Injectable PrEP use leads to zero new HIV infections among gay, trans and non-binary Brazilians

Over the past week at the Conference on Retroviruses and Opportunistic Infections (CROI 2025) in San Francisco, we have been reminded that Latin America is one of the few global regions where new HIV cases have continued to rise since 2010, even though pre-exposure prophylaxis (PrEP) has been available for more than a decade.

However, scaling up of oral PrEP has been slow in this region. Additionally, those on PrEP have experienced adherence challenges. HIV disproportionately impacts young people and sexual and gender minorities, such as gay and bisexual men, transgender and non-binary people in Latin America. Black and mixed race people are also disproportionately impacted.

Find out more: Cabotegravir injections for PrEP

Long-acting injectable PrEP – in the form of cabotegravir (CAB-LA), administered every two months – has the potential to reshape the Latin American epidemic. In the ImPrEP cohort in Brazil, there were no new cases of HIV amongst 1200 people taking CAB-LA. These encouraging results were presented by Dr Beatriz Grinsztejn, Director of the HIV/AIDS Clinical Research Unit at the Evandro Chagas National Institute of Infectious Diseases, Oswaldo Cruz Foundation, and the current president of the International AIDS Society.

ImPrEP CAB Brazil

Gay, bisexual and other men who have sex with men, along with non-binary and trans people, seeking PrEP who tested negative upon rapid HIV testing, and had not used PrEP before, were eligible for enrolment in the ImPrEP CAB Brasil study.

In public PrEP services in six Brazilian cities (Rio de Janeiro, São Paolo, Salvador, Campinas, Florianopolis and Manaus), researchers enrolled 1447 participants who were offered the choice between CAB-LA and oral PrEP, starting in late 2023. Most chose the injectable option (83%). In addition to these two trial arms, researchers compared adherence and outcomes to 2411 people receiving PrEP from 14 Brazilian public health clinics.

To assist participants with their choice of PrEP modality, they either received standard-of-care counselling alone, or access to an educational video combined with standard-of-care information. Most participants who chose CAB-LA cited difficulties with daily medication adherence as a reason (78%).

Participants who watched the educational video were over twice as likely to choose CAB-LA (Odds Ratio: 2.54, 95% Confidence Interval: 1.89-3.44), according to additional poster data. It featured key population community members, providing clear explanations of both oral prep and CAB-LA, side effects, and the importance of adherence. When asked about it, Grinsztejn stated that it was a centrally important component, and aided later persistence even though participants were only presented with it at the time of making the choice at the baseline visit.

Participants who chose CAB-LA underwent further HIV RNA testing prior to the first and any reload injections to ensure that no HIV infections were missed. They were able to have their first injection on the same day.

PrEP coverage was defined as the proportion of days covered by PrEP during follow-up: for oral PrEP, four or more pills a week and for CAB-LA, receiving the first injection providing six weeks of protection, and subsequent injections providing 10 weeks of protection.

Overall, 38% of participants were between 18 and 24 years old. When comparing the three arms, most participants identified as men who have sex with men, while 9% identified as trans, non-binary or queer. Over half of the sample (60%) were Black, mixed race or other race – the highest proportion of people of colour were in the CAB-LA group. The sample mostly had secondary education or more, with people receiving PrEP at public health facilities outside the trial being less well-educated. The median follow-up period was longest for the CAB-LA group at seven months, followed by the oral PrEP choice cohort at around five months and the oral PrEP comparison group at around four months.

Zero HIV transmissions in those on injectable PrEP

CAB-LA injections were considered on time if they were given during a seven-day window, either before or after the target date. Overall, 94% of injections were given on time and adherence remained high even into the later months of the study. CAB-LA participants received a small incentive to assist with transportation on injection days.

The CAB-LA group had 95% PrEP coverage for the duration of the study, with 314,447 person-days of follow-up in total. This was significantly greater than both comparator arms: those receiving oral PrEP in the trial only had 58% coverage for a total of 67,486 person-days of follow-up, while those receiving oral PrEP at public health clinics had 48% coverage for a total of 546,071 person-days.

There were no cases of HIV transmission among those taking injectable PrEP. In the oral PrEP choice cohort, there was one serocoversion over 100.3 person-years of follow-up (incidence rate: 1.0, 95% Confidence Interval: 0.0-5.6 per 100 person-years). In the oral PrEP comparator group, there were nine cases of HIV transmission over 607.1 person-years (incidence rate: 1.5, 95% CI: 0.7-2.8 per 100 person-years).

There was some loss to follow-up in the CAB-LA group, with 71 people receiving at least one injection and not returning for subsequent injections (6%). However, of these, very few only received one injection (nine participants). People with lower education levels and trans and non-binary people were disproportionately represented among those lost to follow-up. Forty participants returned to ‘reload’ – they had had delayed injections during the trial, but later returned for re-dosing.

Twenty-one participants (9%) switched from oral to injectable PrEP during the study, mostly because they preferred not to take pills. There were also those who switched from CAB-LA to oral PrEP, 44 in total (a smaller percentage though, at 4%), largely due to side effects such as injection site reactions.

“In this large implementation study, 83% of participants chose CAB-LA,” Grinsztejn concluded. “CAB-LA significantly improved PrEP coverage and protection, underscoring its potential as a tool in addressing adherence challenges with daily oral PrEP, particularly among young key populations.”

What’s driving oral PrEP non-adherence?

A poster at CROI 2025 by the same research team explored oral PrEP use patterns over time to shed some light on why these groups struggle with maintaining PrEP adherence over time.

This ImPrEP analysis also included Brazilian adult gay men and trans women, followed between early 2018 to the end of 2024. Once again, a minimum of four PrEP pills or more per week was considered as providing adequate coverage. In this instance, discontinuation was defined as not receiving PrEP for at least six months and not returning until the end of the study’s follow-up period.

A total of 3810 participants returned for at least one follow-up visit for a median follow-up time of 5.19 years, or 15,587 person-years. Overall, 47% of participants discontinued oral PrEP, with HIV incidence in this group being much higher than those who stayed on oral PrEP.

Researchers identified three distinct PrEP trajectories, observed over five years of follow-up:

• Adequate adherence: 54% of participants with a consistent probability of having adequate adherence
• Gradual decline in adherence: 38% of participants
• Rapid decline in adherence: 8% of participants. This group only contributed 1309 person-years to the overall study follow-up time.

Younger people were more likely to be in the latter two groups. People with stimulant use had over a 50% higher chance of being in the rapid decliner group than in the adequate adherence group (adjusted odds ratio: 1.52, 95% CI: 1.10-2.09), while those engaging in transactional sex were more likely to be in this group. The HIV incidence rate was 3.66 times higher in rapid decliners than in those with adequate adherence (95% CI: 1.99-10.59; p < 0.001). 

“Higher HIV incidence among those who discontinued PrEP underlines the importance of retention in PrEP services,” the researchers concluded. “Long-acting PrEP technologies may help address the unique challenges faced by these subgroups, enhancing PrEP retention and overall effectiveness.”

During a press conference discussing the ImPrEP CAB Brazil results, Grinsztejn emphasised that increasing availability of injectable PrEP is only the first step: “It’s not only about making injectable PrEP available, but also addressing structural barriers to access and persistence.”