Early-onset colorectal cancer (EOCRC, diagnosed <50 years) incidence has risen approximately 51% since the mid-1990s in high-income countries. Current risk models are limited by a focus on metabolic factors, failing to account for life-course exposures, circadian health, and gene-environment interactions. This project aims to define an integrated "risk architecture" for EOCRC to inform precision prevention.
Research Questions:
1. Do life-course exposures (early-life antibiotic use, ultra-processed food consumption) and objective circadian metrics (sleep regularity, chronotype) confer differential risk for EOCRC compared to late-onset disease?
2. Do genetic factors (Polygenic Risk Scores, FUT2 genotype) modify the risk of these modifiable exposures?
3. Can a multi-domain model integrating lifestyle, genetics, and proteomic biomarkers improve absolute-risk estimation?
Objectives:
1. Characterize associations between early-life antibiotic use and ultra-processed food intake with incident EOCRC.
2. Assess circadian health-operationalized through wrist accelerometry-derived sleep regularity and chronotype-as an independent risk determinant.
3. Evaluate Gene-Environment interactions (e.g., FUT2 genotype x Antibiotics; PRS x Ultra-processed food) to identify high-risk subgroups.
4. Develop and internally validate an absolute-risk prediction model to facilitate risk-stratified screening recommendations.
Scientific Rationale:
Emerging evidence suggests EOCRC etiology involves distinct pathways related to early-life gut dysbiosis and circadian disruption. The UK Biobank is uniquely suited for this investigation due to its prospective design, genomic data, and objective accelerometry subset. We will utilize Cox proportional hazards and Fine-Gray competing risk models to quantify associations, and leverage Olink proteomics to explore inflammatory mediators linking circadian disruption to carcinogenesis.