HER2 Status in Biopsy vs. Resection Specimens of Gastric and Gastroesophageal Adenocarcinoma – Is Tumour Heterogeneity a Problem?
Tao Wang, Eugene T Hsieh, Pauline Henry, Wedad Hanna, Catherine J Streutker, Andrea Grin. University of Toronto, Toronto, ON, Canada; Sunnybrook Health Sciences Centre, Toronto, ON, Canada; St. Michael's Hospital, Toronto, ON, Canada
Background: In advanced gastric and gastroesophageal (GE) adenocarcinomas that overexpress HER2, treatment with trastuzumab confers a survival benefit. To select patients for treatment, HER2 status is evaluated by immunohistochemistry (IHC) and in situ hybridization. It has been noted that gastric and GE adenocarcinomas demonstrate heterogeneity in HER2 expression. Nonetheless, testing is often performed on biopsy specimens alone, which raises the issue of non-representative sampling. In this study, we investigated the correlation of HER2 status between matched biopsy and resection specimens as well as the role of tumour heterogeneity in contributing to discrepancy.
Design: A total of 121 patients with gastric and GE adenocarcinoma had tissue available from an initial biopsy and subsequent resection. HER2 (4B5, Ventana) IHC was performed and evaluated by the criteria established in the ToGA clinical trial. In situ hybridization (Ventana) was performed on equivocal (2+) cases by IHC in either the biopsy or resection. Tumour heterogeneity in positive cases was defined as less than 25% of tumour cells staining 3+ or 2+ in the resection.
Results: Most tumours were intestinal (61%), 18% were diffuse, and 21% were mixed subtype. HER2 was overexpressed in 16 tumours (13%), with an overall biopsy-resection concordance of 96.7%. Overexpression was not seen in any diffuse adenocarcinomas, and only in 1 mixed adenocarcinoma. Four cases were discrepant; 2 were positive on biopsy only, and 2 were positive on resection only. Significant tumour heterogeneity was seen in 2 of 4 discrepant cases, one of which was mixed subtype; heterogeneity was not present in any concordant case (p=0.05). The average biopsy had 5.8 fragments in which 3.9 had tumour. However, in one discrepant case with a negative biopsy, only 1 of 7 fragments contained tumour.
Conclusions: Our study of gastric and GE junction adenocarcinoma demonstrates strong concordance (96.7%) between biopsy and resection specimens for HER2 overexpression. However, discordance may occur in tumours which exhibit significant heterogeneity, especially with biopsies that have poor tumour sampling. Overall, both biopsy and resection specimens are appropriate for HER2 testing, but generous sampling for biopsy specimens is necessary to ensure accurate assessment.
Wednesday, March 6, 2013 1:00 PM
Poster Session VI # 88, Wednesday Afternoon