David O Wilson1, Joel Weissfeld2. 1. Division of Pulmonary, Allergy and Critical Care Medicine, Department of Medicine, University of Pittsburgh, United States. Electronic address: wilsondo@upmc.edu. 2. Department of Epidemiology, University of Pittsburgh, United States.
Abstract
BACKGROUND: A user-friendly method for assessing lung cancer risk may help standardize selection of current and former smokers for screening. We evaluated a simple 4-factor model, the Pittsburgh Predictor, against two well-known, but more complicated models for predicting lung cancer risk. METHODS: Trained against outcomes observed in the National Lung Screening Trial (NLST), the Pittsburgh Predictor used four risk factors, duration of smoking, smoking status, smoking intensity, and age, to predict 6-year lung cancer incidence. After calibrating the Bach and PLCOM2012 models to outcomes observed in the low-dose computed tomography arm of the NLST, we compared model calibration, discrimination, and clinical usefulness (net benefit) in the NLST and Pittsburgh Lung Screening Study (PLuSS) populations. RESULTS: The Pittsburgh Predictor, Bach, and PLCOM2012 represented risk equally well, except for the tendency of PLCOM2012 to overestimate risk in subjects at highest risk. Relative to the Pittsburgh Predictor, Bach and PLCOM2012 increased the area under the receiver operator characteristic curve by 0.007-0.009 and 0.012-0.021 units, respectively, depending on study population. Across a clinically relevant span of 6-year lung cancer risk thresholds (0.01-0.05), Bach and PLCOM2012 increased net benefit by less than 0.1% in NLST and 0.3% in PLuSS. CONCLUSION: In exchange for a small reduction in prediction accuracy, a simpler lung cancer risk prediction model may facilitate standardized procedures for advising and selecting patients with respect to lung cancer screening.
BACKGROUND: A user-friendly method for assessing lung cancer risk may help standardize selection of current and former smokers for screening. We evaluated a simple 4-factor model, the Pittsburgh Predictor, against two well-known, but more complicated models for predicting lung cancer risk. METHODS: Trained against outcomes observed in the National Lung Screening Trial (NLST), the Pittsburgh Predictor used four risk factors, duration of smoking, smoking status, smoking intensity, and age, to predict 6-year lung cancer incidence. After calibrating the Bach and PLCOM2012 models to outcomes observed in the low-dose computed tomography arm of the NLST, we compared model calibration, discrimination, and clinical usefulness (net benefit) in the NLST and Pittsburgh Lung Screening Study (PLuSS) populations. RESULTS: The Pittsburgh Predictor, Bach, and PLCOM2012 represented risk equally well, except for the tendency of PLCOM2012 to overestimate risk in subjects at highest risk. Relative to the Pittsburgh Predictor, Bach and PLCOM2012 increased the area under the receiver operator characteristic curve by 0.007-0.009 and 0.012-0.021 units, respectively, depending on study population. Across a clinically relevant span of 6-year lung cancer risk thresholds (0.01-0.05), Bach and PLCOM2012 increased net benefit by less than 0.1% in NLST and 0.3% in PLuSS. CONCLUSION: In exchange for a small reduction in prediction accuracy, a simpler lung cancer risk prediction model may facilitate standardized procedures for advising and selecting patients with respect to lung cancer screening.
Authors: Margaret R Spitz; Waun Ki Hong; Christopher I Amos; Xifeng Wu; Matthew B Schabath; Qiong Dong; Sanjay Shete; Carol J Etzel Journal: J Natl Cancer Inst Date: 2007-05-02 Impact factor: 13.506
Authors: Martin C Tammemägi; Hormuzd A Katki; William G Hocking; Timothy R Church; Neil Caporaso; Paul A Kvale; Anil K Chaturvedi; Gerard A Silvestri; Tom L Riley; John Commins; Christine D Berg Journal: N Engl J Med Date: 2013-02-21 Impact factor: 91.245
Authors: Margaret R Spitz; Carol J Etzel; Qiong Dong; Christopher I Amos; Qingyi Wei; Xifeng Wu; Waun Ki Hong Journal: Cancer Prev Res (Phila) Date: 2008-09
Authors: David O Wilson; Joel L Weissfeld; Carl R Fuhrman; Stephen N Fisher; Paula Balogh; Rodney J Landreneau; James D Luketich; Jill M Siegfried Journal: Am J Respir Crit Care Med Date: 2008-07-17 Impact factor: 21.405
Authors: Timothy R Church; William C Black; Denise R Aberle; Christine D Berg; Kathy L Clingan; Fenghai Duan; Richard M Fagerstrom; Ilana F Gareen; David S Gierada; Gordon C Jones; Irene Mahon; Pamela M Marcus; JoRean D Sicks; Amanda Jain; Sarah Baum Journal: N Engl J Med Date: 2013-05-23 Impact factor: 91.245
Authors: Hormuzd A Katki; Stephanie A Kovalchik; Lucia C Petito; Li C Cheung; Eric Jacobs; Ahmedin Jemal; Christine D Berg; Anil K Chaturvedi Journal: Ann Intern Med Date: 2018-05-15 Impact factor: 25.391
Authors: Giulia Veronesi; David R Baldwin; Claudia I Henschke; Simone Ghislandi; Sergio Iavicoli; Matthijs Oudkerk; Harry J De Koning; Joseph Shemesh; John K Field; Javier J Zulueta; Denis Horgan; Lucia Fiestas Navarrete; Maurizio Valentino Infante; Pierluigi Novellis; Rachael L Murray; Nir Peled; Cristiano Rampinelli; Gaetano Rocco; Witold Rzyman; Giorgio Vittorio Scagliotti; Martin C Tammemagi; Luca Bertolaccini; Natthaya Triphuridet; Rowena Yip; Alexia Rossi; Suresh Senan; Giuseppe Ferrante; Kate Brain; Carlijn van der Aalst; Lorenzo Bonomo; Dario Consonni; Jan P Van Meerbeeck; Patrick Maisonneuve; Silvia Novello; Anand Devaraj; Zaigham Saghir; Giuseppe Pelosi Journal: Cancers (Basel) Date: 2020-06-24 Impact factor: 6.639
Authors: Vineet K Raghu; Wei Zhao; Jiantao Pu; Joseph K Leader; Renwei Wang; James Herman; Jian-Min Yuan; Panayiotis V Benos; David O Wilson Journal: Thorax Date: 2019-03-12 Impact factor: 9.102