Académique Documents
Professionnel Documents
Culture Documents
DOI 10.1007/s12022-013-9295-2
Abstract In the past decade, the clinico-pathologic characteristics of neuroendocrine tumors (NETs) in the pancreas
have been further elucidated. Previously termed islet cell
tumors/carcinomas or endocrine neoplasms, they are now
called pancreatic NETs (PanNETs). They occur in relatively
younger patients and may arise anywhere in the pancreas.
Some are associated with von HippelLindau, MEN1, and
other syndromes. It is now widely recognized that, with the
exception of tumorlets (minute incipient neoplasms) that occur in some syndromes like MEN1, all PanNETs are malignant, albeit low-grade, and although they have a protracted
clinical course and overall 10-year survival of 6070 %, even
low-stage and low-grade examples may recur and/or metastasize on long-term follow-up. Per recent consensus guidelines
adopted by both European and North American NET Societies (ENETS and NANETs) and WHO-2010, PanNETs are
now graded and staged separately, unlike previous classification schemes that used a combination of grade, stage, and
adjunct prognosticators in an attempt to define benign behavior or malignant categories. For staging, the ENETs
proposal may be more applicable than CAP/AJCC, which is
based on the staging of exocrine tumors. Current grading of
PanNETs is based on mitotic activity and ki-67 index. Other
promising prognosticators such as necrosis, CK19, c-kit, and
others are still under investigation. It has also been recognized
that PanNETs have a rather wide morphologic repertoire
including oncocytic, pleomorphic, ductulo-insular, sclerosing,
and lipid-rich variants. Most PanNETs are diagnosed by fine
M. D. Reid : S. Balci : B. Saka : N. V. Adsay
Department of Pathology, Emory University School of Medicine,
Atlanta, GA, USA
N. V. Adsay (*)
Department of Pathology, Emory University Hospital, 1364 Clifton
Rd NE, Room H178, Atlanta, GA 30322, USA
e-mail: nadsay@emory.edu
Introduction
Many new developments have taken place in our understanding of the biology, morphologic, and molecular characteristics
of neuroendocrine tumors of the pancreas. In 2010, the World
Health Organization (WHO) proposed a new terminology for
these tumors [1], which has had far-reaching effects on their
current diagnosis and management. As a very important and
positive development, the grading and staging parameters for
these tumors have now been separated from their name and
diagnostic sub-categories. In the ensuing article, we will
66
review these new developments, their limitations and controversies, as well as recent advances in the diagnosis, classification, and management of these tumors.
Clinical Aspects
PanNETs are relatively rare tumors, accounting for 2 % of all
pancreatic neoplasms and affecting one to two individuals per
1,000,000/year. However, with improving technology, more
and more cases come to attention as incidentalomas [2].
Tumors are most frequently seen in adults but may rarely
occur in children. Patients are typically 3060 years (with a
mean of 50 years). Men and women are equally affected but
high-grade (G3) carcinomas often occur in older males.
PanNETs more commonly arise in the pancreatic head but
can involve any part of the pancreas.
WDNETs may also arise in a background of familial syndromes including von HippelLindau (VHL) syndrome, tuberous sclerosis complex (TSC), neurofibromatosis type 1
(NF1), and multiple endocrine neoplasia type 1 (MEN1). In
Table 1 Classification system for pancreatic neuroendocrine tumors
Modified from WHO 2010
Differentiation
Grade
Mitosesa/10
|HPF
Ki-67b
proliferation
|index (%)
Well-differentiated PanNET
Grade 1
Grade 2
Grade 3
<2
2-20
>20
<3
3-20
>20
Poorly-differentiated PanNEC
WHO World Health Organization, HPF high power field, PanNET pancreatic neuroendocrine tumor, PanNEC pancreatic neuroendocrine carcinoma (large and small cell type)
a
In at least 50 HPFs
MIB1 antibody
67
Grading
In the current 2010-WHO classification system, pancreatic neuroendocrine tumors are divided into well-differentiated [grade 1
(G1) and 2 (G2)] NETs (WDNETs) and poorly differentiated
(G3) neuroendocrine carcinoma (PDNECs). This system gives
the distinct impression that PDNECs are in continuum with
68
ordinary (well-differentiated) PanNETs, which is not necessarily the case. Evolving evidence strongly suggests that G3
tumors should be regarded separately and as such will be
discussed separately later.
