HER2 expression and response to neoadjuvant chemoradiotherapy in esophagogastric adenocarcinoma.

Abstract

e15145 Background: Targeting HER2 overexpression with trastuzumab in metastatic esophagogastric adenocarcinoma (EGA) improves survival with clinical trials currently examining the impact of HER2 inhibition in early stage disease. This study analyzed the pattern of HER2 overexpression in matched-pair tumor samples of patients (pts) undergoing neoadjuvant chemoradiotherapy (chemoRT). Methods: This analysis included pts with EGA who underwent neoadjuvant chemoRT followed by esophagectomy at the University of Florida from 2001 to 2011 who did not receive anti-HER2 therapy. Demographics, risk factors, tumor features, and outcome data were analyzed. Central blinded IHC was performed on paraffin embedded tumor specimens with HER2 expression scored as negative (-) (0, 1+), indeterminate (2+) or positive (+) (3+). Descriptive statistics, Chi-square exact test, uni and multivariate analyses were used for statistical analyses. Results: 49 pts (41M/8F) were analyzed with median age 65 years (39-87). Most tumors were stage II (55%) or III (43%). 9/49 pts (18%) had pathologic complete response (pCR), another 10/49 had near pCR or not enough tumor (NET) to examine. HER2 expression was more common in lower stages (stage I, II vs III). Baseline HER2 expression did not correlate with treatment response. Univariate analysis did not show significant correlation of age, stage, smoking, pre or post treatment HER2 expression with survival. For matched-pair samples failing to achieve pCR or NET (n=30), pre and post treatment HER2 expression profiles are presented in the Table. HER2 expression appears UP regulated in response to neoadjuvant chemoRT. Conclusions: Our study demonstrated the degree of pre-treatment HER2 expression in EGA does not predict the response to chemoRT. However, HER2 overexpression is more frequent in post treatment samples without pCR. This suggests that HER2 may be upregulated by chemoRT with unreliable concordance between pre and post treatment samples. This has implications for both early and advanced stage therapeutic interventions with targeted therapies. [Table: see text]