. 2025 Oct 1;16(1):8632.

doi: 10.1038/s41467-025-63546-4.

Polygenic risk score for type 2 diabetes shows context-dependent effects across populations

Boya Guo¹, Yanwei Cai², Daeeun Kim^{3

4}, Roelof A J Smit⁵, Zhe Wang^{5

6}, Kruthika R Iyer^{7

8

9}, Austin T Hilliard⁹, Jeffrey Haessler², Ran Tao^{10

11}, K Alaine Broadaway⁴, Yujie Wang³, Nikita Pozdeyev¹², Frederik F Stæger¹³, Chaojie Yang¹⁴, Brett Vanderwerff¹⁵, Amit D Patki¹⁶, Lauren Stalbow⁵, Meng Lin¹², Nicholas Rafaels¹², Jonathan Shortt¹², Laura Wiley¹², Maggie Stanislawski¹², Jack Pattee¹⁷, Lea Davis¹⁸, Peter S Straub^{11

19}, Megan M Shuey^{11

19}, Nancy J Cox^{11

19}, Nanette R Lee²⁰, Marit E Jørgensen^{21

22}, Peter Bjerregaard²², Christina Larsen²², Torben Hansen²³, Ida Moltke¹³, James B Meigs²⁴, Daniel O Stram²⁵, Xianyong Yin^{26

27

28}, Xiang Zhou^{27

28}, Kyong-Mi Chang^{29

30}, Shoa L Clarke^{8

9

31}, Rodrigo Guarischi-Sousa^{8

9}, Joanna Lankester^{8

9}, Philip S Tsao^{8

9}, Steven Buyske³², Mariaelisa Graff³, Laura M Raffield⁴, Quan Sun³³, Lynne R Wilkens³⁴, Christopher S Carlson², Charles B Easton^{35

36}, Simin Liu³⁷, JoAnn E Manson²⁴, Loïc L Marchand³⁴, Christopher A Haiman²⁵, Karen L Mohlke⁴, Penny Gordon-Larsen³⁸, Anders Albrechtsen¹³, Michael Boehnke¹⁵, Stephen S Rich¹⁴, Ani Manichaikul¹⁴, Jerome I Rotter³⁹, Noha A Yousri^{40

41}, Ryan M Irvin¹⁶; biobank at the Colorado Center for Personalized Medicine (CCPM); VA Million Veteran Program (MVP); Chris Gignoux¹², Kari E North³, Ruth J F Loos⁵, Themistocles L Assimes^{8

9}, Ulrike Peters^{2

42}, Charles Kooperberg², Sridharan Raghavan^{12

43}, Heather M Highland³, Burcu F Darst⁴⁴

Collaborators, Affiliations

Collaborators

biobank at the Colorado Center for Personalized Medicine (CCPM):
Heather D Anderson, Christina L Aquilante, Kelsey Arbogast, Christopher H Arehart, Ian M Brooks, Tonya M Brunetti, Judith Brutus-Lestin, Elizabeth E Burke, Emily M Casteel, Joanne B Cole, Curtis R Coughlin 2nd, Kristy Crooks, Jacob Crawford, Erin Culver, Michelle N Edelmann, Matthew J Fisher, Alan W Franklin, Teresa C Frye, Hunter George, Chris Gignoux, Elizabeth K Gilliland, Casey S Greene, Brooke Hawkes, Emily Hearst, Audrey E Hendricks, Randi K Johnson, Colleen G Julian, Dave Kao, Iain Konigsberg, Lisa Ku, Elizabeth L Kudron, Rashawnda Lacy, Ethan M Lange, Yee Ming Lee, Joe A Lesny, Jan T Lowery, Luciana B Vargas, Betzaida L Maldonado, Darcy Marceau, James L Martin, Brianna L Gates, David Mayer, Nicole L McDaniel, Andrew Monte, Ethan Moore, Ann Nadrash, Alaa Radwan, Nick Rafaels, Neda Rasouli, Elise L Shalowitz, Hoda Sherif, Johnathan Shortt, Adrian M Stewart, Kristen J Sutton, Carolyn T Swartz, Anna Tanaka, Matthew R G Taylor, Candace Teague, Emily B Todd, Katy E Trinkley, Laura K Wiley
VA Million Veteran Program (MVP):
Sumitra Muralidhar, Jennifer Moser, Jennifer E Deen, J Michael Gaziano, Elizabeth Hauser, Amy Kilbourne, Michael Matheny, Dave Oslin, Deepak Voora, Jessica V Brewer, Mary T Brophy, Kelly Cho, Lori Churby, Scott L DuVall, Saiju Pyarajan, Robert Ringer, Luis E Selva, Shahpoor Alex Shayan, Brady Stephens, Stacey B Whitbourne, Adriana Hung, Henry Kranzler

