ABSTRACT
BACKGROUND: The increasing prevalence of metabolic dysfunction-associated steatotic liver disease (MASLD) incurs substantial morbidity, mortality and healthcare costs. Detection and clinical intervention at early stages of disease improves prognosis; however, we are currently limited by a lack of reliable diagnostic tests for population screening and monitoring responses to therapy. To address this unmet need, we investigated human invariant Natural Killer T cell (iNKT) activation by fat-loaded hepatocytes, leading to the discovery that circulating soluble CD46 (sCD46) levels accurately predict hepatic steatosis. METHODS: sCD46 in plasma was measured using a newly developed immuno-competition assay in two independent cohorts: Prospective living liver donors (n = 156; male = 66, female = 90) and patients with liver tumours (n = 91; male = 58, female = 33). sCD46 levels were statistically evaluated as a predictor of hepatic steatosis. FINDINGS: Interleukin-4-secreting (IL-4+) iNKT cells were over-represented amongst intrahepatic lymphocytes isolated from resected human liver samples. IL-4+ iNKT cells preferentially developed in cocultures with a fat-loaded, hepatocyte-like cell line, HepaRG. This was attributed to induction of matrix metalloproteases (MMP) in fat-loaded HepaRG cells and primary human liver organoids, which led to indiscriminate cleavage of immune receptors. Loss of cell-surface CD46 resulted in unrepressed differentiation of IL-4+ iNKT cells. sCD46 levels were elevated in patients with hepatic steatosis. Discriminatory cut-off values for plasma sCD46 were found that accurately classified patients according to histological steatosis grade. INTERPRETATION: sCD46 is a reliable clinical marker of hepatic steatosis, which can be conveniently and non-invasively measured in serum and plasma samples, raising the possibility of using sCD46 levels as a diagnostic method for detecting or grading hepatic steatosis. FUNDING: F.B. was supported by the Else Kröner Foundation (Award 2016_kolleg.14). G.G. was supported by the Bristol Myers Squibb Foundation for Immuno-Oncology (Award FA-19-009). N.S. was supported by a Wellcome Trust Fellowship (211113/A/18/Z). J.A.H. received funding from the European Union's Horizon 2020 research and innovation programme (Award 860003). J.M.W. received funding from the Else Kröner Foundation (Award 2015_A10).
