Low-dose spiral computed tomography (LDCT) screening has demonstrated a 20% reduction in mortality from lung cancer in a randomized controlled trial recently. However, challenges, such as high false positive rates of LDCT, remain for the early detection of lung cancer. Risk stratification of populations can help individualized diagnosis and treatment of lung nodule positive people. The construction of lung cancer risk prediction model is the basis for risk stratification. Even though several lung cancer risk models, such as the PLCO-m2012, were built in the past decade, most of the models were based on macro epidemiological risk factors while neglected the contribution of genetic factors.
In our recent study, we constructed a polygenetic risk score (PRS) specific to Chinese population based on the genome-wide association studies (GWAS) of lung cancer in Chinese population. Then, we included both macro epidemiological risk factors as well as the PRS to build a lung cancer risk prediction model. Interestingly, we found a significant improvement of the model when the PRS was included in the lung cancer risk prediction model.
However, taken into account the genetic differences between ethnicity, the effectiveness of PRS was not clear in Caucasian population by now. In this application, we want to build a PRS applicable to Caucasian population and then constructed a comprehensive risk prediction model of lung cancer for Caucasian population. Finally, we want to compare the difference of the PRS as well as the models between Chinese population and Caucasian population. All efforts are expected to be finished in the next 12 months.