Chemoresistance-Related Genes Profile in Specimens Obtained By Endoscopic Ultrasound-Guided Fine-Needle Aspiration (EUS-FNA)

Abstract

Background: Chemotherapy using gemcitabine or 5-FU in advanced pancreatic cancer is often limited by development of side effects and drug resistance. The inter-individual variability of these events may have a genetic basis. DNA microarray systems can measure simultaneous expression of multiple genes in cancer patients. EUS-FNA, which is established as the preferred method to confirm a diagnosis of pancreatic cancer, can also be used to provide pancreatic cancer tissue for microarray analysis. Methods: Pancreatic cancer tissue was obtained from 10 patients with advanced disease by EUS-FNA, using a 22-gauge aspiration needle or 19-gauge trucut needle. Total mRNA was harvested from each sample which was amplified to provide obtain adequate mRNA for the analysis. The gene profile of DNA array was restricted to 133 genes including chemoresistance-related genes. (Focused DNA ArrayTM, Taiho Pharmaceutical Co., Ltd., Tokushima, Japan). Result: In all cases, adequate tissue samples were obtained and DNA microarray assays were successfully achieved. The gene amounts were expressed as the relative values to the individual house keeping gene amount. The mean ± SD of the 10 samples were as follows. Gemcitabine-resistance-related genes: dCMP deaminase, 4.72 ± 1.28; RRM1, 0.45 ± 0.10. 5-FU-resistance-related genes: TS, 7.50 ± 3.38; DPD, 1.41 ± 0.30. One of four patients treated with gemcitabine alone showed partial response (PR) and has survived for 8 months (M). The patient with the highest dCMP deaminase (7.41) among 10 patients and the other patient with the highest RRM1 (0.64) resulted in progress disease (PD) and died within 8 M. One of two patients treated with gemcitabine + S-1 (oral fluorouracil derivative containing a DPD inhibitor) showed PR and has been alive for 11 months. The dCMP deaminase, RRM1, and DPD were almost same and average level in the two patients, however, the TS level (8.07) in PD patient was higher compared to that (4.9) in PR patient. Conclusion: DNA microarray-based gene expression profiling combined with EUS-FNA may be a promising tool to predict chemoresistance in advanced pancreatic cancer. Such information may be useful to select individualizing chemotherapeutic regimens based on each patient's genotype.