ScienceDaily (Jan. 2, 2008)

　"If these results are confirmed, this EGF variation could be used to determine which cirrhotic
patients should be screened more intensively for tumor development," says Kenneth Tanabe, MD,
chief of Surgical Oncology at the MGH Cancer Center, the study's lead author. "In addition, the
molecular pathway controlled by EGF and its receptor EGFR -- which is known to be important in
several types of cancer -- appears to be an excellent target for chemoprevention studies. This is a
deadly cancer and so progress in prevention and early detection is critically important."

　HCC is the sixth most common solid tumor worldwide and most commonly develops in individuals
with cirrhosis, which may be caused by infection with the hepatitis B or C viruses. There are currently
no effective treatments for most HCC patients, so there is considerable interest in strategies that may
prevent development of the tumor.

　EGF's normal function is to stimulate tissue growth. Animal studies have shown that elevated levels
of this protein in the liver lead to tumor development and that blocking the protein's receptor can
prevent development of liver cancer. The current study was designed to determine whether cirrhotic
patients with higher EGF levels are at greater risk for liver cancer and to determine the influence of a
particular inherited gene on EGF levels in cirrhotic patients.

　The researchers focused on a known variation in the EGF gene -- the presence of the nucleotide
guanine (G) instead of the more common adenine (A) in a particular location -- which has been shown
to increase EGF secretion in blood cells and raise the risk for malignant melanoma. Individuals inherit
one copy of the gene from each parent and therefore have this gene with either two copies of A (A/A),
two copies of G (G/G), or one copy of each (A/G). Genetic analysis of liver tumor cell lines showed
that messenger RNA transcribed from DNA strands with the G allele was more stable that that
transcribed from the A version, which could explain why cells with two G copies tend to secrete higher
levels of EGF.

　The researchers then studied tissue samples from all patients in the MGH Cancer Center Tumor
Bank who had cirrhosis. Among the 207 patients with cirrhosis, most of whom were infected with the
hepatitis C virus, 59 also had HCC. Patients with at least one copy of the G nucleotide had a
significantly higher risk of developing HCC than did A/A patients -- ranging from a more than twofold
increase for those with one G to an over fourfold increase for those with two G alleles. In all three
genotypes, tissue analysis showed that EGF levels were highest in the G/G patients, as was activation
of the EGFR receptor. In addition, blood levels of EGF were highest in those with two copies of the G
allele.

　To confirm these finding in a different patient population, the MGH team worked with colleagues
from the Paul Brousse Hospital in Paris. Samples from this group, all of whom had alcoholic cirrhosis,
also showed that patients with the G/G version of the EGF gene had a significantly greater risk of
developing the liver tumor than did the A/A patients, in this instance an almost threefold risk increase.

　In both the MGH and French study groups, controlling for factors such as age and gender did not
change the increased risk associated with the G allele. While both groups primarily consisted of
Caucasian patients, in the MGH group, it was noted that the G allele was more common among Asian
patients; and it is well known that more than half the cases of HCC worldwide occur in China.

　"We now need to prospectively study EGF levels in cirrhotic patients, to see if elevated levels will
correlate with a greater risk of developing HCC, and look at factors such as diet, drugs or ethnicity that
may modulate EGF levels," Tanabe says. "I think this is a terrific opportunity to see if targeting a
specific pathway will prevent HCC in this group of patients, who are at risk for liver cancer because of
their cirrhosis." Tanabe is an associate professor of Surgery at Harvard Medical School.

　This research was published in the January 2 issue of JAMA.
　The study was supported by grants from the National Institutes of Health, the MGH Department of
Surgery, Tucker Gosnell Gastrointestinal Cancer Center, and the Fund for Medical Discovery. Co-
authors of the JAMA article are Dianne Finkelstein, PhD, Hiroshi Kawasaki, Tsutomu Fujii, MD,
PhD, Raymond Chung, MD, Gregory Lauwers, MD, Yakup Kulu, Alona Muzikansky, Darshini
Kuruppu, PhD, Michael Lanuti, MD, Jonathan Goodwin and Bryan Fuch, PhD, of the MGH; and
Antoinette Lemoine, MD, and Daniel Azoulay, MD, PhD, Paul Brousse Hospital, Paris.

　Adapted from materials provided by Massachusetts General Hospital.