This article is mentioned in:

Introduction

Hepatocellular carcinoma (HCC) is the fifth most common malignancy worldwide and the third most frequent cause of global cancer-associated mortality (1–3). Curative therapies for HCC include surgical resection and liver transplantation; however, <30% of patients are candidates for curative surgery (4,5). HCC is a highly aggressive tumor that is characterized by high-grade malignancy, early metastasis, infiltrating growth and a poor prognosis, even following liver transplantation (6–8). Furthermore, treatment options using systemic therapies are very limited. In general, HCC does not respond to classical chemotherapeutics. Due to the extensive molecular and genotypic heterogeneity of HCC, suitable biomarkers for surveillance, diagnosis and prediction of prognosis in patients with HCC remain to be identified, and are currently not ready for introduction into clinical practice. Therefore, novel markers and therapeutic targets are urgently required to develop effective methods for early detection and prognosis of HCC, as well as for therapies to treat advanced HCC.

Transarterial chemoembolization (TACE) is a non-curative, but useful, palliative treatment option for nonsurgical patients with preserved liver function and large or multinodular non-invasive HCC (9). It is also used as a bridging strategy to limit tumor growth during the waiting times for liver transplantation (10,11). Intra-arterial infusion of a chemotherapeutic substance with a viscous emulsion, followed by the embolization of blood vessels with an embolic agent, results in a cytotoxic effect with ischemia (12). TACE has been reported to be effective in downstaging of cancer, leading to improved overall and recurrence-free survival following liver transplantation (13–16). Repeatedly performed TACE has been reported to be capable of selecting biologically favorable tumors and appears to reflect biological properties, such as tumor aggressiveness (17). At present, little is known regarding the molecular mechanisms that are induced during TACE treatment, and TACE-induced prognostic markers may influence the prognosis of HCC following liver transplantation. The identification of useful predictive markers for the prognosis of TACE therapy may be helpful in improving local tumor control, in order to better select patients suitable for liver transplantation, to reduce recurrence, and to prolong survival and quality of life for patients who remain unsuitable for resection (12).

The insulin-like growth factor (IGF) system is physiologically involved in the regulation of cellular proliferation and apoptosis, and is associated with tissue growth (18). The IGF axis is dysregulated in numerous types of cancer and is considered a key driver in hepatocarcinogenesis (19). Whereas the role of the ligands IGF1 and IGF2, as well as the IGF1 receptor (IGF1R) signaling pathway in hepatocarcinogenesis, has been the focus of numerous studies (20–26), few data exist regarding the specific role of the mannose 6-phosphate/IGF2 receptor (IGF2R, CD222). Many types of human cancer, including colorectal carcinoma (27), breast cancer (28) or HCC (29), are associated with a reduced IGF2R function. IGF2R is a transmembrane protein, which is predominantly located in the Golgi apparatus and pre-endosomal compartments (90%), and to a lesser extent at the cell surface (10%). It is ubiquitously expressed in tissues, but also present in the circulation (30,31). IGF2R binds IGF2, as well as proteins bearing mannose 6-phosphate residues (e.g. lysosomal proteins) at distinct sites on the receptor (32). The receptor participates in the internalization and lysosomal degradation of IGF2, a mitogen that normally acts through IGF1R to stimulate cell proliferation (33). As a cell surface protein, IGF2R lacks a tyrosine kinase domain, and was therefore originally assumed to solely act as a scavenger receptor lacking intrinsic signaling (34). Nonetheless, a few studies concerning cardiac pathophysiology (35–37) have indicated that IGF2R contains a putative G-protein binding site within its cytoplasmic domain. Therefore, IGF2R may not only function in degradation of IGF2, but may also trigger an intracellular signaling pathway through coupling with G-protein-coupled receptors. IGF2R also activates the latent precursor of transforming growth factor-β1 (TGF-β1), which is a potent growth inhibitor for several cell types (38,39).

The IGF2R gene is located on chromosome 6q26 in a large 140 kb locus, and is comprised of 48 exons and introns. There is considerable evidence supporting the importance of a genetic predisposition in patients with HCC (40). Numerous single nucleotide polymorphisms (SNPs) have been identified; however, their impact on HCC has yet to be fully elucidated (41). Notably, the SNP rs629849 (transition G → A at position +1,619) in exon 34, which is localized in the IGF2-binding domain, has been reported to be implicated in IGF2-dependent growth (42). Furthermore, the present study analyzed the role of the SNP rs642588 (transition C → T), which is located in the CTCF-binding site of the IGF2R promoter.

These previous findings have suggested that IGF2R possesses various growth inhibitory functions and is therefore considered a candidate tumor suppressor (43). The present study aimed to elucidate the effects of IGF2R expression on survival and tumor recurrence in patients with HCC. The expression levels of IGF2R were measured in HCC and corresponding non-neoplastic tumor-surrounding tissue (TST), and its association with clinicopathological parameters and outcomes was determined. To the best our knowledge, it is currently unknown as to whether the expression and regulation of IGF2R is altered following TACE. Therefore, the effects of TACE pretreatment on IGF2R mRNA expression were investigated. The possible effects of IGF2R gene polymorphisms (rs629849 and rs642588), and their combination, on susceptibility and on the clinicopathological features of HCC were also analyzed.

