Background: Antiretroviral therapy (ART) has evolved rapidly since its beginnings. This analysis describes trends in first-line ART use in Asia and their impact on treatment outcomes.
Methods: Patients in the TREAT Asia HIV Observational Database receiving first-line ART for ≥ 6 months were included. Predictors of treatment failure and treatment modification were assessed.
Results: Data from 4662 eligible patients was analysed. Patients started ART in 2003-2006 (n = 1419), 2007-2010 (n = 2690) and 2011-2013 (n = 553). During the observation period, tenofovir, zidovudine and abacavir use largely replaced stavudine. Stavudine was prescribed to 5.8% of ART starters in 2012/13. Efavirenz use increased at the expense of nevirapine, although both continue to be used extensively (47.5% and 34.5% of patients in 2012/13, respectively). Protease inhibitor use dropped after 2004. The rate of treatment failure or modification declined over time (22.1 [95%CI 20.7-23.5] events per 100 patient/years in 2003-2006, 15.8 [14.9-16.8] in 2007-2010, and 11.6 [9.4-14.2] in 2011-2013). Adjustment for ART regimen had little impact on the temporal decline in treatment failure rates but substantially attenuated the temporal decline in rates of modification due to adverse event. In the final multivariate model, treatment modification due to adverse event was significantly predicted by earlier period of ART initiation (hazard ratio 0.52 [95%CI 0.33-0.81], p = 0.004 for 2011-2013 versus 2003-2006), older age (1.56 [1.19-2.04], p = 0.001 for ≥ 50 years versus <30 years), female sex (1.29 [1.11-1.50], p = 0.001 versus male), positive hepatitis C status (1.33 [1.06-1.66], p = 0.013 versus negative), and ART regimen (11.36 [6.28-20.54], p<0.001 for stavudine-based regimens versus tenofovir-based).
Conclusions: The observed trends in first-line ART use in Asia reflect changes in drug availability, global treatment recommendations and prescriber preferences over the past decade. These changes have contributed to a declining rate of treatment modification due to adverse event, but not to reductions in treatment failure.