Multi-center Genome-wide Association Study Of Renal Cell Carcinoma Identified Susceptibility Loci In Han Chinese Population

Object:Renal cell carcinoma（RCC）is the second most common malignant genitourinary carcinoma with an increasing incidence since 1970s.Etiological studies suggested that genetic factors play an important role in the carcinogenesis of RCC.Therefore,we conducted a large-scale multi-center genome-wide association study（GWAS）of RCC in Han Chinese populations to identify genetic susceptibility loci.Functional annotations and experiments were performed to explore the underlying mechanisms between susceptibility gene and the development of RCC.In our study,evidence showed that dyslipidemia might influence the RCC risk,thus we used Mendelian randomization（MR）to investigate the causal associations between lipid-lowering agents and RCC risk.Methods:Our study was based on case-control study design,including 3,002 cases and4,823 controls recruited from Tianjin,Shanghai and Guangzhou.The discovery stage was consisted of 4 studies,including（1）Tianjin-I GWAS,totally 1,141 cases and 1,334controls;（2）Tianjin-II GWAS,totally 385 cases and 765 controls;（3）Shanghai GWAS,totally 462 cases and 1,010 controls;（4）Guangzhou GWAS,totally 1,014 cases and1,714 controls.Genotype data were processed using standard quality control（QC）process in each study,including individual-level and SNP-level QC.Imputation was performed using 1000 Genomes Project Phase 3 data as reference panel in SHAPEIT and IMPUTE2.Per-allele odds ratio（OR）and 95%confidence interval（CI）were calculated using logistic regression model adjusting for sex,age and significant eigenvectors（P<0.05）in each study.Fixed effect inverse-variance weighted meta-analysis was performed,and Cochran’s Q tests were used to assess the heterogeneity across all studies.For 47 independent SNPs with P<5×10^-5,validation was performed in external datasets.Validation stage was consisted of 3 studies,including（1）Tianjin validation,totally 158 cases and 215 controls;（2）Guangzhou validation,totally 848cases and 1,527 controls.e QTL analysis was used to identify susceptibility genes whose expressions were influenced by susceptibility loci.Regulome DB was used to perform functional annotations and mechanisms prediction.Electrophoretic mobility shift assay（EMSA）was used to verify the capacity of genetic variants binding to the transcription factors.We further investigated the causal influence of circulating lipid levels and commonly prescribed lipid-lowering agents on RCC risk.Instrument variables were constructed using GWAS summary statistics of circulating lipid traits in European ancestry,SNP-outcome associations were obtained from RCC GWAS for European ancestry to estimate the causal effect.Sensitivity analyses were performed using a series of pleiotropy-robust methods.Results:We have identified one RCC susceptibility locus in Han Chinese population,this locus was located at 12q24.31 region and the most significant SNP was rs10846748(OR=1.26,95%CI:1.16-1.36,P=3.86×10^-10).Stratification analyses based on histological types and genders did not suggest significant differences for this SNP.e QTL analysis was performed using TCGA KIRC collection（N=527）,and identified that rs10846753 risk allele G was significantly associated with a higher expression level of SCARB1(P=4.15×10^-9).Gene-level analysis and transcriptome-wide association study（TWAS）further confirmed that SCARB1 was a susceptibility gene for RCC.The estimated heritability explained by common variation was 14.2%.EMSA confirmed that rs77151276 and rs4765621 at SCARB1 locus were specifically binding to the nuclear protein and might be influenced by HIF-α.In drug-target MR analyses,there was no significant association between genetically proxied HMG-Co A reductase inhibition and RCC risk（OR=1.47[95%CI,0.38-5.70],P=0.280）.However,in women,genetically proxied CETP inhibition was associated with a higher risk of RCC（OR=4.31[95%CI,1.19-15.6]）,whereas in men,genetically proxied PCSK9 inhibition was associated with a higher risk of RCC（OR=2.20[95%CI,1.24-3.89]）.Conclusions:Our study identified rs10846748 as the first susceptibility locus of RCC in Han Chinese population,and this SNP might promote lipid accumulation in tumor cells by regulating the expression level of SCARB1.MR analyses suggested that novel lipid-modifying drugs,PCSK9 and CETP inhibitors,were significantly associated with RCC risk in a sex-specific manner.