Affiliations

¹ Public Health Sciences, Fred Hutchinson Cancer Center, Seattle, WA, USA. bguo@fredhutch.org.
² Public Health Sciences, Fred Hutchinson Cancer Center, Seattle, WA, USA.
³ Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA.
⁴ Department of Genetics, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA.
⁵ The Charles Bronfman Institute for Personalized Medicine, Icahn School of Medicine at Mount Sinai, New York, NY, USA.
⁶ Department of Epidemiology, University of Alabama at Birmingham, Birmingham, AL, USA.
⁷ Data Science and Biotechnology, Gladstone Institutes, San Francisco, CA, USA.
⁸ Department of Medicine, Division of Cardiovascular Medicine, Stanford University School of Medicine, Stanford, CA, USA.
⁹ VA Palo Alto Health Care System, Palo Alto, CA, USA.
¹⁰ Department of Biostatistics, Vanderbilt University Medical Center, Nashville, TN, USA.
¹¹ Vanderbilt Genetic Institute, Vanderbilt University of Medical Center, Nashville, TN, USA.
¹² Colorado Center for Personalized Medicine, University of Colorado Anschutz Medical Campus, Aurora, CO, USA.
¹³ Department of Biology, Section for Computational and RNA Biology, University of Copenhagen, Copenhagen, Denmark.
¹⁴ Department of Genome Sciences, University of Virginia, Charlottesville, VA, USA.
¹⁵ Department of Biostatistics and Center for Statistical Genetics, University of Michigan, Ann Arbor, MI, USA.
¹⁶ Department of Epidemiology, School of Public Health, University of Alabama at Birmingham, Birmingham, AL, USA.
¹⁷ Department of Biostatistics and Informatics, Colorado School of Public Health, University of Colorado Anschutz Medical Campus, Aurora, CO, USA.
¹⁸ Department of Medicine, Division of Data-Driven and Digital Medicine, Icahn School of Medicine at Mount Sinai, New York, NY, USA.
¹⁹ Department of Medicine, Division of Genetic Medicine, Vanderbilt University of Medical Center, Nashville, TN, USA.
²⁰ USC-Office of Population Studies Foundation, Inc., University of San Carlos, Cebu City, Philippines.
²¹ Steno Diabetes Center Greenland, Nuuk, Greenland.
²² Centre for Public Health in Greenland, National Institute of Public Health, University of Southern Denmark, Copenhagen, Denmark.
²³ Novo Nordisk Foundation Center for Basic Metabolic Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark.
²⁴ Department of Medicine, Mass General Brigham Healthcare System, Harvard Medical School, Boston, MA, USA.
²⁵ Department of Preventive Medicine, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA.
²⁶ Department of Epidemiology, School of Public Health, Nanjing Medical University, Nanjing, China.
²⁷ Department of Biostatistics, University of Michigan, Ann Arbor, MI, USA.
²⁸ Center for Statistical Genetics, University of Michigan, Ann Arbor, MI, USA.
²⁹ Corporal Michael J. Crescenz VA Medical Center, Philadelphia, PA, USA.
³⁰ Department of Medicine, Division of Gastroenterology and Hepatology, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA.
³¹ Department of Medicine, Stanford Prevention Research Center, Stanford University School of Medicine, Stanford, CA, USA.
³² Department of Genetics, Rutgers University, Piscataway, NJ, USA.
³³ Department of Biostatistics, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA.
³⁴ Epidemiology Program, University of Hawaii Cancer Center, Honolulu, HI, USA.
³⁵ Department of Epidemiology, School of Public Health, Brown University, Pawtucket, RI, USA.
³⁶ Department of Family Medicine, Warren Alpert Medical School, Brown University, Pawtucket, RI, USA.
³⁷ Department of Epidemiology, School of Public Health, Brown University, Providence, RI, USA.
³⁸ Department of Nutrition, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA.
³⁹ Department of Pediatrics, The Institute for Translational Genomics and Population Sciences, The Lundquist Institute for Biomedical Innovation at Harbor-UCLA Medical Center, Torrance, CA, USA.
⁴⁰ College of Health and Life Sciences, Hamad Bin Khalifa University, Doha, Qatar.
⁴¹ Computer and Systems Engineering, Alexandria University, Alexandria, Egypt.
⁴² Department of Epidemiology, School of Public Health, University of Washington, Seattle, WA, USA.
⁴³ Department of Medicine, Division of General Internal Medicine, University of Colorado School of Medicine, Denver, CO, USA.
⁴⁴ Public Health Sciences, Fred Hutchinson Cancer Center, Seattle, WA, USA. bdarst@fredhutch.org.

