Antiretroviral Therapy Switch Rates and Switching Patterns in 16,069 People Living With HIV: A Nationwide, Population-Based Study in Japan

Abstract

To report the status of switch rates and time-to-switch of antiretroviral therapy (ART) regimens by evaluating anchor drug classes and common switching patterns in Japanese people living with human immunode ciency virus (HIV, PLWH). This cross-sectional cohort study extracted data of 28,089 PLWH from the National Database of Health Insurance Claims and Speci c Health Checkups of Japan (NDB), which contains data representing the entire population of Japan. PLWH with rst prescription records of ART administered between January 2011 and March 2019 were identi ed (n = 16,069). The median timeto-switch and switch rates of anchor drug classes were estimated by Kaplan-Meier analysis. Brookmeyer-Crowley and Greenwood methods were used to estimate 95% con dence intervals for switch rates and median days, respectively. Switch rates were compared between anchor drug classes by year using logrank tests. A total of 3,108 (19•3%) PLWH switched anchor drug classes from rst to second regimens. Switch rates increased continuously over eight years for non-nucleoside reverse transcriptase inhibitors (NNRTIs) (14•9%-65•5%) and protease inhibitors (PIs) (13•2%-67•7%), with median time-to-switch of 1,826 and 1,583 days, respectively. Integrase strand transfer inhibitors (INSTIs) maintained a low switch rate (3•0%-7•6%), precluding median-days calculation. The majority of patients treated initially with NNRTIs and PIs switched to INSTIs regardless of switching times (< 1 year: 67•3% and 85•9%, respectively; ≥1 year: 95•5% and 93•6%, respectively). The foremost switching strategies for rst-to-second ART regimens are from NNRTI or PI to INSTI regimens that maintain low switch rates long term. INSTI HIV agents may be the most durable anchor drug class for PLWH receiving ART.