P-053 MET Overexpression and Amplification Define a Distinct Molecular Subgroup for Targeted Therapies in Gastric Cancer

Jia Wei,Nan Wu,Jie Shen,Juan Du,Xuan Sun,Zi-Han Lin,B. Liu
DOI: https://doi.org/10.1093/annonc/mdv233.53
IF: 51.769
2015-01-01
Annals of Oncology
Abstract:Introduction: Gastric cancer is among the leading causes of cancer deaths. Currently, only trastuzumab, ramucirumab, and lapatinib effectively treat gastric cancer. Thus, additional novel targets are required for this disease. Methods: We investigated the immunohistochemical and fluorescence in situ hybridization expression of MET, ROS1, and ALK in four gastric cell lines and a cohort of 98 gastric cancer patients. Crizotinib response was studied using an in vitro histoculture drug response assay and the patient-derived GC xenograft model in vivo. Gene expression status was also analyzed for association with overall survival. Results: Crizotinib potently inhibited in vitro cell growth in only one cell line, which also showed MET amplification. A positive correlation between crizotinib sensitivity and MET overexpression was observed (P = 0.045) in the histoculture drug response assay. Meanwhile, patient-derived tumor xenograft mouse models transplanted with tissues with higher MET protein expression displayed a highly selective sensitivity to crizotinib. In the 98 patients, MET overexpression was found in 42 (42.9%) and MET was amplified in 4 (4.1%). ROS1 and ALK overexpression were found in 25 (25.5%) and 0 patients, respectively. However, none of the patients screened harbored ALK or ROS1 rearrangements. There was no significant association found between overall survival and the MET or ROS1 status. We also observed a case that one advanced GC patient with MET-amplification experienced tumor shrinkage (PR by RECIST) after 3 weeks treated with crizotinib at the time of third-line chemotherapy failure. This patient also showed rapid clinical improvement, with decreased pain and improved performance status. Conclusion: Crizotinib may induce clinically relevant anticancer effects in MET-overexpressing or -amplified gastric cancer patients.
What problem does this paper attempt to address?