PMID: 41034193
PMCID: PMC12488948
DOI: 10.1038/s41467-025-63546-4

Polygenic risk score for type 2 diabetes shows context-dependent effects across populations

Boya Guo et al. Nat Commun. 2025.

. 2025 Oct 1;16(1):8632.

doi: 10.1038/s41467-025-63546-4.

Authors

Collaborators

biobank at the Colorado Center for Personalized Medicine (CCPM):
Heather D Anderson, Christina L Aquilante, Kelsey Arbogast, Christopher H Arehart, Ian M Brooks, Tonya M Brunetti, Judith Brutus-Lestin, Elizabeth E Burke, Emily M Casteel, Joanne B Cole, Curtis R Coughlin 2nd, Kristy Crooks, Jacob Crawford, Erin Culver, Michelle N Edelmann, Matthew J Fisher, Alan W Franklin, Teresa C Frye, Hunter George, Chris Gignoux, Elizabeth K Gilliland, Casey S Greene, Brooke Hawkes, Emily Hearst, Audrey E Hendricks, Randi K Johnson, Colleen G Julian, Dave Kao, Iain Konigsberg, Lisa Ku, Elizabeth L Kudron, Rashawnda Lacy, Ethan M Lange, Yee Ming Lee, Joe A Lesny, Jan T Lowery, Luciana B Vargas, Betzaida L Maldonado, Darcy Marceau, James L Martin, Brianna L Gates, David Mayer, Nicole L McDaniel, Andrew Monte, Ethan Moore, Ann Nadrash, Alaa Radwan, Nick Rafaels, Neda Rasouli, Elise L Shalowitz, Hoda Sherif, Johnathan Shortt, Adrian M Stewart, Kristen J Sutton, Carolyn T Swartz, Anna Tanaka, Matthew R G Taylor, Candace Teague, Emily B Todd, Katy E Trinkley, Laura K Wiley
VA Million Veteran Program (MVP):
Sumitra Muralidhar, Jennifer Moser, Jennifer E Deen, J Michael Gaziano, Elizabeth Hauser, Amy Kilbourne, Michael Matheny, Dave Oslin, Deepak Voora, Jessica V Brewer, Mary T Brophy, Kelly Cho, Lori Churby, Scott L DuVall, Saiju Pyarajan, Robert Ringer, Luis E Selva, Shahpoor Alex Shayan, Brady Stephens, Stacey B Whitbourne, Adriana Hung, Henry Kranzler

