Surgical Oncology 26 (2017) 438e445

Contents lists available at ScienceDirect

Surgical Oncology
journal homepage: www.elsevier.com/locate/suronc

Krukenberg tumors: Seed, route and soil

Annamaria Agnes a, Alberto Biondi a, *, Riccardo Ricci b, Valerio Gallotta c,
Domenico D'Ugo a, Roberto Persiani a
a
Polo Scienze Gastroenterologiche ed Endocrino-Metaboliche, Universita Cattolica del Sacro Cuore Fondazione Policlinico Universitario Agostino Gemelli
Largo F. Vito, 1 00168 Rome, Italy
b  Cattolica del Sacro Cuore Fondazione Policlinico Universitario Agostino Gemelli Largo F. Vito, 1
Polo Scienze Oncologiche ed Ematologiche, Universita
00168 Rome, Italy
c  Cattolica del Sacro Cuore Fondazione Policlinico Universitario Agostino Gemelli Largo F.
Polo Scienze Della Salute Della Donna E Del Bambino, Universita
Vito, 1 00168 Rome, Italy

a r t i c l e i n f o a b s t r a c t

Article history: The aim of this narrative review was to summarize the current evidence on Krukenberg tumors (KTs),
Received 10 June 2017 addressing what is known on their natural history and their impact on the clinical prognosis and which
Received in revised form are the most appropriate management strategies to treat this condition. A literature search was con-
28 August 2017
ducted on Pubmed up to December 2016, selecting the most relevant studies on the basis of the scope of
Accepted 9 September 2017
the review. KTs are ovarian metastases from primary signet-ring cell carcinomas., characterized by the
presence of a sarcoma-like stroma. They have three possible routes of diffusion (lymphatic, peritoneal and
Keywords:
hematogenous), but the preferential one is still unclear. Prognosis is dismal. When KTs are encountered
Krukenberg tumor
Signet ring cells
in the clinical practice, it is reasonable to offer surgical resection to young, fit patients with limited
Ovarian metastases disease. Palliative surgery should be considered for all patients with symptomatic disease. Further
studies should clarify the clinicopathologic characteristics of KTs, their main routes of diffusion, and the
possible role of prophylactic oophorectomy, lymphadenectomy and intraperitoneal chemotherapy. Mo-
lecular and transitional research should parallel the clinical one to help understanding the natural his-
tory of signet-ring cell carcinomas.
© 2017 Elsevier Ltd. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439
2. Search strategy and selection criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439
3. Definition of Krukenberg tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439
4. Epidemiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 440
5. Krukenberg tumors: seed, route and soil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 440
5.1. Seed: the cancer cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 440
5.2. Route . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 440
5.3. Soil: the ovaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441
6. Clinical presentation and diagnosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441
7. Prognosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441
8. Management strategies for KTs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 442
8.1. When the ovarian tumor is diagnosed first . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 442
8.2. When a primary tumor is identified before surgery (synchronous presentation) or has been already resected (metachronous presentation) 442
8.3. General recommendations for surgery and prophylactic oophorectomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443
8.4. When to consider palliative surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443
9. Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443

* Corresponding author.
E-mail address: [email protected] (A. Biondi).

http://dx.doi.org/10.1016/j.suronc.2017.09.001
0960-7404/© 2017 Elsevier Ltd. All rights reserved.
 A. Agnes et al. / Surgical Oncology 26 (2017) 438e445 439

10. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443

Author contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443
Declaration of interests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443
Conflict of interest statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443
Role of funding source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443
Ethics committee approval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 444

1. Introduction stromal proliferation resembling that of a fibrosarcoma. Kruken-

berg attributed these typical features to a primary ovarian tumor,
The “seed and soil” hypothesis was initially formulated by Paget defining it “fibrosarcoma mucocellulare carcinomatodes”. In 1902,
in 1889. This hypothesis states that particular tumor types have a though, Schlagenhaufer [13] recognized this neoplasm as a
predilection for specific secondary sites, independently of purely metastasis of a primary tumor of epithelial origin. In the following
anatomical or vascular factors [1]. In the following years, the seed years, some authors attained to the originally reported character-
and soil hypothesis has been confirmed by repeated observations of istics [14], while others extended the definition of KTs to all
the metastatic pattern of diverse types of cancer; in recent years, it ovarian metastases [15]. For this reason, in 1960, Woodruff and
is being further validated by new discoveries on the cellular basis of Novak proposed KTs to be defined only as those ovarian tumors
the tumors and on their mechanisms of metastasis [2,3]. consistent with Krukenberg's original criteria [16]. Accordingly, in
Krukenberg tumors (KTs) are ovarian metastases secondary to 1973, Serov and Scully established the diagnostic standard for KTs
specific malignancies (signet-ring cell carcinomas), most of which as follows: ovarian tumors with the presence of mucus-filled sig-
derive from the gastrointestinal tract [4,5]. In young women, net-ring cells (SRCs) accompanied by a sarcoma-like proliferation
signet-ring cell (SRC) carcinomas have a predilection for the ovary of the stroma [17].
as a secondary site, rendering KTs a perfect demonstration of the The standard definition of KTs, however, is not universally
seed and soil theory. Nevertheless, KT's mechanisms of metastasis, accepted. In facts, in spite of the relatively well-defined diagnostic
prognostic implications and optimal therapeutic strategies are still criteria, many authors continued to use the term “Krukenberg tu-
a matter of debate. In facts, many doubts exists concerning their mor” to describe all the secondary tumors involving the ovaries
initial origin and their relation to lymphatic and peritoneal disease [18,19], while others, given the high rate of stomach as a primary
[4,6,7], as well as on the choice of the most beneficial treatment site, used it to describe all the ovarian metastases secondary to
when they occur in the natural history of primary SRC malignancies gastric cancer [20e24]. As a result, reviewing the literature may be
[4,8e11]. very disorienting, as epidemiologic and prognostic studies report
The aim of this review is to summarize the current evidence on heterogeneous series. For this reason, some authors advocated that,
KTs, addressing what is known on their natural history and their as KTs are rare and their definition is potentially confusing, the term
impact on the clinical prognosis and which are the most appro- should be abandoned [15]. Nonetheless, the recognition of KTs may
priate management strategies to treat this condition.

2. Search strategy and selection criteria

References for this article were identified through searches of

PubMed with the search terms “Krukenberg”, “ovarian metastases”
and “seed and soil hypothesis”. The search was conducted including
every listed article up to December 2016 and included only English
language studies. Articles were screened by their title and abstract
and selected for full-text review by two authors (AA and AB). The
reference list of the articles examined in full-text was screened for
further eligible articles. The final reference list was generated on
the basis of originality and relevance to the scope of this review. The
flowchart for study selection is shown in Fig. 1.

