Vittoria Castiglioni1, Enrico Radaelli2,3. 1. IDEXX Laboratories Italia S.r.l., Via Guglielmo Silva 36, Milano 20149, Italy. 2. Comparative Pathology Core, Department of Pathobiology, School of Veterinary Medicine, University of Pennsylvania, 3900 Delancey Street, Philadelphia, PA 19104-6051, U.S.A. 3. VIB Center for the Biology of Disease and KU Leuven Center for Human Genetics, Herestraat 49, Leuven 3000, Belgium.
Abstract
Murine hepatoblastoma (HB) is a rare spontaneous tumor with controversial histogenesis. It mainly occurs in aged males, frequently in close association with preexisting hepatocellular neoplasms. The present work describes a spontaneous HB arising within a hepatocellular carcinoma (HCC) in a 22-month-old male C57BL/6J mouse. The mouse also developed pulmonary co-metastases with either tumor components physically associated within the same metastatic foci. Microscopically, the HB consisted of a densely cellular neoplastic growth composed of palisades and perivascular pseudorosettes of poorly differentiated primitive cells, with a scant amount of cytoplasm, elongated hyperchromatic nuclei, and a high mitotic rate, whereas the hepatocellular carcinoma was composed of solid areas of neoplastic hepatocytes. Both in primary tumors and their metastases, β-catenin immunohistochemistry revealed a strong nucleocytoplasmic signal in HB cells, while neoplastic hepatocytes displayed a delicate membranous staining pattern. These findings suggest that the Wnt/β-catenin oncogenic pathway is upregulated in murine HB but not in the co-existing HCC, thus providing some insights into their divergent pathogenesis. Coexisting murine HB and HCC have been demonstrated to be completely distinct entities including origin, mutational landscape, and molecular profile. In this context, they might be regarded as collision tumors because of their intimate association, unique histologic features, and distinct immunohistochemical patterns. Nevertheless, the nature of their coevolution and progression to a co-metastatic phenotype reflects a close interdependence and support the overall idea that HB's origin and progression might be promoted by not otherwise specified paracrine stimuli provided by the concurrent hepatocellular tumor (the so called "interaction theory").
Murinehepatoblastoma (HB) is a rare spontaneous tumor with controversial histogenesis. It mainly occurs in aged males, frequently in close association with preexisting hepatocellular neoplasms. The present work describes a spontaneous HB arising within a hepatocellular carcinoma (HCC) in a 22-month-old male C57BL/6J mouse. The mouse also developed pulmonary co-metastases with either tumor components physically associated within the same metastatic foci. Microscopically, the HB consisted of a densely cellular neoplastic growth composed of palisades and perivascular pseudorosettes of poorly differentiated primitive cells, with a scant amount of cytoplasm, elongated hyperchromatic nuclei, and a high mitotic rate, whereas the hepatocellular carcinoma was composed of solid areas of neoplastic hepatocytes. Both in primary tumors and their metastases, β-catenin immunohistochemistry revealed a strong nucleocytoplasmic signal in HB cells, while neoplastic hepatocytes displayed a delicate membranous staining pattern. These findings suggest that the Wnt/β-catenin oncogenic pathway is upregulated in murine HB but not in the co-existing HCC, thus providing some insights into their divergent pathogenesis. Coexisting murine HB and HCC have been demonstrated to be completely distinct entities including origin, mutational landscape, and molecular profile. In this context, they might be regarded as collision tumors because of their intimate association, unique histologic features, and distinct immunohistochemical patterns. Nevertheless, the nature of their coevolution and progression to a co-metastatic phenotype reflects a close interdependence and support the overall idea that HB's origin and progression might be promoted by not otherwise specified paracrine stimuli provided by the concurrent hepatocellular tumor (the so called "interaction theory").
Hepatoblastoma (HB) is a primitive, poorly differentiated, embryonal hepatic neoplasm. In
mice, HBs are rare spontaneous entities that mainly occur in aged male animals[1] and are usually seen within or adjacent to
preexisting hepatocellular adenomas (HCAs) or carcinomas (HCCs)[2], [3].
The cell of origin of HBs remains unknown, although liver blastemal cells, neoplastic
hepatocytes, oval cells, and biliary epithelial cells have been proposed[3]. Previously considered to be the result of
malignant progression of benign or low-grade hepatocellular tumors, it has been recently
documented that murinehepatoblastoma most likely represents a distinct entity displaying a
completely different mutational spectrum and transcriptome if compared with hepatocellular
tumors[4].In humans, HBs are mainly observed in young children below the age of 5 years and are
exceptionally rare in adults[5],
[6]. In contrast to their murine
counterpart, human HBs are always solitary tumors unassociated with preexisting hepatocellular
neoplasias and are thought to arise de novo from hepatoblast remnants of the
embryonic liver[2], [7], [8]. On the other hand, HBs in both mice and humans share similarities, such
as primitive morphologic features, predominance in the male gender, and a comparable
metastatic rate with lungs as the most common metastatic site[2], [9].
