Malignant Mesothelioma in Urban Dogs

Vet. Pathol. 20: 53 1-540 (I 983) Malignant Mesothelioma in Urban Dogs M. L. HARBISON and J. J. GODLESKI Department of Pathology, Harvard Medical School, Angel1 Memorial Animal Hospital, Brigham and Women s Hospital, Boston, Mass. Abstract. Clinical and postmortem materials from six dogs with a diagnosis of malignant mesothelioma were studied retrospectively. The dogs were urban pets with clinical signs of malignant effusions. Two mesotheliomas were pleural, one pericardial, and one peritoneal. Both pleura and pericardium were involved in one dog, and the pleura and peritoneum in another. On gross examination at necropsy, diffuse granular or velvety plaques covering mesothelial surfaces were found in all dogs; firm discrete pleural nodules also were present in two dogs. Neither distant mestastases nor areas of deep lung invasion were found. The tumors varied histologically, but the most common type was epithelial with a papillary pattern. Ultrastructurally, the neoplastic cells had prominent surface microvilli, numerous desmosomes, and tonofilaments. Lung tissue from these dogs and from control dogs was evaluated for the presence of ferruginous bodies. Asbestos bodies were found in three of five dogs with mesotheliomas but rarely were found in control dogs. As a group, the mesothelioma cases had significantly more asbestos bodies and total ferruginous bodies than controls. The clinical and morphologic appearance of canine mesothelioma is similar to human mesothelioma and also may be associated with exposure to airborne fibers. Malignant mesothelioma is a rare spontaneous tumor in dogs. Individual cases of this neoplasm occumng spontaneously in adult dogs have been reported [ 3, 11, 12, 14, 15, 181. Although the clinical and pathological features of the tumor have been examined, associated etiologic or epidemiologic factors have not been described. Experimentally, malignant mesotheliomas have been induced in dogs exposed to intratracheal instillation of crocidolite asbestos and cigarette smoke [ 131. In man, the association between malignant mesothelioma and asbestos exposure has been well established [2]. Many patients with mesothelioma have had direct occupational exposure, but the tumor also has been found in family members of asbestos workers and in people living in the vicinity of an asbestos mine or factory 121. Ferruginous bodies-fibers coated with ferritin and amorphous protein-are found in human lungs and are markers of asbestos exposure [4, 171. The majority of ferruginous bodies, termed typical ferruginous bodies, have cores of amphibole 531

532 Harbison and Godleski Table I. Doas with mesothelioma Dog number Breed Age (years) Sex Clinical presentation Duration 1 Doberman 8 F Emaciation, pleural and 1 month 2 Cocker spaniel 9 M peritoneal effusions Decreased exercise tol- Not stated 3 Dachshund 15 F erance, cough, pericardial effusion Anorexia, weakness, 10 days 4 English setter I1 M pleural effusions Polypnea, pleural effu- 1 month sion 5 Terrier-cross 10 F Peritoneal effusion 2 months 6 Collie-cross 7 M Weight loss, anorexia, 1 month pleural and peritoneal effusions* * Sixteen months earlier, this dog had benign pericardial and pleural effusions; a pencardotomy was done. asbestos [9]. These asbestos bodies can be found in the lungs of the general urban population [9, 161, but the number of asbestos bodiei per gram lung separates occupational from environmental exposure to asbestos [7, 91. We studied clinical and pathologic findings from a series of cases of canine mesotheliomas. We also examined lung tissue from these dogs and from a group of control dogs for the presence of fermginous bodies to evaluate the role of asbestos in this canine neoplasm. Materials and Methods Records of the Pathology Department of Angel1 Memorial Animal Hospital from 1960 to 1980 were reviewed. Six cases, having the coded (Standard Nomenclature of Disease and Operation-AMA) diagnosis of mesothelioma, were retrieved. Clinical and postmortem materials from these dogs were available for study. The dogs were urban pets referred to the hospital. These cases of mesothelioma represented 0.2% of dogs necropsied during this period. All necropsy material had been fixed in 10% neutral buffered formalin. Parafin-embedded tissues stained with hematoxylin and eosin (HE) were examined by light microscopy. For histochemical studies, tumor tissue also was stained with periodic acid-schiff (PAS) with and without diastase, mucicarmine, and Alcian blue (ph 2.5). For electron microscopy, formalin-fixed tumor or tumor tissue removed from parafin blocks was post-fixed with 2.5% glutaraldehyde and then 1.0% osmium, dehydrated, embedded in epon, and stained with uranyl acetate and lead citrate. The ferruginous body content of lung tissue from five dogs with mesotheliomas was investigated. Lung tissue also was collected from the next necropsied, age and chronologically matched dogs and from eight additional control dogs necropsied in 1981. The formalinfixed tissue was digested according to a described method [6]. One to five grams of lung were dissolved in 5% sodium hypochlorite. After extraction and washing, the resultant sediment was collected on a 0.45-pm filter (Millipore Co., Bedford, MA) which was cleared

