Third-Line Antidiabetic Therapy Intensification Patterns and Glycaemic Control in Patients with Type 2 Diabetes in the USA: A Real-World Study

摘要

Background Third-line antidiabetic drug (ADD) intensification patterns and glycemic control post intensification in type 2 diabetes mellitus (T2DM) have not been thoroughly explored in a real-world setting. Objective This study explored the patterns and risks of third-line ADD intensification post second-line ADDs and the probability of desirable glucose control over 12 months by third-line ADD classes at the population level. Methods We used the electronic medical records of 255,236 patients with T2DM in the USA initiating a second-line ADD post metformin from January 2013 to evaluate the rates and risks of third-line intensification and the probability of desirable glycemic control with different ADDs after addressing inherent heterogeneity using appropriate methodologies. Results Patients had a mean age of 60 years and glycated hemoglobin (HbA1c) of 8.5% at second-line ADD. Over 209,136 person-years (PY) of follow-up, 40% had initiated a third-line ADD at HbA1c of 8.8%. Patients receiving dipeptidyl peptidase-4 inhibitors (DPP-4i) and glucagon-like peptide-1 receptor agonists (GLP-1 RAs) as the second-line ADD had a 7% (95% hazard ratio [HR] confidence interval [CI] 1.05-1.10) and 28% (95% HR CI 1.24-1.33) higher adjusted risk of intensifying with a third-line ADD than did those receiving sulfonylureas as the second-line ADD. Those receiving sodium-glucose cotransporter-2 inhibitors (SGLT-2i) as second-line ADD had a 17% (95% HR CI 0.80-0.87) lower risk. The adjusted probability of reducing HbA1c by >= 1% was similar in those receiving third-line sulfonylureas, thiazolidinediones, GLP-1 RAs, SGLT-2i, and insulin (minimum, maximum 95% CI of probability 0.61, 0.68), whereas those receiving DPP-4i had a significantly lower probability (0.58; 95% CI 0.56-0.59). Similarly, the probability of reducing HbA1c < 7.5% was similar in the sulfonylurea, GLP-1 RA, and SGLT-2i groups (minimum, maximum of 95% CI of probability 0.41, 0.49), whereas those receiving DPP-4i had a significantly lower probability of achieving an HbA1c < 7.5% (0.37; 95% CI 0.36-0.38). Conclusion This study, based on a large representative cohort of patients with T2DM from the USA, suggests the need for revisiting real-world practices in choosing therapeutic intensification pathways and a more proactive strategy to tackle the persistent risk factor burden in patients with T2DM.