In Denver this past February, at the 33rd Conference on Retroviruses and Opportunistic Infections (CROI 2026), ViiV Healthcare arrived with a familiar promise, fewer doses, less burden, more choice for HIV patients. But this time they had a sharper edge than previous years. Long-acting injectables are no longer just a lifestyle upgrade for the already-stable. The company’s headline is aimed at the messier, more consequential problem in HIV care, what to do when daily tablets and daily life don’t reliably line up.
The centrepiece is the LATITUDE study. The final 48-week data, now published in The New England Journal of Medicine, is showing that switching to long-acting injectable cabotegravir + rilpivirine (Cabenuva) outperformed continued daily oral therapy for people with a history of adherence challenges. In plain terms, the injectable strategy cut the cumulative risk of “regimen failure” (a composite of virologic failure and discontinuation) by nearly half through 48 weeks, 22.8% versus 41.2%.
That result matters not because it flatters an already successful regimen, but because it challenges a quiet, uncomfortable assumption that has shaped HIV policy for years. That adherence problems are a patient flaw to be coached around, rather than a predictable outcome of poverty, stigma, mental health strain, unstable housing, chaotic work, side-effects, or just… being human.
When “support” isn’t enough, an injection can be
LATITUDE (ACTG A5359) deliberately recruited people who were not sailing through conventional care. Participants with evidence of sub-optimal adherence or poor retention. The trial design itself is revealing. Before randomisation, participants received adherence support (including conditional economic incentives) to achieve viral suppression on oral therapy, and only then were they randomised to switch to monthly injections or remain on oral ART.
This is precisely why the findings land with such force. Even after additional support aimed at “fixing” adherence, staying on pills still produced a much higher probability of failure than switching to injections. In the injectable arm, virologic failure events were far less frequent than in the oral arm (reported as 6.8% vs 28.2% cumulative probability for VF at week 48 in the company’s summary of the NEJM results).
The study also carries a piece of recent history. An independent monitoring board recommended halting randomisation early (February 2024) and inviting eligible participants to switch, based on interim efficacy. That’s not marketing sheen; it’s the kind of decision committees make when they think equipoise has collapsed.
LATITUDE is the strongest rebuttal yet to the idea that long-acting injectables are merely a convenience product. In a subset of people living with HIV, those most likely to fall out of suppression, the “convenience” framing becomes a category error. This is harm reduction by pharmacology.
And yet, long-acting care doesn’t magically eliminate adherence. It moves adherence from “take a pill every day” to “show up reliably every month (or every two months)”. That shift can be liberating, but it can also be unforgiving. Missed injections carry pharmacologic tail risks, potential resistance concerns, and a need for careful bridging strategies.
Guideline writers have been grappling with exactly this. How to use injectable cabotegravir/rilpivirine as “replacement ART” and how to select and support patients, especially those with prior viraemia or retention challenges, without setting them up to fail on a new schedule.
So the real question isn’t “do injections help people who struggle with pills?” LATITUDE strongly suggests yes. The real question is whether health systems are prepared to deliver an adherence-supportive service around the injection, appointment flexibility, outreach when someone misses, transport support, stigma-aware clinics, and protocols that assume life will interrupt the best-laid dosing calendars.
Ultra-long-acting ViiV’s pipeline hints at a different decade of HIV care
Beyond Cabenuva’s near-term implications, ViiV used CROI 2026 to widen the lens. The company is trying to push HIV treatment into ultra-long-acting (ULA) territory. Multi-month intervals that begin to resemble contraception models more than chronic disease management.
Two pipeline threads stand out:
1) VH184: a “third-generation” integrase inhibitor with ULA intent.
ViiV is presenting first-in-human Phase I data for injectable long-acting formulations of VH184, positioning it as the first “third-generation” INSTI and explicitly framing the work around ULA potential for future regimens. The implication is that ViiV wants the backbone drug itself to be compatible with far longer dosing intervals, rather than simply stretching existing approaches.
2) Antibody-based durability: lotivibart (N6LS) plus cabotegravir LA.
The more intriguing, and arguably more futuristic, concept is combining long-acting small molecules with broadly neutralising antibodies. At CROI 2026, ViiV highlights 12-month data from the Phase IIb EMBRACE study evaluating investigational lotivibart (N6LS) given every four months, paired with monthly long-acting cabotegravir. If you can extend one component to every four months (and later perhaps both), you’re building a path away from monthly clinic dependency.
ViiV has also flagged an early capsid-inhibitor programme (VH499) among the ULA efforts, signalling that the company is not betting on a single mechanism to deliver the next era of dosing flexibility. Long-acting treatment can be an equity lever, or an equity trap, depending on how it’s deployed.
The hopeful version is straightforward for people who cycle in and out of care, who face stigma at home, who have chaotic schedules, who struggle with swallowing pills, or who simply find daily HIV medication psychologically heavy, a clinic-administered injection could reduce failure and keep viral suppression steady, protecting personal health and lowering transmission risk.
The pessimistic version is also plausible if ULA drugs are expensive, rationed, and tied to specialist centres with limited appointment capacity, then the people who benefit most may be the least likely to access them. And if missed-visit safety nets are weak, long-acting regimens could punish instability rather than accommodate it.
LATITUDE’s population, older, disproportionately Black/African American, with a meaningful minority reporting current or prior injection drug use, reads like a reminder of where HIV care still fractures. It is exactly the group that should not be asked to “prove” perfect attendance before receiving the tool designed to help them maintain suppression.
There’s a temptation, especially among budget-strained systems, to look at long-acting drugs and see staffing savings. That would be a misread.
Long-acting HIV care, done properly, is high-touch in a different way. It requires proactive scheduling, missed-appointment tracking, continuity across services, and patient-centred flexibility. In other words, it requires the very operational competence that fragmented health services often lack.
Which is why CROI 2026 is less about a single dataset and more about a challenge to delivery models. The science is sprinting ahead and the services have to catch up.
ViiV’s CROI 2026 story has two speeds. Cabenuva’s LATITUDE data is immediate and pragmatic. It suggests that for people living with HIV who have struggled with daily oral adherence, switching to long-acting injections can improve outcomes in a way counselling alone often can’t.
The ultra-long-acting pipeline is the longer bet. VH184, antibody combinations like lotivibart (N6LS), and capsid-inhibitor work hint at a future where “treatment” could mean a handful of clinic visits per year.
The innovation is real, but the decisive battleground is delivery. If health systems build humane, flexible injection services, long-acting HIV care could become one of the most meaningful adherence interventions of the decade. If they don’t, we’ll end up with brilliant pharmacology trapped inside ordinary bureaucracy.