Materials and methods

Patient tissue samples

HCC tumor samples and corresponding TST were obtained from 92 patients undergoing tumor resection (n=66) or liver transplantation (n=26) between March 2007 and December 2013 at the Department of Hepatobiliary and Transplantation Surgery and the Department of General and Abdominal Surgery (Johannes Gutenberg University Mainz, Mainz, Germany). A total of 26 patients underwent TACE prior to surgery. TACE was performed at 6-week intervals using mitomycin and lipiodol. Written informed consent was obtained from each patient, according to the agreement on transfer and scientific use of excess material of the University Medicine of the Johannes Gutenberg University Mainz. The present study followed the ethical guidelines of the Declaration of Helsinki and was approved by the Ethics Committee of the State of Rhineland-Palatinate Medical Board [Number 847.243.17 (11077)]. Liver tissues were immediately flash frozen following resection and were stored in liquid nitrogen prior to analysis. All cases of HCC were diagnosed or confirmed by histology. Healthy normal and non-cirrhotic control liver tissues (n=31) were obtained from patients undergoing liver surgery for hepatic metastases following colon or breast carcinoma (male, n=21; female, n=10). The median age of the control patients was 63 years old (range, 47–85 years) at the time of surgery.

Tissue homogenization, RNA isolation and reverse transcription-polymerase chain reaction (RT-PCR) analysis

Prior to RNA extraction, 20 mg tissue samples were homogenized using a Precellys 24 homogenizer (Bertin Instruments, Montigny-le-Bretonneux, France) in CK14 tubes at 5,100 rpm for 2×20 sec. RNA was extracted from tissue samples using the PeqGOLD Total RNA kit (VWR International GmbH, Darmstadt, Germany). cDNA was prepared from 1 µg total RNA (20 µl total volume) using the qScript™ XLT cDNA SuperMix (Quantabio, Beverly, MA, USA). All aforementioned kits were used according to the manufacturers' protocols. Semi-quantitative analysis of IGF2R transcripts was performed by RT-PCR. The Absolute Blue QPCR SYBR-Green Mix kit (Thermo Fisher Scientific, Inc., Waltham MA, USA) and the following primers were used: GAPDH, forward 5′-TTTTGCGTCGCCAGCCGAG-3′, reverse 5′-ACCAGGCGCCCAATACGACC-3′; and IGF2R, HS_IGF2R_1_SG Quantitect Primer Assay (Qiagen GmbH, Hilden, Germany). Cycling conditions were as follows: Initial denaturation at 15 min for 95°C, followed by 45 cycles of denaturation at 95°C for 10 sec, annealing at 66°C for GAPDH and 55°C for IGF2R for 30 sec, and elongation at 72°C for 30 sec. Samples were run on a LightCycler® 480 Real-Time PCR system (Roche Diagnostics GmbH, Mannheim, Germany). PCR products were analyzed using the QIAxcel capillary gel electrophoresis system (Qiagen GmbH). The relative expression levels of IGF2R mRNA in HCC and TST samples were calculated by normalization to GAPDH gene expression using LightCycler® 480 Software Release 1.5.0 (Roche Diagnostics GmbH). For examination of IGF2R mRNA regulation, relative IGF2R mRNA expression in HCC tissues was compared with relative IGF2R mRNA expression in the corresponding TST.

Immunohistochemistry

Immunohistochemical staining was performed on formalin-fixed, paraffin-embedded tissue sections (size, 4 µm). For fixation, the tissue sections were incubated with 4% formalin at room temperature for 48 h. Following deparaffinization and rehydration, endogenous peroxidase activity was inhibited with 4% hydrogen peroxide in methanol for 30 min at room temperature. For antigen retrieval, tissue sections were incubated with 10 mM citrate buffer (pH 6.0) for 20 min in a steamer. Cells were permeabilized with 2% saponin (cat. no. 47036; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) in PBS for 20 min at room temperature. For blocking of non-specific antibody binding, tissues were incubated for 30 min with protein blocking buffer [5% normal goat serum (Dako; Agilent Technologies, Inc., Santa Clara, CA, USA), 0.2% Triton X-100, 2% bovine serum albumin (BSA; SERVA Electrophoresis GmbH, Heidelberg, Germany)]. The following primary antibody was used for immunohistochemistry: Rabbit polyclonal anti-human IGF2R antibody (cat. no. NBP1-19465; Novus Biologicals, LLC, Littleton, CO, USA), which was used at a dilution of 1:100, was used to incubate the sections overnight at 4°C in PBS (2% saponin, 5% BSA, 5% normal serum). For control sections, the specific primary antibody was omitted. The following day, sections were washed three times with Tris-buffered saline-0.5% Tween-20 and incubated for 20 min with Pierce Peroxidase Suppressor (Thermo Fisher Scientific, Inc.). After washing, sections were incubated for 1 h at room temperature with a secondary biotinylated goat anti-rabbit antibody (cat. no. E043201-8; Dako; Agilent Technologies, Inc.) at a dilution of 1:100, and were then treated with the avidin-biotin-peroxidase complex-based Vectastain Elite ABC kit (Vector Laboratories, Inc., Burlingame, CA, USA). Following incubation with horseradish peroxidase-conjugated streptavidin (cat. no. P039701-2; Dako; Agilent Technologies, Inc.) at a dilution of 1:300, tissues were stained at room temperature for 10 min with the liquid DAB+ Substrate Chromogen system (cat. no. K346711-2; Dako; Agilent Technologies, Inc.), according to the manufacturer's protocol. Counterstaining was performed with Gill's hematoxylin solution (Polysciences Inc., Warrington, PA, USA) at room temperature for 7 min and slides were mounted for examination under a light microscope.

Tissue homogenization and DNA extraction

Homogenization of 5–9 mg fresh frozen tissue blocks was conducted using a Precellys 24 homogenizer (Bertin Instruments) in CK14 tubes at 5,100 rpm for 2×20 sec following Proteinase K (Qiagen GmbH) digestion at 56°C for 3 h. Genomic DNA was extracted using the QIAamp DNA Micro kit (Qiagen GmbH), according to the manufacturer's protocol, and was quantified by UV absorption via NanoVue (GE Healthcare Life Sciences, Little Chalfont, UK).

