Introduction • Localized form of gastric cancer is currently treated, independently of genetic profiles, by surgical resection with perioperative chemotherapy, radiotherapy and/or targeted therapy. Next-generation sequencing (NGS) recently provided new information regarding gene mutations related to gastric cancer pathogenesis. this information can determine new therapeutic pathways to successfully treat gastric cancer depending on its molecular profiling. Ramucirumab is a humanized monoclonal antibody directed against vascular endothelial growth factor receptor 2. Ramucirumab is approved in second-line therapy for advanced or metastatic gastric cancer as a single agent or combined with chemotherapy in molecularly unselected patients. Some studies determined prognostic value of plasma biomarkers and cytokines in patients treated with Ramucirumab (RAINBOw trial). Other studies have shown no relationship between Ramucirumab therapy and vEGFR-2 mutation (REGARD trial). however, no studies have shown significant survival benefit for certain genetic profiles in patients treated with Ramucirumab therapy. the aim of this study is to assess the value of certain molecular profiles in predicting response to Ramucirumab therapy in advanced or metastatic gastric cancer. Methods • this is a retrospective study to determine molecular profiles for gastric cancer that predict outcomes of treatment with Ramucirumab. Patients’ molecular profiling will be studied using NGS. Included patients are those with advanced or metastatic gastric cancer treated with second-line Ramucirumab therapy combined with Paclitaxel. Patient characteristics are collected retrospectively from medical records including tumor localization, size, stage, type of surgery, definitive histological subtype, number of lymph nodes resected. type and response to first line chemotherapy as well as overall survival (OS) and diseasefree survival (DFS) are acquired for all patients.

Results• kaplan-Meier curves for DFS and OS will be compared between subjects with different somatic mutations using log-rank test. Multivariate Cox regression models for DFS and OS with mutated somatic genes as independent variables were computed. Any predictive factor for response to Ramucirumab, especially molecular, will be reported.

Conclusion • Molecular characteristics of advanced gastric cancer patients will be analyzed to define any predictive value for response to Ramucirumab in second-line based therapy.

Keywords: gastric cancer; Ramucirumab; genomic profile