筛查人群中肺癌发生与死亡风险与试验参与者的比较
研究比较了北卡罗来纳肺癌筛查登记(NCLSR)和BRFSS人群与NLST试验参与者的肺癌发生及死亡风险。结果显示,NCLSR和BRFSS人群年龄更大、合并症更多、肺癌风险评分更高,且未筛查时5年肺癌发生和死亡风险均高于NLST参与者。提示需在一般人群中监测肺癌筛查的获益与危害。
要点
- 一般筛查人群的5年肺癌风险中位数高于NLST试验参与者。
- 一般筛查人群年龄更大、COPD和既往癌症史比例更高。
- 尽管吸烟暴露相似,一般人群未筛查时肺癌死亡风险更高。
核心结论
- NCLSR和BRFSS人群与NLST参与者吸烟暴露相似,但肺癌危险因素、风险评分及未筛查时肺癌死亡风险更高。
研究设计与样本
- 基于NCLSR、2022年BRFSS肺癌筛查模块和NLST数据的比较分析,使用公开宏估计未筛查时5年肺癌发生和死亡概率,并用Cohen's D比较。
主要发现
- BRFSS和NCLSR人群比NLST参与者年龄更大,COPD和既往癌症史比例更高,肺癌发生和死亡风险更高。
- NCLSR人群当前吸烟比例更高,但各组中位吸烟暴露相似。
- NLST参与者5年肺癌风险中位数为21.2/1000(IQR 13.0-35.2/1000),2013年BRFSS为46.1/1000(Cohen's D=0.67515),2021年BRFSS为34.5/1000(Cohen's D=0.47813),2013年NCLSR为36.5/1000(Cohen's D=0.07748),2021年NCLSR为32.3/1000(Cohen's D=0.48289)。
- 未筛查时5年肺癌死亡概率中位数NLST参与者最低,为12.5/1000,BRFSS和NCLSR人群更高。
临床怎么用
- 在一般人群中开展肺癌筛查时,应关注其与试验人群的差异,权衡筛查获益与潜在危害。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND: Trial participants typically are healthier than the general population. Differences in underlying characteristics between the population undergoing lung cancer screening (LCS) and LCS trial participants may alter the benefits of LCS. RESEARCH QUESTION: Does the risk of lung cancer developing and death resulting differ between trial participants and the general population? STUDY DESIGN AND METHODS: Using data from the (1) the North Carolina Lung Screening Registry (NCLSR), (2) the 2022 Centers for Disease Control and Prevention's Behavioral Risk Factor Surveillance System (BRFSS) Lung Cancer Screening Module, and (3) the National Lung Screening Trial (NLST), we estimated the 5-year probability of lung cancer developing and death resulting if patients were not screened using publicly available macros. Using Cohen's D, we compared these metrics in NCLSR and BRFSS populations with those in NLST participants. RESULTS: BRFSS and NCLSR populations were older, had higher rates of COPD and prior history of cancer, and higher risk of lung cancer developing and death resulting compared with NLST participants. Although NCLSR populations were more likely to be currently smoking, median smoking exposure was similar among all groups. Median 5-year lung cancer risk was 21.2 per 1,000 NLST participants (interquartile range,13.0-35.2 per 1,000 NLST participants) compared with 46.1 per 1,000 2013 BRFSS participants (Cohen's D = 0.67515), 34.5 per 1,000 2021 BRFSS participants (Cohen's D = 0.47813), 36.5 per 1,000 2013 NCLSR participants (Cohen's D = 0.07748), and 32.3 per 1000 2021 NCLSR participants (Cohen's D = 0.48289). Median 5-year probability of dying of lung cancer if not screened showed similar patterns, with lowest rates among NLST participants at 12.5 deaths per 1,000 participants vs BRFSS and NCLSR populations. INTERPRETATION: Our results showed that NCLSR and BRFSS populations have similar smoking exposure as NLST participants, but higher prevalence of lung cancer risk factors, lung cancer risk scores, and risk of death resulting from lung cancer if not screened. Monitoring LCS in the general population is crucial because compared with trial participants, individuals in the general population are older and have more comorbidities, raising concerns about potential LCS harms. However, they have a higher risk of dying of lung cancer if not screened.
引用
- Rivera MP, Benefield T, Durham DD, et al. Comparison of Underlying Risk of Developing and Dying From Lung Cancer in Screened Populations. Chest. 2025. PMID: 40912293. doi:10.1016/j.chest.2025.08.015
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。