PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Gastroenterology. Author manuscript; available in PMC 2012 July 1.
Published in final edited form as:
PMCID: PMC3129453
NIHMSID: NIHMS285227

A Functional Polymorphism in the Epidermal Growth Factor Gene is Associated with Risk for Hepatocellular Carcinoma

Abstract

Background & Aims

A single nucleotide polymorphism 61*G (rs4444903) in the Epidermal Growth Factor (EGF) gene has been associated, in 2 case-control studies, with hepatocellular carcinoma (HCC). We tested associations between demographic, clinical, and genetic data and development of HCC, and developed a simple predictive model in a cohort of patients with chronic hepatitis C and advanced fibrosis.

Methods

Black and white subjects from the HALT-C trial (n=816) were followed prospectively for development of a definite or presumed case of HCC for a median time period of 6.1 years. We used the Cox proportional hazards regression model to determine the hazard ratio for risk of HCC and to develop prediction models.

Results

Subjects with EGF genotype G/G had a higher adjusted risk for HCC than those with genotype A/A (hazard ratio, 2.10; 95% confidence interval, 1.05–4.23; P=0.03). After adjusting for EGF genotype, blacks had no increased risk of HCC risk, compared with whites. Higher serum levels of EGF were observed among subjects with at least one G allele (P=0.08); the subset of subjects with EGF G/G genotype and above-median serum levels of EGF had the highest risk of HCC. We developed a simple prediction model that included the EGF genotype to identify patients at low, intermediate, and high risk for HCC; 6-year cumulative HCC incidences were 2.3%, 10.4%, and 26%, respectively.

Conclusions

We associated the EGF genotype G/G with increased risk for HCC; differences in its frequency among black and white subjects might account for differences in HCC incidence between these groups. We developed a model that incorporates EGF genotype and demographic and clinical variables to identify patients at low, intermediate, and high risk for HCC.

Keywords: liver disease, cancer, tumor, prognosis, HCV

Introduction

Hepatocellular carcinoma (HCC) is the sixth most common solid tumor worldwide and the third leading cause of cancer-related deaths.1-2 Cirrhosis is the strongest known risk factor for HCC, particularly cirrhosis related to hepatitis C virus (HCV) and hepatitis B virus (HBV) infections.3-4 Because current therapies for HCC are limited, and because most diagnoses of HCC are made late in the natural history of the disease, early identification of persons at high risk for HCC is of vital importance in order for targeted screening and chemoprevention to be effective. The current screening strategies for HCC are focused on patients with cirrhosis of any cause and patients with long-standing chronic hepatitis B, even in the absence of cirrhosis. Although, traditionally, many clinicians rely for HCC screening on serologic α-fetoprotein (AFP) testing and abdominal ultrasound (US) imaging,5 AFP screening is so nonspecific that it is no longer recommended.6 In fact, the effectiveness of both of these screening modalities is limited. AFP has reported sensitivities in the range of 25-65% and can be elevated falsely in patients with chronic hepatitis C.7-8 US has a reported sensitivity of approximately 60%, and sensitivities in advanced cirrhosis are even lower.9-10

Mounting evidence from animal studies supports a role for epidermal growth factor (EGF) in malignant transformation of hepatocytes and in the pathogenesis of HCC.11-16 Hepatocellular carcinoma develops in transgenic mice with liver-targeted over-expression of the secreted EGF fusion protein,11 and blockade of EGF receptor (EGFR) activity by gefitinib (an EGFR-tyrosine kinase inhibitor) and AEE788 (a novel dual receptor tyrosine kinase inhibitor of the EGF and the VEGF receptor) halt the development and progression of HCC.15-16

A single nucleotide polymorphism involving an A-to-G mutation at position 61 of the 5’ untranslated region (5’ UTR) region of the EGF gene (rs4444903) results in higher EGF levels in persons with EGF genotype G/G and has been associated with an increased risk of multiple human malignancies.17-19

Recently, Tanabe and colleagues20 demonstrated in two independent case-control study populations, one American (207 subjects with 59 HCC cases), the other French (141 subject with 44 HCC cases), 4-fold and 2.9-fold increases, respectively, in the risk of HCC in subjects with the EGF genotype G/G compared to genotype A/A. Furthermore, patients with genotype G/G had significantly higher serum and hepatic EGF levels. In a logistic model adjusted for age, sex, race, cause and severity of cirrhosis, as estimated by biochemical blood tests, these investigators found that the number of copies of the G allele was associated significantly with the risk of HCC. The study by Tanabe and colleagues, however, was retrospective in design; was not controlled for histological severity of the underlying liver disease; did not include a definition of the duration of cirrhosis prior to the emergence of HCC; was not adjusted for other known HCC risk factors, such as smoking; and lacked ethnic diversity, as the majority of the subjects in both populations were white.

