Neuroendocrine markers are important not only for confirming diagnosis, but also as screening tools for future surveillance. The most commonly utilized neuroendocrine markers include chromogranin A (CgA), pancreatic polypeptide (PPP), pancreastatin, and neuron-specific enolase (NSE). CgA is a glycoprotein used commonly as a tumor marker in histopathology but also has elevated circulating levels in patients with both functional and non-functional PNETs[26,27]. However, falsely elevated levels can be observed in patients with chronic renal insufficiency, liver failure and with proton-pump inhibitor (PPI) use[28,29]. Recent recommendations by the North American Neuroendocrine Tumor Society[30] and a Canadian national expert group[29] have recommended utilizing CgA in the diagnosis and surveillance of advanced PNETs. PPP may be elevated in up to 63% of PNETs[31] and has a specificity of 84% when used during surveillance[32]. Pancreastatin may provide additional diagnostic utility, especially in patients on PPIs or with normal CgA levels[33,34]. Laboratory evaluation should also include tests to rule out F-NETs, including insulinoma and gastrinoma, if suspected (Table 2). Screening for MEN1 with serum parathyroid hormone and calcium levels should be performed in appropriate patients (e.g., diagnosis at young age, multifocal tumors, and/or with relevant personal or family history).

Cross-sectional imaging should be performed in all patients suspected of having a PNET. Computed tomography (CT) remains the initial imaging modality of choice given its good sensitivity, specificity and availability. PNETs typically are well-circumscribed lesions that appear hyperenhancing on contrast-enhanced scans. In fact, there is some evidence that hypoenhancement on arterial phase imaging is associated with more aggressive tumors and worse prognosis (Figure 1)[35]. Similarly, the presence of calcifications within these tumors on CT is associated with higher grade and the presence of lymph node metastases (Figure 2)[36]. Magnetic resonance imaging (MRI) is an alternative modality with the advantage of less radiation exposure. PNETs should be low signal intensity on T1 weighted images and high signal intensity on T2 weighted images. In addition, MRI may be more sensitive than CT for detecting smaller pancreatic lesions and liver metastases[37,38]. While ultrasound has a limited role in the diagnosis of PNETs, intraoperative ultrasound (IOUS) is very sensitive in identifying small PNETs[39], and endoscopic ultrasound (EUS) is a valuable technique for detection, localization and diagnosis through fine needle aspiration of identified lesions[40].

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Representative images of the 5 types of pancreatic neuroendocrine tumor enhancement pattern on arterial phase computed tomography. Two images are shown for each type. A: Hyperenhancing, solid; B: Cystic with hyperenhancing rim; C: Isoenhancing or no mass visualized; D: Homogeneously hypoenhancing; E: Heterogeneous but mostly hypoenhancing with some peripheral enhancement. Groups D and E had worse survival after resection compared with groups A, B, and C (From Worhunsky et al[35]. Pancreatic neuroendocrine tumours: hypoenhancement on arterial phase computed tomography predicts biological aggressiveness. HPB 2014; 16: 304-311). Arrows indicate PNET.

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Axial computed tomography images of Pancreatic neuroendocrine tumor with punctate (A, C) and dense/coarse calcifications (B, D). Despite their small size, all lesions were associated with either lymph node metastasis (A-C) or intermediate (G2) grade (B-D) on pathologic evaluation (From Poultsides et al[36]. Pancreatic Neuroendocrine Tumors: Radiographic Calcifications Correlate with Grade and Metastasis. Ann Surg Onc 2012; 19: 2295-2303).

Somatostatin receptor scintigraphy (SRS), also known as an octreotide scan, is a whole body functional imaging study that uses ¹¹¹indium labeled pentetreotide, a somatostatin analogue. Advantages include identification of unknown metastatic sites and providing important information on functional expression of somatostatin receptors which may guide systemic therapy decisions[41]. Although less available at most institutions compared to SRS, newer functional imaging studies utilizing ⁶⁸Gallium labeled 1,4,7,10-tetraazacyclododecane-N,N’,N’’,N’’’-tetraacetic acid-d-Phe(1)-Tyr(3)-octreotide ((68)Ga-DOTA-TOC) show promising results that may be superior to conventional SRS[42,43]. Although standard positron emission technology (PET) with ¹⁸flourodeoxyglucose is not typically useful in the diagnosis of NF-PNETs, PET with newer radiolabeled tracers may prove more advantageous[44,45]. In general, well differentiated tumors are positive on Octreotide scan and negative on PET scan, with the opposite being true for poorly differentiated grade 3 tumors[46].

Surgery remains the only curative treatment for NF-PNETs and is the mainstay of treatment in most cases. Appropriate candidates who undergo surgery have a significant survival advantage compared with those who do not. Hill et al[47] demonstrated a median survival difference of 114 mo vs 35 mo for patients who underwent resection compared to those who did not but were recommended to, across all patients with localized, regional and metastatic disease.

