OBJECTIVES: The aim of this study was to identify biomarkers for predicting the efficacy of docetaxel, cisplatin, and S-1 (DCS) therapy for advanced gastric cancer using microarrays of biopsy specimens before chemotherapy. METHODS: Nineteen samples were taken from 19 patients with unresectable metastatic gastric cancer who received DCS as a first-line therapy. Laser capture microdissection was performed, and total cellular RNA was extracted from each microdissected sample. Whole-gene expression was analyzed by microarray, and the difference in mRNA expression observed with the microarrays was confirmed by quantitative real-time PCR. Immunohistochemical staining was performed using clinical tissue sections obtained by endoscopic biopsy. RESULTS: Eleven patients were identified as early responders and 8 patients as nonresponders to DCS therapy. Twenty-nine genes showed significant differences in relative expression ratios between tumor and normal tissues. A classifier set of 29 genes had high accuracy (94.7%) for distinguishing gene expression between 11 early responders and 8 nonresponders. Decreasing the size of the classifier set to 4 genes (PDGFB, PCGF3, CISH, and ANXA5) increased the accuracy to 100%. Expression levels by real-time PCR for validation were well correlated with those 4 genes in microarrays. CONCLUSION: The genes identified may serve as efficient biomarkers for personalized cancer-targeted therapy.
OBJECTIVES: The aim of this study was to identify biomarkers for predicting the efficacy of docetaxel, cisplatin, and S-1 (DCS) therapy for advanced gastric cancer using microarrays of biopsy specimens before chemotherapy. METHODS: Nineteen samples were taken from 19 patients with unresectable metastatic gastric cancer who received DCS as a first-line therapy. Laser capture microdissection was performed, and total cellular RNA was extracted from each microdissected sample. Whole-gene expression was analyzed by microarray, and the difference in mRNA expression observed with the microarrays was confirmed by quantitative real-time PCR. Immunohistochemical staining was performed using clinical tissue sections obtained by endoscopic biopsy. RESULTS: Eleven patients were identified as early responders and 8 patients as nonresponders to DCS therapy. Twenty-nine genes showed significant differences in relative expression ratios between tumor and normal tissues. A classifier set of 29 genes had high accuracy (94.7%) for distinguishing gene expression between 11 early responders and 8 nonresponders. Decreasing the size of the classifier set to 4 genes (PDGFB, PCGF3, CISH, and ANXA5) increased the accuracy to 100%. Expression levels by real-time PCR for validation were well correlated with those 4 genes in microarrays. CONCLUSION: The genes identified may serve as efficient biomarkers for personalized cancer-targeted therapy.