Sung Eun Song MD, Presenter: Nothing to Disclose

Bo Kyoung Seo MD, PhD, Abstract Co-Author: Nothing to Disclose

Kyu Ran Cho MD, Abstract Co-Author: Nothing to Disclose

Ok Hee Woo MD, Abstract Co-Author: Nothing to Disclose

Ki Yeol Lee MD, PhD, Abstract Co-Author: Nothing to Disclose

To compare the utility of multimodality breast imaging for assessment of tumor extent in breast cancer and to find out the additional role of a computer-aided detection (CAD) system for MR imaging.

A total of 51 patients who underwent breast cancer surgery and preoperative multimodality breast imaging from August 2008 to February 2011 were included. All these patients performed mammography, US, breast MR imaging, and low-dose CT mammography. Two breast radiologists who were blinded to the pathologic results measured the maximum tumor extent with these modalities. The tumor sizes on MR imaging were measured with the CAD system and without the CAD system. The measured sizes on breast imaging were correlated with pathologic results. To find out the additional role of the CAD system, we compared the accuracy of measured tumor size in terms of lesion type (mass or non-mass like enhancement), parenchymal enhancement pattern (homogeneous, heterogeneous, or nodular), and parenahymal composition (fatty or dense) on MR imaging. Spearman Rank Correlation and Kruskal-Wallis tests were used for statistical analysis.  

In tumor extent, all breast imaging modalities yielded statistically significant correlation with the pathologic results, the coefficients were 0.77 for MR imaging without the CAD system, 0.72 for MR imaging with the CAD system, 0.67 for CT mammography, 0.62 for US and 0.37 for mammography (p < 0.05). Breast MR imaging without the CAD system showed the best correlation with the actual pathological size. On breast MR imaging, accurate measurement of tumor size was influenced by lesion type and masses had much better correlation than non-mass like enhancements (p = 0.006). On the other hand, parenchymal enhancement pattern and composition on MR imaging did not affect measuring tumor extent (p > 0.05).  

Mammography, US, MR imaging, and CT mammography are satisfactory imaging modalities to assess tumor extent in breast cancer and MR imaging is the best modality. Additionally, parenchymal enhancement pattern and breast composition on MR imaging do not affect accurate measurement. However, the CAD system for MR imaging may not have an additional role to assess tumor extent in breast cancer.

Breast MR imaging is the best modality to assess tumor extent without interference of breast composition and enhancement pattern.

Song, S, Seo, B, Cho, K, Woo, O, Lee, K, The Utility of Multimodality Breast Imaging for Assessment of Tumor Extent in Breast Cancer: Mammography, US, MR Imaging with a CAD System, and CT Mammography  .  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11010141.html