The novel prodrug tenofovir alafenamide delivers the nucleotide reverse transcriptase inhibitor tenofovir to target cells more efficiently at a lower dose than tenofovir disoproxil fumarate, thereby reducing systemic exposure. We compared the efficacy and safety of the two drugs in patients with HBeAg-negative chronic hepatitis B virus (HBV) infection in a non-inferiority study. In this ongoing randomised, double-blind, phase 3, non-inferiority study in 105 centres in 17 countries, patients with HBeAg-negative chronic HBV were randomly assigned (2:1) by a computer-generated allocation sequence (block size six), stratified by plasma HBV DNA concentration and previous treatment status, to receive once-daily oral doses of tenofovir alafenamide 25 mg or tenofovir disoproxil fumarate 300 mg, each with matching placebo. Participants, investigators, and those assessing outcomes were masked to group assignment. Eligible patients were aged at least 18 years with HBeAg-negative chronic HBV infection (with plasma HBV DNA concentrations of >20 000 IU/mL), serum alanine aminotransferase concentrations of greater than 60 U/L in men or greater than 38 U/L in women and at no more than ten times the upper limit of normal, and estimated creatinine clearance of at least 50 mL/min (by the Cockcroft-Gault method). The primary efficacy endpoint was the proportion of patients who had HBV DNA less than 29 IU/mL at week 48 in those who received at least one dose of study drug; the study was powered to show non-inferiority with a 10% efficacy margin of tenofovir alafenamide compared with tenofovir disoproxil fumarate. Bone and renal safety, and key secondary safety endpoints were assessed sequentially. The study will be conducted for a total of 3 years as a double-blind comparison to assess the longer term response to treatment. This study is registered with ClinicalTrials.gov, number NCT01940341. Between Sept 12, 2013, and Oct 31, 2014, 426 patients were randomly assigned (285 assigned to tenofovir alafenamide and 141 assigned to tenofovir disoproxil fumarate; one patient assigned to tenofovir disoproxil fumarate did not receive the treatment. 268 (94%) of 285 patients receiving tenofovir alafenamide had HBV DNA less than 29 IU/mL at week 48 versus 130 (93%) of 140 patients receiving tenofovir disoproxil fumarate (difference 1·8% [95% CI -3·6 to 7·2]; p=0·47), which demonstrates non-inferiority. Patients receiving tenofovir alafenamide had significantly smaller mean percentage declines in bone mineral density than those receiving tenofovir disoproxil fumarate (hip -0·29% [95% CI -0·55 to -0·03] vs -2·16% [-2·53 to -1·79], adjusted percentage difference 1·87% [95% CI 1·42 to 2·32; p<0·0001]; spine -0·88% [-1·22 to -0·54] vs -2·51% [-3·09 to -1·94], adjusted percentage difference 1·64% [95% CI 1·01 to 2·27]; p<0·0001). At week 48, mean change in serum creatinine was small in both groups (tenofovir alafenamide 0·01 mg/dL [95% CI 0·00 to 0·02] vs tenofovir disoproxil fumarate 0·02 mg/dL [0·00 to 0·04], adjusted percentage difference -0·01 mg/dL [95% CI -0·03 to 0·01]; p=0·32), but patients receiving tenofovir alafenamide had a smaller reduction in creatinine clearance (median change in estimated glomerular filtration rate -1·8 mL/min [IQR -7·8 to 6·0] vs -4·8 mL/min [-12·0 to 3·0]; p=0·004). Most adverse events were mild to moderate in severity in the two treatment groups. The most common adverse events overall were headache (tenofovir alafenamide 40 [14%] patients vs tenofovir disoproxil fumarate 14 [10%] patients), nasopharyngitis (30 [11%] vs 15 [11%]), and upper respiratory tract infection (35 [12%] vs ten [7%]). 14 (5%) patients receiving tenofovir alafenamide and nine (6%) patients receiving tenofovir disoproxil fumarate had serious adverse events, none of which was deemed by investigators to be related to study treatment; one patient in the tenofovir disoproxil fumarate group died, but this was not deemed to be related to study treatment. In patients with HBeAg-negative chronic HBV, the efficacy of tenofovir alafenamide was non-inferior to that of tenofovir disoproxil fumarate, and had improved bone and renal effects. Longer term follow-up is needed to better understand the clinical impact of these changes. Gilead Sciences.