1261-OR: Molecular Risk Scores Enhance Type 2 Diabetes (T2D) Prediction beyond Hemoglobin A1c (HbA1c) and Body Mass Index (BMI) across Diverse Clinical Contexts
Diabetes · 07.06.2026 · 75 (Supplement_1)
DOI: 10.2337/db26-1261-or ↗Kurz gefasst
Fachartikel in der Zeitschrift Diabetes, erschienen am 07.06.2026. Verfasst von MAGDALENA SEVILLA-GONZALEZ und 9 weiteren Autor:innen.
Themen: Diabetes mellitus
Abstract (Originalsprache)
Introduction and Objective: Translating genetic and molecular biomarkers into clinically actionable T2D risk prediction models is challenged by uncertain incremental value beyond standard clinical factors in real-world populations. We assessed metabolomic risk score (MRS) and a polygenic risk score (PRS) for incident T2D prediction in a diverse healthcare system.
Methods
We studied 36,352 adults free of diabetes at metabolite assessment in the Mass General Brigham Biobank and followed for care for ~6 years. We derived the MRS using elastic net regression in UK Biobank (N=233K; 10,707 incident cases; follow-up: ~13 years) profiled on the Nightingale platform, and a PRS constructed from T2D GWAS summary statistics. Cox models to predict incident T2D were sequentially adjusted from demographics to established clinical risk scores (BMI, blood pressure, family history, and lipids), random glucose and HbA1c to mimic varying levels of clinical data availability.
Results
Both MRS and PRS were associated with incident T2D and improved discrimination in all models including those adjusted for HbA1c (P < 10-10, iC-index 0.01). Each standard deviation increase in MRS was associated with a threefold risk (age/sex adjusted model: HR 3.0, 95% CI 2.8-3.2, C-index 0.83; fully adjusted model: HR 1.7, 1.4-2.0, 0.84). Associations were stronger at higher BMI and among GLP-1 receptor agonist users (P interaction < 10-4) adjusting for clinical factors including BMI. Absolute risk estimation showed marked stratification: individuals with high MRS and PRS had ~30% 10-year T2D risk by age 40 and >50% by age 60 whereas lower risk profiles were <3% in all BMI categories. Higher MRS was also associated with increased risk of incident chronic kidney disease among individuals with T2D.
Conclusion
MRS and PRS provide complementary, clinically meaningful prediction of T2D beyond HbA1c and BMI supporting the integration of molecular profiling with standard clinical risk factors for targeted T2D prevention and risk reduction. Disclosure M. Sevilla-Gonzalez: Research Support; Current; Novo Nordisk. A.M. Martinez-Muñoz: None. P.A. Hanson: None. A. Huerta: None. M. Vora: None. E.W. Karlson: None. J.C. Florez: Research Support; Current; Novo Nordisk. Consultant; Current; Alveus Therapeutics. C.J. Patel: None. J. Mercader: None. A. Leong: Other - A close family member was an employee until August 2024.; Ended; Merck & Co., Inc. Funding 1K99DK139461-01A and R01DK137993 from NIH
Zitieren
APA
SEVILLA-GONZALEZ, M., MARTINEZ-MUÑOZ, A. M., HANSON, P. A., HUERTA, A., VORA, M., KARLSON, E. W., FLOREZ, J. C., PATEL, C. J., MERCADER, J., & LEONG, A. (2026). 1261-OR: Molecular Risk Scores Enhance Type 2 Diabetes (T2D) Prediction beyond Hemoglobin A1c (HbA1c) and Body Mass Index (BMI) across Diverse Clinical Contexts. Diabetes, 75(Supplement_1). https://doi.org/10.2337/db26-1261-or
Vancouver
SEVILLA-GONZALEZ M, MARTINEZ-MUÑOZ AM, HANSON PA, HUERTA A, VORA M, KARLSON EW, et al. 1261-OR: Molecular Risk Scores Enhance Type 2 Diabetes (T2D) Prediction beyond Hemoglobin A1c (HbA1c) and Body Mass Index (BMI) across Diverse Clinical Contexts. Diabetes. 2026;75(Supplement_1). doi:10.2337/db26-1261-or
BibTeX
@article{SEVILLAGONZALEZ2026,
author = {SEVILLA-GONZALEZ, MAGDALENA and MARTINEZ-MUÑOZ, ALAN M. and HANSON, PAUL A. and HUERTA, ALICIA and VORA, MAHEAK and KARLSON, ELIZABETH W. and FLOREZ, JOSE C. and PATEL, CHIRAG J. and MERCADER, JOSEP and LEONG, AARON},
title = {1261-OR: Molecular Risk Scores Enhance Type 2 Diabetes (T2D) Prediction beyond Hemoglobin A1c (HbA1c) and Body Mass Index (BMI) across Diverse Clinical Contexts},
journal = {Diabetes},
year = {2026},
volume = {75},
number = {Supplement_1},
doi = {10.2337/db26-1261-or}
}
EUVIMED