| Purpose:Radiotherapy, concurrent radiochemotherapy and postoperative prophylactic radiotherapy are common treatment for esophageal squamous-cell carcinoma (ESCC). However, the efficacy of radiotherapy differs significantly from individual to individual and the reason remains to be clarified. Because there are no biomarkers that can be used to predict who will benefits from the treatment, over-treatment or deficient treatment often occurs in clinic. The aim of this study was to identify genetic variants associated with survival in ESCC patients receiving radiotherapy based on genome-wide association study. In addition, because the ATM gene has been well known to play an important role in response to DNA damage agents such as ionizing irradiation and anticancer drug cisplatin, we also investigated whether single nucleotide polymorphisms (SNPs) in ATM are associated with survival in patients with III/IV ESCC receiving radiotherapy or concurrent radiochemotherapy.Experimental Design:In the first section, we conducted a genome-wide study to identify genetic variants associated with survival in 282 individuals with stage II/III ESCC receiving postoperative prophylactic radiotherapy followed by validation in an independent set of 243 individuals with the same disease and the same treatment. Finally, the significant associations were further validated in two patient sets including 262 individuals with clinical stage II, III and IV ESCC receiving radiotherapy only, and 388 individuals with pathological stage I, II and III ESCC receiving surgery only. Genotyping analysis of the GWAS samples was conducted using Affymetrix GeneChip Human Mapping 6.0, while the genotyping analysis of the validation samples was done by using the MassARRAY system. In addition, four tagSNPs of ATM were genotyped by PCR-RFLP in 412 individuals with clinical stage III or IV ESCC receiving radiotherapy or radiochemotherapy. The survival time of ESCC among patients with different genotypes was estimated by Kaplan-Meier plot and the significance was examined by log-rank test. The association of each SNP with survival was assessed by multivariable Cox proportional hazard regression analysis. We also conducted biochemical assays to investigate the function of the SNP and gene associated with ESCC survival.Results:In the first section, among 525 patients,304 (57.9%) had died of ESCC, with a median survival time (MST) of 30.0 months. We found that rs11722325, located in approximately 22 kb upstream of KIT, was significantly (P=2.82×10"7) associated with the survival time of individuals by multivariable-adjusted Cox regression model, adjusted for age, sex, smoking status, drinking status and stage of disease. MST of cases with the rs11722325 AA, AC or CC genotype was 46.0,29.0 or 19.0 months, respectively. Individuals with the rs11722325 C allele had worse MST compared with the rs11722325 A allele, with the hazard ratio (HR) for death from ESCC in the combined sample being 1.54 (95% confidence interval (CI),1.30-1.81) calculated using multivariable-adjusted Cox additive model. Significant association was found between the rs11722325 and survival of stage II-IV ESCC receiving radiotherapy (P=0.0057), but not for stage I-III ESCC patients receiving surgery only. By fine mapping, we also found a SNP, rs6554199, located in the promoter of the KIT gene and in linkage disequilibrium with rs11722325 (r2= 0.77), significantly associated with disease-free survival (P=0.0004) and overall survival (P= 3.58×10"5) in patients. MST of cases with the rs6554199 GG, GT or TT genotype was 44.0,28.0 or 20.0 months, respectively. Individuals with the rs6554199 T allele had worse MST compared with the rs6554199 G allele, with HR for death from ESCC in the combined sample being 1.50 (95% CI,1.27-1.78). Significant association was found between rs6554199 and survival of ESCC patients receiving radiotherapy only (P=8.60×10-5), but not found in ESCC patients treated with surgery only.Biochemical assays showed that rs6554199 G> T variant disturbed a transcriptional factor, NPM, specifically binding to the site of rs6554199 G allele in the KIT promoter, which might result in down-regulation of KIT mRNA and protein expression. The KIT expression was significantly lower in ESCC tissues than in the paired normal tissues. We found that the KIT expression had no effect on proliferation of ESCC cells when they were not exposed the X-ray irradiation; however, when ESCC cells were treated with X-ray irradiation, overexpression of KIT significantly suppressed ESCC proliferation, which may result from decreased repair of DSBs induced by X-ray. In addition, over-expression of KIT in ESCC cells upregulated E-cadherin but downregulated P-catenin and Slug expression, which were associated with suppressed migration and invasion of the cells.Furthermore, we also found two ATM SNPs, rs664143 and rs664677, significantly associated with survival time of ESCC patients receiving radiotherapy or concurrent radiochemotherapy. Individuals with the rs664143 AA or GA genotype had poorer MST compared with the rs664143 GG genotype (14.0 versus 20.0 months), with the HR for death being 1.45 (95% CI=1.12-1.89). Individuals with the rs664677 CC or TC genotype also had worse MST than the rs664677 TT genotype (14.0 versus 23.5 months), with the HR of 1.57 (95% CI=1.18-2.08). Stratified analysis showed that these associations were also seen in both stage III and IV cancer and different radiotherapy techniques. Significant associations were found between the four tagSNPs and loco-regional progression in patients with Ⅲ/Ⅳ ESCC receiving radiotherapy or concurrent radiochemotherapy (rs664143, P=0.018; rs664677, P=0.014; rs189037, P= 0.024; rs373759, P=0.040), but we did not find these SNPs associated with survival time of ESCC patients receiving surgery only.Conclusions:These findings indicate that rs11722325 and rs6554199 are genetic factors determing associated with ESCC prognosis in patients treated with radiotherapy. The consistent results of molecular epidemiology and biochemical assays strongly support that rs6554199 G> T is a functional genetic variation in regulation of KIT expression and aberrant expression of KIT is associated with sensitivity to radiation. In addition, the rs664143 and rs664677 SNPs in ATM might serve as independent biomarkers for predicting prognosis of radiotherapy for ESCC. This study provided evidence for future studies on individualized radiotherapy of esophageal cancer. |
PDF Full Text Request