The grading of NETs is based on two calculations: (1)
mitotic count and (2) ki-67 labeling index (using the MIB1
antibody). G1 NETs are defined as having a ki-67 index of
<3 % and <2 mitoses/10 high power fields (HPF) or 2 mm2.
G2 NETs have a ki-67 index of 320 % or 220 mitoses/10
HPFs (Table 1) [1]. The WHO recommends that for mitotic
counts, at least 50 HPFs should be counted, and for ki-67
index, a minimum of 500 tumor cells should be counted in
tumor hot spots. For grade-discordant cases (based on differences in mitotic count and ki-67 index), the higher grade
should be used.
There are several areas of ambiguity in this grading system,
which are now being clarified. For example, in the original
guidelines, it was unclear which category (G1 versus G2) to
place ki-67 indices of 23 %, a range in which a substantial
number of cases fell. The cutoff value has now been clarified
by NANETs as <3 %, indicating that cases within this range
are actually G1.
It should be noted that in the current grading scheme,
morphology is, for the most part, unfortunately disregarded.
It has now become clear that a group of morphologically welldifferentiated NETs that has comparatively high ki-67 indices
of >20 % exists. These well-differentiated but proliferatively
high-grade tumors typically have a ki-67 index of <50 %.
Unfortunately, per 2010-WHO guidelines, these cases are
classified as high-grade (G3) neuroendocrine carcinomas,
along with full-blown PDNECs (which have a far worse
prognosis), but they behave in a clinically worse fashion than
typical G2 tumors (see Biologic Behavior) [10]. Additionally, a study by Sorbye et al. recently indicated that true
PDNECs of the gastrointestinal tract, with their aggressive
behavior and platinum sensitivity, may be better defined by ki67 indices of >50 % [11]. Therefore, it is becoming clear that
the current G3 group defined as >20 % will have to be split
into two separate categories in the future in order to distinguish the well-differentiated G3 tumors (ki-67, 2050 %)
from the poorly differentiated carcinomas (G4?; ki-67 >50 %).
At the other end of the spectrum, for the distinction of G1
from G2 tumors, some groups advocate using 5 % as the cutoff,
which may be a more accurate and reproducible cutoff. While
further studies are needed to clarify this issue, the current 2 %
cutoff remains widely used. Additionally, the current cutoff
which is widely used and endorsed by the WHO-2010 is often
referred to as the 2 % cutoff in publications; however, as
recently clarified by NANETS, it is actually applied as <3 %
and 3 % in practice and as such it is the 3 % cutoff.
It should be noted that other proposed grading systems are
also in place. Hochwald and Klimstra had proposed a two-tier
grading system for PanNETs based on necrosis and mitotic
Staging
The staging of PanNETs is now done separately from the
grade, unlike the WHO-2004 system where the two were
combined to determine the category.
There are currently several different staging protocols for
PanNETs. The one from The American Joint Commission on
Cancer (AJCC)/Union of International Cancer Control
(UICC)/College of American Pathologists (CAP) [1921]
differs from that of the European Neuroendocrine Tumor
Society (ENETS) [22] system (Table 2). The AJCC has essentially adopted the staging used for exocrine tumors, which
is problematic for pT3 tumors, which are defined in part by
peripancreatic soft tissue involvement. Because the pancreas
has very irregular lobules, no capsule, and fat placement
throughout, it is often difficult, if not impossible, to assess
peripancreatic soft tissue invasion in these tumors [2325].
Furthermore, most PanNETs protrude from the pancreas even
69
Table 2 AJCC/UICC/CAP and ENETS TNM classification systems for
pancreatic neuroendocrine tumors
T stage
AJCC/UICC
[19, 20]/CAP TNM
T1
T2
T3
T4
when they are small, especially those located in the tail. That is
probably why many studies have failed to identify a correlation between survival and T stage. In contrast, the ENETS
staging system relies predominantly on more reproducible
criteria such as tumor size rather than poorly defined and
irreproducible parameters like peripancreatic extension.
