Does surgical resection improve survival of pancreatic neuroendocrine tumours (panNET) less than 2 cm? a population-based study of 4,114 patients comparing England and the USA

Mohamed Mortagy; El Asmar Marie Line; Ker Tan; White Benjamin E; John Ramage

doi:10.1530/endoabs.114.OC1

OC1

Prev Next

Section Contents Cite

Endocrine Abstracts (2025) 114 OC1 | DOI: 10.1530/endoabs.114.OC1

UKINETS2025 23rd National Conference of the UK and Ireland Neuroendocrine Tumour Society 2025 Oral Communications (4 abstracts)

Does surgical resection improve survival of pancreatic neuroendocrine tumours (panNET) less than 2 cm? a population-based study of 4,114 patients comparing England and the USA

Mohamed Mortagy ^1,2 , Marie Line El Asmar ¹ , Ker Tan ¹ , Benjamin E White ¹ & John Ramage ^1,3

Author affiliations

¹Hampshire Hospitals NHS Foundation Trust, Basingstoke, United Kingdom; ²St. George’s University School of Medicine, Grenada, Grenada; ³University of Winchester, Winchester, United Kingdom

Background: PanNET are rare pancreatic tumours. Many studies include cases from referral centres only. Resection is normally recommended for localised tumours >2 cm. Tumours ≤2 cm are often monitored for progression if asymptomatic (as in the ASPEN study) but there are no randomised studies. This retrospective population-based study aimed to investigate the effect of surgery on survival of PanNET tumors ≤2 cm.

Methods: A total of 685 and 3429 patients with PanNET ≤2 cm were extracted from NCRAS (2012-2021) and SEER (2011-2022) databases respectively. Kaplan-Meier (KM), univariable and Multivariable Cox Regression (MCR) models for Overall survival (OS), cause-specific-survival (CSS), and Fine-Gray model (FG) for cancer-specific-survival (competing-risk model) were generated for comparing patients who had resection with patients who did not have resection.

Results: In NCRAS, median age was 64-years. Most patients were males (53.3%), white (88.0%), living in urban areas (74.5%), had T1 stage (77.1%), N0 stage (77.7%), M0 stage (72%), stage-group 1 (64.1%), and had surgical resection (53%). The most common resection type was partial/Whipple resection (83.8%). In SEER, median age was 64-years. Most patients were males (51.0%), white (65.5%), living in metropolitan areas (91.5%), had N0 stage (91%), M0 stage (94.5%), stage-group 1 (81.7%), and had surgical resection (62.5%). The most common resection type was partial/Whipple resection (78.4%). In both NCRAS and SEER, unadjusted Kaplan-Meier plots showed that resection is associated with better overall-survival as per log-rank tests (P < 0.001). Moreover, Multivariable-Cox regression for overall survival in SEER showed that surgery is associated with better survival (HR=0.58, P < 0.001) unlike NCRAS that showed no statistical significance (HR=0.63, P = 0.1). However, after adjusting for other factors (such as age and stage), both Multivariable-Cox regression for cause-specific survival and Fine-Gray model for cancer -specific survival for competing-risks showed that resection is not statistically different from receiving resection in both cohorts.

Conclusion: This large multi-national retrospective population study shows that surgical resection of PanNET ≤2 cm is not associated with improved cause-specific and cancer-specific survival (although it is associated with better unadjusted overall survival). This agrees with interim results of the ASPEN study and should inform guidelines for surgical management of PanNET.