Article Text
Abstract
Introduction/Background Platinum resistance development in epithelial ovarian cancer (EOC) is still one the major obstacle in successful treatment, due to the fact that ~ 80% of patients develop drug resistance within 5 years after adjuvant chemotherapy. This study aimed to provide insight to co-expression networks in dependency to platinum resistance status, which could help with better understanding of molecular background of resistance development.
Methodology Estimation of expression profile was done by RNA sequencing and co-expression analysis was performed in two independent datasets: i) training set of 60 EOC patients with defined platinum resistant status and ii) validation TCGA-OV set (N=426). Co-expression networks were analyzed by the WGCNA method by Langfelder and Horvath and Cytoscape platform was used for network visualization.
Results Co-expression network analysis in the training set showed ~ 30 gene modules which significantly associated with patients overall survival and platinum resistant status. Module with the strongest correlation with platinum sensitive status were characterized by high expression of immune system related genes (especially T-lymphocyte activation). Module strongly correlating with OS was characterized by adhesion and migratory genes. Gene of interest, NOTCH3, was one the most correlating genes with platinum resistant status of EOC patients. Dominant modules were further observed and validated in the TCGA-OV validation set, especially immune system a migration pathways genes enriched expression modules.
Conclusion Analysis of co-expression networks in regard to platinum resistant status showed ~ 30 gene modules with significant association. Major modules were enriched by immune system related genes and by genes involved in cell migration and adhesion pathways.
Supported by the Czech Health Research Council grant no. NU20–09-00174 and Cooperatio program, research area Maternal and Childhood Care, by Charles University.
Disclosures This study was supported by running projects of the Czech Health Research Council grant no. NU20–09-00174 and Charles University, Cooperatio Program, research area Maternal and Childhood Care.