Reliability and Reproducibility of Breast MRI as A Roadmap in Treatment Planning of Breast Cancer –A Single Centre Experience

Abstract

Aim: To study the significance of MRI in the preoperative work up of patients with newly diagnosed breast cancer and its potential to create a road map for surgical and definitive treatment. To study and define those interpretation criteria of MRI that serve to increase its specificity in pre-treatment breast cancer assessment with emphasis on non mass enhancement (NME). To study its role in predicting the nodal status in diagnosed cases of breast cancer. Material and methods:MRI and pathological data of 156 patients spread over period from September 2011 to JUNE 2014 were studied retrospectively. Morphology and enhancement characteristics of breast lesions, both masses and NME and axillary lymph nodes were analysed on DCE MRI with subsequent histopathological correlation.Results: Our study showed Sensitivity (Sn): 79%, Specificity (Sp): 86%, Positive Predictive value (PPV): 76%, Negative predictive value (NPV): 88% for identification of extratumoral DCIS (Ductal carcinoma in situ) while Sn: 74%, Sp: 76, PPV: 52%, NPV: 89% for extensive intraductal component (EIC) on histopathology in patients having non mass enhancement (NME) on DCE-MR. When NME was correlated with extratumoral DCIS and EIC taken together, we obtained Sn: 89%, Sp: 78%, PPV: 50%, NPV: 97%. Multifocal/multicentric tumours with or without satellite nodules on MRI had Sn: 86%, Sp: 89%, PPV: 60%, NPV: 97% for multiple lesions on histopathology. For detection of axillary nodal metastasis, fatty hilum effacement showed Sn: 65%, Sp: 89%, PPV: 86%, NPV: 73% and cortical thickening with and without fatty hilum effacement showed Sn: 86%, Sp: 69%, PPV: 73%, NPV: 84%. Conclusion:MRI plays an important role in assessment of disease extent in breast cancer patients desirous of breast conservation surgery and has a potential to act as a road map in treatment planning. Following certain diagnostic criteria it helps in identifying the in situ cancers with a reasonable high specificity.