Comparison of the Accuracy of Mammography, Ultrasonography and MRI for Evaluating Residual Tumor after Neoadjuvant Chemotherapy for Breast Cancer

ABSTRACT

PURPOSE: The purpose of our study was to determine the relative accuracy of mammography, ultrasonography and MRI for evaluating residual tumor after neoadjuvant chemotherapy for breast cancer, as compared with the pathological results. MATERIALS AND METHODS: From December 2004 through August 2005, 13 patients who had mammography, ultrasonography and MRI performed for evaluating tumor response were enrolled in our study from a total of 47 patients who received neoadjuvant chemotherapy for breast cancer. The therapy response was defined by ultrasonography that was able to compare the images taken before and after therapy, and each imaging was retrospectively analyzed by two board-certified radiologists who specialized in breast imaging. The presence or absence of residual tumor was investigated and the tumor measurement according to the imaging was divided into underestimating, being equal to or overestimating the size of the residual tumor, compared with that of the pathological results. The relative accuracy of these modalities was then assessed. RESULTS: Eight of 13 patients showed a partial response and 5 patients showed stable lesion. Agreement rates about the residual tumor, as measured by mammography, ultrasonography and MRI and then compared with the pathological results, were 39%, 54% and 77%, respectively. Of the three methods, MRI agreed with the pathological results significantly more often, but it may overestimate (8%) or underestimate (15%) (p = 0.102). When there was disagreement with the pathological results, mammography exhibited a tendency to underestimate (38%) and ultrasonography exhibited a tendency to overestimate (31%). CONCLUSION: MRI had a higher relative accuracy than did mammography and ultrasonography for evaluating the residual tumor in patients receiving neoadjuvant chemotherapy for breast cancer. However, MRI may overestimate (8%) or underestimate (15%) the residual tumor.