In 2014, the Papanicolaou Society of Cytopathology (PSC) published its guidelines for reporting pancreatobiliary cytopathology1 codified in an atlas in 2015.2 As with terminology reporting systems for other organs, this system included 6 diagnostic categories with definitions for each. The definitions for the “atypical” category, “suspicious,” and “positive/malignant” were like the definitions in other systems. Unique to the PSC system is that it defined adequacy for pancreatobiliary specimens, especially pancreatic cysts, and it created categories of “neoplasm: benign” and “neoplasm: other.” Serous cystadenomas (SCA) were classified as “neoplasm: benign and neoplastic mucinous cysts,” which include intraductal papillary neoplasms and mucinous cystic neoplasms as neoplasm: other. A controversial aspect of the PSC system was the inclusion of “pancreatic neuroendocrine neoplasm” (PanNET) and “solid pseudopapillary neoplasm” (SPN) in the neoplasm: other category. This was done to allow for more flexible patient management because PanNET may occasionally be managed with observation.

The proposed World Health Organization (WHO) International System to update the PSC system follows the WHO Classification of Tumours of the Pancreas published in 2019.3 SCA are now classified as “benign/negative for malignancy,” and PanNET and SPN are in the “malignant” category. Two new categories have been created: “pancreatobiliary neoplasm-low-risk/grade” (PaN-Low) and “pancreatobiliary neoplasm-high-risk/grade” (PaN-High). These categories stratify intraductal and cystic mucinous neoplasms by cytological grade using previously published criteria.4

In this months’ issue, Hoda et al5 report on the risk of malignancy (ROM) for each diagnostic category using the WHO International System for Pancreatobiliary Cytology in a previously studied cohort of patients classified prospectively using the PSC system. The study includes 334 endoscopic ultrasound-guided fine-needle aspiration samples derived from 322 patients and retrospectively classifies the samples into the diagnostic categories of the proposed WHO International System. The absolute ROM is established by correlation with histology or clinical follow-up. Notably, high-grade dysplasia in an intraductal or intracystic process is defined as malignant.

The restructuring of the neoplasm category led to improved prediction of ROM. Under the PSC system, the neoplasm: benign category had a 0% ROM and the neoplasm: other category had a 30% ROM. Using the WHO International System, 4 cases previously classified as neoplasm: benign were reclassified as benign/negative for malignancy. Sixty-six cases classified as neoplasm: other were reclassified as PaN-Low (42), PaN-High (5), and malignant (19). The PaN-Low category has a 5% ROM and the PaN-High category has a 60% ROM. The ROM in the other categories remained the same. Two cases in the PaN-Low category, confirmed as mucinous cysts, proved to be malignant. The cytology samples lacked representative cells to assess grade of dysplasia. Two cases in the PaN-High category showed intermediate grade of dysplasia in the resection specimen, a known pitfall in the classification of dysplasia in cytology samples.4

This study demonstrates improved prediction of ROM in the neoplasm category using the proposed WHO International System for reporting pancreatobiliary cytology. It follows the WHO Classification of Pancreatic Tumours, leading to clarity and uniformity in reporting pancreatic neoplasms. This is the first study assessing the ROM using the proposed WHO International System. Limitations are that this study is retrospective and was conducted on samples collected during the 1-year period of 2016, before molecular analysis was routinely added to pancreatic cyst fluid analysis. It establishes a model for future studies of ROM using the proposed WHO International System, which will incorporate the results of pancreatic cyst fluid analysis with the added results of molecular analysis.

Funding Support

No specific funding was disclosed.

Conflict of Interest Disclosures

The authors made no disclosures.