c-MYC Copy-Number Gain Is an Independent Prognostic Factor in Patients with Colorectal Cancer

BACKGROUND The aim of this study was to determine the incidence and clinicopathological significance of c-MYC gene copy-number (GCN) gain in patients with primary colorectal cancer (CRC). METHODS The c-MYC GCN was investigated in 367 consecutive CRC patients (cohort 1) by using dual-color silver in situ hybridization. Additionally, to evaluate regional heterogeneity, we examined CRC tissue from 3 sites including the primary cancer, distant metastasis, and lymph-node metastasis in 152 advanced CRC patients (cohort 2). KRAS exons 2 and 3 were investigated for mutations. RESULTS In cohort 1, c-MYC gene amplification, defined by a c-MYC:centromere of chromosome 8 ratio ≥ 2.0, was detected in 31 (8.4%) of 367 patients. A c-MYC GCN gain, defined by ≥ 4.0 c-MYC copies/nucleus, was found in 63 (17.2%) patients and was associated with poor prognosis (P = 0.015). Multivariate Cox regression analysis showed that the hazard ratio for c-MYC GCN gain was 2.35 (95% confidence interval, 1.453-3.802; P < 0.001). In a subgroup of stage II-III CRC patients, c-MYC GCN gain was significantly associated with poor prognosis by univariate (P = 0.034) and multivariate (P = 0.040) analyses. c-MYC protein overexpression was observed in 201 (54.8%) out of 367 patients and weakly correlated with c-MYC GCN gain (ρ, 0.211). In cohort 2, the c-MYC genetic status was heterogenous in advanced CRC patients. Discordance between GCN gain in the primary tumor and either distant or lymph-node metastasis was 25.7% and 30.4%, respectively. A similar frequency for c-MYC GCN gain and amplification was observed in CRC patients with both wild-type and mutated KRAS. CONCLUSIONS c-MYC GCN gain was an independent factor for poor prognosis in consecutive CRC patients and in the stage II-III subgroup. Our findings indicate that the status of c-MYC may be helpful in predicting the patients' outcome and for managing CRC patients.