低剂量CT筛查后确诊肺癌患者的生存分析

OBJECTIVE: To evaluate survival and mortality in individuals diagnosed with lung cancer after lung cancer screening (LCS), and the association of mortality with patient and cancer characteristics.

METHODS: Retrospective study at an academic medical center including patients receiving LCS between January 1, 2015, and June 30, 2020, and diagnosed with lung cancer through June 30, 2021, were included. Patients with lung cancer were followed in the electronic health record through May 30, 2024, to determine vital status and patient and cancer characteristics. Survival probabilities were estimated using Kaplan-Meier methodology. Mortality rates were calculated as the number of deaths divided by total person-years. Univariable and multivariable Cox proportional hazards regression models were used to evaluate associations between mortality and patient and cancer characteristics.

RESULTS: In all, 141 individuals with lung cancer were included. Mean age (SD) at diagnosis was 68.5 (6.3) years. Most individuals (111 of 141, 78.7%) had screen-detected lung cancer. Stage I cancer was most common (70 of 141, 49.6%). Estimated 5-year survival (95% confidence interval) was 53.4% (45.4-62.9%). Mortality at median follow-up times of 3.1 and 5.2 years was 337 deaths and 317 deaths per 100,000 person-years, respectively. At multivariable analysis, higher cancer stage and non-screen-detected lung cancer status were associated with higher mortality (P < .001 for any stage >1; hazard ratio of screen-detected versus non-screen-detected cancer [95% confidence interval]: 3.16 [1.77-5.66]).

DISCUSSION: Though screen-detected cancers were associated with lower mortality, mortality in screened patients with lung cancer was higher than the National Lung Screening Trial, showing outcomes in our single-center clinical LCS may differ from prior LCS research trials.