Abstract 3413: Assessment of recurrence risk stratification markers after curative resection of stage II or III gastric cancer and potential clinical applications

• Published in: Cancer research (Chicago, Ill.) Ill.), 2015-08, Vol.75 (15_Supplement), p.3413-3413
• Main Authors: Oshima, Takashi, Sakamaki, Kentaro, Miyagi, Yohei, Shiozawa, Manabu, Kunisaki, Chikara, Yoshikawa, Takaki, Rino, Yasushi, Imada, Toshio, Masuda, Munetaka
• Format: Article
• Language: English
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/1538-7445.AM2015-3413

Objective: Standard treatment for stage II or III gastric cancer is curative resection followed by adjuvant chemotherapy. However, treatment outcomes are expected to be further improved by individualized therapy based on biomarker analysis. We extracted mRNA from frozen specimens of gastric cancer to establish a cDNA bank and have analyzed markers for recurrence risk stratification of after curative resection of stage II or III gastric cancer. We report our currently available results. Methods: We studied 255 patients who underwent curative resection of stage II or III gastric cancer and were followed up for at least 5 years after surgery. The patients were divided into 2 cohorts: a training set consisting of 145 patients and a validation set consisting of 110 patients. A total of 104 genes were selected as candidate biomarkers on the basis of comprehensive screening by DNA microarray, extraction by SAGE library analysis, and target molecules and their families, including the results of preclinical research. The relative expression levels of each gene in gastric cancer tissue were measured by quantitative polymerase chain reaction. Genes that were found to be independent predictors of poor outcomes in the training set were verified in the validation set, using Cox proportional-hazards models. Results: In the training set of stage II or III gastric cancer, 38 genes were selected as independent predictors of poor outcomes. On verification analysis using the validation set, the following 11 genes were selected as candidate markers for recurrence risk stratification (P