We sought to replicate the findings of Tanabe and colleagues in the HALT-C Trial cohort, which offered several advantages over previously studied populations for evaluation of the association between EGF gene single nucleotide polymorphism 61*G and the risk of HCC. These advantages included a large cohort size, prospective follow-up over 8 years for the incidence of new cases of HCC, ethnic diversity, and the opportunity to control for other important HCC risk factors (histological severity of the underlying liver disease and smoking status). Furthermore, the availability of specimens collected during the HALT-C Trial allowed us to correlate EGF genotype, serum EGF protein level, and the risk of HCC. Our ultimate goal was to incorporate the EGF genotype and readily available clinical variables into an HCC predictive model that would allow stratification of HCC risk (low, moderate, high). Potentially, such a predictive model could help target more aggressive HCC screening and chemoprevention protocols to the high-risk group.21

Patients and Methods

The HALT-C Trial was a randomized, controlled, multicenter trial of 1,050 subjects with chronic hepatitis C and advanced hepatic fibrosis (Ishak fibrosis score ≥3) who had failed to achieve a sustained virologic response (SVR) after treatment with pegylated interferon and ribavirin 22. At entry, subjects in this trial had no history of clinical events or laboratory markers of hepatic decompensation, a previous diagnosis of HCC, the presence of liver disease other than HCV, or medical and/or psychiatric contraindications to interferon-based therapy. As a requirement for trial enrollment, all subjects were screened for HCC by imaging (ultrasound, computed tomography, or magnetic resonance imaging) and AFP levels (subjects were excluded for an AFP >200 ng/mL, even in the absence of abnormal imaging).

Although entry into the HALT-C Trial was limited to patients with Ishak fibrosis stage ≥3, staging of biopsies by local pathologists at individual sites was reclassified if a central reassessment by all study pathologists examining the specimen together resulted in a different consensus; in addition, subjects with a baseline Ishak fibrosis score of 2 but a fibrosis score of ≥3 on a previous liver biopsy were eligible for enrollment. As a result, 79 enrolled subjects had baseline Ishak fibrosis scores of 2.23

After a 24-week lead-in phase, in which all subjects were treated with peginterferon alfa-2a (180 μg weekly) and ribavirin (1–1.2 g daily), those who remained viremic at week 20 were classified as nonresponders and were eligible for enrollment in the randomized phase of the trial. In addition, week-20 responders (who continued full-dose combination treatment for up to 48 weeks) were eligible for enrollment if they became viremic during continued therapy or experienced virologic relapse after completing therapy. Finally, enrollment was open to subjects who had been treated unsuccessfully outside the trial with pegylated interferon and ribavirin. Of these 1,050 nonresponders, 517 subjects were assigned randomly to receive maintenance therapy with peginterferon alfa-2a, 90 μg weekly, for 3.5 years, and 533 to receive no treatment but followed for the same duration. All patients had a liver biopsy performed prior to enrollment (baseline) and repeated at 1.5 and 3.5 years after randomization. A panel of 12 hepatic pathologists convened together to review and assign a consensus Ishak fibrosis score (0 – 6) to all biopsy specimens.24 Progression of fibrosis, as a HALT-C trial outcome, required a ≥ 2-point increases in Ishak fibrosis stage from baseline on either of the two post-randomization biopsies.

For this ancillary study, we included all randomized black and white subjects from the HALT-C Trial, who consented for genetic testing on entry to the trial. EGF genotyping was performed by Celera Corporation (Alameda, CA) as part of a 150 SNP analysis study that included only black and white HALT-C Trial subjects. Thus, out of the 1050 randomized subjects, 814 were included in this study and were followed prospectively for the development of a definite or presumed case of HCC over the trial's follow-up period.

Because clinical outcomes, including HCC, occurred in an indistinguishable proportion of both the maintenance-treatment and untreated-control groups of the HALT-C Trial,21 we were able to include the entire randomized cohort in this analysis.

HCC surveillance and diagnostic criteria

Subjects were seen every 3 months during the 3.5 years of the randomized trial and every 6 months thereafter, unless consent was withdrawn or a study-ending clinical outcome (including liver transplantation) occurred. Median follow-up time was 6.1 years (range 0-8.7 years). A complete blood count, a liver panel, and an AFP level were obtained at the local clinical center at each visit. Subjects underwent ultrasound examination of the liver every 6-12 months; those with rising AFP and or new lesions on ultrasound were evaluated further with computed tomography or magnetic resonance imaging. Diagnostic liver biopsy and HCC treatment were conducted at the discretion of investigators at each site.

A case of HCC was classified as “definite” if histological confirmation was obtained or a new mass lesion ≥2 cm on imaging was associated with an AFP level >1,000 ng/mL. A case of HCC was classified as “presumed” if a new mass lesion was found on ultrasound in the absence of histological conformation, the AFP level was <1,000 ng/mL, and one of the following criteria were met: (1) two different liver imaging studies showed a mass lesion with characteristics of HCC (vascular enhancement, wash out), (2) imaging lesions became progressively enlarged, leading to death, or (3) one additional imaging study showed a mass lesion with characteristics of HCC that either increased in size over time or was accompanied by increasing AFP levels. All cases of HCC (presumed and definite) were reviewed by an Outcomes Review Panel composed of rotating panels of 3 investigators to ascertain that the predefined criteria were met and to determine the date when these criteria were first met. All patients who met criteria for presumed or definite HCC continued to be followed, and results of subsequent imaging or histology (including surgically resected specimens and liver explants) were submitted to the panel for review. Presumed HCC cases in subjects who survived and were followed for >24 months without meeting subsequent criteria for definite HCC were dropped from the analyses (n = 2).

Genotyping of EGF Gene and serum EGF protein level

DNA was extracted from whole blood (Gentra Systems Puregene kit), immortalized B-lymphocytes, or peripheral blood mononuclear cells (Qiagen DNA purification columns, Qiagen Inc, Valencia, CA). Genotyping of EGF SNP rs4444903 was carried out by allele-specific real-time PCR in a high throughput facility at Celera. 25 For each allele-specific PCR reaction, 0.3 ng of DNA was amplified. Genotypes were assigned automatically by an in-house software program followed by manual curation without knowledge of phenotype. Genotyping accuracy in prior studies was ~99%.26 Primer sequences are available upon request.

