Screening breast MR imaging in women with a history of chest irradiation.

PURPOSE: To assess the utility of screening magnetic resonance (MR) imaging in detecting otherwise occult breast cancers in women with a history of radiation therapy to the chest. MATERIALS AND METHODS: This HIPAA-compliant study was approved by the authors' institutional review board. The need for informed consent was waived. Retrospective review of the radiology department database identified 247 screening breast MR imaging examinations performed between January 1999 and December 2008 in 91 women with a history of chest irradiation. Findings and recommendations for each breast MR study and on the most recent mammogram were reviewed. The number of cancers diagnosed, their method of detection, and tumor characteristics were examined. The exact 95% binomial proportion confidence intervals were calculated by using methods described by Clopper and Pearson. RESULTS: Biopsy was recommended for 32 suspicious lesions on 27 (11%) of 247 MR imaging studies in 21 women. Seven cancers were identified in 30 lesions sampled (23%). Biopsy was recommended in five additional patients on the basis of mammographic findings, and malignancy was detected in three. Ten cancers were detected during the study period: four detected with MR imaging alone, three with MR imaging and mammography, and three with mammography alone. The four cancers detected with MR imaging alone were invasive carcinomas. Two of three cancers detected with mammography alone were ductal carcinoma in situ (DCIS), and the third was DCIS with microinvasion. CONCLUSION: MR imaging is a useful adjunct modality to screen high-risk women with a history of chest irradiation, resulting in a 4.4% (95% confidence interval: 1.2%, 10.9%) incremental cancer detection rate; the sensitivity for detecting breast cancers by using a combination of MR imaging and mammography was higher than that for either modality alone.

Duke Scholars

Author Kevin Charles Oeffinger Medicine, Medical Oncology