Peer-driven delivery of long-acting injectable cabotegravir for key populations: Preliminary findings from an implementation study in Beira, Mozambique

INTRODUCTION

In Mozambique, key populations (KPs), including female sex workers (FSWs), men who have sex with men (MSM), and transgender people, face disproportionate HIV risk and adherence barriers to daily oral pre-exposure prophylaxis (PrEP). This implementation study evaluated the integration of long-acting injectable cabotegravir (CAB-LA) as PrEP into a peer-driven outreach model to assess its uptake and realworld continuation.

METHODS

This convergent parallel mixed-methods study took place in Beira, Mozambique. The quantitative component was a prospective cohort study enrolling KP members (aged ≥18 years) who were HIV-negative to assess PrEP uptake, continuation, and safety. Participants were recruited through community-based peer networks supporting preventive care for KPs and were given the choice between oral PrEP and intramuscular bi-monthly CAB-LA via community drop-in centres (DICs) and clinics. CAB-LA users required follow-up every 2 months. The qualitative component used in-depth interviews and focus groups to explore acceptability and enablers of continuation.

ETHICS

This study (version 1.7) was approved by MSF Ethics Review Board (ID 2450). Version 1.8 of the study protocol was approved following revision by Mozambique’s national committee for health ethics (Ref. 126/CNBS/2024).

RESULTS

Between July and November 2025, 238 participants were enrolled (median age 26 years, range 18–64). Of these, 111

(47%) were FSW, 102 (43%) were MSM, two (1%) were transgender, and 25 (11%) were undisclosed. One positive NAAT result (<1%) was identified among 204 users screening negative for symptoms of acute HIV infection. 127 [53%] participants were recruited via community DICs. At baseline, 188 (79%) participants chose CAB-LA, and 31 (13%) chose oral PrEP. Longitudinal data showed 204 CAB-LA users (six switches from oralPrEP and ten initiations), of whom 147 (72%) returned for a second dose. Subsequent interval continuation remained robust, with 63 (86%) of 73 eligible users receiving a third dose and 16 (73%) of 22 continuing to a fourth dose. 41 (20%) of 204 CAB-LA users reported 70 adverse events (AEs), including site pain, headache, and nausea. 35(50%) AEs were grade 1, 35(50%) grade 2–3, and none grade 4; all the AEs resolved. Qualitative interviews revealed that CAB-LA is preferred due to its “invisibility”, reducing stigma and domestic conflict associated with oral PrEP bottles. Participants unanimously preferred decentralised models, identifying DICs as “safe spaces”: “It is our home!… I prefer to come here” (FSW). While injection-related side-effects were identified as primary barriers of continuation, users viewed them as an acceptable trade-off for protection: “Nothing comes to us without sacrifice...” (FSW).

CONCLUSION

Preliminary findings demonstrate that CAB-LA is an acceptable and preferred prevention tool, particularly when delivered through community-based spaces like DICs. Discretion and no pill burden are the primary drivers of continuation. However, injection-related side-effects may pose an implementation challenge. Peer-driven models can inform scale-up of long-acting PrEP programmes.