Primary renal tumors are diagnosed uncommonly in

J Vet Intern Med 2006;20:1155 1160 Primary Renal Neoplasia of Dogs Jeffrey N. Bryan, Carolyn J. Henry, Susan E. Turnquist, Jeff W. Tyler, Julius M. Liptak, Scott A. Rizzo, Gabriella Sfiligoi, Steven J. Steinberg, Annette N. Smith, and Tarraca Jackson Background: Primary renal tumors are diagnosed uncommonly in dogs. Hypothesis: Signs and survival will differ among different categories of primary renal tumors. Animals: Data were collected from the medical records of 82 dogs with primary renal tumors diagnosed by examination of tissue obtained by ultrasound-guided biopsy, needle aspiration, surgery, or at postmortem examination. Methods: This was a multi-institutional, retrospective study. Results:: Forty-nine dogs had carcinomas, 28 had sarcomas, and 5 had nephroblastomas. The dogs were geriatric (mean 8.1 years; range: 1 17) with a weight of 24.9 kg (range: 4.5 80). Tumors occurred with equal frequency in each kidney with 4% occurring bilaterally. Initial signs included one or more of hematuria, inappetance, lethargy, weight loss, or a palpable abdominal mass. Pain was reported more frequently in dogs with sarcomas (5/28). The most common hematologic abnormalities were neutrophilia (22/63), anemia (21/64), and thrombocytopenia (6/68). Polycythemia was present in 3 dogs and resolved with treatment. Hematuria (28/49), pyuria (26/49), proteinuria (24/50), and isosthenuria (20/56) were the most frequently observed abnormalities on urinalysis. Pulmonary metastases were noted on thoracic radiographs in 16% of dogs at diagnosis. Seventy-seven percent of dogs had metastatic disease at the time of death. Median survival for dogs with carcinomas was 16 months (range 0 59 months), for dogs with sarcomas 9 months (range 0 70 months), and for dogs with nephroblastomas 6 months (range 0 6 months). Conclusions and Clinical Importance: Primary renal tumors in dogs are generally highly malignant with surgery being the only treatment that improves survival. Key words: Canine; Carcinoma; Kidney; Nephroblastoma; Sarcoma. Primary renal tumors are diagnosed uncommonly in dogs. Tumors can arise from the renal epithelium, renal mesenchyme, or from embryonal tissue of mixed origin. Previous published case series have examined primarily carcinomas with very few dogs with sarcomas or nephroblastomas reported. 1 4 Renal lymphoma has been reported, but it is difficult to determine whether these represent primary or multicentric disease. 4 Affected animals are typically middle-aged or older at diagnosis with mean ages of 7.1 8.8 years reported for carcinomas and 8.8 and 8.0 years for sarcomas and nephroblastomas, respectively. 3,4 Male-to-female ratios of 1.8 : 1 and 1.6 : 1 have been reported for dogs with primary renal carcinomas. 3,4 This is similar to the 1.5 : 1 ratio reported for humans. 5 Although no breed predis- From the Department of Veterinary Medicine and Surgery, University of Missouri-Columbia, MO (Bryan, Henry, Turnquist, Tyler, Jackson); Animal Cancer Center, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, CO (Liptak, Rizzo); Department of Veterinary Medicine and Surgery, University of Pennsylvania, PA (Sfiligoi); Veterinary Cancer Registry (Steinberg); and Department of Clinical Sciences, Auburn University (Smith). Dr. Turnquist is currently affiliated with Worldwide Safety Sciences, Kalamazoo, MI. Dr. Liptak is currently affiliated with Ontario Veterinary College, University of Guelph. Dr. Rizzo is currently associated with the University of Tennessee. Dr. Sfiligoi is currently affiliated with the University of California, Davis. Ms. Jackson is currently associated with North Carolina Agricultural and Technical University, Greensboro, NC. This work was presented in part at the 22nd Annual Meeting of the Veterinary Cancer Society, September 12 15, 2002. Reprint requests: Jeffrey N. Bryan, 379 E. Campus Dr., Columbia, MO 65211; e-mail: bryanjn@missouri.edu. Submitted March 3, 2006; Revised December 22, 2005; Accepted March 6, 2006. Copyright E 2006 by the American College of Veterinary Internal Medicine 0891-6640/06/2005-0014/$3.00/0 positions have been identified for renal neoplasia, a syndrome of nodular dermatofibrosis associated with renal cystadenocarcinoma has been identified in German Shepherd dogs. 6 No previous article has examined the effect of adjuvant chemotherapy on renal tumors, and surgery is the only reported therapeutic option. 