Abstract
BACKGROUND: low-dose computed tomography (LDCT) screening strategies for lung cancer are primarily targeted at high-risk groups based on smoking history and age, leading to over-screening of low-risk individuals while delaying timely detection in high-risk groups. This study aimed to develop a precision screening framework integrating plasma proteomics, polygenic risk score (PRS), and the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial 2012 model (PLCOm2012) to enable dynamic and optimized screening strategies. METHODS: A multicenter, multimodal study was conducted across two prospective cohorts. Plasma protein profiling identified 16 biomarkers, which were combined with PRS and PLCOm2012 to develop a prediction model. Protein score, PRS, and PLCOm2012 score were used to develop a combined risk score (CRS). Risk advancement period (RAP) analysis quantified personalized screening initiation ages based on 10-year cumulative risk thresholds. RESULTS: The CRS demonstrated superior predictive accuracy, with 3-, 5-, and 10-year AUCs of 0.85 (95% CI: 0.80–0.93), 0.88 (95% CI: 0.83–0.92), and 0.85 (95% CI: 0.81–0.87), respectively. High-risk individuals yielded HR of 3.18 (95% CI: 2.34–4.31) with thresholds reached 11.71 (RAP: -11.71, 95% CI: –15.58 - −6.16) years earlier than those at medium-risk, while low-risk individuals could delay screening by 11.78 (RAP: 11.78, 95% CI: 6.95 - 15.08) years. Risk stratification revealed distinct differences in cumulative lung cancer incidence across groups. Personalized screening initiation ages suggested that high-risk individuals begin screening before age 40, while low-risk individuals could postpone screening until age 55 or later. CONCLUSION: This study developed a precision lung cancer screening framework by integrating plasma proteomics, genetic profiling, and clinical data. This approach enhanced early detection in high-risk individuals while safely delaying screening for low-risk groups. TRIAL REGISTRATION: Clinical trial number: NCT06422637. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12967-025-07468-1.