Pancreatic Tumor Classification
The 2019 World Health Organization (WHO) classifies pancreatic tumors into epithelial tumors, mature teratoma, mesenchymal tumors, lymphomas, and secondary tumors that are defined as neoplasms that have spread to the pancreas from an extra-pancreatic primary. The WHO 2019 classification of pancreatic epithelial tumors is provided in Table 2a, and the WHO 2017 classification of pancreatic neuroendocrine neoplasms, incorporated into the WHO 2019 classification of digestive system tumors [1], is provided in Table 2b.
History and Current Concepts of Fine-Needle Aspiration Cytology of Pancreatic Tumors
Before the advent of endoscopic ultrasonography (EUS), percutaneous ultrasonography (US)-guided fine-needle aspiration (FNA) biopsies and computed tomography (CT)-guided FNA biopsies were utilized for obtaining tissue for cytological examination. Beginning in the 1970s, US was the first imaging modality used, with the use of CT reported a few years later [11, 12]. EUS was developed in the 1980s to overcome limitations of transabdominal US imaging of the pancreas caused by intervening gas, bone, and fat, and is currently the most common source of obtaining material from pancreatic lesions. EUS provides excellent visualization of the pancreatic head and uncinate process from the duodenum as well as the body and tail of the pancreas from the stomach. With the advent of curvilinear echo endoscopes, both transgastric and transduodenal EUS-guided FNA (EUS-FNA) biopsies of the pancreas have become a possibility. EUS is superior to percutaneous US and CT scan, especially when the pancreatic tumor is <2–3 cm [13, 14]. In hospitals where CT-guided biopsies and/or endoscopic retrograde cholangiopancreatography (ERCP) brushings are still the standard methods for obtaining tissue for diagnosis of pancreatic lesions, false negative rates of >30% have been reported [15]. Advantages of EUS-FNA over other modalities include the close proximity of the target lesion to the endoscope, sampling under direct ultrasound guidance, and avoidance of overlying bowel [16].
Accuracy of FNA in Pancreatic Tumors
The reported diagnostic yield of EUS-FNA for pancreatic tumorous lesions is good, with a diagnostic accuracy of 78–95%, sensitivity of 78–95%, and specificity of 75–100% [17–19]. EUS-FNA has a high specificity but a low sensitivity for diagnosing malignancy in pancreatic cystic tumors.
False positive and false negative rates of cytological diagnosis of pancreatic masses on EUS-FNA are low. The false positive rate is about 2% and results from technical difficulties and sampling or interpretation errors. False positive results due to interpretation errors can occur from contamination of the specimen by an intervening mucosal malignancy, or misinterpretation of reactive changes in chronic pancreatitis as adenocarcinoma [20]. Chronic pancreatitis is the most common benign pathology causing false positive interpretation of a pancreatic cancer.
Rapid On-Site Evaluation
Rapid on-site evaluation (ROSE) performed by the cytopathologist at the time of an endoscopic procedure is necessary to direct the endosonographer as to whether the aspirate obtained is sufficient for a definitive diagnosis. This not only means that there is enough diagnostic material in the smears, but also that there is sufficient material for an ancillary work-up. The direct smears obtained in the endoscopy suite are quickly processed and examined by a light microscope and immediate direct feedback is provided to the endosonographer. This information assists in guiding the endosonographer to the number of EUS-FNA passes required and whether the biopsy technique needs to be adjusted in non-diagnostic aspirates to yield material for a final diagnosis. ROSE is generally performed on solid pancreatic masses and not on cystic pancreatic lesions. Non-ROSE groups were shown to have more repeat EUS-FNA biopsies than ROSE groups and this difference was reported to be statistically significant in some studies [21]. However, many recent studies have reported adequacy rates of pancreatic FNA to be similar with and without ROSE, indicating that, perhaps, at high-volume centres and in expert hands, ROSE may not be indispensable to achieve excellent results [22–24]. Up to four passes are usually recommended to get diagnostic material for pancreatic FNAs as it has been shown that the chances of aspirating diagnostic material decrease after that [25, 26]. However, certain solid neoplasms of the pancreas are desmoplastic and hence may provide few cells that could be insufficient to provide a definitive diagnosis. ROSE is particularly helpful in these cases so that more aspirates can be obtained to increase the yield for a definitive diagnosis and to ensure that the patient does not have to undergo a repeat FNA at a later date.
Table 2. WHO classifications
Papanicolaou Society of Cytopathology Guidelines for Pancreatobiliary Cytology
Summary of the Guidelines
The Papanicolaou Society of Cytopathology (PSC) developed a set of guidelines for pancreatobiliary cytology similar to those developed for other organ systems, which assess indications and techniques for imaging and FNA, terminology and nomenclature of pancreatobiliary disease, and available ancillary testing and post-biopsy management. These were published and summarized in 2014 [27–31]. These guidelines were developed based on the authors’ expertise, review of the literature,