Lung cancer is the leading cause of cancer mortality. While low-dose CT (LDCT) reduces mortality, screening rates remain low (~10%) due to access barriers, unlike breast or colorectal screens (~75%) [NIH, 2024]. Circulating biomarkers offer a promising solution as a pre-screening triage tool: identifying high-risk individuals from a simple blood draw who should then be prioritized for LDCT. Both cfDNA genomic alterations and the host’s systemic functional response can boost sensitivity for a screening assay. The UK Biobank is critical for characterizing such multiomic signatures via a retrospective longitudinal design. Though healthy at recruitment, hundreds of participants developed incident lung cancer, providing pre-diagnostic samples to assess biomarker dynamics in early-stage disease. UKB’s deep phenotyping enables differentiation of true cancer signals from aging or benign pulmonary comorbidities (e.g., COPD) that often confound liquid biopsies. Integrating WGS with functional readouts supports a comprehensive genotype-to-phenotype discovery approach.
Research Questions:
Are specific multiomic features differentially expressed in pre-diagnostic samples >12 months prior to diagnosis, and what are their longitudinal dynamics?
What are the baseline biological profiles of “pulmonary biological noise” (COPD, CHIP) and aging that might confound cancer-associated signals?
Characterize interactions between host immune proteins and germline genomic profiles in high-risk, cancer-free individuals.
Objectives:
Integrate UKB WGS variants and molecular phenotypes for incident lung cancer cases and matched controls to create a unified dataset for association analysis.
Perform high-dimensional statistical analyses to identify candidate multiomic features that distinguish pre-diagnostic cases from controls.
Quantify the biological variance of these candidate features in controls and partition variance between healthy aging and documented pulmonary comorbidities.