Treatment modification in HIV-Infected individuals starting antiretroviral therapy between 2011 and 2014

INTRODUCTION While antiretroviral therapy (ART) has increased the survival of HIV patients and turned HIV infection into a chronic condition, treatment modifications and poor adherence might limit this therapeutic success. METHODS Patients from the Austrian HIV Cohort Study, who started their first ART after Rilpivirine became available in February 2011, were analyzed for factors associated with treatment modification which could be either a change of drugs or a stop of the regimen. A drug was considered as stopped when the regimen was interrupted for more than eight days. Drugs of particular interest were Darunavir (DRV), Atazanavir (ATV), Raltegravir (RAL), Rilpivirine (RPV) and Efavirenz (EFV). RPV and EFV were analyzed only when taken as single tablet regimen. Other drugs were summarized as "other." Proportional hazards regression methods were used to identify predictors of discontinuation and Kaplan-Meier estimates were used to calculate probabilities of discontinuation. Patients who died were censored at the date of death. RESULTS 965 patients started ART, 282 with DRV, 161 with ATV, 96 with RAL, 108 with RPV and 118 with EFV. Median time for taking initial ART is 11.6 months. 322 (33.4%) patients modified their initial ART. The overall probability of modification at one year was 28.7%, at two years 40.0% and at three years 49.8%. In a multivariable proportional hazards regression analysis, AIDS diagnosis at baseline and injecting drug use (IDU) of men compared with men who have sex with men (MSM) have a higher risk of switch/stop. Compared with DRV, RPV showed a much lower and ATV and particularly "other" a higher risk for discontinuation (Table 1). CONCLUSION Rates of modification and interruption were still high in recent years, particularly in the first year of ART. The decreased rate of modification found in patients treated with Rilpivirine may be attributed to selection of patients according to guidelines.