Available online at www.sciencedirect.com
ScienceDirect EJSO 42 (2016) 1471e1477
Localized high-grade gastroenteropancreatic neuroendocrine tumors: Defining prognostic and therapeutic factors for a disease of increasing clinical significance* C. Mosquera a, N.J. Koutlas b, T.L. Fitzgerald a,* a
East Carolina University, Brody School of Medicine, Division of Surgical Oncology, Greenville, NC, USA b East Carolina University Brody School of Medicine, Greenville, NC, USA Accepted 26 July 2016 Available online 6 August 2016
Abstract Background: Due to the limited sample size in the existing series, the natural history and management of high-grade gastroenteropancreatic neuroendocrine tumors (GEP-NET) is poorly understood. In order to better understand high-grade GEP-NET, a large cohort study was undertaken. Objective: To determine the prognostic factors associated with high-grade GEP-NET. Methods: Patients diagnosed with non-metastatic high-grade GEP-NET from 1988 to 2010 were identified in SEER. Results: Incidence of high-grade GEP-NETs increased from 0.03 to 0.19/100,000 over the study period. The median age was 65 years, and the majority of the patients were white and females. The most common primary site was colorectal, and the most frequent T classification was T3. Surgical resection was performed in 89% of patients that varied by site (p < 0.0001). Nodal involvement was frequent and varied by site (p ¼ 0.0002). The 5-year disease-specific survival was 63.3% and was the greatest for small bowel (p ¼ 0.0003). Survival was associated with age, node status and surgery (p < 0.05). On multivariate analysis, the node status, surgery, and site continued to be associated with survival (p < 0.05); however, age (p ¼ 0.08) no longer influenced the patient’s survival. Conclusion: High-grade GEP-NETs are neoplasms with exponentially increasing in incidence. Tumor location and nodal status are predictors of survival. Surgery is associated with a survival advantage and could be considered for localized disease. Ó 2016 Elsevier Ltd, BASO ~ The Association for Cancer Surgery, and the European Society of Surgical Oncology. All rights reserved.
Keywords: Survival; High-grade gastroenteropancreatic neuroendocrine tumors; Prognostic factors; Incidence
Introduction Neuroendocrine tumors (NETs) are a heterogeneous group of neoplasms, a majority of which originate in the gastrointestinal tract and are designated as gastroenteropancreatic-NET (GEP-NET).1 Although GEP-NETs were once rarely observed, multiple recent reports have documented an increase in the incidence of this disease.1e4 Fraenkel recently reported a 3.65-fold incidence increase in the last four decades in the US.5 *
Presented at the 11th Annual Academic Surgical Congress. * Corresponding author. Brody School of Medicine, 4S24, 600 Moye Boulevard, Greenville, NC 27834, USA. Fax: þ1 252 744 5777. E-mail address: [email protected]
The natural history of localized GEP-NETs varies depending largely on the primary location and grade. The well-differentiated (low to intermediate grade) neoplasms are often found to be slow-growing and asymptomatic and can often be treated with limited resection or observation if diagnosed at an early stage. However, poorly differentiated (high-grade) neoplasms have a proclivity to metastatic dissemination and poor prognosis and may require more aggressive medical or surgical therapy.6 In the case of GEP-NET, the grading system seems to be one of the most important determinates of clinical behavior.3,7e9 The prognosis of poorly differentiated GEPNETs is more similar to small bowel carcinoma than to the well-differentiated GEP-NET.10
http://dx.doi.org/10.1016/j.ejso.2016.07.137 0748-7983/Ó 2016 Elsevier Ltd, BASO ~ The Association for Cancer Surgery, and the European Society of Surgical Oncology. All rights reserved.
