Binge-pattern alcohol consumption and genetic risk as determinants of alcohol-related liver disease

Abstract

Alcohol-related liver disease (ARLD) represents a major public health burden. Identification of high-risk individuals would allow efficient targeting of public health interventions. Here, we show significant interactions between pattern of drinking, genetic predisposition (polygenic risk score, PRS) and diabetes mellitus, and risk of incident ARLD, in 312,599 actively drinking adults in UK Biobank. Binge and heavy binge drinking significantly increase the risk of alcohol-related cirrhosis (ARC), with higher genetic predisposition further amplifying the risk. Further, we demonstrate a pronounced interaction between heavy binge drinking and high PRS, resulting in a relative excess risk due to interaction (RERI) of 6.07. Diabetes consistently elevates ARC risk across all drinking and PRS categories, and showed significant interaction with both binge patterns and genetic risk. Overall, we demonstrate synergistic effects of binge drinking, genetics, and diabetes on ARC, with potential to identify high-risk individuals for targeted interventions.

Fig. 1. Risk of advanced ARLD associated with alcohol consumption group.

Data are presented as hazard ratios (HRs) with respective 95% confidence intervals (CIs). All the p values were two-sided and calculated using the Wald test. HRs were derived from Cox model with the following covariates: alcohol consumption group, sex, age, ethnicity, Townsend deprivation index at recruitment, physical activity, smoking, total weekly alcohol intake, beverage type, drinking with/without meal, prevalent diabetes, and BMI categories.

Fig. 2. Risk of alcohol-related cirrhosis by alcohol consumption and polygenic risk group.

Data are presented as hazard ratios (HRs) with respective 95% confidence intervals (CIs). All the p values were two-sided and calculated using the Wald test. HRs were derived from Cox model with following covariates: interaction of alcohol consumption and PRS groups, total weekly alcohol intake, diabetes, BMI, sex, age, ethnicity, Townsend deprivation index, physical activity, smoke, alcohol type, drinking with/without meals, genotyping array, first 10 ancestry principal components.

Fig. 3. Risk of alcohol-related cirrhosis by diabetes status and both alcohol consumption and polygenic risk group.

Data are presented as hazard ratios (HRs) with respective 95% confidence intervals (CIs). All the p values were two-sided and calculated using the Wald test. HRs were derived from Cox models with the following covariates: sex, age, ethnicity, Townsend deprivation index at recruitment, physical activity, smoking, total weekly alcohol intake, beverage type, drinking with/without meal, BMI, genotyping array, first 10 ancestry principal components plus PRS group, the interaction of alcohol consumption group and diabetes for (a), and alcohol consumption group, interaction of PRS group and diabetes for (b). Data are presented as HR and 95% confidence interval.

Fig. 4. Hazard ratios for risk of alcohol-related cirrhosis with contributions of heavy binge, high PRS, diabetes, and their combination (interaction).

Baseline represents the risk in referent participants who reported drinking within daily limits, had low PRS, and were free of diabetes. Source data are provided as a Source Data file.