HER2 heterogeneity in adenocarcinoma of the distal esophagus and stomach.

Abstract

4567 Background: Patients with HER2 positive adenocarcinoma of the esophagus or stomach are eligible for HER2 targeted therapy, which can improve survival in selected patients. Previous research shows that HER2 gene amplification and HER2 overexpression is frequently heterogeneous within these tumors. Biopsies taken from heterogeneous tumors for predictive testing may therefore result in false-negative outcomes. The objective of this study was to assess HER2 amplification and expression in biopsies and paired resection specimens with adenocarcinoma of the esophagus or stomach, from patients who did not receive neoadjuvant systemic therapy. Methods: Paired biopsies and resection specimens of patients with adenocarcinomas of the esophagus or stomach were retrospectively selected. Immunostaining was performed on all samples using antibody 4B5 (Ventana Medical Systems) and Silver-In-Situ-Hybridization was performed in selected cases. Scoring for HER2 was performed according to the method described by Hofmann et al. (2008). Results: We included 378 cases for analysis. In both biopsies and resection specimens 14% of the cases were HER2 positive. Intratumor heterogeneity in HER2 positive tumors was present in 45% ( n= 24/53) in biopsies and 75% ( n= 39/52) in resection specimens. In HER2 positive resection specimens, 65% ( n= 34/52) of paired biopsies were also positive. In the 18 remaining discordant tumors (resection HER2 positive, biopsy negative), intratumor heterogeneity was present in 16/18 cases. For HER2 negative resection specimens all paired biopsies were also HER2 negative. SISH was performed in 110 tumors. Agreement of HER2 gene amplification between biopsy and resection specimens was observed in 86% (n = 95/110). Five HER2 negative biopsies were positive in the resection specimen. Conclusions: The results of this study indicate that predictive HER2 assessment in adenocarcinoma of the esophagus or stomach can lead to false negative results based on biopsies. As a result, patients with HER2 positive tumors can unintentionally be denied neoadjuvant HER2 targeted therapy. The set of patients investigated in this present study is unique because of the absence of any systemic and/or radiation therapy between the biopsy and the resection of the tumor. Hopefully these results can help in developing methods for improved patient selection for HER2 targeted therapy.