C. Mosquera et al. / EJSO 42 (2016) 1471e1477
A multitude of studies have been done on GEP-NETs. Unfortunately, most of these are retrospective singlecentered8,11,12 or large dataset studies.1,3 These studies are often inclusive of all grades of neuroendocrine neoplasms analyzed as a single entity. Given the clinical significance of tumor grade, the further extrapolation of treatment paradigms is problematic. However, only a few retrospective series have reported prognostic factors exclusive to high-grade GEP-NET.7 The lack of sufficient data specific to high-grade GEPNETs has resulted in no clear consensus on its treatment. Extrapolation of therapeutic paradigms from the lowgrade GEP-NET would indicate that surgery is the primary curative therapy.13 Pape reported surgery to be associated with improved long-term outcomes for a group of patients with mixed grade GEP-NETs.8 Similarly, in a comprehensive review of carcinoid tumors, Schinirer (2011) recommended surgery to be the primary treatment option for all patients with loco-regional diseases.13 The optimal surgical treatment was predicated solely on the stage of presentation.8,13e17 The recommendations are contradictory when data was inferred from high-grade pulmonary NETs. In this setting, surgery was not considered as the primary therapy. In fact, surgery alone is rarely curative. The benefits of surgery following neoadjuvant chemotherapy, even in localized disease, are unknown. Given the aggressive metastatic potential of this disease, the platinum-based regimens are considered for primary therapy.18,19 The National Comprehensive Cancer Network (NCCN) provides the therapeutic guidelines for poorly differentiated high-grade neuroendocrine tumors. The guidelines are based on the (WHO G3) NORDIC NEC study7 and therapeutic guidelines for small cell lung cancer.20 The (WHO G3) NORDIC NEC study presents factors associated with the treatment of advanced GEP-NETs (patients with metastatic disease at diagnosis or patients with localized unresectable disease), with little guidance for those with localized disease. As guidelines are inferred from data that is not specific to localized high-grade GEP-NETs, the applicability of the (WHO G3) NORDIC NEC study for these patients is debatable.21 Indeed, the data obtained from contemporary studies suggested different prognosis and response to therapy for extra-pulmonary NET when compared to the pulmonary small-cell cancers. Given the rarity of data and the controversy regarding the optimal treatment paradigms that are largely based on inference, additional studies on high-grade GEP-NET are imperative for distinguishing the low- and high-grade GEP-NETs. The rationale for the present study was to analyze the outcomes from a large national dataset to better determine the prognostic factors associated with high-grade GEP-NET and also define the treatment algorithms unique to this entity.
Materials and methods Data source The SEER tumor registry was used to extract the patient’s data. The National Cancer Institute SEER program collected the data on patient demographics, tumor characteristics, and treatment methods. The registry also provided information on cancer incidence and survival for around 28% of the US population.22 A SEER registry query was performed from 1988 to 2010 to identify the patients with high-grade GEP-NET. ICD-0-3 histologic codes were used to identify the patients. These codes included: (8150) islet cell carcinoma, (8151) insulinoma malignant, (8152) glucagonoma malignant, (8153) gastrinoma malignant, (8154) mixed islet cell and exocrine adenocarcinoma, (8155) vipoma, (8156) somatostatinoma malignant, (8240) carcinoid tumor, (8240) carcinoid tumor malignant, (8241) enterochromaffin cell carcinoid, (8242) enterochromaffin-like cell tumor malignant, (8243) goblet cell carcinoid, (8244) composite carcinoid, (8246) neuroendocrine carcinoma, (8249) atypical carcinoid tumor, and (8238) apudoma. The sites with five or fewer patients (gallbladder, ampulla of Vater and anus) were excluded. Only poorly differentiated and anaplastic (grade 3/4) neoplasms were included. High-grade GEPNET incidence and survival analysis were performed for the overall population as well as independently by grade. The survival analysis censor was determined based on the “SEER cause-specific death classification” and “COD to site recode.” The patients categorized as “N/A not first tumor” in SEER cause-specific death classification were further analyzed. If the “site category” was the same as the primary site, the patients were classified as “dead from the specific disease”. If the death was attributable to other causes, then the patients were classified as “dead of other cause” for the purpose of disease-specific survival. Our main interest was the impact of surgical therapy on high-grade GEP-NET; cases recognized as stage IV (metastatic) were excluded from the study (48.2% in the anaplastic group and 41.8% in the poorly differentiated group). The patient demographics included age, gender, and race. The tumor characteristics were coded using a combination of the SEER staging systems: historic, collaborative, and extent of disease, and included the location of the primary tumor, size, and nodal status. Treatment was classified as a binomial variable based on the use or non-use of surgery. The patients with unknown surgical intervention were excluded from the study. As these data are de-identified the IRB at the Brody School of Medicine, East Carolina University does not consider this human subjects research requiring IRB approval.
