OBJECTIVE: To study the rate of treatment change due to toxicities in patients who achieved viral suppression within 6 months of starting antiretroviral therapy and who have never experienced virological failure. METHODS: Included patients attended the Royal Free Hospital in London, started antiretroviral therapy in 2000-2005, and achieved viral suppression within 6 months. Included follow-up (censored at virological failure) was spent on a regimen of lamivudine or emtricitabine, with a second nucleoside/nucleotide reverse transcriptase inhibitor, and either a protease inhibitor or a nonnucleoside reverse transcriptase inhibitor. RESULTS: In 912 person-years, there were 140 treatment changes due to toxicities (rate = 15.4 per 100 person-years, confidence interval = 12.8-17.9). In the multivariable analysis, factors associated with a higher rate of treatment change due to toxicities included increased age (for every 10 years increase: incidence rate ratio = 1.28, confidence interval = 1.04-1.57), being on stavudine compared with zidovudine (incidence rate ratio = 2.04, confidence interval = 1.28-3.26), and being on lopinavir compared with efavirenz (incidence rate ratio = 1.55, confidence interval = 1.04-2.31), whereas factors associated with a lower rate were being on tenofovir compared with zidovudine (incidence rate ratio = 0.46, confidence interval = 0.29-0.73), and being on atazanavir compared with efavirenz (incidence rate ratio = 0.23, confidence interval = 0.06-0.91). CONCLUSIONS: In patients who have never experienced virological failure, the rate of treatment change due to toxicities is low, and certain regimens are associated with an even lower rate of change. If virological failure is avoided, some regimens are so far proving to be sufficiently stable to suggest that very long-term use is potentially feasible.