The exact surgical management must be individualized for each patient based on their particular tumor and staging. In general, most NF-PNETs should be resected. However, given the increase in incidentally discovered asymptomatic NF-PNETs, there is growing interest in the role of observation for patients with small indolent tumors. Lee et al[48] retrospectively analyzed 77 patients who underwent nonoperative observation of small, sporadic NF-PNETs without evidence of local invasion or metastasis. Median initial size was 1.0 cm and there was no documented disease specific progression or mortality during a median follow-up of 45 mo. In addition, Bettini et al[49] found that of 51 patients with incidentally diagnosed NF-PNETs < 2 cm, only 6% were malignant and there were no disease specific deaths on long term follow-up. Other population-based analyses have attempted to investigate this question but have been limited by methodological concerns[50-52]. Until better data are available, the ENETS guidelines states that intensive observation could be considered for NF-PNETs < 2 cm but risks and benefits must be carefully weighed in each patient[53].

Small low grade PNETs may safely undergo enucleation regardless of location in the pancreas, provided they are far away from the pancreatic duct and the integrity of this structure can be maintained during enucleation[54]. Enucleation may be performed in an open, laparoscopic or robotic fashion; the technique does not have an appreciable impact on morbidity, mortality, length of hospital stay or survival[55]. For larger or more aggressive NF-PNETs, formal resection is recommended. Tumors in the head of the pancreas typically require pancreaticoduodenectomy (PD) while body and tail lesions may be resected via distal pancreatectomy with or without splenic preservation. Distal pancreatectomy can often be performed via minimally invasive techniques which are associated with decreased morbidity, operative blood loss and hospital length of stay with similar rates of negative margins[56]. Minimally invasive PD has been slow to gain popularity given its greater learning curve and longer operating times. However, recent evidence suggests that it is a feasible option at select centers with potential benefits in morbidity and perhaps oncologic outcomes[57,58].

Several reports now have stressed the prognostic importance of lymph node involvement in patients with NF-PNETs[59-61]. Krampitz et al[61] found that positive lymph nodes were associated with a shorter time interval to the development of liver metastases, and in long term follow-up, a shorter disease specific survival. Similarly, Hashim et al[60] found that lymph node positivity was associated with PNETs of greater size, location in the head, high Ki-67 and with lymphovascular invasion. Furthermore, positive lymph nodes were associated with decreased median disease free survival. These data support the use of routine lymphadenectomy during resection for PNETs. Controversy exists over which lesions may forego lymphadenectomy during simple enucleation. Curran et al[59] analyzed the SEER database and found no lymph node metastases in any low grade PNETs < 1 cm. In contrast, Gratian et al[50] found that among tumors < 0.5 cm in the national cancer database, 33% presented with regional lymph node metastases and 11% with distant metastases. Formal resection with adjacent lymphadenectomy, as opposed to enucleation, is the procedure of choice for PNETs greater than 2 cm, of higher grade, or with radiographic calcifications.

Several authors have described the role of aggressive extended resections for advanced PNETs[62-65]. For example, Norton et al[62] describe acceptable morbidity, low mortality, and excellent overall survival rates, albeit high recurrence rates, in patients with advanced PNETs. Norton et al[63] also described good outcomes for patients with PNETs with major vessel involvement undergoing simultaneous vascular reconstruction. Surgeons at experienced hepatopancreaticobiliary centers may follow standard oncologic principles, including multivisceral and vascular resections, in order to accomplish R0 resections.

All patients with liver metastases from PNETs should be considered for surgical intervention. Although resection can be associated with high recurrence rates, it does improve progression free survival as well as symptom control[66-71]. Saxena et al[66] performed a meta-analysis of 1469 GEP-NETs metastatic to the liver and found 3, 5, and 10 year overall survival rates of 83%, 70.5%, and 42%, respectively, following hepatic resection. Predictors of poor outcomes included poor histologic grade, incomplete resection and extrahepatic disease[66]. When patients are not candidates for resection, alternative methods, such as thermal ablation or hepatic artery embolization, are helpful strategies that improve local control and palliate symptoms[67,69,71,72]. Insufficient data exists to recommend one liver-directed strategy over another[73]. Liver transplantation has been described for well selected patients with metastatic GEP-NET[74]. However, liver transplantation for neuroendocrine liver metastases of pancreatic origin is associated with worse overall outcomes and is not typically recommended[75].

In the setting of metastatic disease, controversy remains regarding the role of surgery for the primary tumor[53,76,77]. Capurso et al[77] performed a systematic review on this topic and found improved overall survival in patients undergoing resection of the primary in 2 of 3 retrospective cohort studies identified. One potential benefit of removal of the primary tumor is allowing providers to focus treatment on the liver metastatic sites with hepatic artery therapies. Primary tumors that are symptomatic should generally undergo resection for palliation of symptoms[78].