PCR and pyrosequencing (PSQ)

Primer sets with one 5′-biotinylated primer were used for the amplification of SNP regions. All primers used for PCR and sequencing were generated using PyroMark Assay Design Software 2.0 (Qiagen GmbH). PCR was performed using the PyroMark PCR kit (Qiagen GmbH), according to the manufacturer's protocol. The following primers were used: rs629849, forward 5′-biotin-AAATCCGGCCTGAGCTATAAG-3′, reverse 5′-AGCATGAGTCTTGAGCAATTACTG-3′; and rs642588, forward 5′-biotin-CACATGGGGATTATGGGAACT-3′ and reverse 5′-AGCATGAGTCTTGAGCAATTACTG-3′. PCR reactions were run with 0.15 µM primer under the following thermal cycling conditions: 95°C for 15 min, followed by 45 cycles at 94°C for 30 sec, at the optimized primer-specific annealing temperature (60°C) for 30 sec and at 72°C for 30 sec, followed by a final extension step at 72°C for 10 min. Amplification of the correct DNA product was confirmed by high-resolution capillary electrophoresis using a QIAxcel Advanced system (Qiagen GmbH). Subsequently, a standard PSQ sample preparation protocol was applied: Streptavidin beads (1.5 µl; GE Healthcare Life Sciences), 40 µl PyroMark binding buffer (Qiagen GmbH), 20 µl PCR product and 18.5 µl water were mixed and incubated for 10 min at room temperature with agitation (1,200 rpm). According to the manufacturer's recommendations, amplicons were denaturated using PyroMark denaturation solution (Qiagen GmbH), washed with PyroMark Wash buffer (Qiagen GmbH) and added to 20 µl annealing buffer containing 0.375 µM sequencing primer (rs629849: 5′-TGGGCCTATTGGTTG-3′; rs642588: 5′-AGCAATTACTGATTAATATG-3′) using the PyroMark Q24 Vacuum Workstation (Qiagen GmbH) at room temperature. Primer annealing was performed by incubating the samples at 80°C for 5 min and then cooling to room temperature prior to PSQ. PyroMark Q24 Advanced Reagents (Qiagen GmbH) were used for the PSQ reaction, and the signal was analyzed using the Pyromark Q24 system (Qiagen GmbH). Genotype analysis of SNPs was conducted using PyroMark Q24 Advanced Software version 3.0.0 (Qiagen GmbH).

Statistical analysis

Data management and all statistical analyses were performed using SPSS program (version 23.0; IBM Corp., Armonk, NY, USA). For categorical variables, between-group differences were analyzed by χ2 test or Fisher's exact test. Continuous data were expressed as the median and range. Two independent groups were compared using the Mann-Whitney U test with subsequent Bonferroni correction. For the comparison of multiple independent samples, the Kruskal-Wallis test and Bonferroni correction was performed. Overall survival rates were calculated using the Kaplan-Meier method and were compared using the log-rank test. The distributions of genotypic frequencies between cases and controls were analyzed by χ2 test or Fisher's exact test. Odds ratio (ORs) and 95% confidence intervals (CIs) were calculated. Univariate analysis was performed to assess the association between genotype frequencies and the clinicopathological features of HCC. P≤0.05 (two-sided) was considered to indicate a statistically significant difference.

Results

IGF2R mRNA expression in HCC

To analyze the role of IGF2R in HCC, the present study detected the mRNA expression levels of IGF2R in HCC tissue and corresponding non-neoplastic TST (n=92). Semi-quantitative RT-PCR results demonstrated significant differences in IGF2R mRNA expression between HCC and TST (Fig. 1A). IGF2R mRNA was highly expressed in TST and was significantly downregulated in cancerous tissues (P=0.004). In HCC tissues, median IGF2R mRNA expression was reduced by 20.4% compared to that in the corresponding TST. Normal liver served as a control for the expression of IGF2R mRNA. TST exhibited comparable IGF2R mRNA expression to normal healthy liver tissue (n=31; P=1.000). In the present cohort, IGF2R mRNA was downregulated in 61% of HCC samples (n=56), whereas IGF2R mRNA expression was upregulated in 39% (n=36) (Fig. 1B). Details of clinical and pathological characteristics of patients and tumors are summarized in Table I, according to World Health Organization specifications (44).

Figure 1. IGF2R mRNA expression in HCC tissue, corresponding TST and healthy liver tissues (from patients undergoing liver surgery for hepatic metastases). (A) IGF2R mRNA expression was downregulated in HCC tissues (n=92) compared with in the corresponding non-neoplastic TST (P=0.004) and in healthy liver tissues (n=31) (P=0.423). For statistical analysis, Kruskal-Wallis test followed by Bonferroni correction was used. **P<0.01. (B) Individual regulation pattern of IGF2R mRNA expression in HCC (n=92). A total of 56 HCC samples exhibited downregulated IGF2R mRNA expression compared with TST (61%), whereas 36 HCC samples exhibited upregulated IGF2R mRNA expression (39%). HCC, hepatocellular carcinoma; IGF2R, insulin-like growth factor-2 receptor; TST, tumor-surrounding tissue.

Table I. Patients and tumor characteristics.

Table I.

Patients and tumor characteristics.

Characteristics	Value
Total number, n	92
Median follow-up, days (range)	796 (4–2,615)
Median recurrence-free survival, days (range)	529 (4–2,615)
Male/female, n	79/13
Median age, years (range)	68 (35–86)
Nodules (1–3/multiple), n	75/17
Tumor diameter (<3 cm/≥3 cm), n	19/73
Median tumor diameter, cm (range)	5.0 (1.0–30)
T-classification (T1/T2/T3/T4), n	42/28/21/1
Grading (G1/G2/G3/Gxa), n	14/60/13/5
AFP (>100/<100), nb	15/71
Angioinvasion (yes/no), nc	24/67
Cirrhosis (yes/no), n	54/38
Child-Pugh grade (A/B/C), n	42/6/6
Cirrhosis (viral/non-viral), n	27/27
Pretreatment with chemoembolization, n	26
Liver transplantation/resection, n	26/66

a Gx, grade could not be assessed

b the status of six cases was unknown

c the status of one case was unknown. AFP, α-fetoprotein.

