Familywise error for multiple time-to-event endpoints in a group sequential design.

We investigate the familywise error rate (FWER) for time-to-event endpoints evaluated using a group sequential design with a hierarchical testing procedure for secondary endpoints. We show that, in this setup, the correlation between the log-rank test statistics at interim and at end of study is not congruent with the canonical correlation derived for normal-distributed endpoints. We show, both theoretically and by simulation, that the correlation also depends on the level of censoring, the hazard rates of the endpoints, and the hazard ratio. To optimize operating characteristics in this complex scenario, we propose a simulation-based method to assess the FWER which, better than the alpha-spending approach, can inform the choice of critical values for testing secondary endpoints.

Efficacy and safety of a fixed-ratio combination of insulin degludec and liraglutide (IDegLira) compared with its components given alone: results of a phase 3, open-label, randomised, 26-week, treat-to-target trial in insulin-naive patients with type 2 diabetes.

BACKGROUND: A fixed-ratio combination of the basal insulin analogue insulin degludec and the glucagon-like peptide-1 (GLP-1) analogue liraglutide has been developed as a once-daily injection for the treatment of type 2 diabetes. We aimed to compare combined insulin degludec-liraglutide (IDegLira) with its components given alone in insulin-naive patients. METHODS: In this phase 3, 26-week, open-label, randomised trial, adults with type 2 diabetes, HbA1c of 7-10% (inclusive), a BMI of 40 kg/m(2) or less, and treated with metformin with or without pioglitazone were randomly assigned (2:1:1) to daily injections of IDegLira, insulin degludec, or liraglutide (1·8 mg per day). IDegLira and insulin degludec were titrated to achieve a self-measured prebreakfast plasma glucose concentration of 4-5 mmol/L. The primary endpoint was change in HbA1c after 26 weeks of treatment, and the main objective was to assess the non-inferiority of IDegLira to insulin degludec (with an upper 95% CI margin of 0·3%), and the superiority of IDegLira to liraglutide (with a lower 95% CI margin of 0%). This study is registered with ClinicalTrials.gov, number NCT01336023. FINDINGS: 1663 adults (mean age 55 years [SD 10], HbA1c 8·3% [0·9], and BMI 31·2 kg/m(2) [4·8]) were randomly assigned, 834 to IDegLira, 414 to insulin degludec, and 415 to liraglutide. After 26 weeks, mean HbA1c had decreased by 1·9% (SD 1·1) to 6·4% (1·0) with IDegLira, by 1·4% (1·0) to 6·9% (1·1) with insulin degludec, and by 1·3% (1·1) to 7·0% (1·2) with liraglutide. IDegLira was non-inferior to insulin degludec (estimated treatment difference -0·47%, 95% CI -0·58 to -0·36, p<0·0001) and superior to liraglutide (-0·64%, -0·75 to -0·53, p<0·0001). IDegLira was generally well tolerated; fewer participants in the IDegLira group than in the liraglutide group reported gastrointestinal adverse events (nausea 8·8 vs 19·7%), although the insulin degludec group had the fewest participants with gastrointestinal adverse events (nausea 3·6%). We noted no clinically relevant differences between treatments with respect to standard safety assessments, and the safety profile of IDegLira reflected those of its component parts. The number of confirmed hypoglycaemic events per patient year was 1·8 for IDegLira, 0·2 for liraglutide, and 2·6 for insulin degludec. Serious adverse events occurred in 19 (2%) of 825 patients in the IDegLira group, eight (2%) of 412 in the insulin degludec group, and 14 (3%) of 412 in the liraglutide group. INTERPRETATION: IDegLira combines the clinical advantages of basal insulin and GLP-1 receptor agonist treatment, resulting in improved glycaemic control compared with its components given alone. FUNDING: Novo Nordisk.