Genomic signatures for paclitaxel and gemcitabine resistance in breast cancer derived by machine learning

• Published in: Molecular oncology Vol. 10; no. 1; pp. 85 - 100
• Main Authors: Dorman, Stephanie N., Baranova, Katherina, Knoll, Joan H.M., Urquhart, Brad L., Mariani, Gabriella, Carcangiu, Maria Luisa, Rogan, Peter K.
• Format: Journal Article
• Language: English
• Published: United States Elsevier B.V 01.01.2016; John Wiley & Sons, Inc; John Wiley and Sons Inc
• Subjects: Anthracycline; Antineoplastic Agents - therapeutic use; Artificial intelligence; Breast cancer; Breast Neoplasms - drug therapy; Breast Neoplasms - genetics; Cancer therapies; Cell Line, Tumor; Chemoresistance; Chemotherapy; Copy number; Deoxycytidine - analogs & derivatives; Deoxycytidine - therapeutic use; Deoxyribonucleic acid; DNA; DNA methylation; Drug Resistance, Neoplasm - genetics; Drug sensitivity; Female; Fibroblast growth factor 2; Gemcitabine; Gene expression; Genomes; Genomic profiles; Humans; Learning algorithms; Machine Learning; Medical research; Metabolism; Metabolites; Minimal residual disease; Mutation; Paclitaxel; Paclitaxel - therapeutic use; Patients; Resistance; Support Vector Machine; Tumor cell lines; Tumors; Canada; Resistance; Breast cancer; Genomic profiles; Gemcitabine; Drug sensitivity; Paclitaxel
• ISSN: 1574-7891; 1878-0261
• DOI: 10.1016/j.molonc.2015.07.006

Increasingly, the effectiveness of adjuvant chemotherapy agents for breast cancer has been related to changes in the genomic profile of tumors. We investigated correspondence between growth inhibitory concentrations of paclitaxel and gemcitabine (GI50) and gene copy number, mutation, and expression first in breast cancer cell lines and then in patients. Genes encoding direct targets of these drugs, metabolizing enzymes, transporters, and those previously associated with chemoresistance to paclitaxel (n = 31 genes) or gemcitabine (n = 18) were analyzed. A multi-factorial, principal component analysis (MFA) indicated expression was the strongest indicator of sensitivity for paclitaxel, and copy number and expression were informative for gemcitabine. The factors were combined using support vector machines (SVM). Expression of 15 genes (ABCC10, BCL2, BCL2L1, BIRC5, BMF, FGF2, FN1, MAP4, MAPT, NFKB2, SLCO1B3, TLR6, TMEM243, TWIST1, and CSAG2) predicted cell line sensitivity to paclitaxel with 82% accuracy. Copy number profiles of 3 genes (ABCC10, NT5C, TYMS) together with expression of 7 genes (ABCB1, ABCC10, CMPK1, DCTD, NME1, RRM1, RRM2B), predicted gemcitabine response with 85% accuracy. Expression and copy number studies of two independent sets of patients with known responses were then analyzed with these models. These included tumor blocks from 21 patients that were treated with both paclitaxel and gemcitabine, and 319 patients on paclitaxel and anthracycline therapy. A new paclitaxel SVM was derived from an 11-gene subset since data for 4 of the original genes was unavailable. The accuracy of this SVM was similar in cell lines and tumor blocks (70–71%). The gemcitabine SVM exhibited 62% prediction accuracy for the tumor blocks due to the presence of samples with poor nucleic acid integrity. Nevertheless, the paclitaxel SVM predicted sensitivity in 84% of patients with no or minimal residual disease. [Display omitted] •Cell line genomic signatures of paclitaxel and gemcitabine resistance were derived.•Paclitaxel signature accurately predicts resistant cell lines with 82% accuracy.•Gemcitabine signature accurately predicts resistant cell lines with 84% accuracy.•Paclitaxel signature predicted sensitive patients with 84% accuracy.•Gemcitabine signature predicted response in FFPE tumor blocks with 62% accuracy.