Patient survival and tumor recurrence in patients

According to the RT-PCR results, the patients were divided into two groups: i) Patients with HCC that exhibited downregulated IGF2R mRNA expression (IGF2R mRNA expression in HCC/IGF2R mRNA expression in TST <1; n=56), and ii) patients with HCC that exhibited upregulated IGF2R mRNA expression in HCC (IGF2R mRNA expression in HCC/IGF2R mRNA expression in TST ≥1; n=36).

Overall 5-year survival of the patients with downregulated IGF2R mRNA tumor expression was 34%; conversely, it was 66% for patients with upregulated IGF2R mRNA tumor expression. Despite this pronounced difference, statistical significance could not be achieved (P=0.085; Fig. 2A). Significant downregulation of IGF2R mRNA expression markedly affected the risk of tumor recurrence within 5 years following surgery (Fig. 2B). Recurrence-free 5-year survival was 16% in patients with downregulated IGF2R mRNA expression, whereas it was 76% in patients with upregulated IGF2R mRNA expression (P=0.002).

Figure 2. Survival and tumor recurrence of patients with HCC, according to intratumoral IGF2R mRNA up- or downregulation. The ratio of IGF2R mRNA expression in HCC and IGF2R mRNA expression in TST was used as a cut-off value. Patients with HCC were subdivided into an upregulated group (IGF2R mRNA expression HCC/TST ≥1; n=36) and a downregulated group (IGF2R mRNA expression HCC/TST <1; n=56). (A) Patient survival. The association between IGF2R mRNA expression and overall patient survival did not reach statistical significance (P=0.085). (B) Tumor recurrence-free survival. Downregulation of IGF2R mRNA expression in HCC was associated with significantly worse recurrence-free survival (P=0.002). HCC, hepatocellular carcinoma; IGF2R, insulin-like growth factor-2 receptor; TST, tumor-surrounding tissue.

Protein expression of IGF2R, as determined by immunohistochemistry

To identify the localization of IGF2R expression in tumor tissue, protein expression was assessed by immunohistochemistry. As shown in Fig. 3A, IGF2R staining was detected at the tumor border in HCC tissues with downregulated IGF2R mRNA expression. In addition, predominant IGF2R staining in hepatocytes was detected in the non-neoplastic TST, whereas IGF2R staining was almost completely absent in tumor tissue.

Figure 3. Immunohistochemistry of IGF2R protein expression in HCC and corresponding non-neoplastic TST. (A) IGF2R protein expression was detected at the tumor border in a patient with HCC and downregulated IGF2R mRNA expression. IGF2R exhibited pronounced staining of hepatocytes in non-neoplastic TST, whereas it was almost completely absent in tumor tissue. IGF2R expression was stained brown. (B) For the negative control, the specific antibody was omitted during the staining procedure. HCC, hepatocellular carcinoma; IGF2R, insulin-like growth factor-2 receptor; TST, tumor-surrounding tissue.

Patient and tumor characteristics in association with IGF2R mRNA expression

Upregulation of IGF2R mRNA expression in HCC was associated with a better recurrence-free survival (P=0.017), a lower median age (P=0.013) and a higher occurrence of liver cirrhosis (P=0.05) (Table II). TACE pretreatment prior to resection or liver transplantation was carried out more often in the group with upregulated IGF2R mRNA tumor expression (P=0.032). IGF2R mRNA expression was not associated with any other tumor characteristics.

Table II. Patient and tumor characteristics associated with intratumoral IGF2R mRNA expression.

Table II.

Patient and tumor characteristics associated with intratumoral IGF2R mRNA expression.

	IGF2R mRNA
Characteristics	Downregulation (HCC/TST <1)	Upregulation (HCC/TST ≥1)	P-value
Number of patients, n	56	36
Median follow-up, (range)	796 (5–2,316)	826 (19–2,615)	0.247 (n.s.)
Median recurrence-free survival, days (range)	445 (5–2,316)	645 (19–2,615)	0.017
Male/female, n	48/8	31/5	1.000 (n.s.)
Median age, years (range)	69 (35–86)	66 (46–78)	0.013
1–3 nodules/multiple nodules, n	45/11	30/6	n.s. (0.789)
Tumor diameter (<3 cm/≥3 cm), n	9/47	10/26	n.s. (0.196)
Median tumor diameter, cm (range)	5.4 (1.0–20)	4.5 (2.0–30)	n.s. (0.203)
T-classification (T1/T2/T3/T4), n	22/18/16/0	20/10/5/1	n.s. (0.169)
Grading (G1/G2/G3/Gxa), n	9/39/7/1	5/21/6/4	n.s. (0.233)
AFP (>100/<100), nb	10/41	5/30	n.s. (0.577)
Angioinvasion (yes/no), nc	17/38	7/29	n.s. (0.331)
Cirrhosis (yes/no), n	28/28	26/10	0.050
Child-Pugh grade (A/B/C), n	23/2/3	19/4/3	n.s. (0.614)
Cirrhosis (viral/non-viral), n	10/18	13/13	n.s. (0.409)
Pretreatment with chemoembolization (yes/no), no	11/45	15/21	0.032
Liver transplantation/resection, n	11/45	15/21	0.032

a Gx, grade could not be assessed

b the status of six cases was unknown

c the status of one case was unknown. AFP, α-fetoprotein; HCC, hepatocellular carcinoma; n.s., non-significant; TST, tumor-surrounding tissue. Numbers in bold indicate statistical significance (P≤0.05).

