Comparative effectiveness of low-dose CT lung cancer screening among high-risk non-smoking female subgroups and males in China.

Objective: Lung cancer screening guidelines abroad predominantly rely on smoking pack-years and age, Chinese guidelines uniquely incorporate non-smoking risk factors, though empirical validation remains limited. This study compared low-dose computed tomography (LDCT) screening participation and effectiveness between non-smoking females and males in real-world settings. Methods: This study used data from the Cancer Screening Program in Urban China (CanSPUC) in Zhejiang Province, spanning from April 2019 to December 2024. Eligible participants were aged 45 to 74 and were evaluated for a high risk for lung cancer. We stratified the population into three comparison groups: non-smoking females, non-smoking males and smoking males. Four key screening metrics were compared between groups: LDCT adherence rate, positive rate, lung cancer detection rate and false-positive rate. Multivariable robust (modified) Poisson regression was used to estimate adjusted risk ratios (aRRs) between groups. Results: A total of 88,797 participants aged 45 to 74 years were included in this study, comprising 43,604 non-smoking females and 45,193 males (6988 non-smokers). The overall LDCT participation rate was 61.2%, with a positive rate of 5.6%, lung cancer detection rate was 1.5%, and a false-positive rate of 95.2%. After adjustment for age, BMI, educational level, passive smoking, occupational exposure to hazardous substances, history of tuberculosis; history of COPD; family history of lung cancer, hypertension; hyperlipidemia and diabetes, smoking males showed significantly higher rates than non-smoking females in LDCT adherence (aRR=1.22, 95% CI 1.20-1.24), positive findings (aRR=1.12, 95% CI 1.04-1.20), and lung cancer detection (aRR=1.36, 95% CI 1.17-1.57). Conversely, non-smoking males demonstrated lower adherence (aRR=0.82, 95% CI 0.79-0.85), fewer positive findings (aRR=0.83, 95% CI 0.70-0.98), and reduced lung cancer detection (aRR=0.57, 95% CI 0.37-0.86) compared to non-smoking females. False-positive rates showed no significant differences across groups. During risk stratification of non-smoking females, both passive smoking (aRR=1.46, 95% CI 1.08-2.01) and hormonal abnormalities (aRR=1.39, 95% CI 1.11-1.75) were identified as independent risk factors for lung cancer, with no significant interaction observed between these factors. Multivariable analyses revealed that high-risk non-smoking females with both passive smoking and hormonal abnormalities showed no statistically significant differences in lung cancer detection rates (vs. all males: aRR=0.93, 95% CI 0.77-1.13; vs. smoking males: aRR=0.88, 95% CI 0.72-1.07) or false-positive rates (vs. all males: aRR=0.95, 95% CI 0.85-1.07; vs. smoking males: aRR=0.96, 95% CI 0.85-1.07) compared to male populations. Conclusions: In summary, current lung cancer screening criteria demonstrate limited applicability in female populations, as evidenced by significantly lower LDCT adherence, positive rates, and detection rates among non-smoking females compared to smoking males. Critically, we identified a well-defined subgroup of non-smoking females characterized by combined passive smoking and hormonal abnormalities. This subgroup exhibited screening outcomes that were not statistically significantly different from those observed in male populations. These findings support the development of more precise, risk-stratified screening criteria that better account for sex-specific differences in lung cancer risk and clinical presentation.