The models used to identify those persons at highest risk for lung cancer leave significant room for improvement, according to researchers at the National Cancer Institute Division of Cancer Epidemiology & Genetics, Biostatistics Branch. In a report published in the Annals of Internal Medicine, they report that only 4 models (the Bach model, PLCOM2012, LCRAT, and LCDRAT) are highly accurate at predicting risk in ever-smokers for screening.
Hormuzd Katki, PhD, and colleagues report that ending the epidemic of smoking-related illness requires continued progress in smoking cessation and prevention. However, they have found that effectively and efficiently targeting lung cancer screening to persons at highest risk has been less than optimal. This study revealed that 4 lung cancer risk models appear to perform best in selecting US ever-smokers for screening; however, the models should be further refined to improve their performance in certain subpopulations.
A variety of lung cancer risk models are currently in use. Dr Katki’s team sought to measure the performance of each model in selecting ever-smokers for screening. They searched MEDLINE for studies published between January 1, 2000, and December 31, 2016, using the terms lung-cancer, risk, prediction, and model. Only models that provided a cumulative risk estimate for primary lung cancer or lung cancer mortality for at least 1 time point were included.
The team compared the US screening populations selected by 9 lung cancer risk models: the Bach model; the Spitz model; the Liverpool Lung Project (LLP) model; the LLP Incidence Risk Model (LLPi); the Hoggart model; the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial Model 2012 (PLCOM2012); the Pittsburgh Predictor; the Lung Cancer Risk Assessment Tool (LCRAT); and the Lung Cancer Death Risk Assessment Tool (LCDRAT). For this investigation, the 9 prominent lung cancer risk models were applied to a representative sample of the US population, and the similarities and differences among the ever-smokers selected for CT lung cancer screening by each model were analyzed.