Neuroendocrine Cancer, Therapeutic Strategies in G3 Cancers

TLDR: In NEN G3, the distinction between NET G3 and NEC G3 is clinically and prognostically meaningful, and platinum-based chemotherapy remains the recommended first-line treatment in metastasized NEC patients.

Abstract: Background: According to the latest WHO classification, neuroendocrine neoplasm (NEN) G3 of the gastrointestinal tract is defined by a proliferation index Ki67 ab

Full text

E-Mail karger@karger.com
To the Point Article
Digestion 2017;95:109–114
DOI: 10.1159/000454761
Neuroendocrine Cancer, Therapeutic
Strategies in G3 Cancers
Anja Rinke Thomas M. Gress
Department of Gastroenterology, University Hospital Marburg, Marburg , Germany
Key Points: (1) In NEN G3, the distinction between NET G3
and NEC G3 is clinically and prognostically meaningful. (2)
Platinum-based chemotherapy remains the recommended
first-line treatment in metastasized NEC patients. (3) There is
no established standard for NET G3; treatments estab-
lished for NET G2 such as temozolomide-based chemother-
apy or peptide receptor radiotherapy may be considered. (4)
Specific trials for NET G3 are necessary. (5) New therapies for
NEC are urgently needed. Checkpoint inhibitors should be
evaluated. © 2017 S. Karger AG, Basel
Introduction: Pathological, Clinical and Prognostic
Heterogeneity of Neuroendocrine Neoplasm G3
According to the WHO classification of gastrointesti-
nal neuroendocrine neoplasm (NEN) 2010, high-grade
NEN are defined by a proliferative index of Ki67 exceed-
ing 20% and/or mitotic count above 20 mitoses in
10high-power fields
[1]. Only about 5% of all gastroin-
testinal NEN belong to the G3 category
[2] ; thus, only
limited data exists on appropriate treatment strategies
for this rare disease. Until recently, NEN G3 has been
regarded to be equivalent to poorly differentiated neuro-
endocrine carcinoma (NEC) and therefore, patients with
NEN G3 have been uniformly treated with platinum-
Keywords
Neuroendocrine neoplasm · Neuroendocrine carcinoma ·
Neuroendocrine tumour · Grading · Proliferation
marker Ki67 · Treatment · Chemotherapy
Abstract
Background: According to the latest WHO classification,
neuroendocrine neoplasm (NEN) G3 of the gastrointestinal
tract is defined by a proliferation index Ki67 above 20%. Gas-
trointestinal neuroendocrine carcinoma (NEC) is a rare and
highly aggressive malignancy and despite responsiveness to
chemotherapy, overall survival is poor. In the last 3–4 years,
the heterogeneity of the NEN G3 group has become evident.
Summary: In addition to the proliferative activity, the tu-
mour differentiation seems to play a major role, further di-
viding the NEN G3 group into neuroendocrine tumour (NET)
G3 and NEC. NET G3 often arise in the pancreas, and their
median proliferation rate is lower and prognosis is better as
compared to NEC. However, NET G3 show a limited response
to platinum-based chemotherapy. Lack of specific data for
NET G3 hampers clear therapeutic recommendations. Cis-
platin combined with etoposide is the established standard
regimen for advanced gastrointestinal NEC. Substituting
carboplatin for cisplatin or irinotecan for etoposide is consid-
ered alternative first-line regimen. There is no standard sec-
ond-line treatment; options are discussed within this review.
Received: September 26, 2016
Accepted after revision: November 25, 2016
Published online: February 4, 2017
Anja Rinke
Department of Gastroenterology and Endocrinology
University Hospital Marburg
Baldinger Strasse, DE–35043 Marburg (Germany)
E-Mail sprengea
 @  staff.uni-marburg.de
© 2017 S. Karger AG, Basel
www.karger.com/dig

Rinke/Gress
Digestion 2017;95:109–114
DOI: 10.1159/000454761
110
based chemotherapy. The largest retrospective study on
gastrointestinal NEN G3 – the so-called Nordic NEC tri-
al – reported a lower response rate (RR) to platinum-
based chemotherapy in patients with Ki67 <55% (15%)
compared to patients with high proliferative activity ex-
ceeding Ki67 55% (42%; p < 0.01). Nevertheless, overall
prognosis was superior in the subgroup of patients with
Ki67 <55%
[3].
