Misdiagnosis Still Common in Pancreatic Cystic Neoplasms

CHICAGO – Preoperative clinical diagnoses of pancreatic cystic neoplasms (PCNs) are frequently found to be in error when patients go to surgery as recommended under international guidelines, data from a retrospective study show.

An analysis of all pancreatic resections performed for presumed PCN at the Verona Pancreas Institute, Italy, from 2011 through 2020 showed a high degree of discrepancy between the preoperative clinical diagnosis and the final postoperative pathology, with some lesions being misdiagnosed in nearly two-thirds of cases, reported Anna Burelli, MD, of the department of general and pancreatic surgery at the University of Verona.

“Diagnostic errors are still common for resected PCNs. Morphological and clinical information alone still poorly frame actual targets for surgery, and hopefully the development of new reliable biomarkers will represent the next evolution in pancreatic cystic neoplasm management,” she said in an oral abstract session at the annual Digestive Disease Week® (DDW).

Diagnostic errors are significant issues in care of patients with PCN, because clinicians must balance the need for prompt, definitive treatment when necessary with the need for avoiding the significant morbidity of pancreatic resection for patients with lesions that turn out to be nonmalignant.

The investigators define “misdiagnosis” as a discrepancy between the preoperative clinical diagnosis and the postoperative pathology, and “mismatch” as a discrepancy between the preoperative suspicion of malignant or benign disease and the final pathology.

Checkered history

In previous cases series from Massachusetts General Hospital in Boston (2010) and the Verona Pancreas Institute (2012) – both experienced, high-volume centers – PCN misdiagnosis rates were 30% and 21%, respectively, and results from the current study show that things haven’t changed much since then, Dr. Burelli said.

PCNs are divided into neoplastic and nonneoplastic categories, with mucin-producing subtypes considered to be precancerous lesions that require accurate diagnosis and close monitoring.

Examples of neoplastic PCNs are intraductal papillary mucinous neoplasms (IPMNs) of the main pancreatic duct or side branch and mucinous cystadenomas. In contrast, serous cystadenomas, considered nonneoplastic, are mostly benign lesions discovered incidentally during abdominal imaging for another indication. It is very difficult, however, to distinguish between the two PCN subtypes clinically.

For example, Dr. Burelli showed images from a patient who received a preoperative diagnosis of mixed IPMN that was in fact found to be chronic pancreatitis on postoperative pathology.

Dr. Burelli noted that AGA and joint European guidelines for management of PCNs have been updated over the past decade, with the latest AGA iteration in 2015.

A 2017 study evaluating the 2015 AGA guidelines for management of asymptomatic PCNs found that following the guidelines in a large multicenter cohort “would have resulted in 60 % fewer patients being referred for surgical resection, and accurately recommended surveillance in 95% of patients with asymptomatic PCNs.”

Misdiagnosis and mismatch common

In the current study, Dr. Burelli and colleagues reviewed all pancreatic resections performed for PCNs at their center from 2011 through 2020.

Of 601 patients included in the retrospective study, 301 underwent endoscopic ultrasound (EUS).

The investigators identified misdiagnosis in 19% of cases and mismatch in 34%, and there was no significant improvement in diagnostic accuracy among the 50% of patients who underwent EUS.

The most frequently misdiagnosed lesions were cystic neuroendocrine tumors, in 61% of cases. The least misdiagnosed lesions were pseudopapillary tumors, in 6% of cases.

Many of the diagnostic errors were clinically important. For example, seven cases presumed to be serous cystic neoplasms (an almost always benign lesion) were found on final pathology to have a different, malignant histology.

Mismatch examples included 50 IPMNs with high-risk stigmata that were presumed to be malignant before surgery but were nonmalignant on final pathology, and 38 IPMNs without high-risk stigmata which were thought on clinical examination to be benign but turned out to be malignant on final pathology.

“Our results are in line with the current literature,” Dr. Burelli said, citing a recent meta-analysis showing that among 3,292 patients who underwent resection for mucinous cystic neoplasms (MCNs), the pooled rate of malignancy was 16.1%, yet the 2012 International Association of Pancreatology guidelines recommend surgery for all fit patients with MCNs, and joint European evidence-based guidelines from 2018 recommend surgery for MCNs 40 mm or larger, those with mural nodules, and for patients who are symptomatic.

The 16.1% pooled malignancy rate suggests “that there is space for surveillance in most cases of MCNs,” she said.

In addition, morphologic and clinical evaluation for IPMN with high-risk stigmata have been shown to have low specificity and low sensitivity, “so should guideline recommendations be revised?” Dr. Burelli said.

She pointed to a recent multi-institutional study in Gastroenterology showing that real-time next-generation sequencing of pancreatic cyst fluid “is sensitive and specific for various pancreatic cyst types and advanced neoplasia arising from mucinous cysts, but also reveals the diversity of genomic alterations seen in pancreatic cysts and their clinical significance.”

“This is not the future; this is the present,” she concluded.

Invited discussant R. Matthew Walsh, MD, a surgeon specializing in pancreatic and cancer surgery at the Cleveland Clinic, complimented the contributions of her group.

“The patient that you showed with chronic pancreatitis could have very well benefited from the operation regardless of the diagnosis if they were symptomatic,” he said, addressing Dr. Burelli. “So what is the group that is the regrettable surgical patients, and where are you aiming your studies? Is it really the 24% with high-risk features in IPMN that have low-grade dysplasia, or is it the 58% who we’re not sure why they were operated on because they didn’t have high-grade features who had low-grade dysplasia?”

She replied that “the goal here is to avoid surgery for benign entities, and we know that the only true benign entities are serous cystic neoplasms, and all the others have a malignant potential, but we think at Verona Pancreas Institute there is no reason to operate on low-grade dysplasia free patients. This is what we really would like to avoid.”

Dr. Walsh also asked, given their finding that EUS did not appear to offer a benefit to patients or change decision making, which patients should still get EUS.

“I think that only patients in which the diagnosis is uncertain or in which there are some worrisome features or high-risk stigmata should undergo EUS before surgery, and also to continue follow-up,” Dr. Burelli said. “I don’t think that the conclusion is that EUS is not useful, but it’s not useful in all.”

For example, large, microcystic lesions can be readily identified radiographically, but other, more complex cases may still require EUS to help nail down or refine a diagnosis, she said.

The study was internally funded. Dr. Burelli and Dr. Walsh reported having no conflicts of interest.

DDW is sponsored by the American Association for the Study of Liver Diseases, the American Gastroenterological Association, the American Society for Gastrointestinal Endoscopy, and the Society for Surgery of the Alimentary Tract.

This article originally appeared on MDedge.com, part of the Medscape Professional Network.

Source: Read Full Article