A recent study of 1,072 post-surgical PanNET patients
showed that the ENETS system was superior to the
AJCC/UICC system for stratifying risk of death and creating
risk-based treatment guidelines [26, 27]. Despite the proven
discrepancies between the two staging systems, there appears
to be no adverse effect on diagnosis and management [28]. In
our current practice, we typically provide the stage by both
schemes, separately.
70
Fig. 2 Well-differentiated pancreatic neuroendocrine tumor is seen arising in the pancreatic body as a solitary, well-circumscribed tan yellow
mass. Note the absence of hemorrhage and necrosis
71
Fig. 3 a Well-differentiated
pancreatic neuroendocrine tumor,
grade 1. The tumor is stroma poor
with an organoid arrangement of
bland cells and intervening thinwalled blood vessels
(hematoxylin & eosin stain, 400
magnification). b Tumor cells are
strongly positive for
chromogranin A (400
magnification)
72
survival is typically less than 2 years. Platinum-based therapeutic agents have shown some promise in controlling their
growth; however, their overall prognosis remains grim.
These poorly differentiated, high-grade (G3) carcinomas are
subdivided based on cell size into small- and large-cell variants.
The large-cell variant is more common, often large, and characterized by large cells with prominent nucleoli and variable
cytoplasm. Mitotic activity is brisk (often >40/10 HPFs) and
necrosis is often extensive. The small-cell variant shows small to
intermediate cells with coarse salt and pepper chromatin, high
nucleus-to-cytoplasm ratio, inconspicuous nucleoli, prominent
nuclear molding, and crush artifact (Fig. 8). Mitotic figures are
easily identifiable and there is extensive tumor necrosis.
Better delineation of PDNECs from ordinary G3 PanNETs
has been elucidated in some recent studies. Well-differentiated
but highly proliferative PanNETs typically have a ki-67 index
below 4050 % [10], whereas PDNECs typically show ki-67
>55 %, are characterized by highly aggressive behavior, and
require platinum-based therapy [11].
It should be noted here that PDNECs are very uncommon
tumors, and most cases that are diagnosed as such prove to be
acinar cell carcinomas once studied more carefully [38].
Ancillary Studies
Immunohistochemistry
PanNETs are essentially defined by the production of cytoplasmic neuroendocrine granules, which are commonly
highlighted by immunohistochemical markers chromogranin,
synaptophysin, and CD56 (Fig. 3). Ordinary well-differentiated PanNETs tend to show stronger more diffuse staining
with neuroendocrine markers than PDNECs. Synaptophysin
is highly sensitive but less specific. Chromogranin is the most
specific neuroendocrine marker but is the least sensitive of the
three. CD56 antibody (123c3) which is directed against neural
73
74
Differential Diagnosis
Differential diagnoses of PanNETs include all primary pancreatic neoplasms with diffuse, cellular sheet-like proliferation
as well as metastatic tumors.
Acinar cell carcinoma is one such primary pancreatic tumor
that is derived from exocrine acinar cells and shows acinar
differentiation as evidenced by positivity for trypsin, chymotrypsin, lipase, other pancreatic enzymes, and BCL10. Tumor
cells have abundant, granular, zymogen-rich cytoplasmic granules, coarse chromatin, and prominent, sometimes cherry red,
central nucleoli. A subset of acinar cell carcinomas is morphologically almost indistinguishable from PanNETs. In addition,
most acinar cell carcinomas focally express neuroendocrine
markers either as scattered individual cells or large zones or
even as a very-well-formed distinct component (mixed carcinomas). The presence of exuberant mitotic activity in the setting
of seemingly bland monotonous epithelial cells is more in
keeping with acinar cell carcinoma than a differentiated
PanNET, in which mitotic figures are often few and far between. Additionally, acinar cell carcinomas show prominent
central nucleoli, which are typically less prominent in
PanNETs. Most helpful though is the neuroendocrine chromatin pattern. Morphologic distinction is facilitated by immunohistochemistry. Acinar cell carcinoma is negative or only focally positive for neuroendocrine markers and shows positivity for
pancreatic enzymes, while NETs show opposite results.
