e., protease and/or nuclease, nucleic acids could easily be degraded. Furthermore, since EUS-FNA specimens are long and thin, they may be easily broken down by digestive enzymes. On the other hand, a reagent such as an RNase inhibitor included in RNAlater® may be easy to instill into the tissues and/or their cell components for the same reason. Therefore, EUS-FNA specimens may be suitable for storage with RNAlater® for RNA preparation. In our investigation, the analyzable rate was lower than 50% for EUS-FNA specimens of RNAlater® storage (46%). For further improvement, it will be very important to take as many cell-rich EUS-FNA

specimens as possible. Actually, specimens that we couldn’t obtain from contained much fibrotic tissue or blood instead of cells (data Hydroxychloroquine molecular weight not shown). After EUS-FNA, confirmation of the cell component by microscopic observation and preservation of only cell-rich part with RNAlater®

cutting off from the obtained specimens will be efficient before RNA preparation. In pancreatic juice samples, total RNA and DNA were obtained in good quality and quantity from the directly frozen samples. RNAlater® storage could not improve quality of nucleic acid in pancreatic juice. All those samples involved white pellet. We suspected that the component of white pellet was a contrast agent contained in the pancreatic juice samples. To confirm it, we mixed RNAlater® and the contrast agent Urografin®, and the white pellets like in RNAlater®-stored samples were observed immediately. Copanlisib molecular weight Furthermore, the volume of the white pellets appeared were almost the same as that of Urografin®. The contrast agent is difficult to be dissolved, therefore, when it is mixed with different solution such as RNAlater®, its composition changes and the contrast agent

may precipitate. If we use RNAlater® for pancreatic juice storage, we have to remove the supernatant containing a contrast agent such as Urografin®, for example, by performing centrifuge. After then, only the precipitation including pancreatic cells should be stored with RNAlater®. Furthermore, control experiments with RNase inhibitors other than RNAlater® to exclude the possible vehicle effects will be needed. Pancreatic juice is an ideal specimen for pancreatic cancer biomarkers discovery, only because it is an exceptionally rich source of proteins released from pancreatic cancer cells [16–18]. Gene analysis of pancreatic juice deserves further investigation to determine its utility as a tool for the evaluation of pancreatic lesions. It may be presumed that FNA samples and pancreatic juice samples were classified into different clusters because the cell population is different in the two kinds of samples. The gene expression data obtained in this study succeeded in classifying cancer and non-cancer in the EUS-FNA samples. However, the pancreatic juice samples were not classified as any particular cluster.