REVIEW

CURRENT
OPINION Pancreatic neuroendocrine tumors
Giampaolo Perri, Laura R. Prakash, and Matthew H.G. Katz

Purpose of review
Pancreatic neuroendocrine tumors (pNETs) are a rare, heterogeneous group of pancreatic neoplasms with
a wide range of malignant potential. They may manifest as noninfiltrative, slow-growing tumors, locally
invasive masses, or even swiftly metastasizing cancers.
Recent findings
In recent years, because of the increasing amount of scientific literature available for pNETs, the
classification, prognostic stratification criteria, and available consensus guidelines for diagnosis and
therapy have been revised and updated.
Summary
The vast majority of new pNET diagnoses consist of incidentally discovered lesions on cross-sectional
imaging. The biologic behavior of pNETs is defined by the grade and stage of the tumor. Surgery is the
only curative treatment and it, therefore, represents the first therapeutic choice for any localized pNET;
however, recent evidence suggests that patients with small (<2 cm), nonfunctioning G1 tumors can be
safely observed.
An aggressive surgical approach towards liver metastases is recommended in selected cases, as well as
liver-directed therapies for disease control. In the presence of unresectable progressive disease,
somatostatin analogs, targeted therapies such as everolimus, peptide receptor radionuclide therapy, and
systemic chemotherapy are all useful tools for prolonging survival.
Keywords
neuroendocrine tumor, octreotide, pancreas, pancreatic islet cell tumor, peptide receptor radionuclide therapy

INTRODUCTION, EPIDEMIOLOGY, staining alone is not a defining criterion for tumor

NOMENCLATURE classification, as both F-pNETs and NF-pNETs may
Pancreatic neuroendocrine tumors (pNETs) are a secrete multiple peptides. NF-pNETs account for the
heterogeneous group of neoplasms that arise from majority (50–75%) of pNETs.
the endocrine tissues of the pancreas. Previously Although the vast majority of pNETs are spo-
known as islet cell tumors because of their presumed radic, approximately 10% of all pNETs can be also
origin in the islets of Langerhans, they comprise less associated with hereditary genetic endocrinopa-
than 2% of all pancreatic neoplasms and approxi- thies. These tumors are found in 80–100% of
mately 7% of all neuroendocrine tumors (NETs) patients with multiple endocrine neoplasia type I
across all anatomic sites [1]. These tumors have an (MEN-1), in whom NF-pNET, gastrinomas, and insu-
incidence of 1 or less case per 100 000 individuals linomas are most common. NF-pNETs are found in
per year [2,3] but their incidence has increased approximately 20% of patients with von Hippel
significantly in recent years, along with the wide- Lindau (VHL) syndrome. pNETs are also found,
spread use of cross-sectional imaging and the albeit less commonly, in patients with neurofibro-
increased rate of detection of so-called pancreatic matosis type I and tuberous sclerosis [7–9]. Tumors
incidentalomas [4–6] (Fig. 1). pNETs may manifest
at any age but are most often diagnosed between the
Department of Surgical Oncology, MD Anderson Cancer Center, The
age of 40 and 65 [1]. University of Texas, Houston, Texas, USA
pNETs may be classified as either functional Correspondence to Matthew H.G. Katz, MD, Department of Surgical
(F-pNET) or nonfunctional (NF-pNET), depending Oncology, MD Anderson Cancer Center, The University of Texas, 1400
on their secretion of various peptide hormones, Pressler Street, FCT 17.6058, Unit 1484, Houston, TX 77030-4009,
including insulin, gastrin, glucagon, and vasoactive USA. E-mail: [email protected]
intestinal peptide (VIP), resulting in various Curr Opin Gastroenterol 2019, 35:468–477
clinical syndromes (Table 1). Immunohistochemical DOI:10.1097/MOG.0000000000000571

www.co-gastroenterology.com Volume 35  Number 5  September 2019

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.

 Pancreatic neuroendocrine tumors Perri et al.

CLASSIFICATION SYSTEMS FOR TUMOR

KEY POINTS GRADE AND STAGE
 pNETs are a very heterogeneous group of neoplasms The original WHO criteria for pNETs were first
with a wide range of behavior. established in 2000 [10]. The revised 2010 classi-
fication system [11] officially adopted the label
 Surgery is the only curative treatment.
‘neuroendocrine’ and the European Neuroendo-
 Observation is a reasonable approach for patients with crine Tumor Society (ENETS) grading system. This
small, low-grade NF-pNET. represents a proliferation-based system with stratifi-
 Systemic therapies are useful to control unresectable/ cation into three tiers of proliferative activity deter-
progressive disease. mined on the basis of mitotic count and Ki-67 index.
The prognostic value of this system has been vali-
 The role of preoperative therapy for patients with dated by several large studies [12]. The current WHO
advanced pNETs is still unclear. &&
classification system was adopted in 2017 [13 ].
It raised the cut-off levels of Ki-67 for G2 tumors
to 3% and now discriminates between two subsets of
G3 tumors on the basis of their degree of histopath-
that develop in the context of an inherited syn- ologic differentiation (Table 2).
drome tend to share a more indolent course than Like other tumors, pNETs are staged using the
sporadic tumors. tumor/node/metastasis (TNM) classification.
In general, NF-pNETs have a wide range of clini- Although the WHO classification grades tumors
cal behavior. They may manifest as noninfiltrative, on the basis of their anticipated biological behavior,
slow-growing tumors, locally invasive masses, or the TNM classification estimates prognosis on the
even swiftly metastasizing cancers. The biologic basis of tumors’ anatomic extent. Past TNM staging
behavior of F-pNETs and NF-pNETs is defined by systems for pNETs included a variant proposed by
their grade and stage. the 7th edition of American Joint Committee on

FIGURE 1. Time trends of the overall pancreatic neuroendocrine tumors caseload from 1990 to 2015 in a high-volume center
for pancreatic surgery (previously published) (from [36 ]).
&

0267-1379 Copyright ß 2019 Wolters Kluwer Health, Inc. All rights reserved. www.co-gastroenterology.com 469

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.