C. Mosquera et al. / EJSO 42 (2016) 1471e1477
Statistical analysis The data were represented as mean, median or percentage based on the requirement. Student’s T test or X2 test were used to perform univariate analysis. KaplaneMeier method was used for univariate survival analysis, and the Cox regression hazard model was used for the multivariate survival analysis. The analysis was conducted using JMPÒ Pro version 10.0.0; 2012 (SAS Institute Inc, Cary, NC, USA). Results Dataset characteristics Demographics A total of 600 patients with high-grade GEP-NET were identified, of which 43.16% of patients were classified as Stage IV, 55.8% (335 patients) had local/regional stage
IV disease, and 1% were un-staged. The demographics of 335 patients with localized high-grade GEP-NET, poorly differentiated and anaplastic disease, are presented in Table 1. The mean age was 65 years, and the median age was 64.7 years (range 20e90 years). There was a slight women preponderance (52.2%), and the majority were found to be white (84.5%). Tumor characteristics The most common primary site for tumor occurrence was colorectal (38.2%, n ¼ 128), followed by appendix (19.7%, n ¼ 66), small bowel (15.8%, n ¼ 53), pancreas (14.9% n ¼ 50) and stomach (11.3%, n ¼ 38). Most neoplasms were <4 cm (44.8%, n ¼ 150), followed by 4 cm in 31.9% cases, and unknown in 23.3% of cases. Tumors in the small bowel were more likely to be < 4 cm (84.4%), followed by the stomach (70.8%) and appendix (64.4%). The pancreatic (57.5%) and colorectal (52.4%) neoplasms were more likely to be larger (4 cm),
Table 1 Demographics of patients with high-grade gastroenteropancreatic neuroendocrine tumors, SEER Registry 1988e2012. Factor
Gender Male Female Age Median Mean Race White Black Other Unknown Site Colorectal Appendix Small bowel Pancreas Gastric N stage N0 N1 Unknown Size <4 cm 4 cm Unknown Extension Invades through muscularis propria Invades serosa or adjacent structures Invades lamina propria Pancreas Invades muscularis propria Unknown Surgery Yes No Total
Poorly differentiated No
84.5 10.4 4.2 0.9
283 35 14 3
84.3 10.28 4.25 1.0
238 29 12 3
84.9 11.3 3.7 0
45 6 2 0
38.2 19.7 15.8 14.9 11.3
128 66 53 50 38
37.2 19.8 15.2 14.9 12.7
105 56 43 42 36
43.4 18.8 18.8 15.1 3.7
23 10 10 8 2
50.7 41.2 8.1
170 138 27
49.2 42.9 7.8
139 121 22
58.5 32.0 9.4
31 17 5
44.8 31.9 23.3
150 107 78
46.09 31.2 22.6
130 88 64
37.7 35.8 26.4
20 19 14
33.7 21.5 16.4 14.9 9.0 4.5
113 72 55 50 30 15
33.6 23.7 15.6 14.9 8.1 3.9
95 67 44 42 23 11
33.9 9.4 20.7 15.1 13.2 7.5
18 5 11 8 7 4
298 37 335
251 31 282
47 6 53
C. Mosquera et al. / EJSO 42 (2016) 1471e1477
Figure 1. Incidence of high-grade GEP NETs from 1988 to 2012.
(p < 0.0001). T classification was determined by the depth of tumor invasion; T1: tumor invades lamina propria, T2: tumor invades muscularis propria, T3: tumor invades through muscularis propria, and T4: tumor invades serosa or adjacent structures. T3 was the predominant T stage (33.7%, n ¼ 113), followed by T4 (21.5%, n ¼ 72), T1 (16.4%, n ¼ 55), and T2 (9%, n ¼ 30). Pancreatic and colorectal primaries were more likely to have nodal metastasis (57.8% and 55.1% respectively), whereas stomach (75.8%), appendix (72.6%) and small bowel (56%) were commonly node negative (p ¼ 0.0002). Treatment The majority of the patients (89%) underwent surgery. Surgery was more common for appendiceal primaries (97%), followed by colorectal (94.5%), and small bowel tumors (88.7%). Surgery was less often performed in cases of pancreatic (78%) and gastric neoplasms (71.1%).