Serum EGF protein level was quantified by an enzyme-linked immunosorbent assay (ELISA) (PeproTech, Rocky Hill, NJ). EGF levels were measured in serum as opposed to plasma, because the major source of EGF in the blood is platelet.27 Serum was isolated and stored at -80°C before use. Each ELISA plate well was incubated overnight with 100 μL of capture antibody (1 μg/mL) before blocking with 1% bovine serum albumin in PBS for 1 h. Serum samples diluted 1:5 in PBS containing 25% fetal bovine serum were analyzed in triplicate. Samples were incubated in the plate for 2 h followed by the addition of 100 μL of detection antibody (0.25 μg/mL) for 2 h and 100 μL of avidin peroxidase (1:2,000) for 30 min. Wells were washed four times with PBS containing 0.05% Tween 20 between each step. Color development was monitored after the addition of 100 μL of 2,2’-Azino-bis (3-ethylbenzothiazoline-6-sulfonic acid [ABTS]) (Sigma, St. Louis, MO) in a spectrophotometric plate reader (EMax, Molecular Devices, Sunnyvale, CA). The average concentration of each sample was determined based on a plate-specific EGF standard curve.

Out of the 816 patients genotyped for the EGF gene, protein levels taken at baseline were obtained for 814 subjects; two patients did not have adequate serum samples. Furthermore, EGF serum protein levels from one site were dropped because 30% of the samples from this site had no detectable EGF protein compared to 0%-13% in the remaining nine centers. Therefore, we analyzed serum EGF protein levels in 722 subjects.

Statistical Methods

Cox proportional hazards regression analysis to determine the hazard ratio (HR) for HCC based on EGF genotype was carried on a total of 816 HALT-C Trial subjects, 66 (8%) with confirmed or presumed HCC. Variables with a P value of <0.1 on univariate analysis or that had been associated previously with HCC were included in multivariate model 1. These variables included age, gender, ethnicity, smoking status, baseline Ishak fibrosis stage, and EGF genotype.

Multivariate model 2 was limited to variables readily available in clinical practice (age, gender, smoking status, baseline alkaline phosphatase (ALK-P) level, baseline platelet count) and EGF genotype. This model was used to segregate the cohort into three risk groups based on individual regression model estimates of HCC risk and the overall HCC incidence.21 The cumulative incidence of HCC among these risk groups was determined by Kaplan–Meier analysis and compared with the log-rank test. The regression formula for this model is: AGE*0.058 + FEMALE*-0.265 + SMOKED*1.183 + ALK-P*0.005 + PLATELET*-0.012 + EGF (A/G)*-0.054 + EGF (G/G)*0.730. A score was calculated for each subject, who were then ranked in order of these scores. The 6-year incidence of HCC in the cohort (approximately 8%) was used to select cut points to reflect the risk groups. The low risk cut point was selected such that 2% of the sample developed HCC, while the cut points for intermediate and high risk captured groups in which 2-8% and >8% of subjects developed HCC, respectively. The three cut points selected were: 0 to log10[2.39], log10[2.39] to log10[3.38], and > log10[3.38] ,to reflect low, intermediate, and high risk.

We used a multivariate logistic regression model to assess the association between EGF genotype and a ≥ 2-point increases in Ishak fibrosis stage from baseline on either of two serial liver biopsies 1.5 and 3.5 years post randomization. We used ANOVA to assess differences in serum EGF protein levels between different EGF genotype groups. EGF serum protein levels were log transformed for these analyses. Pearson's chi-square test was used to assess differences in HCC among four different subgroups defined by EGF genotype and serum protein levels. A 2-sided significance level of <0.05 was used for all analyses.

Results

Characteristics of the 816 subjects included in our analysis are shown in Table 1. Of the 816 subjects, HCC developed in 66. The mean age of subjects in whom HCC developed was 52 years old, 79% were male, 27% black, and 91% smokers. The 6-year cumulative incidence of HCC among blacks was 10.8%, compared to 6.7% among whites. On univariate analysis, we found that subjects in whom HCC developed had lower platelet counts, higher ALK-P levels, and more advanced fibrosis on liver biopsy compared to subjects who remained free of HCC (Table 1).

Table 1
Baseline characteristics of subjects with and without HCC.

EGF gene polymorphisms and risk of HCC

The prevalence of the different EGF genotypes in our cohort among blacks was 56% G/G, 33% G/A, and 10% A/A (Hardy-Weinberg equilibrium Chi-Square 4.40, P = 0.04) and among whites was 17% G/G, 49% G/A, and 34% A/A (Hardy-Weinberg equilibrium Chi-Square 0.008, P = 0.93).

Subjects with EGF genotype G/G had a higher risk for HCC than those with genotype A/A (HR 2.12, 95% CI 1.12-4.00) (Table 2). This risk continued to be significant after we adjusted for age, gender, ethnicity, smoking status, and baseline Ishak fibrosis score in a Cox proportional hazards regression model (HR 2.1, 95% CI 1.05-4.23, P = 0.03). Subjects with genotype G/A had no significant increase in risk of HCC as compared to subjects with genotype A/A (HR 0.97, 95% CI 0.51-1.85). Other significant factors associated with increased HCC risk in the Cox proportional hazards regression model included age (HR = 1.07; CI 1.03-1.10), smoking status (HR = 3.21; CI 1.35-7.64), and baseline Ishak fibrosis score (HR = 1.52; CI 1.25-1.86). After adjusting for EGF genotype, we found no increase in the HCC risk of blacks as compared to whites (HR = 0.96, CI 0.71-1.30; Table 2).

