[Association between chronic lung diseases and the risk of lung cancer in UK Biobank: observational and Mendelian randomization analyses].

Objective: To investigate the association between chronic lung diseases and the risk of lung cancer. Methods: Using UK Biobank (UKB) survey data, 472 397 participants who had not previously been diagnosed with cancer and whose self-reported sex was consistent with their genetic sex were studied. Information on the prevalence of previous chronic lung diseases, general demographic characteristics and the prevalence of lung cancer was collected using baseline questionnaires and national health system data. The multivariate Cox proportional risk regression model was used to analyze the association between four previous chronic lung diseases (asthma, chronic obstructive pulmonary disease, idiopathic pulmonary fibrosis and interstitial pulmonary disease) and the risk of lung cancer. A total of 458 526 participants with genotype data in the observational study were selected as research objects, and the closely related and independent genetic loci with four chronic lung diseases were selected as instrumental variables, and the association between four chronic lung diseases and the risk of lung cancer was analyzed by Mendelian randomization (MR). The dose-response relationship between genetic risk score and the risk of lung cancer in different chronic lung diseases was evaluated using a restricted cubic spline function. Results: The age [M (Q1, Q3)] of the subjects was 57 (50, 63) years old, and there were 3 516 new cases of lung cancer (0.74%) during follow-up. The multivariate Cox proportional hazard regression model analysis showed that previous chronic obstructive pulmonary disease and idiopathic pulmonary fibrosis were associated with the risk of lung cancer, about 1.61 (1.49-1.75) and 2.61 (1.24-5.49), respectively. MR Studies showed that genetically predicted chronic obstructive pulmonary disease and idiopathic pulmonary fibrosis were associated with the risk of lung cancer, with HR (95%CI) of 1.10 (1.03-1.19) and 1.04 (1.01-1.08), respectively. The results of restricted cubic spline function analysis showed that the risk of lung cancer increased linearly with the increase of genetic risk scores for chronic obstructive pulmonary disease and idiopathic pulmonary fibrosis (P<0.05). Neither observational studies nor Mendelian randomization analysis found an association between previous asthma or interstitial lung disease and the risk of lung cancer (both P values>0.05). Conclusion: Chronic obstructive pulmonary disease and idiopathic pulmonary fibrosis are potential risk factors for lung cancer.目的： 探讨既往患有慢性肺部疾病与肺癌发病风险的关联情况。 方法： 采用英国生物银行（UKB）调查数据，以既往未被诊断为癌症且自报性别与遗传性别一致的472 397名参与者为研究对象。采用基线问卷和国家健康系统数据收集既往慢性肺部疾病患病情况、一般人口学特征以及肺癌患病情况等信息。采用多因素Cox比例风险回归模型分析哮喘、慢性阻塞性肺病、特发性肺纤维化、间质性肺病等4种既往慢性肺部疾病与肺癌发病风险的关联。以观察性研究中具有基因型数据的458 526名参与者为研究对象，选择与4种慢性肺部疾病密切相关且相互独立的遗传位点作为工具变量，采用孟德尔随机化（MR）分析4种慢性肺部疾病与肺癌发病风险的关联情况。采用限制性立方样条函数评估不同慢性肺部疾病GRS与肺癌发病风险的剂量反应关系。 结果： 研究对象年龄的M（Q1，Q3）为57（50，63）岁，随访期间新发肺癌3 516例（0.74%）。多因素Cox比例风险回归模型分析结果显示：既往患有慢性阻塞性肺病和特发性肺纤维化与肺癌发病风险存在关联，分别为1.61（1.49~1.75）和2.61（1.24~5.49）。MR研究结果显示：遗传预测的慢性阻塞性肺病和特发性肺纤维化与肺癌发病风险存在关联，HR（95%CI）分别为1.10（1.03~1.19）和1.04（1.01~1.08）。限制性立方样条函数分析结果显示：肺癌发病风险随慢性阻塞性肺病和特发性肺纤维化遗传风险评分增加而升高且呈线性关系（P<0.05）。观察性研究和孟德尔随机化分析均未发现既往患有哮喘或间质性肺病与肺癌发病风险存在关联（均P>0.05）。 结论： 既往患有慢性阻塞性肺病和特发性肺纤维化是肺癌发生的潜在危险因素。.