Pancreas

Table 1. Clinical features of functional pancreatic neuroendocrine tumors

Secreted Malignant Clinical Biochemical
Tumor Incidence Location hormone (%) features diagnosis

Insulinoma 40–55% Pancreas (>99%) Insulin <10 Hypoglycemic syndromes Insulin  5 mIU/la
(Whipple’s triad) Glucose 40 mg/dl
C-peptide 0.6 ng/ml
Proinsulin  20 pmol/l
(or >25% of
immunoreactive
insulin)
Gastrinoma 25–50% Duodenum (70%) Gastrin 60–90 Zollinger–Ellison syndrome Serum gastrin level
Pancreas (25%) (abdominal pain, 10 times normal
Others (5%) gastroesophageal reflux, range þ gastric
diarrhea, duodenal pH < 2
ulcers, PUD/GERD)
Glucagonoma Rare Pancreas (100%) Glucagon 50–80 Rash, glucose intolerance, Glucagon > 500 pg/ml
necrolytic migratory
erythema, weight loss
Somatostatinoma Rare Pancreas (55%) Somatostatin >70 Diabetes mellitus, Somatostatin-fasting
Duodenum- cholelithiasis, diarrhea serum level
jejunum (45%)
VIPoma (Verner– Rare Pancreas (90%) Vasoactive 40–70 WHDA VIP fasting serum level
Morrison) Other (10%) intestinal
peptide
ACTHoma Rare Pancreas (4–16% Adreno 95 Cushing’s syndrome
all ectopic Cortico
Cushing’s) Tropic
Hormone
(ACTH)
pNET-causing Rare Pancreas (100%) Serotonin, 60–90 Carcinoid syndrome Urinary 5-HIAA in a
carcinoid tachynins 24-h urine collection
syndrome

GERD, gastroesophageal reflux disease; 5-HIAA, 5-hydroxyindoleacetic acid; PUD, peptic ulcer disease; VIP, vasoactive intestinal peptide; WHDA, watery
diarrhea, hypokalemia, achlorhydria.
a
Should be performed at the time of hypoglycemia during prolonged fasting (up to 72 h).

Cancer (AJCC) staging manual [14] and a second

Table 2. 2017 WHO classification and grading of promoted by the ENETS [12,15]. The current 8th
PanNENs edition of the AJCC staging manual is routinely used
Classification/ Ki-67 proliferation Mitotic for anatomic staging in 2019.
Grade indexa (%) indexa As shown in a recent large population-based
Well-differentiated PanNENs: PanNETs study, tumor grade and stage are crucial parameters
PanNET G1 <3 <2 for stratifying patients with pNET by prognosis. The
PanNET G2 3–20 2–20 5-year survival rate ranges from 55% for patients
PanNET G3 >20 >20 with localized, resected pNETs to 15% for patients
Poorly differentiated PanNENs: PanNECs with nonresectable pNETs [14]. Median overall
PanNEC (G3) >20 >20 survival durations range from around 12 years for
Small-cell type patients with G1-grade pNETs to 10 months for
Large-cell type &
patients with G3 pNETs [16 ].
Mixed neuroendocrine-nonneuroendocrine neoplasm

PanNECs, pancreatic neuroendocrine carcinomas; PanNENs, pancreatic

neuroendocrine neoplasms; PanNETs, pancreatic neuroendocrine tumors. CLINICAL PRESENTATION
Reproduced with permission from Lloyd RV, Osamura RY, Klöppel G, Rosai J. F-pNETs typically present with specific clinical
World Health Organization Classification of tumours of endocrine organs,
4th ed. Lyon, IARC, 2016; p 211. Copyright ß 2017 International Agency syndromes related to hypersecretion of specific
for Research on Cancer. hormones, as described in Table 1. The most
a
The Ki-67 proliferation index is based on the evaluation of at least 500 cells common F-pNETs include insulinomas, which typi-
in areas of higher nuclear labeling (so-called hotspots). The mitotic index is
based on the evaluation of mitoses in 50 high-power fields (0.2 mm2 each) in cally present with Whipple’s triad (hypoglycemia
areas of higher density and is expressed as mitoses per 10 high-power fields especially after fasting, low plasma glucose mea-
(2.0 mm2). The final grade is determined on the basis of whichever index sured at the time of symptoms and relief of symp-
(Ki-67 or mitotic) places the tumor in the highest grade category. For
assessing Ki-67, casual visual estimation (eyeballing) is not recommended; toms when the glucose is raised to normal), and
manual counting using printed images is advocated. gastrinomas, which presents with peptic ulcer

470 www.co-gastroenterology.com Volume 35  Number 5  September 2019

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.

 Pancreatic neuroendocrine tumors Perri et al.