Table 2
Cox proportional hazards model for development of HCC

EGF genotypes, serum EGF protein levels, and risk of HCC

Table 3 shows serum EGF protein levels among the different EGF genotypes and between subgroups with and without HCC. An association between the EGF genotype and higher serum EGF protein levels was seen among subjects with at least one G allele (P = 0.08) (Table 3a). Furthermore, although falling short of statistical significance, subjects with HCC had a median serum EGF protein level 27% higher than that of subjects without HCC, an increase that was more pronounced among those with genotype G/G (median level for subjects with HCC 63% higher than that for subjects without HCC) but less pronounced among those with genotype A/A (median level for subjects with HCC 20% higher than that for subjects without HCC) (Table 3b). Figure 1 shows the portion of subjects who developed HCC among four different subgroups defined by EGF genotype and serum EGF level (above or below median). Among subjects with the G/G genotype and serum EGF level above median, 15% developed HCC compared to 10% among those with G/G and serum level below the median. The proportion of subjects with HCC was similar among those with other genotypes (G/A and A/A) and serum EGF protein level above or below the median and was 6%. These data indicate that a subset of subjects with the G/G genotype and above median serum EGF protein level is at highest risk for HCC development, although HCC risk persisted even among those with the G/G genotype and below median serum EGF protein level compared to other EGF genotypes.

Figure 1
Portion of subjects developing HCC among four different subgroups defined by EGF genotype and serum protein levels above or below the median.
Table 3a
EGF protein serum levels (pg/ml) among the different EGF genotypes
Table 3b
EGF protein serum levels (pg/ml) between subjects with and without HCC

Simplified model to predict HCC risk

A second Cox proportional hazards regression model (Model 2), limited to variables readily available in clinical practice (age, gender, smoking status, ALK-P level, platelet count) plus EGF genotype, was used to predict HCC risk and segregate the cohort into 3 risk groups based on the resulting individual estimates derived from model 2 (Table 4).

Table 4
Cox proportional hazards model for HCC-risk

Based on this model, 57% of the cohort segregated to a low-risk category, 29% to an intermediate-risk category, and 14% to a high-risk category with 6-year cumulative HCC incidences of 2.3%, 10.4%, and 26%, respectively (Figure 2). Table 5 shows the characteristics of each risk group with the predictors that went into the model.

Figure 2
Kaplan–Meier estimates of 6-year cumulative incidence of HCC among subjects predicted to have low (<2% incidence), intermediate (2-8% incidence), and high (>8% incidence) risk of HCC according to regression Model 2.
Table 5
Characteristics of each HCC-risk group*.

Adding EGF genotype in a step-wise fashion to these clinical variables improved the model's segregation power and fit. With EGF in the model, 57% of the cohort segregated to a low-risk category compared to 48% without EGF in (a difference of 72 patients being reclassified to a low-risk category by adding EGF genotype). The fit of the model also improved as measured by the Akaike information criterion that moved from 772 to 768 with EGF in the model. Thus, adding EGF 61*G SNP to this clinically simple model enhances its performance in segregating the cohort according to HCC risk better than any predictive model reported in the literature thus far.

EGF genotype and liver fibrosis progression

In order to assess the association between EGF genotype and liver fibrosis progression, we ran a multivariate logistic regression model predicting ≥ 2-point increases in Ishak fibrosis stage by EGF genotype (G/G vs. G/A and A/A) after controlling for age, gender, and baseline Ishak fibrosis stage. Interestingly, the adjusted odds ratio (OR) for a ≥ 2-point increases in Ishak fibrosis stage in subjects with the G/G genotype was decreased (OR 0.62, 95% CI 0.37 – 1.02) (P = 0.059) compared to G/A and A/A.

Discussion

HCC is a complex, heterogeneous malignancy, the pathogenesis of which involves multiple genetic and epigenetic alterations and modulation of molecular signaling pathways implicated in malignant transformation of hepatocytes and tumor progression.28 Dysregulation of the EGF/EGFR signaling pathway, is thought to be important in early hepatocarcinogenesis.29-30 A functional polymorphism in the 5’ untranslated region of the EGF gene (61*G) that modulates tissue-specific EGF gene expression has been associated with multiple human malignancies including HCC.17, 19-20, 31-32

Tanabe and colleagues20 first showed in two case-control studies that, in patients with alcoholic and hepatitis C-associated cirrhosis, the EGF 61*G allele is associated with an increased risk of HCC compared to the A allele. Whether the observation of this increased risk was confounded by the histologic severity of the underlying liver disease, duration of cirrhosis prior to the development of HCC, or other known HCC risk factors, such as smoking, remained to be determined. In the current study, after adjusting for age, gender, ethnicity, smoking status, and baseline stage of liver fibrosis, we demonstrated in a large cohort followed prospectively for emerging cases of HCC during a period exceeding 6 years that subjects with EGF genotype G/G had more than a twofold increased risk of HCC compared to those with genotype A/A.

With a multivariate Cox regression model derived from the same HALT-C Trial cohort, Lok and colleagues21 showed that the risk of HCC was increased in blacks compared to whites. In the current study, blacks had a higher cumulative 6-year incidence of HCC compared to whites, which corresponded to a higher prevalence of EGF genotype G/G among blacks compared to whites (56% versus 17%). Indeed, the association between black race and HCC diminished in our Cox regression model after adjustment for EGF 61*G genotype, thus indicating that a higher percentage of G/G and lower percentage of A/A among black subjects could contribute to the higher incidence of HCC in this ethnic group.