3. Definition of Krukenberg tumors

In 1896, Friedrich Ernst Krukenberg (1871e1946), while work-

ing as a student in the laboratory of the German pathologist Felix
Jacob Marchand, described 5 cases of a characteristic ovarian tu-
mor [12]. Its clinicopathologic features were the frequent associ-
ation with ascites and its bilateralism; ovaries retained their
general shape, with a ‘‘knobby’’ but smooth surface, and were
microscopically characterized by firm areas alternating with
myxomatous zones, prominent involvement of lymphatics and
from the presence of cells containing mucin that pushed the nu-
cleus on one side (signet ring cells) surrounded by a cellular Fig. 1. The selection process of the studies included in this review.
440 A. Agnes et al. / Surgical Oncology 26 (2017) 438e445

still have a relevant significance in defining the biological behavior from the primary tumor (which had an amplification of the FGFR2
of certain tumors. In facts, they are strictly related to SRC carci- gene) in comparison with the metastases (which had an amplifi-
nomas, which show typical clinicopathological features and may cation of the TGFBR2 gene). Results pointed towards a tumor ge-
have a specific pattern of response to therapy [25]; thus, the correct netic “divergence” occurring in an early phase [38]. Of note, both
identification of KTs may be useful to clarify the metastatic mech- FGFR2 and TGFBR2 are involved in the cancer-stromal interaction.
anisms of SRCs tumors, to assess the patients' prognosis and The importance of cancer-stromal interaction is increasingly being
consequently to select the best therapeutic protocols. recognized, and may have a prominent role in the mechanism of
diffusion of certain tumors (see Soil: the ovary) [39].
4. Epidemiology In summary, it is still not clear if the dispersion of the “seed” is
an early mechanisms in the natural history or tumors with SRC
In most series, ovarian metastases represent 15%e28% of all features. The implications of identifying a common pathway of
ovarian tumors. Of these, non-gynecologic malignancies represent mutations between the primary and the metastatic cancer cell are
the 59e78% (gastrointestinal tract metastases representing the important, as this pathway may be the best target to consider in
34e47% and breast metastases the 14e31%) and gynecologic ma- developing targeted therapeutic agents [38].
lignancies the 18e41% [26e28]. The incidence of KTs is rare, as they
are reported to represent 1$3% of all ovarian tumors and 9% of all 5.2. Route
ovarian metastases in most Western series [28]. In Eastern pop-
ulations, that have a high incidence of gastric cancer, KTs represent Since the original description of Krukenberg, who described the
up to 3$6% of all ovarian tumors and 21$5% of all ovarian metastases frequent occurrence of lymphatic involvement in the “fibrosarcoma
[11,27]. In facts, the most frequent site of origin for KTs is the mucocellulare carcinomatodes”, retrograde lymphatic spread is
stomach (76%), followed by the colorectal tract (11%), the breast believed to be the most likely route for metastasis in KTs [7]. By the
(4%), the biliary tract and the gallbladder (3%), and by other sites lympathic route hypothesis, cancer cells metastasize to peri-gastric
(small intestine, appendix, pancreas, uterus, urinary bladder, renal nodes, forming carcinomatous emboli that block the lymphatic
pelvis) (15%) [4]. In particular, by their original definition, KTs are upward flow. Then, they reach the para-aortic and pelvic lymph
strictly associated with diffuse and schirrous-type gastric cancers nodes along with the lymphatic reflux. As pelvic organs with a rich
(which often harbor a SRCs component) [4,8,17,29], and to colo- network of lymphatic vessels, ovaries would be preferentially
rectal, breast and appendiceal cancers with a SRC component [4]. reached by the cancer cells. Most of the pathological studies pro-
The primary origin of this tumors remains occult in 7e18% of cases, pend for the lymphatic route hypothesis, as lymphovascular inva-
thus the existence of a “primary” KT is still object of debate [11,15]. sion is frequently detected in KTs, and tumor involvement is
KTs are reported to be bilateral in 72e83% of cases, more often reported to occur by lymphatics in the hilum and cortex of the
than other ovarian metastases [4,8,15]. Mean age of women diag- ovary rather than on its surface, which is often free of disease
nosed with KTs ranges from 40 to 45 years, and 35e45% of them are [4,5,7,15,40]. In addition, some authors reported the frequent as-
under 40 years of age [4]. This age of distribution is definitely sociation between the incidence of ovarian metastases and the
younger than that of women with primary gastric, colorectal or extent of lymph node involvement [41e43]. Finally, there were
breast cancers. The first reason is that many SRC carcinomas are reports of patients with early-stage gastric cancer who developed
prominent in younger patients and occurring more frequently in ovarian metastases. In patients with early-stage gastric cancer the
woman [30,31]. A second reason may be related to the greater tro- height of the gastric mucosa is significantly reduced; gastric mu-
phism of the ovaries during the reproductive age, which could cosa and submucosa both contain a rich lymphatic plexus, so in
render them more receptive as a site of metastases. In clinical series, these patients the lymphatic capillaries may be located nearer the
up to 60% of KTs are diagnosed in premenopausal women [8,31]. surface epithelium. Therefore, in this setting, cancer cells are sup-
Moreover, some studies reported the detection of microscopic posed to early involve the lymphatic vessels and thus lead to
ovarian metastasis in 41% of patients with gastric SRC carcinoma [32] ovarian metastasis via the lymphatic route [44].
and in 55% of gastric cancer patients younger than 36 years of age A second accredited theory is that of peritoneal spread [3,49]. In
[33]. In addition, the proportion of ovarian metastases in SRC colo- particular, many clinicians (surgeons and oncologists) consider
rectal carcinoma is even reported to be between 60 and 100% [4]. ovarian metastases as a form of peritoneal involvement, and treat
patients accordingly. This theory is supported by the frequent as-
5. Krukenberg tumors: seed, route and soil sociation between KTs, free peritoneal cancer cells and peritoneal
carcinomatosis [10,15]. SRC carcinomas are well-known to have a
5.1. Seed: the cancer cell high rate of peritoneal diffusion [45]. In addition, detection of free
peritoneal cancer cells is possible even in tumors that do not invade
While SRCs in KTs mirror SRCs from the primary tumor, the real the serosal layer of the stomach and colon [46e48].
relation of the metastatic cell with its primary and the effective A third possibility is that KT is due to hematogenous diffusion,
timing of the initiation of the metastatic process are still unknown. for several reasons. First, KTs are prevalent in premenopausal
One hypothesis is that metastases derive from cells identical to the women, which have a greater vascularity of the ovaries [49]. Sec-
primary tumor, occurring as a stochastic event [34]. Accordingly, ond, the presence of hilar metastases and lymphovascular invasion
many studies demonstrated similarities between the expression corroborate the possibility of hematogenous diffusion in addition to
profiles of primary tumors and metastases [35,36]. Moreover, one a lymphatic one. Third, the occurrence of KTs as a consequence of
recent study specifically identified major concordance between the breast cancer favors an hematogenous more than a lymphatic or
primary gastric tumors and the paired Krukenberg tumors in peritoneal route of diffusion [7]. Fourth, recent investigations on
regards to the expression of a panel of biomarkers (HER2/neu, c- circulating tumor cells (CTCs) [50e52], which may be present even
MET, p53, Ki-67) [37]. However, another recent study characterized when the tumors are far from having invaded the serosal layer, are
both the cells from the primary tumor and the cells from the questioning many of the theories on the mechanisms of metastasis
ovarian metastases in a case of gastric cancer. The primary tumor of certain cancers. In this regard, one study has recently demon-
was a poorly differentiated, diffuse carcinoma and the metastasis a strated that epithelial ovarian CTCs are able to metastasize to the
KT. This study identified different genetic mutations in the cells omentum hematogenously [53].
 A. Agnes et al. / Surgical Oncology 26 (2017) 438e445 441

