Coffee Consumption and Improved Liver Outcomes: Clinical, Imaging, and Proteomic Evidence From the UK Biobank
Clinical Gastroenterology and Hepatology, 2026
Kim H., Rezaee-Zavareh M., Wang Y., Attia A., Kwak M., Burm S., Celtik D., Legaspi D., Khattab O., Kim N., Mengistu B., Larios K., Kim D., Ayoub W., Kuo A., Martin P., Vipani A., Wang Y., Liangpunsakul S., Li D., Lu S., Pandol S., Yang J.
| Disease area | Application area | Sample type | Products |
|---|---|---|---|
Hepatology Nutritional Science | Pathophysiology | Plasma |  Olink Explore 3072/384 |
Abstract
Background & Aims
Coffee consumption has been linked to a reduced risk of liver disease, but large-scale prospective studies integrating imaging biomarkers, proteomics, and intake patterns remain limited. The effects of coffee type, additives, and metabolic modifiers are also unclear.
Methods
We analyzed 354,957 UK Biobank participants without baseline cirrhosis or hepatocellular carcinoma. Coffee consumption, type (caffeinated/decaffeinated), and additives (sugar/sweeteners) were assessed by questionnaire. Incident cirrhosis, hepatocellular carcinoma, and liver-related mortality were ascertained through linked records. In a subcohort (n = 28,961) undergoing magnetic resonance imaging, hepatic fat (proton density fat fraction), iron, and fibroinflammation (iron-corrected T1) were evaluated. Proteomic profiling (n = 44,633) used Olink assays. Models were adjusted for demographic, behavioral, metabolic, and genetic covariates.
Results
Over a median 13-year follow-up, higher coffee intake showed a graded reduction in liver outcomes, with ≥5 cups/day group having risk reductions of cirrhosis (hazard ratio, 0.68; 95% confidence interval, 0.58–0.79), hepatocellular carcinoma (hazard ratio, 0.53; 95% confidence interval, 0.34–0.83), and liver-related mortality (hazard ratio, 0.58; 95% confidence interval, 0.45–0.74). Protective associations were similar for caffeinated and decaffeinated coffee and persisted among those adding sugar or artificial sweeteners, although additive use correlated with modestly higher iron-corrected T1. Higher coffee intake corresponded to lower hepatic fat, iron, and fibroinflammation. Proteomic analysis revealed consistent patterns: higher levels of hepatocellular synthesis and complement proteins (transthyretin, selenoprotein P, complement factor H-related protein4/5) and lower levels of fibrogenic and macrophage-activation markers (microfibril-associated protein 4, colony stimulating factor 1 receptor, ectonucleotide pyrophosphatase/phosphodiesterase 2, transthyretin) with coffee drinking.
Conclusions
Higher coffee intake was associated with favorable clinical, imaging, and proteomic indicators of liver health. These multidimensional findings support moderate unsweetened coffee as a simple strategy for liver disease prevention.