disease (Zollinger–Ellison syndrome), gastroesoph- history plays a major role in the differential
ageal reflux disease, or secretory diarrhea. Less com- diagnosis.
mon F-pNETs include VIPomas, glucagonomas, and MRI represents an alternative to CT, with the
somatostatinomas. advantages of less radiation exposure and possibly a
The vast majority of new pNET diagnoses consist higher sensitivity than CT for detecting smaller
of an NF-pNET incidentally discovered on cross- pancreatic lesions and liver metastases [20,21].
sectional imaging performed for another indication. Somatostatin receptor scintigraphy, also known as
Indeed, NF-pNETs typically remain asymptomatic an octreotide scan, remains useful for its ability
until they reach a significant tumor burden. When to detect radiographically occult metastases and
they become symptomatic, they may present with to characterize the presence or absence functional
vague symptoms, such as nonspecific abdominal expression of somatostatin receptors, which may
pain, early satiety, or weight loss, typically related also guide systemic therapy decisions.
to mass effect. Locally advanced lesions can present PET with CT 68Gallium-labeled somatostatin
with bowel obstruction or obstructive jaundice. NF- analogs (SSAs) ((68)Ga-DOTATATE) has the highest
pNETs frequently synthesize peptides, but they do sensitivity for localizing pNETs [22,23]. For these
not produce symptoms or specific syndromes (either reasons, it is routinely performed in our center for
because they produce hormones at a low enough most patients with a suspected pNET. In contrast,
level to not cause symptoms or hormones that do PET/CT with 18flourodeoxyglucose (FDG) is not
not cause symptoms). typically useful in the diagnosis of NF-pNETs. It
When metastases occur, they are typically mul- may remain useful as a prognostic tool, as well-
tifocal and located in the liver, though other meta- differentiated tumors are generally positive on
static sites including lungs, bone, peritoneum, (68)Ga-DOTATATE scan and negative on FDG scan,
adrenal, brain, and spleen are also observed [17]. whereas the opposite is true for poorly differenti-
ated, grade 3 tumors [24].
Endoscopic ultrasonography (EUS) provides
DIAGNOSIS AND STAGING high-resolution imaging of the pancreas with high
As mentioned above, NF-pNETs are often discovered sensitivity [25]. Although limited by the require-
incidentally on cross-sectional imaging, and, in gen- ment of a highly skilled endoscopist, it can detect
eral, cross-sectional imaging is indicated in all lesions as small as 2–3 mm in diameter. It is there-
patients in whom a pNET is suspected. fore useful in patients who present with a hormonal
syndrome suggestive of a pNET but who lack evi-
dence of disease on conventional imaging. At the
Localization and staging same time, EUS-guided fine needle aspiration (FNA)
Diagnostic imaging for suspected pNETs should aspiration biopsy can secure a nonoperative histo-
typically begin with a multiphase abdominal com- logic diagnosis and quantify tumor grade. The eval-
puted tomography (CT) scan with intravenous con- uation of the tumor grade on EUS–FNA material
trast using a pancreatic protocol given its good should be included in the workup of each patient
sensitivity, specificity, and availability. pNETs with an NF-pNET, as proposed by the recently pub-
&&
characteristically appear as well-circumscribed lished ENETS consensus guidelines update [26 ].
lesions that hyperenhance with contrast in the arte-
rial phase that washes out in the delayed venous
phase. Hypoenhancement in the arterial phase and Biochemical evaluation
the presence of calcifications are associated with The assay of hormones or peptides, including insu-
more aggressive tumors and worse prognosis [18,19]. lin, glucagon, gastric, VIP, and somatostatin, is
On CT images, the differential diagnosis of essential for the diagnosis when signs or symptoms
a hyperenhancing pancreatic mass includes solid of an F-pNET are present. For a suspected insulin-
pseudopapillary tumors, pancreatic renal cell carci- oma, the biochemical assessment of insulin,
noma metastases, and intrapancreatic splenule glucose, C-peptide, and proinsulin should be per-
(if in the pancreatic tail), because of the solid hyper- formed at the time of hypoglycemia during a pro-
enhancing radiographic appearance of all these enti- longed, monitored fast. When gastrinoma is
ties. Solid pseudopapillary tumors may be suspected, fasting serum gastrin levels should be
distinguished from pNETs because of their central evaluated. A serum gastrin level that is 10 times
cystic component, owing to hemorrhagic degenera- greater than the upper limit of the normal range
tion, causing the enhancing solid areas to be along with a gastric pH less than 2 is diagnostic of
typically noted only peripherally. For renal cell car- gastrinoma. The characteristic biochemical findings
cinoma metastases, the patient’s past oncological of the most common F-pNETs are listed in Table 1.

0267-1379 Copyright ß 2019 Wolters Kluwer Health, Inc. All rights reserved. www.co-gastroenterology.com 471

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.

Pancreas

Chromogranin A (CgA) and neuron-specific potential role of observation, especially for tumors
enolase (NSE) are commonly used markers for func- anticipated to have an indolent natural history.
tional and nonfunctional pNETs. CgA, a protein
that is stored in the secretory granules of different
NETs, has been correlated with tumor burden and Localized disease
metastasis of well-differentiated NETs rather than The evidence that small (<2 cm) pNETs often
&&
poorly differentiated NETs and has been used as a behave in an indolent manner [34 ] has altered
prognostic marker for both progression-free and the historic practice of resecting all pNETs.
overall survival and a clinical tool for monitoring Although long-term follow-up data are needed to
therapeutic response [27,28]. Unfortunately, CgA is guarantee the safety of a conservative, nonsurgical
sensitive but not specific [29,30], and its level may approach, ENETS guidelines suggest that observa-
vary day by day and is influenced by food intake and tion can be considered for NF-pNETs less than 2 cm,
other medical conditions, like the use of proton following a careful analysis of the risks and benefits
&&
pump inhibitors, or renal and hepatic insufficiency. of surgery in each patient [26 ]. The potential role
NSE is an enzyme released after neuronal damage of surgery should take tumor grade, functional sta-
caused by different conditions but is also stored in tus, and the presence or absence of symptoms into
&&
the cytoplasm of NET cells. Its sensitivity and speci- account [26 ,35]. In general, in the absence of
ficity, in two series totaling more than 200 patients symptoms, radical resections are reserved for
with gastroentero-pancreatic NETs, were only 39– patients with a favorable performance status and
43 and 65–73%, respectively, for differentiating either a high-grade (G2þ) or a larger, low-grade
&
NET from nonendocrine tumors [31,32]. At our tumor [36 ]. At our center, we typically advise obser-
center, clinical decisions are rarely made on the vation for patients with G1 NF-pNETs less than 1 cm,
basis of either of these two markers and we find typically advise resection for patients with G1 NF-
them of limited utility. pNETs more than 2 cm, and carefully balance possi-
ble risks and benefits of surgery in patients with G1
tumors between 1 and 2 cm. Surgical resection is
SURGICAL MANAGEMENT generally recommended for large or G2 pNETs, or
In general, surgery is the only curative treatment for symptomatic (including functional) tumors (Fig. 2).
pNET, and it therefore represents the treatment of The role of surgery for patients with G3, well-differ-
choice for any patient with localized pNET [33]. Nev- entiated NETs, NEC is controversial [37,38]. Cura-
ertheless, the improvement of cross-sectional imag- tive surgery may be attempted in localized disease,
ing and its extensive use in the general population although retrospective series indicates that it is
have significantly increased the detection of small, rarely curative as a sole therapeutic modality given
asymptomatic NF-pNETs, raising interest in the its high relapse rate following radical surgery. At our

FIGURE 2. MD Anderson Institutional Algorithm for treating G1–G2 NF-pNETs.

