METHODS: DIAMOND was a randomised, double-blind, placebo-controlled crossover trial done at six hospitals in Canada, Malaysia, and the Netherlands. Eligible participants were adult patients (aged 18-75 years) with chronic kidney disease, without a diagnosis of diabetes, with a 24-h urinary protein excretion greater than 500 mg and less than or equal to 3500 mg and an estimated glomerular filtration rate (eGFR) of at least 25 mL/min per 1·73 m2, and who were on stable renin-angiotensin system blockade. Participants were randomly assigned (1:1) to receive placebo and then dapagliflozin 10 mg per day or vice versa. Each treatment period lasted 6 weeks with a 6-week washout period in between. Participants, investigators, and study personnel were masked to assignment throughout the trial and analysis. The primary outcome was percentage change from baseline in 24-h proteinuria during dapagliflozin treatment relative to placebo. Secondary outcomes were changes in measured GFR (mGFR; via iohexol clearance), bodyweight, blood pressure, and concentrations of neurohormonal biomarkers. Analyses were done in accordance with the intention-to-treat principle. This study is registered with ClinicalTrials.gov, NCT03190694.

FINDINGS: Between Nov 22, 2017, and April 5, 2019, 58 patients were screened, of whom 53 (mean age 51 years [SD 13]; 32% women) were randomly assigned (27 received dapagliflozin then placebo and 26 received placebo then dapagliflozin). One patient discontinued during the first treatment period. All patients were included in the analysis. Mean baseline mGFR was 58·3 mL/min per 1·73 m2 (SD 23), median proteinuria was 1110 mg per 24 h (IQR 730-1560), and mean HbA1c was 5·6% (SD 0·4). The difference in mean proteinuria change from baseline between dapagliflozin and placebo was 0·9% (95% CI -16·6 to 22·1; p=0·93). Compared with placebo, mGFR was changed with dapagliflozin treatment by -6·6 mL/min per 1·73 m2 (-9·0 to -4·2; p<0·0001) at week 6. This reduction was fully reversible within 6 weeks after dapagliflozin discontinuation. Compared with placebo, bodyweight was reduced by 1·5 kg (0·03-3·0; p=0·046) with dapagliflozin; changes in systolic and diastolic blood pressure and concentrations of neurohormonal biomarkers did not differ significantly between dapagliflozin and placebo treatment. The numbers of patients who had one or more adverse events during dapagliflozin treatment (17 [32%] of 53) and during placebo treatment (13 [25%] of 52) were similar. No hypoglycaemic events were reported and no deaths occurred.

INTERPRETATION: 6-week treatment with dapagliflozin did not affect proteinuria in patients with chronic kidney disease without diabetes, but did induce an acute and reversible decline in mGFR and a reduction in bodyweight. Long-term clinical trials are underway to determine whether SGLT2 inhibitors can safely reduce the rate of major clinical kidney outcomes in patients with chronic kidney disease with and without diabetes.