Alcohol Liver Cirrhosis May Have a Higher Genetic Risk Score

Introduction

Alcohol, a leading cause of liver disease is associated with significant morbidity and mortality. Early recognition of Alcohol-related liver disease (ARLD) would be useful to encourage alcohol abstinence, minimize the progression of liver fibrosis, and manage cirrhosis-related complications including hepatocellular carcinoma.

Genome-wide association studies (GWAS) have identified several Single nucleotide polymorphisms (SNPs) for ARLD.¹ Genetic risk score (GRS, unweighted and weighted) modelling provides an opportunity to examine the effects of genetic factors on an outcome² because it sums up the genetic risk contributed by each locus.

The SNP-based GRS can be computed to evaluate the inherited genetic risk for ARLD in an individual with alcohol use disorder. We computed the GRS based on SNPs for risk assessment at the individual level.³

Materials and Methods

This study was approved by the Institutional Ethics Committee, St John’s Medical College Hospital and National Institute of Mental Health and Neurosciences. Written informed consent was obtained from all the voluntary participants for the study.

Men with Alcohol Use Disorder (AUD) with Cirrhosis (AUDC +ve, N=131) and AUD without Cirrhosis (AUDC-ve, N= 105) based on International classification of Mental and Behavioural Disorders (ICD) 10 criteria, drawn from the clinical services of St John’s Medical College Hospital (Gastroenterology and Psychiatry) and from National Institute of Mental Health and Neuro Sciences (NIMHANS) – Centre for Addiction Medicine outpatient clinic were studied. Fibroscan and / or sonographic findings were used to rule out fibrosis (Liver Stiffness Measurement, LSM <14kPa/Fib-4 <3.25) in the AUDC-ve group. SAS (N=260) 1000 genome data was used to calculate GRS in the population.

The serum sample was used to estimate biochemical parameters. A total of 10 SNPs from genes involved in alcohol metabolism (ADH2-rs2066701, ADH3- rs1789920, ALDH2-rs2238151); lipid metabolism (PNPLA3-rs738409, TM6SF2-rs58542926, APOC3- rs2854116); cytokine (PPARγ-rs1801282, TNFα-rs361525) and one-carbon metabolism (MTHFR-rs1801131 and rs1801133) known to be associated with the risk of AUDC+ve,^4,5 were assessed in this study. Genotypes were generated for all 10 SNPs using PCR-RFLP (Polymerase Chain Reaction- Restriction Fragment length Polymorphism). GWAS and association studies have shown that these SNPs are associated with alcohol use disorder with cirrhosis.^4,6-8 Unweighted genetic risk score (uGRS) – summation of allele doses of SNPs in the current study was calculated. Also, population standardized weighted genetic risk scores (wGRS) were calculated based on the effect size from a recent study.⁹

Results and Discussion

The demographic and clinical characteristics of participants (N=236) is described in Table 1. The two groups did not differ in age, but those in the AUDC +ve group had used lesser quantities of alcohol (Mean ± SD = 130 ± 68 mL/day) compared to the AUDC-ve group (Mean ± SD = 160 ± 77 mL/day) (p=0.01). They had also been drinking for a comparatively shorter duration (16±7 years vs 18±8 years) and had a later age of onset of drinking (29±8 years vs 23±7 years) (p=<0.001). AUDC+ve group showed higher concentrations of serum total and direct bilirubin, Alkaline Phosphatase (ALP), Gamma- Glutamyl Transferase (GGT) and lower concentrations of total protein, albumin, and haemoglobin levels compared to AUDC-ve group (Table 2).

Unweighted GRS (uGRS) – The calculated risk scores ranged from 0-10. As the numbers were modest at the extremes, we collapsed 0-3 as a single group and also 8-10 as another group. The frequency distribution for the uGRS scores were 5%, 14%, 20%, 23%, 19%, and 17% in the AUDC+ve group compared to 16%, 14%, 10%, 20%, 25%, and 12% in the AUDC-ve group. The uGRS scores were higher in AUDC+ve group compared to AUDC-ve group (p=0.08). But it did not reach statistical significance.

In the regression model, we found higher uGRS (β=0.14; p=0.03) was significantly associated with cirrhosis after controlling for other variables like age, quantity of alcohol, duration of drinking and age at onset (AAO) (β=0.62; p=0.04). 

wGRS calculation showed a significantly higher genetic risk score in the AUDC+ve group (Mean ± SD = 1.3±1.0) and in AUDC-ve group (Mean ± SD = 1.1±1.0) compared to control population group (SAS) (Mean ± SD = 0.82 ± 0.76, p=0.0001).

The area under curve (AUC) for overall (10 SNPs) wGRS was 0.572 with a cutoff of 1.2 (sensitivity, 52% and specificity, 40%) and AUC for uGRS was - 0.546 with a cutoff of 4.5 (sensitivity, 80% and specificity, 60%). For alcohol metabolizing genes (ADH3, ADH2 and ALDH2), GRS was AUC=0.595, p=0.02 with a cutoff of 0.48 (sensitivity, 54% and specificity, 35%). Alc-GRS (Alcohol metabolising genes- GRS) had higher predictability.

Another study by Whitfield et al., found that risk score based on three genetic risk variants and the diabetic status enabled the stratification of heavy drinkers based on their risk of cirrhosis, permitting earlier preventive interventions.¹⁰

GRS significantly contributes to risk prediction in cirrhosis group. GRS might be useful to predict the risk of developing cirrhosis in individuals who have been drinking heavily. This prediction, based on a DNA sample is possible before clinical symptoms are felt, by which time considerable tissue/organ damage would have usually occurred. We need to validate the findings in a larger population, however.

Based on our findings, genetic risk score alone is not contributing to predicting risk of alcohol liver cirrhosis but with potential other variables viz., age, age at onset, quantity of alcohol consumption, and duration of alcohol consumption and to explore further including a larger number of SNPs associated with alcohol liver cirrhosis. 

Conflict of Interest

None

Acknowledgments

We acknowledge comments and feedback received at the Society of Biological Psychiatry Conference 2022 which has helped to improve the manuscript.