Lung Cancer Risk Calculators

This page is made available for public non-profit use, that is, by individuals, universities, research institutes and hospitals and medical practices which are non-for-profit.

This site makes available downloading of a spreadsheet calculator which produces 6-year lung cancer risk (probability) estimates in ever-smokers according to the Tammemägi 2012 (PLCOm2012) lung cancer risk prediction model, the details of which were published in:

Tammemagi MC, Katki HA, Hocking WG, Church TR, Caporaso N, Kvale PA, Chaturvedi AK, Silvestri GA, Riley TL, Commins J, Berg CD. Selection Criteria for Lung-Cancer Screening. New England Journal of Medicine. 2013;368(8):728-36.

We have added a calculator for a lung cancer risk prediction model that is parallel to the PLCOm2012 in that it includes the same predictors and has 6 years of follow-up and was developed in Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial controls and was validated in PLCO intervention arm data, but it also includes never-smokers. This model is referred to as PLCOall2014. The correlation between risks estimated by PLCOm2012 and PLCOall2014 is >0.99.

This model is described in:

Tammemägi M, Church T, Hocking W, GA S, Kvale P, Riley T, Commins J, Berg C. Evaluation of the Lung Cancer Risks at which to Screen Ever- and Never-Smokers: Screening Rules Applied to the PLCO and NLST Cohorts. PLoS Medicine. 2014;11(12): e1001764. doi:10.1371/journal.pmed.1001764.

We have added a calculator for a lung cancer risk prediction model that is parallel to the PLCOm2012 but has been re-calculated (re-parameterized) excluding the race/ethnicity predictor. Several jurisdictions have requested this model as it was thought that the associations between race/ethnicity and lung cancer that are observed in the United States, where the modeling data come from, do not apply in their country, state or province. The predictive accuracy/discrimination as measured by the receiver operator characteristic area under the curves (AUC) of the PLCOm2012 and PLCOm2012noRace models in the Prostate Lung Colorectal and Ovarian Cancer Screening Trial control validation data are 0.797 (95% CI 0.782-0.811), and 0.796 (95% CI 0.780-0.811), respectively (p-value testing difference in AUC’s=0.43). Recommendations for use of the PLCOm2012noRace model are provided in a footnote in the calculator spreadsheet.

This model is described in:

Pasquinelli MM, Tammemägi CM, Kovitz KL, Durham ML, Deliu Z, Rygalski K, et al. Risk Prediction Model versus United States Preventive Services Task Force Lung Cancer Screening Eligibility Criteria – Reducing Race Disparities. Journal Thoracic Oncology. 2020; In press, release on 18 August 2020.

The PLCOm2012 risk prediction model uses baseline sociodemographic, medical and exposure data to predict lung cancer risk. Once an individual has entered into a lung cancer screening program and is not diagnosed with cancer as a consequence of a positive screen, the screening results provide valuable additional information that can improve risk prediction. The PLCO2012results model incorporates PLCOm2012 risk scores with lung cancer screening results to more accurately predict future lung cancer risk. This website provides a spreadsheet calculator that provides PLCOm2012 and PLCO2012results risk estimates given an individual’s predictor values and screening results. The reference for the development and validation of the PLCO2012results model is:

Tammemägi MC, ten Haaf K, Toumazis I, Kong CY, Han SS. Jeon J, Commins J, Riley T, Meza R. Development and validation of a multivariable lung cancer risk prediction model that includes low-dose computed tomography screening results – A secondary analysis of data from the National Lung Screening Trial. JAMA Netw Open. 2019;2(3):e190204. doi:10.1001/jamanetworkopen.2019.0204

Also, this site makes possible the downloading of two separate spreadsheet calculators which are based on a full (extended) model and a parsimonious model both of which predict the probability that a nodule detected on computed tomography (CT) screen is malignant. Details of these models are published in:

McWilliams A, Tammemagi M, Mayo J, Roberts H, Liu G, Soghrati K, Yasufuku K, Martel S, Laberge F, Gingras M, Atkar-Khattra S, Berg CD, Evans KG, Finley R, Yee J, English JC, Nasute P, Goffin J, Puksa S, Stewart L, Tsai S, Johnston MR, Manos D, Nicholoas G, Goss GD, Seely JM, Amjadi K, Tremblay A, Burrowes P, MacEachern P, Bhatia R, Tsao M-S, Lam S. Probability of cancer in pulmonary nodules detected on first screening computed tomography. New England Journal of Medicine. 2013;369;10:910-9.

Before downloading one or more spreadsheet calculators, please complete the following form. This information will allow us to track use of the models and enable us to contact you and provide you with information on updates, corrections or alternative models. Please also indicate in the boxes provided which calculators you are downloading.