Comparative effectiveness of pioglitazone and sitagliptin added to metformin and sulfonylurea in patients with type 2 diabetes

Dipeptidyl peptidase-4 inhibitors and thiazolidinediones are commonly used as add-ons in type 2 diabetes mellitus (T2DM) treatment, but comparative data are limited. This study emulated a target trial to compare sitagliptin and pioglitazone in patients on metformin and sulfonylurea, focusing on changes in glycemic control and clinical outcomes. This emulated target trial comprised 68,372 patients with T2DM identified from 4 university hospitals in South Korea (2001-2024). Patients with T2DM receiving both metformin and sulfonylurea were included and assigned to groups receiving either pioglitazone or sitagliptin as add-on therapy using 1:1 propensity score matching. Clinical indicators, including glycemic parameters, were evaluated for differences between groups using unpaired t-tests during the 24-week follow-up. To assess the possibility of achieving intensive glycemic control (HbA1c < 6.5%), Cox proportional hazards analysis was performed, reporting adjusted hazard ratio (aHR) with 95% confidence intervals (CIs). Following 1:1 propensity score matching, 1742 participants were assigned to each of the pioglitazone (mean age, 58.75 years; 64.7% male) and sitagliptin (mean age, 58.95 years; 64.7% male) groups. Both pioglitazone and sitagliptin reduced HbA1c over 24 weeks with no significant between-group difference (pioglitazone: -0.48% [95% CI, -0.55 to -0.41]; sitagliptin: -0.55% [95% CI, -0.61 to -0.48]; P-value = .182 for the between-group difference). However, sitagliptin was associated with a higher possibility of achieving HbA1c < 6.5% (aHR, 1.28 [95% CI, 1.08-1.53]), and showed greater reductions in body weight (pioglitazone: 0.26 kg [95% CI, -0.37 to 0.89]; sitagliptin: -0.91 kg [95% CI, -1.65 to -0.17]; P-value = .018 for the between-group difference) and LDL cholesterol (pioglitazone: -2.22 mmol/L [95% CI, -7.41 to 2.96]; sitagliptin: -11.42 mmol/L [95% CI, -16.95 to -5.89]; P-value = .017 for the between-group difference). Our findings suggest that sitagliptin showed more favorable metabolic profiles and a higher likelihood of achieving intensive glycemic targets compared with pioglitazone, beyond average glycemic control.