Pancreatoblastomas are morphologically similar to acinar cell
carcinoma and may also be confused with PanNETs. However,
immunohistochemistry as well as the finding of classical
squamoid morules should facilitate distinction.
Morphologic distinction of mixed acinarneuroendocrine
carcinoma, an acinar cell neoplasm that shows over 30 %
neuroendocrine differentiation, can be especially challenging
both on morphology as well as immunohistochemistry as
tumor cells stain with both acinar and neuroendocrine
markers. The key to distinguishing this variant from PanNET
is the mixed neuroendocrine and acinar immunophenotype as
well as the distribution of the neuroendocrine component,
which should not be significant in these mixed tumors.
Another key differencel is SPN. Both PanNET and SPN can
show solid sheet-like pattern and nested growth (Fig. 9). Typically, the nesting is more vague in SPNs, and both patterns
blend, rather imperceptibly, with each other, creating a more
mesenchymal-like appearance. Striking overlapping of tumor
nuclei is more prominent in SPN and the presence of large
cytoplasmic vacuoles also favors SPN. On close examination,
one should note the absence of the salt and pepper chromatin
of neuroendocrine tumors and the presence of fine, open,
powdery chromatin along with longitudinal nuclear grooves,
which are distinctive findings in SPN. In addition, the presence
of PAS-positive, diastase-resistant hyaline globules is supportive of SPN, although similar globules can occasionally be seen
75
Cytology
Fine needle aspiration (FNA) has been used in the cytologic
diagnosis of primary and metastatic PanNETs for many years. It
is typically performed by endoscopic ultrasound-guided FNA,
which is successful at identifying NETs in up to 90 % of cases.
The cytologic features of these tumors are distinctive. For
ordinary, well-differentiated (G1 and 2) NETs, tumor cells are
classically singly dispersed or clustered as bland-appearing
monotonous cells with round to oval nuclei and variable cytoplasm, which may or may not have distinctive plasmacytoid
features (Fig. 10). In addition, multiple cytoplasmic vacuoles
may be seen. On Papanicolaou stain, tumor cells show the
classical salt-and-pepper chromatin distribution, which supports the diagnosis (Fig. 10). Pseudorosettes are visible not only
on cellblock but also on smears. Immunohistochemical analysis
of cellblock material often demonstrates tumor cell positivity
for keratin as well as neuroendocrine markers.
In poorly differentiated neuroendocrine carcinoma of smallcell type, the cytologic features are similar to the lung counterpart. Tumor cells are often crushed and small to intermediate in
size with irregular nuclear borders, high nucleus-to-cytoplasm
Fig. 10 Fine needle aspiration of a well-differentiated pancreatic neuroendocrine tumor showing singly dispersed, bland tumor cells with
plasmacytoid morphology, oval nuclei, and salt and pepper chromatin
(Papanicolaou stain, 400 magnification)
76
Biologic Behavior
Treatment Options
Fig. 11 Well-differentiated pancreatic neuroendocrine tumor (grade 3 by ki67 index) in contrast to the poorly differentiated neuroendocrine carcinoma
in Fig. 8. a By mitotic count (2/10 high power fields), this tumor was
classified as grade 1 per 2010 WHO grading scheme (hematoxylin & eosin
stain, 400 magnification). b The corresponding ki-67 stain yielded a proliferation index of 25 %, consistent with a grade 3 tumor (400 magnification)
77
References
1. Klimstra D, Arnold R, Capella C, Hruban R, Kloppel G, Komminoth
P, Solcia E, Rindi G (2010) Neuroendocrine neoplasms of the pancreas. In: Bosman F, Carneiro F, Hruban R, Theise N (eds) WHO
classification of tumours of the digestive system, 3rd edn. IARC,
Lyon, pp 322-326
2. Vagefi PA, Razo O, Deshpande V, McGrath DJ, Lauwers GY, Thayer
SP, Warshaw AL, Fernandez-Del Castillo C (2007) Evolving patterns
in the detection and outcomes of pancreatic neuroendocrine neoplasms:
the Massachusetts General Hospital experience from 1977 to 2005.
