The role of specimen radiography in breast-conserving therapy of ductal carcinoma in situ

• Published in: Breast (Edinburgh) Vol. 26; pp. 73 - 79
• Main Authors: Lange, M, Reimer, T, Hartmann, S, Glass, Ä, Stachs, A
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier Ltd 01.04.2016
• Subjects: Breast - diagnostic imaging; Breast - pathology; Breast - surgery; Breast cancer; Breast Neoplasms - diagnostic imaging; Breast Neoplasms - pathology; Breast Neoplasms - surgery; Breast-conserving therapy; Carcinoma, Intraductal, Noninfiltrating - diagnostic imaging; Carcinoma, Intraductal, Noninfiltrating - pathology; Carcinoma, Intraductal, Noninfiltrating - surgery; Ductal carcinoma in situ; Female; Free surgical margin; Hematology, Oncology and Palliative Medicine; Humans; Intraoperative Care - methods; Logistic Models; Mammography - methods; Margins of Excision; Mastectomy, Segmental - methods; Middle Aged; Neoplasm, Residual - diagnostic imaging; Retrospective Studies; Risk Factors; ROC Curve; Sensitivity and Specificity; Specimen radiography; Ductal carcinoma in situ; Breast cancer; Breast-conserving therapy; Free surgical margin; Specimen radiography
• ISSN: 0960-9776; 1532-3080
• DOI: 10.1016/j.breast.2015.12.014

Abstract Background To assess the role of intraoperative specimen radiography (SR) and to define risk factors for positive margins in breast-conserving therapy (BCT) of ductal carcinoma in situ (DCIS). Methods In a retrospective study in calcification-associated DCIS treated with BCT between January 2009 and December 2011, digital mammographs and SR were reviewed and radiological margin width was determined. Clinical, radiological, and histological data were correlated with surgical histological data, and a histologically free margin of at least 2 mm was taken as evidence of successful BCT. Results 47/91 patients (51.6%) fulfilling the inclusion criteria had histologically involved surgical margins. Univariate analyses revealed DCIS size, mammographic extension of calcification, presence of comedo necrosis, negative progesterone receptor status, and a small radiological margin on SR to be risk factors for unsuccessful BCT. Receiver Operating Characteristic (ROC) analysis showed a radiological margin width of 4 mm to be optimal, with a sensitivity of 72.3% and specificity of 52.3%. The likelihood of surgical free margins was increased 2.9-fold with a radiological margin width ≥4 mm. On multivariate logistic regression analysis, only histological DCIS size >20 mm clearly emerged as an independent predictive factor for surgically involved margins (p < 0.001), while an SR margin <4 mm trended toward significance (p = 0.066). Conclusions SR is a reliable method for predicting free surgical margins in non-invasive breast cancer where a minimum radiological free margin of 4 Fmm is achieved. However, histological DCIS size remains the most important factor determining successful BCT.