Original Study Usefulness of Magnetic Resonance in Patients With Invasive Cancer Eligible for Breast Conservation: A Comparative Study Alessandro Fancellu, 1 Daniela Soro, 2 Paolo Castiglia, 3 Vincenzo Marras, 4 Marcovalerio Melis, 5 Pietrina Cottu, 1 Alessandra Cherchi, 1 Angela Spanu, 6 Silvia Mulas, 1 Claudio Pusceddu, 7 Luca Simbula, 2 Giovanni B. Meloni 2 Abstract In this comparative study, magnetic resonance imaging (MRI) of the breast altered the treatment plan in 16.5% of patients with breast cancer who were candidates for breast conservation. MRI was associated with higher mastectomy rates compared with a conventional preoperative workup but did not reduce reoperation rates for positive margins. Lymph node positivity and tumor size > 15 mm may predict an MRI-triggered change of surgical plan. Background: The role of magnetic resonance imaging (MRI) in newly detected breast cancer remains controversial. We investigated the impact of preoperative MRI on surgical management of infiltrating breast carcinoma (IBC). Methods: We reviewed data of 237 patients with IBC who were suitable for breast-conserving surgery (BCS) between 2009 and 2011. Of these patients, 109 underwent preoperative MRI (46%; MRI group) and 128 did not (54%; no-MRI group). We analyzed MRI-triggered changes in surgical plan and compared differences in rates of positive margins and mastectomy. Results: Tumor size was larger in the MRI group (16.8 mm vs. 13.9 mm; P < .001). MRI changed the initial surgical planning in 18 of 109 patients (16.5%) because of detection of larger tumor diameter requiring wider resection (8 patients [7.3%]) or additional malignant lesions in the ipsilateral (9 patients [8.2%]) or contralateral breast (1 patient [0.9%]). MRI-triggered treatment changes included mastectomy (n ¼ 12), wider excision (n ¼ 5), and contralateral BCS (n ¼ 1). Reoperation rates for positive margins after BCS appeared higher in the no-MRI group (4.1% vs. 8.6%), but the difference missed statistical significance (P ¼ .9). Overall mastectomy rates were higher in the MRI group (13.7% vs. 7.0%; P < .05). The likelihood of having a change of treatment resulting from MRI was significantly higher for patients with tumors > 15 mm and for those with positive lymph nodes. Conclusion: Lymph node positivity and tumor size > 15 mm may predict an MRI-triggered change in surgical plan. Preoperative MRI resulted in higher mastectomy rates justified by biopsy-proven additional foci of carcinoma and did not significantly reduce reoperation rates for positive margins. Clinical Breast Cancer, Vol. 14, No. 2, 114-21 ª 2014 Elsevier Inc. All rights reserved. Keywords: Breast-conserving surgery, Breast cancer treatment, Invasive carcinoma, Magnetic resonance imaging, Margins Introduction Breast-conserving surgery (BCS) followed by whole-breast irra- diation is currently considered the standard of care for the local management of early breast cancer, because evidence exists that such an approach is as effective as mastectomy with regard to oncologic outcomes. 1,2 Appropriate surgical planning has traditionally been performed through the “triple assessment,” ie, clinical examination, radiologic investigations (mammography and ultrasonography 1 Unit of General Surgery II, Clinica Chirurgica 2 Department of Radiology 3 Department of Biomedical Sciences 4 Department of Pathology University of Sassari, Sassari, Italy 5 New York University School of Medicine and Department of Surgery, NY Harbor Healthcare System VAMC, New York, NY 6 Department of Nuclear Medicine, University of Sassari, Sassari, Italy 7 Department of Radio-oncology, Oncological Hospital of Cagliari, Cagliari, Italy Submitted: Jun 22, 2013; Accepted: Oct 2, 2013; Epub: Oct 25, 2013 Address for correspondence: Alessandro Fancellu, MD, PhD, Unit of General Surgery II, Clinica Chirurgica, University of Sassari, V.le San Pietro 43, 07100 Sassari, Italy Fax: þ39-079-228394; e-mail contact: afancel@uniss.it 114 - Clinical Breast Cancer April 2014 1526-8209/$ - see frontmatter ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.clbc.2013.10.002