Vet Pathol 42:1, 2005

Brief Communications and Case Reports

77

Vet Pathol 42:77–81 (2005)

Invasive Epithelial Mesothelioma in a Dog
F. REGGETI, B. BRISSON, K. RUOTSALO, E. SOUTHORN,
AND

D. BIENZLE

Abstract. This report describes the gross, microscopic, and immunohistochemical features of an invasive epithelial mesothelioma in an 11-year-old neutered male Golden Retriever. The tumor involved the pericardium, pleura, mediastinum, and peritoneum and invaded into submesothelial tissues. Neoplastic cells in the thoracic fluid showed prominent features of malignancy in a background of mixed inflammatory cells and scattered erythrocytes. Histologically, the tumor consisted of nests of epithelioid cells with frequent mitotic figures and multinucleation that infiltrated submesothelial tissues. Neoplastic cells strongly coexpressed vimentin and cytokeratin intermediate filaments, which assisted in the differentiation from other epithelial tumors of nonmesothelial origin. Key words: Canine; cytokeratin; effusion; immunohistochemistry; mesothelioma; vimentin. study, we describe the macroscopic, cytologic, histopathologic, and immunohistochemical findings of an invasive epithelioid mesothelioma in a dog with pericardial, pleural, and peritoneal effusions. An 11-year-old neutered male Golden Retriever was referred to the Veterinary Teaching Hospital, University of Guelph, Ontario, Canada, with a 2-week history of collapse, exercise intolerance, and pericardial effusion. At presentation, thoracic radiographs and echocardiography confirmed the effusion. Masses were not noted, and about 400 ml of serosanguinous fluid was removed from the pericardial space. Cytologic assessment of the fluid suggested an exfoliating neoplasm, and exploratory surgery was recommended. The dog was readmitted 3 months later with similar clinical signs and pleural and peritoneal effusion. A complete blood cell count showed mild anemia (hematocrit 0.38 liter/ liter, reference limits 0.39–0.56 liter/liter), neutrophilia (10.9

Mesothelioma is a rare neoplasm arising from the lining cells of the peritoneal, pleural, and pericardial cavities or the tunica vaginalis of the testis.13 Spontaneous mesothelioma has been reported in humans as well as in many species of animals, including dogs, cattle, goats, horses, rats, and hamsters,7 but is most common in cattle, where it may be congenital.8 Mesothelioma is associated with exposure to asbestos in humans3,6–10,12,13 and potentially in dogs.3,13 This association is not restricted to pleural mesotheliomas because peritoneal mesotheliomas also occur in humans exposed to asbestos. The tumor has been experimentally induced in rats and hamsters by inhalation or injection of asbestos, glass fibers, and aluminum oxide.7 Mesothelioma has also been induced in rats by intracavitary injections of the simian virus 40 (SV40).10 There are distinct histologic patterns: epithelioid (resembling carcinoma), sarcomatoid or sclerosing (resembling fibrosarcoma), and biphasic mesothelioma.5 In this

78

Brief Communications and Case Reports

Vet Pathol 42:1, 2005

Fig. 1. Thoracic cavity; canine. Thoracoscopy showing mediastinal masses, ulceration of the parietal pleura, and a small amount of serosanguinous fluid. No lesions are apparent in the lungs. Fig. 2. Thoracic fluid; canine. Direct smear of thoracic fluid showing clusters of hyperchromatic cohesive cells in a background of inflammatory cells and red blood cells. Occasional erythrocytophagia is evident. Wright-Giemsa stain. Bar 100 m. Fig. 3. Thoracic wall; canine. Nests of neoplastic cells are present within the pleura (arrows) and infiltrate the adjacent muscle fibers (arrowheads). HE. Bar 100 m. Fig. 4. Thoracic wall; canine. Clusters of neoplastic cells show marked immunoreactivity for pancytokeratin. Immunoperoxidase stain, hematoxylin counterstain. Bar 20 m.

