The Circulating IGF System in Hepatocellular Carcinoma: The Impact of Liver Status and Treatment

Highlights

• HCC is a condition characterized by IGF-II overproduction

• We therefore measured intact and pro-IGF-II in serum and well as IGF bioactivity in patients with HCC before and after treatment

• HCC patients were compared with patients with liver cirrhosis patients and healthy controls

• Our study showed that liver status has paramount effect on the circulating IGF-system, whereas the impact of HCC in much less

• Thus, our data question the use of IGF-II as a marker of HCC.

Abstract

Background

Previous studies have demonstrated an elevated IGF-II mRNA expression and protein levels in tumors and blood from patients with hepatocellular carcinoma (HCC), hereby suggesting a role of IGF-II as a pathogenic marker of HCC. We hypothesized that in HCC, an increased IGF-II secretion would translate into an elevated circulating IGF bioactivity, which would normalize following treatment.

Methods

Patients with HCC (n = 39) were studied before and after radio-frequency ablation and/or transarterial chemo-embolization. Baseline data were compared to healthy subjects (n = 150) and patients with liver cirrhosis (n = 41). Serum levels of IGF ligands and IGF binding proteins (IGFBPs) were determined using gold standard methods as well as novel assays and compared to liver function tests and HCC treatment status.

Results

At baseline, HCC patients differed from cirrhosis patients and healthy controls regarding IGF-I (29 [23–37] vs. 12 [7–19] vs. 109 [103–116] μg/l), IGF-II (254 [224–288] vs. 118 [102–137] vs. 545 [525–566] μg/l) and IGF bioactivity (0.53 [0.41–0.68] vs. 0.29 [0.24–0.34] vs. 1.43 [1.33–1.53] μg/l) (mean [95% confidence interval], all age-adjusted P < 0.001). All variables but IGFBP-2 were strongly associated with liver status (MELD score), and accordingly, differences were either attenuated or disappeared when adjusted for MELD score. There was no effect of treatment on any IGF variables.

Conclusions

The marked differences in IGF and IGFBP levels between patients with HCC, liver cirrhosis and healthy subjects are mainly explained by variations in liver status. Therefore, this study questions the clinical utility of circulating IGF variables as markers of HCC.

Introduction

Hepatocellular carcinoma (HCC) is a common disease on a global scale and associated with chronic viral hepatitis, exposure to dietary aflatoxin B1, and liver cirrhosis [1]. Irrespective of etiology, the common trait in HCC consists of an imbalance between pro-apoptotic and anti-apoptotic signaling. Anti-apoptotic signaling in hepatocytes is maintained through the PI3-K/Akt and Ras/Raf/MEK/MAPK pathways [2], two major downstream targets of the IGF-I receptor (IGF-IR) [3]. As both IGF-I and IGF-II activate the IGF-IR, both growth factors may influence hepatocyte survival and enrich hepatocellular cancer stem cell populations [4].

The physiology, pathophysiology, and clinical utility of IGF-II have recently been reviewed [5]. Excess IGF-II stimulates cancer cell proliferation and increases cancer cell survival both in vitro and in vivo [6], including in HCC cells [7]. IGF-II is essential for intrauterine growth and placental development. Within the first weeks after birth, there is a shift from potent fetal promoters to a weaker adult promoter, resulting in a very low expression of the igf2 gene in the normal adult liver [8]. By contrast, an elevated igf2 gene expression has been demonstrated in neoplastic tissues from patients with HCC and in animal models of HCC [8], and this over-expression is linked to reactivation of the fetal promoters and deactivation of the adult promoter [9]. Indeed, reducing IGF-II mRNA levels with anti-sense oligonucleotides resulted in decreased cell proliferation in human hepatoma cell lines with high IGF-II secretion [10]. At the molecular level hepatitis B virus (HBV) and hepatitis C virus (HCV) increase the production of IGF-II, thus promoting growth and viability of malignant cells [8].

Several clinical studies have demonstrated elevated IGF-II levels (mRNA and protein) in both tumors and blood from HCC patients, hereby suggesting a role of IGF-II as a marker of HCC [11], [12], [13], [14], [15]. By contrast, others have found sparse evidence of an association between IGF-II and HCC [16]. The reasons for this discrepancy have not been thoroughly investigated. Furthermore, it remains to be clarified whether serum IGF-II may be used as a biomarker of the response to HCC treatment.

We hypothesized that IGF-II would be elevated in HCC and normalize with radio-frequency ablation (RFA) or transarterial chemo-embolization (TACE). IGF-II exerts its biological effects via interaction with the IGF-IR [3], and accordingly, we found it of interest to investigate the ability of serum from HCC patients to activate the IGF-IR under physiological conditions. To this end, an IGF-IR kinase receptor activation (KIRA) assay, developed and validated in our laboratory [17], was applied to compare serum obtained from well-characterized patients with HCC, patients with liver cirrhosis and healthy control subjects. This measure of IGF bioactivity was compared with measurements of IGF-I, IGF-II, and pro-IGF-II, using size exclusion gel chromatography at low pH, which is the gold standard method to eliminate IGFBP interference [18]. Finally, we assessed the impact of the marked reduction in tumor burden following RFA and TACE of HCC tumors on the circulating IGF system.