gms | German Medical Science

Introduction: Hepatic steatosis is one of the most common chronic liver diseases and a major public health concern. The prevalence of steatotic liver disease has been estimated to be 23.4% worldwide, with a sharply increasing prevalence in the past years [1]. In recent animal experiments, hepatic iron has been implicated to promote steatosis and exacerbate fibrotic conditions [2]. Thus, pathways implied in iron metabolism could be relevant targets for therapeutic interventions in steatotic liver disease. Metabolomics have emerged as a powerful tool to characterize pathophysiological pathways in metabolic disease, including liver disease [3]. In the current study, we aimed to use population-based data to identify serum metabolites that are associated with hepatic phenotypes, including hepatic fat and iron content derived by magnetic resonance imaging (MRI).

Methods: The analysis is based on the KORA-MRI study, a sample from a population-based cohort including N=400 individuals without history of cardiovascular disease who underwent whole-body MRI [4]. Hepatic fat content was assessed as proton density fat fraction in % and iron content as relaxation rate in s-1. Hepatic steatosis was defined as liver fat content ≥ 5.56% and iron overload as values ≥ 41.0 s-1. The fatty liver index (FLI) was calculated from BMI, waist circumference, triglycerides and GGT [5]. Targeted serum metabolites were quantified from fasted samples by the Biocrates AbsoluteIDQTM p180 kit. Associations between metabolites as exposure variables and hepatic phenotypes as outcomes were evaluated by linear or logistic regression models, adjusted for potential confounders and corrected for multiple testing. Pathway analyses were conducted to reveal different pathways between individuals with and without steatosis, and individuals with and without iron overload, respectively.

Results: The final sample comprised 217 men and 159 women (mean age 56 years). Overall, 50.8% of participants had hepatic steatosis and 43.6% had iron overload. After adjustment for confounders, 6 metabolites (three amino acids, alpha-aminoadipic acid, one lysophosphatidylcholine and one acylalkylphosphatidylcholines) were significantly associated with hepatic fat content, and 12 metabolites (two carnitines, alpha-aminoadipic acid, one lysophosphatidylcholine, four diacylphosphatidylcholines, two acylalkylphosphatidylcholines, two sphingomyelins) were associated with hepatic iron content. Performance of metabolites to predict hepatic steatosis and iron overload was superior to that of the FLI in men (AUC of 0.917 vs 0.826, and 0.798 vs 0.607 for steatosis and iron overload, respectively, p<0.01), and non-inferior to that of the FLI in women (AUC of 0.879 vs 0.881, and 0.664 vs 0.593 for steatosis and iron overload, respectively, p>0.3). Pathway analysis showed overlapping pathways in hepatic steatosis and iron overload such as phenylalanine metabolism; phenylalanine, tyrosine and tryptophan biosynthesis; and alanine, aspartate and glutamate metabolism.

Conclusion: In a sample from a population-based cohort, circulating metabolites that were associated with hepatic fat and iron content predicted steatosis and iron overload. Moreover, these metabolites shared common pathways, underlining the potential role of iron in the progression of hepatic disorders, and the potential role of iron-related treatment targets.

The authors declare that they have no competing interests.

The authors declare that a positive ethics committee vote has been obtained.