Abstract and Introduction
Abstract
Background & Aims: Genetic variants, abdominal obesity and alcohol use are risk factors for incident liver disease (ILD). We aimed to study whether variants either alone or when aggregated into genetic risk scores (GRSs) associate with ILD, and whether waist-hip ratio (WHR) or alcohol use interacts with this risk.
Methods: Our study included 33 770 persons (mean age 50 years, 47% men) who participated in health-examination surveys (FINRISK 1992–2012 or Health 2000) with data on alcohol use, WHR and 63 genotypes associated with liver disease. Data were linked with national health registers for liver-related outcomes (hospitalizations, malignancies and death). Exclusions were baseline clinical liver disease. Mean follow-up time was 12.2 years. Cox regression analyses between variants and ILD were adjusted for age, sex and BMI.
Results: Variants in PNPLA3, IFNL4, TM6SF2, FDFT1, PPP1R3B, SERPINA1 and HSD17B13 were associated with ILD. GRSs calculated from these variants were not associated with WHR or alcohol use, but were exponentially associated with ILD (up to 25-fold higher risk in high versus low score). The risk of ILD in individuals with high GRS and high WHR or alcohol use compared with those with none of these risk factors was increased by up to 90-fold. GRSs provided new prognostic information particularly in individuals with high WHR.
Conclusions: The effect of multiple genetic variants on the risk of ILD is potentiated by abdominal obesity and alcohol use. Simple GRSs may help to identify individuals with adverse lifestyle who are at a particularly high risk of ILD.
Introduction
Abdominal obesity and alcohol use are the most common modifiable causes of chronic liver disease, predisposing to non-alcoholic fatty liver disease (NAFLD) and alcoholic liver disease (ALD) respectively.[1] NAFLD and ALD together affect up to a third of the population[1] and both conditions cover a spectrum of disease from steatosis to steatohepatitis to advanced liver disease including cirrhosis and hepatocellular carcinoma (HCC).[2] Abdominal obesity and high alcohol use can coexist, rendering individuals with both risk factors particularly vulnerable to advanced liver disease.[3–5] However, even with known lifestyle risk factors, the individual susceptibility to progress from steatosis to advanced liver disease remains highly variable.[6]
This variability might be explained by genetics.[7] At least 63 genetic variants to date have been shown to modify the risks of NAFLD and/or ALD.[8–13] Some of these, such as patatin-like phospholipase domain-containing protein 3 (PNPLA3) rs738409-G, transmembrane 6 superfamily member 2 (TM6SF2) rs58542926-T and hydroxysteroid 17-beta dehydrogenase 13 (HSD17B13) rs72613567-TA, affect both NAFLD and ALD, suggesting a shared pathogenesis underlying these conditions.[14,15] However, while the association between many variants and liver disease is well-validated, some variants were initially discovered in relatively small cohorts and not independently replicated.[8] Moreover, most of the studies have relied either on plasma aminotransferases, imaging or liver histology as the liver disease-related readout,[8] and it is unclear whether these variants associate with clinical liver-related outcomes, such as hospitalization owing to cirrhosis or hepatocellular cancer or liver-related death.
Recent evidence suggests that the impact of risk-modifying variants may not be fixed but rather depend on the presence of lifestyle risk factors. The impact of variants in PNPLA3, TM6SF2 and GCKR on hepatic triglyceride content was reported to be amplified by obesity.[16] Similarly, the impact of the PNPLA3 variant on ALT levels and risk of cirrhosis was modified by obesity.[16]
The impact of one risk-modifying variant may also depend on the presence of others. This was demonstrated in a recent study where the ALT-lowering effect of the HSD17B13 variant depended on the presence of the risk-increasing PNPLA3 and TM6SF2 variants, as well as on obesity and alcohol use.[17] To this end, genetic risk scores (GRS) have emerged as powerful instruments to investigate the cumulative effect of multiple risk-modifying variants.[18] Very recently, a GRS calculated using the PNPLA3, TM6SF2 and HSD17B13 variants was shown to increase the risk of cirrhosis and HCC by up to 12- and 29-fold respectively.[18] However, the relative importance of genetic risk in relation to modifiable lifestyle factors remains unclear. This knowledge could have marked clinical implications in identifying the individuals at the highest risk, as well as in defining the importance of lifestyle modification in the prevention and treatment of liver disease.
In the present study, we sought to validate 63 variants that have been previously associated with NAFLD or ALD using a candidate variant approach in a prospective, population-based cohort of 33 770 individuals with registry-linked liver-related outcome data and a mean follow-up of 12.2 years. Furthermore, we calculated GRSs and assessed the additive impact of abdominal obesity and alcohol use on the genetic risk for incident advanced liver disease (ILD).
Liver International. 2023;43(5):1035-1045. © 2023 Blackwell Publishing
