). But, as we’ve learnt from contraception, having a range of options would certainly be a huge advance in the field.
In terms of treatment, where are we now?
In the short term, I think the combination of cabotegravir and rilpivirine will be the first product to be submitted for approval; the 96-day update for the LATTE study into this combination was presented at IAS 2017 in Paris this July. These results really look quite promising and phase 3 studies are also underway. However, I don’t think it’s going to be a long-term solution because it requires intramuscular injection and I am also doubtful about how broad a population it will really appeal to. Nowadays, we only see our successfully-treated patients twice a year. With the cabotegravir/rilpivirine combination, patients would have to come to the clinic every one to two months to get their injections, since you can’t self-administer an intramuscular injection. So I think it would only appeal to a narrow segment of the population.
Perhaps a more interesting approach, although probably 5-10 years away from becoming available, would be implantables. These would probably only require 6-monthly visits to exchange the implant. Of these, the Merck nucleoside reverse transcriptase translocation inhibitor (NRTTI), MK8591, is the most promising so far. For treatment, it still needs a partner drug, but for prevention MK8591 would be fine. In fact, at IAS 2017, Martin Markowitz presented a poster showing that injection of MK8591 protected against an SHIV challenge in monkeys.
Do long-acting formulations offer more opportunity in prevention than treatment?
I think there’s a huge opportunity in prevention, where the biggest challenge is adherence, for example in adolescents girls. But also among MSM, I’m sure there’s a group that would prefer monthly injection or infrequent administration of a product to protect themselves. However, I think there’s a role in treatment too, particularly for non-adherent patients. For example, in our clinic, 80% of the patients are undetectable, but the 20% who are not include people who don’t come to their visits because they have chaotic lives, psychiatric disease, substance abuse problems, etc. If we could see these people twice a year and give them an injection or implant with a long-acting formulation and have them suppressed for the next six months, that would be a huge advance.
What about broadly neutralising antibodies?
The area of broadly neutralising antibodies is very interesting. Broadly neutralising antibodies are intrinsically long-acting and they’ve been further modified to make them even longer acting. One example is the long-acting VRC01-LS, which has a half-life of months. However, as a single antibody that’s not going to be sufficient. Some companies are therefore also working on genetically engineered antibodies that are bi-specific or tri-specific. Although those are just entering into clinical trials now, I think they have huge potential for treatment. The issue, however, will be cost of goods and logistics of delivery.
What about side-effects with long-acting formulations?
Beyond injection-site reactions, one of the challenges with the cabotegravir/rilpivirine approach is that there are not enough safety data to allow trials to go directly to delivery of the long-acting formulation. To make sure there’s no hypersensitivity there has to be an initial lead-in phase with an oral formulation. In this respect, one of the advantages of implantables is that they can be removed in the case of a hypersensitivity reaction.
Broadly neutralising antibodies are not expected to have any side effects because, in contrast to some of the monoclonal antibodies used to target cellular factors, they only target the virus. Nonetheless, as with almost any antibody, you will still have rare hypersensitivity reactions.
How close are we to a long-acting PrEP formulation?
Cabotegravir is probably going to be the first long-acting PrEP formulation that will become available. There are two large cabotegravir phase 3 trials going on at the moment, but it will be a few years before they’re completed and I really think it will depend on what the results are. The study in men is set up as a non-inferiority trial and, to me, if the injectable isn’t better than daily tenofovir, why would you do it? The study in women, on the other hand, is set up as a superiority study, partly because oral Truvada® has been less successful in this population, so there they have a chance of making the case.
The ASPIRE follow-on study with the dapivirine ring is also coming to an end and will provide extended efficacy and safety data. So I think with that they’ll probably get approval, although I don’t know how much uptake there will be.
Beyond toxicity, what other concerns are there with long-acting therapy?
For the small molecules, the concern is drug resistance. Particularly in treatment, and hypothetically in prevention. It’s fine if people keep receiving their injections, or switch to oral medication if they discontinue the injections, but for both cabotegravir and rilpivirine, the tail is very long. Drug can be detected in the plasma for up to a year after the last injection. In terms of the rate of drug decay, we don’t know how long the window is during which replication is no longer inhibited, but the drug level is high enough to select for resistance. A very steep fall-off in plasma levels would be fine, because then the window is quite narrow, but if it’s a long slow slope, you could have weeks or months during which the virus resumes replication and you could start to see selection for drug resistance. Once this happens, there’s the possibility of transmitting a resistant virus. People are doing modelling to address this question, but we need empirical data to show us what the risk is.
»View Prof. Kuritzkes' NCHIV slides here