472 www.co-gastroenterology.com Volume 35  Number 5  September 2019

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.

 Pancreatic neuroendocrine tumors Perri et al.

center, we often offer patients with G3 pNETs sys- reserve nonanatomic operations for patients with
temic therapy prior to consideration for resection. insulinomas and small G1 NF-pNETs, where a stan-
Small, low-grade pNETs and insulinomas may dard lymphadenectomy may not be required.
often be enucleated from the adjacent parenchyma Several reports have stressed the prognostic
safely as long as they are distant from the pancreatic importance of lymph node involvement in resected
duct and the integrity of this structure can be guaran- pNETs, confirming that the finding of positive
teed. Endoscopic placement of a pancreatic stent to lymph nodes was associated with a shorter time
protect the main pancreatic duct can be considered interval to metastatic liver recurrence, shorter
preoperatively. Although enucleation and nonana- median disease free-survival, and shorter survival
& &
tomic resections (e.g., central pancreatectomy) have in long-term follow-up [36 ,40,41,42 ]. A positive
the advantage of a decreased long-term endocrine/ nodal status can be observed in the case of G1 and
exocrine impairment when compared with standard G2 tumors, with a range in the literature between 17
&
resections, they are associated with a relatively high and 53% [36 ]. However, as recently observed by
&
rate of pancreatic fistulae and postoperative acute Marchegiani et al. [42 ] nonfunctioning G1 pNETs
pancreatitis [39]. Furthermore, a lymphadenectomy 20 mm or less without nodal metastasis or vascular
is not typically performed concurrently with these invasion have a negligible risk of developing recur-
limited operations. Therefore, we typically favor for- rence and may be considered cured by surgery
mal pancreatectomy with lymphadenectomy and (Fig. 3). Other commonly recognized prognostic

FIGURE 3. Kaplan–Maier curves for predicted cumulative disease-free survival comparing: (a) pancreatic neuroendocrine
tumors (pNET) less than 20 mm (1) versus pNET more than 21 mm (2); (b) N0 (1) and N1 (2); (c) G1 (1), G2 (2), and G3 (3);
(d) absence (1) and presence of vascular infiltration (2) (previously published) (from [42 ]).
&

0267-1379 Copyright ß 2019 Wolters Kluwer Health, Inc. All rights reserved. www.co-gastroenterology.com 473

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.

Pancreas

factors include tumor size, presence of metastases, survival of patients treated with hepatic resection
the status of the surgical margins, and microscopic may be as long as 60% [54].
vascular involvement. At the time of presentation, around 80% of
In selected patients, aggressive, extended resec- patients with metastatic disease have unresectable
tion of locally advanced tumors including en bloc liver metastases. For most other malignancies, there
removal of adjacent vascular structures and viscera, is little rationale to resect the primary site when
and even simultaneous surgical resection of the widespread, unresectable metastases are present.
primary tumor and synchronous liver metastases However, because a prolonged life expectancy
have been described to be associated with acceptable may be associated with slow-growing pNETs, resec-
morbidity, low mortality, and survival benefit in tion of the primary tumor for local control may be
several studies [43–47]. However, such resections beneficial if the primary site is causing symptoms or
should be offered only to well-selected patients in to avoid local complications.
high-volume, experienced hepatopancreatobiliary When patients are not a candidate for liver
centers following standard oncologic principles resection, alternative methods such as radiofre-
and a multidisciplinary approach. quency ablation, hepatic artery embolization, or
Enucleation and distal pancreatectomy (with or radioembolization can be considered as an alterna-
without splenic preservation) may be performed tive to systemic therapy to improve local control
using a laparoscopic or robotic fashion, with some and palliate symptoms [55].
possible advantages such as reduced length of hos- Liver transplantation may represent an option
pitalization [48,49]. Robotic pancreaticoduodenec- for well-selected patients with unresectable liver
tomy appears to provide similar perioperative metastases and a low proliferation rate, who are
and oncologic outcomes in selected patients when not responsive to medical therapy and are free from
performed at experienced, high-volume centers extrahepatic metastases. However, the experience is
[50,51]. limited and liver transplantation for pNETs metas-
tases has been associated with worse overall out-
comes and early recurrence, possibly because of
Nonsporadic pancreatic neuroendocrine postoperative immunosuppressive treatment and/
tumors or undiagnosed extrahepatic metastases [56].
In the context of VHL patients, given their better
long-term outcome, lesions 15 mm or less can be
safely observed. SYSTEMIC THERAPIES
In MEN-1 patients with pNETs less than 2 or
3 cm, routine surgical resection is not recom- Somatostatin analogs
mended, whereas in MEN-1 patients with pNETs SSAs are used for syndrome control in nonresectable
more than 2 cm surgical excision may improve sur- F-pNETs (such as in carcinoid syndrome, VIPoma,
vival, preventing or delaying distant metastases and somatostatinoma). Given their antiproliferative
&&
[26 ]. Parenchymal-preserving operations are gen- effects, they are also an established first-line therapy
erally recommended, whereas formal pancreatec- to control the growth of nonresectable, well-
&&
tomy is reserved for specific selected cases [26 ]. differentiated NF-pNETs, as many of them express
Typically in our center, we offer resection only for somatostatin receptors. Two placebo-controlled tri-
3 cm or larger tumors, symptomatic tumors, and/or als (PROMID and CLARINET) have found signifi-
rapidly growing tumors. We always attempt to cantly prolonged progression-free survival with
perform parenchymal sparing operations when long-acting release octreotide and long-acting lan-
possible, even in the setting of multifocal disease. reotide (14 versus 6 months and 32 versus 18
months, respectively) [57,58].

