Summary: In a trial in type 2 diabetes inadequately controlled on metformin and glyburide, adding either sitagliptin or bedtime NPH insulin produced similar HbA1c reductions but distinct changes in circulating microRNAs, with sitagliptin affecting a broader set, supporting effects beyond glucose lowering.
PICO Summary
| Element | Detail |
|---|---|
| Population | 32 patients with type 2 diabetes inadequately controlled on metformin and glyburide; 6-month randomised trial, Brazil. |
| Intervention | Addition of sitagliptin (a DPP-4 inhibitor). |
| Comparison | Addition of bedtime NPH insulin (the two were separate randomised arms). |
| Outcome | Both treatments produced a similar HbA1c reduction; only sitagliptin increased postprandial GLP-1. Sitagliptin upregulated six microRNAs (including miR-24-3p, miR-27a-3p, let-7d-5p) and NPH upregulated four (including miR-193b-3p, miR-320a-3p); both raised miR-92a-3p and miR-30c-5p. Pathways related to insulin signalling, inflammation, senescence, and atherosclerosis were enriched. |
Expert Commentary
This is a mechanistic study best understood as a comparison of two intensification strategies that achieve the same glycaemic result by different molecular routes, rather than a test of a combination, and getting that design right matters for interpretation. The clinically grounding finding is that sitagliptin and bedtime NPH lowered HbA1c similarly, so the interest lies in the divergent microRNA signatures, with sitagliptin modulating a broader panel and uniquely raising postprandial GLP-1, consistent with the idea that DPP-4 inhibition exerts pleiotropic effects on insulin signalling, inflammation, and vascular biology beyond glucose. I would treat the molecular results as hypothesis-generating in the strongest sense. The sample is very small at 32, microRNA expression is a surrogate far upstream of any clinical outcome, and the enriched pathways are computational predictions rather than demonstrated effects, so none of this should be read as proof of differential patient benefit. Can I use this with my patients? Not directly. It does not change whether I choose a DPP-4 inhibitor or basal insulin for a given patient, a decision still driven by efficacy needs, hypoglycaemia and weight considerations, cost, and comorbidity, but it adds biological texture to why agents with similar HbA1c effects may differ in their wider actions, pending much larger outcome-focused work.
References
Santos AS, Bando SY, Correa Giannella MLC, da Silva MER. MicroRNAs modulated by DPP-4 inhibitor and bedtime NPH insulin therapy in individuals with type 2 diabetes. Front Endocrinol (Lausanne). 2025;16:1706951. doi:10.3389/fendo.2025.1706951