4 Response of human renal cell carcinoma to chemotherapy has been disappointing, with the combinations of gemcitabine and 5-fluorouracil or tamoxifen and vinblastine showing the most promise. 7 Immunotherapy using interleukin-2 and interferon- remains the mainstay of treatment for this disease in people. 7 The purpose of this study was to further characterize the clinical signs, laboratory findings, diagnostic imaging results, histopathologic diagnoses, and the effect of treatment on survival in dogs with primary renal tumors. Materials and Methods The medical databases of the veterinary teaching hospitals at the University of Missouri-Columbia, Animal Cancer Center at Colorado State University, Auburn University, the University of Pennsylvania, the Donaldson-Atwood Cancer Clinic of the Animal Medical Center, Florida Veterinary Specialists, Rowley Memorial Animal Hospital, Pacific Veterinary Specialists, as well as the databases of the Veterinary Medical Diagnostic Laboratory of the University of Missouri-Columbia and the Vet Cancer Registry were searched for cases of canine primary renal neoplasia. Cases were included if they were diagnosed by examination of tissue obtained by biopsy, needle aspiration, or postmortem examination, and did not have evidence of an alternative primary site. Cases with the diagnosis of lymphoma were excluded as it is very difficult to definitively demonstrate that the disease has arisen primarily in the kidney. Data regarding the breed, sex, age, weight, diagnosis, presenting complaint, results of complete blood count (CBC), serum chemistry profile, urinalysis, thoracic and abdominal radiographs, and ultrasound, method of diagnosis, location and laterality of tumor, type of treatment, survival time, outcome, and the presence or absence of metastasis at diagnosis or death was collected. Survival time was defined as the time from presentation

1156 Bryan et al Table 1. Demographic and clinical characteristics of dogs with renal cancer. Carcinoma Sarcoma Nephroblastoma Number of cases in series 49 28 5 Age (mean, range) 8.4 (2 15) 8.2 (1 17) 5.2 (1 12) Sex a MI: 7 MI: 0 MI: 2 MC: 19 MC: 9 MC: 1 FI: 5 FI: 1 FI: 0 FS: 17 FS: 18 FS: 2 Weight (kg) (mean, range) 25.2 (4.5 80) 23.6 (5 80) 40.4 (9.3 53) Laterality Right: 23 Right:11 Right: 1 Left: 23 Left: 17 Left: 4 Bilateral: 3 Bilateral: 0 Bilateral: 0 Metastasis at diagnosis 6/34 3/21 1/5 Metastasis at death 20/29 14/16 3/4 Survival (median, range) 16 months (0 59 months) 9 months (0 70 months) 6 months (0 6 months) MI, intact male; MC, castrated male; FI, intact female; FS, spayed female. a Sex of one dog was not recorded. for treatment until death because of the tumor or euthanasia. Histopathology review was performed by one pathologist (SET) in all cases for which samples were available (n 5 29). Cases without complete information were included for descriptive purposes, but excluded from categorical variable and survival analysis. Statistical analyses were performed using Chi-squared test or the Fisher s exact test for categorical variables and the Wilcoxon log-rank test for survival analysis using commercial software. a Oneway analysis of variance or Kruskal-Wallis one-way analysis of variance on ranks was used to compare nonparametric integral data. Comparisons were made between presenting complaint, CBC, serum chemistry, or urinalysis results and diagnosis or category of neoplasia to assess for significant association. Comparisons were also made between survival and diagnosis, category of disease, and whether or not the animal was treated with surgery or received chemotherapy. Animals were censored for survival analysis at the point