Patient and tumor characteristics in association with TACE pretreatment

As shown in Table I, 26 patients in the present cohort were pretreated with TACE prior to liver transplantation or resection, whereas 66 patients did not receive any TACE pretreatment prior to surgery.

The present study analyzed the associations between patient and tumor characteristics, and preoperative TACE treatment; significant differences are presented in Table III. Notably, TACE-pretreated patients exhibited a significantly longer median follow-up time following surgery (P=0.023) and a prolonged median recurrence-free survival time (P=0.002). Furthermore, patients receiving TACE prior to surgery displayed less advanced HCC stages, since TACE-pretreated HCCs presented with smaller tumor diameters (P<0.001), with less advanced T-stage tumors (P=0.034) and with more well differentiated tumors (P<0.001). These findings coincided with a significantly higher expression of IGF2R mRNA expression in the tumor tissues of TACE-pretreated patients (P=0.028; Table III).

Table III. Patient and tumor characteristics associated with preoperative TACE pretreatment.

Table III.

Patient and tumor characteristics associated with preoperative TACE pretreatment.

Characteristic	No TACE	TACE	P-value
Number of patients, n	66	26
Median follow-up, days (range)	724 (5–2,588)	1,198 (85–2,615)	0.023
Median recurrence-free survival, days (range)	418 (5–25,88)	1,127 (85–2,615)	0.002
Male/female, n	55/11	24/2	n.s. (0.337)
Median age, years (range)	71 (35–86)	60 (47–71)	<0.001
1–3 nodules/multiple nodules, n	56/10	19/7	n.s. (0.235)
Tumor diameter (<3 cm/≥3 cm), n	6/60	13/13	<0.001
Median tumor diameter, cm (range)	6.2 (2.0–30)	2.4 (1.0–18)	<0.001
T-classification (T1/T2/T3/T4), n	31/15/19/1	11/13/2/0	0.034
Grading (G1/G2/G3/Gxa), n	8/46/12/0	6/14/1/5	<0.001
AFP (>100/<100), nb	47/13	24/2	n.s. (0.137)
Angioinvasion (yes/no), nc	45/20	22/4	n.s. (0.189)
Cirrhosis (yes/no), n	29/37	25/1	<0.001
Child-Pugh grade (A/B/C), n	27/1/1	15/5/5	<0.001
Cirrhosis (viral/non-viral), n	8/21	15/10	n.s. (0.027)
Liver transplantation/resection, n	2/64	24/2	<0.001
IGF2R mRNA (down-/upregulated), n	45/21	11/15	0.032
Median IGF2R mRNA expression TST	0.53 (0.19–4.3)	0.48 (0.22–2.7)	n.s. (0.233)
Median IGF2R mRNA expression HCC	0.36 (0.71–3.5)	0.49 (0.14–1.1)	0.028
Median IGF2R mRNA expression HCC/TST	0.59 (0.17–4.8)	1.1 (0.27–2.2)	0.019

a Gx, grade could not be assessed

b the status of six cases was unknown

c the status of one case was unknown. AFP, α-fetoprotein; HCC, hepatocellular carcinoma; IGF2R, insulin-like growth factor-2 receptor; n.s., non-significant; TACE, transarterial chemoembolization; TST, tumor-surrounding tissue. Numbers in bold indicate statistical significance (P≤0.05).

Notably, significant downregulation of IGF2R mRNA expression in HCC tissue compared with in TST could only be detected in the group of patients that did not receive TACE pretreatment (P=0.004; Fig. 4A). There was no difference between IGF2R mRNA expression in TST and HCC samples from TACE-pretreated patients (P=1.000). Furthermore, tumor tissues of TACE-pretreated patients exhibited a significantly higher median IGF2R mRNA compared with non-treated patients (P=0.019; Fig. 4B). A higher proportion of tumors in the TACE-pretreated patient group exhibited upregulation of IGF2R mRNA expression (P=0.032). Notably, 58% of TACE-pretreated patients exhibited upregulated IGF2R mRNA expression in HCC compared with only 32% of patients that did not receive TACE prior to surgery (P=0.032; Fisher's exact test; Table IV).

Figure 4. IGF2R mRNA expression in HCC and corresponding TST, and its association with TACE treatment prior to surgery. IGF2R mRNA expression was analyzed in HCC and TST in patients who did not receive TACE (n=66) compared with in TACE-pretreated patients (n=26). (A) IGF2R mRNA expression was significantly downregulated in HCC tumor tissues from the non-TACE-pretreated group (P=0.004). For statistical analysis, Mann-Whitney-U test with subsequent Bonferroni correction was used. (B) TACE-pretreated HCC samples exhibited significantly increased IGF2R mRNA expression compared with in non-pretreated HCC samples (P=0.019). For statistical analysis, Mann-Whitney-U test was used. *P<0.05, **P<0.01. HCC, hepatocellular carcinoma; IGF2R, insulin-like growth factor-2 receptor; TACE, transarterial chemoembolization; TST, tumor-surrounding tissue.

Table IV. Association of IGF2R mRNA expression in HCC with TACE pretreatment.

Table IV.

Association of IGF2R mRNA expression in HCC with TACE pretreatment.

	IGF2R mRNA expression in HCC
Group	Downregulated, n (%)	Upregulated, n (%)	P-value
No TACE	45 (68)	21 (32)	0.032
TACE	11 (42)	15 (58)

[i] HCC, hepatocellular carcinoma; IGF2R, insulin-like growth factor-2 receptor; TACE, transarterial chemoembolization.

Association of TACE pretreatment with patient survival and tumor recurrence

Patients with HCC that received TACE therapy prior to surgery did not exhibit a significantly better overall 5-year survival compared with non-TACE-treated patients (P=0.097; Fig. 5A). Conversely, TACE pretreatment induced a significantly increased recurrence-free 5-year survival in patients with HCC (P=0.001; Fig. 5B). In the TACE group, 80% of TACE-pretreated patients exhibited recurrence-free survival 5 years after surgery, whereas only 32% of patients who were not treated with TACE exhibited no recurrence after 5 years.

