Christine Möller

Background: Development of cirrhosis and hepatocellular carcinoma (HCC) in patients with high alcohol intake is modulated by genetic predispositions. Genetic variation in angiotensin II type 1 receptor (AGTR1) has been described as a risk factor for non-alcoholic fatty liver disease in Asian patients. Methods: We analysed Caucasian patients with alcohol–associated cirrhosis without (n = 238) and with (n = 339) HCC, healthy controls (n = 200), and HCV–infected cirrhotic patients with and without HCC (n = 263) for association with the polymorphisms rs3772622 and rs2276736 in AGTR1. Results: Rs2276736 in AGTR1 was associated with both low–density lipoprotein (LDL) cholesterol levels and hepatic steatosis in patients with alcohol–associated liver disease. The distribution of genotypes for both rs3772622 and rs2276736 in AGTR1 were comparable between controls, cirrhosis patients, and those with HCC. Minor allele frequencies were 32% (44%) in healthy controls, 35%/34% (46%/45%) in alcohol–associated liver disease without/with HCC and 31%/38% (43%/39%) in HCV cirrhosis and HCV HCC, respectively. The genotype of the most important genetic risk factor for fatty liver disease, PNPLA3 I148M, did not interact with the AGTR1 polymorphisms. Conclusion: Genetic variation in AGTR1, although associated with blood lipid levels and hepatic steatosis, is not a risk factor for alcohol–associated cirrhosis or HCC in Caucasians.

Hepatocellular carcinoma (HCC) is a severe complication of advanced alcoholic liver disease, which is modulated by genetic predisposition. Identifying new genetic loci might improve screening. Genetic variation of SAMM50 was linked to HCC. We aimed to validate this finding in a large cohort of patients with advanced alcoholic liver disease (ALD). A large, well-characterised cohort of patients with alcoholic cirrhosis without (n = 674) and with (n = 386) HCC, as well as controls with HCC due to viral hepatitis (n = 134), controls with heavy alcohol abuse without liver disease (n = 266) and healthy subjects (n = 237), were genotyped for SAMM50 rs3827385 and rs3761472 and for PNPLA3 rs738409. Genotype frequencies were compared between patients with alcohol-associated cirrhosis with and without HCC by uni- and multivariate analysis. Minor variants in both SAMM50 rs3827385 and rs3761472 were significantly more frequent in patients with alcoholic HCC versus alcoholic cirrhosis and versus the control cohorts. An even stronger association was noted for PNPLA3 rs738409. The univariate analysis resulted in an odds ratio (OR) of 1.8 for carriers of at least one minor variant of SAMM50 rs3827385 and rs3761472 (each p < 0.001), but this association was lost in multivariate analysis with age (OR 1.1/year), male sex (OR 3.2), diabetes (OR 1.9) and carriage of PNPLA3 148M (OR 2.1) remaining in the final model. Although minor variants of both SAMM50 loci are strongly associated with alcoholic HCC, this association is not independent of carriage of the well-known risk variant PNPLA3 148M.