- Email: [email protected]
Su1279 Factors of Endoscopic Ultrasound Fine-Needle Aspiration for Successful Next-Generation Sequencing in Pancreatic Cancer
Abstracts
pancreatic cysts is exceedingly important yet remains difficult. The relative viscosity of pancreatic cyst fluid has proved useful for distinguishing mucinous from non-mucinous cysts and is an indirect measure of cyst-fluid mucin, glycoproteins, and DNA content. In previous studies, the viscosity was assessed by the string sign (elongation viscosity), a surrogate marker of cyst fluid viscosity. However, the method lacks a theoretical framework for predicting the viscoelastic nature of the cyst fluid, a rheological property of complex physiologic fluids having two components, viscosity and elasticity. The aim of the present study was to assess the utility of the rheological properties of pancreatic cyst fluid, as compared with cytology and biochemical analysis in differentiating pancreatic cyst types. Patients and Methods: EUS-FNA was performed on consecutive subjects with pancreatic cysts. In addition to routine cyst fluid analyses (biochemical, cytological and viscosity assessment using the string sign), a rheological behavior curve of the cyst fluid was generated (steady-state shear viscosity at different shear rates measured by a rotational rheometer (Discovery DHR-2, TA Instruments, USA)). Pancreatic cysts were classified as mucinous (mucinous cystadenoma, mucinous adenocarcinoma, IPMN) or non-mucinous (serous cystadenoma, pseudocyst) based on surgical and/or clinical outcome (presentation, follow-up, imaging and fluid analyses). Results: Fifteen subjects were evaluated (9 females, median age 53y). Mean cyst size was 33.1 mm (range 20-60 mm). Six mucinous and 9 nonmucinous cysts were diagnosed. Four mucinous cysts were diagnosed surgically. Median follow-up was 18 months. Three different patterns of rheological behavior curves were identified (figure): Pattern I (constant viscosity <1.6 units), Pattern II (rapidly decreasing viscosity to <1.6 units) and Pattern III (slowly decreasing viscosity >1.6 units). Patterns I, II were correlated with non-mucinous cysts and Pattern III was correlated with mucinous cysts. This curve pattern analysis was evaluated as a new diagnostic test. The concordance between the string sign and the measured relative viscosity was low (pZ0.62, chi-square). The sensitivity, specificity, positive and negative predictive values and accuracy of string sign, elevated CEA (>192 ng/ml), cytology and rheology (curve patterns I, II-nonmucinous, curve pattern III-mucinous) are presented in the table. Limitation: Small cohort, surgical diagnosis was not available in all cases. Conclusion: The accuracy of the string sign, CEA and cytology for mucinous cysts are comparably inadequate. Cyst fluid rheological analysis appears to accurately differentiate pancreatic cyst types. Results:
Su1279 Factors of Endoscopic Ultrasound Fine-Needle Aspiration for Successful Next-Generation Sequencing in Pancreatic Cancer Dong Hyo Noh, Joo Kyung Park, Kyu Taek Lee, Kwang Hyuck Lee*, Jong Kyun Lee, Keum Bit Hwang, Ju-Il Yang, Jun Young Kim Medicine, Samsung Medical Center, Seoul, Korea (the Republic of) Background: Recent developments in the understanding of the genetics of pancreatic ductal adenocarcinoma (PDAC) have led to the potential for a personalized approach in PDAC patients. Several studies have described the feasibility of generating genetic profiles of pancreatic tumor with next generation sequencing (NGS) of endoscopic ultrasound-fine needle aspiration (EUS-FNA) samples. Aim: The aim of this study was to find the best way of EUS-FNA for successful NGS in PDAC. Methods: We tried to perform NGS with tissues from one-hundred ninety eight patients who was histologically proven PDAC by EUS-FNA at Samsung Medical Center between November 2011 and February 2015. Clinical targeted seqeuncing of 83 genes was perfomed with DNA from EUS-FNA samples by Illumina HiSeq 2500 sequencing platform. Retrospectively, the medical records of these patients were reviewed including tumor factors (size, location, T stage), EUS-FNA factors (gauge of needle, needle type, number of needle passes) and histologic factors (cellularity, blood contamination). Results: NGS could be done successfully in one-hundred nine (55.1%). In univariate analysis, larger gauge of needle (Odds ratio (OR) 2.621, 95% Confidential interval (CI) 1.234w5.191; pZ0.007) and location of tumor in body/tail (OR 2.336, 95% CI 1.296w4.210; pZ0.003) was associated with successful NGS. Procore needle was associated with failed NGS (OR 0.356, 95% CI 0.156w0.811; pZ0.014). Multivariate logistic regression analysis revealed that the gauge of needle was independent factor of successful NGS (OR 2.553, 95% CI 1.193w5.460; pZ0.015) and Procore needle reduced the success rate of NGS (OR 0.365, 95% CI 0.145w0.923; pZ0.033). Conclusion: The gauge of needle is an independent factor associated with successful NGS. Also, selection of needle type affects success rate of NGS. This finding suggests that the quantity of cells obtained from FNA specimens is important for successful NGS. Keyword: Next-generation sequencing, Endoscopic ultrasound-fine needle aspiration, Pancreatic cancer.
