Abstract
BACKGROUND: Bladder cancer (BC) screening contributes to better patient outcomes. However, existing diagnostic methods for BC are not suitable for large-scale screening. Serum microRNAs (miRNAs) are expected to be simple, cost-effective, and non-invasive screening tools for BC. This study aims to identify whether relevant serum miRNAs as diagnostic markers for BC. METHODS: This research covered 112 BCs and 112 healthy controls (HCs), utilizing quantitative reverse transcription polymerase chain reaction (RT-qPCR) to pinpoint miRNAs viable for diagnosing BC in serum across three phases. In addition, bioinformatics analyses were conducted to predict target genes and provide functional annotations. RESULTS: Six serum miRNAs showed significantly abnormal expression levels compared with the HC group. A highly efficient screening model, comprising four serum miRNAs (hsa-miR-107, hsa-miR-146a-5p, hsa-miR-15b-5p, and hsa-miR-218-5p) was developed, with an area under the receiver operating characteristic curve (AUC) value of 0.924, 95.24% sensitivity, and 78.57% specificity. The predicted target genes of these serum miRNAs, including ADAMTSL3, CACNB2, FOXP2 and GAS7, may play a role in the initiation and progression of BC. CONCLUSIONS: A panel of four serum miRNAs (hsa-miR-107, hsa-miR-146a-5p, hsa-miR-15b-5p, and hsa-miR-218-5p) exhibits high specificity and sensitivity. It shows potential as an efficient, non-invasive, and cost-effective biomarker for BC screening.