MINING THE CELLMINER DATABASE TO IDENTIFY SHARED BIOMARKERS OF 5-FU AND OXALIPLATIN RESPONSE.

Abstract

Background: 5-Fluorouracil and Oxaliplatin form backbone of colorectal cancer, yet resistance limits their efficacy. Understanding the molecular determinants of sensitivity and resistance may guide potential biomarker discovery and inform drug repurposing strategies.

Methods: We performed an integrative pharmacogenomic analysis of the NCI-60 cancer cell line panel using CellMiner Database. Pathway enrichment was performed using PANTHER. Individual drug and molecular biomarker correlations were explored to identify potential therapeutic vulnerabilities and repurposing opportunities.

Results: Genetic variants in ALDH9A1 were negatively associated with both 5-FU and Oxaliplatin. Protein function-affecting variants in CAMSAP3, LUM, and LRIG2 correlated negatively. DNA methylation of FERMT3 was negatively correlated with drug response, suggesting epigenetic silencing as a resistance mechanism. Copy number variation in COL1A1 also predicted resistance but correlated positively with statin sensitivity, highlighting repurposing potential. Transcriptomic signatures revealed cytoskeletal/adhesion genes (CNN3, ACTN1, DUSP10) as resistance markers, with pathway enrichment pointing to folate metabolism, MAPK signaling, and cytoskeletal remodeling. RNA-seq confirmed NT5E and HIF1A as resistance drivers. Several microRNAs including let-7e, miR-30a, and miR-22, were negatively correlated with drug activity, positioning them as potential biomarkers. Drug-drug correlation showed several cytotoxics positively associated with 5-FU/Oxaliplatin.

Conclusion: This integrative analysis identify potential biomarkers associated with 5-FU and Oxaliplatin response, nominating ALDH9A1, FERMT3, NT5E, HIF1A, and specific microRNAs as resistance biomarkers, while GRIN1, MTHFD2, and miR-7 emerge as sensitizers. Importantly, repurposing opportunities were identified, with statins and kinase inhibitors showing context-dependent associations that may help overcome resistance. These findings may provide a framework for potential biomarkers guided therapy optimization and may inform rational combination strategies in colorectal cancer.