The median disease-specific survival for all the stages was 42 months (range 0e290 months) with a 5y-DDS of 44.7%. The survival was much lower for stage IV disease (13 months) (range 0e215 months) with 5y-DDS of 19.5%. The results were much better for the subgroup of patients with loco/regional disease, 149 months (range 0e290 months) median survival and 5y-DDS of 63.3%. Survival was found to be slightly better for women (153 months vs. 149 months; p value ¼ 0.88). On univariate analysis, age, site of primary tumor, node status, and surgical treatment were found to be associated with 5-y diseasespecific survival (DSS) (p < 0.05); however, race and tumor size were not (Table 2). The younger patients (<65 years old) had greater 5y-DDS of 70.7% compared to the older patients (56.1%) (p ¼ 0.0034). The 5-y DDS was greatest for small bowel primaries (81.1%) followed by gastric (76.9%), appendiceal (75.5%), colorectal (54.2%), and pancreatic (45.4%) primaries (p < 0.0001). The node status also had a profound impact on survival. The patients with lymph node metastasis had a shorter median survival (78 months) and lower 5-year DSS at 54.2%, compared to 73.6% of 5-year DSS in the node negative (N0) cases (p ¼ 0.012). The median survival for patients undergoing surgery was 153 months vs. 71 months for those not undergoing surgery. The 5-y DSS for patients undergoing surgery was 64.6% and decreased to 51.5% for patients not undergoing surgery (p ¼ 0.058). The definitive conclusion in the non-surgical cohort must be viewed with caution, due to the small numbers (n ¼ 37). On Cox regression analysis, node status N0 (OR 1.91, p ¼ 0.0091), surgical treatment (OR 5.25, p 0.0010) and the primary site continued to be associated with survival. Specifically, the small bowel (OR 5.41, p ¼ 0.003) and gastric tumors (OR 3.9, p ¼ 0.075) had superior survival rate compared to the appendix (OR 1.89, p ¼ 0.16), colorectal (OR 1.16, p ¼ 0.79), and pancreatic neoplasms (referent). Age no longer influenced the patient’s survival. Discussion
Incidence To understand the epidemiology of high-grade GEPNETs, its incidence was calculated with the occurrence and SEER registry population. There was a clear emergence of the epidemiologic trend (Fig. 1). The incidence of high-grade GEP-NET in 1988 was 0.026 in 100,000 individuals, and by 2012, it had increased to 0.186 per 100.000 individuals, indicating a 700% increase over a period of 14 years. R2 for this trend was 0.793 and p < 0.0001. The trend was also noticeable when the occurrences were independently evaluated by site, in which the NETs of the appendix increased by 745% followed by colorectal (622%), pancreas and small bowel (>200%), and gastric (135%) primaries.
The natural history of high-grade GEP-NETs is poorly understood. A clear understanding of this entity is imperative given the increasing incidence of this aggressive disease. In this study, a 700% increase in incidence from 1988 to 2012 of high-grade GEP-NET was noted. The overall survival for all groups included in the present study (49 months) was similar to other contemporary series.5 A large percentage of patients presented with metastatic disease (43%) and had poor survival (only 13 months). However, the survival for the loco-regional disease was much longer (149 month’s median and 63.3% of 5y-DSS). Our data on loco-regional diseases suggested the primary location and node status to be useful survival predictors. Finally, there is an association between surgical intervention and survival.
C. Mosquera et al. / EJSO 42 (2016) 1471e1477
Table 2 Survival in Patients with High-Grade GEP-NET’s overall and by grade, SEER Registry 1988e2012. Factor
High-grade GEP-NET Univariate
Gender Male Female Age <65 65 Race White A.A* Other Site Small bowel Gastric Appendix Colorectal Pancreas N stage N0 N1 Size <4 cm 4 cm Surgery Yes No 5y-DSS
65.1 59.3 40.4
81.1 76.9 75.5 54.2 45.4
64.6 51.5 63.3
Poorly differentiated GEP-NET
63.3 40.0 0
51.4 50.0 100 49.5 43.7
59.7 27.7 56.6
Ref. 1.6 1.8
65.4 57.9 49.38
Ref. 1.4 2.9 4.7 5.4
87.8 79.5 71.36 55.1 45.4
0.62 0.071 <0.001 0.003
65.46 54.7 64.3
Ref 1.4 3.9 5.0 5.7
0.57 0.013 <0.001 <0.001
A.A*: African American.
