尿性外来小胞由来非コードRNAの糖尿病性腎症における系統的レビュー：発現プロファイル、臨床相関、および診断性能

Diabetic nephropathy (DN) is a primary contributor to end-stage renal disease, and non-coding RNAs (ncRNAs) extracted from urinary extracellular vesicles (uEVs) are being investigated as prospective biomarkers. This study reviewed the expression patterns, clinical associations, and diagnostic efficacy of them in DN cases. Following the PRISMA 2020 guidelines, PubMed, Embase, and Web of Science were systematically searched up to March 15, 2025 (PROSPERO: CRD420251039771). Eligible studies included primary human observational studies comparing uEV-derived ncRNAs in DN patients with non-DN diabetic subjects and healthy controls. Information on ncRNA expression, how it relates to clinical factors, and its performance as a diagnostic test was included. Bias assessment was done using the ROBINS-I tool. Twenty-four studies met the inclusion criteria. A total of 197 uEV-derived miRNAs were identified, while only one study evaluated other ncRNA species (four lncRNAs and two circRNAs). These miRNAs exhibited significant relationships with renal function markers, such as uACR, eGFR, serum creatinine, and BUN. Different miRNA signatures, such as miR-30a, miR-24-3p, and miR-27b-3p, were associated with stages of albuminuria. There were also associations between some miRNAs and glycemic indices, blood pressure, and lipid profiles. Individual miRNAs showed strong diagnostic efficacy, with miR-636, miR-34a, and miR-15b achieving AUCs of over 0.88. Combination biomarker panels significantly improved the ability to distinguish between DN stages. These results validate uEV-derived ncRNAs as sensitive indicators of DN development and prospective noninvasive biomarkers. However, standardized analytical procedures are necessary to make results more reproducible.