Recurrent transcriptional signatures of acquired chemoresistance across cancer models

Chemoresistance is the innate or acquired ability of cancer cells to evade the cytotoxic effects of chemotherapeutic agents. This hypothesis-generating study presents a systematic comparative analysis of public transcriptomic datasets associated with acquired chemoresistance to major drugs including doxorubicin, cisplatin, and gemcitabine. The aim was to identify recurrent drug-specific gene expression patterns in drug-resistant cancer cells relative to their drug-sensitive parental cells. Publicly available microarray datasets of breast, ovarian, neuroblastoma, bladder, and gallbladder chemoresistant cell lines were retrieved from the Gene Expression Omnibus (GEO) database. Differential expression analysis using R/Bioconductor revealed distinct and overlapping signatures of chemoresistance across different cancer types. In doxorubicin-resistant breast cancer, recurrently dysregulated genes included IL-18, CD44, FN1, EGFR, and the histones H2BC10 and H2BC7. Neuroblastoma and ovarian cancer cell datasets shared Differentially Expressed Genes (DEGs) including ABCB1, RUNDC3B, CROT, and TP53TG1. TMEM47 was the only common DEG among all three (breast cancer, neuroblastoma, and ovarian cancer) doxorubicin-resistant datasets, although with tissue-dependent regulation. Gemcitabine resistance signatures in bladder and gallbladder cells include EPCAM, AGR2, NRP2, FGF1, CTH, TENM2, and the long non-coding RNAs APCDD1L-DT and LINC01234. Combined gemcitabine/5-FU resistance revealed common patterns of resistance in gallbladder and colon cancer cell lines. In cisplatin-resistant ovarian cancer cells, LAMA2 and SOX8 were differentially expressed with a consistent direction of change in all datasets in which they were measured. These recurrent associations do not establish causality; rather, they prioritize candidate genes for functional validation in future studies of chemoresistance.