Nearly 80% of NF-NETs express somatostatin receptors, making them a suitable target for therapy with somatostatin analogues. In addition to their favorable safety profile and effectiveness in controlling symptoms, recent evidence has suggested improvements in oncologic outcomes as well. The PROMID trial[79] was a placebo-controlled double blinded randomized controlled trial (RCT) of long acting release (LAR) octreotide in patients with metastatic well differentiated midgut NETs. Median progression free survival was 14.3 mo in patients receiving octreotide vs 6 mo in the placebo group. More recently, the CLARINET trial randomized 204 patients with enteropancreatic NETs to receive long acting lanreotide or placebo and found significantly prolonged progression free survival in the lanreotide group (65.1% vs 33.0% at 24 mo); this finding was confirmed in a subset of patients with PNETs (Figure 3)[80].

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Results of the Clarinet trial which randomized patients with enteropancreatic neuroendocrine tumors to lanreotide vs placebo. From: Caplin et al[80]. Lanreotide in Metastatic Enteropancreatic Neuroendocrine Tumors. N Engl J Med 2014; 371: 224-233.

Peptide receptor radionuclide therapy (PRRT) also makes use of PNETs’ octreotide receptor expression by coupling radionuclides to somatostatin analogues. Typical radionuclides include ⁹⁰yttrium and ¹⁷⁷lutetium. Response rates range only between 10%-40% with toxicity (primarily bone marrow and renal) rates in a similar range so PRRT should be reserved for cases not responsive to less toxic therapies[81-83]. Furthermore, having been pioneered at the Erasmus Medical Center in the Netherlands, PRRT is still only available at select centers in Europe and North America and randomized data are lacking.

Indolent and well differentiated NETs are typically resistant to traditional systemic chemotherapy which is therefore reserved for patients with high grade, poorly differentiated tumors or with rapidly progressive unresectable disease[84]. However, NETS of pancreatic origin, generally respond better to chemotherapy than other GEP-NETs. Streptozocin was one of the first agents shown to have activity patients with metastatic PNETs, either as monotherapy or in combination with doxorubicin or fluorouracil[85-87]. Currently, platinum based therapy remains the standard of care for patients with high grade metastatic PNETs. Various combinations exist but the most common regimen utilized consists of cisplatin and etoposide. Nevertheless, data supporting the use of this regimen is limited and more evidence is needed to clarify its role as aggressive first line therapy[88].

More recent research has focused on the use of oral temozolamide with or without capecitabine given its ease of administration and favorable side effect profile. A retrospective review of 30 patients with well or moderately differentiated PNETs treated with this regimen demonstrated a 70% response rate and a median PFS of 18 mo[89]. Additional research is becoming available regarding the safety and efficacy of this regimen[90,91].

Increasingly, PNETs have been found to be responsive to targeted therapies. The purpose of these molecular agents is to stabilize disease progression in metastatic unresectable cases[92]. Much focus has been placed on Everolimus, an oral mTOR (mammalian target of rapamycin) inhibitor. Previously, Everolimus had been found to have clinical benefit in patients who progressed while on systemic cytotoxic chemotherapy[93]. The RADIANT-3 trial was an international, multisite, RCT comparing daily Everolimus to placebo in patients with low or moderate grade NF-PNETs. Although response rates were low, PFS was longer in the Everolimus group (11.0 mo vs 4.6 mo) (Figure 4)[94]. Similarly, the RADIANT-2 trial evaluated Everolimus in conjunction with long acting octreotide and found improved PFS in the Everolimus plus octreotide LAR group vs octreotide LAR alone(16.4 mo vs 11.3 mo)[95]. Common adverse effects included stomatitis, rash and diarrhea. Some have suggested this regimen should be first line therapy for most NETs[96].

Open in New Tab   Full Size Figure   Download Figure

Figure 4 Results of the RADIANT-3 trial which randomized patients with nonfunctional neuroendocrine tumors of the pancreas to Everolimus vs placebo. From Yao et al[94]. Everolimus for Advanced Pancreatic Neuroendocrine Tumors. N Engl J Med 2011; 364: 514-523.

Sunitinib is an oral, small-molecule, tyrosine kinase inhibitor with activity against vascular endothelial growth factor (VEGF) and platelet-derived growth factor (PDGF) both of which are expressed abundantly in PNETs[97]. A placebo controlled, double blind, RCT of daily sunitinib in patients with well-differentiated PNETs with documented disease progression found improved PFS in patients receiving Sunitinib (11.4 mo vs 5.5 mo) (Figure 5)[98]. Finally, newer therapies that target the mTOR (e.g., temsirolimus) and VEGF (e.g., bevacizumab) pathways are currently being investigated and hold promise both as single agents and in combination[92,99].

Open in New Tab   Full Size Figure   Download Figure

Figure 5 Results of a randomized controlled trial of Sunitinib vs placebo for well-differentiated pancreatic neuroendocrine tumors demonstrating (A) progression free survival and (B) overall survival. From: Raymond et al[98]. Sunitinib Malate for the Treatment of Pancreatic Neuroendocrine Tumors. N Engl J Med 2011; 364: 501-513.