Return To Abstract Listing

Scientific Session 28 — Breast - Screening/Pathology

Friday, May 5, 2017

Abstracts 2225-3260

3260. Impact of Reinterpretation and Repeat of Breast Imaging Studies

Chang Sen L*,  Mayo C,  Lesslie M,  Yang W,  Leung J. The University of Texas MD Anderson Cancer Center, Houston, TX

Address correspondence to L. Chang Sen (

Objective: Our objective was to study the impact of second opinion interpretation and repeat breast imaging for studies submitted from outside facilities to a quaternary cancer center.

Materials and Methods: This is a retrospective institutional review board-approved study. From January 1, 2010, to June 30, 2014, 2574 studies from outside facilities were submitted to our center for a second opinion. Of these 2574 studies, 178 (6.9%) were excluded due to the following reasons: a known concurrent diagnosis of cancer, outside facility diagnosis unavailable, postsurgical excision at an outside facility, did not undergo additional imaging studies recommended by breast imagers at our institution, or presentation for screening ultrasound or high-risk screening MRI without abnormal findings on mammograms from outside facilities. When two second opinion interpretations for both the screening mammogram and diagnostic mammogram were rendered, it was only counted as one case. This resulted in 2396 cases.

Results: Of 2396 studies in our cohort, 2371 were from the United States and 25 were international studies. Of the 2371 U.S. studies, 218 were from academic centers (AC) and 2153 were from community-based centers (CC). Also of the 2371 studies, 670 were from American College of Radiology–accredited Breast Imaging Centers of Excellence (BICOE) and 1701 were not. The overall malignancy rate in the entire cohort was 11.3% (271/2396). Of 2396 studies reinterpreted, repeat or additional imaging was not recommended or performed in 235 (9.8%) cases. Additional biopsy was recommended in 191 (8.0%) cases, and 189 biopsies were performed revealing 23 cancers, 15 high-risk lesions, and 151 benign lesions. One of the 15 high-risk lesions was upgraded to invasive ductal carcinoma and ductal carcinoma in situ (DCIS) on excision, resulting in a total of 24 cancers. Five of 24 cancers were diagnosed by additional clinical evaluation. In three cases with BI-RADS category 3 assessment (probably benign) at an OSF, biopsy was also recommended by the outside facility radiologists. One case was metastatic ovarian cancer. The remaining 15 cancers reflected true additional cancer detection resulting from second opinion interpretation: 10 invasive carcinomas (two high-grade, one intermediate-to-high, one low-to-intermediate, and six low-grade lesions; median size, 0.8 cm; range, 0.4–1.7 cm) and five DCIS cases (one high-grade, three intermediate, and one low-to-intermediate-grade lesion). This translates into a positive predictive values (PPV) of 7.9% for biopsy recommended (PPV2) and biopsy performed (PPV3). Thirteen of 15 patients were referred from non-BICOE community centers, and the other two patients were referred from BICOE community centers. All 15 patients had repeat breast imaging.

Conclusion: Second opinion interpretation of breast imaging studies performed at outside facilities has a positive impact on additional cancer detection.