A concurrence of the different mechanisms of metastasis is [60]. In this regards, it may be hypothesized that certain organs
the most accepted possibility. In facts, the association between have particular stromal features that renders them an ideal soil for
lymphatic and peritoneal tropism is a well-known feature of SRC the growth of SRCs. An extension of this concept derives from the
tumors [54]. Nevertheless, the typical histologic features of KTs recent discovery of the “premetastatic niches”, modifications in the
seem to point out to a specific biologic behavior. In this regards, microenvironment of the target organ induced by tumor associated
another possibility may be that KTs metastasize through the cells and specific soluble factors [61], which may be interpreted as
lymphatic route, but via the subperitoneal lymphatics and not via the receptiveness of specific soils to be “fertilized” by specific tu-
the retroperitoneal lymphatic route [55]. Even if not completely mors [62].
understood, the subperitoneal space seem to be extremely impor-
tant in the diffusion of tumor cells between the ovaries and the 6. Clinical presentation and diagnosis
abdominal cavity, and to have a major role in the peritoneal
diffusion of various diseases [56]. Further investigations on the role Krukenberg tumors are diagnosed before the primary neo-
of subperitoneal space may clarify this issue and help merging the plasms in up to 65% of cases [63]. KTs may present before the
lymphatic and the peritoneal diffusion theories. Another possibility clinical diagnosis of the primary tumor, be detected during staging
is that KTs may be the consequence of a lymphatic or hematoge- (synchronous presentation) or after resection of the primary tumor
nous diffusion, representing a reservoir for SRCs which secondarily (metachronous presentation). The metachronous presentation oc-
diffuse to the peritoneum during ovulation or through the sub- curs even years after the primary diagnosis [4].
peritoneal lymphatics as well. KTs present both asymptomatically or symptomatically. When
To date, reports have been sparse and no study specifically present, symptoms of the ovarian disease could be prominent even
focused on the timing and real association between KTs, lymphatic in respect to the primary disease. In particular, symptoms of Kru-
involvement, positive cytology and peritoneal carcinomatosis, kenberg tumors are due to the mass effect or to hormonal imbal-
leaving many obscure points. ances caused by the malignant growth. Patients usually complain of
abdominal or pelvic pain, dyspareunia, weight loss, bloating and
5.3. Soil: the ovaries abdominal distension (as a consequence of ascites). In other cases,
the ovarian stroma may be triggered by the presence of cancer cells
The ovaries have several reasons to be a preferential site of causing a hormonal imbalance which may result changes in the
metastases in patients with gastrointestinal carcinomas. menstrual cycle, hirsutism/virilization (“ovarian tumor with func-
According to the lymphatic diffusion hypothesis, they are one of tioning stroma”) and in vaginal bleeding, even in postmenopausal
the first organs encountered by the lymphatic backflow. According women [4,5,28,59].
to the peritoneal diffusion hypothesis, their anatomical position is At imaging, in comparison with other ovarian metastases, KTs
favorable to the contact with free intraperitoneal cancer cells. In are often bilateral. They are more often solid than cystic, but well-
premenopausal women, the monthly ovulations cause an inflam- demarcated intramural cysts could be present as well. At ultraso-
mation site and a small amount of blood loss followed by clotting; nography, they are often homogeneously hyperechoic and exhibit
these factors would aid the entrapment of free intraperitoneal the “lead vessel sign”, consisting of a large lead vessel penetrating
cancer cells (“tumor cell entrapment hypothesis”) [6]. In post- the tumor from the periphery and then branching in a tree pattern
menopausal women, instead, metastases could be due to cancer [64e66]; CT shows lobulated, mostly solid tumors with homoge-
cells which colonize the ovary before menopause, remaining latent neous enhancement of the solid portion, and MRI shows a solid
for years. Another possibility is that free intraperitoneal cancer cells component with an hypointense signal density at T2-weighting
may be trapped by the milky spots (focal aggregations of mesen- [66e68].
chymal cells surrounding lymphatic and vascular channels) present Pathologically, ovarian metastases are well-known to mimic
on the ovarian surface, or by the rough ovarian surface of the primary epithelial ovarian tumors in up to 25% of cases [28]. This
postmenopausal ovary [6]. Lastly, according to the hematogenous problem is frequently encountered with colorectal metastases of
diffusion hypothesis [49], the high vascularity of the ovaries during the cystic type [5]. KTs, instead, are mostly solid and quite char-
the fertile age exposes them to a great number of CTCs, increasing acteristic due to their SRC component [4,15]. Nevertheless, diffi-
the probability of a metastatic deposit. culties may be encountered in diagnosing these tumors as well, as
The role of hormones in rendering ovaries more receptive to their histopathologic aspect may resemble that of other tumors
metastases is non-trascurable as well. In facts, KTs are especially containing SRCs or SRCs containing non-mucinous material (i.e. the
common in young women, and a high association with pregnancy is signet-ring stromal tumor) [7,69]. In addition, a variant of KTs,
reported [4]. The growth of SRC gastric cancers has been reported to called tubular KT, may be confused with a Sertoli-Leydig cell tumor
partially rely on estrogens [57] and the role of estrogen receptors is due to a similar cellular pattern and to the frequent stromal
current under investigation as a possible molecular feature of luteinization typical of both tumors [4,7]. This is particularly risky
certain types of gastric cancer [58]. when KTs occur near pregnancy, as many ovarian tumors occurring
Finally, there may also be a prominent role of the ovarian in this period tend to exhibit stromal luteinization [59]. Moreover,
stroma. KTs, especially those detected near pregnancy, may show a when SRCs are sparse and in low number, these tumors may be
typical stromal luteinization [4,59], which is often accompanied by confused with fibromas [55]. Finally, as in KTs glands or neuroen-
clinical hormone-associated manifestations. Then, one of the most docrine cells may be present together with SRCs, this variations
striking features of KTs is their fibrous stroma, which may be pre- should be accounted for as well during pathological examination
ponderant even in regards to SRCs [7]. SRCs, especially those of [4].
gastric origin, are often accompanied by a fibrous reaction even in
the primary sites (especially the stomach). In addition, breast 7. Prognosis
cancers harboring SRCs (which are often of the lobular histotype),
have a particular tropism for the stomach as a secondary site of There are many retrospective studies evaluating the prognosis
implant, causing a stromal reaction which may be indistinguishable of ovarian metastases in general [11,21,23,24,70e72], but few are
from primary gastric cancer. Also, there are reports of primary the studies focusing on prognosis of KTs defined by their original
gastrointestinal SRC metastasizing to the breast and to the ovaries characteristics [4,8e10].
442 A. Agnes et al. / Surgical Oncology 26 (2017) 438e445