Some recent publications stressed the importance of
morphological differentiation and suggested that the
NEN G3 group should be stratified into morphologically
well-differentiated neuroendocrine tumours (NET) with
an elevated proliferation rate (NET G3) and into poorly
differentiated NEC of large and small cell types
[4–6] .
Although prognosis in patients with NET G3 was slight-
ly worse compared to that in patients with NET G2
[6] ,
it was clearly more favourable when compared to NEC.
Forthe NET G3 subgroup, a median overall survival of
54–99 months was reported contrasting to 11–17 months
for the NEC subgroup
[5–7] . NET G3 were more often
somatostatin receptor positive (about 90% of cases ver-
sus 40–50% of cases in NEC), and the primary tumour
was often located in the pancreas (about two thirds of
cases). Functional activity is uncommon in the whole G3
group but still found more often in patients with NET
G3 (5–41%) as compared to NEC patients (0–6%)
[4–6] .
Response to platinum-based treatment was inferior in
NET G3 patients with an objective RR between 0 and
17% as compared to 31–35% in NEC patients in these
series
[4, 5] . Some studies reported a younger age at di-
agnosis for patients with NET G3 as compared to pa-
tients with NEC
[5] . Although the proliferation rate was
usually lower in NET G3 (median 30–50%) than that in
NEC (median 70–80%) and a Ki67 index above 55% was
rarely documented in NET G3, no clear cut-off value that
distinguishes NET from NEC could be defined
[7] .
Therefore, even the subgroup of NEC may be heteroge-
neous with a very aggressive subtype resembling small
cell lung cancer (SCLC; “NEC G4”) and a lower prolif-
erative NEC (“NEC G3”) with potential therapeutic im-
plications.
The concept of at least 2 different neoplasms within
the group of NEN G3 is supported by recent genetic
analyses: the typical mutations of well-differentiated
pancreatic NET such as DAXX/ATRX mutations
[8]
are found in about half of the pancreatic NET G3 cases
but not in pancreatic NEC, whereas large and small cell
NEC of the pancreas often display the inactivation of
the TP53 and the Rb/p16 pathways that are not found
in NET
[7, 9] .
First-Line Treatment in NEC
Evidence for treatment recommendations in gastroin-
testinal NEC is limited to mainly retrospective series and
a few non-controlled small trials. Patients are often treat-
ed in analogy to the more common SCLC.
For 25 years, cytotoxic treatment with cisplatin and
etoposide has been used as standard treatment for meta-
static NEC
[10] . Moertel and colleagues described an
overall remission rate of 67% (which may be an overesti-
mate compared to objective RRs based on actual radio-
logic criteria such as RECIST) and a median survival of
19 months. Response to treatment was independent of
primary tumour localization. Mitry et al.
[11] reported
comparable results with an RR of 42% and a median sur-
vival of 15 months using a modified cisplatin and etopo-
side regimen in a cohort of 41 patients.
In SCLC, substitution of carboplatin for cisplatin was
not inferior and better tolerated with regard to gastroin-
testinal and renal toxicity
[12] . For extrapulmonal NECs,
2 small studies with carboplatin etoposide reported a me-
dian survival of 20 and 12.7 months respectively
[13] .
Further support for the equal efficacy of carboplatin com-
pared to cisplatin originates from the Nordic NEC trial:
median survival was similar in patients treated with first-
line cisplatin and etoposide (12 months) and carboplatin
and etoposide (11 months)
[3] . The combination of iri-
notecan and cisplatin may be an alternative. A large
Japanese retrospective study reported higher RRs and
survival for this combination compared to cisplatin and
etoposide (RR 50 vs. 28% and survival 13 vs. 7 months).
However, in this study, unequal allocation of patients to
the 2 treatment groups may have influenced these obser-
vations, since location of the primary and elevated levels
of lactate dehydrogenase were independent prognostic
factors for overall survival, whereas the treatment sched-
ule was not
[14] .
Adding a third drug such as paclitaxel
[15] or vincris-
tin
[3] to platinum plus etoposide did not result in a sur-
vival advantage.
Although a subgroup of NEC patients expresses soma-
tostatin receptors, there is no indication to use somatosta-
tin analogues for tumour control in NEC. In addition,
highly proliferative NEC patients usually do not benefit
from somatostatin receptor-based peptide receptor ra-
diotherapy (PRRT)
[2] .
