1651P - Evaluation of tumor MET expression in chinese patients (pts) with advanced gastric or gastroesophageal junction (g/gej) cancer

Date 28 September 2014
Event ESMO 2014
Session Poster Display session
Topics Oesophageal Cancer
Gastric Cancer
Pathology/Molecular Biology
Basic Scientific Principles
Presenter Lin Shen
Citation Annals of Oncology (2014) 25 (suppl_4): iv564-iv573. 10.1093/annonc/mdu359
Authors L. Shen1, Z. Peng1, Z. Li2, J. Gao1, M. Lu1, J. Gong1, K.S. Oliner3, Y. Hei4, H. Zhou5
  • 1Department Of Gastrointestinal Oncology, Peking University Cancer Hospital & Institute, 100142 - Beijing/CN
  • 2Department Of Pathology, Peking University Cancer Hospital & Institute, Beijing/CN
  • 3Molecular Sciences, Amgen Inc., Thousand Oaks/US
  • 4Global Development, Amgen Inc., Shanghai/CN
  • 5Medical Department, Amgen Inc., Shanghai/CN



Gastric cancer accounts for >325,000 deaths in China annually. MET and its ligand hepatocyte growth factor are promising targets in G/GEJ cancer. This study evaluated overall survival (OS) by tumor MET expression and by treatment disposition in Chinese pts with advanced G/GEJ cancer.


Baseline formalin-fixed, paraffin-embedded tumor samples from pts with stage IV unresectable G/GEJ cancer enrolled in clinical trials during 2008–2010 were assessed for MET protein levels and MET gene copy numbers by immunohistochemistry (IHC; Dako) and fluorescence in situ hybridization (FISH; Dako), respectively. MET-positive: membrane protein staining in ≥25% tumor cells. MET-amplified: MET:centromere ratio ≥2.0. We tested the association between Kaplan-Meier OS and MET status (log-rank test) and clinical parameters (Fisher's exact), and the association between MET expression and amplification (Fisher's exact).


168 pts were eligible; 137 had evaluable IHC samples. OS by MET status is shown (Table). Of 113 pts with evaluable FISH samples, 8 were MET-amplified. MET expression and amplification were not associated with sex, primary tumor site, prior surgery, lung or liver metastases, or number of metastases. Lauren classification was associated with MET amplification (P=0.038) but not MET expression (P=0.446). In 95 pts with evaluable IHC and FISH samples, MET expression was associated with MET amplification (P<0.001); all MET-amplified pts had ≥90% MET expression.

Patients Evaluable for IHC MET-positive MET-negative P-valued
Overall, n 53 84
OS analysis, n 51 82
Median OS, mo 12.3 10.9 0.799
First-line platinum-based,a n 35 47
OS analysis, n 33 47
Median OS, mo 11.3 11.8 0.367
First-line platinum-based only,b n 30 34
OS analysis, n 28 34
Median OS, mo 10.6 11.9 0.119
First-line taxane-based,a,c n 23 50
OS analysis, n 23 48
Median OS, mo 12.6 10.8 0.259

aIncludes platinum- and taxane-based triplet (n=18). bExcludes platinum- and taxane-based triplet (n=18). cIncludes paclitaxel alone (n=1). dLog-rank test.


Tumor MET expression was not significantly associated with OS in this population of Chinese pts with advanced G/GEJ cancer. The study was limited by the small sample size and unbalanced treatment disposition between MET-positive and MET-negative pts. Prospective analysis of potential prognostic markers in Chinese pts with advanced G/GEJ cancer is needed.


K.S. Oliner: employment by and stock ownership in Amgen Inc.; Y. Hei: employment by and stock ownership in Amgen Inc.; H. Zhou: employment by and stock ownership in Amgen Inc. All other authors have declared no conflicts of interest.