of last contact if they died or were euthanized because of other causes or were lost to follow-up. Differences were considered to be of statistical significance at P #.05. Results Eighty-two dogs with primary renal neoplasia between 1986 and 2002 were identified. Forty-nine dogs had carcinoma including renal carcinoma (n 5 16), transitional cell carcinoma (n 5 9), tubular adenocarcinoma (n 5 7), tubular and papillary adenocarcinoma (n 5 6), renal adenocarcinoma (n 5 5), renal tubular carcinoma (n 5 4), clear cell renal carcinoma (n 5 1), and papillary cystadenocarcinoma (n 5 1) (Table 1). Twenty-eight dogs had sarcoma including hemangiosarcoma (n 5 12), renal sarcoma (n 5 7), leiomyosarcoma (n 5 4), malignant fibrous histiocytoma (n 5 3), spindle cell sarcoma (n 5 1), and fibroleiomyosarcoma (n 5 1). Five dogs had nephroblastoma. Histologic examination of archived tissue was performed in 29 cases including 13 carcinomas, 13 sarcomas, and 3 nephroblastomas. Cases that had not been examined by immunohistochemistry were stained with smooth muscle actin for leiomyosarcomas, cytokeratin to rule out carcinoma for undifferentiated sarcomas, Factor VIII for hemangiosarcoma, and vimentin in the mesenchymal population and cytokeratin in the epithelial populations of nephroblastomas. Of the 29 cases reviewed, one diagnosis was changed from nephroblastoma to sarcoma. For dogs with a diagnosis of carcinoma, there was no statistical difference in proportion of males and females (P 5.177). For dogs with a diagnosis of sarcoma, males were affected less commonly than females, particularly spayed females (P 5.027). The difference in the sex distribution between tumor categories was not statistically significant (P 5.105). Age at presentation (81 recorded) ranged from 1 to 17 years (mean 5 8.1 years). Age was statistically similar among tumor categories (P 5.276). Body weight was statistically similar among tumor categories (P 5.216). Diagnosis was made by several methods in this case series. Sixty-four dogs underwent unilateral nephrectomy with subsequent histologic evaluation of excised tissues. Five were diagnosed by incisional biopsy and one by ultrasound-guided needle biopsy. One carcinoma was diagnosed by cytology from ultrasound-guided fine needle aspiration. Seven were diagnosed at postmortem examination, and 1 was diagnosed by samples collected at surgery after being euthanized. Where localization was recorded (n 5 81), 53% of tumors occurred in the right kidney, 43% in the left, and 4% bilaterally. All three dogs with bilateral disease had carcinomas, 2 with renal cell carcinomas and 1 with cystadenocarcinoma. Breeds affected included mix-breed dogs (n 5 17), Cocker Spaniel (n 5 7), Labrador Retriever (n 5 7), Rottweiler (n 5 5), Golden Retriever (n 5 5), German Shepherd (n 5 3), Miniature Poodle (n 5 3), Malamute (n 5 2), Springer Spaniel (n 5 2), Shetland Sheepdog (n 5 2), Boxer (n 5 2), and 1 each of Great Dane, Collie, Bernese Mountain Dog, Greyhound, Blue-Tick Hound, Siberian Husky, Standard Poodle, Belgian Sheepdog, Puli, Chow Chow, Border Collie, Pointer, Old English Sheepdog, Beagle, Shih Tzu, Jack Russell Terrier, Lhasa Apso, West Highland White Terrier, Cavalier King Charles Spaniel, Miniature Schnauzer, Chesapeake Bay Terrier, Airedale Terrier, Boston Terrier, Basset Hound, Pomeranian, and Standard Schnauzer. Common signs reported by the owners were: hematuria in 32% (26/82) of dogs; a palpable abdominal mass in