Liver metastases
According to ENETS guidelines, an aggressive Targeted therapies
approach toward liver metastases is recommended pNETs have been found to be responsive to the
whenever a radical resection (>90% of tumor targeted agents everolimus (an oral mTOR inhibitor)
removal) is possible, in the absence of extraabdomi- and sunitinib (an oral tyrosine-kinase inhibitor).
nal disease, and in the presence of a low Ki-67 (G1– Therefore, based on the result of two different pla-
G2 at cytology) and somatostatin receptors (to cebo-controlled trials attesting their efficacy [59,60],
deliver radiolabeled therapies after cytoreductive they are recommended for progressive G1–G2 pNET
surgery). Even in selected cases, the rate of tumor as a second line after the failure of SSA or systemic
recurrence is high, up to 76% [52,53]. But, the 5-year chemotherapy, irrespective of Ki-67 or tumor

474 www.co-gastroenterology.com Volume 35  Number 5  September 2019

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.

 Pancreatic neuroendocrine tumors Perri et al.

burden. The median progression-free survival was and streptozocin (STZ), doxorubicin, and 5-fluoro-
around 11 months with either of the drugs, whereas uracil (FAS) [62–65].
tumor remission was superior for sunitinib versus A cooperative group trial (ECOG 2211) recently
placebo. Targeted agents should be used with the randomized 144 patients with progressive pNETs to
intent of controlling disease, not to downsize or received temozolomide monotherapy versus temo-
downstage it in anticipation for surgical resection. zolomide and capecitabine, with progression-free
survival (PFS) as the primary endpoint [66]. The
temozolomide and capecitabine regimen was asso-
Peptide receptor radionuclide therapy ciated with one of the longest PFS durations
Peptide receptor radionuclide therapy (PRRT) is a reported for pNET-directed therapy (median PFS
tumor-targeted strategy that uses radiation to 22.7 months for temozolomide and capecitabine
induce tumor cell death by coupling radionuclides versus 14.4 months). Our institution has previously
to somatostatin analogs. Typical radionuclides shown how a combination regimen FAS (5-florour-
include 90yttrium and 177lutetium. PRRT represents acil, STZ and doxorubicin) is associated with a
a therapeutic option as a second or third line in response rate as high as 55% among patients with
progressive pNETs with somatostatin receptor posi- primarily metastatic disease [67,68]. Based on these
tivity and homogenous expression [61]. To date, no data, temozolomide and capecitabine can be con-
phase 3 trial of PRRT has been conducted in patients sidered a standard of care regimen in advanced, well-
with NET with a pancreatic primary site, and the differentiated pNETs, although intravenous strepto-
optimal sequencing with targeted drugs and/or che- zocin-based regimens remain a valid option. In G3
motherapy needs to be defined in pNETs. pNECs, cisplatin-based chemotherapy is the stan-
dard first-line therapy, followed by second-line
folinic acid, fluorouracil and oxaliplatin (FOLFOX)
Chemotherapy or folinic acid, fluorouracil and irinotecan [69].
Systemic chemotherapy is currently indicated for
patients’ locally advanced or progressive G1-G2
pNETs, next to SSA and targeted drugs therapy, or Preoperative therapy
as a first-line therapy in case of G3 pNETs and small Based on its cytotoxic activity, systemic chemother-
or large-cell pancreatic neuroendocrine carcinomas apy has recently been explored as preoperative ther-
(pNECs). The most widely used regimens include apy for patients with locoregionally advanced or
combinations of temozolomide with capecitabine, borderline-resectable pNET, alone or in association

FIGURE 4. Percentage change in size of target lesion(s) for each patient treated with FAS (n ¼ 29). Patients that underwent
noncurative pancreatectomy for palliation years after therapy; # patients that developed pulmonary metastases while on
chemotherapy (previously published) (from [74]).

0267-1379 Copyright ß 2019 Wolters Kluwer Health, Inc. All rights reserved. www.co-gastroenterology.com 475

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.

