Development of an miRNA signature for individual prognostic assessment in seminoma patients.

DOI: 10.1200/jco.2025.43.5_suppl.645 Publication Date: 2025-02-18T14:32:12Z
ABSTRACT
645 Background: Due to the risk of long-term toxicities in young non-metastatic seminoma patients, surveillance regimens nowadays are standard care. However, up 20% patients will relapse and undergo further treatment, requiring even greater efforts terms adjuvant therapy. miRNAs known be very stable robust biomarkers. This study aims establish a prognostic miRNA profile utilizing bioinformatics tools differentiate between metastatic (met) (nmet) seminomas. Methods: MiRNA was extracted from formalin-fixed paraffin-embedded sections. Using microarray analysis (Agilent Technology), two cohorts with (24 each, 12 nmet met) were examined. The following criteria used preselect relevant miRNAs: absolute median log2 fold change >0.5, distance correlation accuracy single-miRNA classifier trained leave-one-out cross-validation (met vs. nmet) >0.7. Validation performed 88 (47 41 through qPCR using TaqMan assay, RNU48, miR-191-5p, miR-361-5p serving as references. Statistical analyses Mann-Whitney U test machine learning algorithms. Results: microarrays identified 16 exhibiting an >70% differentiating met tumors ones. validation 6 by revealed significant expression differences for miR-371a-5p, miR-512-3p, miR−509−3−5p, miR-517a-3p (p<0.05), dependent on reference used. All 4 higher expressed tumors. Synchronous exhibit stronger above-mentioned compared metachronous tumors, except miR-509-3-5p. miR-371a-5p found most distinguish both synchronous Classifiers PCR data different combinations investigated show medium performance (average 0.59) including one model based high 0.99. Conclusions: By machine-learning algorithms data, we have established signature that can seminomas reliability. confirms value patients. Currently, examine suitability this panel liquid biopsy.
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES (0)
CITATIONS (0)