In 1968, Hale [8] reported a series of 81 patients with KT of metastasectomy, especially when the presentation is metachro-
various origins. Of 81 patients, 68 were followed-up, and their nous and the disease clinically resectable without residual disease.
median survival from the time of diagnosis was 7$1 months. In
1992, Petru et al. [9] reported a series of 82 patients with non- 8. Management strategies for KTs
gynecological malignancies metastatic to the ovaries. In this se-
ries 13 patients had a histologically confirmed KT; their median Optimal management strategies for KTs have not been estab-
overall survival was 6 months (1e46 months). In 2006, Kiyokawa lished yet. The recommendations valid for ovarian metastases in
et al. [4] reported a series of 120 patients with KT. Of these, the general may be applied to KTs, but literature should be interpreted
survival outcomes of 69 patients were recorded up to one year and with caution when considering the natural history of primary SRC
the survival of 44 patients up to two years. The median overall carcinomas. In this section, we address the main situations that
survival was 13$4 months (1e67), with a 1-year OS of 36% and a 2- could be faced in the management of KTs.
year OS of 21%. In 2007, Yook et al. [10] reported a series of 37
patients with ovarian metastases from gastric carcinoma, of which 8.1. When the ovarian tumor is diagnosed first
35 were KTs. In this series, the median survival was 17 months. It
was significantly affected only by the presence of peritoneal seed- As soon as an ovarian metastasis is suspected preoperatively
ing (p ¼ 0$013), while no survival differences were detected be- (bilateralism, suggestive imaging, and occurrence of associated
tween patients with synchronous versus metachronous gastrointestinal symptoms) the first step in management should be
metastases. Finally, in 2010, Kim et al. [11] reported a series of 158 a thorough staging, with the aim of identifying the primary tumor
patients with ovarian metastases, of which 34 had a pathologically or to assess the presence of other secondary localizations. This
confirmed KT, and 25 a pathologically confirmed gastric cancer. The guarantees the opportunity to plan the best treatment strategy, and
median survival of gastric cancer patients with KT was 12 months, to avoid delays in management of the primary neoplasm and un-
and no survival difference was found in comparison with the me- necessary surgical procedures, as hysterectomy [6] or pelvic and
dian survival of patients with non-KT metastases (12 months, para-aortic lymphadenectomy [9], that are part of the surgical
p ¼ 0$486). Patients were analyzed in regards to the entity of management of primary ovarian cancers but have an unclear role in
cytoreduction and the use of postoperative chemotherapy. A sig- the resection of secondary ovarian tumors.
nificant survival benefit was detected with the administration of However, there may be cases in which a KT is diagnosed by
chemotherapy (p ¼ 0$012), but optimal cytoreduction had no sig- pathology after oophorectomy is performed, and the primary is still
nificant role (p ¼ 0$169). unknown. In these cases, due to the still unclear existence of “pri-
Previous retrospective series on the prognostic factors of mary KTs”, the maximal effort should be done to identify the pri-
ovarian metastases in general are more detailed. Median survivals mary neoplasm.
of patients with nongynecological ovarian metastases range be- In both cases, it should be considered that primary SRC carci-
tween 15 and 42 months [11,18,19,21,72]. In these patients, survival nomas of the stomach and breast may be particularly challenging to
is significantly associated with the primary site of the tumor (pa- diagnose. In particular, gastric SRCs are difficult to diagnose even
tients with a gastric primary have a median survival of 13e19$2 with repeated endoscopies [84]. Therefore, a high threshold of
months, patients with colorectal primary of 27$3e48 months and suspect and knowledge on the natural history of SRC tumors is
patients with a breast primary of 16e54 months) [18,19,21,72,73], essential in these cases.
with the timing of diagnosis (with a survival advantage detected in
patients diagnosed with metachronous disease) [18], and with the 8.2. When a primary tumor is identified before surgery
surgical resection of the metastasis (with an overall survival benefit (synchronous presentation) or has been already resected
for patients treated with radical resection of the tumor) [21,60,72]. (metachronous presentation)
As for ovarian metastases in gastric cancer patients, studies spe-
cifically conducted in this population reported a survival benefit for Even when ovarian metastases are correctly identified and
radical resection in metachronous tumors [24,70] or both in syn- deemed resectable, many surgeons consider KTs as a definite
chronous and in metachronous tumors [20,22]. Others reported a marker of incurable metastatic disease, and are dissuaded by
benefit of surgery plus chemotherapy in comparison to palliative attempting a surgical treatment, due to the belief that KT is a
surgery or surgery alone, for both patients with synchronous and marker of peritoneal dissemination and unresectable disease at
metachronous tumors [74,75]. One series reported a survival surgical laparotomy and an indicator of overall dismal prognosis
advantage for patients with synchronous and metachronous [7,85]. Others, instead, express the need for considering more
ovarian metastases treated with cytoreduction plus HIPEC [23] and aggressive treatments, due to the young age of the patients, to the
another a benefit of HIPEC in adjunct to cytoreduction for patients occasional long-term survival of KT patients when a meta-
with metachronous tumors [76]. In particular, median survival after stasectomy is performed, and to the need of ameliorating the
complete cytoreduction was reported to reach 34 months [24]. As quality of life in patients with symptomatic pelvic disease [4,9,86].
for ovarian metastases in colorectal cancer patients, focused studies When KTs present synchronously with the primary tumor, the
suggest that complete cytoreduction plays an important role as value of surgery is unclear. Theoretically, the optimal treatment
well. In facts, in two previous studies, complete metastasectomy strategy should be determined considering the preferential route of
resulted in prolonged survival compared with palliative surgery diffusion. According to the peritoneal diffusion theory, HIPEC may
[77,78], and other studies confirmed a survival benefit in patients be a valuable strategy in managing KTs associated to peritoneal
treated with radical surgery and aggressive cytoreduction [79e82]. disease or in preventing future peritoneal dissemination after oo-
In particular, in cases of CC0 resection, 5-year survival were re- phorectomy. According to the lymphatic route theory, instead,
ported to reach the 46$6%, with a 56 months median survival, and there may be a rationale in performing oophorectomy plus lym-
even patients with CC1 survived longer than those with CC282. phadenectomy for the treatment of KTs. If, on one hand, pelvic or
Finally, there are reports of colorectal cancer patients undergoing even para-aortic lymphadenectomy may be performed if KT is
pelvic exenteration who survived more than 5 years [83]. Therefore, misdiagnosed as a primary ovarian tumor, on the other this
general recommendations for surgeons faced with ovarian metas- adjunctive procedure may have a role in staging or treating the
tases from gastric and colorectal cancer include consideration for disease in a KT is correctly identified before surgery. However, there
 A. Agnes et al. / Surgical Oncology 26 (2017) 438e445 443