Others19-20, 33 have demonstrated, in a subset of subjects selected randomly from their larger cohorts, significantly higher serum and hepatic EGF levels in those with genotype G/G compared to those with genotype A/A. In the liver, higher hepatic EGF levels correlated with higher levels of phosphorylated EGF receptors.20 In these studies by others, however, subject numbers were too small to permit evaluation of an association between higher serum EGF levels and risk of HCC.

In this study, we assayed serum EGF protein levels in 722 subjects and showed that subjects with HCC had median serum EGF protein levels 27% higher than those without HCC; this increase in serum EGF protein levels in subjects with HCC was more pronounced among those with genotype G/G than those with genotype A/A (63% versus 20%, respectively). When we assessed the association between serum EGF protein levels and HCC stratified by EGF genotype, the subset of subjects with G/G genotype and above median serum EGF protein levels was at highest risk for HCC development (Figure 1), although HCC risk remained elevated in those with G/G genotype and below median serum EGF protein levels compared to other EGF genotypes. These data suggest an additive risk for HCC development between EGF genotype and serum EGF protein levels. Although some of these associations did not meet statistical significance, the trends were apparent even in this study of small numbers of HCC cases. Larger studies will be necessary to confirm the validity of these observations.

The EGF/EGFR signaling pathway has been shown to be an important mediator of hepatocyte proliferative capacity and liver regeneration in response to chronic injury.34-36 Modulation of EGF levels rather than alteration in EGF receptor expression has been suggested to be the mediator of this regenerative liver response.13, 30 These studies lend biological plausibility to our observation of lower liver fibrosis progression rates among subjects with the 61*G functional polymorphism in the 5’ untranslated region of the EGF gene, which predicts increased EGF mRNA expression in hepatocytes and stability in serum. They argue that the observed association between the EGF 61*G functional polymorphism and HCC is not mediated by a more aggressive liver fibrosis course, but perhaps more likely by the inability to downregulate the EGF pathway once cirrhosis has developed, leading to early hepatocarcinogenesis and uncontrolled progression of early HCC.

Rat models of liver fibrosis/cirrhosis induced by dimethylnitrosamine (DMN) and bile duct ligation have demonstrated, that both the proliferative capability of hepatocytes and EGF mRNA expression decrease with progression to cirrhosis.37-38 Failure to attenuate hepatic EGF expression in surrounding non-tumor hepatic tissue has also been associated with poor survival in patients with HCC. 39 Thus, in light of our finding higher serum EGF levels in subjects with HCC and genotype G/G than in those with HCC and genotype A/A, we speculate that EGF genotype G/G might also be associated with poor HCC survival and rapid progression as well.

Given the relatively strong association between EGF genotype and risk of HCC confirmed now in three large, independent studies, and given the important role of the EGF/EGFR signaling pathway in hepatocytes proliferation in response to chronic liver injury and in early hepatocarcinogenesis, we sought to incorporate the EGF genotype into a simple predictive model for HCC based on readily available clinical characteristics that have been shown previously to be associated with the risk of HCC.21 With this model, we were able to categorize approximately three quarters of the cohort to low-risk or high-risk categories with an associated 6-year cumulative HCC incidence of 2.3% and 26% respectively. Therefore, potentially, this model could be applied clinically to stratify HCC risk in patients with advanced fibrosis and cirrhosis; those with low risk would be triaged to less frequent surveillance imaging protocols, while those with high risk would be considered for protocols involving more frequent surveillance intervals, more sensitive imaging techniques, and, possibly, chemoprevention strategies.28, 40

The strengths of this study include its large sample size, prospective nature, availability of high-quality data on covariate and potential confounders, and a clear, stringent definition of HCC as a trial outcome according to predefined criteria. Questions that remain unanswered by our study and merit future inquiry are a) whether the EGF genotype is a risk factor for HCC in other groups of patients at risk, e.g., patients with cirrhosis of other causes and patients with long-standing chronic hepatitis B in the absence of cirrhosis, and b) whether our clinical prediction model of HCC risk can be validated in other cohorts. Furthermore, whether EGF expression in the liver and EGF levels in serum correlate or are regulated independently remains to be determined.

In conclusion, we have shown that single nucleotide polymorphism 61*G in the EGF gene correlates with the risk of HCC, independent of several known or purported clinical risk factors and could contribute to the discrepancy in HCC risk between blacks and whites. Furthermore, the EGF genotype can be incorporated in a clinical prediction model that can be used to stratify relative HCC risk in patients with chronic hepatitis C and advanced fibrosis/cirrhosis.

Acknowledgments

This study was supported by the National Institute of Diabetes & Digestive & Kidney Diseases (contract numbers are listed below). Additional support was provided by the National Institute of Allergy and Infectious Diseases (NIAID), the National Cancer Institute, the National Center for Minority Health and Health Disparities and by General Clinical Research Center and Clinical and Translational Science Center grants from the National Center for Research Resources, National Institutes of Health (grant numbers are listed below). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center for Research Resources or the National Institutes of Health. Additional funding to conduct this study was supplied by Hoffmann-La Roche, Inc., (now Genentech) and Celera Corporation through Cooperative Research and Development Agreements (CRADAs) with the National Institutes of Health. B. C. Fuchs is supported by a K award from the NCI (5 K01 CA140861-02). R.T. Chung is supported by DK078772. The authors express appreciation for the efforts of Joe Catanese and the Celera High Throughput group for generating genotyping data, to Dr. David Ross and Computational Biology group for bioinformatics support, and to Charles Rowland and Robert Lagier for preliminary statistical analysis.

