AI-driven integration of genomic and exposome data for cancer risk: the combined risk score (CRS)
Genetic risk scores work best for identifying at-risk populations for prevention trials rather than individual clinical decisions, requiring careful validation across diverse populations.
This review critically evaluates polygenic risk scores and combined genomic-exposome risk scores (CRS) for cancer, finding robust relative risk stratification but modest discrimination metrics and limited external validation across ancestry groups. The authors argue CRS are currently best used for population-level risk enrichment and prevention trial design rather than clinical decision support, requiring explicit calibration and equity governance.
What the study was
- Study design
- Integrated systematic narrative review with structured quantitative summaries
- Population
- General population cancer risk stratification context; pan-cancer
- Category
- Genomics/Precision Medicine
- Maturity
- Exploratory
- Journal
- Human Genomics
Why it surfaced
Balanced review noting both promise and limitations of CRS; useful for context-setting but no new primary data.
A plain-language summary of published research — not medical advice. Talk to a clinician about your own care.