Switch to long-acting cabotegravir and rilpivirine in virologically suppressed adults with HIV in Africa (CARES): week 48 results from a randomised, multicentre, open-label, non-inferiority trial

Background: Long-acting injectable cabotegravir and rilpivirine is licensed for individualised treatment of HIV-1 infection in resource-rich settings. Additional evidence is required to support use in African treatment programmes where demographic factors, viral subtypes, previous treatment, and delivery and monitoring approaches differ. The aim of this study was to determine whether switching to long-acting therapy with injections every 8 weeks is non-inferior to daily oral therapy in Africa. Methods: CARES is a randomised, open-label, non-inferiority trial being conducted at eight sites in Uganda, Kenya, and South Africa. Participants with HIV viral load below 50 copies per mL on oral antiretroviral therapy and no history of virological failure were randomly assigned (1:1; web-based, permuted blocks) to receive cabotegravir (600 mg) and rilpivirine (900 mg) by intramuscular injection every 8 weeks, or to continue oral therapy. Viral load was monitored every 24 weeks. The primary outcome was week 48 viral load below 50 copies per mL, assessed with the Food and Drug Administration snapshot algorithm (non-inferiority margin 10 percentage points) in the intention-to-treat exposed population. This trial is registered with the Pan African Clinical Trials Registry (202104874490818) and is ongoing up to 96 weeks. Findings: Between Sept 1, 2021, and Aug 31, 2022, we enrolled 512 participants (295 [58%] female; 380 [74%] previous non-nucleoside reverse transcriptase inhibitor exposure). Week 48 viral load was below 50 copies per mL in 246 (96%) of 255 participants in the long-acting therapy group and 250 (97%) of 257 in the oral therapy group (difference –0·8 percentage points; 95% CI –3·7 to 2·3), demonstrating non-inferiority (confirmed in per-protocol analysis). Two participants had virological failure in the long-acting therapy group, both with drug resistance; none had virological failure in the oral therapy group. Adverse events of grade 3 or greater severity occurred in 24 (9%) participants on long-acting therapy and ten (4%) on oral therapy; one participant discontinued long-acting therapy (for injection-site reaction). Interpretation: Long-acting therapy had non-inferior efficacy compared with oral therapy, with a good safety profile, and can be considered for African treatment programmes. Funding: Janssen.