In addition to the authors of this manuscript, the following individuals were instrumental in the planning, conduct and/or care of patients enrolled in this study at each of the participating institutions as follows:

University of Massachusetts Medical Center, Worcester, MA: (Contract N01-DK-9-2326) Gyongyi Szabo, MD, Barbara F. Banner, MD, Maureen Cormier, RN, Donna Giansiracusa, RN

University of Connecticut Health Center, Farmington, CT: (Grant M01RR-06192) Herbert L. Bonkovsky, MD, Gloria Borders, RN, Michelle Kelley, RN, ANP

Saint Louis University School of Medicine, St Louis, MO: (Contract N01-DK-9-2324) Adrian M. Di Bisceglie, MD, Bruce Bacon, MD, Brent Neuschwander-Tetri, MD, Elizabeth M. Brunt, MD, Debra King, RN

Massachusetts General Hospital, Boston, MA: (Contract N01-DK-9-2319, Grant M01RR-01066; Grant 1 UL1 RR025758-01, Harvard Clinical and Translational Science Center) Andrea E. Reid, MD, Atul K. Bhan, MD, Wallis A. Molchen, David P. Lundmark

University of Colorado Denver, School of Medicine, Aurora, CO: (Contract N01-DK-9-2327, Grant M01RR-00051, Grant 1 UL1 RR 025780-01) Gregory T. Everson, MD, Thomas Trouillot, MD, Marcelo Kugelmas, MD, S. Russell Nash, MD, Jennifer DeSanto, RN, Carol McKinley, RN

University of California - Irvine, Irvine, CA: (Contract N01-DK-9-2320, Grant M01RR-00827) Timothy R. Morgan, MD, John C. Hoefs, MD, John R. Craig, MD, M. Mazen Jamal, MD, MPH, Muhammad Sheikh, MD, Choon Park, RN

University of Texas Southwestern Medical Center, Dallas, TX: (Contract N01-DK-9-2321, Grant M01RR-00633, Grant 1 UL1 RR024982-01, North and Central Texas Clinical and Translational Science Initiative) William M. Lee, MD, Thomas E. Rogers, MD, Peter F. Malet, MD, Janel Shelton, Nicole Crowder, LVN, Rivka Elbein, RN, BSN, Nancy Liston, MPH

University of Southern California, Los Angeles, CA: (Contract N01-DK-9-2325, Grant M01RR-00043) Karen L. Lindsay, MD, MMM, Sugantha Govindarajan, MD, Carol B. Jones, RN, Susan L. Milstein, RN

University of Michigan Medical Center, Ann Arbor, MI: (Contract N01-DK-9-2323, Grant M01RR-00042, Grant 1 UL1 RR024986, Michigan Center for Clinical and Health Research) Anna S. Lok, MD, Robert J. Fontana, MD, Joel K. Greenson, MD, Pamela A. Richtmyer, LPN, CCRC, R. Tess Bonham, BS

Virginia Commonwealth University Health System, Richmond, VA: (Contract N01-DK-9-2322, Grant M01RR-00065) Mitchell L. Shiffman, MD, Richard K. Sterling, MD, MSc, Melissa J. Contos, MD, A. Scott Mills, MD, Charlotte Hofmann, RN, Paula Smith, RN

Liver Diseases Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD: Marc G. Ghany, MD, T. Jake Liang, MD, David Kleiner, MD, PhD, Yoon Park, RN, Elenita Rivera, RN, Vanessa Haynes-Williams, RN

National Institute of Diabetes and Digestive and Kidney Diseases, Division of Digestive Diseases and Nutrition, Bethesda, MD: James E. Everhart, MD, MPH, Leonard B. Seeff, MD, Patricia R. Robuck, PhD, Jay H. Hoofnagle, MD, Elizabeth C. Wright, PhD

University of Washington, Seattle, WA: (Contract N01-DK-9-2318) Chihiro Morishima, MD, David R. Gretch, MD, PhD, Minjun Chung Apodaca, BS, ASCP, Rohit Shankar, BC, ASCP, Natalia Antonov, M. Ed.

New England Research Institutes, Watertown, MA: (Contract N01-DK-9-2328) Kristin K. Snow, MSc, ScD, Anne M. Stoddard, ScD, Teresa M. Curto, MSW, MPH

Inova Fairfax Hospital, Falls Church, VA: Zachary D. Goodman, MD, PhD

Data and Safety Monitoring Board Members: (Chair) Gary L. Davis, MD, Guadalupe Garcia-Tsao, MD, Michael Kutner, PhD, Stanley M. Lemon, MD, Robert P. Perrillo, MD

Abbreviations

AFP
alpha fetoprotein
ALK-P
alkaline phosphatase
EGF
epidermal growth factor
EGFR
epidermal growth factor receptor
ELISA
enzyme-linked immunosorbent assay
HBV
hepatitis B virus
HCC
hepatocellular carcinoma
HCV
hepatitis C virus
SVR
sustained virologic response
US
ultrasound
5’ UTR
5’ untranslated region

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

We are enclosing two sections relating to disclosures of potential conflicts of interest. Disclosures that pertain to the industrial sponsors who have partnered with the NIDDK to support this study, which are also listed in the text of the manuscript, are listed in section A.

In addition, many of the HALT-C Trial investigators have other associations with industry relating to the area of hepatitis C, and, to achieve the highest level of disclosure, we list these for you as well in section B.

Authors with no financial relationships related to this project are: B. K. Abu Dayyeh, M. Yang, B. C. Fuchs, D. L. Karl, S. Yamada, T. O'Brien, J. L. Dienstag, and K. K. Tanabe.

Section A.

