Prevention And Control Management And Association Analysis Of Lifestyle And Genetic Predisposition With Lung Cancer

ObjectivesLung cancer is one of the most common cancers in the world.At present,its mortality ranks first in all cancers,which brings heavy economic and disease burden to the world’s population.Lung cancer is attributed a variety of risk factors,but the effect of lifestyle factors and genetic factors on the risk of lung cancer needs to be further quantified to provide suggestions for the prevention and control of lung cancer.The development of medical information technology and big data technology provides new thinking and opportunities for the comprehensive understanding of lung cancer risk factors and the precise prevention of lung cancer.We focused on the association analysis of lifestyle and genetic predisposition with lung cancer,and used multi-source database to identify the key factors of lung cancer and carry out hierarchical classification.A large cohort study was conducted to investigate the association between healthy lifestyle index,genetic predisposition and lung cancer risk,and the interaction between healthy lifestyle index and genetic susceptibility to lung cancer.Based on genetic factors,lifestyle factors and other common risk factors of lung cancer,the risk prediction model of lung cancer was constructed and its prediction efficiency was evaluated,so as to provide a more accurate method for identifying high-risk groups or individuals of lung cancer.To compare and analyze the effect of lifestyle intervention on lung cancer prevention among different genetic risk groups,to build a precise prevention and control management model for high-risk groups of lung cancer,and to put forward policy recommendations for optimization of lung cancer prevention.MethodsBased on the multi-source databases of LD hub and UK Biobank（UKB）,genetic variance analyzer（GNOVA）was used to identify and classify the key risk factors of lung cancer through genetic correlation analysis.Cox proportional risk regression model were used to analyze the survey data and genotype data of the large-scale UKB cohort from 2006 to 2017 and evaluate the hazard ratio（HR）of lifestyle factors and genetic factors for the risk of lung cancer.Using genome-wide association analysis（GWAS）to identify lung cancer risk loci.Based on fast GWA model and genetic additive effect model,linear regression model was used to fit the relationship between single nucleoside polymorphism（SNP）and lung cancer.The polygenic risk score（PRS）for lung cancer through the polygenic prediction model（PRS-CS）was used to evaluate the genetic susceptibility of individual lung cancer.Based on smoking,drinking,diet,physical activity,and central obesity indicators,healthy lifestyle index scores were constructed to evaluate individual lifestyle and compliance with healthy lifestyle.The interaction between lifestyle variables and genetic risk was added to Cox proportional regression model to evaluate the joint effect and interaction of lifestyle and genetic risk on lung cancer risk.Based on Cox proportional risk regression model,lifestyle and genetic factors were combined to build a risk prediction model of lung cancer.Using counterfactual analysis and population attributable fraction（PAF）to evaluate the impact of lifestyle on the risk of lung cancer in different genetic risk profile and the burden of attributable risk factors.Population attributable risk was used to evaluate the preventive effect of lifestyle intervention.Through the scene simulation analysis,the prevention effect of lifestyle on different lung cancer genetic risk profile was estimated.ResultsBased on multi-source database,this study identified 199 key factors of lung cancer.Among the behavioral factors,smoking and lung cancer had the most significant correlation(the absolute value of genetic correlation coefficient|r|=0.48,P=3.4×10^-52).Physical activity,diet and drinking had significant correlation with lung cancer.Among the metabolic risk factors,waist circumference was the most significantly correlated with lung cancer(|r|=0.18,P=1.1×10^-14).The absolute value of correlation coefficient was as high as 0.91(P=9.5×10^-18).GWAS of lung cancer showed six previously unreported loci of lung cancer:rs145484481(OR:1.01,P=5.6×10^-9)in RP11-202K23.1,rs139804501(OR:1.01,P=5.6×10^-10)in CDH12,and HYKK rs11852372(OR:1.001,P=2.1×10^-10),rs116253640(OR:1.05,P=1.5×10^-8)in CEMP,rs145788203 in LINC00511(OR:1.01,P=7.9×10^-9),rs5360667(OR:1.02,P=9.2×10^-9)in CCBE1 gene.The liability heritability was 12%.The results of Cox proportional risk regression analysis adjusted by multiple confounding factors showed that non-smoking,regular physical activity,healthy diet,moderate drinking and non-center obesity could reduce 80%（HR:0.20,95%CI:0.18,0.22）,13%（HR:0.87,95%CI:0.78,0.96）,23%（HR:0.77,95%CI:0.67,0.86）,13%（95%CI:0.78,0.98）,21%（HR:0.79,95%CI:0.22）,respectively.Healthy lifestyle index score and lung cancer polygenic risk score were independently associated with lung cancer incidence.There was a linear relationship between the healthy lifestyle score and the risk of lung cancer（P for trend<0.001）.The risk of lung cancer in the individuals with the highest healthy lifestyle score was 88%lower than that in the individuals with the lowest healthy lifestyle score（HR:0.12,95%CI:0.08,0.17）.The risk of lung cancer in the population with high genetic risk to lung cancer was 1.98-fold as high as that in the population with low genetic susceptibility（HR:1.98,95%CI:1.66,2.37）.The risk of lung cancer was 8.03 times higher in high genetic risk combined with adverse lifestyle than in low genetic risk population（HR:9.03,95%CI:6.39,12.77）.No significant interaction between lifestyle and genetic risk was found（P>0.05）.Adding genetic information can improve the prediction accuracy of lung cancer risk（AUC=0.79）.Five lifestyle factors can explain 87.6%of lung cancer in high genetic risk group（PAF:87.6%,95%CI:76.5%,94.1%）,71.6%in mediate genetic risk group（PAF:71.6%,95%CI:58.3%,82.4%）,and 75.0%in low genetic risk group（PAF:75.0%,95%CI:38.6%,89.7%）.Adherence to healthy lifestyle can reduce 76.0%（95%CI:67.1%,82.6%）,70.6%（95%CI:63.7%,76.1%）,and 67.0%（95%CI:46.0%,79.9%）of lung cancer in low,intermediate,and high genetic risk group,respectively.The results of scenario simulation analysis showed that if individuals can follow a healthy lifestyle,the risk of lung cancer can be significantly reduced regardless of genetic risk.ConclusionBased on multi-source data,199 risk factors of lung cancer were identified by genetic correlation analysis,which further verified that lung cancer was the result of multiple factors.Non-smoking,moderate alcohol,regular physical activity,healthy diet and non-central obesity can significantly reduce the risk of lung cancer,which are independent protective factors of lung cancer.Genome wide association analysis of lung cancer identified six previously unreported lung cancer risk gene loci.The estimated heritability of lung cancer tendency was 12%,which supported the genetic susceptibility of lung cancer.There was no significant interaction between lifestyle and genetic susceptibility to lung cancer.Lifestyle and genetic susceptibility to lung cancer were independently related to the risk of lung cancer.At every genetic risk level,a healthy lifestyle is related to a similar degree of lung cancer risk reduction,emphasizing that everyone should adhere to a healthy lifestyle.To some extent,healthy lifestyle can offset the risk of lung cancer caused by high genetic susceptibility.High genetic risk individuals can get the biggest risk reduction from lifestyle intervention,and high genetic risk population can be regard as the primary target population of lifestyle intervention.Precise intervention and improvement of healthy lifestyle compliance have important public health significance and application value for the prevention of lung cancer.