Surgery often is the first-line treatment in localized
disease, but cure by surgery alone is rare. Therefore,
most oncologists would recommend platinum-based ad-
juvant treatment, although data supporting this concept

NEN G3
Digestion 2017;95:109–114
DOI: 10.1159/000454761
111
are scarce. After curatively intended surgery, prophylac-
tic cranial irradiation is not recommended, since brain
metastases are much rarer than in patients with SCLC
[2,
3]
.
Second- and Third-Line Treatment Options in NEC
In patients demonstrating progressive disease after
platinum-based, first-line chemotherapy, no standard
treatment is available.
The limited data for second-line options from small
retrospective series are summarized in Table1 . In con-
trast to SCLC, topotecan did not induce any remissions
and resulted in a poor median PFS of only 2.1 months in
a small series of gastrointestinal NECs
[16] . We therefore
do not recommend this drug for second-line treatment in
gastrointestinal NEC.
The most commonly used options comprise: (i) the
reintroduction of platinum and etoposide when a durable
response is achieved in the first line and when progres-
sion occurs more than 3 months after the end of first-line
treatment, (ii) irinotecan-based treatment (FOLFIRI)
[17] , and (iii) oxaliplatin-based treatments (FOLFOX or
XELOX)
[2, 18] . The Uppsala group suggested the use of
temozolomide alone or in combination with capecitabine
and/or bevacizumab as second-line treatment
[19] . Pa-
tients with Ki67 <60% and positive somatostatin receptor
scintigraphy responded more often, probably corre-
sponding at least in part to patients with NET G3 rather
than NEC. Another small retrospective study in NEC pa-
tients failed to demonstrate the efficacy of temozolomide
[20] .
Treatment Strategies in Neuroendocrine Tumors G3
As the WHO classification 2010 did not define a G3
subgroup with retained differentiation, there are no spe-
cific treatment data available for this subgroup. Because
of the lower RRs to platinum-based chemotherapy ob-
served in several retrospective trials
[3–5] and the fact
that biology and prognosis of NET G3 are closer to NET
G2 than to NEC (see Introduction), considering treat-
ment options established for progressive NET G2 seems
appropriate. According to the ENETS guidelines
[2, 21],
therapeutic alternatives comprise temozolomide-based
chemotherapy and – in case of a pancreatic primary (two
thirds of cases) – streptozocin-based chemotherapy.
PRRT can also be successful in receptor-positive cases,
although generally RR and duration of response in G3
are lower when compared to G1/2 NET
[22] . The pos-
sible activity of everolimus and sunitinib for NET G3
needs further investigation. For the extremely rare dif-
ferentiated NET G3 with high proliferative activity ex-
ceeding Ki67 55%, most clinicians will probably still de-
Table 1. Second- and third-line treatments in NEC
Author, year Entities/primaries Schedule n ORR, % DCR, % mPFS mOS
Welin et al. [19], 2011 48% ≤30% Ki67
GEP/UK/lung
Temozolomide ±
capecitabine ±
bevacizumab
25 33 71 6 22
Hentic et al. [17], 2012 GEP NEC
mKi67: 50%
FOLFIRI 19 31 57 4 18
Olsen et al. [20], 2012 NEC mKi67: 50%
GEP/UK/lung/GU
Temozolomide 28 0 38 2.4 3.5
Sorbye et al. [3], 2013 NEC GEP/UK Reintroduction
platin + etoposide
26 15 27 ns 19 from initial
treatment
Olsen et al. [16], 2014 NEC GEP/UK
mKi67: 95%
Topotecan 22 0 23 2.1 3.2
Yamaguchi et al. [14], 2014 NEC GEP Platin + etoposide 23 17 ns 1.9 5
irinotecan + cisplatin 5 40 ns 4.8 8.7
Hadoux et al. [18], 2015 NEC GEP/UK/lung (65% LC, 30% SC) FOLFOX 20 29 64 4.5 9.9
ORR, objective response rate; DCR, disease control rate; mPFS, median progression free survival; mOS, median overall survival; GEP, gastroenteropancreatic;
UK, unknown; Cap, capecitabine; Bev, bevacizumab; NEC, neuroendocrine carcinoma; GU, genitourinary; ns, not stated; SC, small cell; LC, large cell.

Rinke/Gress
Digestion 2017;95:109–114
DOI: 10.1159/000454761
112
cide to use a platinum-based chemotherapy as first-line
treatment.
