Abstract
BACKGROUND: The epidermal growth factor receptor (EGFR) is a potential target of anticancer therapy in gastric cancer. However, its prognostic role in metastatic gastric or gastroesophageal junction (GE) cancer has not been established yet. METHODS: EGFR status was analyzed by immunohistochemistry (IHC) in paraffin-embedded samples from 357 patients who received chemotherapy in 4 first-line trials. Automated RNA extraction from paraffin and RT-quantitative PCR were additionally used to evaluate EGFR mRNA expression in 130 patients. RESULTS: EGFR protein expression (any grade) and overexpression (3+) were observed in 43% and 11% of patients, respectively. EGFR positivity correlated with intestinal type histology (p = 0.05), but not with other clinicopathologic characteristics. Median follow-up was 18.2 months. Median overall survival (OS) was similar in patients with EGFR positive vs. those with EGFR negative tumors, regardless whether positivity was defined as ≥1+ (10.6 vs. 10.9 months, p = 0.463) or as 3+ (8.6 vs. 10.8 months, p = 0.377). The multivariate analysis indicated that EGFR status is not an independent prognostic factor (hazard ratio 0.85, 0.56 to 1.12, p = 0.247). There were also no significant differences in overall survival when patients were categorized according to median (p = 0.116) or quartile (p = 0.767) distribution of EGFR mRNA gene expression. Similar distributions of progression-free survival according to EGFR status were observed. CONCLUSIONS: Unlike different cancer types where EGFR-positive disease is associated with an adverse prognostic value, EGFR positivity is not prognostic of patient outcome in metastatic gastric or GE cancer.