Vet Pathol 42:1, 2005

Brief Communications and Case Reports

79

109/liter, reference limits 2.9–10.6 109/liter), monocy109/liter, reference limits 0.0–1.1 109/liter), tosis (1.5 9/liter, reference limits 117– 10 and thrombocytosis (467 109/liter). Changes in the chemistry profile included 418 hyponatremia (129 mmol/liter, reference limits 140–154 mmol/liter), hypochloremia (99 mmol/liter, reference limits 104–119 mmol/liter), hyperkalemia (7.0 mmol/liter, reference limits 3.8–5.4 mmol/liter), hyperphosphatemia (2.37 mmol/liter, reference limits 0.90–1.85 mmol/liter), and hypermagnesemia (1.3 mmol/liter, reference limits 0.7–1.0 mmol/liter). Abdominal ultrasound was unremarkable. Thoracoscopy was also performed and approximately 5 liters of serosanguinous fluid was removed from the pleural space. The mediastinum showed multiple white to tan protruding nodules, and the parietal pleura was irregular, with focal areas of redness and ulceration. No abnormalities were observed in the lungs (Fig. 1). A pericardial ‘‘window’’ incision was performed. Thoracic fluid was submitted for cytologic examination, and bacterial culture and biopsy samples from the thoracic wall, mediastinum, and pericardium were submitted for histopathology. Direct smears of the thoracic fluid were stained with Wright-Giemsa for cytologic examination. Formalin-fixed, paraffin-embedded sections (5 m) from samples of pericardium, parietal pleura, and mediastinum were stained with hematoxylin and eosin for routine histopathology. For immunohistochemistry, deparaffinized sections were treated with hydrogen peroxide and heated in retrieval solution (Target Retrieval Solution, DakoCytomation, Mississauga, Ontario, Canada). Sections were then incubated with antibodies against pancytokeratin (clone AE1/AE3, 1 : 100) and vimentin (clone V9, 1 : 200), followed by peroxidase-conjugated antibodies (EnVision Labelled Polymer, peroxidase; all from DakoCytomation). After exposure to an appropriate chromogen (Vector NovaRed, Vector Laboratories, Burlington, Ontario, Canada), the slides were counterstained with hematoxylin. The thoracic fluid was dark yellow and turbid. The nucle109/liter (reference limits 0.0– ated cell count was 84.7 9/liter), and the protein concentration was 30 g/liter 3.0 10 (reference limits 10–25 g/liter). Direct smears showed clusters of cohesive cells in a background of mixed inflammatory cells and scattered erythrocytes (Fig. 2). There was marked anisocytosis. Cells were hyperchromatic, and cytoplasmic intercellular borders were distinct. The nuclear to cytoplasmic ratio was high, but cells with abundant vacuolated clear cytoplasm with ‘‘signet ring’’ appearance were also present. Some cells had short cytoplasmic projections resembling those observed in normal mesothelial cells. There was up to sixfold anisokaryosis, and nuclei were round with coarse chromatin, containing single to multiple prominent round to angular nucleoli of different sizes. Ab-

normal mitotic figures were frequent. Numerous neutrophils were present within the clusters of neoplastic epithelial cells, and erythrocytophagia by macrophages was evident. No organisms were noted on smears or culture of the thoracic fluid. The cytologic interpretation was exfoliating carcinoma. Biopsies from the thoracic wall included pleura and underlying muscle. Microscopically, the parietal pleura was thickened by granulation tissue, and there was focal mesothelial hypertrophy. Within the granulation tissue and infiltrating into muscle fibers were nests of epithelioid cells with distinct cell borders, abundant pale eosinophilic cytoplasm, and round, centrally located nuclei with one to several prominent nucleoli. There was marked anisocytosis and anisokaryosis with frequent multinucleation (Fig. 3). Samples from the mediastinum consisted of solid sheets of neoplastic epithelioid cells with an average of two mitotic figures per high-power field (400 ). There were multiple foci of suppurative inflammation (not shown). Immunohistochemistry showed strong expression of both pancytokeratin and vimentin by tumor cells (Figs. 4, 5). The histologic diagnosis was invasive mesothelioma. Thoracocentesis was periodically performed to alleviate clinical signs in the dog, and chemotherapy with intrathoracic carboplatin (300 mg/m2) every 4 weeks was instituted. Nevertheless, the pleural effusion persisted, and 6 months after the initial diagnosis the dog was euthanatized. Necropsy was performed, and tissues from the abdominal wall showed prominent cords of neoplastic epithelioid cells surrounded by abundant mature collagenous fibrous tissues (Fig. 6). These findings confirmed spread of the tumor into the peritoneal cavity. Mesothelial cells are undifferentiated cells of mesodermal origin that line the coelomic cavities as a monolayer. Unique features of these cells are constitutive high expression of the tumor suppressor gene p53,1,12 acquisition of epithelial or fibroblast morphology with specific inflammatory stimuli, and phagocytic potential. Malignant mesothelial tumors in humans are aggressive and highly associated with exposure to asbestos, a group of silicate minerals with different physical and carcinogenic properties.1 Crysotyle fibers are the most common type of asbestos but have been questioned as carcinogens. In contrast, amphibole fibers include crocidolite, the most oncogenic form of asbestos. Inhaled amphibole fibers are distributed throughout the pleura by way of lymphatics and may contribute to transformation of mesothelial cells by mechanical disruption of the cellular spindle apparatus during mitosis, leading to chromosomal instability, missegregation, and aneuploidy.6 This may account for the loss of chromosome 22, described by some authors as the most common numerical cytogenetic abnormality in humans.9 Other genetic lesions include mutations that are associated with loss or inactivation of tumor suppressor genes.

