| Title: | Unified genetic risk score for prostate cancer enables improved risk stratification for clinical decision-making. |
| Journal: | Journal of Medical Genetics |
| Published: | 29 Jul 2026 |
| Pubmed: | https://pubmed.ncbi.nlm.nih.gov/42527158/ |
| DOI: | https://doi.org/10.1136/jmg-2026-111746 |
Publication 21278
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
Abstract
BACKGROUND: Current clinical approaches to inherited prostate cancer (PCa) risk rely on binary classification of pathogenic variant (PV) carrier status without accounting for gene-specific heterogeneity or polygenic risk. We developed an integrated genetic risk model reflecting a continuum of inherited susceptibility.</p>
METHODS: In the UK Biobank (UKB; n=218 484), we evaluated associations of PVs in 11 clinically recommended genes and a polygenic risk score (PRS) with incident PCa using Cox models. A continuum model (GenProb-PCa) incorporating gene-specific PVs and PRS was developed and compared with binary PV-based models. Model performance was assessed using discrimination, calibration and continuous net reclassification index (cNRI). External validation was performed in a health system cohort, the Genomic Health Initiative (GHI; n=6590).</p>
RESULTS: Five genes (ATM, BRCA2, CHEK2, HOXB13, MSH2) and the PRS were independently associated with PCa risk (all p<0.001) in UKB. Compared with binary models, GenProb-PCa demonstrated superior discrimination (C-index 0.69 vs 0.52; p<0.001), with significant improvement in reclassification (cNRI 0.58; p<0.001). Findings were validated in GHI using the UKB-derived coefficients (C-index 0.64 vs 0.54, p<0.001; cNRI 0.31, p<0.001). Compared with binary PV-based models, 20% of non-PV carriers were reclassified to higher risk groups and 60% of carriers to lower risk groups. The model identified individuals at markedly elevated lifetime risk, with cumulative incidence exceeding 20% by age 75 among the top 8% of the distribution.</p>
CONCLUSIONS: An integrated continuum genetic risk model, GenProb-PCa, improves PCa risk stratification beyond binary approaches by capturing heterogeneity in inherited risk. This framework may enable more precise risk-based screening strategies.</p>
10 Authors
- Zhuqing Shi
- Ashley J Mulford
- Jun Wei
- Huy Tran
- Annabelle Ashworth
- Siqun Lilly Zheng
- Jim Lu
- Alan R Sanders
- Brian Helfand
- Jianfeng Xu
1 Application
| Application ID | Title |
|---|---|
| 50295 | Validity and performance of inherited risk assessment for common diseases using polygenic risk score, family history, and high-penetrance genes |
Enabling scientific discoveries that improve human health