Predicting risk of lung cancer in different populations

Lung cancer is often diagnosed late and has a poor prognosis. Smoking is by far the most important cause of lung cancer, and in some countries current or former smokers are offered screening using low-dose computed tomography (CT). However, a substantial proportion of patients with lung cancer have never smoked. Some risk factors have been identified which are associated with higher risk in ever- and never smokers, and several models have been developed aiming to predict future risk of lung cancer. The aim of this project is to evaluate and compare these models, investigating their predictive performance in diverse populations, such as the UK Biobank and the China Kadoorie Biobank, where the distribution of lung cancer subtypes and smoking differ significantly. 

The extent to which predicting risk of lung cancer can be improved may be explored in particular by investigating the contribution of genetic and non-genetic risk factors, by smoking status, sex, and ethnicity. The proposed project may ultimately inform screening and prevention strategies for lung cancer in different populations.