1193P - Ethical lung cancer screening selection? Comparison of the European 4 - In The Long Run (4ITLR) criteria with the 2021 USPSTF (US), and HUNT lung cancer risk model in a large Norwegian prospective population-based study

Background

Exclusion of high-risk individuals may jeopardize an ethical lung cancer screening. Fixed criteria with high pack-years and age have excluded large populations that are at true high risk, e.g. the NLST criteria excluded 74% of those that would eventually develop lung cancer (Pinsky and Berg 2012 J Med Screen; Markaki et al 2018 EBioMedicine), and the NELSON criteria would exclude 62.6% of lung cancers from screening in Norway (Nguyen et al 2024 JTO Clin Res Rep). Here we aimed to compare the European Horizon 2020 4 - In The Long Run (4ITLR) criteria, with 2021 USPSTF criteria and the HUNT Lung Cancer Model (HUNT LCM).

Methods

4ITLR criteria was age 60 - 80 years old, >=40 pack-years, currently smoking or quit smoking <10 years. 2021 USPSTF criteria was age 50 - 80 years old, >20 pack-years currently smoking or quit smoking <15 years. These two criteria and the HUNT LCM were tested in 10 Norwegian prospective population-based studies of ever-smokers, age 24-94, n=44 831 (the CONOR study) with mean follow-up of 12.5 years.

Results

In 6 years 222 lung cancers were diagnosed. The 4ITLR selected 796 / 44831 (1.8%) of the th population and predicted 33/222 (14.86% sensitivity) of the lung cancers. By selecting the same number of individuals with the HUNT-LCM this predicted 43/222 (19,4% sensitivity) of the lung cancers. The 2021 USPSTF selected (5327/44831)11.9% of the population and predicted 132/222 (59.46% sensitivity) of the lung cancers. Selecting the same number of individuals by HUNT LCM, 148/222 (66.7% sensitivity) of the lung cancers were predicted (p=0.0253).

Conclusions

The use of 4ITLR criteria results in the exclusion of 85.3% of future lung cancers in a high-risk population. It is ethically dubious to screen only a minor fraction of high-risk individuals. Using a risk model as the HUNT LCM including as many as the 2021 USPSTF criterial, would include 66.7% of cases. The use of the HUNT Lung Cancer Model will significantly increase the rate of lung cancer diagnosis.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

Levanger Hospital, Nord Trøndelag Hospital Trust, Levanger, Norway.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.