Renal Neoplasia of Dogs 1157 Fig 1. Kaplan-Meier plot of survival by tumor type. Carcinomas (n 5 49) are designated by a solid line, sarcomas (n 5 28) designated by long dashed line, and nephroblastomas (n 5 5) by short dashed line. Censored animals are designated by a vertical line ( ). There is no statistically significant difference between the curves by Wilcoxon survival analysis (P 5.788). 20% (16/82); polydipsia or polyuria in 10% (8/82); inappetance (27%; 22/82); lethargy (26%; 21/82); weight loss (20%; 16/82); vomiting (13%; 11/82); pain (7%; 6/82); and behavior changes (5%; 4/82). Pain was recorded as the presenting complaint for 6 dogs, and was significantly associated with a diagnosis of sarcoma (P 5.016). Changes recorded in hematologic examinations were nonspecific. Abnormalities were noted in 51 of 70 dogs (73%). Differential white cell counts, platelet count, or red cell parameters were not recorded for all evaluations. Neutrophilia was the most common change and was present in 22 of 63 (35%) ranging from 13,000 cells/ml to 266,500 cells/ml. Four dogs had neutrophil counts.30,000 cells/ml (46,800 266,500 cells/ml). Hematologic abnormalities included: anemia in 21 of 64 dogs (33%); leukocytosis (20%; 13/64); lymphopenia (9%; 6/63); thrombocytopenia (9%; 6/68); and monocytosis (6%; 4/63). Polycythemia was present in 3 dogs, all with carcinomas and which survived 7.5 32 months. Abnormalities in serum chemistry values were minor and nonspecific. Serum chemistry results were normal in 16 of 65 (25%) recorded results. High serum alkaline phosphatase (ALP) was present in 21 of 65 dogs (22%). Blood urea nitrogen (BUN) was high in 14 of 65 dogs (22%), and serum creatinine was high in 13 of 65 (20%). Serum alanine aminotransferase (15%; 10/65), aspartate aminotransferase (9%; 6/65), cholesterol (6%; 4/65), phosphorous (10%; 6/64), c-glutaryl transferase (9%; 6/64), and globulins (5%; 3/62) were also high. Serum albumin concentration was decreased in 12 of 63 dogs (19%) with a statistically significant association with a diagnosis of sarcoma compared to carcinoma or nephroblastoma (P 5.043). Abnormalities were commonly identified by urinalysis. Twenty-eight of 49 dogs (57%) had hematuria and 26 of 49 dogs (53%) had pyuria. Proteinuria was reported in 24 of 50 dogs (48%). Isosthenuria was present in 20 of 56 dogs (36%). Thirty-three of 82 dogs (36%) did not have a urinalysis recorded. Results of thoracic radiographs were recorded for 61 dogs with10 dogs (16%) having evidence of metastasis at diagnosis. Of these dogs, 6 had carcinomas, 3 had sarcomas, and 1 had a nephroblastoma. A further 37 dogs (61%) had metastasis identified at the time of death. Twenty of these dogs had carcinomas, 14 had sarcomas, and 3 had nephroblastomas. Data were available documenting metastatic status at death for 49 dogs, yielding an overall metastatic rate of 69% (21/30) for carcinomas, 88% (13/15) for sarcomas, and 75% (3/4) for nephroblastomas. Abdominal radiograph results were recorded for 37 dogs. A mass was identified on radiographs in 33 of these dogs (89%). Abdominal ultrasound results were recorded for 55 dogs with a renal mass identified in 51 of these dogs (93%). Suspected metastatic lesions were identified by ultrasound in the contralateral kidney (n 5 1), spleen (n 5 1), and liver (n 5 2) in 4 dogs. Of 73 dogs for which the information was available, 7 had other neoplastic diseases in addition to the renal tumor. These included a soft-tissue sarcoma (grade I), insulinoma, plasma cell tumor, adrenocortical carcinoma, mammary carcinoma, anaplastic sarcoma, and 1 dog with renal hemangiosarcoma had concurrent leiomyosarcoma, leiomyoma, melanoma, ameloblastoma, and interstitial carcinoma. There was no statistical difference among survival times by diagnosis (P 5.788) (Table 1, Fig 1). Sixtyeight of 82 dogs underwent surgery. Median survival for all dogs undergoing surgery, regardless of tumor type, was 16 months (range, 0 70 months), compared to,1 month (range, 0 32 months) for dogs not undergoing surgery (P,.001). Chemotherapy was administered to 23 dogs, 21 of which had a previous nephrectomy. In dogs with renal carcinoma, 13 received a doxorubicinbased protocol, alone or in combination with cyclophosphamide, carboplatin, mitoxantrone, paclitaxel, or prednisone. An additional 3 dogs with renal carcinoma received protocols of carboplatin, mitoxantrone and piroxicam, or carprofen alone. In dogs with renal sarcoma, 3 dogs received a doxorubicin-based protocol, alone or with doxycycline, vincristine, or cyclophosphamide. In the same group, 1 dog each received mitoxantrone and carboplatin. For dogs with an antemortem diagnosis, median survival time without chemotherapy was 7.5 months (range, 0 70 months) compared to 12.0 months (range, 0 32 months) with chemotherapy (Fig 2). These were not statistically different (P 5.39). Discussion Primary renal tumors are uncommon in dogs. Previous reports have examined primarily carcinomas with only a series of 6 sarcomas and case reports of a spindle cell sarcoma, osteosarcoma, leiomyosarcoma, hemangioma, hemangiosarcoma, giant cell tumor, oncoctyoma, and mixed mesenchymal tumor in the literature. 