Department of Technology, Faculty of Health and Technology, University College Copenhagen, Copenhagen, Denmark.
Department of Clinical Biochemistry, Herlev and Gentofte Hospital, Copenhagen University Hospital, Herlev, Denmark.
Lung Cancer. 2023 Jul;181:107229. doi: 10.1016/j.lungcan.2023.107229. Epub 2023 May 3.
Screening reduces lung cancer mortality of high-risk populations. Currently proposed screening eligibility criteria only identify half of those individuals, who later develop lung cancer. This study aimed to develop and validate a sensitive and simple model for predicting 10-year lung cancer risk.
Using the 1991-94 examination of The Copenhagen City Heart Study in Denmark, 6,820 former or current smokers from the general population were followed for lung cancer within 10 years after examination. Logistic regression of baseline variables (age, sex, education, chronic obstructive pulmonary disease, family history of lung cancer, smoking status and cumulative smoking, secondhand smoking, occupational exposures to dust and fume, body mass index, lung function, plasma C-reactive protein, and AHRR(cg05575921) methylation) identified the best predictive model. The model was validated among 3,740 former or current smokers from the 2001-03 examination, also followed for 10 years. A simple risk chart was developed with Poisson regression.
Age, sex, education, smoking status, cumulative smoking, and AHRR(cg05575921) methylation identified 65 of 88 individuals who developed lung cancer in the validation cohort. The highest risk group, consisting of less educated men aged >65 with current smoking status and cumulative smoking >20 pack-years, had absolute 10-year risks varying from 4% to 16% by AHRR(cg05575921) methylation.
A simple risk chart including age, sex, education, smoking status, cumulative smoking, and AHRR(cg05575921) methylation, identifies individuals with 10-year lung cancer risk from below 1% to 16%. Including AHRR(cg05575921) methylation in the eligibility criteria for screening identifies smokers who would benefit the most from screening.
筛查可降低高危人群的肺癌死亡率。目前提出的筛查资格标准仅能识别出一半的人，而这些人后来患上了肺癌。本研究旨在开发和验证一种用于预测 10 年肺癌风险的敏感且简单的模型。
使用丹麦哥本哈根城市心脏研究 1991-1994 年的检查数据，对来自普通人群的 6820 名曾经或现在的吸烟者进行了 10 年内的肺癌随访。对基线变量（年龄、性别、教育程度、慢性阻塞性肺疾病、肺癌家族史、吸烟状态和累计吸烟量、二手烟、职业性粉尘和烟雾暴露、体重指数、肺功能、血浆 C 反应蛋白和 AHRR(cg05575921) 甲基化）进行逻辑回归，确定了最佳预测模型。该模型在来自 2001-2003 年检查的 3740 名曾经或现在的吸烟者中进行了验证，随访时间也为 10 年。使用泊松回归开发了一个简单的风险图表。
年龄、性别、教育程度、吸烟状态、累计吸烟量和 AHRR(cg05575921) 甲基化确定了验证队列中 88 名发展为肺癌的个体中的 65 名。最高风险组由受教育程度较低的男性组成，年龄>65 岁，当前吸烟状态，累计吸烟量>20 包年，AHRR(cg05575921) 甲基化的 10 年绝对风险从 4%到 16%不等。
一个简单的风险图表，包括年龄、性别、教育程度、吸烟状态、累计吸烟量和 AHRR(cg05575921) 甲基化，可识别出 10 年肺癌风险从低于 1%到 16%的个体。将 AHRR(cg05575921) 甲基化纳入筛查资格标准，可以识别出最受益于筛查的吸烟者。
