A new study suggests that polygenic risk scores (PRS) based on next-generation sequencing data may offer a more accurate method for predicting the risk of non–small cell lung cancer (NSCLC), the most common type of lung cancer. Researchers leveraged Optum’s Clinicogenomics database to analyze over 20 000 patients, aiming to develop a predictive model that integrates genetic variants and demographic data.

Traditional risk models for NSCLC often miss critical genetic factors, limiting their utility in early detection. By contrast, the PRS approach used in this study aggregates the relative contributions of multiple genetic mutations based on their allele frequency, offering a more nuanced risk assessment.

Key genetic drivers—including epidermal growth factor receptor (EGFR), mesenchymal epithelial transition factor receptor (MET), and anaplastic lymphoma kinase (ALK)—were identified as having the highest cumulative PRS scores in the NSCLC cohort, with a maximum score of 270. Notably, the EGFR L858R mutation was observed in 32% of patients with NSCLC, marking it as a strong genetic indicator of risk. For comparison, the highest PRS in a control group of patients with breast cancer was 150, centered around BReast CAncer (BRCA) gene mutations.

The model was validated using logistic regression and tested with standard performance metrics. Results showed a moderate association between PRS and NSCLC status, with an area under the curve (AUC) of 0.65.

Researchers concluded that incorporating PRS into clinical assessments may enhance genetic risk stratification and support earlier detection strategies. While further refinement is needed to improve predictive power, this approach holds promise for tailoring screening and treatment plans to individuals based on their genetic profile.

Reference

Mohanty P, Kukreja I, Kumari R, et al. Predicting non-small cell lung cancer risk: utilizing a polygenic approach with us cohort data. Presented at: AMCP 2025; March 31-April 3; Houston, TX; Abstract C9.