2,4,8 16 Twenty-seven of the dogs presented here were renal cell

1158 Bryan et al Fig 2. Kaplan-Meier plot of survival by whether an animal did or did not receive chemotherapy. Animals receiving chemotherapy (n 5 22) are designated by a solid line and animals not receiving chemotherapy (n 5 49) by a dashed line. Censored animals are designated by a vertical line ( ). There was no statistically significant difference between the groups by Wilcoxon survival analysis (P 5.39). carcinomas or a tubular, tubular and papillary, or clear cell variant of renal carcinoma. Twelve had more differentiated tumors classified as adenocarcinomas. Nine dogs had tumors arising from the transitional epithelium. One dog had bilateral papillary cystadenocarcinomas. No survival difference could be identified among these tumor groups, making the epithelial cell of origin a poor predictor of outcome. The sarcomas were more varied in tissue of origin, and could not be grouped for analysis. No survival difference was observed based on histologic tumor type in sarcomas. All groups of tumors exhibited malignant behavior in this series, regardless of their histologic diagnosis. No breed group was apparently overrepresented among these 82 dogs. Previous reports have generally found no breed predilection for renal tumors. 3,4,17 The one exception to this is the occurrence of hereditary multifocal renal cystadenocarcinomas in German Shepherd dogs in association with nodular dermatofibrosis. 6 Dogs in this report were geriatric with ages of 8.4 +/2 3.4 years and 8.2 +/2 3.7 years, respectively, for dogs with carcinomas and sarcomas. This is similar to previous reports. 3,4 This is also similar to the geriatric presentation of renal carcinomas and sarcomas in humans. 5,20 In humans, nephroblastomas typically affect children and have been reported to affect dogs younger than 2 years of age. 5,21 In this series, although 3 of the 5 dogs with nephroblastoma were #2 years, the mean and median ages did not differ statistically from the ages of dogs with carcinoma or sarcomas. Of the 3 nephroblastomas in this study that were available for histopathologic confirmation, the two that were confirmed by the presence of vimentin staining in the mesenchymal population and cytokeratin staining of the epithelial population were from the two oldest dogs with nephroblastoma, aged 9 and 12 years. 22 This finding corroborates the previous reports of nephroblastomas in dogs up to 8 years of age. 4,17,23 As in previous reports, presenting complaints and clinical signs were nonspecific and included polyuria/ polydipsia, vomiting, and behavior changes. 4 It has been previously reported that an abdominal mass could be palpated in as many as 43% of cases of renal tumors. 4 In this series, only 20% presented with a palpable abdominal mass. An equal proportion of dogs presented for weight loss with a higher proportion presenting for inappetance (27%) and lethargy (26%). Similar complaints were reported in another case series of renal carcinomas. 3 Hematuria was the most common presenting complaint, occurring in 32% of the cases. This is similar to the frequency previously reported. 4 Pain was recorded as the presenting complaint for 6 dogs. There was a significant association between this complaint and the diagnosis of sarcoma. A similar phenomenon has been observed in 81% of humans presenting with angiosarcoma. 