TY - JOUR
T1 - American Gastroenterological Association guidelines are inaccurate in detecting pancreatic cysts with advanced neoplasia
T2 - A clinicopathologic study of 225 patients with supporting molecular data
AU - Singhi, Aatur D.
AU - Zeh, Herbert J.
AU - Brand, Randall E.
AU - Nikiforova, Marina N.
AU - Chennat, Jennifer S.
AU - Fasanella, Kenneth E.
AU - Khalid, Asif
AU - Papachristou, Georgios I.
AU - Slivka, Adam
AU - Hogg, Melissa
AU - Lee, Kenneth K.
AU - Tsung, Allan
AU - Zureikat, Amer H.
AU - McGrath, Kevin
N1 - Publisher Copyright:
© 2016 American Society for Gastrointestinal Endoscopy.
PY - 2016/6/1
Y1 - 2016/6/1
N2 - Background and Aims The American Gastroenterological Association (AGA) recently reported evidence-based guidelines for the management of asymptomatic neoplastic pancreatic cysts. These guidelines advocate a higher threshold for surgical resection than prior guidelines and imaging surveillance for a considerable number of patients with pancreatic cysts. The aims of this study were to assess the accuracy of the AGA guidelines in detecting advanced neoplasia and present an alternative approach to pancreatic cysts. Methods The study population consisted of 225 patients who underwent EUS-guided FNA for pancreatic cysts between January 2014 and May 2015. For each patient, clinical findings, EUS features, cytopathology results, carcinoembryonic antigen analysis, and molecular testing of pancreatic cyst fluid were reviewed. Molecular testing included the assessment of hotspot mutations and deletions for KRAS, GNAS, VHL, TP53, PIK3CA, and PTEN. Results Diagnostic pathology results were available for 41 patients (18%), with 13 (6%) harboring advanced neoplasia. Among these cases, the AGA guidelines identified advanced neoplasia with 62% sensitivity, 79% specificity, 57% positive predictive value, and 82% negative predictive value. Moreover, the AGA guidelines missed 45% of intraductal papillary mucinous neoplasms with adenocarcinoma or high-grade dysplasia. For cases without confirmatory pathology, 27 of 184 patients (15%) with serous cystadenomas (SCAs) based on EUS findings and/or VHL alterations would continue magnetic resonance imaging (MRI) surveillance. In comparison, a novel algorithmic pathway using molecular testing of pancreatic cyst fluid detected advanced neoplasias with 100% sensitivity, 90% specificity, 79% positive predictive value, and 100% negative predictive value. Conclusions The AGA guidelines were inaccurate in detecting pancreatic cysts with advanced neoplasia. Furthermore, because the AGA guidelines manage all neoplastic cysts similarly, patients with SCAs will continue to undergo unnecessary MRI surveillance. The results of an alternative approach with integrative molecular testing are encouraging but require further validation.
AB - Background and Aims The American Gastroenterological Association (AGA) recently reported evidence-based guidelines for the management of asymptomatic neoplastic pancreatic cysts. These guidelines advocate a higher threshold for surgical resection than prior guidelines and imaging surveillance for a considerable number of patients with pancreatic cysts. The aims of this study were to assess the accuracy of the AGA guidelines in detecting advanced neoplasia and present an alternative approach to pancreatic cysts. Methods The study population consisted of 225 patients who underwent EUS-guided FNA for pancreatic cysts between January 2014 and May 2015. For each patient, clinical findings, EUS features, cytopathology results, carcinoembryonic antigen analysis, and molecular testing of pancreatic cyst fluid were reviewed. Molecular testing included the assessment of hotspot mutations and deletions for KRAS, GNAS, VHL, TP53, PIK3CA, and PTEN. Results Diagnostic pathology results were available for 41 patients (18%), with 13 (6%) harboring advanced neoplasia. Among these cases, the AGA guidelines identified advanced neoplasia with 62% sensitivity, 79% specificity, 57% positive predictive value, and 82% negative predictive value. Moreover, the AGA guidelines missed 45% of intraductal papillary mucinous neoplasms with adenocarcinoma or high-grade dysplasia. For cases without confirmatory pathology, 27 of 184 patients (15%) with serous cystadenomas (SCAs) based on EUS findings and/or VHL alterations would continue magnetic resonance imaging (MRI) surveillance. In comparison, a novel algorithmic pathway using molecular testing of pancreatic cyst fluid detected advanced neoplasias with 100% sensitivity, 90% specificity, 79% positive predictive value, and 100% negative predictive value. Conclusions The AGA guidelines were inaccurate in detecting pancreatic cysts with advanced neoplasia. Furthermore, because the AGA guidelines manage all neoplastic cysts similarly, patients with SCAs will continue to undergo unnecessary MRI surveillance. The results of an alternative approach with integrative molecular testing are encouraging but require further validation.
UR - http://www.scopus.com/inward/record.url?scp=84956938576&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84956938576&partnerID=8YFLogxK
U2 - 10.1016/j.gie.2015.12.009
DO - 10.1016/j.gie.2015.12.009
M3 - Article
C2 - 26709110
AN - SCOPUS:84956938576
SN - 0016-5107
VL - 83
SP - 1107-1117.e2
JO - Gastrointestinal endoscopy
JF - Gastrointestinal endoscopy
IS - 6
ER -