Abstract
INTRODUCTION: Randomized controlled trial evidence suggests that long-acting injectable (LA-I) cabotegravir and rilpivirine (CAB+RPV) has similar virological failure (VF) rates to daily oral therapy, but clinical practice evidence is lacking. Integrase inhibitor (INI) resistance may limit future therapy. The optimal regimen is uncertain. METHODS: We synthesized evidence from PubMed, EMBASE, Cochrane and conference abstract databases through 18 November 2024, to identify observational cohort studies (OCS) that reported on VF events in virally suppressed individuals who switched to LA-I CAB+RPV. We extracted data on VF, resistance-associated mutations (RAMs) at VF, post-VF regimen choice and re-suppression. We assessed the risk of bias using a modified Downs and Black tool. RESULTS: VF definitions differed considerably among OCS, with 172 individuals experiencing VF across 79 cohorts that included 13 899 individuals. Twenty-eight cohorts (n = 7987) reported genotypic information at VF. Out of the 80 VF events with genotypic information available at the time of the VF event, NNRTI mutations were identified in 45 cases, INIs in 40 cases, and dual-class resistance in 33 cases. Notably, 28 VF events were not accompanied by resistance. Post-VF regimen choices were reported for 92 VF events. Regimens used were protease inhibitor (PI)-based, oral INI-based and some physicians continued LA-I CAB+RPV post-VF. Re-suppression occurred in 87.8% (65/74) of VF events in which it was described. CONCLUSIONS: In OCS, VF was a very uncommon occurrence and comparable with clinical trials. However, when it did occur, it was frequently accompanied by resistance. Post-VF regimens used to achieve suppression varied, including LA-I CAB+RPV maintenance and were highly successful.