Furthermore, genetic alterations have been associated with this malignancy, with human and
mouse HBs both harboring oncogenic mutations in Ctnnb1[10], [11] leading to constitutive activation of the Wnt/β-catenin pathway.Considering nonhuman species, hepatoblastomas have been described in equine fetuses,
neonates, and occasionally young adult horses [12], [13],
[14], in young and adult
sheep[15], and in a cat[16], llama[17] and dog[18]. In all these
species, HBs have been mainly reported in young animals, paralleling the human counterpart,
with only a minority of cases occurring in adults. With reference to laboratory animals, there
are rare reports of this lesion in rats[3]. A
single case was described in a 10-day-preterm fetus of a male cynomolgus macaque
(Macaca fascicularis)[19].This study aimed to describe pathological features of a spontaneous murine condition
characterized by the co-occurrence of HB and HCC and development of pulmonary co-metastases
consisting of closely associated neoplastic cells from both the hepatic cancers.The animal considered in this study was an experimentally naïve 22-month-old male C57BL/6J
mouse included as control in an aging study. The mouse was sacrificed because of generalized
deterioration of clinical conditions associated with marked abdominal enlargement and dyspnea.
A complete necropsy and histopathological examination were performed following the same
protocols and procedures as described by Castiglioni et al.[20]. For immunohistochemistry (IHC), 4 µm
formalin-fixed paraffin-embedded sections were dewaxed and rehydrated. Endogenous peroxidase
was blocked by incubating sections in 3% H2O2 for 15 min. For antigen
retrieval, sections were immersed in citrate buffer (pH 6.0), heated, and cooled at room
temperature. β-catenin (rabbit monoclonal, clone 6B3, dilution 1:150, Cell Signaling) specific
antibody was applied for 1.5 hours at room temperature. The reaction was amplified with an
avidin-biotin method (Vectastain Elite ABC Kit PK-6100, Vector Laboratories Inc., Burlingame,
CA, USA) and visualized with 3,3′-diaminobenzidine. Sections were counterstained with
hematoxylin. A negative immunohistochemical control was prepared by replacing the primary
antibody with an irrelevant one, and a known positive control section was included in the
immunolabeling assay. Procedures involving animals were performed in accordance with the
guidelines of the Catholic University of Leuven (KU Leuven) Animal Care and Use Ethical
Committee, which specifically approved this study (reference number 072/2015).At necropsy, ascites was detected, and the entire left hepatic lobe and part of the median
were markedly enlarged by a large, mottled gray-brown multinodular mass measuring
approximately 1.5 × 1.5 × 1 cm. The lungs were characterized by multiple small brownish
nodules (measuring approximately 1 to 3 mm in diameter) that were particularly evident along
the margins of the lobes.Histologically, the affected hepatic parenchyma was almost completely effaced by two distinct
neoplastic growths (Fig. 1). The first one was characterized by multiple
coalescing HCAs consisting of expansile masses that were moderately cellular, unencapsulated,
relatively well demarcated and composed of well-differentiated neoplastic hepatocytes arranged
in cords or thin trabeculae separated by a delicate sinusoidal vascular network. Scattered
throughout the HCAs there were foci of increased atypia with higher N/C ratios, hyperchromatic
nuclei, with neoplastic hepatocytes arranged in thicker trabeculae (more than 3 cells in
thickness) separated by irregularly distended sinusoids often forming pseudocystic spaces, and
with foci of coagulative necrosis. These were interpreted as foci of progression suggesting a
morphological diagnosis of trabecular type HCC arising within an HCA (Fig. 2). The
second neoplastic entity arose within the previously described hepatocellular tumor as a
multicentric process. These growths were infiltrating, unencapsulated, and composed of dense
palisades and perivascular pseudorosettes of poorly differentiated primitive cells, with a
scant amount of cytoplasm, elongated hyperchromatic nuclei, and a high mitotic rate (Fig. 3). Scattered throughout there were foci of coagulative necrosis and hemorrhages and
few foci of squamous differentiation (Fig. 3,
inset). Morphologically, the tumor was consistent with a diagnosis of HB. Histopathology of
the pulmonary lesions revealed disseminated neoplastic aggregates multifocally plugging small
and mid-sized pulmonary vessels (tumor emboli) and invading the surrounding parenchyma (tumormetastases). Pulmonary lesions recapitulated the two different neoplastic growths observed
within the liver, with neoplastic hepatocytes closely associated with primitive, poorly
differentiated neoplastic cells (interpreted as pulmonary metastases of HB arising within an
HCC) (Fig. 4).