Arch Surg 142 (4):347-354. doi:10.1001/archsurg.142.4.347
3. Anlauf M, Schlenger R, Perren A, Bauersfeld J, Koch CA, Dralle H,
Raffel A, Knoefel WT, Weihe E, Ruszniewski P, Couvelard A,
Komminoth P, Heitz PU, Kloppel G (2006) Microadenomatosis of
the endocrine pancreas in patients with and without the multiple
endocrine neoplasia type 1 syndrome. Am J Surg Pathol 30 (5):
560-574. doi:10.1097/01.pas.0000194044.01104.25
4. Lubensky IA, Pack S, Ault D, Vortmeyer AO, Libutti SK, Choyke
PL, Walther MM, Linehan WM, Zhuang Z (1998) Multiple neuroendocrine tumors of the pancreas in von Hippel-Lindau disease
patients: histopathological and molecular genetic analysis. Am J
Pathol 153 (1):223-231. doi:10.1016/S0002-9440(10)65563-0
5. Hoang MP, Hruban RH, Albores-Saavedra J (2001) Clear cell endocrine pancreatic tumor mimicking renal cell carcinoma: a distinctive
neoplasm of von HippelLindau disease. Am J Surg Pathol 25 (5):
602-609
6. Fryer E, Serra S, Chetty R (2012) Lipid-rich (clear cell) neuroendocrine tumors of the pancreas in MEN I patients. Endocr Pathol 23
(4):243-246. doi:10.1007/s12022-012-9221-z
7. Whipple AO, Frantz VK (1935) Adenoma of islet cells with hyperinsulinism: a review. Ann Surg 101 (6):1299-1335
8. Shi C, Siegelman SS, Kawamoto S, Wolfgang CL, Schulick RD,
Maitra A, Hruban RH (2010) Pancreatic duct stenosis secondary to
small endocrine neoplasms: a manifestation of serotonin production?
Radiology 257 (1):107-114. doi:10.1148/radiol.10100046
9. Kenney B, Singh G, Salem RR, Paterno F, Robert ME, Jain D (2011)
Pseudointraductal papillary mucinous neoplasia caused by microscopic
periductal endocrine tumors of the pancreas: a report of 3 cases. Hum
Pathol 42 (7):1034-1041. doi:10.1016/j.humpath.2010.09.018
10. Basturk O, Yang Z, Tang LH, Hruban R, McCall C, Adsay V,
Krasinskas A, Jang KT, Bellizzi A, Shi C, Klimstra DS (2013)
Increased (> 20 %) Ki67 proliferation index in morphologically well
differentiated pancreatic neuroendocrine tumors (PanNETs) correlates with decreased overall survival (abstract). Modern Pathology
26:423A-423A
11. Sorbye H, Welin S, Langer SW, Vestermark LW, Holt N, Osterlund P,
Dueland S, Hofsli E, Guren MG, Ohrling K, Birkemeyer E, ThiisEvensen E, Biagini M, Gronbaek H, Soveri LM, Olsen IH, Federspiel
B, Assmus J, Janson ET, Knigge U (2013) Predictive and prognostic
factors for treatment and survival in 305 patients with advanced
gastrointestinal neuroendocrine carcinoma (WHO G3): the
NORDIC NEC study. Annals of Oncology: Official Journal of the
European Society for Medical Oncology/ESMO 24 (1):152-160.
