Nuclear Overxpression of Phosphorylated Epidermal Growth Factor Receptor as an Independent Poor Prognosticator in Gallbladder Carcinoma
CF Li, FM Fang, JM Wang, HY Huang. Chi-Mei Hosp., Tainan, Taiwan; Chang Gung Memorial Hosp., Kaohsiung, Taiwan; NCKU, Tainan, Taiwan
Background: Advanced gallbladder carcinoma (GBCA) is scarcely curable, requiring better understanding of molecular aberrations to aid in its management. Despite small-molecule inhibitors of EGFR proved useful in treating lung adenocarcinomas, littlie is known about the precise mechanism of deregulated EGFR signaling in GBCA. Moreover, prior study of GBCA never examined the occurrence and significance of EGFR nuclear import, a novel pathway found in other cancers to regulate gene expression.
Design: Immunoexpression localization and level of EGFR and phosphorylated EGFR (pEGFR) were assessable for 104 GBCAs on tissue microarrays, 76 of which were successfully analyzed for EGFR gene status by both chromogenic in situ hybridization and mutant-enriched PCR targeting exons 19 and 21, respectively. The results were correlated with clinicopathological factors and disease-specific survival (DSS). Fractional western blotting and confocal immunofluoresce were used to substantiate the genuine nuclear localization of EGFR in 3 GBCA cell lines (SNU308, RCB1129, and RCB1130).
Results: EGFR and pEGFR were overexpressed in the cytoplasm (C) in 22 and 72 cases and in the nuclei (N) in 25 and 62 cases, respectively. EGFR amplification (amp) was identified in 11 cases (15%), while no case showed deletion of exon 19 or L858R mutation of exon 21. To varying extent, all cell lines demonstrated endogenous nuclear localization of EGFR. EGFR amp and overexpressed EGFR and pEGFR were variably associated with significant clinicopathological prognosticators, including AJCC stage II-IV with C-EGFR (p=0.036) and N-EGFR (p=0.010), pT3-4 stage with N-EGFR (p=0.015) and EGFR amp (p=0.045), and vascular invasion with N-EGFR (p=0.003) and N-pEGFR (p=0.031). Notably, EGFR amp was strongly related to overexpressed C-EGFR alone (p<0.001) but not correlated with DSS. Although both N-EGFR (p=0.0014) and N-pEGFR (p=0.0017) were adverse poor prognosticators univariately, only pEGFR independently predict worse DSS (p=0.0468, HR=2.024), together with AJCC stage II-IV (p=0.0132, HR=2.216) and old age (p=0.0428, HR=1.785).
Conclusions: EGFR amplification is present in a minor subset of GBCA but not prognostically useful. It is not rational to treat refractory GBCA with small-molecule EGFR inhibitor, given absence of mutation in exons 19 and 21. Nuclear import of EGFR truly occurs in GBCA and confers clinical aggressiveness with N-pEGFR overexpression being identified as an independent prognosticator.
Tuesday, March 23, 2010 9:30 AM
Poster Session III # 131, Tuesday Morning