Association Study Of Candidate Gene And MiRNA Gene Polymorphisms With Genetic Predisposition To Lung Cancer

BackgroundLung cancer is the most common cancer.Apart from the environmental exposure（such as tobacco use）,genetic factors also play an important role in the genesis of lung cancer,and the difference of genetic susceptibility between individuals showed a significant association with predisposition to lung caccer.Over the last two decades,candidate-gene association study is one of the major approaches to study the genetic susceptibility to lung cancer,and there are more than 1000 publications based on candidate-gene approach to study the associations between the genetic polymorphisms and lung cancer susceptibility.These candidate-gene association studies have found many genetic variants showed significant associations with lung cancer risk,but there are a number of conflicting results among these studies,which lead to the challenge to confirm the true susceptibility genetic variants for lung cancer risk.At present,there is a lack of comprehensive systematic review,meta-analysis,and assessment of epidemiological evidence for such amount of research evidence.Moreover,previous genetic association studies on lung cancer were mainly focus on the genetic variants in protein-coding genes,such as,DNA damage repair genes,metabolic enzymes genes,immune/inflammatory system related genes,and apoptosis pathway genes.Genetic variants in non-protein-coding genes have received little attention.However,there is a lot of non-protein-coding transcripts（termed as non-coding RNAs,nc RNAs）in the human genome.And,accumulating evidence showed that many nc RNAs,especially some micro RNAs（mi RNAs）,play an important role in the initiation and progression of a variety of tumors,including lung cancer.Recently,studies suggests that single nucleotide polymorphisms（SNPs）in mi RNAs are also associated with the risk of some tumors.However,only few studies have focused on the associations of SNPs in mi RNAs with lung cancer risk.It need to explore more SNPs in mi RNA associated with lung cancer susceptibility to provide more new clues for the studies on the etiology of lung cancer.Objective1.Based on the comprehensive systematic review,meta-analysis,and assessment of epidemiological evidence,we analysed all of the published candidate-gene association studies in lung cancer at present,and then identified,consolidated,and interpreted genetic associations of common variants with lung cancer.2.Based on the bioinformatics analyses,we selected several potentially functional SNPs in mi RNA,and then performed a case-control study to evaluate the associations of these selected potentially functional SNPs with lung cancer risk,which could lead to identify new SNPs in mi RNAs showed significant association with the susceptibility of lung cancer.Method1.We conducted a comprehensive literature search and review of previous published candidate-gene association studies in lung cancer up to November 1,2015.Meta-analyses were performed for the variants investigated by at least three independent studies,and subgroup analyses were also performed based on the ethnicity,pathological types,and smoking status.For the variants with significant association by the meta-analysis,the epidemiological credibility of cumulative evidence was assessed using the Venice criteria.We further conducted bioinformatics analysis to annotate the potential function of variants showed significant associations to lung cancer susceptibility with strong cumulative evidence.Additionally,for the non-significant associations revealed by all meta-analyses,we also evaluated the quality of evidence for the identified associations based on three aspects,including heterogeneity across the results of studies,potential bias,and statistical power.2.Based on bioinformatics analyses,we selected 13 potentially functional polymorphisms in different mi RNAs.Then,we genotyped these 13 SNPs by using SNPscanTM system in a case-control population,including 626 lung cancer cases and 736 controls whose age and gender were matched to cases.Multivariate logistic regression analysis was used to evaluate the association between lung cancer risk and each SNP,and subgroup analyses were also performed by pathological types and smoking status to explore the association between each SNP and lung cancer in different subgroup.ResultsPart Ⅰ1.After a systematic literature search and screen,a total of 1018 publications with 2910 genetic variants in 754 different genes or chromosomal locis were eligible for inclusion.2.Meta-analyses were performed on 246 genetic variants（in 138 different genes）from the reported variants.Based on the meta-analyses and evaluations of Venice criteria,we identified 22 variants（from 21 unique