A randomized phase II trial comparing temozolo-
mide+ capecitabine with cisplatin + etoposide (EA2142)
as first-line treatment in NEN G3 excluding small cell
NEC will provide further information on the best treat-
ment choice.
Our treatment strategy in patients with NEN G3 is
summarized in Figure 1 .
Fig. 1. Diagnostic and therapeutic algorithm in patients with gas-
troenteropancreatic NEN G3.
1
If Ki67 ≤ 55%, always consult a pa-
thologist: NET G3or NEC? In doubtful cases, reference pathology
is recommended.
2
Staging includes CT of chest and abdomen (or
FDG-PET-CT), bone scintigraphy, MRI of the brain, tumour
marker neurone specific enolase (NSE), hormone markers only if
clinically suspectedsyndrome.
3
Staging includes CT of the chest,
CT or MRI of theabdomen, somatostatin receptor imaging (pref-
erentially
68
Ga-DOTATOC-PET-CT), chromogranin A, if nega-
tive:NSE, hormone markers only if clinically suspected syndrome.
4
Adjuvant treatment: 4 cycles of cisplatin + etoposide ± radiatio.
NEN, neuroendocrine neoplasm; NEC, neuroendocrine carcino-
ma; NET, neuroendocrine tumour; OP, operation; Strepto/5FU,
chemotherapy with streptozocin and 5 fluorouracil; pNET, pan-
creatic neuroendocrine tumour; PRRT, peptide receptor radio-
therapy; Tem/Cap, chemotherapy with temozolomide and
capecitabine; Carbo/Eto, chemotherapy with carboplatin and eto-
poside; TACE, transarterial chemoembolization; SIRT, selective
internal radiotherapy.
Color version available online
NEN G3
NEC NET G3
Staging
2
Staging
3
Localized/resectable
OP
R1/2 R0
Follow-up
Progression
Second line chemotherapy
e.g. Tem/Cap; FOLFIRI or
FOLFOX or paclitaxel
Adjuvant treatment
4
or
Follow-up
Metastasized or
locally irresectable
Cisplatin/carboplatin
and etoposide
Metastasized Localized
Follow-up
Progression
Second line chemotherapy
e.g. FOLFOX or Carbo/Eto
consider TACE/SIRT
consider PRRT
OP
R1/2 R0
1
Follow-up
Temozolomide ±
capecitabine
or strepto/5FU
(pNET only ) or PRRT
(receptor positive cases)

NEN G3
Digestion 2017;95:109–114
DOI: 10.1159/000454761
113
Perspectives
A better characterization of the heterogeneous NEN
G3 group and specific therapeutic trials for subgroups is
urgently needed.
The Scandinavian group started a phase II trial for
first-line treatment using a combination of temozolo-
mide and everolimus (NCT02248012) in patients with
NEN G3 Ki67 <55%.
New treatment options in NEC patients whose prog-
nosis has remained dismal for the last decades are urgent-
ly needed. Checkpoint inhibitors, such as nivolumab or
pembrolizumab, represent fascinating new compounds
that need to be studied in NEN G3. Kim et al.
[23] have,
for example, shown that PD-L1 expression in NEN is as-
sociated with G3 grading and poor prognosis. In Merkel
cell carcinoma, which also represents a high-grade neu-
roendocrine cancer, pembrolizumab as first-line treat-
ment resulted in an objective RR of 56%
[24] . In 12% of
gastric and colonic NEC, a mismatch repair deficiency
was found – a condition associated with response to
checkpoint inhibitors in colon cancer. Thus, rationale ex-
ists to further investigate immunotherapy in NEN G3 and
to identify patients who could benefit from these novel
treatment options.
Another approach is the evaluation of new targeted
treatments. For example, an increased expression of the
protein delta-like 3 (DLL3) was discovered in high-
grade pulmonary NEC
[25] , and a DLL3-targeted anti-
body-drug conjugate will be evaluated in several DLL3
expressing malignancies, including gastrointestinal
NEC.
Key Points
1. Within the group of NEN G3, the distinction between
NET G3 and NEC G3 is clinically and prognostically
meaningful.
2. Chemotherapy with cis- or carboplatin combined with
etoposide remains the recommended first-line treat-
ment in metastasized NEC patients.
3. There is no established standard for NET G3; treat-
ments established for NET G2 such as temozolomide-
based chemotherapy or PRRT may be considered.
