Usefulness of digital breast tomosynthesis (DBT) in evaluation of pathological response after neoadjuvant chemotherapy (NAC) for breast cancer.

Abstract

119 Background: Digital breast tomosynthesis (DBT) has been developed recently and it provides an advantage in detection of breast masses compared to 2D mammography (MMG) since it allows to separate the tissue layers and to noticeably reduce occlusions caused by overlapping anatomical structures. Neoadjuvant chemotherapy (NAC) has been used increasingly for the treatment of breast cancer. The imaging methods that have been used until now to assess tumor response to NAC have serious limitations. Attaining pathological complete response (pCR) is the most important predictor of long-term survival in patients receiving NAC. We assessed the radiological findings and capability of DBT to predict response to NAC in comparison with 2D MMG. Methods: 33 woman (mean age, 47.3 y.o.) having 36 lesions were recruited. Image utilizing adjunction of DBT to 2D MMG were taken for diagnosis from January, 2010 to March, 2015. Pathological subtypes were IDC (n = 34), ILC (n = 1), and IMPC (n = 1). Pathological responses of the lesions to NAC were Grade1 (n = 16), Grade2 (n = 11), and Grade3 (n = 9). The clinical image data sets were acquired with a DBT system (MAMMOMAT Inspiration, Siemens, Germany).For each patient, CC and MLO views were taken for diagnosis. With one-view 2D MMG and DBT, those radiation dose, utilizing the ACR phantom 156, were 1.20mGy and 1.80mGy. Results: The clinical response to NAC was classified into the following categories: 1) Fairly Response (FR): reduction in size of the tumor by more than 70% including Complete Response (CR), 2) Partial Response (PR): reduction in size of the tumor by 30% or more but less than 70%, 3) Stable disease (SD): reduction in size of the tumor by less than 30%. DBT findings of pathological Grade2-3 cases were all suggested FR (28/28, 100.0%) Grade3 cases were all suggested CR (9/9,100.0%). Regarding the Grade1b cases, all lesions were suggested PR (8/8, 100.0%). Regarding the Grade1a cases, 3 out of 8 (36.5%) lesions were suggested CR, and 2 out of 8 (25%) lesions suggested PR, and 3 out of 8 lesions suggested SD. Conclusions: DBT will contribute to improved diagnosis accuracy with regard to NAC response to breast cancer in comparison with 2D MMG.