Journal of the American Heart Association ·12 Nov 2026 ·Added 12 Nov 2026

Clonal Hematopoiesis and Cardiovascular and Microvascular Outcomes in Type 2 Diabetes: A Retrospective Analysis of the UK Biobank.

Yang Gao, Chaowei Zhang, Donghui Chen, Zhuocheng Shi, Qingbo Shi, Yushuo Gu, Boyi Shi, Cao Ma, Quan Guo, Yunxia Wang, Zhiwen Zhang, Bingchuan Geng, Muwei Li

Proteomics & Metabolomics

DOI ↗

Abstract

BACKGROUND: Clonal hematopoiesis of indeterminate potential (CHIP) has been associated with cardiovascular disease, but its associations across cardiovascular and microvascular complications in type 2 diabetes remain incompletely characterized. METHODS: We retrospectively analyzed 17 893 UK Biobank participants with type 2 diabetes, whole-exome sequencing, and no prevalent cardiovascular or microvascular complications. CHIP was defined from the curated UK Biobank call set by ≥1 driver variant with a variant allele fraction ≥2%. Associations with incident outcomes were evaluated using Fine-Gray models with multiple imputation for missing covariates. Driver-gene, clone-size, proteomic, and metabolomic analyses were exploratory. RESULTS: CHIP was present in 670 participants (3.7%) over a median follow-up of 14.6 years. CHIP was associated with the composite cardiovascular outcome (subdistribution hazard ratio [SHR], 1.19 [95% CI, 1.00-1.42]), with heart failure showing the clearest component association (SHR, 1.43 [95% CI, 1.13-1.81]); acute coronary syndrome and stroke were not significantly associated. CHIP was also associated with the composite microvascular outcome (SHR, 1.34 [95% CI, 1.15-1.56]), broad chronic kidney disease (SHR, 1.28 [95% CI, 1.05-1.57]), and retinopathy (SHR, 1.43 [95% CI, 1.15-1.77]) but not neuropathy or strict diabetes-specific nephropathy. Ten-year adjusted risk differences were 1.94% and 3.46% for the cardiovascular and microvascular composites, respectively. Among 2919 proteins, ZBP1 (Z-DNA-binding protein 1) was the only protein retained after false-discovery-rate correction; no metabolite module was associated with CHIP after correction, whereas selected lipoprotein and amino acid/lactate modules were associated with cardiovascular outcomes, particularly heart failure. CONCLUSIONS: In adults with type 2 diabetes, CHIP was associated with a higher risk of incident microvascular complications and, among cardiovascular outcomes, most consistently with heart failure. Exploratory multiomic findings, including a protein-wide ZBP1 signal and selected metabolic associations with cardiovascular outcomes, provided additional biological context but require independent validation.