The impact of failure to achieve symptom control after resection of functional neuroendocrine tumors: An 8-institution study from the US Neuroendocrine Tumor Study Group

Abstract

Background: The goals of resection of functional neuroendocrine tumors (NETs) are two-fold: Oncological benefit and symptom control. The interaction between the two is not well understood.

Methods: All patients with functional NETs of the pancreas, duodenum, and ampulla who underwent curative-intent resection between 2000 and 2016 were identified. Using Cox regression analysis, factors associated with reduced recurrence-free survival (RFS) were identified.

Results: Two-hundred and thirty patients underwent curative-intent resection. Fifty-three percent were insulinomas, 35% gastrinomas, and 12% were other types. Twenty-one percent had a known genetic syndrome, 23% had lymph node (LN) positivity, 80% underwent an R0 resection, and 14% had no postoperative symptom improvement (SI). Factors associated with reduced RFS included noninsulinoma histology, the presence of a known genetic syndrome, LN positivity, R1 margin, and lack of SI. On multivariable analysis, only the failure to achieve SI following resection was associated with reduced RFS. Considering only those patients with an R0 resection, failure to achieve SI was associated with worse 3-year RFS compared with patients having SI (36% vs 80%; P = 0.006).

Conclusions: Failure to achieve symptomatic improvement after resection of functional NETs is associated with worse RFS. These patients may benefit from short-interval surveillance imaging postoperatively to assess for earlier radiographical disease recurrence.

Keywords: VIPoma; gastrinoma; glucagonoma; insulinoma; recurrence; symptomatic.

FIGURE 1

Kaplan-Meier plots assessing the association between symptom improvement and recurrence-free survival. For all patients within the study cohort (A), the failure to achieve postoperative symptom improvement was associated with a significantly reduced recurrence-free survival. This remained true when examining only patients with R0 resection status on final pathologic assessment (B)