The incidence of high-grade GEP-NET is increasing. In this study, the incidence increased 700% in the period from 1988 to 2012. The results obtained from the present study were in consistence with other epidemiologic studies. A review of The Netherlands Cancer Registry demonstrated an incidence increase of 300% in high-grade NETs.20 In 2011, a national (US) epidemiologic review was undertaken that considered the data obtained from the Third National Cancer Survey, The End Results Group, and SEER for evaluation. The results demonstrated an increase in the incidence of GEP-NET, from 1 to 3.65/100,000 for the period of 1973 to 2007.2 Unfortunately, the data was not independently analyzed by grade. The survival of patients with high-grade GEP-NET is unfavorable when compared to other GEP-NETs. With survival of more than 80% at five years, the low-grade GEPNET (carcinoids tumor) is often considered as an indolent cancer.13 However, the high-grade NETs have a reported 5-year survival of less than 10%.18,24 We report a 63.3% 5y-DSS for loco-regional diseases, and the results were similar to 10y-DSS reported by Pape. This study, however, includes both high- and low-grade GEP-NETs,8 and thus, it is the likely explanation for our 5y-DSS rate to be similar to that of the 10y-DSS. When our data was compared to other studies restricted to high-grade neoplasms, the survival rates were found to be higher.23,24 A two-decade long
European data analysis including both high-grade GEPNETs and pulmonary NETs reported a 20% 5-year survival for Grade 3 large cell neuroendocrine tumors and 6% for small cell neuroendocrine tumors.23 Similar rates were reported by Lepage with a 6e11% of 5y-DSS in poorly differentiated GEP-NET.24 The survival rate reported in these studies was similar to Stage IV rates of our data. However, on an overall basis, we found survival to be profoundly better for earlier stages of the disease. This may be secondary to the heterogeneous nature of high-grade GEPNET. The classification of high-grade GEP-NET has engendered some controversy. Many would argue that Ki67 should be unitized to classify patients in two groups those 20e55% (well-differentiated grade 3 neuroendocrine tumors) and those >55% (neuroendocrine carcinoma.)25 These data are not available in the SEER registry. The optimal therapeutic paradigm in the high-grade GEP-NETs has engendered significant controversy. As noted above, the current treatment regimens are extrapolated from different entities such as low-grade GEP-NET and small cell lung cancer. Based on the reported data, we recommended surgical resection be considered for localized high-grade GEP-NETs. The patients undergoing surgery had prolonged median disease-free survival significantly compared to those not undergoing resection. Our results were in concordance with a 1988 report that evaluated
C. Mosquera et al. / EJSO 42 (2016) 1471e1477
the gastrointestinal carcinoid tumors and recommended to “resect primary tumor to negative margins with the goal to improve outcomes and quality of life”.9 Similar results were also presented by Pape et al.,8,11 in which they reported that any oncologic procedure independent of its results would be beneficial for the evaluation of survival. However, the benefit would be greater if complete resection (R0) was achieved due to disease recurrence impact on survival. Surgery was found to be the only treatment option for the potential cure of loco-regional neoplasms. Oberg’s report on the results of clinical practice guidelines was in support with our findings as it presented an 80e100% 5yDSS in localized/resectable GEP-NET. However, he also reported that surgery played an important role in extensive disease and recommended to add medical therapy either before or concomitantly with surgical intervention.17 The primary site and node status are considered to be the useful prognostic factors for high-grade GEP-NET. We found >75% of 5y-DSS in the patients with tumors in the small bowel, stomach, and appendix, and 54.2% and 45.4% for colorectal and pancreatic neoplasms, respectively. The results differed from the studies that were not exclusively high-grade GEP-NET. In these studies, the appendiceal neoplasms were reported to have the best median survival rate for GEP-NET.3 Our experience is very similar to the one presented by Lawrence (2011). In this series of GEP-NETs, the pancreatic neoplasms had the lowest survival, followed by colon, small bowel, appendix, and gastric