24 Arising from mesenchymal structures, thus avoiding intraluminal urinary obstruction or bleeding, these tumors may cause fewer clinical signs than carcinomas or nephroblastomas before growing large enough to cause pain. Recorded CBC abnormalities in these dogs were nonspecific, as no abnormality had a statistical association with any type of renal tumors. Neutrophilia was the most commonly reported abnormality. This was likely associated with a stress response, corticosteroid administration, or a response to infection in most of the dogs. However, for dogs with neutrophilia greater than 40,000 cells/ml, paraneoplastic syndrome was a possible explanation. In a previous report, paraneoplastic neutrophilia resolved after nephrectomy for a renal carcinoma. 25 Two dogs (1 with hemangiosarcoma and 1 with carcinoma) died in the early diagnostic period, 1 lived 2 months with hemangiosarcoma and 1 lived 4 months with carcinoma. Both short-term survivors died with metastasis. Neither dog had a recorded re-evaluation of the CBC. Paraneoplastic polycythemia was present in 3 dogs, and resolved in each case after nephrectomy. 5 Polycythemia was not associated with a shorter survival time. Serum chemistry profile changes were also nonspecific. Azotemia was the most common abnormality, and was not statistically significantly associated with any tumor type. The only statistically significant association was a decreased likelihood for dogs with sarcoma to have hypoalbuminemia compared to carcinoma or nephroblastoma patients. There was not, however, a concomitant significant increase in the proteinuria of dogs with carcinomas or nephroblastomas. Proteinuria and hematuria were detected with nearly equal frequency. Both were present in approximately half of the urinalyses recorded. Virchow s triad of hematuria, flank pain, and flank mass, however, was not described for any dog in this study. 26 Pyuria was detected in as many dogs as was hematuria, raising the possibility that secondary infection is common in dogs with renal tumors. This must be identified and addressed

Renal Neoplasia of Dogs 1159 if surgery is planned or cytotoxic chemotherapy is to be instituted. The detection of pulmonary metastasis at presentation was recorded in 16% of cases compared to 48% previously reported. 4 The lower rate of detection may suggest diagnosis earlier in the course of disease in the animals in this study compared to the older study. Earlier stage is supported by the lower rate of palpation of abdominal masses in this study, as well. Abdominal ultrasound studies were performed more frequently than abdominal radiographic studies in this series. With either examination, a renal mass was detected in approximately 90% of cases. However, intra-abdominal metastasis was strongly suspected in 4 dogs based on ultrasound findings and confirmed at surgery in three of these dogs. Abdominal radiographs did not reveal strong evidence for metastasis in any of the dogs. The overall metastatic rate was 77% at death. These were proportionally distributed through the 3 types of tumors. All forms of primary renal neoplasia reported in this study exhibited malignant biologic behavior. Metastatic sites commonly include the lungs or any of the abdominal organs. Metastasis or local extension to the adrenal gland occurs in approximately 4% of human renal cell carcinoma cases. 26 In this series, it was seen in only 1 case. However, extension into the vena cava and the mesentery occurred in other cases. These structures should be evaluated carefully for such extension or metastasis in preoperative imaging studies. In human patients, the development of one form of neoplasia is a risk factor for further neoplastic events. 27 Seven (9%) of the dogs in this study were diagnosed with neoplasia of other body systems either before, concurrently, or after the diagnosis of renal neoplasia. The most striking of these dogs was one who had five other forms of neoplasia diagnosed over a period of several years before the renal hemangiosarcoma. This raises the possibility of a mutation in a tumor suppressor gene giving rise to multiple neoplastic changes. A retrospective case series of humans with urologic malignancies identified second malignancies in 5 patients with renal tumors. These included colon, thyroid, uterus, and breast cancer. 