Identification of the Treatment Survivability Gene Biomarkers of Breast Cancer Patients via a Tree-Based Approach

Studying breast cancer survivability among different patients who received various treatments may help to understand the relationship between the survivability and treatment therapy based on the gene expression. In this work, we built a classifier system that predicts whether a given breast cancer patient who underwent some form of treatment (which is either hormone therapy (H), radiotherapy (R), or surgery (S)) will survive beyond five years after the treatment therapy. Our classifier is a tree-based hierarchical approach which partitions breast cancer patients according to survivability classes; each node in the tree is associated to a treatment therapy and finds a predictive subset of genes that can best predict whether a given patient will survive after that particular treatment. We applied our tree-based method to a gene expression dataset consisting of 347 treated breast cancer patients and identified potential biomarker subsets with accuracies ranging from 80.9% to 100%. We have investigated the roles of many biomarkers through the literature.