Brazilian randomized study: Impact of preoperative magnetic resonance in the evaluation for breast cancer conservative surgery (BREAST-MRI Trial)

Date 03 May 2019
Event ESMO Breast Cancer 2019
Session Poster lunch
Topics Imaging
Breast Cancer
Presenter Bruna Mota
Citation Annals of Oncology (2019) 30 (suppl_3): iii39-iii44. 10.1093/annonc/mdz098
Authors B.S. Mota1, Y.N. Reis1, M.T. Doria1, M.D. Ricci1, C. Shimizu2, V. Ferreira2, T. Tucunduva2, N. de Barros2, E.C. Baracat1, J.R. Filassi1
  • 1Gynecology, ICESP - Instituto do Cancer do Estado de Sao Paulo, 01246-000 - Sao Paulo/BR
  • 2Radiology, ICESP - Instituto do Cancer do Estado de Sao Paulo, 01246-000 - Sao Paulo/BR



Breast magnetic resonance imaging (MRI) has a high sensitivity (95-100%) in detecting invasive neoplasms. Controversy still exists whether preoperative staging with breast MRI has an impact on surgical outcomes. The aim of this randomized controlled trial is to evaluate the ability of MRI in selecting patients for breast conserving surgery (BCS).


BREAST-MRI is a randomized, open label, unblinded trial conducted at ICESP, São Paulo, Brazil. Patients were stratified for mammary density and randomized on a 1:1 basis. Sample size was calculated assuming a recurrence rate of 10% for conservative surgery and 1% for mastectomies for invasive breast cancer. All patients were female, older than 18 years old, diagnosed with breast cancer stage 0 to III and candidates for BCS were evaluated by physical exam, breast ultrasound (US) and mammography. Breast MRI were performed on a 1.5T MRI system (Signa HDxt®, GE Healthcare). Breast density on mammography were assessed using ACR BI-RADS®. The surgery was modified when MRI showed an increase of more than 50% of the tumor size. Primary end-point was accuracy of MRI and rate of mastectomies.


815 patients were eligible, 446 patients were included. MRI was performed in 49,1% (219) patients. Groups were similar according to clinical and tumor characteristics (Table). The MRI group showed surgical change management in 68 patients (31,1%). Surgical changes were made in the ipsilateral breast in 49 cases (72,1%), at contralateral breast in 13 cases (19,1%) and at both breasts in 6 patients (8,8%). MRI correctly modified the surgical procedure in 44 out of 55 patients (80%) for ipsilateral breast and 7 out of 13 patients (53,8%) from contralateral breast. There was no difference at the accuracy rate of MRI according to the mammary density (p 0,74). Conversion to mastectomy occurred in 19 patients at the ipsilateral breast.


Preoperative MRI changed surgical management in 31,1% of the cases, to either wide BCS or mastectomy, with an accuracy rate of 80%. However, it did not change reoperation rates.

Editorial acknowledgement

Clinical trial identification


Legal entity responsible for the study



Has not received any funding.


All authors have declared no conflicts of interest.

Table: 125P

Baseline characteristics

MRI Group (n = 219)Control Group (n = 227)p
Age (median)57,357,90,45
Body Mass Index (kg/m2)29,329,40,35
Yes20 (9,2%)25 (11%)
No198 (90,8%)202 (89%)
Missing information10
Menopausal status0,26
Premenopausal64 (29,4%)56 (24,7%)
Postmenopausal154 (70,6%)171 (75,3%)
More than 5 years2 (11,1%)6 (26,1%)
Less than 5 years16 (88,9%)17 (73,9%)
Mammary density0,68
A12 (5,5%)12 (5,3%)
B93 (42,5%)99 (43,6%)
C101 (46,1%)108 (47,6%)
D13 (5,9%)8 (3,5%)
Clinical Stage0,28
028 (12,8%)34 (15%)
I97 (44,3%)108 (47,6%)
II91 (41,6%)85 (37,4%)
III3 (1,4%)0
BCS168 (76,7%)226 (99,5%)
Wide BCS35 (16%)0
Mastectomy13 (5,9%)1 (0,5%)
NSM3 (1,4%)0
Re operation rates0,646
Further wide local excision12 (5,5%)11 (4,8%)
Mastectomy5 (2,3%)5 (2,2%)
Yes103 (49%)83 (37,2%)
No107 (51%)140 (62,8%)
Yes199 (95,2%)209 (95%)
No10 (4,8%)11 (5%)
Hormone therapy0,59
Yes181 (86,6%)196 (88,3%)
No28 (13,4%)26 (11,7%)
Follow-up (months)23,6 (1-45)23,6 (1-53)0,99
Local recurrence01 (0,4%)
Distant recurrence4 (1,8%)3 (1,3%)
Breast cancer death01 (0,4%)
Any death2 (0,9%)1 (0,4%)

HRT: Hormone replacement therapy.