Abstract P2-02-09: Accuracy of MRI for estimating residual tumor size after neoadjuvant endocrine therapy in early stage breast cancer with ER positive and HER2 negative

Abstract

Abstract Back ground Neoadjuvant endocrine therapy (NET) is increasingly being used in the management of breast cancer patients with ER positive and HER2 negative. A major clinical benefit is that patients with large operable tumors may be offered breast-conserving surgery (BCS). Therefore, success of BCS following NAT depends on accurate assessment of the residual tumor extent after NAT. The correlation between the extent of the tumor at pathology and preoperative assessments of the size obtained by physical examination and conventional imaging is impaired by hormonal therapy-induced necrosis and fibrosis. However, the most superior method to assess the extent of residual disease after NAT was unknown. The aim of this study was to evaluate the relative accuracies of mammography(MMG), ultrasound(US), specimen mammography(SMMG) and MRI in predicting residual tumor size and pathological response after neoadjuvant endocrine therapy (NET) in patients with ER positive and HER2 negative. Each prediction method (EPM) was compared with the gold standard of surgical pathology. Patients and methods From April 2010 to December 2012, 50 post menopausal patients with breast cancer at clinical stages T1c-T2N0 were prospectively enrolled in this study on NET. Fifty patients (age range, 52-75 years; mean age, 64.1 years) who received NET with 24 weeks of letrozole were evaluated with EPM before and after NET. We compared the predicted residual tumor size and the predicted response on imaging with residual tumor size and response on the pathologically determined size. We defined the tumor sizes measured by EPM and pathology as in agreement when the greatest tumor dimension measured by EPM was within 70_120% of the measurement determined by microscopic pathology. The EPM assessment was categorized as an underestimation when the longest diameter on imaging was <70% that at pathology and as an overestimation when the longest diameter on imaging was>120% that at pathology. Statistical analysis was performed using logistic regression analysis. Results The correlation coefficient between the residual tumor sizes determined by pathology and the predicted tumor size was 0.721(p = 0.651) for MMG, -0.111(p = 0.642) for US, 0.723(P = 0.001) for SMMG and 0.714 (p<0.000001) for MRI. The rate of agreement between the final response predictions and the responses measured by pathology were 50% for MMG, 40% for US, 35% for SMMG and 78% for MRI. The MMG,US, SMMG and MRI measurement disagreed with the pathologically determined size in 50%,60%,65%,22%, overestimating the size in 0%,12%, 11%,2% and underestimating the size in 50%, 48%,53%,20% respectively. Conclusions Predictions of response and residual tumor size made on MRI were best correlated with the assessments of response and residual tumor size made on pathology than were predictions made on the basis of clinical examination, mammography or ultrasound. MRI is one of the most accurate methods for predicting the extent of residual tumor after NET. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P2-02-09.