Information from the abstract
Prediabetes is a heterogeneous condition, with individuals exhibiting varying risks of progression to type 2 diabetes (T2D). We applied data-driven clustering to metabolic variables to identify prediabetes subgroups, evaluated their progression risks, and evaluated single nucleotide polymorphisms (SNPs)–cluster interactions. Among 1,016 individuals with prediabetes, k‑prototypes clustering was applied to age, sex, BMI, waist circumference, blood pressure, lipid parameters, and HbA1c to derive phenotypic clusters. Incident T2D risk was compared across clusters using Cox proportional‑hazards regression. Interactions between four T2D risk variants ( THADA, CDKN2B, SLC30A8, VPS26A ) and cluster membership were tested. Cluster‑specific genetic risk was quantified using a polygenic risk score (PRS). Four prediabetes sub-phenotypes were identified: low-risk prediabetes (LORPD), characterized by younger, predominantly female individuals with near-normal BMI and lipid profiles; mild hyperglycemia prediabetes (MIHPD), characterized by modest hyperglycemia with otherwise favorable metabolic characteristics; mild obesity-related prediabetes (MORPD), characterized by obesity-related metabolic abnormalities; and metabolic syndrome prediabetes (MESPD), the most severe group with the highest BMI, waist circumference, blood pressure, and the most adverse lipid profile. Progression to T2D varied markedly by cluster: MESPD cluster had the highest incidence (58.1 per 1,000 person-years) and roughly 24-fold (95% CI 12.88–45.68) compared to controls. In contrast, the LORPD cluster had the lowest progression risk, at 3.35-fold (95% CI 1.15–9.80). The MIHPD (HR 12.88; 95% CI 7.06–23.50) and MORPD (HR 16.26; 95% CI 9.91–26.70) cluster showed intermediate rates of progression to T2D. Significant SNP–cluster interactions were detected. In MESPD, THADA rs7578597 T, CDKN2B rs7018475 G, and VPS26A rs1802295 T increased T2D hazard (HRs 6.77, 4.06, 4.98); in MORPD, SLC30A8 rs13266634 T was associated (HR 3.25). PRS were used to refine risk stratification, with the MESPD cluster showing a significant genetic predisposition to T2D. PRS further improved T2D risk prediction within the LORPD cluster. These results indicate that although our clusters were delineated solely by clinical phenotype, integrating genetic–cluster interaction data can substantively enhance preventive strategies. Moreover, these findings support precision prevention by tailoring monitoring and interventions to individuals’ combined phenotypic and genetic risk profiles.
Why this record is monitored
This record has an Impact Signal of 77/100 based on recency, source, collaboration, and bibliographic signals. It prioritizes monitoring and is not a judgment of research quality.
Related topics: Genetic Associations and Epidemiology · Diabetes, Cardiovascular Risks, and Lipoproteins · Adipokines, Inflammation, and Metabolic Diseases
Thai researcher and institutional participation
Apinya Surawit · Phongthana Pasookhush · Sureeporn Pumeiam · Pichanun Mongkolsucharitkul · Sophida Suta · Bonggochpass Pinsawas · Suphawan Ophakas · Nutchavadee Vorasan · Parit Voharnsuchon · Korapat Mayurasakorn · Siriraj Hospital
Data limitations
This page is a bibliographic record based on abstract-level information, not a full analysis or quality assessment. Verify the DOI and original article before citation.