are no reports focusing on the possible role of lymphadenectomy in parallel this course of research by a contemporary analysis of the
KTs, and even reports on this procedure as a treatment for ovarian genomic pattern of primary and metastatic tumors from the same
metastases in general are scarce [10,78]. Therefore, given its non- patient.
negligible morbidity [87], it is not currently recommended. With the aim of clarifying the main route of diffusion of KTs,
promising areas of study seem to be both the investigation on the
8.3. General recommendations for surgery and prophylactic role of CTCs in SRC carcinomas of gastrointestinal and breast origin,
oophorectomy and the characterization of the properties of the subperitoneal
space.
As KTs are rare and patients are often managed by different Lastly, the use of novel biomarkers as CTCs, cDNA and miRNAs
surgical specialist (surgical oncologist and gynecologist), multi- [95] may represent a method to identify recurrence in a timely way,
disciplinary collaboration, with revision of the different series and to stratify prognosis to determine which patients could benefit
and creation of institutional guidelines, should be of the utmost from a prophylactic oophorectomy.
importance. General recommendations for cross-specialty situa-
tions have already been proposed. When operating on ovarian
10. Conclusions
tumors, gynecological surgeons should conduct a complete exam-
ination of the gastrointestinal tract (especially of the stomach)
KTs are ovarian metastases of SRC carcinomas of different origin.
whenever possible [88]. They should also consider appendectomy,
These tumors are rare, their mechanism of metastasis is still un-
as the appendix may harbor the primary focus of SRC carcinoma in
clear and evidences on their optimal treatment are scarce. Further
some cases [4]. Similarly, while performing surgery for gastric,
clarification of the clinicopathologic characteristics of KTs and of
colonic and appendiceal cancers, general surgeons should conduct
their main route of diffusion would be both important in defining
a meticulous pelvic examination with an accurate visual and
which this optimal treatment could be. Given the current knowl-
manipulator assessment of the aspect and trophism of the ovaries,
edge, when KTs are encountered in the clinical practice, it is
and consider oophorectomy whenever an anomaly is detected [88].
reasonable to offer surgical resection to young, fit patients with
Unfortunately, affected ovaries are not macroscopically enlarged
limited disease. Palliative surgery should be considered for all pa-
or abnormal in all cases [5]. For this reason, during surgery for
tients with symptomatic disease, after a thorough risk-benefit
gastrointestinal malignancies, some authors have advocated the
assessment. Prospective clinical studies are needed to clarify the
routine performance of a surface scraping of the ovary [6], or a
possible role of prophylactic oophorectomy, pelvic lymphadenec-
prophylactic oophorectomy in all cases of post-menopausal disease
tomy and HIPEC. Molecular and transitional research should par-
and in selected cases of premenopausal ones [6,10,42]. Prophylactic
allel the clinical one to help clarifying the natural history of the
oophorectomy is safe and feasible in adjunct to standard surgery, as
diverse SRC carcinomas.
complications are uncommon [89], and would have the advantage
of avoiding a second laparotomy [6]. Nevertheless, KTs are rare
(1$6% of all ovarian tumors 11,27,28), and prophylactic oophorectomy Author contributions
is documented to cause consistent detrimental long-term effects
(increase in all-cause mortality, cardiovascular diseases, cognitive Agnes A, Biondi A, Ricci R, Gallotta V, D'Ugo D and Persiani R
and neurologic impairment, depression and anxiety, osteoporosis, made substantial contributions to conception and design of the
sexual dysfunction), especially in premenopausal women. Post- paper; Agnes A performed the literature search and drafted the
menopausal woman may be affected as well [89]. Therefore, pro- article; Biondi A, Ricci R, Gallotta V, D'Ugo D and Persiani R revised
phylactic oophorectomy should be accurately weighted case by critically the paper and all Authors gave the final approval of the
case, and probably reserved only to selected patients, until further definitive version of the article.
studies are conducted.
Declaration of interests
8.4. When to consider palliative surgery
The authors declare no financial or personal relationship with
SRCs have been associated to detrimental outcomes and scarce other people or organizations that could inappropriately influence
response to chemotherapy in many tumor types [25,90,91]. More- or bias this work.
over, ovaries has been suggested to be “metastatic sanctuaries”, due
to the limited effects of chemotherapy in patients with ovarian
metastases of colorectal cancer [92]. As a consequence, the Conflict of interest statement
administration of upfront chemotherapy may be of low efficacy in
treating symptomatic disease in patients with KTs. For this reason, The authors declare no conflict of interest related to the publi-
palliative surgery prior to chemotherapy should be contemplated cation of this manuscript.
in symptomatic patients, especially considering the fact that the
quality of life of patients may be significantly hampered by the Role of funding source
symptoms of KTs. Unfortunately, there are no studies comparing
quality of life changes after chemotherapy versus palliative sur- Not financially supported.
gery, neither in the ovarian metastases patients nor in the KTs
population.
Ethics committee approval
9. Perspectives
Not applicable.
Molecular characterization of primary gastric, colorectal and
breast cancers, which is currently under the spotlight [93,94], may Acknowledgments
be a promising approach to understand the biological behavior of
SRC carcinomas; the characterization of ovarian metastases should None.
444 A. Agnes et al. / Surgical Oncology 26 (2017) 438e445

