Determinants of long-acting cabotegravir plus rilpivirine discontinuation in real-world HIV care: insights from the relativity cohort

OBJECTIVES: To evaluate real-world persistence with long-acting cabotegravir plus rilpivirine (CAB+RPV LA) in a nationwide cohort of virologically suppressed adults, describe the main clinical and structural reasons for discontinuation, and identify predictors of early interruption in routine practice. DESIGN: Ambispective, multicenter cohort study assessing the durability, safety, adherence, and virological outcomes of CAB+RPV LA implementation across 58 Spanish healthcare settings. METHODS: Adults initiating CAB+RPV LA therapy were enrolled using standardized electronic tools. Baseline demographics, comorbidities, HIV and virological variables, and injection timing were recorded. Permanent discontinuations were attributed to systemic adverse events, injection-site reactions, virological failure (VF), or nonclinical causes. The virological endpoints were aligned, although not identical, to the CONSENSUS-LAI criteria. Kaplan-Meier curves and multivariate Cox models were used to identify the independent predictors. RESULTS: Among 3146 virologically suppressed participants, 199 (6.3%) permanently discontinued therapy over a median 13.8-month follow-up. Discontinuation was more frequent among women (21.6 vs. 14.4%; P = 0.007), foreign-born individuals (42.9 vs. 29.5%; P < 0.001), those with a higher BMI, and those with shorter prior suppression periods. injection-site reaction- and systemic adverse event-related discontinuations were rare (1.2 and 0.7%, respectively), whereas structural reasons accounted for 3.5%. VF occurred in 20 participants (0.6%). Adjusted analyses showed that age ≥70 years [hazard ratio (HR) 5.41; 95% confidence interval (CI) 1.91-15.32] and foreign nationality (HR 2.06; 95% CI 1.48-2.86) independently increased discontinuation risk, whereas male sex (HR 0.65; 95% CI 0.44-0.96) and longer suppression (HR/day 0.99; P = 0.036) were protective. CONCLUSIONS: In this large national cohort, CAB+RPV LA demonstrated strong persistence, excellent tolerability, and very low rates of VF. Discontinuations were mainly structural, highlighting the need to reinforce continuity-of-care pathways and system-level support for optimal CAB+RPV LA scale-up.