Background Glucagon-like peptide-1 (GLP-1) receptor agonists and sodium-glucose co-transporter-2 (SGLT2) inhibitors improve glycaemic control and reduce bodyweight in patients with type 2 diabetes through different mechanisms. We assessed the safety and efficacy of the addition of the once-weekly GLP-1 receptor agonist dulaglutide to the ongoing treatment regimen in patients whose diabetes is inadequately controlled with SGLT2 inhibitors, with or without metformin. Methods AWARD-10 was a phase 3b, double-blind, parallel-arm, placebo-controlled, 24-week study done at 40 clinical sites in Austria, Czech Republic, Germany, Hungary, Israel, Mexico, Spain, and the USA. Eligible adult patients (>= 18 years) with inadequately controlled type 2 diabetes (HbA(1c) concentration >= 7.0% [ 53 mmol/mol] and >= 9.5% [ 80 mmol/mol]), a BMI of 45 kg/m(2) or less, and taking stable doses (>3 months) of an SGLT2 inhibitor (with or without metformin) were randomly assigned (1: 1: 1) via an interactive web-response system to subcutaneous injections of either dulaglutide 1.5 mg, dulaglutide 0.75 mg, or placebo once per week for 24 weeks. Patients and investigators were masked to dulaglutide and placebo assignment, and those assessing outcomes were masked to study drug assignment. The primary objective was to test for the superiority of dulaglutide (1.5 mg or 0.75 mg) versus placebo for change in HbA(1c) concentration from baseline at 24 weeks. All analyses were done in the intentionto- treat population, defined as all randomly assigned patients who received at least one dose of study drug. This study is registered with ClinicalTrials. gov, number NCT02597049. Findings Between Dec 7, 2015, and Feb 3, 2017, 424 patients were randomly assigned to dulaglutide 1.5 mg (n=142), dulaglutide 0.75 mg (n=142), and placebo (n=140). One patient in the dulaglutide 0.75 mg group was excluded from the analysis because they did not receive any dose of the study drug. The reduction in HbA(1c) concentration at 24 weeks was larger in patients receiving dulaglutide (least squares mean [LSM] for dulaglutide 1.5 mg -1.34% [SE 0.06] or -14.7 mmol/mol [0.6]; dulaglutide 0.75 mg -1.21% [0.06] or -13.2 mmol/mol [0.6]) than in patients receiving placebo (-0.54% [0.06] or -5.9 mmol/mol [0.6]; p<0.0001 for both groups vs placebo). The LSM differences were -0.79% (95% CI -0.97 to -0.61) or -8.6 mmol/mol (-10.6 to -6.7) for dulaglutide 1.5 mg and -0.66% (-0.84 to -0.49) or -7.2 mmol/mol (-9.2 to -5.4) for dulaglutide 0.75 mg (p<0. 0001 for both). Serious adverse events were reported for five (4%) patients in the dulaglutide 1.5 mg group, three (2%) patients in the dulaglutide 0.75 mg group, and five (4%) patients in the placebo group. Treatment-emergent adverse events were more common in patients treated with dulaglutide than in patients who received placebo, mainly because of an increased incidence of gastrointestinal adverse events. Nausea (21 [15%] patients in the dulaglutide 1.5 mg group vs seven [5%] in the dulaglutide 0.75 mg group vs five [4%] in the placebo group), diarrhoea (eight [6%] vs 14 [10%]vs four [3%]),and vomiting (five [4%] vs four [3%] vs one [1%]) were more common with dulaglutide than with placebo. One episode of severe hypoglycaemia was reported in the dulaglutide 0.75 mg group. Two (1%) patients receiving dulaglutide 1.5 mg died, but these deaths were not considered to be related to study drug; no deaths occurred in the other groups. Interpretation Dulaglutide as add-on treatment to SGLT2 inhibitors (with or without metformin) resulted in significant and clinically relevant improvements in glycaemic control, with acceptable tolerability that is consistent with the established safety profile of dulaglutide.