References predictors for breast cancer, N. Engl. J. Med. 355 (6) (2006) 560e569.
[36] B. Weigelt, A.M. Glas, L.F. Wessels, A.T. Witteveen, J.L. Peterse, L.J. van't Veer,
Gene expression profiles of primary breast tumors maintained in distant
[1] S. Paget, The distribution of secondary growths in cancer of the breast, Lancet
metastases, Proc. Natl. Acad. Sci. U. S. A. 100 (26) (2003) 15901e15905.
133 (3421) (1889) 571e573.
[37] B. Wang, K. Sun, Y. Zou, Comparison of a panel of biomarkers between gastric
[2] I.J. Fidler, G. Poste, The “seed and soil” hypothesis revisited, Lancet Oncol. 9 (8)
primary cancer and the paired krukenberg tumor, Appl. Immunohistochem.
(2008) 808.
Mol. Morphol. (2016, Mar 3) [Epub ahead of print].
[3] G.P. Gupta, J. Massague, Cancer metastasis: building a framework, Cell 127 (4)
[38] L.D. Nadauld, S. Garcia, G. Natsoulis, et al., Metastatic tumor evolution and
(2006) 679e695.
organoid modeling implicate TGFBR2 as a cancer driver in diffuse gastric
[4] T. Kiyokawa, R.H. Young, R.E. Scully, Krukenberg tumors of the ovary: a
cancer, Genome Biol. 15 (8) (2014) 428.
clinicopathologic analysis of 120 cases with emphasis on their variable
[39] M. Yashiro, K. Hirakawa, Cancer-stromal interactions in scirrhous gastric
pathologic manifestations, Am. J. Surg. Pathol. 30 (3) (2006) 277e299.
carcinoma, Cancer Microenviron. 3 (1) (2010) 127e135.
[5] R.H. Young, From krukenberg to today: the ever present problems posed by
[40] F. Holtz, W.R. Hart, Krukenberg tumors of the ovary: a clinicopathologic
metastatic tumors in the ovary: part I. Historical perspective, general princi-
analysis of 27 cases, Cancer 50 (11) (1982) 2438e2447.
ples, mucinous tumors including the krukenberg tumor, Adv. Anat. Pathol. 13
[41] T.C. Chang, C.C. Changchien, C.W. Tseng, et al., Retrograde lymphatic spread: a
(5) (2006) 205e227.
likely route for metastatic ovarian cancers of gastrointestinal origin, Gynecol.
[6] P.H. Sugarbaker, Peritoneal Carcinomatosis, Kluwer Academic Publishers,
Oncol. 66 (3) (1997) 372e377.
Boston, 1996.
[42] N.K. Kim, H.K. Kim, B.J. Park, et al., Risk factors for ovarian metastasis
[7] O.M. Al-Agha, A.D. Nicastri, An in-depth look at Krukenberg tumor: an over-
following curative resection of gastric adenocarcinoma, Cancer 85 (7) (1999)
view, Arch. Pathol. Lab. Med. 130 (11) (2006) 1725e1730.
1490e1499.
[8] R.W. Hale, Krukenberg tumor of the ovaries. A review of 81 records, Obstet.
[43] Q. Feng, W. Pei, Z.X. Zheng, J.J. Bi, X.H. Yuan, Clinicopathologic characteristics
Gynecol. 32 (2) (1968) 221e225.
and prognostic factors of 63 gastric cancer patients with metachronous
[9] E. Petru, H. Pickel, M. Heydarfadai, et al., Nongenital cancers metastatic to the
ovarian metastasis, Cancer Biol. Med. 10 (2) (2013) 86e91.
ovary, Gynecol. Oncol. 44 (1) (1992) 83e86.
[44] N. Kakushima, T. Kamoshida, S. Hirai, et al., Early gastric cancer with Kru-
[10] J.H. Yook, S.T. Oh, B.S. Kim, Clinical prognostic factors for ovarian metastasis in
kenberg tumor and review of cases of intramucosal gastric cancers with
women with gastric cancer, Hepatogastroenterology 54 (75) (2007) 955e959.
Krukenberg tumor, J. Gastroenterol. 38 (12) (2003) 1176e1180.
[11] W.Y. Kim, T.J. Kim, S.E. Kim, et al., The role of cytoreductive surgery for non-
[45] I. Thomassen, Y.R. van Gestel, B. van Ramshorst, et al., Peritoneal carcino-
genital tract metastatic tumors to the ovaries, Eur. J. Obstet. Gynecol. Reprod.
matosis of gastric origin: a population-based study on incidence, survival and
Biol. 149 (1) (2010) 97e101.
risk factors, Int. J. Cancer 134 (3) (2014) 622e628.
[12] F. Krukenberg, Ueber das Fibrosarcoma ovarii mucocellulare (carcinomat-
[46] B. Huang, Z. Sun, Z. Wang, et al., Factors associated with peritoneal metastasis
odes), Arch. Gynecol. Obstet. 50 (2) (1896) 287e321.
in non-serosa-invasive gastric cancer: a retrospective study of a
[13] F. Schlagenhaufer, XXX. Ueber das metastatische Ovarialcarcinom nach Krebs
prospectively-collected database, BMC Cancer 13 (2013) 57.
des Magens, Darmes und anderer Bauchorgane (Part 1 of 2), Gynecol. Obstet.
[47] M. Yoshida, T. Sugino, K. Kusafuka, et al., Peritoneal dissemination in early
Invest. 15 (4) (1902) 485e506.
gastric cancer: importance of the lymphatic route, Virchows Arch. 469 (2)
[14] P.J. Reel, Krukenberg tumor of the ovary, Ann. Surg. 73 (4) (1921) 481e486.
(2016) 155e161.
[15] J.M. Leffel Jr., J.C. Masson, M.B. Dockerty, Krukenberg's tumors: a survey of
[48] J.M. Lloyd, C.M. McIver, S.A. Stephenson, P.J. Hewett, N. Rieger,
forty-four cases, Ann. Surg. 115 (1) (1942) 102.
J.E. Hardingham, Identification of early-stage colorectal cancer patients at risk
[16] J.D. Woodruff, E.R. Novak, The Krukenberg tumor: study of 48 cases from the
of relapse post-resection by immunobead reverse transcription-PCR analysis
ovarian tumor registry, Obstet. Gynecol. 15 (1960) 351e360.
of peritoneal lavage fluid for malignant cells, Clin. cancer Res. Off. J. Am. Assoc.
[17] S.F. Serov, R.E. Scully, L.H. Sobin, Histological Typing of Ovarian Tumours,
Cancer Res. 12 (2) (2006) 417e423.
1973.
[49] B.E. Miller, B. Pittman, J.Y. Wan, M. Fleming, Colon cancer with metastasis to
[18] Y.J. Jeung, H.J. Ok, W.G. Kim, S.H. Kim, T.H. Lee, Krukenberg tumors of gastric
the ovary at time of initial diagnosis, Gynecol. Oncol. 66 (3) (1997) 368e371.
origin versus colorectal origin, Obstet. Gynecol. Sci. 58 (1) (2015) 32e39.
[50] Y.F. Sun, X.R. Yang, J. Zhou, S.J. Qiu, J. Fan, Y. Xu, Circulating tumor cells: ad-
[19] R. Jiang, J. Tang, X. Cheng, R.Y. Zang, Surgical treatment for patients with