Tumor pattern Dog Location number (gross/histologic) 1 Diffuse/well differentiated 2 Diffuse/poorly differentiated 3 Diffuse and nodular/ poorly differentiated 4 Diffuse and nodular/ well differentiated 5 Diffuse/moderatel y differentiated 6 Diffuse/moderately differentiated Table 11. Characteristics of canine mesothelioma Extent Bilateral pleura, pen- Superficial invasion of toneum pleural and peritoneal aspects of diaphragm and surface of abdominal organs Pericardium Invasion of myocardium and regional lymph nodes Bilateral pleura Only parietal pleura involved Bilateral pleura Extensive parietal pleural invasion; superficial invasion of visceral pleura and diaphragm Peritoneum Superficial invasion of surface of abdominal organs and diaphragm Bilateral pleura, pen- Extensive right parietal cardium pleural involvement; local invasion of epicardium, esophageal adventitia, re- No No No No No Distant metastasis Microscopic tumor in periadrenal lymphatics, sinus of one mesenteric gional lymph nodes lymph node Other pathologic findings - Severe necrotizing pericarditis Chronic interstitial nephritis; parathyroid hyperplasia Pleural plaques; interstitial cell tumor of testis Chronic passive congestion of liver and lungs

534 Harbison and Godleski Fig. 1: Visceral pleural plaques in mesothelioma. with xylene, mounted on slides, and then examined by light microscopy. Morphologic types and numbers of ferruginous bodies per gram of wet lung were recorded. Results The dogs studied and their clinical features are listed in table I. The most prominent sign in all dogs with mesotheliomas was malignant effusion. The mean age was 10 f 2.8 years, and there was no sex or breed predilection. In all instances, the owners eventually elected euthanasia for these dogs after a relatively short, terminal clinical course, which averaged one month. One dog had a previous history of benign pleural and pericardial effusions. Two neoplasms were pleural, one pericardial, and one peritoneal (table 11). Both pleura and pericardium were involved in one dog, and the pleura and peritoneum in another. The tumors were similar in gross appearance with a diffuse pattern most common. These diffuse lesions were characterized by granular or velvety plaques covering mesothelial surfaces. Firm discrete nodules, ranging from 0.3 to 4.0 cm in diameter, also were found in two dogs. The mesotheliomas invaded locally and superficially. The diaphragm was involved in three dogs. Neither distant metastases nor deep lung invasion was found in any dog. In one instance, neoplastic cells were identified microscopically in distant lymphatics. No parietal pleural plaques, similar to those associated with asbestos exposure in man, were described in these dogs; however, fibrous visceral pleural plaques were present in one dog (fig. 1). There were few other significant lesions in any of these dogs. The tumors varied histologically but were predominantly epithelial. None of the cases were spindle cell tumors. A histologic spectrum could be seen in individual tumors as well as from dog to dog. The most common pattern was papillary with a vascular stroma covered by one or several layers of neoplastic mesothelial cells (fig. 2). The cuboidal or columnar cells were uniform and had abundant cytoplasm, distinct borders, and prominent nucleoli. Occasionally the neoplastic cells formed a less organized proliferation which was associated with a scanty stroma (fig. 3).