Affiliations

¹ Public Health Sciences, Fred Hutchinson Cancer Center, Seattle, WA, USA. bguo@fredhutch.org.
² Public Health Sciences, Fred Hutchinson Cancer Center, Seattle, WA, USA.
³ Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA.
⁴ Department of Genetics, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA.
⁵ The Charles Bronfman Institute for Personalized Medicine, Icahn School of Medicine at Mount Sinai, New York, NY, USA.
⁶ Department of Epidemiology, University of Alabama at Birmingham, Birmingham, AL, USA.
⁷ Data Science and Biotechnology, Gladstone Institutes, San Francisco, CA, USA.
⁸ Department of Medicine, Division of Cardiovascular Medicine, Stanford University School of Medicine, Stanford, CA, USA.
⁹ VA Palo Alto Health Care System, Palo Alto, CA, USA.
¹⁰ Department of Biostatistics, Vanderbilt University Medical Center, Nashville, TN, USA.
¹¹ Vanderbilt Genetic Institute, Vanderbilt University of Medical Center, Nashville, TN, USA.
¹² Colorado Center for Personalized Medicine, University of Colorado Anschutz Medical Campus, Aurora, CO, USA.
¹³ Department of Biology, Section for Computational and RNA Biology, University of Copenhagen, Copenhagen, Denmark.
¹⁴ Department of Genome Sciences, University of Virginia, Charlottesville, VA, USA.
¹⁵ Department of Biostatistics and Center for Statistical Genetics, University of Michigan, Ann Arbor, MI, USA.
¹⁶ Department of Epidemiology, School of Public Health, University of Alabama at Birmingham, Birmingham, AL, USA.
¹⁷ Department of Biostatistics and Informatics, Colorado School of Public Health, University of Colorado Anschutz Medical Campus, Aurora, CO, USA.
¹⁸ Department of Medicine, Division of Data-Driven and Digital Medicine, Icahn School of Medicine at Mount Sinai, New York, NY, USA.
¹⁹ Department of Medicine, Division of Genetic Medicine, Vanderbilt University of Medical Center, Nashville, TN, USA.
²⁰ USC-Office of Population Studies Foundation, Inc., University of San Carlos, Cebu City, Philippines.
²¹ Steno Diabetes Center Greenland, Nuuk, Greenland.
²² Centre for Public Health in Greenland, National Institute of Public Health, University of Southern Denmark, Copenhagen, Denmark.
²³ Novo Nordisk Foundation Center for Basic Metabolic Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark.
²⁴ Department of Medicine, Mass General Brigham Healthcare System, Harvard Medical School, Boston, MA, USA.
²⁵ Department of Preventive Medicine, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA.
²⁶ Department of Epidemiology, School of Public Health, Nanjing Medical University, Nanjing, China.
²⁷ Department of Biostatistics, University of Michigan, Ann Arbor, MI, USA.
²⁸ Center for Statistical Genetics, University of Michigan, Ann Arbor, MI, USA.
²⁹ Corporal Michael J. Crescenz VA Medical Center, Philadelphia, PA, USA.
³⁰ Department of Medicine, Division of Gastroenterology and Hepatology, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA.
³¹ Department of Medicine, Stanford Prevention Research Center, Stanford University School of Medicine, Stanford, CA, USA.
³² Department of Genetics, Rutgers University, Piscataway, NJ, USA.
³³ Department of Biostatistics, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA.
³⁴ Epidemiology Program, University of Hawaii Cancer Center, Honolulu, HI, USA.
³⁵ Department of Epidemiology, School of Public Health, Brown University, Pawtucket, RI, USA.
³⁶ Department of Family Medicine, Warren Alpert Medical School, Brown University, Pawtucket, RI, USA.
³⁷ Department of Epidemiology, School of Public Health, Brown University, Providence, RI, USA.
³⁸ Department of Nutrition, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA.
³⁹ Department of Pediatrics, The Institute for Translational Genomics and Population Sciences, The Lundquist Institute for Biomedical Innovation at Harbor-UCLA Medical Center, Torrance, CA, USA.
⁴⁰ College of Health and Life Sciences, Hamad Bin Khalifa University, Doha, Qatar.
⁴¹ Computer and Systems Engineering, Alexandria University, Alexandria, Egypt.
⁴² Department of Epidemiology, School of Public Health, University of Washington, Seattle, WA, USA.
⁴³ Department of Medicine, Division of General Internal Medicine, University of Colorado School of Medicine, Denver, CO, USA.
⁴⁴ Public Health Sciences, Fred Hutchinson Cancer Center, Seattle, WA, USA. bdarst@fredhutch.org.