vances in detection methods, biological issues, and clinical relevance, J. Cancer
different origins of Krukenberg tumors: outcomes and prognostic factors, Eur.
Res. Clin. Oncol. 137 (8) (2011) 1151e1173.
J. Surg. Oncol. J. Eur. Soc. Surg. Oncol. Br. Assoc. Surg. Oncol. 35 (1) (2009)
[51] S.H. Lim, K.J. Spring, P. de Souza, S. MacKenzie, L. Bokey, Circulating tumour
92e97.
cells and circulating nucleic acids as a measure of tumour dissemination in
[20] W. Peng, R.X. Hua, R. Jiang, et al., Surgical treatment for patients with Kru-
non-metastatic colorectal cancer surgery, Eur. J. Surg. Oncol. J. Eur. Soc. Surg.
kenberg tumor of stomach origin: clinical outcome and prognostic factors
Oncol. Br. Assoc. Surg. Oncol. 41 (3) (2015) 309e314.
analysis, PLoS One 8 (7) (2013) e68227.
[52] H. Takeuchi, Y. Kitagawa, Circulating tumor cells in gastrointestinal cancer,
[21] H.K. Kim, D.S. Heo, Y.-J. Bang, N.K. Kim, Prognostic factors of Krukenberg's
J. Hepatobiliary Pancreat. Sci. 17 (5) (2010) 577e582.
tumor, Gynecol. Oncol. 82 (1) (2001) 105e109.
[53] S. Pradeep, S.W. Kim, S.Y. Wu, et al., Hematogenous metastasis of ovarian
[22] L.C. Lu, Y.Y. Shao, C.H. Hsu, et al., Metastasectomy of Krukenberg tumors may
cancer: rethinking mode of spread, Cancer Cell 26 (1) (2014) 77e91.
be associated with survival benefits in patients with metastatic gastric cancer,
[54] S. Pernot, T. Voron, G. Perkins, C. Lagorce-Pages, A. Berger, J. Taieb, Signet-ring
Anticancer Res. 32 (8) (2012) 3397e3401.
cell carcinoma of the stomach: impact on prognosis and specific therapeutic
[23] F. Rosa, D. Marrelli, P. Morgagni, et al., Krukenberg tumors of gastric origin:
challenge, World J. Gastroenterol.Gastroenterol. 21 (40) (2015) 11428e11438.
the rationale of surgical resection and perioperative treatments in a multi-
[55] H. Miyashita, Studies on gastric carcinoma: the mechanism of its metastasis to
center western experience, World J. Surg. 40 (4) (2016) 921e928.
ovaries, Nihon Ganchiryo Gakukai Shi J. Jpn. Cancer Th) 4 (1969) 469e481.
[24] J.H. Cheong, W.J. Hyung, J. Chen, J. Kim, S.H. Choi, S.H. Noh, Surgical man-
[56] R.M. Gore, G.M. Newmark, K.H. Thakrar, U.K. Mehta, J.W. Berlin, Pathways of
agement and outcome of metachronous Krukenberg tumors from gastric
abdominal tumour spread: the role of the subperitoneal space, Cancer Im-
cancer, J. Surg. Oncol. 87 (1) (2004) 39e45.
aging 9 (2009) 112e120.
[25] G. Piessen, M. Messager, E. Leteurtre, T. Jean-Pierre, C. Mariette, Signet ring
[57] X.F. Yang, L. Yang, X.Y. Mao, D.Y. Wu, S.M. Zhang, Y. Xin, Pathobiological
cell histology is an independent predictor of poor prognosis in gastric
behavior and molecular mechanism of signet ring cell carcinoma and
adenocarcinoma regardless of tumoral clinical presentation, Ann. Surg. 250
mucinous adenocarcinoma of the stomach: a comparative study, World J.
(6) (2009) 878e887.
Gastroenterol. 10 (5) (2004) 750e754.
[26] M.J. Webb, D.G. Decker, E. Mussey, Cancer metastatic to the ovary: factors
[58] M.S. Ur Rahman, J. Cao, Estrogen receptors in gastric cancer: advances and
influencing survival, Obstet. Gynecol. 45 (4) (1975) 391e396.
perspectives, World J. Gastroenterol. 22 (8) (2016) 2475e2482.
[27] N. Yada-Hashimoto, T. Yamamoto, S. Kamiura, et al., Metastatic ovarian tu-
[59] R.H. Young, Ovarian tumors and tumor-like lesions in the first three decades,
mors: a review of 64 cases, Gynecol. Oncol. 89 (2) (2003) 314e317.
Semin. Diagn Pathol. 31 (5) (2014) 382e426.
[28] Y.R. de Waal, C.M. Thomas, A.L. Oei, F.C. Sweep, L.F. Massuger, Secondary
[60] Q. Jiang, S. Liu, H. Xiong, et al., Metastatic signet ring cell carcinoma of the
ovarian malignancies: frequency, origin, and characteristics, Int. J. Gynecol.
bilateral breasts and ovaries from gastrointestinal tract in a young breast-
Cancer 19 (7) (2009) 1160e1165.
feeding female - a case report, J. Cancer 2 (2011) 484e489.
[29] K. Fujiwara, Y. Ohishi, H. Koike, S. Sawada, T. Moriya, I. Kohno, Clinical im-
[61] R.N. Kaplan, R.D. Riba, S. Zacharoulis, et al., VEGFR1-positive haematopoietic
plications of metastases to the ovary, Gynecol. Oncol. 59 (1) (1995) 124e128.
bone marrow progenitors initiate the pre-metastatic niche, Nature 438 (7069)
[30] H. Goldvaser, O. Purim, Y. Kundel, et al., Colorectal cancer in young patients: is
(2005) 820e827.
it a distinct clinical entity? Int. J. Clin. Oncol. 21 (4) (2016) 684e695.
[62] J.P. Sleeman, N. Cremers, New concepts in breast cancer metastasis: tumor
[31] F. Wu, X. Zhao, B. Mi, et al., Clinical characteristics and prognostic analysis of
initiating cells and the microenvironment, Clin. Exp. Metastasis 24 (8) (2007)
Krukenberg tumor, Mol. Clin. Oncol. 3 (6) (2015) 1323e1328.
707e715.
[32] I. Duarte, O. Llanos, Patterns of metastases in intestinal and diffuse types of
[63] F.A. Tavassoli, P. Devilee, Pathology and Genetics of Tumours of the Breast and
carcinoma of the stomach, Hum. Pathol. 12 (3) (1981) 237e242.
Female Genital Organs, IARC, 2003.
[33] P.Y. Tamura, C. Curtiss, Carcinoma of the stomach in the young adult, Cancer
[64] A.C. Testa, G. Ferrandina, D. Timmerman, et al., Imaging in gynecological
13 (1960) 379e385.
disease (1): ultrasound features of metastases in the ovaries differ depending
[34] R. Bernards, R.A. Weinberg, A progression puzzle, Nature 418 (6900) (2002)
on the origin of the primary tumor, Ultrasound Obstet. Gynecol. 29 (5) (2007)
823.
505e511.
[35] C. Fan, D.S. Oh, L. Wessels, et al., Concordance among gene-expression-based
 A. Agnes et al. / Surgical Oncology 26 (2017) 438e445 445

