Abstract 6211: Personalized risk models based on lung cancer screening history: post hoc analysis of the National Lung Screening Trial (NLST)

Abstract Introduction: Our group previously revealed (PMID27509046) that lung cancer (LC) screening history is an independent prognostic factor whereby incidentally-detected lung cancer cancers with one or more preceding negative screens is associated with poorer survival outcomes compared to those preceded by only positive screens. As a follow-up study, we sought to develop personalized risk models that considered screening history for incidence cancers, interval LCs, and prevalence LCs. Methods: Using data from the CT-arm of the NLST, we conducted our independent case-control analyses to develop parsimonious risk models. Controls (n=26, 038) were those never diagnosed with LC through the end of follow-up. The four LC case groups were 270 prevalence LCs (diagnosed at baseline screen), 44 interval LCs (diagnosed following a negative screen at any time point prior to the next screening), 206 screen-detected LCs (SDLCs) that had a baseline positive screen, and 164 SDLCs that had a baseline negative screen. For each case-control analysis, univariable analyses identified statistically significant covariates from 47 variables including demographics, smoking history, alcohol use, medical history, family history of LC, and occupational exposures. Then, for each analysis, statistically significant covariates were included into a stepwise backwards selection approach to identify a model with the most informative covariates. Results: Age was the only covariate found in all four models; pack-years smoked, BMI, and smoking status were found in three of models. For prevalence LCs, the model (AUC=0.711) included age, pack-years smoked, BMI, smoking status, smoking onset age, personal history of cancer, family history of LC, frequency of alcohol consumption, and milling occupation. For interval LCs, the model (AUC=0.734) included age, smoking status, smoking onset age, cigar smoking, marital status, and asbestos occupation. For baseline positive SDLCs, the model (AUC=0.685) included age, pack-years smoked, BMI, emphysema, chemicals/plastics exposure, and milling occupation. For baseline negative SDLCs, the model (AUC=0.701) included age, pack-years smoked, BMI, smoking status, emphysema, sarcoidosis, and sandblasting occupation. Conclusions: We deployed a rigorous model building approach to identify unique personalized risk models that considered screening history and timing of LC diagnosis. Besides smoking and age, which are inclusion criteria for screening, these models identified other important risk factors including occupational exposures, medical history, alcohol use, personal history of cancer, and family history of LC. These models could be used to provide personalized risk assessment and screening management in screening. Including additional biomarkers (e.g., radiomics, circulating, etc.) could improve the statistical performance of the models. Citation Format: Sofia Haddan, Matthew Schabath, Asim Waqas, Ghulam Rasool. Personalized risk models based on lung cancer screening history: post hoc analysis of the National Lung Screening Trial (NLST) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 6211.