Kelley KA, Crawford JD, Thomas K, Gardiner SK, Johnson NG. A Comparison of Breast-Specific Gamma Imaging of Invasive Lobular Carcinomas and Ductal Carcinomas. JAMA Surg. 2015;150(8):816-818. doi:10.1001/jamasurg.2015.0965
Copyright 2015 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
The practice of preoperatively evaluating the breast tissue of patients with newly diagnosed breast cancer has evolved, with increased use of advanced imaging. The evaluation of breast tissue using functional breast-specific gamma imaging (BSGI) has proven to be a valuable method of imaging as an adjunct to mammography and ultrasonography for early detection of breast cancer.1 The use of BSGI specifically in imaging invasive lobular carcinomas (ILCs), which are known to be difficult to detect, has not been well characterized.2,3 We reviewed our experience with BSGI as an adjunct imaging method used specifically for detecting ILCs.
This is a retrospective review of a prospectively maintained imaging registry evaluating all patients who underwent BSGI during the period from 2006 to 2012 at our institution. It was approved by the institutional review board of the Breast Health Center at the Legacy Cancer Institute in Portland, Oregon. All data were deidentified. Any patient with a new, biopsy-proven diagnosis of breast cancer who underwent BSGI for a preoperative evaluation was included. Women underwent diagnostic mammography and a breast biopsy confirming the diagnosis of cancer. Lesions newly identified by use of BSGI were further assessed using ultrasonography and/or magnetic resonance imaging (MRI). Patients with invasive ductal carcinoma (IDC) and ductal carcinoma in situ were included in the IDC group, and this group was compared with the patients with ILC. Invasive mammary carcinomas and other lesions, which accounted for 7.6% of the carcinomas in our population, were not included in the analysis. Sensitivity and specificity were calculated and compared between the ILC and IDC groups using the χ2 test of proportions, with P < .05 considered statistically significant.
A total of 699 patients with breast cancer underwent BSGI during the study period. The baseline patient and tumor characteristics are represented in Table 1. We identified 44 patients with ILC lesions, 4 of whom had bilateral breast disease. The ILC group tended to be older and have larger tumors and more nodal involvement. The known lesion was identified using BSGI in 40 of the 44 patients (90.9%) in the ILC group and in 485 of the 602 patients (80.6%) in the IDC group. In cases in which the known ILC was not visualized, the mean size of these lesions was 1.2 cm (range, 0.7-1.7 cm). New ipsilateral or contralateral cancers were identified using BSGI in 7 of the 44 patients (15.9%) in the ILC group and in 34 of the 602 patients (5.6%) in the IDC group.
The sensitivity and specificity of BSGI for all cancers was 85.4% and 81.3%, respectively (Table 2). The sensitivity of BSGI for ILCs was 88.9%, and the sensitivity of BSGI for IDCs was 85.1% (P = .64; 95% CI, 0.038-0.078). The specificity of BSGI for ILCs was 78.7%, and the specificity of BSGI for IDCs was 81.8% (P = .74; 95% CI, −0.101 to 0.163).
To our knowledge, this appears to be the largest series of functional imaging of ILCs reported to date. Breast-specific gamma imaging has a slightly higher sensitivity for the detection of ILCs than for the detection of IDCs (88.9% vs 85.1%). The specificities of BSGI for ILCs and IDCs are comparable. Smaller series have reported a BSGI sensitivity for ILCs near 100%.2,4 In addition, with the use of BSGI, a greater proportion of additional ILC lesions (15.9%) were detected compared with IDC lesions (5.6%). This finding is clinically significant given that 20% of patients with ILC have contralateral breast cancers, and it suggests that this subgroup should routinely undergo advanced breast imaging.
In comparing advanced breast imaging tools, such as MRI, we found that BSGI is simple to interpret, better tolerated by patients, and far less costly. Furthermore, BSGI has a higher sensitivity for ILC than does MRI.5 The limitations of BSGI are increased radiation exposure and a lower sensitivity for subcentimeter lesions and axillary nodal metastasis.6 Unlike other breast imaging modalities, BSGI reporting is not yet standardized by the Breast Imaging-Reporting and Data System classification. In addition, very few patients with ILC in our study also underwent MRI, prohibiting any comparisons between BSGI and MRI.
Our subgroup of patients with ILC had a higher incidence of synchronous cancer and therefore should routinely be considered for advanced breast imaging. Breast-specific gamma imaging has a high sensitivity and specificity and is an excellent adjunctive imaging tool for patients with newly diagnosed ILCs.
Corresponding Author: Nathalie G. Johnson, MD, Breast Health Center, Legacy Cancer Institute, 1040 NW 22nd, Ste 560, Portland, OR 97210 (email@example.com).
Published Online: June 24, 2015. doi:10.1001/jamasurg.2015.0965.
Author Contributions: Drs Kelley and Johnson had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Crawford.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Kelley, Johnson.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Kelley, Gardiner.
Administrative, technical, or material support: Crawford, Johnson.
Study supervision: Thomas.
Conflict of Interest Disclosures: None reported.
Previous Presentation: This paper was presented at the 86th Annual Meeting of the Pacific Coast Surgical Association; February 20, 2015; Monterey, California.