In in vitro experiments, high hENT1 mRNA levels have been shown to be associated with GEM sensitivity, as represented by IC50 values [20, 21]. In cells, GEM is phosphorylated to its active metabolites by dCK. Several reports have suggested that high dCK enzyme activity may contribute to GEM sensitivity in experimental settings [5] and surgical samples [6]. However, GEM is inactivated by deamination, as catalyzed by DCD. CDA and 5′-NT are also a catabolic enzymes of GEM. Therefore, resistance to GEM

may be induced by increased activity of DCD, CDA or 5′-NT [3, 5, 22]. Ribonucleotide reductase, which consists of dimerized large and small RRM1 and RRM2 subunits, is the rate-limiting enzyme for DNA synthesis, as it is the only known enzyme that converts

ribonucleotides to deoxyribonucleotides. GEM exerts BB-94 its cytotoxicity by inhibiting ribonucleotide reductase. High expression of RRM1 and RRM2 has been suggested to be a mechanism of GEM resistance [22–26]. Thus, several metabolic enzymes and nucleoside transporters have been suggested to affect GEM sensitivity. FDA analysis may therefore be suitable to identify predictors of GEM efficacy by using a very small quantity of samples taken by EUS-FNA from unresectable pancreatic cancer, as it can simultaneously assess the expression of multiple mRNAs related to GEM sensitivity. Our results suggested that high dCK mRNA expression is a predictor of GEM efficacy. In these experimental settings, RNA from most samples were subjected to FDA analysis Necrostatin-1 molecular weight and were not subjected to further assessment. However, to confirm the relationship between dCK mRNA expression Thiamet G and GEM efficacy, quantitative measurement of expression by real-time reverse transcription-polymerase chain reaction is required. In this study, other GEM sensitivity-related gene expressions including hENT-1 could not be proved to be predictors for GEM efficacy. However, these gene expressions may not be totally denied as predictors of GEM efficacy by the present study using small number of samples.

The contamination of Stem Cells inhibitor normal tissue into tumor tissue obtained by EUS-FNA may also be a major obstacle to an accurate analysis. Microdissection technique for EUS-FNA sample might be required to avoid the normal tissue contamination. Conclusion In conclusion, dCK mRNA expression in EUS-FNA biopsy specimens may be a predictor for response to GEM in patients with unresectable pancreatic cancer. The FDA used in this study also contained molecular target genes that may be promising for the treatment of pancreatic cancer. These data may be helpful for future cancer treatments that target specific molecules. Acknowledgements We would like to thank Masakazu Fukushima of the Tokushima Research Center for his scientific advice. This study is supported by Ministry of Education, Culture, Sports, Science and Technology of Japan, Grant-in-Aid for Scientific Research (C) 19590317. References 1.