Tobacco Smoking and Lung Cancer Risk After Negative Baseline Low-Dose Computed Tomography Findings

Lung cancer (LC) remains the leading cause of cancer-related mortality worldwide, with tobacco smoking being the primary risk factor. However, the long-term LC risk among individuals with negative low-dose computed tomography (LDCT) findings and the role of tobacco smoking in risk stratification remain poorly understood, limiting evidence-based guidance for subsequent screening intervals.

To evaluate the association of tobacco smoking with long-term LC risk after a negative baseline LDCT finding and to inform optimized screening strategies.

This population-based, prospective cohort study was conducted under the Cancer Screening Program in Urban China. Individuals aged 40 to 74 years with negative baseline LDCT findings (October 1, 2013, to December 31, 2021) were included, with follow-up until December 2023. All participants were monitored for LC incidence.

Self-reported smoking status, pack-years, and time since quitting.

The primary outcome was LC incidence, analyzed using Kaplan-Meier methods and multivariable Cox proportional hazards regression models. The association between smoking exposure and LC risk was assessed, with time-stratified analyses and dose-response associations.

Among 30 565 participants (14 761 never smokers and 15 804 smokers; mean [SD] age, 57.1 [7.7] years; 15 693 [51.3%] female), 76 LC cases occurred during 139 011.51 person-years (crude incidence rate, 54.67 of 100 000 person-years). Smokers had higher LC risk than never smokers (adjusted hazard ratio [AHR], 2.73; 95% CI, 1.49-5.01), driven by those with a smoking history of 20 pack-years or more (eg, ≥30 pack-years: AHR, 3.22; 95% CI, 1.85-5.58). There was no elevated risk at 2 years (AHR, 2.07; 95% CI, 0.91-4.69), but risk was significantly increased at 3 years (AHR, 2.54; 95% CI, 1.19-5.41) and onward. A nonlinear dose-response association was found between pack-years and LC risk, with risk surpassing clinically relevant thresholds at approximately 20 pack-years (eg, 20 to <30 pack-years: AHR, 2.48; 95% CI, 1.14-5.40). Females exhibited higher susceptibility than males at comparable exposure (≥30 pack-years: AHR, 5.78 [95% CI,1.87-17.83] for females vs 1.36 [95% CI, 0.18-10.39] for males). Significant risk was seen in those aged 50 to 54 years (≥30 pack-years) and 55 to 74 years (≥20 pack-years). Short-term cessation (<15 years) was not significantly associated with reduced LC risk.

In this cohort study, smokers with negative baseline LDCT findings exhibited a substantially elevated long-term LC risk, which became significant only after 2 years after screening. These findings suggest support for extending the initial screening interval and implementing personalized long-term monitoring based on smoking history.