Figure 5. Survival and tumor recurrence of patients with HCC, and their association with TACE pretreatment. (A) TACE-pretreated patients with HCC (n=26) did not exhibit significantly improved survival compared with the non-treated patients (n=66). (B) TACE pretreatment of patients with HCC resulted in a significantly better recurrence-free survival (P=0.001). HCC, hepatocellular carcinoma; TACE, transarterial chemoembolization.

IGF2R SNPs as a mechanism for IGF2R mRNA expression

Two SNPs of the IGF2R gene, rs629849 and rs642588, were investigated to analyze their association with HCC pathological characteristics. Frequency distributions were studied in 83 patients with HCC and were compared with those of 100 healthy controls; 12 patients with HCC could not be analyzed due to insufficient HCC tissue for DNA extraction. The alleles with the highest distribution frequency for rs629849 and rs642588 were GG (76.5%) and CC (64.5%), respectively. Differences in the frequencies of IGF2R genotypes were not statistically significant between patients with HCC and healthy controls (P=0.145 and P=0.068, respectively; Table V).

Table V. Distribution frequency of IGF2R genotypes in 100 healthy controls and 83 patients with HCC.

Table V.

Distribution frequency of IGF2R genotypes in 100 healthy controls and 83 patients with HCC.

Characteristic	Control, n=100 (%)	HCC, n=83 (%)	Total, n=183 (%)	P-value
IGF2R rs629849
GG	79 (79.0)	61 (73.5)	140 (76.5)	0.145a
AG	21 (21.0)	19 (22.9)	40 (21.9)
AA	0	3 (3.6)	3 (1.6)
GG	79 (79.0)	61 (73.5)	140 (76.5)	0.388b
AG/AA	21 (21.0)	22 (26.5)	43 (23.5)
IGF2R rs642588
CC	60 (60.0)	58 (69.9)	118 (64.5)	0.068c
CT	38(38.0)	20 (24.1)	58 (31.7)
TT	2 (2.0)	5 (6.0)	7 (3.8)
CC	60 (60.0)	58 (69.9)	118 (64.5)	0.214d
CT/TT	40 (40.0)	25 (30.1)	65 (35.5)

a P-value between the control and HCC groups with regards to GG, AG and AA genotypes, as determined by χ² test

b P-value between the control and HCC groups with regards to GG and AG/AA genotypes, as determined by Fisher's test

c P-value between the control and HCC groups with regards to CC, CT and TT genotypes, as determined by χ² test

d P-value between the control and HCC groups with regards to CC and CT/TT genotypes, as determined by Fisher's test. HCC, hepatocellular carcinoma; IGF2R, insulin-like growth factor-2 receptor.

To estimate the ORs and 95% CIs of each IGF2R SNP in HCC, patients with HCC and healthy controls were classified into two subgroups: Those with at least one mutated allele and those with homozygous wild-type alleles. Results indicated no significant difference in any allele frequency distribution between patients with HCC and healthy controls (Table V). ORs with 95% CIs were estimated for each gene polymorphism for pathological characteristics, including T classification, grading and disease history of liver cirrhosis. The results revealed that at least one mutated A allele in rs629849 IGF2R gene polymorphism had a significantly lower risk for developing non-viral liver cirrhosis (OR=0.25, 95% CI=0.06–0.98) (Table VI). Nevertheless, no significant difference between the IGF2R genotype frequencies tested and any other clinicopathological variables was observed.

Table VI. ORs and 95% CIs of clinical status and IGF2R genotype frequencies in patients with hepatocellular carcinoma (n=83).

Table VI.

ORs and 95% CIs of clinical status and IGF2R genotype frequencies in patients with hepatocellular carcinoma (n=83).

	rs629849				rs642588
Characteristics	GG n (%)	AG or AA n (%)	OR (95% CI)	P-value	CC n (%)	CT or TT n (%)	OR (95% CI)	P-value
Number of patients, n	61	22			58	25		0.21
T classification
T1/T2	45 (73.8)	18 (81.8)	0.63 (0.18–2.13)	0.57	43 (74.1)	20 (80.0)	0.72 (0.23–2.25)	0.78
T3/T4	16 (26.2)	4 (18.2)			15 (25.9)	5 (20.0)
Gradinga
G1/G2	51 (85.0)	16 (84.2)	1.06 (0.26–4.40)	1.00	44 (81.5)	23 (92.0)	0.38 (0.08–1.90)	0.32
G3	9 (15.0)	3 (15.8)			10 (18.5)	2 (8.0)
Cirrhosis
Negative	25 (41.0)	9 (40.9)	1.00 (0.37–2.70)	1.00	25 (43.1)	9 (36.0)	1.35 (0.51–3.55)	0.63
Positive	36 (59.0)	13 (59.1)			33 (56.9)	16 (64.0)
Cirrhosis
Viral	13 (36.1)	9 (69.2)	0.25 (0.06–0.98)	0.05	16 (48.5)	6 (37.5)	1.57 (0.46–5.32)	0.55
Non-viral	23 (63.9)	4 (30.8)			17 (51.5)	10 (62.5)
Child-Pugh grade
A	28 (77.8)	11 (84.6)	0.64 (0.12–3.48)	0.71	25 (75.8)	14 (87.5)	0.45 (0.08–2.40)	0.46
B or C	8 (22.2)	2 (15.4)			8 (24.2)	2 (12.5)
Tumor diameter (cm)
<3	10 (16.4)	6 (27.3)	0.52 (0.16–1.66)	0.35	11 (19.0)	5 (20.0)	0.94 (0.29–3.05)	1.00
≥3	51 (83.6)	16 (72.7)			47 (81.0)	20 (80.0)
Nodules
1-3	49 (80.3)	19 (86.4)	0.65 (0.16–2.54)	0.75	48 (82.8)	20 (80.0)	1.20 (0.36–3.96)	0.76
Multiple	12 (19.7)	3 (13.6)			10 (17.2)	5 (20.0)
Angioinvasionb
Negative	46 (75.4)	13 (61.9)	1.89 (0.66–5.43)	0.27	39 (68.4)	20 (80.0)	0.54 (0.18–1.67)	0.42
Positive	15 (24.6)	8 (38.1)			18 (31.6)	5 (20.0)
AFPc
<100	44 (81.5)	18 (81.8)	0.98 (0.27–3.53)	1.00	43 (81.1)	19 (82.6)	0.91 (0.25–3.25)	1.00
>100	10 (18.5)	4 (18.2)			10 (18.9)	4 (17.4)
Surgery
Liver transplantation	16 (26.2)	6 (27.3)	0.95 (0.32–2.84)	1.00	16 (27.6)	6 (24.0)	1.21 (0.41–3.57)	0.79
Resection	45 (73.8)	16 (72.7)			42 (72.4)	19 (76.0)
Pretreatment
No TACE	46 (75.4)	14 (63.6)	1.75 (0.62–4.99)	0.41	42 (72.4)	18 (72.0)	1.02 (0.36–2.91)	1.00
TACE	15 (24.6)	8 (36.4)			16 (27.6)	7 (28.0)

