Summary Background Our objective was to assess therapeutic non-inferiority of dual treatment with lopinavir–ritonavir and lamivudine to triple treatment with lopinavir–ritonavir plus two nucleos(t)ides for maintenance of HIV-1 viral suppression. Methods In this randomised, open-label, non-inferiority trial, we recruited patients from 32 HIV units in hospitals in Spain and France. Eligible patients were HIV-infected adults (aged ≥18 years) with HIV-1 RNA of less than 50 copies per mL, for at least 6 months on triple treatment with lopinavir–ritonavir (twice daily) plus lamivudine or emtricitabine and a second nucleos(t)ide, with no resistance or virological failure to these drugs, and no positive hepatitis B serum surface antigen. Investigators at each centre randomly assigned patients (1:1; block size of four; stratified by time to suppression <1 year or >1 year and nadir CD4 cell count <100 cells per μL or >100 cells per μL; computer-generated random sequence) to continue triple treatment or switch to dual treatment (oral lopinavir 400 mg and oral ritonavir 100 mg twice daily plus oral lamivudine 300 mg once daily). The primary endpoint was response to treatment in the intention-to-treat population (all randomised patients) at 48 weeks. The non-inferiority margin was 12%. This study is registered with ClinicalTrials.gov , number NCT01471821. Findings Between Oct 1, 2011, and April 1, 2013, we randomly assigned 250 participants to continue triple treatment (127 51% patients) or switch to dual treatment (123 49% patients). In the intention-to-treat population, 110 (86·6%) of 127 patients in the triple-treatment group responded to treatment versus 108 (87·8%) of 123 in the dual-treatment group (difference −1·2% 95% CI −9·6 to 7·3; p=0·92), meeting the criteria for non-inferiority. Serious adverse events occurred in eight (7%) patients in the triple-treatment group and five (4%) in the dual-treatment group (p=0·515), and study drug discontinuations due to adverse events occurred in four (3%) in the triple-treatment group and one (1%) in the dual-treatment group (p=0·223). Interpretation Dual treatment with lopinavir–ritonavir plus lamivudine has non-inferior therapeutic efficacy and is similarly tolerated to triple treatment. Funding AbbVie and Red Temática Cooperativa de Investigación en Sida.