Direct MS enabled discovery of lipid signatures with diagnostic implications in colorectal cancer.

Lipid metabolites are promising biomarkers for detection of colorectal cancer (CRC). However, current methods often involve complex sample preparation and long analysis times, which can affect the stability and integrity of clinical samples, especially tissues. This creates a need for faster, more accurate lipid profiling approaches. Comprehensive analysis of lipid changes and related metabolic pathways is key to understanding CRC development and improving diagnostic accuracy. Herein, we employed internal extractive electrospray ionization mass spectrometry (iEESI-MS) to analyze lipid profiles from CRC and healthy tissue groups combined with multivariate statistical analyses. Our analysis identified 40 significantly altered lipid species spanning nine major classes, comprising glycerophospholipids (phosphatidylcholines (PC), phosphatidylethanolamines (PE), phosphatidylglycerols (PG), phosphatidic acids (PA), and phosphatidylserines (PS), sphingomyelins (SM) and ceramides (Cer), and triacylglycerols (TG) and diacylglycerols (DG)). The lipid signature PC (36:4) demonstrated efficacy in discriminating CRC from normal tissues with highest area under the curve (AUC) values of 0.95. The optimal model selected included 10 lipid metabolites with high AUC value of 0.985, a sensitivity of 0.967, and a specificity of 0.95, suggesting that the lipidomic model has potential clinical applications. Pathway enrichment analysis further revealed statistically significant perturbations (p < 0.01) in glycerophospholipid metabolism pathways and glycerolipid metabolism pathways, suggesting their potential involvement in CRC pathogenesis. The iEESI-MS approach demonstrated the capability to rapidly detect CRC-specific lipid biomarkers and perturbed metabolic pathways, offering a promising tissue diagnostic tool that bridges lipid metabolism research with clinical applications in precision medicine for CRC detection.