Association between Epidermal Growth Factor ‎‎+61A/G (rs4444903) Gene Polymorphism with ‎Risk of Hepatocellular Carcinoma in Cirrhotic ‎Hepatitis C Patients

INTRODUCTION

Worldwide, HCC represents the fifth commonest cancer. Mostly, HCC is diagnosed in association with liver cirrhosis [1].Among the Egyptian population, HCC is the fourth most frequent cancer [2], furthermore, lately in Egypt, there is a twofold surge in HCC incidence among patients suffering from chronic liver disease [3].

The earlier the detection of HCC, the better prognosis. Nowadays, EASL guidelines don't recommend serum AFP for HCC surveillance, owing to its poor sensitivity in detecting early lesions [4]. This necessitates the exploration of an alternative sensitive, reliable and non-invasive biomarker tool for early detection of HCC.

Comprehension of HCC biology has revealed that hepatocarcinogenesis is affected by accumulations of genetic mutations [5].

The receptor of epidermal growth factor (EGFR) is overexpressed in many cancers and is increased in the liver. Cell survival, cell cycle progression, tumor invasion, and angiogenesis are all regulated by EGFR in carcinogenesis [6].

The polymorphism of EGF + 61 A/G (rs4444903) is a frequent SNP (single nucleotide polymorphism) in the EGF gene's 5′-untranslated region that affects EGF gene transcription [6].

PATIENTS AND METHODS

• Study design

The Department of Tropical Medicine at Main University Hospital of Alexandria enlisted 90 volunteers; including 60 cirrhotic HCV infected patients and 30 seemingly healthy age and sex-matched controls.

In this case-control study; cirrhotic HCV patients were divided into 30 HCC and 30 non-HCC patients; HCV infection was confirmed in all patients by anti-HCV antibodies and HCV RNA. 

Exclusion criteria included patients with: Co-infection with HBV, Malignancy other than HCC, Immunosuppression, Organ transplantation, and autoimmune diseases.

All the patients in HCC group did not receive directly acting antiviral drugs (DAAs) for HCV infection.

• Methods:

Materials:

After gaining informed consent from 90 individuals, whole blood samples on EDTA were taken. Until DNA extraction, all samples were stored at -80 C.

Laboratory and Radiological workup:

All patients and controls were given a complete medical history and clinical examination. Complete blood count and ESR, blood glucose level (fasting), serum level of urea and creatinine, liver enzymes and function tests including serum levels of aspartate transaminase, alanine transaminase, albumin, bilirubin, alkaline phosphatase, prothrombin time and INR,  HCV Ab and HBs Ag, as well as estimation of serum AFP were all performed on all subjects. All test subjects were subjected to radiological examination, which included abdominal and pelvic ultrasonography. All patients whose ultrasound abdomen confirmed the presence of localised hepatic lesions also had a Triphasic CT abdomen done.

Liver cirrhosis was diagnosed by abdominal ultrasonography.

Quantification of HCV RNA level:

The blood HCV RNA level was measured.

The detection limit was 12 IU/mL using TaqMan® 48 automatic fluorescent quantitative polymerase chain reaction (q-PCR) kits and a Roche COBAS® AmpliPrep®/COBAS® TaqMan® 48 Analyzer (Roche Diagnostics, Mannheim, Germany).

Genotyping of EGF gene:

Collected whole blood samples from all participants, were initially utilized to extract DNA using Qiagen DNA extraction kit ® QIAmp DNA extraction Trademark Thermo Fisher Scientific Inc. Mini kit spin column. To confirm the availability of enough DNA yield, we used a nanodrop to estimate the extracted DNA. The fluorogenic 5'-nuclease assay was used to genotype the EGF 61 A > G polymorphisms using allele-specific primers and following the manufacturer's instructions. TaqMan Genotyping Assay Applied Biosystems ThermoFischer kit was utilized, which detects SNP rs444903. The following oligonucleotide primer sequences were utilized for PCR amplification, EGF forward:

CTTTCAGCCCCAATCCAAGGGTTGT [A/G] reverse:

GCTGGAACTTTCCATCAGTTCTTCC.

A total of 15 ul of reaction mixture was made up of 5.0 ul 2X TaqMan® Master Mix, 0.5 ul 20X Assay Working Stock, 4.5 ul extracted DNA, and 4 ul Nuclease free water. Stepone Applied Biosystems provided the thermal cycler, which was set to a holding temperature of 95°C for 10 minutes for polymerase to be activated, then 40 cycles of DNA denaturation at 95°C for 15 seconds each, followed by 40 cycles of 60°C annealing/extension for 1 minute each.

Allelic discrimination plots

TaqMan® Genotyper Software, Thermo Fisher Cloud Genotyping Application, and real-time PCR equipment software display allelic discrimination results as a plot of Allele 1 (VICTM dye) versus Allele 2 (FAMTM dye). Each sample is designated as a separate point on the allelic discrimination plot as a homozygote, heterozygote, or no template control.

• Data Analysis (statistical):

The current work's data was analysed by IBM SPSS software version 20.0. (Armonk, NY: IBM Corp). We utilised number and percent for qualitative data, and range (minimum and maximum), mean, median, and standard deviation for quantitative data. The significance of the findings was calculated at a 5% level of significance.

The Chi-square test was used to analyse a comparison between research groups in terms of categorical features. For quantitative variable distributions, the Kolmogorov-Smirnov test, Shapiro-Wilk test, and D'Agstino test were employed, while for the vision test, the Histogram and QQ plot were utilised. Parametric tests were used if the data were evenly distributed; non-parametric tests were used if the data were irregularly distributed.