neoplasms.2 In addition, regional lymph node metastases are common and are found to be associated with diminished survival for high-grade GEP-NETs.1,14 We found a 20% 5yDSS in the patients with node-positive disease. To the best of our knowledge, there are no previous studies that have evaluated the incidence and impact of regional nodal metastasis in high-grade GEP-NETs. However, the previous series evaluating other NETs reported the relevance of node disease upon survival. Yao reported that the node status is associated with a significantly decreased median survival in patients diagnosed with NETs when compared to the patients with only localized or node negative disease. They stated the presence of positive nodes to be associated with a 50% decrease in the median survival. These findings have been supported by several similar studies.8,13 Unfortunately, the SEER registry does not report chemotherapy data. This seems to be a significant weakness of this study because we can provide little insight into the adjuvant treatment of high-grade GEP-NETs. However, previous oncological studies supported the routine use of adjuvant chemotherapy after surgical resection.16,19 Smith extrapolated data from the high-grade pulmonary NETs to conclude that chemotherapy offers significant disease control benefit.17 Based on similar data, Fazio suggested that chemotherapy was the main therapeutic treatment modality for high-grade GEP-NETs.19 Further data analysis is needed to better understand the benefit of adjuvant therapy
for this rare entity. This study has several limitations, many of which are inherent to large SEER registry studies, including the inability to classify patients according to the 2010 WHO system or Ki-67 proliferation index data. Currently SEER classifies solid neoplasm as Grade I (well-differentiated), Grade II (moderately differentiated), Grade III (poorly differentiated), and Grade IV (undifferentiated) without reference to Ki-67. There was also a lack of adjuvant therapy data, which may significantly influence survival data. The major concerns are the small number of patients managed without surgical intervention and the inability to further characterize patient data. A greater number of nonoperatively managed patients would probably bring useful information on the prognosis and survival of this disease. Conclusion In conclusion, the high-grade GEP-NET is a disease that has experienced an exponential increase in its incidence. For those with localized disease, the survival seems to be surprisingly good. Although the numbers in this study are small and conclusions are thus limited, surgery appears to impart a significant survival advantage. The incidence of regional nodal metastasis is high and associated with a survival disadvantage. Additional data is needed to define the optimal treatment for high-grade GEP-NETs. Conflict of interest Nothing to disclose.
References 1. Modlin IM, Lye KD, Kidd M. A 5-decade analysis of 13,715 carcinoid tumors. Cancer 2003;97(4):934–59. http://dx.doi.org/10.1002/cncr.11105. 2. Lawrence B, Gustafsson BI, Chan A, Svejda B, Kidd M, Modlin IM. The epidemiology of gastroenteropancreatic neuroendocrine tumors. Endocrinol Metab Clin North Am 2011;40(1):1–18. http://dx.doi.org/ 10.1016/j.ecl.2010.12.005 vii. 3. Yao JC, Hassan M, Phan A, et al. One hundred years after “carcinoid”: epidemiology of and prognostic factors for neuroendocrine tumors in 35,825 cases in the United States. J Clin Oncol 2008;26(18):3063–72. http://dx.doi.org/10.1200/JCO.2007.15.4377. 4. Fitzgerald TL, Dennis SO, Kachare SD, Vohra NA, Zervos EE. Increasing incidence of duodenal neuroendocrine tumors: incidental discovery of indolent disease? Surgery 2015;158(2):466–71. http:// dx.doi.org/10.1016/j.surg.2015.03.042. 5. Fraenkel M, Kim M, Faggiano A, de Herder WW, Valk GD. Knowledge NETwork. Incidence of gastroenteropancreatic neuroendocrine tumours: a systematic review of the literature. Endocr Relat Cancer 2014;21(3):R153–63. http://dx.doi.org/10.1530/ERC-13-0125. 6. Sorbye H, Strosberg J, Baudin E, Klimstra DS, Yao JC. Gastroenteropancreatic high-grade neuroendocrine carcinoma. Cancer 2014; 120(18):2814–23. http://dx.doi.org/10.1002/cncr.28721. 7. Sorbye H, Welin S, Langer SW, et al. Predictive and prognostic factors for treatment and survival in 305 patients with advanced gastrointestinal neuroendocrine carcinoma (WHO G3): the NORDIC NEC study. Ann Oncol 2013;24(1):152–60. http://dx.doi.org/10.1093/annonc/mds276.
