Abstract P5-22-02: Surgical strategy after neoadjuvant therapy in patients with operable breast cancer can be optimized by knowledge of the level of agreement of measured tumor size on MRI after neoadjuvant therapy and final pathologic assessment

Abstract

Abstract Introduction Neoadjuvant chemotherapy (NAC) is increasingly applied in the treatment of patients with operable breast cancer. Wide local excision after NAC aims to remove the complete residual tumor with no tumor on ink, without compromising cosmetics. Therefore, surgical strategies are determined based on the remaining tumor size after NAC visualized on magnetic resonance imaging (MRI). Recent studies on the correlation between preoperative MRI and pathological response demonstrate inconsistent results. A need remains to adequately guide surgical decisions after NAC. The aim of this study was to not only investigate the correlation between MRI and pathological evaluation but to gain knowledge of the exact level of agreement, specified for different tumor subtypes, which could further guide surgical decision making. Methods All patients operated for breast cancer after NAC between January 2013 and July 2016 in a large teaching hospital were retrospectively included. Longest residual tumor diameter was determined with MRI and correlated with postoperative pathological findings. Tumors were subdivided based on estrogen receptor (ER) status and human epidermal growth factor receptor 2 (HER2). Spearman correlation was used to correlate MRI and pathological tumor size findings. Bland-Altman method was used to evaluate the agreement between both measurements. Results 193 patients with 195 breast cancers were included. The correlation between tumor size at MRI and pathology was 0.63 for the whole group, 0.39 for tumors with subtype ER+/HER2-, 0.55 for ER+/HER2+, 0.63 for ER-/HER2+ and 0.85 for ER-/HER2-. The correlation for lobular carcinomas was 0.44. The mean difference and limits of agreement (LoA), between tumor size measured at MRI and pathological size was 4.6 mm (LoA -27.0 to 36.3 mm, n=195). Mean difference and LoA for subtype ER+/HER2- was 7.6 mm (LoA -31.3 to 46.5 mm, n=100), for ER+/HER2+ 0.9 mm (LoA -8.5 mm to 10.2 mm, n=33), for ER-/HER2+ -1.2 mm (LoA -5.1 to 7.5 mm, n=21), for ER-/HER- -0.4 mm (LoA -8.6 to 7.7 mm, n=41) and 19.4 mm (LoA -16.8 to 55.6 mm, n=14) for lobular carcinoma. Conclusion The correlation and agreement between the post-NAC MRI and postoperative pathological assessment of residual tumor size for ER+/Her2- and lobular tumors is weak. The agreement shows a wide variation in over- and underestimation of tumor size in mm by MRI and thus surgical strategy can still be poorly guided. It does however provide us with exact information on the possible range of margins we should take into account at surgery. As demonstrated in other studies ER+/HER2+, ER-/HER2+ and ER-/HER2- tumors demonstrate a clear correlation. Also, the level of agreement in these tumor subtypes shows that we can use MRI evaluation as a reliable predictive tool of tumor residual size to base our surgical strategy upon with a small variation in over- and underestimation of tumor size. Because the lobular carcinoma and HER2+ subgroups were small, conclusions should of course be viewed cautiously. Citation Format: Boersma C, van Veen JLC, Maaskant JM, Van der Starre-Gaal J, Van 't Veer-Ten Kate M, Francken AB, Noorda EM. Surgical strategy after neoadjuvant therapy in patients with operable breast cancer can be optimized by knowledge of the level of agreement of measured tumor size on MRI after neoadjuvant therapy and final pathologic assessment [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P5-22-02.