a The status of four cases could not be assessed (Gx)

b the status of one case was unknown

c the status of seven cases was unknown. AFP, α-fetoprotein; CI, confidence interval; IGF2R, insulin-like growth factor-2 receptor; OR, odds ratio. Numbers in bold indicate statistical significance (P≤0.05).

The present study also estimated the association of various combinations of these IGF2R SNPs with HCC susceptibility. As shown in Table VIII, no significant difference was observed in SNPs between healthy controls and HCC patients.

Discussion

It has been suggested that IGF2R may act as a tumor suppressor; therefore, the present study evaluated its role as a biomarker in the pathology of HCC. To the best of our knowledge, the present study is the first to analyze the expression profile of IGF2R in a large series of human HCC samples, and to investigate its association with clinical and tumor-specific data. In the human HCC samples analyzed in the present study, the mRNA expression levels of IGF2R were significantly downregulated in cancerous tissue compared with in the corresponding non-neoplastic TST. In addition, downregulation of IGF2R protein was detected in HCC tissues from one patient, which indicated that the results of RT-PCR were concordant with the protein expression experiment. The present study predominantly focused on the role of IGF2R mRNA expression as a potential marker in the pathogenesis of HCC; therefore, solely HCC tissue and TST samples were collected. One of the limitations of the present study is that tissue samples containing a tumor border, which would have been required for complete immunohistochemical analysis and subsequent statistical analysis, were not available. Further experiments are required to confirm the downregulation of IGF2R protein in HCC. The loss of IGF2R has already been described in some tumor types and its expression is associated with tumor suppression (28,45–47). Chen et al revealed that a decreased expression of IGF2R in human breast cancer cells, via infection with an adenovirus carrying a ribozyme targeted against IGF2R mRNA, enhances IGF2-induced proliferation and reduces susceptibility to tumor necrosis factor-induced apoptosis (28). In addition, radioimmunotherapy targeted to IGF2R in osteosarcoma cells suppresses tumor growth in a murine osteosarcoma xenograft model (45). Transfection experiments using breast cancer cells and choriocarcinoma cells demonstrated that overexpression of IGF2R decreases cellular growth rates in vitro and decreases tumor growth in nude mice (46,47). Whether the reduction of IGF2R levels in HCC may provide HCC cells with an important selective growth advantage should be investigated further and has not been reported in the present study. Furthermore, this study demonstrated that patients with upregulated IGF2R mRNA expression were younger than those with downregulated expression; currently, we do not have any explanation for this. Therefore, it requires confirmation in further studies using a different and larger cohort. Nevertheless, the present study demonstrated a strong association between IGF2R mRNA expression and the risk of tumor recurrence within 5 years following surgery, indicating a functional consequence of IGF2R expression for patients with HCC. However, whether IGF2R serves a direct role in tumor suppression, or merely an indirect role as a transporter for ligands designated for degradation in the lysosomes, remains to be further elucidated.

In order to increase the number of informative cases, both resected and transplanted livers were included for analysis, and approximately one-third of patients in the present cohort were pretreated with TACE prior to surgery. These patients most likely exhibited a good response to TACE, leading to subsequent liver transplantation in the majority of cases. The present study demonstrated that the TACE-pretreated patients possessed significantly higher mRNA expression levels of IGF2R compared with non-treated patients. To the best of our knowledge, this study is the first to describe this relationship; however, at present, it cannot answer whether IGF2R mRNA expression is directly influenced by the TACE procedure, because IGF2R mRNA expression was solely analyzed in liver tumor explants following TACE. It may be speculated that during TACE some factor could be induced, which may be responsible for disease stability. IGF2R may be considered a good candidate gene to further analyze in this context, as this study revealed that TACE-pretreated patients with HCC had a significantly higher expression of IGF2R mRNA compared with non-pretreated patients. It would be interesting to know whether tumors that respond to TACE therapy possess an overexpression of IGF2R mRNA already prior to TACE therapy or whether IGF2R mRNA expression is induced by TACE, e.g. by hypoxia- and/or chemotherapy-induced mechanisms. To analyze that, the expression of IGF2R mRNA in TACE-treated HCC samples has to be compared with pre-TACE biopsies taken at the time of diagnosis. This approach could determine whether IGF2R mRNA is induced by the TACE procedure or not.

