Abstract P1-01-05: Effects of preoperative MRI on rate of ipsilateral and contralateral recurrence of breast cancer

Abstract

Abstract Introduction: Preoperative MRI of the breast is the most sensitive imaging modality in the detection of multifocal or multicentric breast cancer, as well as simultaneous contralateral breast cancer. The aim of this retrospective review was to evaluate the effect of preoperative MRI on local control for patients with breast cancer. Methods: The Enterprise Data Warehouse of Northwestern Medicine was searched for women who underwent breast conserving surgery for ductal carcinoma in situ (DCIS) or primary invasive breast cancer in the interval of 2004-2010. The use of preoperative MRI, and the clinical and therapeutic details of the patients thus identified were extracted by direct review of the electronic medical record. A breast event was defined as a local recurrence in the treated breast more than six months after completion of treatment (ipsilateral) or a new breast cancer in the untreated breast (contralateral). Differences in the frequency of all events (local and distant), for ipsilateral breast events, and for contralateral breast events was evaluated with Cox proportional hazards model, adjusting for patient age, tumor size, nodal status, the presence of triple negative disease, and the use of radiotherapy and systemic therapy. Results: In our cohort of 1097 patients, 526 had preoperative MRI and 571 had no MRI. The patients who had preoperative MRI were younger (59 vs. 66 years, p<0.0001), were more commonly premenopausal (37.8% vs. 27.3%, p=0.0004), were more likely to present with palpable tumors (34.8% vs. 26.6%, p=0.004), were more likely to have invasive lobular disease (16% vs. 11.6%), and less likely to have DCIS (16.5% vs. 29%, p=0.001 for differences in histologic pattern). Mean tumor size was equivalent in the two groups (17.5 and 17.3 mm), but nodes were more frequently positive in the MRI group (23.9% vs. 19.1%, p=0.045). Triple negative tumors were more frequent in the MRI group (14.1% vs. 7.5%, p=0.0003). Mean follow up was 51.5 months in the MRI group and 59.4 months in the no MRI (p<0.0001). The number of events was 49 in the MRI group and 68 in the no MRI group. The Cox hazard ratio (HR) for all events (adjusted for follow-up duration and factors described in the Methods) was equivalent between the two groups (HR 0.90, 95% CI 0.59-1.36, p=0.61). The HR for ipsilateral (HR 0.93, 95% CI 0.57-1.51, p=0.76) and contralateral events (HR 1.22, 95% CI 0.57-2.62, p=0.61) was equivalent between the two groups. Conclusions: In analyses adjusted for important prognostic features, the use of preoperative breast MRI was not associated with a reduced hazard of any breast cancer event, or of in-breast events (ipsilateral or contralateral). However, the MRI group had a more adverse tumor and patient profiles; a propensity score analysis will be performed, to further adjust for these differences. These findings add weight to the position that routine use of preoperative MRI for all breast cancer patients is not beneficial. Citation Format: Amanda L Amin, Irene B Helenowski, Thomas E Kmiecik, Shruti R Zaveri, Nora M Hansen, Kevin P Bethke, Seema A Khan. Effects of preoperative MRI on rate of ipsilateral and contralateral recurrence of breast cancer [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P1-01-05.