Abstract Archives of the RSNA, 2013
SSA01-04
Sonographic Findings in 691 Pure Ductal Carcinoma in Situ Lesions with Histopathologic and Biologic Correlation
Scientific Formal (Paper) Presentations
Presented on December 1, 2013
Presented as part of SSA01: Breast Imaging (Diagnostic Ultrasound)
Marion Elizabeth Scoggins MD, Presenter: Nothing to Disclose
Gaiane M. Rauch MD, PhD, Abstract Co-Author: Nothing to Disclose
Patricia Sue Fox MS, Abstract Co-Author: Nothing to Disclose
Ana Paula Benveniste MD, Abstract Co-Author: Nothing to Disclose
Henry M. Kuerer MD, Abstract Co-Author: Nothing to Disclose
Wei Tse Yang MD, Abstract Co-Author: Nothing to Disclose
Young Mi Park MD, PhD, Abstract Co-Author: Nothing to Disclose
Sara Lari, Abstract Co-Author: Nothing to Disclose
Savitri Krishnamurthy MD, Abstract Co-Author: Nothing to Disclose
Sonographic (US) findings in 691 pure ductal carcinoma in situ (DCIS) lesions were retrospectively analyzed by estrogen receptor (ER) status, nuclear grade, and comedonecrosis to evaluate the prognostic value of US as an adjunct to mammography (M).
An institutional review board approved retrospective single institution database search performed for patients with pure DCIS evaluated from January 1996 to July 2009 who underwent pre-operative M and whole-breast US. Images were reviewed per ACR BI-RADS lexicon. Pathologic features recorded were ER status, nuclear grade, and comedonecrosis. ER+ was defined as nuclear staining in at least 1% of cells. Statistical comparisons were made using t-test, Chi-square, Fisher’s exact test, Kruskal-Wallis or Wilcoxon rank-sum test.
There were 1911 pure DCIS patients identified; those with incomplete data (n=5), lacking pre-operative US (n=1214) or M (n=1) were excluded leaving 691 patients for analysis. Of 691 lesions, 304 (44%) were visible on M and US, 315 (46%) visible on M only, 58 (8%) visible on US only, and 14 (2%) visible on neither M nor US. There were 425 (62%) ER+, 104 (15%) ER-, and 162 (23%) lesions with unknown ER. Comedonecrosis was present in 296 (43%) lesions, absent in 395 (57%). There were 334 (48%) non-high grade (nuclear grade I/II) lesions, 353 (51%) high-grade (III), and 4 (1%) of unknown grade. ER- lesions were more frequently visible on US than ER+ lesions (62% vs. 48%, p<0.05). An irregular non-circumscribed hypoechoic mass with no posterior features or enhancement was the most common US finding, regardless of histopathologic features. A shadowing US mass was more frequently high grade or ER- (p<0.05). A nonmass US lesion or calcifications were more frequently detected in high grade or comedo DCIS (p<0.0001). A round or oval US mass was more likely non-high grade (p<0.05). Average US lesion size was larger in ER-, high grade, and necrotic lesions (p<0.05).
ER- DCIS is more likely visible on US than ER+ DCIS. A shadowing US mass is more frequently high grade or ER-. While a mass is the most common US finding of DCIS regardless of histopathologic features, nonmass lesions are more likely to be associated with high-grade tumors and comedonecrosis.
A shadowing mass on US should raise suspicion for ER- DCIS which provides imaging-based prognostic and biologic information during cancer diagnosis and work-up.
Scoggins, M,
Rauch, G,
Fox, P,
Benveniste, A,
Kuerer, H,
Yang, W,
Park, Y,
Lari, S,
Krishnamurthy, S,
Sonographic Findings in 691 Pure Ductal Carcinoma in Situ Lesions with Histopathologic and Biologic Correlation. Radiological Society of North America 2013 Scientific Assembly and Annual Meeting, December 1 - December 6, 2013 ,Chicago IL.
http://archive.rsna.org/2013/13010844.html