LETTERS to the EDITOR

Prediction of outcome in fulminant hepatic failure by serum human hepatocyte growth factor

The goal of this study was to evaluate whether pretransplant serum hyaluronic acid (HA) levels can predict outcomes after adult-to-adult living donor liver transplantation (LDLT).

In study I, 21 patients who underwent LDLT (March 2002-February 2004) were divided into 2 groups: the H-I group (HA ≥500 ng/mL; n = 12) and the L-I group (HA <500 ng/mL; n = 9). The influence of pretransplantation HA levels on short-term surgical outcome was investigated. In study II, 77 LDLT patients (May 2004-December 2014) were also divided into 2 groups: the H-II group (HA ≥500 ng/mL; n = 40) and the L-II group (HA <500 ng/mL; n = 37). We compared long-term survival and investigated prognostic factors.

In study I, HA levels significantly decreased after LDLT, and those in the H-I group were significantly higher compared with the L-I group at 1, 3, 5, and 7 days after LDLT. There were significant differences in postoperative peak total bilirubin levels (H-I vs L-I, 17.2 vs 6.2 mg/dL; P = .013), peak ascitic fluid volume (1327 vs 697 mL/d; P = .005), and the hepatocyte growth factor levels at 3 days after LDLT (1879 vs 1092 pg/mL; P = .03). In study II, the 1- and 5-year survival rates were significantly lower in the H-II group than in the L-II group (H-II vs L-II, 65.0% and 48.5% vs 86.5% and 80.8%; P = .004). In multivariate analysis, significant prognostic factors were preoperative HA ≥500 ng/mL (P = .004) and graft to recipient body weight ratio <0.8 (P = .042).

Preoperative HA level can be a prognostic risk factor. Patients with high HA levels are vulnerable and should be carefully managed after LDLT.

Background & Aims: Neovascularization, which is vital to the healing of injured tissues, recently has been found to include both angiogenesis, which involves in mature endothelial cells, and vasculogenesis, involving endothelial progenitor cells. The aim of this study was to clarify the possible roles of endothelial progenitor cells during postnatal liver regeneration. Methods: To determine how endothelial progenitor cells participate in liver regeneration, human or mouse endothelial progenitor cells were transplanted into the mice with carbon tetrachloride-induced acute liver injury. Survival rate of the mice in endothelial progenitor cell-transplanted and control groups was calculated. Separately, livers removed temporally from both groups were examined. Results: At an early stage, transplanted human endothelial progenitor cells were seen mainly surrounding hepatic central veins where hepatocytes showed extensive necrosis; later, the transplanted cells formed tubular structures. More of these cells were observed along hepatic sinusoids. Transplantation of human or mouse endothelial progenitor cells improved survival of the mice following liver injury (from 28.6% to 85.7%, P < .0005 and from 33.3% to 80.0%, P < .001, respectively), accompanied by greater proliferation of hepatocytes. Human endothelial progenitor cells produced several growth factors, such as hepatocyte growth factor, transforming growth factor-α, heparin-binding epidermal growth factor–like growth factor, and vascular endothelial growth factor, and also elicited endogenous growth factors. Conclusions: Endogenous and exogenous growth factors and direct neovascularization after endothelial progenitor cell transplantation promoted liver regeneration, thus improving survival after liver injury. Transplantation of endothelial progenitor cells could represent a new therapeutic strategy for promoting liver regeneration.

Measurement of serum human hepatocyte growth factor (HGF) by enzyme-linked immunosorbent assay (ELISA) is useful for the early diagnosis and prediction of prognosis of patients with acute liver failure (ALF). This ELISA methodology, however, is neither rapid nor convenient for use at the bedside. In this study, we have developed a rapid semi-quantitative immunochromatographic (IC) assay and evaluated its usefulness in assessing patients with acute hepatic injury. Only 100 μl of serum is required; the assay can be easily completed in 20 min. The values obtained using this novel assay correlated well with the values obtained using the standard ELISA protocol. In addition, the values obtained in the IC assay correlated with clinical course; increased serum HGF levels were associated with an increased frequency of ALF and death. These results indicate that this rapid semi-quantitative IC assay for HGF is useful for the early diagnosis of ALF and prediction of clinical outcome in acute hepatic injury.

We investigated the direct effect of hepatocyte growth factor (HGF) on the expression of type I collagen in normal and scleroderma dermal fibroblasts, and analyzed the mechanisms underlying the effect in vitro. HGF did not change the protein expression of type I procollagen in the medium of normal human fibroblasts, whereas it reduced the expression in scleroderma fibroblasts. But mRNA levels and the promoter activity of α2(I) collagen gene were not significantly affected by HGF in either of the cells. On the other hand, matrix metalloproteinase-1 expression or activity was increased by HGF in both cells, but HGF had stronger effects in scleroderma fibroblasts than normal fibroblasts. Scleroderma fibroblasts overexpressed c-met protein, the receptor for HGF. The overexpression in scleroderma fibroblasts was abolished by the addition of antisense transforming growth factor (TGF)-β1 oligonucleotide. Our study indicated that HGF may reduce type I collagen accumulation only in scleroderma fibroblasts by enhancing collagenolysis activity, probably because of the overexpression of c-met because of autocrine TGF-β signaling. Thus, further investigation of the effects of HGF on collagen metabolism may contribute to the treatment of fibrosis in scleroderma.

Hepatocyte growth factor (HGF) stimulates liver regeneration and has the potential to be a therapeutic agent for fatal liver diseases, including fulminant hepatic failure and liver cirrhosis. In this study, we investigated the pharmacokinetics of recombinant human HGF, which will be soon available for clinical applications. When recombinant human HGF (0.1 mg/kg) was administered intravenously to normal rats, serum levels of human HGF increased to 89.7 ± 20.6 ng/ml 5 min after the bolus injection, followed by a decrease with a half-life of 2.4 min. Recombinant HGF administered intravenously was distributed primarily to the liver and induced c-Met tyrosine phosphorylation in liver tissues. In comparison, rats given recombinant human HGF via the portal vein exhibited lower serum HGF and an increase in hepatic distribution. Additionally, when compared with normal rats, those with 70% partial hepatectomy or liver cirrhosis showed an increase in serum levels of human HGF with a prolonged half-life. These results suggest that, despite a short half-life, bolus injection of recombinant human HGF may induce therapeutic effect in patients with fatal live disease, and that the dose of this recombinant protein should be modulated according to the degree of liver injury.