Kl. Burns et al., MOLECULAR-GENETIC CORRELATES OF P16, CDK4, AND PRB IMMUNOHISTOCHEMISTRY IN GLIOBLASTOMAS, Journal of neuropathology and experimental neurology, 57(2), 1998, pp. 122-130
The vast majority of glioblastomas have CDKN2A, CDK4, or RE gene alter
ations that perturb the p16-cdk4-pRb cell cycle regulatory cascade. To
explore whether immunohistochemical methods provide an alternative me
ans of assessing this pathway, we studied 25 glioblastomas using a com
bination of molecular genetic and immunohistochemical assays. Homozygo
us deletion of the CDKN2A gene was detected in 12 of 25 (48%) cases, C
DK4 amplification in 4 of 25 (16%) tumors, and loss of heterozygosity
at the RE gene in 8 of 22 (36%) informative cases. Five of 25 (20%) gl
ioblastomas had diffuse p16 immunohistochemical positivity. Significan
tly, all of these had either CDK4 amplification or RE LOH, suggesting
that p16 immunopositivity only occurs in those tumors with alterations
of another component in the pathway. Nineteen (76%) cases were unifor
mly immunonegative for p16, and 12 (48%) had CDKN2A homozygous deletio
ns, but the remaining 7 cases lacked CDKN2A deletions, mutations and p
romoter methylation. All glioblastomas stained diffusely for cdk4, irr
espective of CDK4 gene amplification status. Extensive pRb staining wa
s present in most cases that maintained both RE alleles, and absent in
most cases with RE loss. but there were notable discrepancies. Thus,
p16 and pRb immunohistochemistry cannot replace molecular genetic anal
ysis of this critical regulatory cascade; instead, the combined result
s hint at complex regulation of this cell cycle checkpoint. From a pra
ctical point of view, although p16 immunonegativity does not necessari
ly indicate CDKN2A deletion, diffuse positive p16 immunostaining stron
gly suggests either CDK4 amplification or RE loss and excludes CDKN2A
deletion.