Objective: Among treated patients with drug-resistant viremia, structured treatment interruptions often result in the re-emergence of drug-susceptible HIV-1. Theoretically, this may allow for a more durable response to salvage therapy. We therefore studied the long-term treatment outcome to antiretroviral therapy in a cohort of patients who had previously interrupted therapy, focusing on the determinants of treatment success versus failure. Design: A prospective observational study of the response to antiretroviral therapy in patients resuming therapy after a treatment interruption. Virological and immunological studies were performed every month for 3 months and then every 3 months. Results: Twenty-four patients underwent a structured treatment interruption and resumed therapy after a variable period of time (median 20 weeks). The median duration of treatment after the treatment interruption was 109 weeks. A transient virological response was observed in all patients who resumed a regimen containing no drug to which their pre-interruption virus was fully susceptible. Virus isolated during virological failure was genotypically and phenotypically identical to the preinterruption virus, exhibited reduced replicative capacity, and replicated in vivo at levels similar to the pre-interruption baseline. In contrast, durable viral suppression (< 200 copies/ml) was observed in patients who initiated a regimen containing only one drug to which their pre-interruption virus was fully susceptible. Despite viral suppression, the pre-interruption drug-resistant virus population remained detectable in two patients. Conclusion: Although drug-resistant HIV-1 persists at low levels during and after the interruption of therapy, durable suppression of this virus population may be achieved with a combination regimen containing only one fully active agent. (C) 2003 Lippincott Williams Wilkins.