Durability of insulin degludec plus liraglutide versus insulin glargine U100 as initial injectable therapy in type 2 diabetes (DUAL VIII): a multicentre, open-label, phase 3b, randomised controlled trial

Background: Durability of glycaemic control might reduce disease burden and improve long-term outcomes. DUAL VIII investigated the durability of insulin degludec plus liraglutide (IDegLira) versus insulin glargine 100 units/mL (IGlar U100) in patients with type 2 diabetes with the use of a visit schedule that mirrored routine clinical practice.

Methods: In this 104-week international, multicentre, open-label, phase 3b randomised controlled trial, insulin-naive patients aged 18 years and older, with HbA_1c between 7·0-11·0% (53-97 mmol/mol), BMI of 20 kg/m² or higher, on stable doses of oral antidiabetic drugs, were recruited from outpatient clinics. Patients were randomly assigned 1:1, with a simple sequential allocation randomisation schedule (block size of four), to IDegLira or IGlar U100, each treatment being an add-on to existing therapy. The internal safety committee, the independent external committee, and the personnel involved in defining the analysis sets were masked until the database was released for statistical analysis. Patients and all other investigators were not masked. In the IDegLira group, patients were given degludec 100 units/mL plus liraglutide 3·6 mg/mL in a 3 mL prefilled PDS290 pen for subcutaneous injection; in the IGlar U100 group, patients were given IGlar U100 solution, in a 3 mL prefilled Solostar pen for subcutaneous injection. Both treatments were given once daily at any time of day and it was recommended that the time of day remained the same throughout the trial. The primary endpoint was time from randomisation to need for treatment intensification (HbA_1c ≥7·0% [53 mmol/mol] at two consecutive visits, including week 26). Once patients met this criterion, the trial product was permanently discontinued and patients were not withdrawn from trial but rather remained on follow-up for the entire treatment and follow-up period. The primary analysis was in the intention-to-treat population. This study is registered with ClinicalTrials.gov, number NCT02501161.

Findings: From Jan 8, 2016, to Oct 3, 2018, 1345 patients were screened, of which 1012 (75·2%) were eligible and randomly assigned to either IDegLira (n=506) or IGlar U100 (n=506). 484 (96%) of 506 in the IDegLira group and 481 (95%) of 506 in the IGlar U100 group completed the trial. Baseline characteristics were similar and representative of patients eligible for basal insulin intensification (overall mean diabetes duration 10 years; HbA_1c 8·5% [69 mmol/mol]; fasting plasma glucose 10 mmol/L). Patients in the IDegLira group had significantly longer time until intensification was needed than those in the IGlar U100 group (median >2 years vs about 1 year). Fewer patients in the IDegLira group needed treatment intensification over 104 weeks than those in the IGlar U100 group (189 [37%] of 506 vs 335 [66%] of 506). The preplanned sensitivity analyses of the primary endpoint were in agreement with the primary analysis (hazard ratio 0·45 [95% CI 0·38-0·54]) in the proportional hazards regression model and the generalised log-rank test was also in favour of IDegLira (p<0·0001). No new or unexpected safety and tolerability issues were identified and there were no treatment-related deaths.

Interpretation: In patients with uncontrolled type 2 diabetes on oral antidiabetic drugs, initial injectable therapy with IDegLira resulted in fewer patients reaching the treatment intensification criterion during 104 weeks versus IGlar U100, with longer durability of the treatment effect with IDegLira.

Funding: Novo Nordisk.