Abacavir and tenofovir alafenamide offer reduced bone toxicity compared with tenofovir disoproxil fumarate. We aimed to compare safety and efficacy of tenofovir alafenamide plus emtricitabine with that of abacavir plus lamivudine. In this randomised, double-blind, active-controlled, non-inferiority phase 3 trial, HIV-1-positive adults (≥18 years) were screened at 79 sites in 11 countries in North America and Europe. Eligible participants were virologically suppressed (HIV-1 RNA <50 copies per mL) and on a stable three-drug regimen containing abacavir plus lamivudine. Participants were randomly assigned (1:1) by a computer-generated allocation sequence (block size 4) to switch to fixed-dose tablets of tenofovir alafenamide (10 mg or 25 mg) plus emtricitabine (200 mg) or remain on abacavir (600 mg) plus lamivudine (300 mg), with matching placebo, while continuing to take the third drug. Randomisation was stratified by the third drug (boosted protease inhibitor vs other drug) at screening. Investigators, participants, and study staff giving treatment, assessing outcomes, and collecting data were masked to treatment group. The primary endpoint was the proportion of participants with virological suppression (HIV-1 RNA <50 copies per mL) at week 48 (assessed by snapshot algorithm), with a 10% non-inferiority margin. We analysed the primary endpoint in participants enrolled before May 23, 2016 (when target sample size was reached), and we analysed safety in all enrolled participants who received at least one dose of study drug (including patients enrolled after these dates). This study was registered with ClinicalTrials.gov, number NCT02469246. Study enrolment began on June 29, 2015, and the cutoff enrolment date for the week 48 primary endpoint analysis was May 23, 2016. 501 participants were randomly assigned and treated. At week 48, virological suppression was maintained in 227 (90%) of 253 participants receiving tenofovir alafenamide plus emtricitabine compared with 230 (93%) of 248 receiving abacavir plus lamivudine (difference -3·0%, 95% CI -8·2 to 2·0), showing non-inferiority. Few participants discontinued treatment because of adverse events: 12 (4%) of 280 participants in the tenofovir alafenimide plus emtricitabine group and nine (3%) of 276 in the abacavir plus lamivudine group. Three participants had serious, treatment-related adverse events: one each with renal colic and neutropenia in the tenofovir alafenamide plus emtricitabine group, and one myocardial infarction in the abacavir plus lamivudine group. There were no treatment-related deaths. Tenofovir alafenamide, in combination with emtricitabine and various third drugs, maintained high efficacy with a renal and bone safety profile similar to that of abacavir. In virologically suppressed patients, a regimen containing tenofovir alafenamide could be an alternative to those containing abacavir, without concern for new onset of renal or bone toxicities or hyperlipidaemia. Gilead Sciences Inc.