PMID: 41034193
PMCID: PMC12488948
DOI: 10.1038/s41467-025-63546-4

Abstract

Polygenic risk scores hold prognostic value for identifying individuals at higher risk of type 2 diabetes. However, further characterization is needed to understand the generalizability of type 2 diabetes polygenic risk scores in diverse populations across various contexts. We systematically characterize a multi-ancestry type 2 diabetes polygenic risk score among 244,637 cases and 637,891 controls across diverse populations from the Population Architecture Genomics and Epidemiology Study and 13 additional biobanks and cohorts. Polygenic risk score performance is context dependent, with better performance in those who are younger, male, without hypertension, and not obese or overweight. Additionally, the polygenic risk score is associated with various diabetes-related cardiometabolic traits and type 2 diabetes complications, suggesting its utility for stratifying risk of complications and identifying shared genetic architecture between type 2 diabetes and other diseases. These findings highlight the need to account for context when evaluating polygenic risk score as a tool for type 2 diabetes risk prognostication and the potentially generalizable associations of type 2 diabetes polygenic risk score with diabetes-related traits, despite differential performance in type 2 diabetes prediction across diverse populations. Our study provides a comprehensive resource to characterize a type 2 diabetes polygenic risk score.

PubMed Disclaimer

Conflict of interest statement

Competing interests: R.J.F.L. has acted as a member of advisory boards and as a speaker for Ely Lilly and the Novo Nordisk Foundation, for which she has received fees. L.M.R. is a consultant for the NHLBI TOPMed Administrative Coordinating Center (through Westat). U.P. was a consultant with AbbVie, and her husband holds individual stocks for the following companies: BioNTech SE—ADR, Amazon, CureVac BV, Google/Alphabet Inc Class C, NVIDIA Corp, Microsoft Corp. The remaining authors declare no competing interests.

Figures

**Fig. 1. Workflow of T2D PRS construction and evaluation.**
Four T2D PRS were constructed and evaluated in the Population Architecture using Genomics and Epidemiology (PAGE) study to identify the best-performing score for downstream analyses. The best-performing PRS across populations was used for stratified analyses across demographic, medical, and lifestyle and behavioral characteristics, and association analyses with diabetes-related traits in PAGE. Stratified and association analyses were replicated in 11 independent cohorts. Phenome-wide association studies (PheWAS) were conducted across five independent biobanks to assess the broader clinical impact of the T2D PRS.

**Fig. 2. Performance of T2D PRS across self-identified populations.**
A Area under the curve (AUC) for four T2D PRS methods in PAGE. PRS Methods 1–4 correspond to those in Fig. 1. Data are presented as AUC with 95% confidence interval (CI). B Association between each SD unit increase in the Ge et al. T2D PRS with T2D risk in PAGE and the additional biobanks and cohorts. Data are presented as odds ratios (OR) with 95% CI. Sample sizes for A and B are provided in Supplementary Data 1.

**Fig. 3. Distribution of age at T2D diagnosis by PRS decile and population in All of Us.**
Data are presented as mean age at diagnosis with standard error. Sample sizes are provided in Supplementary Data 6.

