Non-invasive diagnostic evaluation of urinary exosomal let-7c cluster expression in bladder cancer using machine learning approaches.

BACKGROUND: Bladder cancer (BCa) diagnosis typically relies on invasive cystoscopy, which is effective but costly and uncomfortable. Urinary microRNAs (miRNAs), especially exosomal ones, are promising non-invasive biomarkers due to their stability in biological fluids and disease specificity. The let-7c cluster, owing to its tumor-suppressive function and frequent dysregulation in BCa, has emerged as a promising candidate for diagnostic evaluation. This study assessed the diagnostic potential of the urinary exosomal let-7c cluster through Machine Learning (ML)-integrated miRNA profiling, offering early proof-of-concept for its role in BCa classification. METHODS: Urine samples were collected from 66 participants, including 50 BCa patients and 16 healthy controls (HC). Exosomal miRNAs were isolated and quantified using Quantitative Real-Time-Polymerase-Chain-Reaction (qRT-PCR). Statistical analysis and hypothesis tests were conducted to explore the nature and diagnostic relevance of individual biomarkers. A logistic regression classifier was applied to evaluate both the combined and differential diagnostic capabilities of the selected biomarkers. Accuracy, precision, recall and AU-ROC scores were used to assess model performance. Bioinformatics analysis was performed to identify pathways associated with the features prioritized by the ML models, ensuring their relevance to BCa. RESULTS: Statistically significant differentiation was observed between BCa patients and HC, with miR-99a-5p (p = 0.013, AU-ROC = 0.71) demonstrating preliminary diagnostic performance. In contrast, let-7c-5p (AU-ROC = 0.65, p > 0.1) and miR-125b-5p (p = 0.087, AU-ROC = 0.64) exhibited non-significant trends, showing weaker discrimination. The logistic regression ML model achieved 80.0% accuracy (AU-ROC = 0.86, recall = 90%) in distinguishing cancer from HC and 53.3% (AU-ROC = 0.63) for miRNA-only grade classification. With clinical variables added, accuracy increased to 73.3%, but the AU-ROC remained modest at 0.61, indicating limited discriminatory ability in high-grade versus low-grade. Across Ta-T2, miR-99a-5p showed clearer separation, whereas let-7c-5p and miR-125b-5p exhibited weak stage-related differences. Bioinformatics analysis confirmed the biological relevance of these miRNAs in BCa-related pathways, including PI3K-Akt, p53, NF-κB and RAS/MAPK signaling, with hub genes such as TP53, MYC, EGFR and CCND1 identified. CONCLUSION: Urinary let-7c cluster miRNAs show preliminary diagnostic potential when evaluated using ML models and may serve as a non-invasive supportive tool to conventional diagnostic approaches. However, these results should be regarded as proof-of-concept evidence that has to be validated in larger, independent cohorts because of the small sample size, low subgroup power and class imbalance.