Search

Background: Endoscopic ultrasound with fine-needle aspiration (EUS-FNA) is recommended for diagnosis of pancreatic cystic lesions (PCLs). Its role in surveillance is unclear. Our goal was to determine if a second EUS-FNA changes diagnosis or management of PCLs., Methods: A retrospective analysis of an EUS database, searching for EUS-FNAs in PCLs from 2007 to 2017 was performed. Demographics, cyst characteristics, and FNA results were compared in patients under surveillance, performing a single or two consecutive EUS-FNAs., Results: Of 203 PCLs referred for EUS-FNA, surveillance was decided in 128 (63%). Data of 105 (82%) patients with a single EUS-FNA were compared with 23 (18%) with two EUS-FNAs during surveillance. Patients were younger in this latter group (P = .055), whereas CEA levels were marginally higher (P = .078) and a mass/nodule were more frequent (P = .006). The mean time between EUS-FNAs was 38 months (4.7-118.8) for 18 patients maintaining surveillance vs 18 months (2.9-56.9) in the four referred for surgery (P = NS) after two EUS-FNAs (two NETs, one IPMN-HGD, and one MCN-LG). A high correlation in CEA level between consecutive EUS-FNAs (r 2 = 0.945, P < .01) was present, with a change of category observed (cut-off level = 192 ng/mL) in two patients only. Of four patients with a second EUS-FNA with conclusive cytology, two had NETs confirmed on resection., Conclusions: Repeating EUS-FNA in surveillance of PCLs with clinical suspicion of malignancy increased neoplasm diagnoses, changing decision toward surgery in almost 20% of patients while excluding IPMNs with mucin nodules from unnecessary resections. A second EUS-FNA for cytology appears justified in some PCLs, particularly for diagnosing NETs., (© 2020 Wiley Periodicals LLC.)

Approximately half of all pancreatic cysts are neoplastic, mainly comprising intraductal papillary mucinous neoplasms (IPMN), which can progress to invasive carcinoma. Current Fukuoka guidelines have limited sensitivity and specificity in predicting progression of asymptomatic pancreatic cysts. We present first results of the prospective ZYSTEUS biomarker study investigating (i) whether detection of driver mutations in IPMN by liquid biopsy is technically feasible, (ii) which compartment of IPMN is most suitable for analysis, and (iii) implications for clinical diagnostics. Twenty-two patients with clinical inclusion criteria were enrolled in ZYSTEUS. Fifteen cases underwent endoscopic ultrasound (EUS)-guided fine-needle aspiration and cytological diagnostics. Cellular and liquid fraction of the cysts of each case were separated and subjected to deep targeted next generation sequencing (NGS). Clinical parameters, imaging findings (EUS and MRI), and follow-up data were collected continuously. All IPMN cases (n = 12) showed at least one mutation in either KRAS (n = 11) or GNAS (n = 4). Three cases showed both KRAS and GNAS mutations. Six cases harbored multiple KRAS/GNAS mutations. In the three cases with pseudocysts, no KRAS or GNAS mutations were detected. DNA yields were higher and showed higher mutation diversity in the cellular fraction. In conclusion, mutation detection in pancreatic cyst fluid is technically feasible with more robust results in the cellular than in the liquid fraction. Current results suggest that, together with imaging, targeted sequencing supports discrimination of IPMN from pseudocysts. The prospective design of ZYSTEUS will provide insight into diagnostic value of NGS in preoperative risk stratification. Our data provide evidence for an oligoclonal nature of IPMN., (© 2018 Wiley Periodicals, Inc.)

Introduction: The aim of this study was to determine if neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) were predictive of malignancy in mucin-producing pancreatic cystic neoplasms (MpPCN)., Methods: One hundred and twenty patients with MpPCN were retrospectively reviewed. Malignant neoplasms were defined as neoplasms harbouring invasive carcinoma or high grade dysplasia. A high NLR and PLR were defined as ≥2.551 and ≥208.1, respectively., Results: High NLR was significantly associated with symptomatic tumors, larger tumors, solid component, main-duct IPMN, and Sendai high risk category. High PLR was significantly associated with jaundice and Sendai high risk category. On univariate analyses, symptomatic tumors, jaundice, solid component, dilated pancreatic duct, and both a high NLR and PLR were significant predictors of malignant and invasive MpPCN. On multivariate analyses, solid component and dilated pancreatic duct were independent predictors of malignant and invasive MpPCN. PLR was an independent predictor for invasive MpPCN. When MpPCN were stratified by the Fukuoka and Sendai Guidelines, both a high NLR and PLR were significantly associated with malignant neoplasms within the high risk categories., Conclusions: PLR is an independent predictor of invasive carcinoma. The addition of PLR as a criterion to the FCG and SCG significantly improved the predictive value of these guidelines in detecting invasive neoplasms., (© 2015 Wiley Periodicals, Inc.)

Distinguishing mucinous from nonmucinous cystic lesions of the pancreas often constitutes a diagnostic dilemma. The clinical management differs between such lesions; therefore it is important to make an accurate preoperative diagnosis. Various centers have reported conflicting results regarding their ability to detect mucin-producing neoplastic cells and appropriately reach a diagnosis based on endoscopic ultrasound (EUS) guided FNA. The aim of this study is to assess the ability of EUS-FNA cytology to diagnose and differentiate mucinous from nonmucinous pancreatic cystic lesions. We reviewed records of patients who underwent EUS of pancreatic cystic lesions. If FNA was performed and mucinous neoplasm was suspected, aspirate was evaluated for cytomorphology and presence of mucin. FNA results were compared to final histologic diagnosis if surgery was performed. Cytologic diagnosis was provided for 28/30 (93%). By comparing EUS-FNA diagnoses with final surgical pathology, FNA accurately diagnosed in 10/11 cases with sensitivity and specificity for detection of malignancy of 100 and 89, respectively, while the accuracy for identification of mucinous cystic neoplasms was 100%. Our results indicate that in the appropriate clinical and imaging setting, EUS-FNA cytology with analysis for mucin production by tumor cells is an important test in distinguishing pancreatic cystic lesions and guiding further management.

Most pancreatic cysts (90%) are inflammatory pseudocysts. Approximately 10% of pancreatic cysts are neoplasms, including serous cystadenomas, and mucinous tumors, some of which are malignant. Analysis of pancreatic cyst fluid obtained by percutaneous or endoscopic fine-needle aspiration is increasingly being used for the preoperative diagnosis of pancreatic or peripancreatic cysts. However, cyst fluid chemical and cytologic features in less common types of pancreatic cysts have not been reported. Lymphoepithelial cyst of the pancreas is exceedingly rare, and only occasional individual reports have described cyst fluid findings. We report on a case of lymphoepithelial cyst of the pancreas developing in a middle-aged man. Cyst fluid aspirated under radiological guidance showed elevated levels of carcinoembryonic antigen (CEA), CA19-9, CA 125, and amylase, and a viscosity greater than that of serum. A cell block preparation of a fine-needle aspiration showed tissue fragments with a keratinized squamous lining and a lymphocytic infiltrate in the wall, and abundant background keratinous debris. The cytologic and biochemical findings in this case exhibit similarities to the findings reported in other reports, and may represent a recognizable pattern on cyst fluid analysis., (Copyright 1999 Wiley-Liss, Inc.)