← Fig. 5. Thoracic wall; canine. Neoplastic cells stain strongly for vimentin. Immunoperoxidase stain, hematoxylin counterstain. Bar 20 m. Fig. 6. Abdominal wall; canine. Cords of neoplastic epithelioid cells (E) surrounded by collagenous fibrous tissues (F). HE. Bar 100 m. Inset: Neoplastic cells show marked anisokaryosis, prominent nucleoli, and mitosis (arrow). HE. Bar 50 m.

80

Brief Communications and Case Reports

Vet Pathol 42:1, 2005

Induction of the proto-oncogenes c-fos and c-jun and DNA damage due to production of reactive oxygen species likely further contribute to loss of proliferative regulation in mesothelial cells and to additional mutations.9 Accumulation of such genetic changes may account for the long latency period of 20–40 years observed in humans,1,9 indicating that transformation of mesothelium is likely a gradual multistep process. Widespread introduction of SV40 into the human population through contaminated poliovirus vaccine has recently been acknowledged as a likely contributor to mesothelioma.10,12 The virus is present in an episomal location in 40–50% of human mesotheliomas in the USA, and mesothelial cells are highly susceptible to in vitro transformation after infection. SV40 viral antigens, in particular the ‘‘large T-antigen,’’ bind and inactivate p53 and members of the retinoblastoma tumor suppressor family, thus rendering the cells more susceptible to transformation.1 Mesothelioma in humans is difficult to treat and typically has a rapidly progressive disease course. In the USA, 2000 new human cases are reported every year,1,9,12 providing great impetus for development of more effective therapies. Mesothelioma in dogs is rare, and understanding the natural history of the tumor is limited by few detailed reports. The anatomic distribution and clinical features of canine mesotheliomas are similar to those in humans and the dog in this report.13 Thoracoscopy permitted observation of the tumor and access to adequate biopsy samples while reducing the morbidity typically associated with a full thoracotomy. Examination of fluid from effusions typically yields a suggestive but not definitive diagnosis. Cytologically, mesothelial neoplasia is not readily distinguished from epithelial neoplasia or marked mesothelial hyperplasia.13 The dysplastic features of cells in this case excluded mesothelial hyperplasia but did not allow distinction from a primary pulmonary carcinoma or a metastatic carcinoma. The histologic diagnosis of mesothelioma is commonly based on detection of abnormal multilayered polygonal to spindle-shaped cells that are invading submesothelial tissues. Histopathology in this case showed cords of epithelioid cells infiltrating from the pleura into adjacent muscle, which was convincing of mesothelioma, but did not rule out carcinoma of nonmesothelial origin. Normal mesothelial cells coexpress cytokeratin and vimentin, which may allow differentiation from other epithelial tumors.10 In this case, tumor cells strongly coexpressed both cytokeratin and vimentin intermediate filaments, thus supporting the diagnosis of mesothelioma. Cytokeratin and vimentin are also coexpressed in other tumors including some anaplastic carcinomas, amelanotic melanomas, renal carcinomas, and Sertoli cell tumors, but there was no evidence of any other potential primary masses in this patient. In humans, detection of the calcium-binding protein calretinin in tumor tissue may be diagnostically useful,11 but in a canine sclerosing peritoneal mesothelioma,2 this was not found to be a useful marker. The histologic appearance of mesothelioma may range from epithelioid to sarcomatoid, necessitating in some cases immunohistochemistry or electron microscopy to distinguish reactive mesothelium from lung or metastatic adenocarcinomas, sarcomas, or melanomas. Association of mesothelioma in dogs with asbestos has