Fig.
1.
Liver, 22-month-old male C57BL/6J mouse. The hepatic parenchyma is
expanded and effaced by two different neoplastic growths that are in close proximity: a
hepatocellular carcinoma (black asterisk) and a hepatoblastoma (white asterisk).
Hematoxylin and eosin. Scale bar: 2 mm.
Fig. 2.
Liver, 22-month-old male C57BL/6J mouse.
Hepatocellular carcinoma arising within a hepatocellular adenoma. Note the carcinomatous
focus (arrows) characterized by increased trabecular thickness and findings of atypia in
tumor cells including a higher N/C ratio and hyperchromasia. Hematoxylin and eosin.
Scale bar: 100 µm.
Fig. 3.
Liver, 22-month-old male
C57BL/6J mouse. The hepatoblastoma consists of poorly differentiated primitive cells
with elongated hyperchromatic nuclei, scant cytoplasm, and a high mitotic rate. Tumor
cells are typically arranged in dense palisades and perivascular pseudorosettes. Small
scattered foci of squamous differentiation (inset) are also evident. Hematoxylin and
eosin. Scale bar: 100 µm.
Fig. 4.
Lung,
22-month-old male C57BL/6J mouse. Metastatic focus consisting of combined hepatoblastoma
and hepatocellular carcinoma. Hematoxylin and eosin. Scale bar: 50
µm.
Liver, 22-month-old male C57BL/6J mouse. The hepatic parenchyma is
expanded and effaced by two different neoplastic growths that are in close proximity: a
hepatocellular carcinoma (black asterisk) and a hepatoblastoma (white asterisk).
Hematoxylin and eosin. Scale bar: 2 mm.Liver, 22-month-old male C57BL/6J mouse.
Hepatocellular carcinoma arising within a hepatocellular adenoma. Note the carcinomatous
focus (arrows) characterized by increased trabecular thickness and findings of atypia in
tumor cells including a higher N/C ratio and hyperchromasia. Hematoxylin and eosin.
Scale bar: 100 µm.Liver, 22-month-old male
C57BL/6J mouse. The hepatoblastoma consists of poorly differentiated primitive cells
with elongated hyperchromatic nuclei, scant cytoplasm, and a high mitotic rate. Tumor
cells are typically arranged in dense palisades and perivascular pseudorosettes. Small
scattered foci of squamous differentiation (inset) are also evident. Hematoxylin and
eosin. Scale bar: 100 µm.Lung,
22-month-old male C57BL/6J mouse. Metastatic focus consisting of combined hepatoblastoma
and hepatocellular carcinoma. Hematoxylin and eosin. Scale bar: 50
µm.Immunohistochemistry of β-catenin revealed that neoplastic hepatocytes in both hepatic and
pulmonary locations showed mild to moderate membranous immunoreactivity, whereas a strong
nucleocytoplasmic signal was evident in HB tumor cells (Fig. 5
and 6).
Fig. 5.
Liver, 22-month-old male C57BL/6J mouse.
Hepatoblastoma (left) and adjacent hepatocellular adenoma (right). β-catenin
immunohistochemistry reveals a strong nucleocytoplasmic signal in hepatoblastoma tumor
cells, while neoplastic hepatocytes in the surrounding adenoma display a delicate
membranous staining pattern. β-catenin immunohistochemistry (IHC); hematoxylin
counterstain. Scale bar: 100 µm.
Fig. 6.
Lung,
22-month-old male C57BL/6J mouse. The same metastatic focus is depicted as in Fig. 4. β-catenin immunohistochemistry reveals a
strong nucleocytoplasmic signal in the hepatoblastoma fraction of the metastatic lesion,
while the hepatocarcinomatous component maintains a delicate membranous staining
pattern. β-catenin immunohistochemistry (IHC); hematoxylin counterstain. Scale bar: 50
µm.
Liver, 22-month-old male C57BL/6J mouse.
