Abstract / Summary
Abstract Background In colorectal cancer (CRC), liver metastasis is the primary factor leading to death, yet effective biomarkers for liver metastatic colorectal cancer (LMCRC) remain scarce. Ferroptosis, a regulated type of cell death spurred by iron, has been demonstrated to have a significant role in tumor progression, metabolic adjustment, and immune regulation. This study centered on discovering ferroptosis-related molecular alterations in LMCRC and establishing a relevant diagnostic framework. Method Genes that exhibited differential expression between primary colorectal tumors and liver metastases were identified by utilizing the GSE6988 dataset and cross-checked with ferroptosis-related genes obtained from FerrDb V3 and a published ferroptosis gene set. Functional enrichment analysis was performed employing Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), Gene Set Enrichment Analysis (GSEA), and Gene Set Variation Analysis (GSVA) methodologies.Key biomarkers were discerned employing an integrative analytical framework that incorporated logistic regression, Support Vector Machine-Recursive Feature Elimination (SVM-RFE), and Least Absolute Shrinkage and Selection Operator (LASSO) regression. The diagnostic efficacy of the model was subsequently validated in the external GSE81558 dataset employing Receiver Operating Characteristic (ROC) curve and decision curve analysis. Single-sample Gene Set Enrichment Analysis (ssGSEA) was employed to quantify the levels of immune cell infiltration. Finally, computational strategies, including virtual screening and molecular docking, were applied to screen for potential compounds targeting the protein of interest. Results In the GSE6988 dataset, investigators identified 272 genes that had notable expression alterations. Among them, 173 genes had elevated expression, and 99 genes had reduced expression. By cross-checking with ferroptosis-related gene sets (FRGs), 28 ferroptosis-associated genes with changed expression (FRDEGs) were found. These genes were primarily associated with processes like oxidative stress reaction, iron/copper ion regulation, ferroptosis, mineral absorption, fatty acid routes, and bile acid metabolism.The amalgamation of logistic regression, SVM-RFE, and LASSO regression identified four crucial genes: WNT5A, LCN2, TRPA1, and AKR1C2. A risk-scoring model developed from these genes displayed outstanding accuracy in the initial cohort (AUC > 0.90) and sustained strong performance in external validation sets (AUC falling within the range of 0.70 to 0.90). The ssGSEA analysis revealed statistically significant disparities in 14 immune cell populations between the LMCRC and CRC cohorts.. WNT5A demonstrated the strongest positive correlation with mast cells (r = 0.60, p < 0.05), while AKR1C2 exhibited the most pronounced negative correlation with gamma delta T cells (r =−0.433, p < 0.05).Molecular docking investigations proposed promising predicted binding capabilities of Z9405216322 and D481−2213 for AKR1C2. Conclusion This study examined the molecular characteristics related to ferroptosis in LMCRC and established a diagnostic model built on four crucial genes that hold promising clinical implications. The findings suggest potential biomarkers and treatment approaches for further experimental and clinical validation.