From the Journals

Branch duct intraductal papillar mucinous neoplasms confer increased malignancy risk


Key clinical point: Branch duct intraductal papillary mucinous neoplasms conferred a markedly increased risk of malignancy even when they lacked worrisome features at baseline.

Major finding: At 5 years, the standardized incidence ratio for malignancy was 18.8 compared with the general population.

Data source: A retrospective study of 577 patients with suspected branch duct intraductal papillary mucinous neoplasms.

Disclosures: The investigators did not disclose external funding sources. They reported having no relevant conflicts of interest.

The appropriate surveillance strategy for branch duct IPMNs is a point of debate, and numerous guidelines have offered recommendations for managing these potentially malignant neoplasms. Among the contested topics is the appropriateness of ceasing imaging surveillance of lesions that are stable over years. In 2015, an American Gastroenterological Association guideline made a conditional recommendation for cessation of imaging surveillance of pancreatic cysts that have remained stable after 5 years, noting that only very low-quality evidence was available. Given the paucity of data on this topic, this recommendation has been debated.

Dr. Pergolini and colleagues shed new light on this question with this retrospective review. Their study demonstrates that a dramatically increased risk of developing pancreatic malignancy persists even when a branch duct IPMN demonstrates no worrisome features or growth after 5 years of imaging surveillance. In fact, in their cohort, the risk of malignancy not only persisted among patients with branch duct IPMNs compared to population-based controls, but in fact, the risk was even greater after 5 years of follow-up. The risk persisted even after 10 years of follow-up. This study lends credibility to the opinion that branch duct type IPMNs should undergo ongoing surveillance even after 5 years of stability on imaging. Furthermore, it invites further study on smaller (less than 1.5 cm) branch duct IPMNs that remain stable over 5 years, as they appear to be very low risk and may represent a category of IPMNs that do not require indefinite surveillance.

Anthony Gamboa, MD, is assistant professor of medicine, program director of advanced endoscopy fellowship, division of gastroenterology, hepatology and nutrition, Vanderbilt University, Nashville, Tenn. He has no conflicts of interest.



Patients with branch duct intraductal papillary mucinous neoplasms were about 19 times more likely to develop malignancies over 5 years compared with the general population, although they lacked worrisome features of malignancy at baseline.

Few studies have explored branch duct intraductal papillary mucinous neoplasms (BD-IPMNs), which the researchers defined as unilocular or multilocular pancreatic cysts with a nondilated main pancreatic duct (smaller than 5 mm). To begin filling this gap, they retrospectively studied 577 patients with suspected or presumed BD-IPMNs followed at Massachusetts General Hospital. Patients underwent cross-sectional imaging 3 months or more after initial diagnosis at least once thereafter. Standardized incidence ratios were calculated based on population-level data for the United States from the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) program.

Patients tended to be in their mid-60s at diagnosis (range, 21-90 years) and 59% were female, said the researchers. Median follow-up time was 82 months and ranged between 6 and 329 months, but 63% of patients were followed for at least 5 years (median, 107 months), and about one in five were followed for more than a decade. Fully 83% of patients were asymptomatic at initial diagnosis, of which 10% subsequently became symptomatic. Most patients underwent diagnostic CT, but nearly half underwent MRI/MRCP and about a third underwent endoscopic ultrasound. At diagnosis, median cyst size was 14 mm (range, 2-54 mm) and 9% of patients had cysts measuring at least 3 cm. By the end of follow-up, 55% had larger cysts than at baseline, and cysts grew by a median of 0.9 mm per year.

At diagnosis, only 1% of patients had high-risk stigmata while 12% had worrisome features such as acute pancreatitis, cysts measuring at least 3 cm, thickened or enhancing cyst walls, nonenhancing mural nodules, main pancreatic duct size of 5-9 mm, an abrupt change in caliber of the main pancreatic duct, and lymphadenopathy. During follow-up, another 13% of patients developed new worrisome features and 9% developed high-risk stigmata, while 2% experienced regression of a nodule. In all, 36% of patients had cysts with either worrisome features, high-risk stigmata, or both at some point during the study.

Among 363 patients followed for at least 5 years, 20 (5.5%) were diagnosed with high-risk dysplasia or invasive neoplasms and 4.4% developed invasive cancer, for a standardized incidence ratio of 18.8 (95% confidence interval, 9.7-32.8; P less than .001). Among 108 patients who had cysts measuring 1.5 cm or less, only one individual developed a distinct ductal adenocarcinoma during 5 or more years of follow-up. But of 255 patients with larger cysts, the 5-year rate of malignancy was 7.5% (P = .01).

“The absence of worrisome features or high-risk stigmata at a 5-year time point does not exclude the development of pancreatic malignancy, and the risk in these patients is 18.8 times higher than that of the general population,” the researchers concluded. “Because of this, we strongly support continued surveillance after 5 years from the initial diagnosis.” Cysts that remain 1.5 cm or smaller for at least 5 years are probably low risk, they said. “This is an important issue for further investigation, since it may help reduce costs related to surveillance and improve patients’ quality of life.”

The investigators did not disclose external funding sources. They reported having no conflicts of interest.