Baseline factors associated with glycaemic response to treatment with once-weekly dulaglutide in patients with type 2 diabetes.

Dulaglutide glycaemic efficacy has been demonstrated in the AWARD clinical trial programme. The objective of the present analysis was to determine the major baseline factors associated with the reduction in glycated haemoglobin (HbA1c) in response to dulaglutide. Baseline covariates from patients receiving dulaglutide in six phase III studies (n = 2806) were analysed using a gradient-boosting method to assess their relative influence on the change in HbA1c after 26 weeks of treatment. Influential variables (relative influence >5%) were further evaluated in univariate and multivariable modelling. The gradient-boosting analysis showed that the top influential baseline factors associated with HbA1c reduction were: HbA1c (48.8%), age (9.1%), fasting serum glucose (FSG; 8.2%), fasting serum insulin (FSI; 6.7%) and estimated glomerular filtration rate (eGFR; 5.4%). Multivariable regression showed that higher baseline HbA1c was the major factor associated with greater HbA1c reduction [coefficient estimates: -0.6% (-6.6 mmol/mol); p < 0.0001]. Age ≤65 years, lower FSG level, FSI level ≤55 pmol/L and eGFR ≤100 mL/min/1.73 m were associated with greater decreases in HbA1c, but the effect was very small [coefficient estimates: -0.05% to -0.2% (-0.6 to -2.2 mmol/mol)]. These data indicate that higher baseline HbA1c, reflecting poor glycaemic status, is the major factor associated with greater reduction in HbA1c in response to dulaglutide treatment.