Performance of lung cancer risk prediction models in different racial and ethnic groups in the United States: results from the Lung Cancer Cohort Consortium

A new study led by scientists from the International Agency for Research on Cancer (IARC) compared the performance of lung cancer risk prediction models against the criteria currently used to determine eligibility for lung cancer screening in the USA. The study found that all the included risk-based strategies achieved better average estimated screening efficiency than the current criteria, and also reduced racial and ethnic differences in efficiency. The study was published in Annals of Internal Medicine.

The researchers assessed 16 lung cancer risk models in 641 830 Asian, Hispanic, non-Hispanic Black, and non-Hispanic White adults aged 50–80 years with a history of smoking in the USA. They examined how well the models could predict lung cancer risk and their impact on screening eligibility, sensitivity, and efficiency across the different racial and ethnic groups. When each model selected the same number of individuals for screening as the United States Preventive Services Task Force criteria (USPSTF-2021), all models performed more efficiently overall and reduced racial and ethnic differences in screening efficiency compared with USPSTF-2021. However, the existing risk models – which were developed using data from mainly non-Hispanic Whites – tended to underestimate risk in non-Hispanic Blacks and had reduced predictive ability in Asians and non-Hispanic Blacks.

These findings suggest that using risk prediction models for lung cancer screening eligibility could yield a better balance of benefits and harms, with smaller differences across racial and ethnic groups. However, further research is needed to optimize prediction models for the diverse population in the USA.

Feng X, Guida F, Guernon A, Alcala K, Aldrich MC, Arslan AA, et al.
Performance of lung cancer risk prediction models in different racial and ethnic groups in the United States: results from the Lung Cancer Cohort Consortium
Ann Intern Med. Published online 30 June 2026;
https://doi.org/10.7326/ANNALS-25-03816

Read the article