Background The Notch signaling pathway is involved in the progression of esophageal squamous cell carcinoma (ESCC), although the roles of single nucleotide polymorphisms (SNPs) of the Notch signaling pathway genes in the process remain unknown. Methods The present study included 1009 patients with histopathologically diagnosed ESCC at Fudan University Shanghai Cancer Center. Two‐stage multivariate Cox proportional hazards regression analysis was used to estimate associations between 13,248 SNPs in 103 Notch signaling pathway genes and overall survival of the patients. Results We found that overall survival of the patients was significantly associated with genotypes of HDAC9 rs1729318 (AT+TT vs. AA: hazard ratio = 1.44, 95% confidence interval = 1.16–1.80, pcombined = 0.001) and HDAC9 rs1339555498 (GT + TT vs. GG: hazard ratio = 1.38, 95% confidence interval = 1.10–1.74, pcombined = 0.005). Further receiver operator characteristic (ROC) curve analysis indicated that the model with both available clinical factors and these two SNPs improved the area under the ROC curve compared to the model with clinical factors only (1‐year: 0.66 vs. 0.64, p = 0.034). Additional expression quantitative trait loci analysis showed that the rs1729318 T variant genotypes were associated with increased mRNA expression levels of HDAC9 in normal esophageal muscular tissue (p = 0.003). Conclusions The results suggest that these two potential functional SNPs on HDAC9 may serve as biomarkers for predicting survival of ESCC patients. However, further studies are needed to confirm these findings.

BackgroundCancers arising within the gastrointestinal tract are complex disorders involving genetic events that cause the conversion of normal tissue to premalignant lesions and malignancy. Shared genetic features are reported in epithelial-based gastrointestinal cancers which indicate common susceptibility among this group of malignancies. In addition, the contribution of rare variants may constitute parts of genetic susceptibility.MethodsA cross-cancer analysis of 38,171 shared rare genetic variants from genome-wide association assays was conducted, which included data from 3,194 cases and 1,455 controls across three cancer sites (esophageal, gastric and colorectal). The SNP-level association was performed by multivariate logistic regression analyses for single cancer, followed by association analysis for SubSETs (ASSET) to adjust the bias of overlapping controls. Gene-level analyses were conducted by SKAT-O, with multiple comparison adjustments by false discovery rate (FDR). Based on the significant genes indicated by SKATO analysis, pathways analysis was conducted using Gene Ontology (GO), the Kyoto Encyclopedia of Genes and Genomes (KEGG) and Reactome databases.ResultsMeta-analysis in three gastrointestinal (GI) cancers identified 13 novel susceptibility loci that reached genome-wide significance (PASSET< 5×10-8). SKAT-O analysis revealed EXOC6, LRP5L and MIR1263/LINC01324 to be significant genes shared by GI cancers (Padj<0.05, PFDR<0.05). Furthermore, GO pathway analysis identified significant enrichment of synaptic transmission and neuron development pathways shared by all three cancer types.ConclusionRare variants and the corresponding genes potentially contribute to shared susceptibility in different GI cancer types. The discovery of these novel variants and genes offers new insights for the carcinogenic mechanisms and missing heritability of GI cancers.