Efficacy and safety of linagliptin according to patient baseline characteristics: A pooled analysis of three phase 3 trials.

BACKGROUND AND AIMS

We aimed to determine if patient baseline characteristics affect responses to linagliptin and identify relevant predictors of glycated hemoglobin (HbA1c) reduction in patients with type 2 diabetes mellitus (T2DM).

METHODS AND RESULTS

Data were pooled from three 24-week, placebo-controlled trials of similar design (linagliptin, n = 1651; placebo, n = 607). Patients were categorized according to baseline characteristics: age, T2DM duration, gender, body mass index (BMI), Homeostasis Model Assessment of Insulin Resistance (HOMA-IR), and metabolic syndrome (MetS). Changes from baseline in HbA1c after 24 weeks were assessed with analysis of covariance (ANCOVA). The proportion of patients with baseline HbA1c >7% achieving HbA1c of ≤7% at week 24 were evaluated. Independent predictors of HbA1c response with linagliptin were analyzed in a multivariate analysis with ANCOVA. Linagliptin treatment led to significant mean (SE) placebo-corrected reductions from baseline in HbA1c across all subgroups (-0.42% [±0.11] to -0.79% [0.08]; all p < 0.001). Within subgroups, HbA1c reduction was more pronounced in patients without MetS (-0.74% [0.06]; treatment interaction p = 0.0489). The proportion of patients with baseline HbA1c >7% achieving a target HbA1c ≤7% was greater with linagliptin versus placebo (30.2% vs 11.5%; odds ratio 3.82; 95% CI 2.82 to 5.17; p < 0.001). Characteristics significantly predicting HbA1c reductions after 24 weeks were fasting plasma glucose and race (both p < 0.05).

CONCLUSION

This post-hoc analysis supports that linagliptin achieved clinically meaningful improvements in hyperglycemia in patients with diverse clinical characteristics. These improvements were more pronounced in patients without MetS.