Correlation between kinetic features on breast MRI computer-aided detection (CAD) and prognostic factors of breast cancer.

Abstract

#4015 The purpose of this study was to correlate the kinetic features of breast MR computer-aided detection (CAD) with the prognostic factors of breast cancer.
 A total of 65 patients were included in our study from December 2007 to March 2008. We reviewed kinetic features of MR CAD such as proportion of initial rapid rise enhancement, percentage of washout, and enhancement patterns. We also recorded prognostic factors such as tumor size, estrogen receptor, progesterone receptor, HER-2/neu status, lymphovascular invasion, and histologic grade on pathology. We correlated the kinetic features on MR CAD with prognostic factors. Pearson and Spearman correlation test were used to evaluate correlation between kinetic features and tumor size. Fisher exact test also used for the statistical analysis between kinetic features and other prognostic factors.
 In 65 cases, 59 cases showed rapid washout enhancement patterns, 5 cases showed rapid plateau pattern and only one case showed medium plateau. Tumor size showed correlation with the initial rapid rise enhancement (p=0.03) on Pearson correlation test and the delayed washout percentage (p=0.05) on Spearman correlation test.
 Positive progesterone receptor status was significantly associated with rapid early enhancement (p=0.023) and delayed washout (p=0.05), respectively. HER-2/neu status showed trend with the early enhancement (p=0.063). Other prognostic factors such as estrogen receptor, lymphovascular invasion and histologic grade showed no significant correlation with the kinetic features on MR CAD.
 Several kinetic features on MR CAD were correlated with the specific prognostic factors of breast cancer. Recognition of several kinetic features correlated with prognostic factors might help to predict of patients' prognosis and to decide the treatment plan. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 4015.