Background HER2 and topoisomerase 2 alpha (TOP2A) genomic status was previously reported to predict benefit from anthracyclines in breast cancer. We sought to define the prognostic impact and possible pitfalls related to these biomarkers in resectable gastroesophageal adenocarcinoma. Methods HER2 and TOP2A gene amplification by fluorescent in situ hybridization and HER2 protein expression by immunohistochemistry (IHC) were assessed on whole tissue sections from 101 patients receiving peri-or postoperative epirubicin-based chemotherapy. In a subgroup of patients, at least two matched tumor blocks, originating either from surgical procedures (n = 88) or diagnostic biopsies (n = 32), were available for HER2 analyses by IHC. Results Eighteen of 101 patients (17.8 %) were HER2 positive, whereas TOP2A was amplified in 4 of 84 patients (4.7 %). HER2 positivity was significantly associated with improved disease-free survival [HR = 0.47 (95 % CI 0.22-0.99), P = 0.046] and overall survival [HR = 0.33 (95 % CI 0.13-0.83), P \ 0.018], independent of clinicalpathologic features. HER2 expression in matched tumor blocks from the same resection specimen was discordant in up to 11.8 % of pairs, while this rate increased up to 27.2 % when diagnostic biopsies and paired surgical samples were compared. Conclusions HER2 status is an independent prognostic biomarker in gastroesophageal adenocarcinomas receiving epirubicin-based chemotherapy. Compared to diagnostic biopsies, HER2 assessment in multiple resection specimens might lower the risk of sampling errors. These findings have several implications with respect to the optimal choice of the sample to be submitted to IHC testing of HER2.