The following are disclosures that pertain to the industrial sponsors who have partnered with the NIDDK to support this study, which are also listed in the text of the manuscript.

Financial relationships of the authors with Hoffmann-La Roche, Inc. (now Genentech), are as follows:

R. T. Chung receives research support.

Financial relationships of the authors with Celera Corporation are as follows:

J. J. Sninsky is an employee and has equity interest.

Section B.

In addition, many of the HALT-C Trial investigators have other associations with industry relating to the area of hepatitis C, and, to achieve the highest level of disclosure, we list these for you as well.

J. L. Dienstag: Research Support: Vertex Pharmaceuticals; Serves on Data Monitoring Committee: Schering-Plough Research Institute, Human Genome Sciences, and Medtronic; Ad hoc Hepatitis Advisory Board: Boehringer-Ingelheim; Antiviral Advisory Board: Gilead Sciences; Ad hoc Consultant (stock options): Achillion; Clinical Advisory Board (stock options): Nucleonics; Scientific Advisory Board (stock options): Metabasis. K. K. Tanabe: Vertex: Honoraria for scientific lecture on January 1, 2008. R. T. Chung: Research Support: Schering-Plough (now Merck), Novartis, and Romark; Consultant: Merck, Pfizer, and Gilead.

Disclosures: Financial relationships of the authors with Hoffmann-La Roche, Inc. (now Genentech), are as follows: R. T. Chung receives research support. Financial relationships of the authors with Celera Corporation are as follows: J. J. Sninsky is an employee and has equity interest. Authors with no financial relationships related to this project are: B. K. Abu Dayyeh, M. Yang, B. C. Fuchs, D. L. Karl, S. Yamada, T. O'Brien, J. L. Dienstag, and K. K. Tanabe.

