Injectable HIV therapy beat pills for people with adherence barriers

Monthly injections of long-acting cabotegravir plus rilpivirine kept HIV suppressed better than daily pills in people with a documented history of missed doses, according to the randomised LATITUDE trial published in the New England Journal of Medicine [s1]. Regimen failure by week 48 occurred in 22.8% of those on the injections versus 41.2% on standard oral therapy — a difference large enough that the trial's data monitors stopped the randomised comparison early [s1].

The result matters because it tests long-acting HIV treatment in exactly the group excluded from the studies that first brought it to market. Cabotegravir–rilpivirine was approved in the United States in 2021 only for adults who were already virologically suppressed on a stable oral regimen — that is, people who had shown they could take medication reliably [s2]. LATITUDE asked the opposite question: does removing the daily pill help those who struggle with one?

What the trial did

LATITUDE, funded by the National Institute of Allergy and Infectious Diseases, enrolled people with HIV who had inadequate adherence to antiretroviral therapy, defined as a persistent HIV-1 RNA level above 200 copies per millilitre or loss to follow-up [s1]. The design ran in two steps. In step 1, 453 participants received up to 24 weeks of adherence support, conditional economic incentives, and standard care with oral drugs [s1]. The enrolled group was younger and more marginalised than a typical treatment trial: median age 40 years, 63% Black, and 29% assigned female sex at birth [s1].

Participants who reached an HIV-1 RNA level of 200 copies per millilitre or lower during step 1 then entered step 2, where 306 were randomly assigned in a 1:1 ratio to either continue standard oral care or switch to monthly injections of long-acting cabotegravir plus rilpivirine [s1]. That structure is important: everyone in the randomised comparison had first been brought to at least partial suppression, so the trial measures whether injections hold that gain better than pills do, not whether they can rescue uncontrolled disease from the start.

The primary outcome was regimen failure — either confirmed virologic failure, meaning two consecutive HIV-1 RNA measurements above 200 copies per millilitre, or discontinuation of treatment during step 2 [s1].

What it found

After a median follow-up of 48 weeks, a prespecified analysis showed the injection group doing better on secondary outcomes, and step-2 randomisation was stopped early on that basis [s1]. The cumulative incidence of regimen failure by week 48 was 22.8% with cabotegravir–rilpivirine and 41.2% with standard care, a difference of 18.4 percentage points (98.4% confidence interval −32.4 to −4.3; P=0.002) [s1].

Safety looked similar between the groups. The cumulative incidence of an adverse event was 43.5% with the injections and 42.4% with standard care, a difference of 1.1 percentage points (95% CI −12.7 to 15.0) [s1]. Resistance-associated mutations, the long-standing worry with any injectable regimen because drug lingers in the body for months after a missed dose, developed in 2 participants with confirmed virologic failure in each group [s1].

How to read it

An early stop for benefit tends to inflate an effect size, so the 18.4-point gap is best treated as real in direction but uncertain in exact magnitude. The wide confidence interval — from about 4 to 32 percentage points — says the same thing: the injections were clearly better, but by how much is not pinned down.

Two features bound the conclusion. The trial was open-label, so participants and clinicians knew who was getting injections, and the extra clinic contact that monthly dosing requires is itself a form of adherence support that a pill-based arm does not get. And every randomised participant had already been coaxed to suppression in step 1 with incentives and support that are not universally funded, so the trial speaks to maintaining control in a supported setting, not to starting from scratch. That resistance emerged equally in both arms is reassuring but rests on small numbers.

Why it is a departure

Long-acting antiretrovirals have reshaped HIV prevention faster than treatment: twice-yearly and daily options have moved to the centre of the prevention conversation, as covered in the approval of twice-yearly lenacapavir for pre-exposure prophylaxis and in the CDC's clinical recommendation for injectable PrEP. Treatment has been more cautious, precisely because the approved indication fenced injectables off from people with adherence problems — the group in whom a missed injection could seed resistance [s2]. LATITUDE is the first randomised evidence that, handled within a supported programme, the same tool can help the people it was originally kept away from [s1].

What to watch

The open questions are about the wrapper, not the drug. LATITUDE bundled injections with incentives and intensive support, and whether the benefit survives outside a trial — in clinics without that scaffolding, and in health systems where a monthly injection visit is hard to guarantee — is what implementation studies will have to settle [s1]. Guideline committees will also weigh whether the resistance signal stays flat as larger numbers accrue, since that risk is the reason the cautious label was written in the first place [s1][s2].

This article describes research and regulatory context and is not medical advice. Decisions about antiretroviral therapy are for patients and their treating clinicians.

Sources

• Cabotegravir plus Rilpivirine for Persons with HIV and Adherence Challenges — New England Journal of Medicine, 18 February 2026
• Drugs@FDA: Cabenuva (cabotegravir; rilpivirine), NDA 212888 — U.S. Food and Drug Administration, approved 21 January 2021