Mesothelioma in Dogs 535 Fig. 2: Papillary pattern with vascular stroma covered by uniform, neoplastic mesothelial cells. HE. Bar = 55 pm. Fig. 3: Less organized proliferation of cells with scanty stroma. HE. Bar = 20 pm. Fig. 4: Anaplastic cells forming a solid pattern. HE. Bar = 15 pm. Fig. 5: Ultrastructural features of malignant mesothelial cells. Bar = 3 pm.

536 Harbison and Godleski 6a Fig. 6: Morphologic types of ferruginous bodies found in digested lung tissue: a. typical ferruginous body or asbestos body with straight, transparent core; b. atypical fermginous body with broad core and flared ends; c. atypical segmented and bent ferruginous body. Bar = 10gm. Table 111. Ferruginous bodies/gram wet lung in control dogs and dogs with mesotheliomas Dog number 2 3 4 5 6 Mean Matched controls (n = 5)' 198 1 controls (n = 8)' Total controls (n = 13)' bodies (fig. 6a) 2 0 6 27 0 0 0 5 7 65 3.0** 19.4 0 0 < 0.1 < 0.1 < 0.1** < 0.1 Atypical Atypical broad segmented Total bodies bodies bodies (fig. 6b) (fig. 6c) 5 8 43 78 4 27.6 18.1 5.6 10.3 ' Values for controls are mean values. * Mean f standard deviations are significantly different (p <.01). ** Significantly different (p <.05). 7 41 43 83 76 49.6 f 30.1* 18.1 f 15.7 5.6 f 4.6 10.4 f 11.6* Another epithelial pattern had more anaplastic cells with indistinct borders forming a dense mass (fig. 4). The predominant epithelial pattern of each tumor, in terms of differentiation, is listed in table 11. Mitotic figures were seen infrequently in the neoplasms. Four of six mesotheliomas stained positively with Alcian blue. This staining could be seen at cell borders, in the stroma, and in a few cells with

Mesothelioma in Dogs 5 37 vacuolated cytoplasm. One mesothelioma demonstrated light mucicarmine staining, and two cases stained positively with PAS, but negatively with PAS-diastase. The mesotheliomas had prominent ultrastructural features on electron microscopy. The neoplastic cells had long slender surface microvilli, numerous desmosomes, tonofilaments, and uniform round or oval nuclei (fig. 5). Intracellular organelles were difficult to evaluate since the material had been fixed initially with formalin at necropsy and, in two instances, recovered from paraffin blocks. Histologic examination of the lungs from these dogs did not reveal interstitial pulmonary fibrosis or the presence of fermginous bodies. Figure 6 illustrates the appearance and table I11 lists the quantity of fermginous bodies found in digested lung tissue from the dogs in this study. Dogs with mesothelioma had greater numbers and different morphologic types of ferruginous bodies compared with controls. Three morphologic types of bodies were found in the lung samples. The bodies were typically 40 to 70 pm in length. The first type, typical fermginous bodies or asbestos bodies, was characterized by optically transparent straight cores. These were present in three of five dogs with mesotheliomas but were a very rare finding in both the age-matched (mean age & 9.8 years) and the 198 1 control dogs. In dogs with mesotheliomas, 2, 6, and 7 bodies/gram wet weight were found. Most controls had no asbestos bodies. Only three of thirteen had any bodies, and each of these three dogs had a single asbestos body14 grams of digested lung tissue. Atypical fermginous bodies composed of pale yellow, broad, and regular cores with flared ends also were found. The majority of fibers with this appearance were uncoated. These bodies resemble sheet-silicate ferruginous bodies [8]. Three of five dogs with mesotheliomas had these atypical bodies which also were present very rarely in control dogs. Two dogs with tumors (dogs 3 and 5) had the highest numbers of both asbestos bodies and these atypical fermginous bodies. A third type of ferruginous body was present in the lungs of all dogs, both controls and dogs with mesotheliomas. These structures were very broad, segmented, and bent fermginous bodies which are thought to contain diatomaceous earth [ 101. These bodies could be distinguished readily by light microscopy from asbestos bodies. When numbers of ferruginous bodies were compared by the z test for equal means, the dogs with mesotheliomas had significantly more total fermginous bodies than controls (p <.Ol) and more asbestos bodies than controls (p <.05). Discussion Canine malignant mesotheliomas are similar to human mesotheliomas in clinical and morphologic appearance. People with mesotheliomas usually die from the complications of their primary lesions [ 11 as did the dogs in this study. The disease has a rapidly progressive clinical course with a grave prognosis in both species. A clinical picture highlighted by malignant effusions was present in our dogs and in other reported cases of canine mesotheliomas [3, 1 1, 12, 14, 15, 181.