The F-test (ANOVA) and Post Hoc test (Tukey) were employed to compare the groups if the data were evenly distributed. In the event of anomalous data distribution, the Mann-Whitney test was used to compare two independent populations, while the Kruskal Wallis test was used to compare the three groups, and the Mann-Whitney test was used to analyse pairwise comparisons. Correlations between two quantitative variables were determined using the Spearman coefficient.

RESULTS:

1- Demographic, AFP serum level (of the studied groups), triphasic CT findings, and BCLC (Barcelona Clinic Liver Cancer) for staging of HCC group: (Table I)

There was a male predominance in all the groups; 73.3%, 80%, and 66.7% in HCV with HCC, HCV without HCC, and control groups respectively without significant difference. The mean age tends to be in the fifties in all studied groups at which the difference was not significant.

For the HCC, non-HCC, and control groups, the mean AFP was 102.244.28 ngml, 3.040.66 ngml, and 3.080.94 ngml, respectively. The HCC group had a statistically significant difference from the cirrhotic non-HCC group (p = 0.001), as well as a difference that is statistically significant from the controls (p = 0.001).

In the group of HCC, Triphasic CT abdomen showed that 33.3% have a single focal lesion, 50% have two focal lesions and 16.7% have three focal lesions, the mean of focal lesion size was 3.19 ± 1.18, the patients who have portal vein thrombosis were 43.3%. 43.3% of the HCC group patients have metastasis most commonly in the lung (61.5%).

In the HCC group, according to BCLC staging, stage (A) was found in 20%, stage (B) BCLC was found in 16.7%, stage (C) was detected in 13.3%, while stage (D) was presented in 50%.

2- EGF

• Distribution of EGF genotypes and alleles in the examined groups (Table II)

EGF genotypes distribution was as following: AA genotype was found in 30%,13.3% and 36.7% for HCC group, non HCC group and controls respectively. AG genotype was detected in 33.3%, 53.3%, and 60.0% in the HCC group, non HCC group and controls respectively. GG genotype was found in 36.7%, 33.3%, and 3.3% respectively. There were statistically significant differences between the HCC and control groups (p 0.004), as well as between the cirrhotic HCV and control groups (p 0.005).

EGF A allele was discovered in 46.7% of HCC patients, 40.0% of cirrhotic HCV patients, and 66.7% of control, respectively, and G allele was found in 53.3%, 60.0%, and 33.3% of HCC patients, cirrhotic HCV patients, and control, respectively. There were statistically significant differences between HCC patients and controls (p 0.027) as well as cirrhotic HCV patients and controls (p 0.00).

• Polymorphism of EGF+61 A/G Assessment as a hazard factor for HCC development in cirrhotic patients (Table III)

Patients carrying EGF GG genotype or G allele are 13.44-fold, 2.28-fold (respectively) increased risk for HCC development than those carrying AG genotype or A allele.

• Relation between Epidermal growth factor alleles and BCLC staging (Table IV)

Regarding epidermal growth factor A allele, BCLC stage D was found in 35.7% followed by stage A which present in 32.1. While in EGF G allele, BCLC stage D was detected in 62.5% followed by stage B which present in 15.6%. Between the two alleles, there was no statistically significant difference regarding BCLC staging.

• Relation between EGF alleles and triphasic CT findings in HCC group (Table V):

Regarding focal lesions number; in patients with A allele, one focal lesion was presented in 25%, two focal lesions were presented in 53.6%, while three focal lesions were found in 21.4%. In patients with G allele, one focal lesion was found in 40.6%, two focal lesions were detected in 46.6% and three lesions were detected in 12.5%.

In patients with A allele, the mean of the focal lesion size was 3.11 ± 1.21cm, while in patients with G allele the mean lesion size was 3.26 ± 1.14.

As regard portal vein thrombosis, it was detected in 35.7% of patients with A allele, while in patients with G allele portal vein thrombosis was found in 50.0%.

In patients with A allele, tumor metastasis was found in 32.1% with the high percentage in the lung (44.4%), then in the lung and abdominal LN (33.3%) and finally in the head of the pancreas (22.2%). Regarding patients with G allele; tumor metastasis was found in 53.1% with a high percentage in the lung (70.6%) then in the bone (32.5%) and finally in the lung and abdominal LN (5.9%).

There were no significant differences between the 2 alleles and the different triphasic CT findings of HCC.

Table (I): Demographic data AFP serum level (of the three studied groups) , triphasic CT findings and BCLC staging in HCC group:

c²:  Chi-square test            F: ANOVA test   U: Mann Whitney test      H: Kruskal Wallis test,

p: Comparing between the different studied groups

p₁: Comparing between the HCC and Non-HCC groups

p₂: Comparing between the HCC and Control groups

p₃: Comparing between the Non-HCC and Control groups.  

*: Statistically significant at p ≤ 0.05

Table (II): Comparison of the three groups studied in terms of EGF  genotypes and alleles 

c²:  Chi-square test

p: Comparing between the different studied groups

p₁: Comparing between the HCV+HCC and HCV groups

p₂: Comparing between the HCV+HCC and Control groups

p₃: Comparing between the HCV and Control groups

*: Statistically significant at p ≤ 0.05 

Table (III): A comparison between HCC and control groups based on EGF +61 A/G gene

®: Reference

OR: Odd`s ratio 

L.L: Lower limit

U.L: Upper limit

C.I: Confidence interval

*: Statistically significant at p ≤ 0.05

Table (IV): Relation between an allele and BCLC staging in HCC group

c²:  Chi-square test           MC: Monte Carlo

p: Association between an allele and different parameters p-value.

Table (V): HCC group; EGF alleles and triphasic CT findings relation:

c²:  Chi-square test            U: Mann Whitney test      MC: Monte Carlo        

p: Association between an allele and different parameters p-value.

DISCUSSION