Otto et al (48) proposed that TACE pretreatment may select patients with biologically less aggressive tumors. In this previous study, patients who experienced tumor response to TACE had a significantly prolonged disease-free survival compared to those with tumor progression during TACE. These findings are in accordance with the present study, which demonstrated that the TACE-pretreated patient cohort exhibited a significantly increased recurrence-free 5 year survival. Furthermore, Otto et al reported that the freedom of recurrence was not influenced by classification of the patient according to the Milan criteria or by downstaging, but rather by the stability of the disease during pretreatment with TACE (48). At present, it has not yet been analyzed as to whether IGF2R mRNA expression is associated with the prognosis of HCC following TACE pretreatment. Further investigations are required to address whether IGF2R mRNA expression before or after TACE is associated with the prognosis of HCC.

The molecular mechanisms that are induced through TACE therapy remain to be completely elucidated. It is well known that cells under stress, such as hypoxia, promote IGF2R expression (49). Whether IGF2R expression in HCC is affected by TACE-induced hypoxia remains to be analyzed. Nevertheless, the present data indicated that IGF2R may be a promising candidate marker, which may serve a role in TACE-induced hypoxia, as well as in tumor response. Most hypoxia-induced pathways not only promote tumor growth, but also induce apoptosis (50). Since IGF2R induces apoptosis, this is an interesting aspect to consider. If it could be shown that IGF2R is induced through hypoxia, and that IGF2R expression is associated with the prognosis of HCC following TACE pretreatment, this would be a further goal in improving TACE treatment by potentially increasing the proportion of patients able to undergo liver resection, reduce recurrence, and prolong survival and quality of life of patients who remain unsuitable for resection (12).

The genotypes of the SNPs rs629849 (located in the binding site for IGF2) and rs642588 (located in the CTCF-binding site of the IGF2R promoter) were not significantly associated with HCC risk in the present study. Weng et al (41) and Rashad et al (51) reported that the combination of IGF2 rs10840452 (AA) and IGF2R rs629849 (GG) homozygosity exhibited a significant protective effect against HCC occurrence. The authors concluded that IGF2 and IGF2R polymorphisms are significant IGF-system-associated factors in HCC development. The present findings concerning the SNP rs629849 are supported by Rezgui et al (42); this previous study analyzed the structure and function of the human SNP rs629849 and revealed that it fails to alter gene expression, protein half-life and cell membrane distribution of IGF2R, thus suggesting that the polymorphism has no direct effect on receptor function. Furthermore, the comparison of binding kinetics of ‘wild-type’ and ‘mutated’ (rs629849) IGF2R to IGF2 revealed no differences in ‘on’ and ‘off’ rates, concluding that the rs629849 polymorphism is non-functional.

Notably, Rashad et al (51) demonstrated that the homozygous IGF2R rs629849 GG genotype is significantly associated with worse Child-Pugh grades. In the present study, no significant association was detected between the SNPs tested and the Child-Pugh classification grade. However, patients with a homozygous IGF2R rs629849 GG genotype were associated with a significantly elevated risk of non-viral liver cirrhosis (P=0.05). Furthermore, upregulation of IGF2R mRNA expression was associated with a higher occurrence of cirrhosis in patients with HCC (P=0.05) de Bleser et al (52) revealed that IGF2R is upregulated during liver fibrosis, which may be caused by the overexpression of IGF2R in hepatic stellate cells, which are the major cell type involved in liver fibrosis (53). Fibrogenesis is stimulated by TGF-β1, which is known to be activated by IGF2R (38). Activated stellate cells are also responsible for secreting collagen scar tissue, which can lead to cirrhosis (53). These findings support the role of IGF2R in fibrogenesis and the development of cirrhosis. However, data reporting the influence of rs642588 on the functionality of the CTFC-binding site in the IGF2R promoter region are currently not available.

The present study suggested an important role for IGF2R expression in HCC, particularly with regards to TACE pretreatment. These findings indicated that IGF2R may be considered a good candidate for further investigation of TACE inducibility and as a marker for improved recurrence-free patient survival. Furthermore, this study suggested a pivotal role for IGF2R in the development of liver cirrhosis. Further studies are required to investigate the precise mechanisms underlying the effects of IGF2R on the progression of HCC, and to evaluate possible diagnostic and therapeutic consequences.

Acknowledgements

The authors would like to thank Mrs. Larissa Herbel (1st Department of Internal Medicine, University Medicine of the Johanes Gutenberg University Mainz) and Mrs. Ulrike Suessdorf (Department of General, Visceral and Transplantation Surgery, University Medicine of the Johannes Gutenberg University Mainz), for excellent technical assistance.

Funding

Research funding from the University Medicine of the Johannes Gutenberg University Mainz for AL, FS and TZ was used for this study.

Availability of data and materials

The datasets used and/or analyzed during the present study are available from the corresponding author on reasonable request.

Authors' contributions

AL, FS and TZ made major contributions to the conception, analysis and interpretation of the data, and were major contributors in writing, drafting and revising the manuscript. HL, GO, JM, CD and PRG collected the clinical samples and the corresponding clinical data. AL, FS, JV and MHL performed the experiments and analyzed data. MHL conducted the clinical data collection and performed the statistical analysis. AS performed the histological evaluation. HL, TZ, GO, PRG and CD critically revised the manuscript. All authors read and approved the final version of the manuscript, and agree to be accountable for all aspects of the research in ensuring that the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Ethics approval and consent to participate

The present study followed the ethical guidelines of the Declaration of Helsinki and was approved by the Ethics Committee of the State of Rhineland-Palatinate Medical Board [Number 847.243.17 (11077)]. All patients provide written informed consent, according to the agreement on transfer and scientific use of excess material of the University Medicine of the Johannes Gutenberg University Mainz prior to data or specimen collection.

Patient consent for publication

Not applicable.

Competing interest

The authors declare that they have no competing interests.

Glossary

Abbreviations

Abbreviations:

References