Pharmacogenomically Selected Treatment for Gastric and Gastroesophageal Junction (GEJ) Tumors

Gastric and gastroesophageal junction (GEJ) cancers are a leading cause of cancer mortality. Despite the development of newer chemotherapies, the response rates and median survival in patients with these tumors has remained essentially stagnant. Defining host and molecular/biologic tumor characteristics to customize treatment may lead to improved survival outcomes. Retrospective studies have identified genetic markers that predict treatment outcome. However, there have been no prospective studies in gastric and GEJ cancer evaluating the clinical utility of these genetic factors. We hypothesize that genomically based treatment will improve the expected response rate in patients with gastric and GEJ cancers. We propose a prospective, multi-institutional Phase II clinical trial testing a germline polymorphism in the thymidylate synthase (TS) gene, the number of tandem repeats in the TS enhancer region (TSER) as a treatment selection marker. The polymorphic variant conferring three tandem repeats (TSER\*3) has been associated with 5-FU resistance due to high tumor TS expression in comparison to the TSER\*2 variant (two tandem repeats). The TSER\*3 polymorphism is common (allelic frequency of 0.5-0.8). In the proposed study, we will prospectively genotype patients with gastric and GEJ cancers. Patients who are expected to be 5-FU sensitive (carrying a TSER\*2 allele) will receive a 5-FU containing regimen (5-FU, leucovorin, oxaliplatin). Patients who are expected to be 5-FU resistant (homozygous for TSER\*3) will not be included in the study. In completing this study, we will determine whether treatment selection based on germline TSER polymorphism status improves the response rate in patients with metastatic gastric and GEJ tumors. Additional correlative studies are proposed to identify confounding factors that may alter the expected outcomes of this treatment approach.