Although clinical and radiological findings of patients with GM may mimic those of breast carcinoma, our study showed that women of childbearing age, especially among Hispanic ethnicity with a recent history of pregnancy or high prolactin level and newly tender mass-like lesion, in addition to new focal asymmetry on mammogram and heterogeneous hypoechoic irregular-shaped mass on ultrasound exam, should raise concern for GM. Non-invasive approach and clinical follow-up were the preferred treatment method.

Radiation therapy is one of the cornerstones for the treatment of thoracic malignancies. Although advances in radiation therapy technology have improved the delivery of radiation considerably, adverse effects are still common. Postirradiation changes affect the organ or tissue treated and the neighboring structures. Advances in external-beam radiation delivery techniques and how these techniques affect the expected thoracic radiation-induced changes are described. In addition, how to distinguish these expected changes from complications such as infection and radiation-induced malignancy, and identify treatment failure, that is, local tumor recurrence, is reviewed. ©

Abstract. Breast cancer management could be improved by developing real-time imaging tools to assess tissue architecture without extensive processing. We sought to determine whether confocal fluorescence microscopy (CFM) provides sufficient information to identify neoplasia in breast tissue. Breast tissue specimens were imaged following proflavine application. Regions of interest (ROIs) were selected in histologic slides and in the corresponding region on confocal images, and then divided into sets for training and validation. Readers reviewed images in the training set and evaluated images in the validation set for the presence of neoplasia. Accuracy was assessed using histologic diagnosis as the gold standard. Seventy tissue specimens from 31 patients were imaged; 235 ROIs were identified and diagnosed as neoplastic or non-neoplastic. A training set was assembled using 23 matched ROIs; 49 matched ROIs were assembled into a validation set. Neoplasia was identified in histologic images: 93% sensitivity, 97% specificity [area under the curve (AUC ¼ 0.987)] and in confocal images: 93% sensitivity 93% specificity (AUC ¼ 0.957). CFM produced images of architectural features in breast tissue comparable with conventional histology, while requiring little processing. Potential applications include assessment of excised tissue margins and evaluation of tissue adequacy for bio-banking and genomic studies. © The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI.

Tissue sampling is a problematic issue for inflammatory breast carcinoma, and immediate evaluation following core needle biopsy is needed to evaluate specimen adequacy. We sought to determine if confocal fluorescence microscopy provides sufficient resolution to evaluate specimen adequacy by comparing invasive tumor cellularity estimated from standard histologic images to invasive tumor cellularity estimated from confocal images of breast core needle biopsy specimens. Grayscale confocal fluorescence images of breast core needle biopsy specimens were acquired following proflavine application. A breast-dedicated pathologist evaluated invasive tumor cellularity in histologic images with hematoxylin and eosin staining and in grayscale and false-colored confocal images of cores. Agreement between cellularity estimates was quantified using a kappa coefficient. 23 cores from 23 patients with suspected inflammatory breast carcinoma were imaged. Confocal images were acquired in an average of less than 2 min per core. Invasive tumor cellularity estimated from histologic and grayscale confocal images showed moderate agreement by kappa coefficient: κ = 0.48 ± 0.09 (p < 0.001). Grayscale confocal images require less than 2 min for acquisition and allow for evaluation of invasive tumor cellularity in breast core needle biopsy specimens with moderate agreement to histologic images. We show that confocal fluorescence microscopy can be performed immediately following specimen acquisition and could indicate the need for additional biopsies at the initial visit.

Added SE does not improve the diagnostic ability of conventional ultrasound when evaluating abnormal axillary lymph nodes.

Purpose To determine the relationship of microcalcification morphology and distribution with clinical, histopathologic, biologic features and local recurrence (LR) in patients with pure ductal carcinoma in situ (DCIS) of the breast. Methods All patients with pure DCIS who underwent preoperative mammography at our institution from 1996 through 2009 were identified. Mammographic findings were classified according to the ACR BI-RADS lexicon. Associations between mammographic findings and clinical, histopathologic, biologic characteristics and LR were analyzed. Statistical inference used multiple logistic regression and Cox proportional hazards regression adjusted for age and confounding due to bias from nonrandomized selection of radiation therapy. Results We identified 1657 patients with microcalcifications visualized on mammography. The mean age at diagnosis was 55 years (SD, 11). The mean follow-up was 7 years (range, 1–16). Ipsilateral LR was 4% in segmentectomy (987) and 1.5% in mastectomy (670) patients. Increased LR risk was seen in patients with dense breast tissue (p<0.05) and larger DCIS size (p<0.01). Radiation therapy was associated with a 2.8-fold decrease in the LR risk. Fine linear (branching) microcalcifications were associated with 5.2-fold increase in LR. Extremely dense breast tissue was associated with positive/close margins (p=0.04) and multicentricity (p<0.01). Younger women were more likely to have extremely dense breast tissue (p<0.0001), multicentric disease (p<0.0004), and undergo mastectomy (p<0.0001). Conclusions Dense breast tissue, large DCIS size, and fine linear (branching) microcalcifications were associated with increased LR, yet overall LR rates remained low. Extremely dense breast tissue was a risk factor for multicentricity and positive margins in DCIS.

Background The clinicopathologic, mammographic, and sonographic findings in patients with pure ductal carcinoma in situ (DCIS) were assessed by estrogen receptor (ER) expression. Methods After institutional review board approval, patients with pure DCIS evaluated from January 1996 to July 2009 with known ER status and available imaging were identified. Images were reviewed per the ACR BI-RADS® lexicon (4th edition). Clinical, pathologic, and imaging characteristics were analyzed by ER status using t-test, chi-square test, and Fisher’s exact test. Results Of 1219 patients with pure DCIS and known ER status identified, 1187 with complete data were included. Mammography was performed in all 1187 patients and sonography in 519 (44%). There were 972 (82%) patients with ER-positive and 215 (18%) with ER-negative disease. ER-negative DCIS was more likely to be high grade (93% vs 44%, p<0.0001), associated with comedonecrosis (64% vs 29%, p<0.0001), and multifocal (23% vs 15%, p=0.009). On sonography, ER-negative DCIS was more likely to be visible (61% vs 46%, p=0.004), larger (mean size, 2.3 vs 1.6 cm, p=0.006), and show posterior shadowing (53% vs 28%, p=0.006). Mastectomy was more frequently performed for ER-negative DCIS (47% vs 37%, p=0.008). Palpable DCIS was visible on sonography in 55% of cases and mammography in 81%. Compared with ER-positive palpable DCIS, ER-negative palpable DCIS was larger and more likely to be visible on sonography. Compared with ER-positive noncalcified DCIS, ER-negative noncalcified DCIS was less likely to be visible on mammography. Conclusion ER-positive and ER-negative pure DCIS have different clinicopathologic and imaging characteristics. ER-negative DCIS is associated with worse prognostic factors then ER-positive DCIS. On sonography, ER-negative DCIS is more frequently visible than ER-positive DCIS, tends to be larger, and more frequently demonstrates posterior shadowing.

Of the 691 pure DCIS lesions, 362 (52%) were visible on sonography and presented most commonly as a mass. Lesion visibility of DCIS on sonography was not related to nuclear grade or the presence of comedonecrosis.