RSNA 2014 

Abstract Archives of the RSNA, 2014


SSE02-02

Tomosynthesis 3D Mammography Compared with Contrast Enhanced Breast MRI in Pre-operative Evaluation of Patients Diagnosed with Breast Cancer

Scientific Papers

Presented on December 1, 2014
Presented as part of SSE02: Breast Imaging (Tomosynthesis Diagnostics)

Participants

Mary Woo Yamashita MD, Presenter: Nothing to Disclose
Sandy Chia-En Lee MD, Abstract Co-Author: Nothing to Disclose
Qinghua Min, Abstract Co-Author: Nothing to Disclose
Lingyun Ji MS, Abstract Co-Author: Nothing to Disclose
Pulin Arun Sheth MD, Abstract Co-Author: Nothing to Disclose
Susan Groshen PhD, Abstract Co-Author: Nothing to Disclose
Ingrid He, Abstract Co-Author: Nothing to Disclose
Akshara Singareeka Raghavendra MBBS, MS, Abstract Co-Author: Nothing to Disclose
Linda Hovanessian-Larsen MD, Abstract Co-Author: Nothing to Disclose

PURPOSE

To evaluate the performance of 3D Tomosynthesis (3D Tomo) versus Contrast Enhanced MRI (CE-MRI) as an adjunct to the standard 2D mammography (2D) in detection of additional cancers in women diagnosed with breast cancer (BC).

METHOD AND MATERIALS

We retrospectively reviewed available imaging, surgical, and medical records of 29 women diagnosed with BC and underwent 3D Tomo as part of their initial screening study or diagnostic imaging workup between September 2012 and January 2014. All these women had a CE-MRI prior to any surgical, medical, or radiation therapy. 3 dedicated breast imagers independently interpreted each study acquired by: 1) 2D, 2) 2D + 3D Tomo, and 3) 2D + 3D Tomo with addition of clinical history. 10 negative 2D + 3D Tomo cases were added as controls; the order of studies was randomized for each radiologist. The presence of a suspicious lesion (agreement by 2/3 radiologists) was recorded. A retrospective analysis of the CE-MRI results was performed. The gold standard was histopathology obtained by needle core biopsy and/or surgery.

RESULTS

Among 29 women diagnosed with BC, 48 lesions had histopathology results: 36 malignant and 12 benign. CE-MRI identified all 36 cancers but falsely identified 8 out of 12 benign lesions as suspicious. 2D identified 23 of 36 cancers (64% sensitivity) and excluded all 12 benign lesions (100% specificity). 2D + 3D Tomo identified 27 of 36 cancers (75% sensitivity) and excluded all 12 benign lesions (100% specificity). 2D + 3D Tomo with clinical information identified 32 of 36 cancers (89% sensitivity) and excluded 10 of 12 benign lesions (83% specificity).  

CONCLUSION

CE-MRI is highly sensitive in detecting invasive BC; however, its low specificity leads to unnecessary biopsies. In our pilot series, 3D Tomo mammography was found to be a valuable imaging modality for identifying additional cancers in newly diagnosed breast cancer patients with 89% sensitivity and 83% specificity.

CLINICAL RELEVANCE/APPLICATION

Further investigation with a larger cohort may prove that 3D Tomo can be an alternative method of evaluating additional cancers in newly diagnosed BC patients. 

Cite This Abstract

Yamashita, M, Lee, S, Min, Q, Ji, L, Sheth, P, Groshen, S, He, I, Raghavendra, A, Hovanessian-Larsen, L, Tomosynthesis 3D Mammography Compared with Contrast Enhanced Breast MRI in Pre-operative Evaluation of Patients Diagnosed with Breast Cancer.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14011723.html