538 Harbison and Godleski The gross lesions of the neoplasm in man and dog are similar. Tumor growth occurs primarily by local extension, and distant metastases are not a prominent feature. Histologically the mesotheliomas in these dogs were predominantly of the epithelial type and resembled this type in man. The fibrous or spindle cell variety of mesothelioma, found also in man, was not seen. The histologic appearance of other reported canine mesotheliomas also has been epithelial [3, 14, 181. Finally, human and canine mesotheliomas share the same ultrastructural characteristics. Our electron microscopic findings are similar to those found in a previous case report of canine mesothelioma [ 181. Fibers coated with iron and protein were found in digested lung tissue from all dogs. As a group, dogs with malignant mesothelioma had significantly more ferruginous bodies in their lungs than did controls. This quantitative difference suggests that exposure to airborne fibers was not the same in the two groups of dogs. Our findings also suggest a qualitative difference in the types of fibers inhaled by controls and dogs with mesotheliomas. Asbestos bodies were present in three dogs with tumors but were virtually absent in controls. Similarly, atypical pale yellow ferruginous bodies were found in three dogs with mesotheliomas. Asbestos bodies have been found in the lungs of nearly everyone in the general population [ 101 and the prevalence of lung ferruginous bodies is greater in residents of urban than rural communities [2]. Asbestos bodies are markers of exposure, and numbers of bodies in urban patients correlate in a general way with occupation. In one study, most women and male office workers had less than 50 asbestos bodies per gram wet lung and were considered to have environmental exposure to asbestos. On the other hand, many steel and construction workers had greater than 100 bodies per gram and were considered to have additional occupational exposure [7]. Asbestos workers themselves had many times more bodies (lo5 to lo7 per gram) [9]. Considering people with the highest exposures, numbers of asbestos bodies correlate generally with the presence of diffuse interstitial pulmonary fibrosis [5]. In people with low exposure, however, no correlations between numbers of bodies and disease have been made [5, 8, 91. Dogs in this study had fewer asbestos bodies in their lungs than have been found in people. Most control dogs had no asbestos bodies, unlike the human findings. The higher prevalence and the greater numbers of asbestos bodies in dogs with mesotheliomas may signal a difference in residence or life-style. It also is possible that a dog could have a bystander exposure as a family pet of an asbestos worker. Interestingly, broad segmented atypical ferruginous bodies were present in all dogs but are observed very infrequently in man [lo]. The significance of these bodies is not known. Our discussion of fibers found in these dogs is based on structures seen by light microscopy. We assume that morphologically typical ferruginous bodies found in dogs contain asbestos as they do in people. Whether the numbers of these asbestos bodies found in the dogs with mesotheliomas reflect exposures to asbestos which