Hepatoblastoma (left) and adjacent hepatocellular adenoma (right). β-catenin
immunohistochemistry reveals a strong nucleocytoplasmic signal in hepatoblastoma tumor
cells, while neoplastic hepatocytes in the surrounding adenoma display a delicate
membranous staining pattern. β-catenin immunohistochemistry (IHC); hematoxylin
counterstain. Scale bar: 100 µm.Lung,
22-month-old male C57BL/6J mouse. The same metastatic focus is depicted as in Fig. 4. β-catenin immunohistochemistry reveals a
strong nucleocytoplasmic signal in the hepatoblastoma fraction of the metastatic lesion,
while the hepatocarcinomatous component maintains a delicate membranous staining
pattern. β-catenin immunohistochemistry (IHC); hematoxylin counterstain. Scale bar: 50
µm.The complex hepatic changes observed in this aged C57BL/6J mouse recapitulate the
well-characterized scenario in which HB is usually seen within or adjacent to preexisting
hepatocellular tumors[1]. The presence of areas
of HCC within HCA supports the overall idea that hepatocellular carcinogenesis in the mouse
often follows the preneoplastic pathway, with preneoplastic lesions evolving into adenomas and
later on into foci of carcinomatous progression arising within adenomas (hence the diagnostic
designation of HCC arising within HCA)[3]. The
hepatoblastoma here examined showed foci of squamous differentiation, in agreement with
current literature that frequently reports small areas of bone and squamous differentiation
along with foci of extramedullary hematopoiesis[3]. Both murine HCC and HB exhibit metastatic potential, with the lung being
the most frequently involved distant organ[2].
To the authors’ knowledge, evidence of pulmonary co-metastases has been reported only once
prior to this report[2].The different patterns of β-catenin expression observed in both hepatic and pulmonary
locations, with a strong nucleocytoplasmic signal in HB and delicate membranous
immunoreactivity displayed by the hepatocellular proliferations, are in line with the
well-established notion that the Wnt/β-catenin oncogenic pathway drives murine HB but not HCC
tumorigenesis[4]. The divergent oncogenic
mechanisms activated by these two tumors further support the hypothesis that HB does not
progress from the concurrent hepatocellular tumor and that it should rather be regarded as a
separate neoplasm. Comparative transcriptomic analysis demonstrated a high degree of
concordance among murine HB, embryonic mouse hepatoblasts, and hepatic pluripotent stem cells,
suggesting that these poorly differentiated and primitive tumors likely originate from a
primordial hepatic precursor cell[4],
[21]. Recent studies also showed
that coexisting HCC and HB do not share the same mutational spectrum, thus providing
compelling evidence that they are indeed distinct tumor entities[4], [10].
In light of these considerations, we propose to consider the entity here described as a
collision tumor, that is defined as two distinct neoplasms with diverse origins and phenotypes
coinciding at the same location[22].
Furthermore, as suggested by the frequent formation of HB within (or near) a well-developed
hepatocellular neoplasm, the molecular events that lead to hepatic precursor cell
transformation and HB formation are facilitated by not otherwise specified paracrine stimuli
provided by the preexisting neoplastic hepatocytes. In this context, we support the idea that
the microenvironment created by the hepatocellular tumors promotes the development of HB via a
still unclear paracrine mechanism. Such a type of interaction, known as the “interaction
theory”, has already been proposed for a series of co-developing neoplasms in humans[23], [24]. In this context, the evidence of co-metastases, in which both tumor cell
types are physically associated within the same metastatic focus, possibly indicates that the
close biological interdependence between HB and HCC also contributes to tumor progression and
combined metastatic spread.In conclusion, to the authors’ knowledge, this is the first report characterizing
immunohistochemically both hepatic and pulmonary co-metastasis and gives potentially
interesting insights into the pathogenesis and coevolution of these frequently associated
cancers, raising attention concerning whether they might be regarded as collision tumors,
although they have not yet been classified as such academically.Disclosure of Potential Conflicts of Interest: The authors declare that they
have no potential conflicts of interest with respect to the research, authorship, and/or
publication of this article. The authors disclose that they received no financial support for
the research, authorship, and/or publication of this article.
Authors: Laurence de Leval; Noëlle Hardy; Manuel Deprez; Jean Delwaide; Jacques Belaïche; Jacques Boniver Journal: Virchows Arch Date: 2002-08-29 Impact factor: 4.064
Authors: Bob Thoolen; Robert R Maronpot; Takanori Harada; Abraham Nyska; Colin Rousseaux; Thomas Nolte; David E Malarkey; Wolfgang Kaufmann; Karin Küttler; Ulrich Deschl; Dai Nakae; Richard Gregson; Michael P Vinlove; Amy E Brix; Bhanu Singh; Fiorella Belpoggi; Jerrold M Ward Journal: Toxicol Pathol Date: 2010-12 Impact factor: 1.902
Authors: Sachin Bhusari; Arun R Pandiri; Hiroaki Nagai; Yu Wang; Julie Foley; Hue-Hua L Hong; Thai-Vu Ton; Michael DeVito; Keith R Shockley; Shyamal D Peddada; Kevin E Gerrish; David E Malarkey; Michelle J Hooth; Robert C Sills; Mark J Hoenerhoff Journal: Toxicol Pathol Date: 2015-08-18 Impact factor: 1.902