Validation of a Multi-ancestry Polygenic Risk Score and Age-Specific Risks of Prostate Cancer: A Meta-analysis Within Diverse Populations

Fei Chen, Burcu F. Darst, Ravi Madduri, Alex A Rodriguez, Xin Sheng, Christopher T. Rentsch, Caroline Andrews, Wei Tang, Adam S. Kibel, Anna Plym, Kelly Cho, Mohamed Jalloh, Sérigne Maguèye Gueye, Lamine Niang, Olufemi J. Ogunbiyi, Olufemi Popoola, Akindele O. Adebiyi, Oseremen I. Aisuodionoe-Shadrach, Hafees O. Ajibola, Mustapha A. Jamda, Olabode P. Oluwole, Maxwell Nwegbu, Ben Adusei, Sunny Mante, Afua O.d. Abrahams, James E. Mensah, Andrew A. Adjei, Halimatou Diop‐Ndiaye, Joseph Lachance, Timothy R. Rebbeck, Stefan Ambs, J. Michael Gaziano, Amy C. Justice, David V. Conti, Christopher A. HaimanmedRxiv 2022 · DOI

TL;DR

Researchers have validated a multi-ancestry polygenic risk score (PRS) that effectively predicts prostate cancer risk across European, African, and Hispanic populations. The study reveals that men in the highest genetic risk group reach critical risk thresholds nearly a decade earlier than those with average risk, suggesting that genetic testing could significantly personalize the timing and frequency of clinical screenings.

What Problem Does This Paper Address?

Prostate cancer is a leading cause of death, but it doesn't affect all groups equally. In the U.S., men of African ancestry face significantly higher incidence and mortality rates. While clinical guidelines currently use age and family history to determine screening schedules, these are blunt instruments.

Most existing genetic risk models—known as Polygenic Risk Scores (PRS)—were developed using data primarily from men of European descent, often leading to poor performance when applied to diverse populations. This paper seeks to bridge that gap by validating a "multi-ancestry" PRS, ensuring that the benefits of genomic medicine are accessible and accurate for men of various backgrounds.

Methods

The authors validated a previously developed PRS composed of 269 genetic variants associated with prostate cancer.

  1. PRS Construction: The score is calculated by summing the "weighted" dosages of these 269 variants. These weights were derived from a massive prior trans-ancestry genome-wide association study (GWAS).
  2. Statistical Validation: They used logistic regression to calculate the Odds Ratio (OR)—a measure of how much more likely someone in a high-score category is to have cancer compared to someone in the middle (40-60%) of the distribution.
  3. Meta-analysis: They combined results from several independent studies using a fixed-effects inverse-variance-weighted approach to ensure the findings were robust across different cohorts.
  4. Absolute Risk Estimation: By combining the genetic data with population-level incidence (from SEER) and mortality rates (from the CDC), they estimated the "absolute risk"—the actual percentage chance a man has of developing prostate cancer by a certain age.

Datasets

The study is impressive in its scale, involving 31,925 cases and 490,507 controls. Key datasets included:

  • The VA Million Veteran Program (MVP): The primary validation cohort, providing a large sample of European, African, and Hispanic men.
  • UK Biobank: A massive UK-based cohort used for European ancestry replication.
  • MADCaP & CA UG: Specialized networks focusing on men of African descent in the U.S. and Sub-Saharan Africa.
  • MGB Biobank & NCI-MD: Additional independent clinical and case-control studies.

Key Insights and Ideas

  • The "Age Gradient": One of the most striking findings is that genetic risk is a much stronger predictor for younger men. For men under 55 in the top 1% of genetic risk, the odds of having prostate cancer were 17 times higher than average. This suggests that genetics plays a dominant role in early-onset cases, while environmental factors or accumulated "wear and tear" might dilute the genetic signal in older age.
  • A 10-Year Head Start: Men in the top 10% of the PRS reached a 5% absolute risk of prostate cancer roughly 10 to 12 years earlier than men with average genetic risk. This provides a concrete window for clinical intervention: a high-risk man might need to start screening at 45 or 50, while a low-risk man could potentially wait much longer.
  • Ancestry Portability: While the PRS performed best in European ancestry men, it remained a powerful and valid predictor for African and Hispanic men. This is a significant step forward in reducing the "genomic divide" in precision medicine.
  • The Aggression Paradox: Interestingly, the PRS was great at predicting if a man would get cancer, but it didn't reliably distinguish between "aggressive" (fast-growing) and "non-aggressive" (slow-growing) types.

