[Performances of breast magnetic resonance imaging in the context of neoadjuvant chemotherapy for breast cancer to predict pathological complete response].

PURPOSE:The objective of this work was to estimate the reliability of MRI after neoadjuvant chemotherapy (NAC) for breast cancer to detect a residual tumour by comparing the tumoral size measured by MRI with the histological size. We also estimated the concordance of diagnosis of complete pathological response between histological examination and MRI. MATERIALS AND METHODS:We included all the patients who received a neoadjuvant chemotherapy for breast cancer in the university hospital of Tours from January, 2008 to December 31st, 2012 and in the comprehensive cancer centre of Rennes from January, 2008 till May 31st 201. We considered that the pathological response was complete (pCR) when there was no residual invasive tumour in the mammary surgical specimen. RESULTS:Two hundred and fifty-one women who received NAC for a non-metastatic breast cancer were included in the study: 103 in Tours and 148 in Rennes. Two women (0.8%) refused breast surgery whatever the type. One hundred and twenty-three (49%) women had a breast conservative surgery. One hundred and fifteen (45.8%) had a mastectomy and 11 (4.4%) had breast conservative surgery followed by mastectomy for positive margins. A complete pathological response was present in 54 cases (21.5%). We did not found any significant difference between characteristics of patients with pCR or not. CONCLUSION:Breast MRI remains the most performing examination to evaluate the initial tumoral size and the residual tumour after NAC, but does not add any value at mid or at the end of treatment for the patients to whom a mastectomy is decided at presentation. The correlation between the breast MRI and the histology size is not perfect, but at the moment, MRI stills of the most performing examination to predict the pCR.