LAUSR

Nonalcoholic fatty liver disease (NAFLD), also defined as metabolic dysfunctionassociated fatty liver disease (MAFLD),1 is the leading cause of chronic liver disease worldwide, accounting for up to 30% based on recent global estimates.2 NAFLD comprises a broad spectrum of liver conditions, spanning from hepatic fat accumulation, progressing to inflammation (nonalcoholic steatohepatitis [NASH]), fibrosis and ultimately leading to cirrhosis, endstage liver disease and hepatocellular carcinoma.3 Obesity and type 2 diabetes are the major acquired risk factors for NAFLD. Indeed, up to 65% individuals with type 2 diabetes have NAFLD and this proportion becomes even higher in individuals with morbid obesity undergoing weightloss procedures.4,5 Similarly, genetic risk factors represent key determinants of liver disease development and progression.6,7 NAFLD may also coexist with other endocrinopathies.8 Within this context, mounting epidemiological evidence suggests that sex hormones may play an important role in determining liver fat content.9,10 Specifically, crosssectional populationbased cohort studies have reported that lower testosterone levels were associated with NAFLD in older men and inversely associated with NAFLD in women, whereas lower sex hormonebinding globulin (SHBG) levels were associated with NAFLD in both sexes.11 Similarly, oestrogen deficiency in postmenopausal women is associated with an increased prevalence of NAFLD as compared with premenopausal women.12 Along this line, elevated androgen levels in women with polycystic ovary syndrome (PCOS) are associated with a markedly increased risk of fatty liver, as well as insulin resistance and obesity.13 In this issue of Liver International, Smyk and colleagues went further and investigated the impact of common genetic variants predisposing to or protecting against NAFLD on liver phenotypes in two independent European cohorts of adult women with PCOS (German cohort, n = 42; Polish cohort, n = 143).14 They found that in the German cohort characterized by older age and more severe liver damage, carriers of the patatinlike phospholipase domaincontaining 3 (PNPLA3) p.I148M variant, the main genetic risk variant for NAFLD, had higher liver stiffness. Interestingly, in the Polish cohort, homozygous carriers of the hydroxysteroid 17beta dehydrogenase 13 (HSD17B13) rs72613567 T>TA lossoffunction variant, the main protective variant against fibrotic NASH, showed lower concentrations of total testosterone, 17OH progesterone and androstenedione (all p < .05). In the German cohort, no differences were found among carriers of this variant, sex hormones and SHBG levels. Limitations of this study include the use of different techniques to assess NAFLD across the cohorts, the lack of liver histology and the low sample size. However, it suggests the possibility that the protective association between the HSD17B13 variant and progressive NASH may be mediated by the modulation of sex hormones metabolism. This information has translational relevance, because silencing of hepatic HSD17B13 expression by RNA interference is a promising approach for the treatment of this condition.15,16 HSD17B13 belongs to the family of 17βhydroxysteroid dehydrogenases which comprises enzymes involved in sex hormone metabolism. Specifically, 17βhydroxysteroid dehydrogenases interconvert the inactive 17ketosteroid and the more active 17βhydroxy forms of oestrogens and androgens in a pivotal molecular process orchestrating the tissuespecific control of oestrogen and androgen levels.17 However, the role of HSD17B13 protein in human liver disease and the molecular mechanisms behind the protective impact of the HSD17B13 rs72613567 genetic variant on the risk of liver disease are still poorly understood.18 The results by Smyk et al.14 together with the putative biological role of HSD17B13 prompted us to examine the impact of the HSD17B13 rs72613567 variant on circulating SHBG, total and estimated free testosterone levels19 in more than 300,000 European participants from the UK Biobank populationbased cohort stratified by sex (Table 1). As expected, in both sexes, the minor TA allele was associated with lower ALT levels. In women, the minor allele was associated with lower SHBG levels (p = 8.9E−25), but higher free testosterone (p = 3.5E−09), whereas it had no significant impact on total testosterone. In men, the minor allele was associated with lower SHBG (p = 1.9E−41), lower total testosterone (p = 6.1E−24) and with no effect on free testosterone. The effect size of the reduction in ALT and SHBG was similar in both sexes. These results are discordant with the epidemiological evidence in individuals with NAFLD11 and show that NAFLD is not inevitably associated with lower SHBG levels in both sexes and lower total testosterone levels in men. To understand whether these results were specific for the HSD17B13 variant, we examined the association of the other common variants associated with the risk of NAFLD in PNPLA3, TM6SF2, MBOAT7, GCKR, MARC1, GPAM and APOE genes.20 In both sexes, all these variants, except for the one in MBOAT7 in women, associated with SHBG, a protein secreted exclusively by hepatocytes.