Gene amplified in oesophageal cancer 1 (GAEC1) amplification in colorectal cancers and its impact on patient's survival.

GAEC1 (gene amplified in oesophageal cancer 1) is located at 7q22.1, first identified in oesophageal cancer. Initial work indicated that GAEC1 can act as an oncogene. Our pilot study found ∼80% of colorectal cancers showing amplification of GAEC1. In this research, we will study GAEC1 copy number in colon cancer cell lines and colorectal tissues, and its prognostic significance. Two human colon cancer cell lines (SW480 and SW48) and one normal colonic epithelial cell line (FHC) were obtained from American Type Culture Collection. Culturing conditions for these cell lines were as published previously. Tissues were collected from 283 patients (213 Australian; 70 Japanese) diagnosed with colorectal cancers. Ninety surgically removed non-cancer colorectal tissues (diverticular diseases, hyperplastic polyps and volvulus) were used as controls. H&E stained sections from each cancer were checked to select a block with sufficient cancer tissue and representative morphological features for each patient for DNA extraction. Cancers were staged according to the Union for International Cancer Control (UICC) for TNM (tumour, node and metastasis) classification. Only primary adenocarcinomas were included. Actuarial survival rates of patients were calculated from the date of surgical resection of the colorectal cancers to the date of death or last follow-up. DNA extraction and PCR reactions for GAEC1 copy number were performed as in our previous studies. 4 A ΔCt of <−1 was considered as gain of GAEC1 copies, Ct value >1 was considered as reduced GAEC1 copies and a Ct between these ranges was defined as normal/no change in GAEC1 copies. GAEC1 DNA was detectable in all colon cell lines tested. The colon cancer cell lines (SW480 and SW48) showed reduced GAEC1 copies compared with normal colonic epithelial cells (FHC) (figure 1A). At the tissue level, GAEC1 gene copies were higher in colorectal cancer compared with non-cancer tissues (p<0.0001; table 1). In Australian samples, there was no significant correlation between GAEC1 copy number and clinical/pathological parameters (table 2). Japanese patients showed increased percentage of GAEC1 deletion in colorectal cancers with lymph node metastasis compared with cancers without any lymph node metastasis (10% vs 0%, p=0.046) (table 3).GAEC1 deletion was less common in early stage (Stages I and II) colorectal cancers compared with advanced stage (Stages III and IV) cancers in Japanese patients (0% vs 10%, p=0.046) (table 3). In addition, the proportion of cancers with GAEC1 amplification was slightly higher in the Japanese compared with the Australian populations (80% vs 59%). GAEC1 copy number distribution was significantly different, with GAEC1 copy number higher among Japanese cancers compared with Australian (p=0.003). Median follow-up period for Australian and Japanese patients was 5 years. Survival data was available in 188 Australian patients (134 survived and 54 died of cancer). Among Japanese patients, 64 patients had follow-up data (43 patients survived and 21 died of cancer). Kaplan-Meier analysis indicated that mean survival of Australian and Japanese patients with colorectal cancer depended on pathological stage of cancer (p<0.0001). For Australian and Japanese patients with colorectal cancer, patients with high GAEC1 copies had longer survival compared with patients with low/