Background: Oral semaglutide is the first oral glucagon-like peptide-1 (GLP-1) receptor agonist for glycaemic control in patients with type 2 diabetes. Type 2 diabetes is commonly associated with renal impairment, restricting treatment options. We aimed to investigate the efficacy and safety of oral semaglutide in patients with type 2 diabetes and moderate renal impairment. Methods: This randomised, double-blind, phase 3a trial was undertaken at 88 sites in eight countries. Patients aged 18 years and older, with type 2 diabetes, an estimated glomerular filtration rate of 30–59 mL/min per 1·73 m², and who had been receiving a stable dose of metformin or sulfonylurea, or both, or basal insulin with or without metformin for the past 90 days were eligible. Participants were randomly assigned (1:1) by use of an interactive web-response system, with stratification by glucose-lowering medication and renal function, to receive oral semaglutide (dose escalated to 14 mg once daily) or matching placebo for 26 weeks, in addition to background medication. Participants and site staff were masked to assignment. Two efficacy-related estimands were defined: treatment policy (regardless of treatment discontinuation or rescue medication) and trial product (on treatment without rescue medication) in all participants randomly assigned. Endpoints were change from baseline to week 26 in HbA_1c (primary endpoint) and bodyweight (confirmatory secondary endpoint), assessed in all participants with sufficient data. Safety was assessed in all participants who received at least one dose of study drug. This trial is registered on ClinicalTrials.gov, number NCT02827708, and the European Clinical Trials Registry, number EudraCT 2015-005326-19, and is now complete. Findings: Between Sept 20, 2016, and Sept 29, 2017, of 721 patients screened, 324 were eligible and randomly assigned to oral semaglutide (n=163) or placebo (n=161). Mean age at baseline was 70 years (SD 8), and 168 (52%) of participants were female. 133 (82%) participants in the oral semaglutide group and 141 (88%) in the placebo group completed 26 weeks on treatment. At 26 weeks, oral semaglutide was superior to placebo in decreasing HbA_1c (estimated mean change of −1·0 percentage point (SE 0·1; −11 mmol/mol [SE 0·8]) vs −0·2 percentage points (SE 0·1; −2 mmol/mol [SE 0·8]); estimated treatment difference [ETD]: −0·8 percentage points, 95% CI −1·0 to −0·6; p<0·0001) and bodyweight (estimated mean change of −3·4 kg [SE 0·3] vs −0·9 kg [SE 0·3]; ETD, −2·5, 95% CI −3·2 to −1·8; p<0·0001) by the treatment policy estimand. Significant differences were seen for the trial product estimand: mean change in HbA_1c −1·1 percentage points (SE 0·1; −12 mmol/mol [SE 0·8] versus −0·1 percentage points (SE 0·1; −1 mmol/mol [SE 0·8]; ETD −1·0 percentage points, 95% CI −1·2 to −0·8; p<0·0001); mean change in bodyweight −3·7 kg (SE 0·3) versus −1·1 kg (SE 0·3; ETD −2·7 kg, 95% CI −3·5 to −1·9; p<0·0001). More patients taking oral semaglutide than placebo had adverse events (120 [74%] of 163 vs 105 [65%] of 161), and discontinued treatment as a result (24 [15%] vs eight [5%]). Gastrointestinal events, mainly mild-to-moderate nausea, were more common with oral semaglutide than with placebo. Three deaths occurred during the treatment period that were not condsidered to be treatment related, one in the semaglutide group and two in the placebo group. Interpretation: Oral semaglutide was effective in patients with type 2 diabetes and moderate renal impairment, potentially providing a new treatment option for this population. Safety, including renal safety, was consistent with the GLP-1 receptor agonist class. Funding: Novo Nordisk A/S.