4. Specific trials for NET G3 are necessary.
5. New therapeutic options for NEC are urgently needed.
Checkpoint inhibitors should be evaluated.
Disclosure Statement
The authors declare that they have no potential conflicts of in-
terest to disclose related to this review.
References
1 Rindi G, Arnold R, Bosman FT, Capella C,
Klimstra DS, Klöppel G, Komminoth P:
Nomenclature and classification of neuroen-
docrine neoplasms of the digestive system; in
Bosman FT, Carneiro F, Hruban RH, Theise
ND (eds): WHO Classification of Tumors of
the Digestive System. Lyon, IARC, 2010, pp
S13–S14.
2 Garcia-Carbonero R, Sorbye H, Baudin E,
Raymond E, Wiedenmann B, Niederle B, Sed-
lackova E, Toumpanakis C, Anlauf M, Cwikla
JB, Caplin M, O’Toole D, Perren A; Vienna
Consensus Conference Participants: ENETS
consensus guidelines for high-grade gastro-
enteropancreatic neuroendocrine tumors and
neuroendocrine carcinomas. Neuroendocri-
nology 2016;
103: 186–194.
3 Sorbye H, Welin S, Langer SW, Vestermark
LW, Holt N, Osterlund P, Dueland S, Hofsli
E, Guren MG, Ohrling K, Birkemeyer E, Thi-
is-Evensen E, Biagini M, Gronbaek H, Soveri
LM, Olsen IH, Federspiel B, Assmus J, Janson
ET, Knigge U: Predictive and prognostic fac-
tors for treatment and survival in 305 patients
with advanced gastrointestinal neuroendo-
crine carcinoma (WHO G3): the NORDIC
NEC study. Ann Oncol 2013;
24: 152–160.
4 Vélayoudom-Céphise FL, Duvillard P, Fou-
can L, Hadoux J, Chougnet CN, Leboulleux S,
Malka D, Guigay J, Goere D, Debaere T, Cara-
mella C, Schlumberger M, Planchard D, Elias
D, Ducreux M, Scoazec JY, Baudin E: Are G3
ENETS neuroendocrine neoplasms heteroge-
neous? Endocr Relat Cancer 2013;
20: 649–657.
5 Heetfeld M, Chougnet CN, Olsen IH, Rinke
A, Borbath I, Crespo G, Barriuso J, Pavel M,
O’Toole D, Walter T; Other Knowledge
Network Members: Characteristics and treat-
ment of patients with G3 gastroenteropancre-
atic neuroendocrine neoplasms. Endocr Relat
Cancer 2015;
22: 657–664.
6 Basturk O, Yang Z, Tang LH, Hruban RH,
Adsay V, McCall CM, Krasinskas AM, Jang
KT, Frankel WL, Balci S, Sigel C, Klimstra DS:
The high-grade (WHO G3) pancreatic neuro-
endocrine tumor category is morphologically
and biologically heterogenous and includes
both well differentiated and poorly differenti-
ated neoplasms. Am J Surg Pathol 2015;
39:
683–690.
7 Tang LH, Basturk O, Sue JJ, Klimstra DS: A
practical approach to the classification of
WHO grade 3 (G3) well-differentiated neuro-
endocrine tumor (WD-NET) and poorly
differentiated neuroendocrine carcinoma
(PD-NEC) of the pancreas. Am J Surg Pathol
2016;
40: 1192–1202.
8 Jiao Y, Shi C, Edil BH, de Wilde RF, Klimstra
DS, Maitra A, Schulick RD, Tang LH, Wolf-
gang CL, Choti MA, Velculescu VE, Diaz LA
Jr, Vogelstein B, Kinzler KW, Hruban RH, Pa-
padopoulos N: DAXX/ATRX, MEN1, and
mTOR pathway genes are frequently altered
in pancreatic neuroendocrine tumors. Sci-
ence 2011;
331: 1199–1203.
9 Yachida S, Vakiani E, White CM, Zhong Y,
Saunders T, Morgan R, de Wilde RF, Maitra A,
Hicks J, Demarzo AM, Shi C, Sharma R, La-
heru D, Edil BH, Wolfgang CL, Schulick RD,
Hruban RH, Tang LH, Klimstra DS,
Iacobuzio-Donahue CA: Small cell and large
cell neuroendocrine carcinomas of the pancre-
as are genetically similar and distinct from
well-differentiated pancreatic neuroendocrine
tumors. Am J Surg Pathol 2012;
36: 173–184.