Factor of Successful NGS
String Sign CEA Cytology Rheology
Sensitivity
Specificity
PPV
NPP
Accuracy
P value
0.67 0.67 0.50 1.00
0.67 0.89 1.00 0.89
0.57 0.80 1.00 0.86
0.75 0.80 0.75 0.83
0.67 0.80 0.80 0.93
0.32 0.09 0.05 0.001
Needle gauge 25gauge 19gauge or 22gauge Needle type Conventional FNA Procore needle
Odds ratio
95% confidential interval
P-value
Reference 2.553
1.193w5.460
0.016
Reference 0.365
0.145w0.923
0.033
Multivariate logistic regression
Flow chart
Su1280 Alterations in EUS morphology, cyst fluid cytology, DNA quality and DNA quantity in patients treated with EUS-guided ethanol lavage with paclitaxel injection Kook Hyun Kim*1,2, Kathleen A. McGreevy1, Kristin La Fortune1, Harvey Cramer1, John M. DeWitt1 1 Indiana university, Indianapolis, IN; 2Yeungnam university, Daegu, Korea (the Republic of) Background: Evaluation of efficacy for endoscopic ultrasound (EUS)-guided pancreatic cyst ablation (PCA) is mainly based on decrease in size of the treated cyst. The aim of this study is to evaluate changes of cyst fluid cytology, DNA quality, DNA quantity and sonographic morphology following EUS-PCA with ethanol lavage and paclitaxel injection. Methods: In a retrospective single center study, subjects with
AB334 GASTROINTESTINAL ENDOSCOPY Volume 83, No. 5S : 2016
www.giejournal.org
pancreatic cysts is exceedingly important yet remains difficult. The relative viscosity of pancreatic cyst fluid has proved useful for distinguishing mucinous from non-mucinous cysts and is an indirect measure of cyst-fluid mucin, glycoproteins, and DNA content. In previous studies, the viscosity was assessed by the string sign (elongation viscosity), a surrogate marker of cyst fluid viscosity. However, the method lacks a theoretical framework for predicting the viscoelastic nature of the cyst fluid, a rheological property of complex physiologic fluids having two components, viscosity and elasticity. The aim of the present study was to assess the utility of the rheological properties of pancreatic cyst fluid, as compared with cytology and biochemical analysis in differentiating pancreatic cyst types. Patients and Methods: EUS-FNA was performed on consecutive subjects with pancreatic cysts. In addition to routine cyst fluid analyses (biochemical, cytological and viscosity assessment using the string sign), a rheological behavior curve of the cyst fluid was generated (steady-state shear viscosity at different shear rates measured by a rotational rheometer (Discovery DHR-2, TA Instruments, USA)). Pancreatic cysts were classified as mucinous (mucinous cystadenoma, mucinous adenocarcinoma, IPMN) or non-mucinous (serous cystadenoma, pseudocyst) based on surgical and/or clinical outcome (presentation, follow-up, imaging and fluid analyses). Results: Fifteen subjects were evaluated (9 females, median age 53y). Mean cyst size was 33.1 mm (range 20-60 mm). Six mucinous and 9 nonmucinous cysts were diagnosed. Four mucinous cysts were diagnosed surgically. Median follow-up was 18 months. Three different patterns of rheological behavior curves were identified (figure): Pattern I (constant viscosity <1.6 units), Pattern II (rapidly decreasing viscosity to <1.6 units) and Pattern III (slowly decreasing viscosity >1.6 units). Patterns I, II were correlated with non-mucinous cysts and Pattern III was correlated with mucinous cysts. This curve pattern analysis was evaluated as a new diagnostic test. The concordance between the string sign and the measured relative viscosity was low (pZ0.62, chi-square). The sensitivity, specificity, positive and negative predictive values and accuracy of string sign, elevated CEA (>192 ng/ml), cytology and rheology (curve patterns I, II-nonmucinous, curve pattern III-mucinous) are presented in the table. Limitation: Small cohort, surgical diagnosis was not available in all cases. Conclusion: The accuracy of the string sign, CEA and cytology for mucinous cysts are comparably inadequate. Cyst fluid rheological analysis appears to accurately differentiate pancreatic cyst types. Results:
Su1279 Factors of Endoscopic Ultrasound Fine-Needle Aspiration for Successful Next-Generation Sequencing in Pancreatic Cancer Dong Hyo Noh, Joo Kyung Park, Kyu Taek Lee, Kwang Hyuck Lee*, Jong Kyun Lee, Keum Bit Hwang, Ju-Il Yang, Jun Young Kim Medicine, Samsung Medical Center, Seoul, Korea (the Republic of) Background: Recent developments in the understanding of the genetics of pancreatic ductal adenocarcinoma (PDAC) have led to the potential for a personalized approach in PDAC patients. Several studies have described the feasibility of generating genetic profiles of pancreatic tumor with next generation sequencing (NGS) of endoscopic ultrasound-fine needle aspiration (EUS-FNA) samples. Aim: The aim of this study was to find the best way of EUS-FNA for successful NGS in PDAC. Methods: We tried to perform NGS with tissues from one-hundred ninety eight patients who was histologically proven PDAC by EUS-FNA at Samsung Medical Center between November 2011 and February 2015. Clinical targeted seqeuncing of 83 genes was perfomed with DNA from EUS-FNA samples by Illumina HiSeq 2500 sequencing platform. Retrospectively, the medical records of these patients were reviewed including tumor factors (size, location, T stage), EUS-FNA factors (gauge of needle, needle type, number of needle passes) and histologic factors (cellularity, blood contamination). Results: NGS could be done successfully in one-hundred nine (55.1%). In univariate analysis, larger gauge of needle (Odds ratio (OR) 2.621, 95% Confidential interval (CI) 1.234w5.191; pZ0.007) and location of tumor in body/tail (OR 2.336, 95% CI 1.296w4.210; pZ0.003) was associated with successful NGS. Procore needle was associated with failed NGS (OR 0.356, 95% CI 0.156w0.811; pZ0.014). Multivariate logistic regression analysis revealed that the gauge of needle was independent factor of successful NGS (OR 2.553, 95% CI 1.193w5.460; pZ0.015) and Procore needle reduced the success rate of NGS (OR 0.365, 95% CI 0.145w0.923; pZ0.033). Conclusion: The gauge of needle is an independent factor associated with successful NGS. Also, selection of needle type affects success rate of NGS. This finding suggests that the quantity of cells obtained from FNA specimens is important for successful NGS. Keyword: Next-generation sequencing, Endoscopic ultrasound-fine needle aspiration, Pancreatic cancer.
Factor of Successful NGS
String Sign CEA Cytology Rheology
Sensitivity
Specificity
PPV
NPP
Accuracy
P value
0.67 0.67 0.50 1.00
0.67 0.89 1.00 0.89
0.57 0.80 1.00 0.86
0.75 0.80 0.75 0.83
0.67 0.80 0.80 0.93
0.32 0.09 0.05 0.001
Needle gauge 25gauge 19gauge or 22gauge Needle type Conventional FNA Procore needle
Odds ratio
95% confidential interval
P-value
Reference 2.553
1.193w5.460
0.016
Reference 0.365
0.145w0.923
0.033
Multivariate logistic regression
Flow chart
Su1280 Alterations in EUS morphology, cyst fluid cytology, DNA quality and DNA quantity in patients treated with EUS-guided ethanol lavage with paclitaxel injection Kook Hyun Kim*1,2, Kathleen A. McGreevy1, Kristin La Fortune1, Harvey Cramer1, John M. DeWitt1 1 Indiana university, Indianapolis, IN; 2Yeungnam university, Daegu, Korea (the Republic of) Background: Evaluation of efficacy for endoscopic ultrasound (EUS)-guided pancreatic cyst ablation (PCA) is mainly based on decrease in size of the treated cyst. The aim of this study is to evaluate changes of cyst fluid cytology, DNA quality, DNA quantity and sonographic morphology following EUS-PCA with ethanol lavage and paclitaxel injection. Methods: In a retrospective single center study, subjects with
AB334 GASTROINTESTINAL ENDOSCOPY Volume 83, No. 5S : 2016
www.giejournal.org