**Fig. 4. Effect of the T2D PRS on T2D risk stratified by demographic, medical, and lifestyle and behavioral factors meta-analyzed across PAGE and the additional biobanks and cohorts.**
Data are presented as odds ratio (OR) with 95% confidence interval (CI). P-values of heterogeneity from a two-sided Cochrane Q-test are indicated when differences were statistically significant (P < 0.05). A Demographic characteristics and medical history factors. B Behavioral and lifestyle factors. C Medication use. D Lipids. Sample sizes for A–D are provided in Supplementary Data 9.

**Fig. 5. Effect of T2D PRS on diabetes-related traits meta-analyzed across PAGE and the additional biobanks and cohorts.**
The X-axis represents the beta estimates and 95% confidence intervals (CIs) of PRS for continuous traits and odds ratios (OR) and 95% CIs of PRS for binary traits. P-values shown on each plot are exact values derived from two-sided linear regression models for continuous traits and two-sided logistic regression models for binary traits. Solid circles indicate significant associations that passed the Bonferroni-adjusted P-value threshold of P < 2.50 × 10⁻³, while open circles indicate associations that were not significant. Panels show associations for the T2D PRS and A glycemic traits in T2D controls and B cardiometabolic traits in T2D cases (left), controls (middle), and individuals with prediabetes (right). The units of continuous traits are listed in Supplementary Data 13. Sample sizes are provided in Supplementary Data 14. Results for vascular disease, kidney function related traits, and inflammatory biomarkers are shown in Supplementary Fig. 4.

**Fig. 6. T2D PRS PheWAS results meta-analyzed across all five biobanks and populations.**
The X-axis represents phecodes color-coded by their corresponding phenotype category and is ordered by the category with the most to the category with the least significant hits. The Y-axis represents the −log₁₀(p-value) derived from two-sided logistic regression models. The red horizontal line represents Bonferroni-adjusted p-value threshold P < 2.75 × 10⁻⁵ (732 phecodes meet this threshold), and the blue horizontal line represents an unadjusted p-value threshold of P < 0.05 (1120 phecodes meet this threshold). Upward triangles indicate positive associations, while downward triangles indicate negative associations. The top ten most significant associations from the endocrine/metabolic category are annotated, with the single most significant association shown in black and others shown in gray, while the single most significant association from each other category is annotated.

**Fig. 7. Comparison of significant associations and effect sizes of T2D PRS PheWAS results by population.**
A Venn diagram of significant PheWAS results across populations. B Comparison of odds ratios from T2D PRS PheWAS results across populations.

See this image and copyright information in PMC

References

1. International Diabetes Federation. IDF Atlas 2021 10th edn (International Diabetes Federation, 2021).
1. Bommer, C. et al. Global economic burden of diabetes in adults: projections from 2015 to 2030. Diabetes Care41, 963–970 (2018). - PubMed
1. Khera, A. V. et al. Genome-wide polygenic scores for common diseases identify individuals with risk equivalent to monogenic mutations. Nat. Genet.50, 1219–1224 (2018). - PMC - PubMed
1. Frieser, M. J., Wilson, S. & Vrieze, S. I. Behavioral impact of return of genetic test results for complex disease: systematic review and meta-analysis. Health Psychol.37, 1134–1144 (2018). - PMC - PubMed
1. Natarajan, P. et al. Polygenic risk score identifies subgroup with higher burden of atherosclerosis and greater relative benefit from statin therapy in the primary prevention setting. Circulation135, 2091–2101 (2017). - PMC - PubMed

MeSH terms

Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions
Actions

Grants and funding

LinkOut - more resources

Full Text Sources
- Nature Publishing Group
- PubMed Central
Medical
- MedlinePlus Health Information

Save citation to file

Email citation

Add to Collections

Add to My Bibliography

Your saved search

Create a file for external citation management software

Your RSS Feed

Polygenic risk score for type 2 diabetes shows context-dependent effects across populations

Collaborators

Affiliations

Polygenic risk score for type 2 diabetes shows context-dependent effects across populations

Authors

Collaborators

Affiliations

Abstract

Conflict of interest statement

Figures

References

MeSH terms

Grants and funding

LinkOut - more resources

Full Text Sources

Medical