Abstract Background: Gastric cancer is among the leading causes of cancer-related deaths worldwide, notably in Asia. Though targeted therapies have been under clinical evaluation, the biological complexity of this cancer type may confer resistance to single-targeted therapies. Especially, scirrhous phenotype is known to be affected by multiple growth factors. To overcome this biological complexity, multi-targeted kinase inhibitors (MTKIs) have been clinically tested and were reported to be effective in a subset of patients. We wanted to elucidate the kinase activity signature of gastric cancer and to explore predictive biomarkers of MTKIs using gastric cancer cell lines by tyrosine kinase activity profiling. Material and methods: The growth-inhibitory effect of the MTKIs sunitinib, sorafenib, pazopanib and the Akt inhibitor MK-2206 was evaluated by the MTT assay. 11 scirrhous and 14 non-scirrhous gastric cancer cell lines were grown until semi-confluent, lysed, aliquoted and stored at -80°C until use. The basal tyrosine kinase activity of the lysates was measured on PamChip® peptide micro-arrays, containing 144 peptides derived from known human phosphorylation sites. Quantification of peptide phosphorylation and data analysis were performed using BioNavigator and SIMCA P+ software. Results: HSC-39, HSC-40A, KATO-III and HSC-43 scirrhous cells and SNU-16 non-scirrhous cells were sensitive to all three MTKIs with IC50 values < 1 μmol/L. Orthogonal partial least square-discriminant analysis (OPLS-DA) of basal tyrosine kinase activity profiles showed a clear segregation between sensitive cell lines and insensitive cell lines. Levels of putative target proteins such as PDGFR, VEGFR, Raf and Kit did not correlate with the sensitivity. All sensitive cell lines turned out to harbor an FGFR2 amplification suggesting that FGFR2 amplification may be a predictive biomarker of response to MTKI treatment. Kinases such as EGFR, EPHA, FGFR, IGF1R, INSR and Src contributed to this segregation suggesting that these are relevant to the sensitivity to MTKIs. MET overexpressing cell lines were resistant to treatment with the four inhibitors. Scirrhous cells were segregated from non-scirrhous cells by OPLS-DA among cell lines with signet-ring cell carcinoma cytology with R2, 0.994; Q2, 0.615. EGFR, ERBB2, 3, 4, EPHAs, FAK, IGF1R, INSR, MAP2Ks, RET and SRC seemed to be responsible for the segregation suggesting that they play an important role in scirrhous gastric cancer biology. Conclusions: Kinase activity profiles reflect the biological complexity of gastric cancer. FGFR2 status correlates with sensitivity to sunitinib, pazopanib and sorafenib, whereas MET status does not. These data imply that tyrosine kinase activity profiling with PamChip® peptide arrays can be utilized to identify molecular targets and predictive biomarkers in gastric cancer and this should be further explored. Citation Format: Yasuhiro Koh, Masakuni Serizawa, Rik de Wijn, Riet Hilhorst, Takako Nakajima, Hirofumi Yasui, Kazuyoshi Yanagihara, Martijn Dankers, Rob Ruijtenbeek, Narikazu Boku. In vitro tyrosine kinase activity profiling to identify molecular targets and predictive biomarkers in gastric cancer cell lines. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 4655. doi:10.1158/1538-7445.AM2013-4655