27 Mutations of genes such as UROC28 have been identified in cancers of multiple organs. Mutations of the p53 tumor suppressor gene have been reported in osteosarcoma, melanoma, and mammary and squamous cell carcinomas in dogs. 28 31 Occurrences of several forms of aggressive neoplasia, such as the dog in the present case series, raise the likelihood that mutations in genes like p53 may result in syndromes similar to Li-Fraumeni in humans. Definitive diagnosis in this series of dogs was most often made after nephrectomy. Surgical resection of the renal tumor had the most positive impact on survival compared to other treatment modalities. Dogs that underwent nephrectomy outlived dogs that were untreated (12 dogs) or dogs that received only medical therapy (2 dogs). Nephrectomy has been advocated with both curative- and palliative-intent for managing human renal tumors. 32 This procedure is most beneficial in unilateral disease where a curative procedure is possible. Clinical signs can be relieved by removing the mass causing pain or morbidity when renal lymphoma has been excluded as a diagnosis with aspiration cytology or needle biopsy. It is clear in this series that there is a survival benefit to surgical removal of the primary renal mass. What is less clear is the role of adjuvant chemotherapy in dogs with renal neoplasia. Dogs treated with chemotherapy in this series did not survive a detectably longer time than dogs not treated with chemotherapy. As this is a retrospective series of many types and classes of renal neoplasia with no uniformity of chemotherapy protocol, this may be because of a lack of power to detect a difference. This retrospective study was not designed to determine a therapeutic outcome, and multiple chemotherapy combinations were used in these patients. What is clear from this series of dogs is that each of the tumor types represented has a high risk of metastasis. Prospective, randomized trials are necessary to elucidate the benefit of adjuvant chemotherapy in treating dogs with these tumors. Primary renal neoplasia remains an infrequent diagnosis in dogs. Clinical signs and clinicopathologic findings are nonspecific, so careful early evaluation of patients with clinical signs referable to the urinary tract is necessary to detect these tumors at a manageable stage. Successful management is dependent on complete excision of the mass. Adjuvant chemotherapy may play a role in management of a number of these tumors, but that role has not been defined. Footnotes a SigmaStat for Windows 3.10. Systat Software Inc., Richmond, CA Acknowledgments This retrospective study was completed with the cooperation and assistance of the Veterinary Cooperative Oncology Group. The authors thank Stephen J. Withrow, DVM, DACVS, DACVIM (Onc), Philip J. Bergman, DVM, MS, PhD, DACVIM(Onc), William S. Dernell, DVM, MS, DACVS, Tracy LaDue, DVM, DACVIM(Onc), Jeff D. Bay, DVM, DACVIM, Amy Hipple, DVM, DACVIM(Onc), Christine Simcik, DVM, Terry Allen, DVM, Michael Slawienski, DVM, DACVS, Randy Jones, DVM, Linda Fineman, DVM, DACVIM(Onc), and Wendi Carter for their participation through referral and contribution of cases. References 1. Frimberger AE, Moore AS, Schelling SH. Treatment of nephroblastoma in a juvenile dog. J Am Vet Med Assoc 1995; 207:596 598. 2. Rudd RG, Whitehair JG, Leipold HW. Spindle cell sarcoma in the kidney of a dog. J Am Vet Med Assoc 1991;198:1023 1024.

