Background: Thai patients with HIV have higher exposure to HIV protease inhibitors than do white people and dose reduction might be possible. We compared the efficacy of low-dose with standard-dose ritonavir-boosted atazanavir in virologically suppressed Thai patients with HIV.

Methods: In this randomised, open-label, non-inferiority trial, we recruited patients aged 18 years or older who were receiving ritonavir-boosted protease-inhibitor-based antiretroviral therapy (ART) with HIV plasma viral loads of less than 50 copies per mL, an alanine aminotransferase concentration of less than 200 IU/L, and a creatinine clearance of at least 60 mL/min from 14 hospitals in Thailand. We excluded patients who had active AIDS-defining disease or opportunistic infections, had a history of an HIV viral load of 1000 copies per mL or more after 24 weeks of any ritonavir-boosted protease-inhibitor-based ART, used concomitant medications that could interact with the study drugs, were pregnant or lactating, had illnesses that might change the effect of the study drugs, or had a history of sensitivity to the study drugs. A biostatistician at the study coordinating centre randomly allocated patients (1:1) to switch the protease inhibitor for oral atazanavir 200 mg and ritonavir 100 mg or for atazanavir 300 mg and ritonavir 100 mg once daily, both with two nucleoside or nucleotide reverse transcriptase inhibitors at recommended doses. Randomisation was done with a minimisation schedule, stratified by recruiting centre, use of tenofovir, and use of indinavir as a component of the preswitch regimen. The primary endpoint was the proportion of patients with viral loads of less than 200 copies per mL at week 48, and we followed up patients every 12 weeks. Treatments were open label, the non-inferiority margin was −10%, and all patients who received at least one dose of study medication were analysed. This trial is registered with ClinicalTrials.gov, number NCT01159223.

Findings: Between July 6, 2011, and Dec 23, 2013, we randomly assigned 559 patients: 279 to receive atazanavir 200 mg and ritonavir 100 mg (low dose) and 280 to atazanavir 300 mg and ritonavir 100 mg (standard dose). At week 48, 265 (97·1%) of 273 in the low-dose group and 267 (96·4%) of 277 in the standard-dose group had viral loads of less than 200 copies per mL (difference 0·68; 95% CI −2·29 to 3·65). Seven (3%) of 273 in the low-dose group and 21 (8%) of 277 in the standard-dose group discontinued their assigned treatment (p=0·01). 46 (17%) of 273 participants in the low-dose group and 97 (35%) of 277 in the standard-dose group had total bilirubin grade 3 or higher toxicity (≥3·12 mg/dL; p<0·0001).

Interpretation: A switch to low-dose atazanavir should be recommended for Thai patients with well controlled HIV viraemia while on regimens based on boosted protease inhibitors.