REFERENCES

1. Parkin DM. The global health burden of infection-associated cancers in the year 2002. Int J Cancer. 2006 Jun 15;118(12):3030–3044. [PubMed]
2. Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002. CA Cancer J Clin. 2005 Mar-Apr;55(2):74–108. [PubMed]
3. Velazquez RF, Rodriguez M, Navascues CA, et al. Prospective analysis of risk factors for hepatocellular carcinoma in patients with liver cirrhosis. Hepatology. 2003 Mar;37(3):520–527. [PubMed]
4. Tsukuma H, Hiyama T, Tanaka S, et al. Risk factors for hepatocellular carcinoma among patients with chronic liver disease. N Engl J Med. 1993 Jun 24;328(25):1797–1801. [PubMed]
5. El-Serag HB, Marrero JA, Rudolph L, Reddy KR. Diagnosis and treatment of hepatocellular carcinoma. Gastroenterology. 2008 May;134(6):1752–1763. [PubMed]
6. Bruix J, Sherman M. Management of hepatocellular carcinoma. Hepatology. 2005 Nov;42(5):1208–1236. [PubMed]
7. Trevisani F, D'Intino PE, Morselli-Labate AM, et al. Serum alpha-fetoprotein for diagnosis of hepatocellular carcinoma in patients with chronic liver disease: influence of HBsAg and anti-HCV status. J Hepatol. 2001 Apr;34(4):570–575. [PubMed]
8. Bayati N, Silverman AL, Gordon SC. Serum alpha-fetoprotein levels and liver histology in patients with chronic hepatitis C. Am J Gastroenterol. 1998 Dec;93(12):2452–2456. [PubMed]
9. Kim CK, Lim JH, Lee WJ. Detection of hepatocellular carcinomas and dysplastic nodules in cirrhotic liver: accuracy of ultrasonography in transplant patients. J Ultrasound Med. 2001 Feb;20(2):99–104. [PubMed]
10. Bolondi L, Sofia S, Siringo S, et al. Surveillance programme of cirrhotic patients for early diagnosis and treatment of hepatocellular carcinoma: a cost effectiveness analysis. Gut. 2001 Feb;48(2):251–259. [PMC free article] [PubMed]
11. Tonjes RR, Lohler J, O'Sullivan JF, et al. Autocrine mitogen IgEGF cooperates with c-myc or with the Hcs locus during hepatocarcinogenesis in transgenic mice. Oncogene. 1995 Feb 16;10(4):765–768. [PubMed]
12. Singletary SE, Baker FL, Spitzer G, et al. Biological effect of epidermal growth factor on the in vitro growth of human tumors. Cancer Res. 1987 Jan 15;47(2):403–406. [PubMed]
13. Blanc P, Etienne H, Daujat M, et al. Mitotic responsiveness of cultured adult human hepatocytes to epidermal growth factor, transforming growth factor alpha, and human serum. Gastroenterology. 1992 Apr;102(4 Pt 1):1340–1350. [PubMed]
14. Giannelli G, Sgarra C, Porcelli L, Azzariti A, Antonaci S, Paradiso A. EGFR and VEGFR as potential target for biological therapies in HCC cells. Cancer Lett. 2008 Apr 18;262(2):257–264. [PubMed]
15. Okamoto K, Neureiter D, Alinger B, et al. The dual EGF/VEGF receptor tyrosine kinase inhibitor AEE788 inhibits growth of human hepatocellular carcinoma xenografts in nude mice. Int J Oncol. 2008 Oct;33(4):733–742. [PubMed]
16. Schiffer E, Housset C, Cacheux W, et al. Gefitinib, an EGFR inhibitor, prevents hepatocellular carcinoma development in the rat liver with cirrhosis. Hepatology. 2005 Feb;41(2):307–314. [PubMed]
17. Shahbazi M, Pravica V, Nasreen N, et al. Association between functional polymorphism in EGF gene and malignant melanoma. Lancet. 2002 Feb 2;359(9304):397–401. [PubMed]
18. Almeida LO, Custodio AC, Santos MJ, et al. The A61G EGF polymorphism is associated with development of extraaxial nervous system tumors but not with overall survival. Cancer Genet Cytogenet. 2010 Apr 1;198(1):15–21. [PubMed]
19. Lanuti M, Liu G, Goodwin JM, et al. A functional epidermal growth factor (EGF) polymorphism, EGF serum levels, and esophageal adenocarcinoma risk and outcome. Clin Cancer Res. 2008 May 15;14(10):3216–3222. [PMC free article] [PubMed]
20. Tanabe KK, Lemoine A, Finkelstein DM, et al. Epidermal growth factor gene functional polymorphism and the risk of hepatocellular carcinoma in patients with cirrhosis. JAMA. 2008 Jan 2;299(1):53–60. [PubMed]
21. Lok AS, Seeff LB, Morgan TR, et al. Incidence of hepatocellular carcinoma and associated risk factors in hepatitis C-related advanced liver disease. Gastroenterology. 2009 Jan;136(1):138–148. [PMC free article] [PubMed]
22. Di Bisceglie AM, Shiffman ML, Everson GT, et al. Prolonged therapy of advanced chronic hepatitis C with low-dose peginterferon. N Engl J Med. 2008 Dec 4;359(23):2429–2441. [PMC free article] [PubMed]
23. Everhart JE, Wright EC, Goodman ZD, et al. Prognostic value of Ishak fibrosis stage: findings from the hepatitis C antiviral long-term treatment against cirrhosis trial. Hepatology. 2010 Feb;51(2):585–594. [PMC free article] [PubMed]
24. Ishak K, Baptista A, Bianchi L, et al. Histological grading and staging of chronic hepatitis. J Hepatol. 1995 Jun;22(6):696–699. [PubMed]
25. Germer S, Higuchi R. Homogeneous allele-specific PCR in SNP genotyping. Methods Mol Biol. 2003;212:197–214. [PubMed]
26. Grupe A, Li Y, Rowland C, et al. A scan of chromosome 10 identifies a novel locus showing strong association with late-onset Alzheimer disease. Am J Hum Genet. 2006 Jan;78(1):78–88. [PubMed]
27. Oka Y, Orth DN. Human plasma epidermal growth factor/beta-urogastrone is associated with blood platelets. J Clin Invest. 1983 Jul;72(1):249–259. [PMC free article] [PubMed]
28. Llovet JM, Bruix J. Molecular targeted therapies in hepatocellular carcinoma. Hepatology. 2008 Oct;48(4):1312–1327. [PMC free article] [PubMed]
29. Morimitsu Y, Hsia CC, Kojiro M, Tabor E. Nodules of less-differentiated tumor within or adjacent to hepatocellular carcinoma: relative expression of transforming growth factor-alpha and its receptor in the different areas of tumor. Hum Pathol. 1995 Oct;26(10):1126–1132. [PubMed]
30. Komuves LG, Feren A, Jones AL, Fodor E. Expression of epidermal growth factor and its receptor in cirrhotic liver disease. J Histochem Cytochem. 2000 Jun;48(6):821–830. [PubMed]
31. Xu W, Li Y, Wang X, et al. Association between EGF promoter polymorphisms and cancer risk: a meta-analysis. Med Oncol. 2009 Dec 24; [PubMed]
32. Price PM, Megyesi J, Saggi S, Safirstein RL. Regulation of transcription by the rat EGF gene promoter in normal and ischemic murine kidney cells. Am J Physiol. 1995 Apr;268(4 Pt 2):F664–670. [PubMed]
33. Li Y, Xie Q, Lu F, et al. Association between epidermal growth factor 61A/G polymorphism and hepatocellular carcinoma susceptibility in Chinese patients. Liver Int. 2010 Jan;30(1):112–118. [PubMed]
34. Noguchi S, Ohba Y, Oka T. Influence of epidermal growth factor on liver regeneration after partial hepatectomy in mice. J Endocrinol. 1991 Mar;128(3):425–431. [PubMed]
35. Jones DE, Jr., Tran-Patterson R, Cui DM, Davin D, Estell KP, Miller DM. Epidermal growth factor secreted from the salivary gland is necessary for liver regeneration. Am J Physiol. 1995 May;268(5 Pt 1):G872–878. [PubMed]
36. Natarajan A, Wagner B, Sibilia M. The EGF receptor is required for efficient liver regeneration. Proc Natl Acad Sci U S A. 2007 Oct 23;104(43):17081–17086. [PubMed]
37. Bissig KD, Marti U, Solioz M, et al. Epidermal growth factor is decreased in liver of rats with biliary cirrhosis but does not act as paracrine growth factor immediately after hepatectomy. J Hepatol. 2000 Aug;33(2):275–281. [PubMed]
38. Kuriyama S, Yokoyama F, Inoue H, et al. Sequential assessment of the intrahepatic expression of epidermal growth factor and transforming growth factor-beta1 in hepatofibrogenesis of a rat cirrhosis model. Int J Mol Med. 2007 Feb;19(2):317–324. [PubMed]
39. Hoshida Y, Villanueva A, Kobayashi M, et al. Gene expression in fixed tissues and outcome in hepatocellular carcinoma. N Engl J Med. 2008 Nov 6;359(19):1995–2004. [PMC free article] [PubMed]
40. Furuse J. Growth factors as therapeutic targets in HCC. Crit Rev Oncol Hematol. 2008 Jul;67(1):8–15. [PubMed]