Pancreas

5. Kimura W, Kuroda A, Morioka Y, et al. Clinical pathology of endocrine

with PRRT [70–72]. In a recent study, we hypothe- tumors of the pancreas: analysis of autopsy cases. Dig Dis Sci 1991;
sized that FAS may be used to reduce the size or 36:933–942.
6. Yao JC, Hassan M, Phan A, et al. One hundred years after ‘carcinoid’:
anatomic extent of locoregionally advanced pNET epidemiology of and prognostic factors for neuroendocrine tumors in
prior to anticipated pancreatectomy. However, 35,825 cases in the United States. J Clin Oncol 2008; 26:3063–3072.
7. Krampitz GW, Norton JA. Pancreatic neuroendocrine tumors. Curr Probl Surg
using objective measures of radiographic response 2013; 50:509–54520.
(Modified Response Evaluation Criteria in Solid 8. Jensen RT, Berna MJ, Bingham DB, et al. Inherited pancreatic endocrine
tumor syndromes: advances in molecular pathogenesis, diagnosis, manage-
Tumors, RECIST version 1.1) [73], we showed that ment, and controversies. Cancer 2008; 113:1807–1843.
a clinically significant change in either the size of 9. Mao C, Shah A, Hanson DJ, et al. Von Recklinghausen’s disease associated
with duodenal somatostatinoma: contrast of duodenal versus pancreatic
the primary tumor or its relationship with adjacent somatostatinomas. J Surg Oncol 1995; 59:67–73.
mesenteric vasculature is uncommon in response to 10. Solcia E, Kloppel G, Alhman H, et al. Histological typing of endocrine tumors:
WHO international histological classification of tumors. 2nd ed. Berlin:
FAS therapy (Fig. 4) [74]. The role of preoperative Springer; 2000.
therapy for patients with advanced pNETs 11. Rindi G, Arnold R, Bosman F, et al. Nomenclature and classification of
neuroendocrine neoplasms of the digestive system. In: WHO classification
remains experimental. of tumours of the digestive system. 4th ed. Lyon: IARC Press; 2010.
12. Rindi G, Klöppel G, Alhman H, et al. TNM staging of foregut (neuro)endocrine
tumors: a consensus proposal including a grading system. Virchows Arch
CONCLUSION 2006; 449:395–401.
13. Klöppel G, Klimstra DS, Hruban RH, et al. Pancreatic neuroendocrine tumors:
pNETs are a very heterogeneous group of neoplasms && update on the new World Health Organization classification. AJSP Rev Rep
that have increased in incidence in recent years. 2017; 22:233–239.
The currently adopted WHO classification.
They are often slow growing, but they harbor a wide 14. Sobin LH, Gospodarowicz MK, Wittekind C. TNM classification of malignant
range of behavior and may present at an incurable tumours. 7th ed. Oxford: Wiley-Blackwell; 2009.
15. Rindi G, Falconi M, Klersy C, et al. TNM staging of neoplasms of the endocrine
stage. Therapeutic management is mainly guided by pancreas: results from a large international cohort study. J Natl Cancer Inst
symptoms, tumor grade, and tumor stage. Surgery is 2012; 104:764–777.
16. Dasari A, Schen C, Halperin D, et al. Trends in the incidence, prevalence, and
the only curative treatment, whereas systemic ther- & survival outcomes in patients with neuroendocrine tumors in the United
apies can only control disease progression. Observa- States. JAMA Oncol 2017; 3:1335–1342.
The importance of tumor grading for NET prognosis.
tion may represent a reasonable approach for 17. Zerbi A, Falconi M, Rindi G, et al. Clinicopathological features of pancreatic
patients with small, low-grade NF-pNET. In the endocrine tumors: a prospective multicenter study in Italy of 297 sporadic
cases. Am J Gastroenterol 2010; 105:1421–1429.
presence of unresectable progressive disease, 18. Worhunsky DJ, Krampitz GW, Poullos PD, et al. Pancreatic neuroendocrine
somatostatin analogs, targeted therapies such as tumours: hypoenhancement on arterial phase computed tomography predicts
biological aggressiveness. HPB (Oxford) 2014; 16:304–311.
everolimus, PRRT, and systemic chemotherapy are 19. Poultsides GA, Huang LC, Chen Y, et al. Pancreatic neuroendocrine tumors:
all useful. The role of preoperative therapy for radiographic calcifications correlate with grade and metastasis. Ann Surg
Oncol 2012; 19:2295–2303.
patients with advanced pNETs is still unclear. 20. Sundin A, Vullierme MP, Kaltsas G, et al. ENETS consensus guidelines for the
standards of care in neuroendocrine tumors: radiological examinations.
Neuroendocrinology 2009; 90:167–183.
Acknowledgements 21. Dromain C, de Baere T, Lumbroso J, et al. Detection of liver metastases from
endocrine tumors: a prospective comparison of somatostatin receptor scinti-
None. graphy, computed tomography, and magnetic resonance imaging. J Clin
Oncol 2005; 23:70–78.
22. Hofman MS, Kong G, Neels OC, et al. High management impact of Ga-68
Financial support and sponsorship DOTATATE (GaTate) PET/CT for imaging neuroendocrine and other
None. somatostatin expressing tumours. J Med Imaging Radiat Oncol 2012;
56:40–47.
23. Gabriel M, Decristoforo C, Kendler D, et al. 68Ga-DOTA-Tyr3-octreotide PET
Conflicts of interest in neuroendocrine tumors: comparison with somatostatin receptor scintigra-
phy and CT. J Nucl Med 2007; 48:508–518.
There are no conflicts of interest. 24. Squires MH, Volkan Adsay N, Schuster DM, et al. Octreoscan versus FDG-
PET for neuroendocrine tumor staging: a biological approach. Ann Surg
Oncol 2015; 22:2295–2301.
25. James PD, Tsolakis AV, Zhang M, et al. Incremental benefit of preoperative
REFERENCES AND RECOMMENDED EUS for the detection of pancreatic neuroendocrine tumors: a meta-analysis. J
Gastrointest Endosc 2015; 81:848.e1–856.e1.
READING 26. Falconi M, Eriksson B, Kaltsas G, et al. ENETS consensus guidelines update
Papers of particular interest, published within the annual period of review, have && for the management of patients with functional pancreatic neuroendocrine
been highlighted as: tumors and non-functional pancreatic neuroendocrine tumors. Vienna Con-
& of special interest sensus Conference participants. Neuroendocrinology 2016; 103:153–171.
&& of outstanding interest
Updated version of European Consensus Guidelines.
27. Han X, Zhang C, Tang M, et al. The value of serum Chromogranin A as a
1. Lawrence B, Gustafsson BI, Chan A, et al. The epidemiology of gastroenter- predictor of tumor burden, therapeutic response, and nomogram-based
opancreatic neuroendocrine tumors. Endocrinol Metab Clin North Am 2011; survival in well moderate nonfunctional pancreatic neuroendocrine tumors
40:1–18; vii. with liver metastases. Eur J Gastroenterol Hepatol 2015; 27:527–535.
2. Bilimoria KY, Bentrem DJ, Merkow RP, et al. Application of the pancreatic 28. Massironi S, Rossi RE, Casazza G, et al. Chromogranin A in diagnosing and
adenocarcinoma staging system to pancreatic neuroendocrine tumors. J Am monitoring patients with gastroenteropancreatic neuroendocrine neoplasms:
Coll Surg 2007; 205:558–563. a large series from a single institution. Neuroendocrinology 2014; 100:
3. Goh BK, Chow PK, Tan YM, et al. Validation of five contemporary prognos- 240–249.
tication systems for primary pancreatic endocrine neoplasms: results from a 29. Pulvirenti A, Rao D, Mcintyre CA, et al. Limited role of Chromogranin A as
single institution experience with 61 surgically treated cases. ANZ J Surg clinical biomarker for pancreatic neuroendocrine tumors. HPB (Oxford) 2019;
2011; 81:79–85. 21:612–618.
4. Hallet J, Law CH, Cukier M, et al. Exploring the rising incidence of neuro- 30. Yang X, Yang Y, Li Z, et al. Diagnostic value of circulating Chromogranin A for
endocrine tumors: a population-based analysis of epidemiology, metastatic neuroendocrine tumors: a systematic review and meta-analysis. PLoS One
presentation, and outcomes. Cancer 2015; 121:589–597. 2015; 10:e0124884.