genes）showed significant associations with lung cancer susceptibility with strong credibility of cumulative epidemiological evidence,including APEX1 rs1130409,APEX1 rs1760944,ATM rs664677,AXIN2 rs2240308,CHRNA3 rs6495309,CHRNA5 rs16969968,CLPTM1 L rs402710,CXCR2 rs1126579,CYP1A1 rs4646903,CYP2E1 rs6413432,ERCC1 rs11615,ERCC2 rs13181,FGFR4 rs351855,HYKK rs931794,MIR146 A rs2910164,MIR196A2 rs11614913,OGG1 rs1052133,PON1 rs662,REV3 L rs462779,SOD2 rs4880,TERT rs2736098,and TP53 rs1042522.3.Bioinformatics analyses showed that,most of these 22 variants have evidence of an expression and/or functional impact on their host gene by either changing protein sequence,or modifying regulatory elements.4.Meta-analyses suggested that the other 150 variants（from 98 unique genes）showed non-significant associations with lung cancer risk.The assessment of evidence showed that meta-analyses investigated 7 genetic variants（ERCC1 rs16979802,ERCC1 rs2298881,ERCC1 rs735482,POLI rs3730668,PPARG rs1801282,PTGS2 rs20417,and TNF rs1799724）have strong quality of cumulative epidemiological evidence.Part Ⅱ1.Based on bioinformatics analyses,we selected 13 potentially functional SNPs in different mi RNAs（rs11597888 G/A,rs12803915 G/A,rs16867808 T/C,rs2292879 A/G,rs45530340 C/T,rs5997893 G/A,rs61747536 C/T,rs62085660 C/G,rs6464546 G/A,rs6717413 A/G,rs7247237 C/T,rs745666 G/C and rs999665 G/A）,and then all of these SNPs showed no-significant associations with lung cancer risk in general（P > 0.05）identified by association-analyses.2.In the subgroup analyses by pathological types,we found that the heterozygous genotype GA of rs11597888（G>A）had a significant association with an increased risk of lung adenocarcinoma(co-dominant model: OR_adjusted = 1.34,95%CI = 1.00-1.79,P = 0.046);the variant allele G of rs62085660（C>G）had a significant association with a decreased risk of lung squamous cell carcinoma(additive model: OR_adjusted = 0.79,95%CI = 0.62-1.00,P = 0.049;dominant model: OR_adjusted = 0.72,95%CI = 0.53-0.98,P = 0.035).3.In the subgroup analyses by smoking status,we found variant genotypes（GA+AA）of rs11597888（G>A）could significantly increase the risk of lung cancer in smokers(co-dominant model: OR_adjusted = 1.44,95%CI = 1.02-2.02,P = 0.037;dominant model: OR_adjusted = 1.42,95%CI = 1.03-1.96,P = 0.033);rs16867808（T>C）showed significant association with lung cancer risk based on multiple genetic models(additive model: OR_adjusted = 0.58,95%CI = 0.40-0.84,P = 0.004;co-dominant model: OR_adjusted = 0.55,95%CI = 0.35-0.85,P = 0.007;dominant model: OR_adjusted = 0.53,95%CI = 0.35-0.81,P = 0.004).4.Annotation by Haplo Reg data showed that these three significant variants（rs11597888 G/A,rs62085660 C/G,and rs16867808 T/C）identified from subgroup analyses were located within DNA regulatory elements（enhancer and/or promoter region）,DNase I hypersensitivity region,some transcription factors binding sites in multiple cells derived from multiple organs including lung cancer,and might alter regulatory motifs for genes.Haplo Reg data also showed that rs62085660 C/G and rs16867808 T/C have been identified as expression quantitative trait locu（e QTLs）of a variety of genes.Conclusion1.Based on the comprehensive systematic review,meta-analyses,and assessment of epidemiological evidence,we found that,in all of the reported candidate-gene association studies in lung cancer,approximately 9%（22/246）of genetic variants showed strong cumulative epidemiological evidence of an association with lung cancer risk at present.Meanwhile,previous functional studies and bioinformatics analyses suggested that most of these variants also have evidence of a functional impact on their host genes by either changing protein sequence or modifying regulatory elements.2.According to the association analysis based on case-control study design,we found that rs11597888 G/A（mi RNA AL391839.1）,rs62085660 C/G（mi RNA AC145343.1）,and rs16867808 T/C（mi RNA AL021918.2）were significantly associated with lung cancer risk.While these significant associations for these three variants can only be identified in a specific subgroup,including rs11597888（G>A）was significantly associated with lung adenocarcinoma risk,rs62085660（C>G）was significantly associated with lung squamous cell carcinoma,both rs11597888（G>A）and rs16867808（T>C）were significantly associated with lung cancer risk only in smoking population.The results of our study can provide the latest and comprehensive evidence of genetic risk on lung cancer.Moreover,the new SNPs in mi RNA were identified to be significantly associated with lung cancer susceptibility in our study,which can provide strong clues for further studies on the risk effects on lung cancer and theoretical references for personal prevention and treatment of lung cancer.