OBJECTIVES: The aim of the study was to investigate the frequency and severity of adverse events (AEs) and laboratory abnormalities of interest over 96 weeks of treatment with etravirine or placebo in the pooled TMC125 DUET (Demonstrate Undetectable viral load in patients Experienced with ARV Therapy) trials. METHODS: Treatment-experienced, HIV-1-infected patients randomly received etravirine 200 mg twice a day (bid) or placebo, plus a background regimen. The frequency and severity of neuropsychiatric, rash, hepatic and lipid AEs were analysed; frequencies were also adjusted for total patient-years of exposure (PYE). RESULTS: A total of 599 and 604 patients received etravirine and placebo, respectively (median treatment duration 96.0 and 69.6 weeks, respectively). There was no significant difference between the treatment groups in the frequency of neuropsychiatric AEs. However, a significant difference in the frequency of rash was observed (20.5% vs. 11.8%, respectively; P < 0.0001); rash was generally mild to moderate in severity; the rate of discontinuation because of rash was low (2.2% vs. 0% in the etravirine and placebo groups, respectively). The frequency of hepatic AEs was low and similar between the treatment groups (8.7% vs. 7.1%, respectively; P = 0.3370); hepatic enzyme levels did not increase over time. Lipid-related laboratory abnormalities and changes over time in lipid levels were generally comparable between treatment groups. Adjusting for treatment exposure, the frequency of AEs remained similar between treatment groups, with the exception of rash [13.7 vs. 9.3 per 100 PYE; relative risk (95% confidence interval) 1.48 (1.02-1.95)]. CONCLUSIONS: The frequency of AEs of interest was generally similar between the treatment groups, both overall and when adjusted for treatment exposure, with the exception of rash which was more frequent in the etravirine group.