Associations Of Genetic Variations And Survival Of Patients With Esophageal Or Small-Cell Lung Cancer

Accumulating evidence has suggested that genetic variants play important roles in survival length of patients with cancer. Genome-wide association studies (GWAS) have identified two SNPs rs1050631and rs7242481located in SLC39A6(also known as LIV-1), one of subfamilies of ZIP transporter, significantly associated with survival time of esophageal squamous-cell carcinoma (ESCC) patients (Wu et al, Nat Genet2013). The rs7242481G>A mutation, located in the5’ untranslated region of SLC39A6, changes the binding pattern of transcriptional factors and results in up-regulation of the gene expression. In the present study, we further examined the association of SLC39A6expression with survival length of ESCC patients and its underlying mechanism. Using real-time PCR analysis, we found that72.4%(55/76) of primary ESCC had significantly higher SLC39A6mRNA expression compared to their adjacent normal esophageal tissues. We randomly selected26pairs of ESCC tumor and normal tissues for immunoblotting determination of SLC39A6protein levels and found that69.2%(18/261samples had SLC39A6protein over-expressed in ESCC tissues compared with adjacent normal epithelia. Immuohistochemical staining of SLC39A6protein on tissue microarrays of100ESCC samples showed that higher SLC39A6expression was significantly associated with shorter survival time in patients with advanced ESCC (stage III/IV). In patients with stage III/IV ESCC, the median survival time (MST) was12months for patients with immunoreactive score (IRS)≥3compared with41months for those with IRS≤2, with the hazard ratio (HR) for patients with IRS≥3being2.38(95%CI,1.17-4.84; P=0.0173). In order to further characterize the underlying mechanism of SLC39A6in ESCC survival, siRNA-mediated SLC39A6silencing was performed in SLC39A6over-expressed cells. Down-regulation of SLC39A6expression significantly repressed cell proliferation (all P<0.05for72hours and96hours), cell migration and invasion (P <0.05); however, transiently reduced SLC39A6levels neither significantly affected ESCC apoptotic rates nor significantly affected cell cycle arrest.We conducted an association analysis between22GWAS-identified lung cancer susceptibility/survival-associated SNPs and overall survival of874small-cell lung cancer patients. HRs were calculated using multivariate Cox models for association analysis of different genotypes and survival. For cumulative analysis, three genetic models additive, dominant and recessive were used to assess the association between individual SNP and overall survival. The most statistically significant model was considered to be the best one. Patients were divided into low-(0-1unfavorable genotype), medium-(2unfavorable genotypes) and high-risk (3-4unfavorable genotypes) groups by counting the number of unfavorable genotypes. After FDR-adjusted for multiple comparisons, rs4809957, rs36600, rs716274and rs1878022SNPs were shown significantly associated with overall survival of small-cell lung cancer patients. Minor allele of rs716274showed an increased risk of death for small-cell lung cancer. The other three SNPs were significantly associated with better survival. We found a significant gene-dosage effect on overall survival with the increasing of unfavorable genotypes. Compared with patients in the low-risk groups’patients, HRs for patients in medium-and high-risk groups were1.46(95%CI,1.10-1.93; P=0.0080) and1.74(95%CI,1.33-2.29; P=6.11×10-5). patients in low-risk groups had MST of36months, which was significantly longer that those in medium risk (MST=27months) and high-risk (MST=22months) groups (Plog-rank=0.0005).In conclusion, these results showed that over-expressed of SLC39A6may promote ESCC cell proliferation, migration and invasion, and thus was associated with shorter survival time in patients with advanced ESCC (stage Ⅲ/Ⅳ). Some SNPs identified to be associated with susceptibility to lung cancer are also significantly associated with survival of patients with small-cell lung cancer. These findings support our hypothesis that genetic variations play impotent roles in determining survival time of patients with cancer.