Hepatocyte growth factor levels in livers and serum at Kasai-portoenterostomy are not predictive of clinical outcome in infants with biliary atresia

Growth Factors Assessed during Kasai Procedure in Liver and Serum Are Not Predictive for the Postoperative Liver Deterioration in Infants with Biliary Atresia

Background: Biliary atresia (BA) is a neonatal cholangiopathy characterized by progressive destruction of the biliary system resulting in liver cirrhosis. Residual bile drainage can temporarily be achieved through Kasai portoenterostomy (KPE) and some children show long-term survival with their native liver. However, most children eventually require liver transplantation (LTX). As several growth factors (GF) and chemokines have been shown to promote fibrogenesis in the liver, we assessed whether GF are predictive for the course of disease. Material and Methods: Liver and sera samples were collected from 49 infants with BA during KPE. Levels of 13 different GF were measured by multiplex immunoassay. Patient outcomes were stratified into favorable (bilirubin < 20 µmol/L at 2-year follow-up) and unfavorable (LTX). GF levels were compared between groups by a t-test, correlation coefficients were calculated, and principal component analyses performed. Results: Twenty-two patients showed a favorable and 27 an unfavorable disease course. No relation of GF and outcome could be established. In both groups, high levels of SDF-1alpha/CXCL12 (1473.0 ± 497.5 pg/mL), FGF2 (301.2 ± 207.8 pg/mL), and VEGF-a (209.0 ± 146.4 pg/mL) levels were measured within the liver, followed (in descending order) by PDGF-bb, LIF, GM-CSF, BDNF, VEGF-d, beta-NGF, IL-7, SCF, PIGF-1, and EGF. Serum marker levels showed much higher mean variation compared to hepatic values and no correlation to the protein microenvironment in the liver. Conclusions: Our study demonstrates high amounts of GF in livers from infants with BA at KPE, but no correlation to the outcome or serum values could be established. Our data suggest that local or systemic GF levels are unsuitable for prediction of the disease course. Collectively, we conclude that in BA the degree of proliferative activity caused by GF is a dismissible factor for the further course of disease.

Hepatic expression of HGF/C-met and native liver survival in biliary atresia

BACKGROUND

The prognosis of biliary atresia (BA) remains difficult to predict. This study evaluated the roles of hepatocyte growth factor (HGF) and its receptor (C-met) towards clinical outcome and native liver survival.

METHODS

Hepatic HGF and C-met expression were determined using immunohistochemistry from liver biopsies of 41 BA patients during Kasai operation, and 17 non-cholestatic patients. The HGF and C-met expression was visually scored as per its intensity and percentage of stained area. BA patients were classified as high- and low-HGF and C-met receptor status. Native liver survival was compared between the two groups at 3-year follow-up. Data are shown as median and range.

MAIN RESULTS

Median age of BA patients was 2 (1-6) months. Hepatic HGF and C-met staining scores of BA patients were higher than those of non-cholestatic patients (P < 0.0001). There was a correlation between HGF and C-met staining scores (spearman r = 0.77, P < 0.0001). However, there was no association between their expression and early outcome at 6 months post-op. Mean follow-up time was 68.6 months. Survival analysis revealed that native liver survival at 1 year and 3 years were 88% and 77%, respectively. Additionally, 82.6% (19/23) of patients in the low-HGF group survived with native liver, compared with 66.7% (10/15) of those in high-HGF group (P = 0.436). For C-met expression, 78.6% (22/28) of low-score and 70% (7/10) of high score groups survived with native liver (P = 0.673).

CONCLUSIONS

Strong expression of hepatic HGF and its receptor in BA patients was demonstrated. However, the expression was not associated with the early outcome and native liver survival. These results suggest that HGF involved in the liver pathology of BA but its expression cannot be used as a prognostic indicator. Small sample size of patients was a main limitation. Further studies are warranted to validate our findings.