Levels of Growth Differentiation Factor 15 Correlated with Metabolic Dysfunction-Associated Steatotic Liver Disease in Children

Abstract

Metabolic dysfunction-associated steatotic liver disease (MASLD) is the most common chronic progressive hepatopathy in children, and the identification of non-invasive biomarkers is urgently needed. Growth differentiation factor 15 (GDF15) was associated with MASLD in adults. In this study, we investigated the circulating and hepatic levels of GDF15 and their association with liver damage in pediatric MASLD and in a murine model. This observational study included 158 children with biopsy-proven MASLD. Patients with MASLD were categorized into two groups based on steatohepatitis (MASH) presence and evaluated for GDF15 circulating levels, while GDF15 hepatic levels were assessed only in a subset of patients. Children with MASLD exhibited higher levels of circulating GDF15 compared to the controls. Moreover, the MASH subgroup had significantly higher values of GDF15 compared to the Not-MASH subgroup. The GDF15 levels in the MASH subgroup showed a positive correlation with fibrosis. Finally, the hepatic expression of the GDF15 gene correlated with GDF15 circulating levels and with the hepatic expression of the COL1A1 and COL3A1 genes in 15 children with MASLD. In conclusion, our study demonstrated that GDF15 levels are associated with liver damage, reinforcing the potential role of GDF15 as a biomarker for MASLD-related fibrosis in children.

Keywords: GDF15; MASH; MASLD; children; hepatic steatosis; insulin resistance; liver fibrosis; metabolic syndrome; obesity; tissue injury.

Figure 1

Plasma levels of GDF15 in the study population. The violin plot shows the comparison of the plasma GDF15 levels between 24 healthy subjects (controls) and 158 patients with MASLD (MASLD). Mann–Whitney U-test, *** p < 0.001.

Figure 2

Plasma levels of GDF15 in MASLD patients according to MASH diagnosis. (A) The violin plot shows the comparison of the plasma GDF15 levels between patients without (Not-MASH) and with a MASH diagnosis (MASH). Mann–Whitney U-test, *** p < 0.001. (B) The violin plot shows the comparison of the plasma GDF15 levels between different stages of fibrosis in patients with MASH.

Figure 3

Plasma and hepatic expression of GDF15 in a subgroup of children with MASLD. (A) The graph shows the relative hepatic gene expression of GDF15 in 9 children with MASH diagnosis (MASH) vs. 6 children without MASH (Not-MASH), normalized to β-actin transcript expression. Unpaired t-test with Welch’s correction, * p < 0.05. (B) Scatter plot representation of the plasma and corresponding hepatic GDF15 levels in 9 MASH and 6 Not-MASH children.

Figure 4

Hepatic expression of GDF15 and correlated genes in a gene expression dataset from adult patients. (A) Violin plot shows the fold-change (log2) hepatic expression of the GDF15 transcript in control subjects (CTRLs) and in patients with simple fatty liver (FL) or MASH (MASH) by using data retrieved from the GSE126848 dataset. One-way Anova with Bonferroni correction ** p < 0.001 (B) Pathways enrichment analysis by Reactome for genes significantly correlated with GDF15. Correlation analysis between the hepatic gene expression of GDF15 and that of (C) COL1A1 and (D) COL3A1 in the GSE126848 dataset.

Figure 5

Hepatic gene expression levels of COL1A1 and COL3A1 and correlation with GDF15 gene expression in a subgroup of children with MASLD. The graphs show the relative hepatic gene expression of (A) COL1A1 and (B) COL3A1 in 9 children with a MASH diagnosis (MASH) and 6 children without MASH (Not-MASH) analyzed by qRT-PCR and normalized to the levels of the β-actin transcript. Unpaired t-test with Welch’s correction, * p < 0.05. (C) Heatmap of the correlations between GDF15, COL1A1, and COL3A1 fold inductions in a subgroup of children with MASH (n = 9), * p < 0.05; ** p < 0.01.