476 www.co-gastroenterology.com Volume 35  Number 5  September 2019

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.

 Pancreatic neuroendocrine tumors Perri et al.

31. Baudin E, Gigliotti A, Ducreux M, et al. Neuron-specific enolase and Chro- 53. House MG, Cameron JL, Lillemoe KD, et al. Differences in survival for patients
mogranin A as markers of neuroendocrine tumours. Br J Cancer 1998; with resectable versus unresectable metastases from pancreatic islet cell
78:1102–1110. cancer. J Gastrointest Surg 2006; 10:138–145.
32. Nobels FR, Kwekkeboom DJ, Coopmans W, et al. Chromogranin A as serum 54. Saxena A, Chua TC, Perera M, et al. Surgical resection of hepatic metastases
marker for neuroendocrine neoplasia: comparison with neuron-specific en- from neuroendocrine neoplasms: a systematic review. Surg Oncol 2012;
olase and the alpha-subunit of glycoprotein hormones. J Clin Endocrinol 21:e131–e141.
Metab 1997; 82:2622–2628. 55. Lesurtel M, Nagorney DM, Mazzaferro V, et al. When should a liver resection
33. Hill JS, McPhee JT, McDade TP, et al. Pancreatic neuroendocrine tumors: the be performed in patients with liver metastases from neuroendocrine tumours?
impact of surgical resection on survival. Cancer 2009; 115:741–751. A systematic review with practice recommendations. HPB (Oxford) 2015;
34. Bettini R, Partelli S, Boninsegna L, et al. Tumor size correlates with malignancy 17:17–22.
&& in nonfunctioning pancreatic endocrine tumor. Surgery 2011; 150:75–82. 56. Le Treut YP, Grégoire E, Belghiti J, et al. Predictors of long-term survival after
The importance of tumor size for surgical decision-making. liver transplantation for metastatic endocrine tumors: an 85-case French
35. Partelli S, Bartsch DK, Capdevila J, et al. ENETS Consensus Guidelines for multicentric report. Am J Transplant 2008; 8:1205–1213.
Standard of Care in neuroendocrine tumours: surgery for small intestinal and 57. Rinke A, M€ uller HH, Schade-Brittinger C, et al. Placebo-controlled, double-
pancreatic neuroendocrine tumours. Neuroendocrinology 2017; 105: blind, prospective, randomized study on the effect of octreotide LAR in the
255–265. control of tumor growth in patients with metastatic neuroendocrine midgut
36. Landoni L, Marchegiani G, Pollini T, et al. The evolution of surgical strategies tumors: a report from the PROMID Study Group. J Clin Oncol 2009;
& for pancreatic neuroendocrine tumors (Pan-NENs): time-trend and outcome 27:4656–4663.
analysis from 587 consecutive resections at a high-volume institution. Ann 58. Caplin ME, Pavel M, Ćwikła JB, et al. Antitumour effects of lanreotide for
Surg 2017; 269:725–732. pancreatic and intestinal neuroendocrine tumours: the CLARINET open-label
Past and current trends in pNET surgical management. extension study. Endocr Relat Cancer 2016; 23:191–199.
37. Deng BY, Liu F, Yin SN, et al. Clinical outcome and long-term survival of 150 59. Yao JC, Shah MH, Ito T, et al. Everolimus for advanced pancreatic neuroen-
consecutive patients with pancreatic neuroendocrine tumors: a comprehen- docrine tumors. N Engl J Med 2011; 364:514–523.
sive analysis by the World Health Organization 2010 grading classification. 60. Raymond E, Dahan L, Raoul JL, et al. Sunitinib malate for the treatment of
Clin Res Hepatol Gastroenterol 2018; 42:261–268. pancreatic neuroendocrine tumors. N Engl J Med 2011; 364:501–513.
38. Brenner B, et al. Small cell carcinomas of the gastrointestinal tract: clinico- 61. Hicks RJ, Kwekkeboom DJ, Krenning E, et al. ENETS Consensus Guidelines
pathological features and treatment approach. Semin Oncol 2007; for the standards of care in neuroendocrine neoplasia: peptide receptor
34:43–50. radionuclide therapy with radiolabeled somatostatin analogues. Neuroendo-
39. Paiella S, De Pastena M, Faustini F, et al. Central pancreatectomy for benign crinology 2017; 105:295–309.
or low-grade malignant pancreatic lesions—a single-center retrospective 62. Dilz LM, Denecke T, Steffen IG, et al. Streptozocin/5-fluorouracil chemother-
analysis of 116 cases. Eur J Surg Oncol 2019; 45:1125. apy is associated with durable response in patients with advanced pancreatic
40. Hashim YM, Trinkaus KM, Linehan DC, et al. Regional lymphadenectomy is neuroendocrine tumours. Eur J Cancer 2015; 51:1253–1262.
indicated in the surgical treatment of pancreatic neuroendocrine tumors 63. Strosberg JR, Fine RL, Choi J, et al. First-line chemotherapy with capecitabine
(PNETs). Ann Surg 2014; 259:197–203. and temozolomide in patients with metastatic pancreatic endocrine carcino-
41. Partelli S, Javed AA, Andreasi V, et al. The number of positive nodes accurately mas. Cancer 2011; 117:268–275.
