Back to Annual Meeting Program

Context
Overexpression of epidermal growth factor (EGF) in the liver induces transformation to hepatocellular carcinoma in animal models. Polymorphisms in the EGF gene modulate EGF levels.

Objective
To assess the relationship among human EGF gene single-nucleotide polymorphism, EGF expression, and risk of hepatocellular carcinoma.

Design, Setting, and Participants
Molecular mechanisms linking the 61*G allele polymorphism to EGF expression were examined in human hepatocellular carcinoma cell lines and human liver tissue. A case-control study involving 207 patients with cirrhosis was conducted at the Massachusetts General Hospital (1999-2006) and a validation casecontrol study involving 121 patients with cirrhosis was conducted at Hôpital Paul Brousse (1993-2006). Restriction fragment-length polymorphism was used to determine the EGF gene polymorphism genotype. Logistic regression analysis was used to assess the association between the EGF polymorphism and hepatocellular carcinoma risk.

Main Outcome Measures
Mechanisms by which the EGF gene polymorphism modulates EGF levels and associations among EGF gene polymorphism, EGF levels, and hepatocellular carcinoma.

Results
Transcripts from the EGF 61*G allele exhibited more than a 2-fold longer halflife than those from the 61*A allele, and EGF secretion was 2.3-fold higher in G/G hepatocellular carcinoma cell lines than A/A cell lines. Serum EGF levels were 1.8-fold higher in G/G patients than A/A patients, and liver EGF levels were 2.4-fold higher in G/G patients than A/Apatients. Among the 207 patients with cirrhosis in the Massachusetts study population, 59 also had hepatocellular carcinoma. Analysis of the distribution of allelic frequencies revealed that there was a 4-fold odds of hepatocellular carcinoma in G/G patients compared with A/A patients in the Massachusetts study population (odds ratio, 4.0;95% confidence interval [CI], 1.6-9.6; P=.002). Logistic regression analysis demonstrated that the number of copies of G was significantly associated with hepatocellular carcinoma after adjusting for age, sex, race, etiology, and severity of cirrhosis (G/G or A/G vs A/A; hazard ratio, 3.49; 95% CI, 1.29-9.44; P=.01). The significant association was validated in the French patients with alcoholic cirrhosis and hepatocellular carcinoma.

Conclusion
The EGF gene polymorphism genotype is associated with risk for development of hepatocellular carcinoma in liver cirrhosis through modulation of EGF levels.