Key points are not available for this paper at this time.
Negative tumor markers are suggested to distinguish germinoma from non-germ cell tumors (NGGCT) but not non-intracranial germ cell tumor (non-iGCT). Accurate and non-invasive diagnostic markers are crucial to avoid surgery-related damage. Metabolites responding to real-time pathophysiological changes may offer new insights into germinoma diagnosis. Using metabolomics and a multi-step machine learning approach, potential diagnostic metabolic biomarkers in the cerebrospinal fluid of 65 patients (26 germinoma, 31 non-iGCT, and 8 NGGCT) were identified. A subset of these samples (13 plus 8/9 random samples) was used as test data to evaluate classification models. Additionally, a comprehensive multi-omics strategy was applied to investigate germinoma pathogenesis using various tissue samples, including transcriptome, metabolome, and DNA methylation data, focusing on dysregulated metabolic pathways and immune cell infiltration. 51 differential metabolites were detected between germinoma and non-iGCT. A three-marker signature including N-acetylaspartylglutamate (NAAG), inosine, and β-HCG showed potential in distinguishing germinoma from non-iGCT across four different classification algorithms (the area under the receiver operating characteristic curve of SVM, random forest, naive Bayes, and K-nearest neighbors: 85.48%, 82.45%, 78.28%, 86.19%, respectively). Multi-omics analysis revealed activation of nicotinamide metabolism in germinoma, with fewer immunosuppressive cells and more anti-tumor immune cells compared with NGGCT. A combination of decreased NAAG, elevated inosine, and β-HCG has been suggested as a potential diagnostic signature of germinoma. Our findings imply distinct tumor pathogenesis between germinoma and NGGCT, providing new insights into therapeutic strategies for iGCT.
Yin et al. (Fri,) studied this question.