1160 Bryan et al 3. Lucke VM, Kelly DF. Renal carcinoma in the dog. Vet Pathol 1976;13:264 276. 4. Klein MK, Campbell GC, Harris CK, et al. Canine primary renal neoplasms: A retrospective review of 54 cases. J Am Anim Hosp Assoc 1987;24:443 452. 5. Linehan WM, Bates SE, Yang JC. Cancer of the kidney. In: DeVita VT Jr, Hellman S, Rosenberg SA, eds. Cancer Principles and Practice of Oncology, 7th ed., Philadelphia: Lippincott, Williams, and Wilkins; 2005:1139 1167. 6. Lium B, Moe L. Hereditary multifocal renal cystadenocarcinomas and nodular dermatofibrosis in the German shepherd dog: Macroscopic and histopathologic changes. Vet Pathol 1985;22: 447 455. 7. Glaspy JA. Therapeutic options in the management of renal cell carcinoma. Semin Oncol 2002;29:41 46. 8. Munday JS, Egins J, Selcer BA, Stedman NL. Renal osteosarcoma in a dog. J Small Anim Pract 2004;45:618 622. 9. Sato T, Aoki K, Shibuya H, et al. Leiomyosarcoma of the kidney in a dog. J Vet Med A Physiol Pathol Clin Med 2003; 50:366 369. 10. Ohler C, Mughannam A, Reinke J, Spangler W. Transient hemi-inattention in a dog with metastatic renal hemangiosarcoma. J Am Anim Hosp Assoc 1994;30:207 212. 11. Mott JC, McAnulty JF, Darien DL, Steinberg H. Nephron sparing by partial median nephrectomy for treatment of renal hemangioma in a dog. J Am Vet Med Assoc 1996;208:1274 1276. 12. Haziroglu R, Kul O, Tunca R, Guvenc T. Osteoclast-like giant tumour arising from the kidney in a dog. Acta Vet Hung 2005;53:225 230. 13. Dempster AG, Delahunt B, Malthus AW, Wakefield JS. The histology and growth kinetics of canine renal oncocytoma. J Comp Pathol 2000;123:294 298. 14. Buergelt CD, Adjiri-Awere A. Bilateral renal oncocytoma in a Greyhound dog. Vet Path 2000;37:188 192. 15. Eddlestone S, Taboada J, Senior D, Paulsen DB. Renal haemangioma in a dog. J Small Anim Pract 1999;40:132 135. 16. Robison RL, Grosenstein PA, Argentieri GJ. Mixed mesenchymal tumor in the kidney of a young beagle dog. Toxicol Pathol 1997;25:326 328. 17. Hayes HM Jr, Fraumeni JF Jr. Epidemiological features of canine renal neoplasms. Cancer Res 1977;37:2553 2556. 18. Jonasdottir TJ, Mellersh CS, Moe L, et al. Genetic mapping of a naturally occurring hereditary renal cancer syndrome in dogs. Proc Natl Acad Sci USA 2000;97:4132 4137. 19. Everitt JI, Goldsworthy TL, Wolf DC, Walker CL. Hereditary renal cell carcinoma in the Eker rat: A unique animal model for the study of cancer susceptibility. Toxicol Lett 1995;82 83:621 625. 20. Cerilli LA, Huffman HT, Anand A. Primary renal angiosarcoma: A case report with immunohistochemical, ultrastructural, and cytogenetic features and review of the literature. Arch Pathol Lab Med 1998;122:929 935. 21. Crow S. Urinary tract neoplasms in dogs and cats. Comp Cont Educ Pract Vet 1985;7:607 618. 22. Meuten DJ. Tumors of the urinary system. In: Meuten DJ, ed. Tumors in Domestic Animals, 4th ed. Ames, IA: Iowa State Press; 2002:209 546. 23. Seaman RL, Patton CS. Treatment of renal nephroblastoma in an adult dog. J Am Anim Hosp Assoc 2003;39:76 79. 24. Tsuda N, Chowdhury PR, Hayashi T, et al. Primary renal angiosarcoma: A case report and review of the literature. Pathol Int 1997;47:778 783. 25. Peeters D, Clercx C, Thiry A, et al. Resolution of paraneoplastic leukocytosis and hypertrophic osteopathy after resection of a renal transitional cell carcinoma producing granulocyte-macrophage colony-stimulating factor in a young bull terrier. J Vet Intern Med 2001;15:407 411. 26. Russo P. Renal cell carcinoma: Presentation, staging, and surgical treatment. Semin Oncol 2000;27:160 176. 27. Mydlo JH, Gerstein M. Patients with urologic cancer and other nonurologic malignancies: Analysis of a sample and review of the literature. Urology 2001;58:864 869. 28. Koenig A, Bianco SR, Fosmire S, et al. Expression and significance of p53, rb, p21/waf-1, p16/ink-4a, and PTEN tumor suppressors in canine melanoma. Vet Pathol 2002;39: 458 472. 29. Lee CH, Kweon OK. Mutations of p53 tumor suppressor gene in spontaneous canine mammary tumors. J Vet Sci 2002;3: 321 325. 30. Loukopoulos P, Thornton JR, Robinson WF. Clinical and pathologic relevance of p53 index in canine osseous tumors. Vet Pathol 2003;40:237 248. 31. Murakami Y, Tateyama S, Rungsipipat A, et al. Immunohistochemical analysis of cyclin A, cyclin D1 and P53 in mammary tumors, squamous cell carcinomas and basal cell tumors of dogs and cats. J Vet Med Sci 2000;62:743 750. 32. Flanigan RC, Yonover PM. The role of resection for patients with renal carcinoma. Curr Oncol Rep 2001;3:424 432.