Cabotegravir has a substantial trial programme covering both prevention and treatment, and the prevention results were among the more striking in recent HIV research.
What did HPTN 083 find?
This trial compared injectable cabotegravir every two months with daily oral PrEP in men who have sex with men and transgender women. Injectable PrEP produced significantly fewer infections and the study was stopped early once the difference became clear, since continuing would have been unjustifiable.
What did HPTN 084 find?
The parallel trial in women in sub-Saharan Africa found a similar and larger difference in favour of the injection, and was also stopped early. Together the two trials established injectable PrEP across quite different populations and settings.
Why was the injection better?
Almost entirely adherence. Participants receiving injections were reliably protected because the drug was administered rather than remembered. Analyses of oral arm participants with detectable drug levels showed high protection there too, which is the same pattern seen in earlier PrEP trials.
What did the trials show about resistance?
A small number of infections occurred, and where they did, integrase resistance was sometimes found. These cases clustered around undiagnosed infection at enrolment or infection during the period when drug levels were declining. This finding shaped current practice on testing before every injection.
What is known about the tail?
Pharmacokinetic studies show cabotegravir remaining detectable for many months after the last injection, declining gradually. Modelling and observed cases indicate a window in which levels are present but insufficient to prevent infection, which is the period of resistance risk.
What does that mean practically?
That anyone stopping injectable PrEP should switch to oral PrEP or another method to cover the tail rather than simply stopping. This is one of the clearest practical lessons from the trial programme.
What established it in treatment?
The ATLAS and FLAIR studies compared monthly injectable cabotegravir with rilpivirine against continued daily oral therapy in people already suppressed. Both found the injectable regimen maintained suppression comparably. ATLAS-2M subsequently showed that two-monthly dosing worked as well as monthly.
What did participants report?
High satisfaction, with most preferring injections to tablets despite injection site reactions being common. That preference is consistent across studies and reflects how much the daily reminder of a diagnosis weighs on some people.
What about injection site reactions?
Reported by the large majority of participants at some point, usually mild to moderate pain lasting a few days, and diminishing with successive injections. Discontinuation because of them was uncommon.
What questions remain?
How the tail is best managed in real-world settings where people disengage from care without telling anyone. How well the every-two-month schedule holds up outside trial conditions where appointments are chased. And how commissioning will scale the service demands, since injections require considerably more clinic capacity than dispensing tablets.
What is the overall picture?
Strong evidence for effectiveness in both uses, with the practical demands and the resistance risk around the tail as the recognised limitations rather than anything about the drug's activity.
How long has it been in use?
Since the early 2020s, so real-world experience is shorter than for oral drugs. Trial data are strong and the practical questions about running an injectable service at scale are still being worked through in NHS settings.
What are services learning?
Mainly that the logistics matter as much as the drug: appointment capacity, reminder systems, and what happens when someone disengages. Programmes that manage those well reproduce the trial results; those that do not see gaps appear.
Where can you read the evidence yourself?
NAM aidsmap summarises HIV research reliably and without overstatement, and the British HIV Association publishes the guidelines UK clinicians actually follow. Both are freely available and are considerably more useful than a general search, which tends to surface material from other countries with different practice.
Does the evidence apply to you?
Trials recruit particular populations, and their results transfer best to people who resemble the participants. Where a study excluded a group, that is a gap rather than a negative finding. It is a fair question to ask a clinician which evidence their recommendation for you actually rests on.