Associations between lncRNA-related polymorphisms and hepatocellular carcinoma risk: A two-stage case-control study

Abstract

Background and aim: Single-nucleotide polymorphisms (SNPs) in long non-coding RNAs (lncRNAs) are potential biomarkers for cancer risk, but their association with hepatocellular carcinoma (HCC) is unclear. We examined the association of lncRNA-related SNPs with HCC susceptibility and explored the optimal genetic models for SNPs.

Methods: Five candidate SNPs linked with digestive tumors were first genotyped in a screening population of 700 HCC and 2800 control cases. The association between each SNP and HCC risk was estimated by multivariate logistic regression adjusted by sex and age and recorded as odds ratio (OR) with 95% confidence interval. Significant associations were further tested in a validation population with 1140 HCC and 5115 control cases. Finally, the most appropriate genetic models for HCC-associated SNPs were identified using pairwise allele differences; the overall gene effects of each SNP were further evaluated based on optimal genetic models.

Results: Three candidate SNPs, rs7315438, rs6983267, and rs10795668, showed statistical connections with HCC risk in the discovery stage. Among these, rs7315438 remained steadily significant in the validation stage; rs7315438 and rs10795668 both reached statistical threshold in the combined analysis of both stages. SNP rs7315438 (TC vs TT/CC, OR = 1.410, P < 0.001) was associated with increased risk of HCC in a complete overdominant model, whereas rs10795668 (AG vs AA/GG, OR = 0.892, P = 0.035) exerted a protective effect on HCC risk in a complete overdominant model.

Conclusions: Long non-coding RNA-related SNPs rs7315438 and rs10795668 are potential biomarkers for HCC susceptibility, especially when evaluated based on their optimal genetic models.

Keywords: genetic model; hepatocellular carcinoma; lncRNA; single nucleotide polymorphism; susceptibility.