The p16(INK4a)/CDKN2A gene (p16(INK4a)) is frequently altered by homozygous
deletion, mutation, or methylation in many nonendocrine tumors, and these
alterations may be predictive of recurrence, tumor growth, or aggressivenes
s. Whether this is true of neuroendocrine tumors such as gastrinomas is unc
lear. To address this question we analyzed the gastrinomas from 44 patients
for p16(INK4a) gene mutations and correlated the results to the tumor's bi
ological behavior, growth pattern, and aggressiveness. No gastrinomas had m
utations of exon 1 or exon 2 of the p16(INK4a) gene, although polymorphisms
were found in 54%. No homozygous deletions were found. In 52% of the gastr
inomas, hypermethylation of a 5'-CpG island of the p16(INK4a) gene promoter
was found. To assess the growth behavior of the gastrinomas, all patients
were assessed yearly with at least three conventional imaging studies (comp
uted tomography scan, magnetic resonance imaging, and ultrasound), and sinc
e 1994 have been assessed with radionuclide scanning using [In-111-diethyle
netriamine pentaacetic acid,DPhe(1)] octreotide. The mean follow-up was 5.1
+/- 0.4 yr (range, 1.2-11.7). The presence or absence of methylation of th
e p16(INK4a) gene did not correlate with clinical characteristics of the ga
strinoma, biological behavior (gastrin release and basal or maximal acid ou
tput), the presence or absence of known prognostic factors (tumor size, gas
trinoma location, lymph node metastases, liver metastases, and curability),
or growth pattern of the gastrinoma postresection. These results indicate
that methylation of the p16(INK4a) gene is the most common gene alteration
described to date in gastrinomas. Furthermore, because it is independent of
disease stage it is probably an early event in the pathogenesis and becaus
e it is independent of the primary gastrinoma location, which is now though
t to have different origins, methylation of the p16(INK4a) gene is probably
a central process in the molecular pathogenesis of these tumors.