34P - Predicting tissue c-Met status using plasma soluble c-Met levels in patients with advanced NSCLC

Date 19 December 2015
Event ESMO Asia 2015 Congress
Session Poster presentation 1
Topics Non-Small-Cell Lung Cancer, Metastatic
Pathology/Molecular Biology
Translational Research
Presenter Hongfei Gao
Citation Annals of Oncology (2015) 26 (suppl_9): 8-15. 10.1093/annonc/mdv518
Authors H. Gao1, Z. Chen2, A. Li2, J. Yang2, X. Zhang2, Z. Xie2, J. Su2, J. Lin1, Y. Li1, Y. Wu2, Z. Dong2
  • 1Guangdong Lung Cancer Institute,guangdong Cardiovascular Institute, Guangdong General Hospital, 510080 - Guangzhou/CN
  • 2Guangdong Lung Cancer Institute, Guangdong General Hospital and Guangdong Academy of Medical Sciences, 510080 - Guangzhou/CN



Immunohistochemistry (IHC) and fluorescent in situ hybridization (FISH) are reliable ways to identify overexpression of c-Met protein or amplification of the c-Met gene, but each technique requires a high-quality tissue sample, which may not be available. The aim of this study is to investigate whether plasma soluble c-Met level can be used to predict tissue c-Met status in patients with advanced non-small cell lung cancer (NSCLC).


In 198 patients with advanced NSCLC, we determined c-Met expression by IHC analysis on tumors. The expression was scored according to MetMAb scoring criteria. Plasma concentration of soluble c-Met protein was measured with a human soluble c-Met quantitative enzyme-linked immunosorbent assay (ELISA) kit, and the predictive values were determined based on the receiver-operating characteristic (ROC) curves analysis.


Of the total 198 patients, 103 (52%) patients were tissue c-Met negative, 95(48%) were tissue c-Met positive. The average plasma c-Met concentration was 682.8 ± 155.2 ng/mL in the tissue c-Met-negative group, significantly lower than 886.5 ± 307.4 ng/mL in tissue c-Met- positive group (P < 0.001). ROC curve analysis showed 66.9% specificity and 64.2% sensitivity for predicting tissue c-Met positivity at 755ng/mL of plasma c-Met. AUC for plasma c-Met was 0.724 (95% confidence interval; 0.654–0.794, P < 0.001).


We suggest 755 ng/mL in the plasma as a cutoff for predicting tissue c-Met status with moderate specificity and sensitivity. This information may be useful when the tumour tissue is not available.

Clinical trial identification


All authors have declared no conflicts of interest.