Pancreatic Cystic Endocrine Tumors: A Different Morphological Entity Associated with a Less Aggressive Behavior

Abstract

Background: Cystic pancreatic endocrine tumors (CPETs) are rare lesions and their biological features have been scarcely investigated. Aim: To compare clinical and pathological features of resected non-functioning sporadic CPETs (NF-CPETs) with solid pancreatic endocrine tumors (SPETs) in a single-institution experience. Methods: All patients with a pathologically confirmed diagnosis of sporadic non-functioning pancreatic endocrine tumors who underwent curative resection between 1990 and 2008 were included. A comparison of demographic, clinical and pathological characteristics between CPETs and SPETs was made. Univariate and multivariable analyses were performed to identify preoperative predictors of carcinoma (non-functioning pancreatic endocrine carcinoma). Results: Twenty-one (11.5%) patients with a histological diagnosis of NF-CPET were identified. The median age was 60 years (IQR 46.5–73.5 years) and a diagnosis of carcinoma (non-functioning pancreatic endocrine carcinoma) was made in 3 (14.3%) cases. In the comparison with NF-SPETs, no differences were found in terms of gender (p = 0.75), age (p = 0.81), presenting symptoms (p = 0.43), localization of the tumors (p = 0.46) and type of resection (p = 0.31). The incidence of non-functioning pancreatic endocrine carcinoma was significantly lower in the NF-CPET versus the NF-SPET group (14.3 vs. 40.4%, p = 0.04). By univariate analysis, preoperative predictors of non-functioning pancreatic endocrine carcinoma included the presence of symptoms (OR 3.96, 95% CI 2.06–7.63) and an increase in the absolute value of radiological diameter (OR 1.05, 95% CI 1.03–1.07). A cystic morphology of the lesion turned out to be a negative predictor of carcinoma (OR 0.25, 95% CI 0.70–0.87). These results were confirmed by multivariable analysis. Conclusions: NF-CPETs have a measurable propensity to be benign. In those patients affected by small and asymptomatic NF-CPET a more conservative surgical approach or a follow-up policy could be considered.

Are Cystic Pancreatic Neuroendocrine Tumors an Indolent Entity Results from a Single-Center Surgical Series

“…Hence, as even the negativity at Ga 68 DOTA-PET cannot exclude the presence of a CPanNET [6,18], it cannot be considered as a routine imaging tool for preoperative diagnosis. If we look at the literature, our preoperative diagnostic accuracy is lower than previously reported (50 vs. 80%) [4,5,19]. The most common misdiagnosis in our series was mucinous cystadenoma (15% of cases).…”

“…Our series reinforces this concept, as in the setting of a high-volume center a correct preoperative diagnosis for pancreatic neoplasms was reached only in half of the cases. In retrospective, the identification of a hypervascular peripheral rim at either contrast-enhanced CT scan or MRI might be of help in suspecting a CPanNET, as already reported in other series [5,12,17]. Noteworthy, in 4 out of 27 cases in which a hypervascular peripheral rim was present at cross-sectional imaging, the pancreatic surgeon did not hypothesize a CPanNET, demonstrating that the rarity of this diagnosis in clinical practice can lower the awareness of CPanNETs, eventually leading to misdiagnosis.…”

“…While in most of the cases PanNETs present with a solid morphology, cystic variants of PanNETs might occur. These cystic PanNETs (CPanNETs) might encompass a variable spectrum of morphologies [3][4][5], and they represent only 10-17% of all resected PanNETs [6]. CPanNETs are usually nonfunctioning and sporadic, as few cases of functioning PanNETs have been reported within the clinical context of multiple endocrine neoplasia type 1 (MEN-1) syndrome [5].…”

“…According to some authors, they might represent the effect of both necrotic [2,7,8] and/or hemorrhagic [2,9] phenomena taking place in the setting of a solid PanNET. Another recognized hypothesis is that CPanNETs are independent morphological and clinical entities within the family of NETs of the pancreas [3][4][5][10][11][12]. As a matter of fact, there is scarcity of data with regard to actual differences in terms of epidemiology, biology, and long-term outcomes following surgical resection between cystic and solid pancreatic neuroendocrine tumors (SPanNETs) [5][6][7]11].…”

“…Another recognized hypothesis is that CPanNETs are independent morphological and clinical entities within the family of NETs of the pancreas [3][4][5][10][11][12]. As a matter of fact, there is scarcity of data with regard to actual differences in terms of epidemiology, biology, and long-term outcomes following surgical resection between cystic and solid pancreatic neuroendocrine tumors (SPanNETs) [5][6][7]11]. Although several surgical series report that CPanNETs can sometimes harbor malignancy [5,9,11,13], a recent metaanalysis claimed that, due to their mostly benign biology, parenchyma-sparing pancreatic resections can be advocated as the treatment of choice of CPanNETs [14].…”