Search

Radiogenomics investigates the relationship between imaging phenotypes and genetic expression. Breast cancer is a heterogeneous disease that manifests complex genetic changes and various prognosis and treatment response. We investigate the value of machine learning approaches to radiogenomics using low-dose perfusion computed tomography (CT) to predict prognostic biomarkers and molecular subtypes of invasive breast cancer. This prospective study enrolled a total of 723 cases involving 241 patients with invasive breast cancer. The 18 CT parameters of cancers were analyzed using 5 machine learning models to predict lymph node status, tumor grade, tumor size, hormone receptors, HER2, Ki67, and the molecular subtypes. The random forest model was the best model in terms of accuracy and the area under the receiver-operating characteristic curve (AUC). On average, the random forest model had 13% higher accuracy and 0.17 higher AUC than the logistic regression. The most important CT parameters in the random forest model for prediction were peak enhancement intensity (Hounsfield units), time to peak (seconds), blood volume permeability (mL/100 g), and perfusion of tumor (mL/min per 100 mL). Machine learning approaches to radiogenomics using low-dose perfusion breast CT is a useful noninvasive tool for predicting prognostic biomarkers and molecular subtypes of invasive breast cancer. [ABSTRACT FROM AUTHOR]

Background: Insulin receptor substrate 1 (IRS-1) has been known to be an associated factor with breast cancer progression. However, there has been little study with respect to the relationship between the expression of IRS-1 and breast cancer prognosis in clinical practice. In this study, we evaluated the impact of the estrogen receptor (ER) and IRS-1 on the recurrence and survival of breast cancer patients.Methods: We analyzed the pathologic finding of 376 tissue samples from breast cancer patients who received proper treatment between January 1990 and December 2006 using the tissue microarray. We measured the expression of ER and IRS-1 by immunohistochemistry staining and analyzed the difference of recurrence and survival rate in each subgroup of ER and IRS-1.Results: Our results show that there is a significant difference of disease-free survival (DFS) according to ER and IRS-1 subgroups with both univariate and multivariate analyses. Specifically, ER-positive and IRS-1-positive breast cancer samples showed improved DFS compared to ER-positive and IRS-1-negative breast cancer (adjusted hazard ratio: 2.17; 95% confidence interval: 1.15-4.09; P = 0.01). There was a difference of overall survival according to ER and IRS-1 subgroups by univariate analysis (P = 0.01), but not by multivariate analysis (P = 0.36).Conclusion: ER and IRS-1 subgroups appear to be critical factors for the prediction of breast cancer recurrence. In particular, we suggest that the patients who have ER-positive and IRS-1-negative breast cancer undergo more aggressive treatment because they have poorer prognoses. [ABSTRACT FROM AUTHOR]