Probing Folate-Responsive and Stage-Sensitive Metabolomics and Transcriptional Co-Expression Network Markers to Predict Prognosis of Non-Small Cell Lung Cancer Patients.

Tumour metabolomics and transcriptomics co-expression network as related to biological folate alteration and cancer malignancy remains unexplored in human non-small cell lung cancers (NSCLC). To probe the diagnostic biomarkers, tumour and pair lung tissue samples (n = 56) from 97 NSCLC patients were profiled for ultra-performance liquid chromatography tandem mass spectrometry (UPLC/MS/MS)-analysed metabolomics, targeted transcriptionomics, and clinical folate traits. Weighted Gene Co-expression Network Analysis (WGCNA) was performed. Tumour lactate was identified as the top VIP marker to predict advance NSCLC (AUC = 0.765, Sig = 0.017, CI 0.58-0.95). Low folate (LF)-tumours vs. adjacent lungs displayed higher glycolytic index of lactate and glutamine-associated amino acids in enriched biological pathways of amino sugar and glutathione metabolism specific to advance NSCLCs. WGCNA classified the green module for hub serine-navigated glutamine metabolites inversely associated with tumour and RBC folate, which module metabolites co-expressed with a predominant up-regulation of LF-responsive metabolic genes in glucose transport ( GLUT1 ), de no serine synthesis ( PHGDH , PSPH , and PSAT1 ), folate cycle ( SHMT1/2 and PCFR ), and down-regulation in glutaminolysis ( SLC1A5 , SLC7A5 , GLS , and GLUD1 ). The LF-responsive WGCNA markers predicted poor survival rates in lung cancer patients, which could aid in optimizing folate intervention for better prognosis of NSCLCs susceptible to folate malnutrition.