been suggested from historical exposure of owners3 and from detection of fibers in the lungs of affected animals.3,4 However, identification of asbestos fibers in lung tissues from unaffected dogs suggests that caution should be taken when interpreting such results. The contribution of asbestos to canine mesothelioma is circumstantial at this time, and the interaction of different asbestos fibers with canine mesothelial cells is unknown. Possibly other viral or chemical carcinogenic agents are important in the development of this malignancy in dogs. Therapeutic options for mesothelioma include surgical resection, radiotherapy, and chemotherapy. Because of the distribution of the lesions and tissue toxicity, chemotherapy is usually the treatment of choice, but no single agent or combination of drugs has yielded satisfactory results in people. This dog was treated with intrathoracic carboplatin, which yielded a limited response. This may reflect poor efficacy of carboplatin in the treatment of mesothelioma or the aggressive nature of the tumor. Drug combinations might also be useful in dogs, but the low incidence of mesothelioma in this species renders objective assessment of specific therapies challenging. Experimental approaches proposed for mesothelioma treatment in humans include signal transmission blockade of the vascular endothelial growth factor receptor, platelet-derived growth factor receptor and epidermal growth factor receptor because these molecules are overexpressed by neoplastic mesothelial cells.12 These therapies are at the experimental stage at this point.

Acknowledgements
We thank Dr. Robert Foster for his critical review.

References
1 Carbone M, Kratzke RA, Test JR: The pathogenesis of mesothelioma. Semin Oncol 29:2–17, 2002 2 Geninet C, Bernex F, Rakotovao F, Crespeau FL, Parodi AL, Fontaine JJ: Sclerosing peritoneal mesothelioma in a dog—a case report. J Vet Med A 50:402–405, 2003 3 Glickman LT, Domanski LM, Maguire TG, Dubielzig RR, Churg A: Mesothelioma in pet dogs associated with exposure of their owners to asbestos. Environ Res 32: 305–313, 1983 4 Harbison ML, Godleski JJ: Malignant mesothelioma in urban dogs. Vet Pathol 20:531–540, 1983 5 Head KW, Else RW, Dubielzig RR: Tumors of the alimentary tract. In: Tumors in Domestic Animals, ed. Meuten DJ, 4th ed., pp. 477–478. Iowa State Press, Ames, IA, 2002 6 Hesterberg TW, Barrett JC: Induction by asbestos fibers of anaphase abnormalities: mechanism for aneuploidy induction and possibly carcinogenesis. Carcinogenesis 6: 473–475, 1985 7 Kannerstein M, Churg J: Mesothelioma in man and experimental animals. Environ Health Perspect 34:31–36, 1980 8 Lopez A: The respiratory system. In: Thomson’s Special Veterinary Pathology, ed. Schrefer JA, Duncan LL, and Merchant T, 3rd ed., pp. 1–195. Mosby, St. Louis, MO, 2001

Vet Pathol 42:1, 2005

Brief Communications and Case Reports

81

9 Murthy SS, Testa JR: Asbestos, chromosomal deletions, and tumor suppressor gene alterations in human malignant mesothelioma. J Cell Physiol 180:150–157, 1999 10 Mutsaers SE: The mesothelial cell. Int J Biochem Cell Biol 36:9–16, 2004 11 Ordonez NG: Immunohistochemical diagnosis of epithe´˜ lioid mesotheliomas: a critical review of old markers, new markers. Hum Pathol 33:953–967, 2002 12 Vogelzang NJ: Emerging insights into the biology and

therapy of malignant mesothelioma. Semin Oncol 29(Suppl 18):35–42, 2002 13 Wilson DW, Dungworth DL: Tumors of the respiratory tract. In: Tumors in Domestic Animals, ed. Meuten DJ, 4th ed., pp. 398–399. Iowa State Press, Ames, IA, 2002 Request reprints from Dr. F. Reggeti, Department of Pathobiology, University of Guelph, Guelph, Ontario N1G 2W1 (Canada). E-mail: freggeti@uoguelph.ca.