Key Results

  • Odds Ratios (Top 10% vs. Average):
    • European Ancestry: 3.8-fold increased risk.
  • Hispanic Ancestry: 3.2-fold increased risk.
  • African Ancestry: 2.8-fold increased risk.
  • Discrimination: Adding the PRS to a standard model (age + ancestry components) improved the Area Under the Curve (AUC)—a measure of predictive accuracy—by an average of 0.136 across populations.
  • Absolute Risk Thresholds: In the top 1% of risk, men reached the 5% absolute risk threshold by age 51-53 across all three ancestry groups.

Limitations and Future Work

  • The Performance Gap: The PRS still performs slightly less effectively in men of African ancestry compared to those of European ancestry. The authors emphasize the urgent need for even larger GWAS studies focusing on African populations to refine these scores further.
  • Clinical Utility: While the statistical validation is clear, we don't yet have a clinical trial showing that using this PRS actually improves patient outcomes or reduces overdiagnosis. The authors call for clinical studies to determine how to integrate these scores into standard care.
  • Aggressiveness Prediction: Because the PRS doesn't distinguish between aggressive and indolent disease, there is a risk that high-risk screening could lead to the over-treatment of slow-growing tumors that might never have caused harm. Future research into "aggression-specific" genetic markers is needed.

Summary updated

Abstract

Abstract Background We recently developed a multi-ancestry polygenic risk score (PRS) that effectively stratifies prostate cancer risk across populations. In this study, we validated the performance of the PRS in the multi-ancestry Million Veteran Program (MVP) and additional independent studies. Methods Within each ancestry population, the association of PRS with prostate cancer risk was evaluated separately in each case-control study and then combined in a fixed-effects inverse-variance-weighted meta-analysis. We further assessed the effect modification by age and estimated the age-specific absolute risk of prostate cancer for each ancestry population. Results The PRS was evaluated in 31,925 cases and 490,507 controls, including men from European (22,049 cases, 414,249 controls), African (8,794 cases, 55,657 controls), and Hispanic (1,082 cases, 20,601 controls) populations. Comparing men in the top decile (90-100% of the PRS) to the average 40-60% PRS category, the prostate cancer OR was increased 3.8-fold in European ancestry men (95% CI=3.62-3.96), 2.8-fold in African ancestry men (95% CI=2.59-3.03), and 3.2-fold in Hispanic men (95% CI=2.64-3.92). The PRS did not discriminate risk of aggressive versus non-aggressive prostate cancer. However, the OR diminished with advancing age (European ancestry men in the top decile: ≤55 years, OR=7.11; 55-60 years, OR=4.26; >70 years, OR=2.79). Men in the top PRS decile reached 5% absolute prostate cancer risk ∼10 years younger than men in the 40-60% PRS category. Conclusions Our findings validate the multi-ancestry PRS as an effective prostate cancer risk stratification tool across populations. A clinical study of PRS is warranted to determine if the PRS could be used for risk-stratified screening and early detection. Funding This work was supported by the National Cancer Institute at the National Institutes of Health (grant numbers U19 CA214253 to C.A.H., U01 CA257328 to C.A.H., U19 CA148537 to C.A.H., R01 CA165862 to C.A.H., K99 CA246063 to B.F.D, and T32CA229110 to F.C), the Prostate Cancer Foundation (grants 21YOUN11 to B.F.D. and 20CHAS03 to C.A.H.), the Achievement Rewards for College Scientists Foundation Los Angeles Founder Chapter to B.F.D, and the Million Veteran Program-MVP017. This research has been conducted using the UK Biobank Resource under application number 42195. This research is based on data from the Million Veteran Program, Office of Research and Development, and the Veterans Health Administration. This publication does not represent the views of the Department of Veteran Affairs or the United States Government.
BibTeX
@misc{chen2022validation,
  author = {Fei Chen and Burcu F. Darst and Ravi Madduri and Alex A Rodriguez and Xin Sheng and Christopher T. Rentsch and Caroline Andrews and Wei Tang and Adam S. Kibel and Anna Plym and Kelly Cho and Mohamed Jalloh and Sérigne Maguèye Gueye and Lamine Niang and Olufemi J. Ogunbiyi and Olufemi Popoola and Akindele O. Adebiyi and Oseremen I. Aisuodionoe-Shadrach and Hafees O. Ajibola and Mustapha A. Jamda and Olabode P. Oluwole and Maxwell Nwegbu and Ben Adusei and Sunny Mante and Afua O.D. Abrahams and James E. Mensah and Andrew A. Adjei and Halimatou Diop‐Ndiaye and Joseph Lachance and Timothy R. Rebbeck and Stefan Ambs and J. Michael Gaziano and Amy C. Justice and David V. Conti and Christopher A. Haiman},
  title = {{Validation of a Multi-ancestry Polygenic Risk Score and Age-Specific Risks of Prostate Cancer: A Meta-analysis Within Diverse Populations}},
  year = {2022},
  doi = {10.1101/2022.05.03.22274606},
  howpublished = {medRxiv},
}
Figure