doi:10.1093/annonc/mds276
12. Hochwald SN, Zee S, Conlon KC, Colleoni R, Louie O, Brennan
MF, Klimstra DS (2002) Prognostic factors in pancreatic endocrine
neoplasms: an analysis of 136 cases with a proposal for low-grade
and intermediate-grade groups. J Clin Oncol 20 (11):2633-2642
13. Deshpande V, Fernandez-del Castillo C, Muzikansky A, Deshpande
A, Zukerberg L, Warshaw AL, Lauwers GY (2004) Cytokeratin 19 is
a powerful predictor of survival in pancreatic endocrine tumors. Am J
Surg Pathol 28 (9):1145-1153
14. Jain R, Fischer S, Serra S, Chetty R (2010) The use of Cytokeratin 19
(CK19) immunohistochemistry in lesions of the pancreas,
78
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
79
54. Movahedi-Lankarani S, Hruban RH, Westra WH, Klimstra DS
(2002) Primitive neuroectodermal tumors of the pancreas: a report
of seven cases of a rare neoplasm. Am J Surg Pathol 26 (8):10401047
55. Bachmann J, Kleeff J, Bergmann F, Shrikhande SV, Hartschuh W,
Buchler MW, Friess H (2005) Pancreatic metastasis of Merkel cell
carcinoma and concomitant insulinoma: case report and literature
review. World J Surg Oncol 3:58. doi:10.1186/1477-7819-3-58
56. Fung AD, Cohen C, Kavuri S, Lawson D, Gao X, Reid MD (2013)
Phosphohistone h3 and ki-67 labeling indices in cytologic specimens
from well-differentiated neuroendocrine tumors of the gastrointestinal tract and pancreas: a comparative analysis using automated image
cytometry. Acta Cytol 57 (5):501-508. doi:10.1159/000351475
57. Larghi A, Capurso G, Carnuccio A, Ricci R, Alfieri S, Galasso D,
Lugli F, Bianchi A, Panzuto F, De Marinis L, Falconi M, Delle Fave
G, Doglietto GB, Costamagna G, Rindi G (2012) Ki-67 grading of
nonfunctioning pancreatic neuroendocrine tumors on histologic samples obtained by EUS-guided fine-needle tissue acquisition: a prospective study. Gastrointest Endosc 76 (3):570-577. doi:10.1016/j.
gie.2012.04.477
58. Yang Z, Tang LH, Klimstra DS (2011) Effect of tumor heterogeneity
on the assessment of Ki67 labeling index in well-differentiated neuroendocrine tumors metastatic to the liver: implications for prognostic stratification. Am J Surg Pathol 35 (6):853-860. doi:10.1097/PAS.
0b013e31821a0696
59. Velayoudom-Cephise FL, Duvillard P, Foucan L, Hadoux, J,
Chougnet CN, Leboulleux, S, Malka D, Guigay J, Goere D,
Debaere T, Caramella C, Schlumberger M, Planchard D, Elias D,
Ducreux M, Scoazec JY, Baudin E (2013) Are G3 ENETS neuroendocrine neoplasms heterogeneous? Endocr Relat Cancer 20 (5):649657. doi:10.1530/ERC-13-0027
60. McCall CM, Shi C, Cornish TC, Klimstra DS, Tang LH, Basturk O,
Mun LJ, Ellison TA, Wolfgang CL, Choti MA, Schulick RD, Edil
BH, Hruban RH (2013) Grading of well-differentiated pancreatic
neuroendocrine tumors is improved by the inclusion of both Ki67
proliferative index and mitotic rate. Am J Surg Pathol 37 (11):16711677. doi:10.1097/PAS.0000000000000089
61. Kulke MH, Anthony LB, Bushnell DL, de Herder WW, Goldsmith
SJ, Klimstra DS, Marx SJ, Pasieka JL, Pommier RF, Yao JC, Jensen
RT (2010) NANETS treatment guidelines: well-differentiated neuroendocrine tumors of the stomach and pancreas. Pancreas 39 (6):735752. doi:10.1097/MPA.0b013e3181ebb168
62. Metz DC, Jensen RT (2008) Gastrointestinal neuroendocrine tumors:
pancreatic endocrine tumors. Gastroenterology 135 (5):1469-1492.
doi:10.1053/j.gastro.2008.05.047
63. Yao JC, Phan AT, Chang DZ, Wolff RA, Hess K, Gupta S, Jacobs C,
Mares JE, Landgraf AN, Rashid A, Meric-Bernstam F (2008) Efficacy
of RAD001 (everolimus) and octreotide LAR in advanced low- to
intermediate-grade neuroendocrine tumors: results of a phase II study. J
Clin Oncol 26 (26):4311-4318. doi:10.1200/JCO.2008.16.7858