C. Mosquera et al. / EJSO 42 (2016) 1471e1477 8. Pape UF, Berndt U, Muller-Nordhorn J, et al. Prognostic factors of long-term outcome in gastroenteropancreatic neuroendocrine tumours. Endocr Relat Cancer 2008;15(4):1083–97. http://dx.doi.org/10.1677/ ERC-08-0017. 9. Shebani KO, Souba WW, Finkelstein DM, et al. Prognosis and survival in patients with gastrointestinal tract carcinoid tumors. Ann Surg 1999;229(6):815–21. discussion 822e3. 10. Yachida S, Vakiani E, White CM, et al. Small cell and large cell neuroendocrine carcinomas of the pancreas are genetically similar and distinct from well-differentiated pancreatic neuroendocrine tumors. Am J Surg Pathol 2012;36(2):173–84. http://dx.doi.org/10.1097/ PAS.0b013e3182417d36. 11. Pape UF, Bohmig M, Berndt U, Tiling N, Wiedenmann B, Plockinger U. Survival and clinical outcome of patients with neuroendocrine tumors of the gastroenteropancreatic tract in a german referral center. Ann N. Y Acad Sci 2004;1014:222–33. 12. de Miguel Novoa MP, Fernandez Capel F, Redondo Sedano JV, et al. Gastroenteropancreatic neuroendocrine tumors: clinical characteristics, diagnosis and prognosis at hospital universitario clinico san carlos (madrid). Endocrinol Nutr 2014;61(5):234–41. http://dx.doi.org/ 10.1016/j.endonu.2013.12.001. 13. Schnirer II , Yao JC, Ajani JA. Carcinoidea comprehensive review. Acta Oncol 2003;42(7):672–92. 14. Fendrich V, Bartsch DK. Surgical treatment of gastrointestinal neuroendocrine tumors. Langenbecks Arch Surg 2011;396(3):299–311. http://dx.doi.org/10.1007/s00423-011-0741-7. 15. Alexandraki KI, Kaltsas G. Gastroenteropancreatic neuroendocrine tumors: new insights in the diagnosis and therapy. Endocrine 2012; 41(1):40–52. http://dx.doi.org/10.1007/s12020-011-9562-2. 16. Akerstrom G, Hellman P. Surgical aspects of neuroendocrine tumours. Eur J Cancer 2009;45(Suppl. 1):237–50. http://dx.doi.org/10.1016/ S0959-8049(09)70039-5.
17. Oberg K, Akerstrom G, Rindi G, Jelic SESMO Guidelines Working Group. Neuroendocrine gastroenteropancreatic tumours: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol 2010;21(Suppl. 5):v223–7. http://dx.doi.org/10.1093/annonc/ mdq192. 18. Fazio N, Spada F, Giovannini M. Chemotherapy in gastroenteropancreatic (GEP) neuroendocrine carcinomas (NEC): a critical view. Cancer Treat Rev 2013;39(3):270–4. http://dx.doi.org/10.1016/ j.ctrv.2012.06.009. 19. Smith J, Reidy-Lagunes D. The management of extrapulmonary poorly differentiated (high-grade) neuroendocrine carcinomas. Semin Oncol 2013;40(1):100–8. http://dx.doi.org/10.1053/j.seminoncol. 2012.11.011. 20. Strosberg JR, Coppola D, Klimstra DS, et al. The NANETS consensus guidelines for the diagnosis and management of poorly differentiated (high-grade) extrapulmonary neuroendocrine carcinomas. Pancreas 2010;39(6):799–800. http://dx.doi.org/10.1097/MPA.0b013e3181ebb56f. 21. National Comprehensive Cancer Network. http://www.nccn.org/ professionals/physician_gls/pdf/neuroendocrine.pdf. Updated 11/11/ 14. [accessed 11.09.15], 2015. 22. Surveillance Research Program. http://seer.cancer.gov/. Updated 2015. [accessed 11.09.15], 2015. 23. Korse CM, Taal BG, van Velthuysen ML, Visser O. Incidence and survival of neuroendocrine tumours in The Netherlands according to histological grade: experience of two decades of cancer registry. Eur J Cancer 2013; 49(8):1975–83. http://dx.doi.org/10.1016/j.ejca.2012.12.022. 24. Lepage C, Ciccolallo L, De Angelis R, et al. European disparities in malignant digestive endocrine tumours survival. Int J Cancer 2010; 126(12):2928–34. http://dx.doi.org/10.1002/ijc.24698. 25. Coriat R, Walter T, Terris B, Couvelard A, Ruszniewski P. Gastroenteropancreatic well-differentiated Grade 3 neuroendocrine tumors: review and position statement 2016;July 8. (Epub ahead of print).