Mesothelioma in Dogs 539 have had a carcinogenic effect is not known. In people, the number of asbestos bodies is considerably less than the total number of asbestos fibers in the lungs [5]. Most fibers are uncoated and invisible by light microscopy, but the number of coated fibers is usually a reliable indication of the total number of fibers. In addition to total numbers, the type of asbestos fiber and its physical dimensions are important in asbestos-related human diseases. For example, mesothelioma appears to be associated with increased numbers of amphibole fibers [5]. Thus, an analysis of both coated and uncoated fibers with electron optical techniques is an ideal means to describe the total retention of inhaled fibers in an animal's lungs. However, since these specialized techniques are difficult and time consuming and are not readily available in all centers, counting fibers from lung digests can be a useful technique to gain preliminary information in previously unstudied areas. Acknowledgements We wish to thank Rebecca Stearns for technical assistance. This work was supported by Public Health Service Grant 5T32 RR07000 in veterinary and comparative pathology. References 1 ANTMAN, K.H.: Malignant mesothelioma. New Engl J Med 303:200-202, 1980 2 BECKLAKE, M.R.: Asbestos-related diseases of the lung and other organs: their epidemiology and implications for clinical practice. Am Rev Respir Dis 114: 187-227, 1976 3 BREEZE, R.G.; LAUDER, I.M.: Pleural mesothelioma in a dog. Vet Rec 96:243-246, 1975 4 CASEY, K.R.; ROM, W.N.; MOATAMED, F.: Asbestos-related disorders. Clin Chest Med 2: 179-202, 1981 5 CHURG, A.: Fiber counting and analysis in the diagnosis of asbestos-related disease. Hum Pathol 13:381-392, 1982 6 CHURG, A,: SAKODA, N.; WARNOCK, M.L.: A simple method for preparing ferruginous bodies for electron microscopic examination. Am J Clin Pathol 68: 5 13-5 17, 1977 7 CHURG, A,; WARNOCK, M.L.: Correlation of quantitative asbestos body counts and occupation in urban patients. Arch Pathol Lab Med 101:629-634, 1977 8 CHURG, A,; WARNOCK, M.L.: Number of asbestos bodies in urban patients with lung cancer and gastrointestinal cancer and in matched controls. Chest 76: 143-149, 1979 9 CHURG, A.M.; WARNOCK, M.L.: Asbestos and other ferruginous bodies. Am J Pathol 102~447-456, I98 1 10 CHURG, A.; WARNOCK, M.L.; GREEN, N.: Analysis of the cores of ferruginous (asbestos) bodies from the general population: 11. True asbestos bodies and pseudoasbestos bodies. Lab Invest 40:3 1-38, I979 11 DUBIELZIG, R.R.: Sclerosing mesothelioma in five dogs. J Am Anim Hosp Assoc 15:745-748, 1979 12 GEIB, L.W.: DENARVAEZ, F.; EBY, C.H.: Pleural mesothelioma in a dog. J Am Vet Med ASSOC 140:1317-1319, 1962 13 HUMPHREY, E.W.; EWING, S.L.; WRIGLEY, J.V.; NORTHRUP, W.F.; KERSTEN, T.E.; MAYER, J.E.; VARCO, R.L.: The production of malignant tumors of the lung and pleura in dogs from intratracheal asbestos instillation and cigarette smoking. Cancer 47: 1994-1999, 1981 14 IKEDE, B.O.; ZUBAIDY, A.; GILL, C.W.: Pericardial mesothelioma with cardiac tamponade in a dog. Vet Pathol 17:496-499, 1980

540 Harbison and Godleski 15 KASBOHN, C.; GEMBARDT, C.: Das mesotheliom des hundes aus zytologischer und histologischer sicht. Tierarztl Prax 5:379-388, 1977 16 ROSEN, P.; MELAMED, M.; SAVINO, A.: The fenuginous body content of lung tissue: a quantitative study of 86 patients. Acta Cytol (Baltimore) 16:207-2 11, 1972 17 SUZUKI, Y.; CHURG, J.: Formation of the asbestos body: a comparative study with three types of asbestos. Environ Res 3: 107-1 18, 1969 18 TRIGO, F.J.; MORRISON, W.B.; BREEZE, R.G.: An ultrastructural study of canine mesothelioma. J Comp Pathol91:531-537, 1981 Request reprints from Dr. Margaret Harbison, Department of Physiology, Harvard School of Public Health, 665 Huntington Avenue, Boston, MA 02 1 15 (USA).