predicts recurrence after pancreaticoduodenectomy for nonfunctioning neu- 64. Krug S, Boch M, Dainel H, et al. Streptozocin-based chemotherapy in patients
roendocrine neoplasms. Eur J Surg Oncol 2018; 44:778–783. with advanced neuroendocrine neoplasms: predictive and prognostic mar-
42. Marchegiani G, Landoni L, Andrianello S, et al. Patterns of recurrence after kers for treatment stratification. PLoS One 2015; 10:e0143822.
& resection for pancreatic neuroendocrine tumors: who, when, and where? 65. Fjallskog ML, Janson ET, Falkmer UG, et al. Treatment with combined
Neuroendocrinology 2019; 108:161–171. streptozotocin and liposomal doxorubicin in metastatic endocrine pancreatic
Current predictors of pNET recurrence after surgery. tumors. Neuroendocrinology 2008; 88:53–58.
43. Birnbaum DJ, Turrini O, Vigano L, et al. Surgical management of advanced 66. Kunz PL, Catalano PJ, Nimeiri H, et al. A randomized study of temozolomide or
pancreatic neuroendocrine tumors: short-term and long-term results from an temozolomide and capecitabine in patients with advanced pancreatic neu-
international multiinstitutional study. Ann Surg Oncol 2014; 22:1000–1007. roendocrine tumors: a trial of the ECOG ACRIN Cancer Research Group
44. Norton JA, Harris EJ, Chen Y, et al. Pancreatic endocrine tumors with major (E2211). J Clin Oncol 2018; 36(15_suppl):4004–14004.
vascular abutment, involvement, or encasement and indication for resection. 67. Rivera E, Ajani JA. Doxorubicin, streptozocin, and 5-fluorouracil chemotherapy
Arch Surg 2011; 146:724–732. for patients with metastatic islet-cell carcinoma. Am J Clin Oncol 1998;
45. Abu Hilal M, McPhail MJ, Zeidan BA, et al. Aggressive multivisceral pancreatic 21:36–38.
resections for locally advanced neuroendocrine tumours. Is it worth it? JOP 68. Kouvaraki MA, Ajani JA, Hoff P, et al. Fluorouracil, doxorubicin, and
2009; 10:276–279. streptozocin in the treatment of patients with locally advanced and
46. Almond LM, Hodson J, Ford SJ, et al. Role of palliative resection of the primary metastatic pancreatic endocrine carcinomas. J Clin Oncol 2004; 22:
tumour in advanced pancreatic and small intestinal neuroendocrine 4762–4771.
tumours: a systematic review and meta-analysis. Eur J Surg Oncol 2017; 69. Garcia-Carbonero R, Sorbye H, Baudin E, et al. ENETS consensus guidelines
43:1808–1815. for high-grade gastroenteropancreatic neuroendocrine tumors and neuroen-
47. Gaujoux S, Gonen M, Tang L, et al. Synchronous resection of primary and docrine carcinomas. Neuroendocrinology 2016; 103:186–194.
liver metastases for neuroendocrine tumors. Ann Surg Oncol 2012; 70. Basu S, Parghane RV, Ostwal V, et al. Neoadjuvant strategies for advanced
19:4270–4277. pancreatic neuroendocrine tumors: should combined chemotherapy and
48. Crippa S, Bassi C, Salvia R, et al. Enucleation of pancreatic neoplasms. Br J peptide receptor radionuclide therapy be the preferred regimen for maximizing
Surg 2007; 94:1254–1259. outcome? Nucl Med Commun 2018; 39:94–95.
49. Venkat R, Edil BH, Schulick RD, et al. Laparoscopic distal pancreatectomy is 71. Ambe CM, Nguyen P, Centeno BA, et al. Multimodality management of
associated with significantly less overall morbidity compared to the open ‘borderline resectable’ pancreatic neuroendocrine tumors: report of a sin-
technique: a systematic review and meta-analysis. Ann Surg 2012; gle-institution experience. Cancer Control 2017; 24:1073274817729076.
255:1048–1059. 72. Partelli S, Bertani E, Bartolomei M, et al. Peptide receptor radionuclide
50. Watkins AA, Kent TS, Gooding WE, et al. Multicenter outcomes of robotic therapy as neoadjuvant therapy for resectable or potentially resectable
reconstruction during the early learning curve for minimally-invasive pancrea- pancreatic neuroendocrine neoplasms. Surgery 2018; 16:142.
ticoduodenectomy. HPB (Oxford) 2018; 20:155–165. 73. Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria
51. Kendrick ML, van Hilst J, Boggi U, et al. Minimally invasive pancreatoduode- in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009;
nectomy. HPB (Oxford) 2017; 19:215–224. 45:228–247.
52. Touzios JG, Kiely JM, Pitt SC, et al. Neuroendocrine hepatic metastases: does 74. Prakash L, Bhosale P, Cloyd J, et al. Role of fluorouracil, doxorubicin, and
aggressive management improve survival? Ann Surg 2005; 241:776–783. streptozocin therapy in the preoperative treatment of localized pancreatic
(Discussion 783–785). neuroendocrine tumors. J Gastrointest Surg 2017; 21:155–163.

0267-1379 Copyright ß 2019 Wolters Kluwer Health, Inc. All rights reserved. www.co-gastroenterology.com 477

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.