[65] A.C. Testa, A. Licameli, A. Di Legge, et al., Color Doppler sonographic features of (1981) 272e275.
a Krukenberg tumor in pregnancy, J. Ultrasound Med. 28 (5) (2009) 695e698. [82] K. Xu, H. Gao, Z. Lian, L. Ding, M. Li, J. Gu, Clinical analysis of Krukenberg
[66] J.Y. Cho, C.K. Seong, S.H. Kim, Krukenberg tumor findings at color and power tumours in patients with colorectal cancerda review of 57 cases, World J.
Doppler US; correlation with findings at CT, MR imaging, and pathology. Case Surg. Oncol. 15 (1) (2017) 25.
reports, Acta Radiol. 39 (3) (1998) 327e329. [83] C. Sakakura, A. Hagiwara, J. Yamazaki, et al., Management of postoperative
[67] T. Koyama, Y. Mikami, T. Saga, K. Tamai, K. Togashi, Secondary ovarian tumors: follow-up and surgical treatment for Krukenberg tumor from colorectal
spectrum of CT and MR features with pathologic correlation, Abdom. Imaging cancers, Hepato-Gastroenterol. 51 (59) (2003) 1350e1353.
32 (6) (2007) 784e795. [84] S.H. Lee, K.H. Lim, S.Y. Song, et al., Occult gastric cancer with distant metas-
[68] H.K. Ha, S. Baek, S.H. Kim, H. Kim, E. Chung, K. Yeon, Krukenberg's tumor of tasis proven by random gastric biopsy, World J. Gastroenterol. 22 (16) (2016)
the ovary: MR imaging features, AJR Am. J. Roentgenol. 164 (6) (1995) 4270e4274.
1435e1439. [85] J.H. Cheong, W.J. Hyung, J. Chen, J. Kim, S.H. Choi, S.H. Noh, Survival benefit of
[69] L.M. Roth, I. Ramzy, Perspectives on signet ring stromal cell tumor and related metastasectomy for Krukenberg tumors from gastric cancer, Gynecol. Oncol.
signet ring cell lesions of the gonads, Adv. Anatomic Pathol. 21 (6) (2014) 94 (2) (2004) 477e482.
443e449. [86] L.S. McGoogan, K.D. Hatch, Krukenberg tumorereport of two cases, Nebr Med.
[70] S.Y. Jun, J.K. Park, Metachronous ovarian metastases following resection of the J. 57 (10) (1972) 409e415.
primary gastric cancer, J. Gastric Cancer 11 (1) (2011) 31e37. [87] T. Gauthier, C. Uzan, D. Lefeuvre, et al., Lymphocele and ovarian cancer: risk
[71] R. Omranipour, A. Abasahl, Ovarian metastases in colorectal cancer, Int. J. factors and impact on survival, Oncologist 17 (9) (2012) 1198e1203.
Gynecol. Cancer 19 (9) (2009) 1524e1528. [88] F. McGill, D.B. Ritter, C. Rickard, R.N. Kaleya, S. Wadler, W.M. Greston, Man-
[72] A. Ayhan, T. Guvenal, M.C. Salman, O. Ozyuncu, M. Sakinci, M. Basaran, The agement of Krukenberg tumors: an 11-year experience and review of the
role of cytoreductive surgery in nongenital cancers metastatic to the ovaries, literature, Prim. Care Update Ob. Gyns 5 (4) (1998) 157e158.
Gynecol. Oncol. 98 (2) (2005) 235e241. [89] Susan D Reed BG. Elective oophorectomy or ovarian conservation at the time
[73] Y. Gagnon, B. Tetu, Ovarian metastases of breast carcinoma. A clinicopatho- of hysterectomy. In: UpToDate, Post TW (Ed), UpToDate, Waltham, MA (Accessed
logic study of 59 cases, Cancer 64 (4) (1989) 892e898. on 20 January 20 2016).
[74] J. Wang, Y.K. Shi, L.Y. Wu, et al., Prognostic factors for ovarian metastases from [90] R. Thota, X. Fang, S. Subbiah, Clinicopathological features and survival out-
primary gastric cancer, Int. J. Gynecol. Cancer 18 (4) (2008) 825e832. comes of primary signet ring cell and mucinous adenocarcinoma of colon:
[75] J.H. Cho, J.Y. Lim, A.R. Choi, et al., Comparison of surgery plus chemotherapy retrospective analysis of VACCR database, J. Gastrointest. Oncol. 5 (1) (2014)
and palliative chemotherapy alone for advanced gastric cancer with kruken- 18e24.
berg tumor, Cancer Res. Treat. 47 (4) (2015) 697e705. [91] X. Wu, Z. Zhang, X. Li, et al., Poorer prognosis of primary signet-ring cell
[76] X.J. Wu, P. Yuan, Z.Y. Li, et al., Cytoreductive surgery and hyperthermic carcinoma of the breast compared with mucinous carcinoma, PLoS One 11 (9)
intraperitoneal chemotherapy improves the survival of gastric cancer patients (2016) e0162088.
with ovarian metastasis and peritoneal dissemination, Tumour Biol. 34 (1) [92] J. Ribeiro Gomes, M. Belotto, R. D'Alpino Peixoto, The role of surgery for un-
(2013) 463e469. usual sites of metastases from colorectal cancer: a review of the literature,
[77] D. Rayson, E. Bouttell, F. Whiston, L. Stitt, Outcome after ovarian/adnexal Eur. J. Surg. Oncol. J. Eur. Soc. Surg. Oncol. Br. Assoc. Surg. Oncol. 43 (1) (2017)
metastectomy in metastatic colorectal carcinoma, J. Surg. Oncol. 75 (3) (2000) 15e19.
186e192. [93] E. Papadopoulou, V. Metaxa-Mariatou, G. Tsaousis, et al., Molecular predictive
[78] M. Morrow, W.E. Enker, Late ovarian metastases in carcinoma of the colon and markers in tumors of the gastrointestinal tract, World J. Gastrointest. Oncol. 8
rectum, Arch. Surg. 119 (12) (1984) 1385e1388. (11) (2016) 772e785.
[79] K. Ganesh, R.H. Shah, E. Vakiani, et al., Clinical and genetic determinants of [94] J. Verigos, A. Magklara, Revealing the complexity of breast cancer by next
ovarian metastases from colorectal cancer, Cancer (2016). generation sequencing, Cancers (Basel) 7 (4) (2015) 2183e2200.
[80] A. Fujiwara, S. Noura, M. Ohue, et al., Significance of the resection of ovarian [95] M. Tsujiura, D. Ichikawa, H. Konishi, S. Komatsu, A. Shiozaki, E. Otsuji, Liquid
metastasis from colorectal cancers, J. Surg. Oncol. 102 (6) (2010) 582e587. biopsy of gastric cancer patients: circulating tumor cells and cell-free nucleic
[81] S.L. Blamey, F.T. McDermott, E. Pihl, E.S. Hughes, Resected ovarian recurrence acids, World J. Gastroenterol. 20 (12) (2014) 3265e3286.
from colorectal adenocarcinoma: a study of 13 cases, Dis. Colon Rectum 24 (4)