Canine histiocytic sarcoma (HS) is a rare neoplasm,

Transcription

1 J Vet Intern Med 2007;21: CCNU for the Treatment of Dogs with Histiocytic Sarcoma Katherine A. Skorupski, Craig A. Clifford, Melissa C. Paoloni, Ana Lara-Garcia, Lisa Barber, Michael S. Kent, Amy K. LeBlanc, Aarti Sabhlok, Elizabeth A. Mauldin, Frances S. Shofer, C. Guillermo Couto, and Karin U. Sørenmo Background: Histiocytic sarcoma is an aggressive neoplasm of dendritic cells that carries a grave prognosis. The efficacy of chemotherapy against this disease is unknown. The purpose of this study was to determine the efficacy of 1-(2-chloroethyl)-3- cyclohexyl-1-nitrosourea (CCNU) in dogs with incompletely resected or metastatic histiocytic sarcoma, to describe the clinical characteristics of these dogs, and to identify factors affecting prognosis. Hypothesis: Our hypothesis is that CCNU has activity against canine histiocytic sarcoma and can improve survival in dogs with advanced disease. Animals: Included in analysis are dogs diagnosed with histiocytic sarcoma who had gross measurable or residual microscopic disease and who received CCNU. Methods: A multi-institutional, retrospective, single-arm cohort study was conducted. Available biopsy samples were tested with an antibody against CD18 when possible to confirm the diagnosis of histiocytic sarcoma. Results: Fifty-nine dogs were treated at 8 institutions. Twenty-three tumor specimens were confirmed to be CD18 positive. Treatment with CCNU at 60 to 90 mg/m 2 resulted in an overall response rate of 46% in the 56 dogs with gross measurable disease. All 3 dogs with minimal residual disease experienced tumor relapse but lived 433 days or more after starting CCNU. The median survival of all 59 dogs was 106 days. Thrombocytopenia (,100,000 platelets/ml) and hypoalbuminemia were found to be negatively associated with prognosis and were predictive of,1 month survival. Conclusions and Clinical Importance: Results suggest that CCNU is active against canine histiocytic sarcoma and may be useful in the treatment of dogs without negative prognostic factors. Key words: CD18; Dendritic cell neoplasia; Lomustine; Malignant histiocytosis. Canine histiocytic sarcoma (HS) is a rare neoplasm, representing less than 1% of canine cancers of the lymphoreticular system. 1 The cell of origin recently was determined to be the antigen presenting dendritic cell by extensive immunohistochemical evaluation with antibodies against various subunits of the major adhesion molecule family of leukocytes. 2 Cluster of differentiation (CD) 18 is the b 2 subunit of the adhesion molecule that From the Departments of Clinical Studies (Skorupski, Shofer, Sørenmo) and Pathobiology, (Mauldin), Matthew J. Ryan Veterinary Hospital, University of Pennsylvania Philadelphia, PA; Department of Veterinary Clinical Sciences and Veterinary Teaching Hospital (Lara-Garcia, Couto), College of Veterinary Medicine, The Ohio State University, Columbus, OH; Department of Oncology and Radiation Therapy (Clifford), Red Bank Veterinary Hospital, Tinton Falls, NJ; Department of Medical Sciences (Paoloni), School of Veterinary Medicine, University of Wisconsin, Madison, WI; Department of Clinical Sciences (Barber), Tufts Cummings School of Veterinary Medicine, North Grafton, MA; Department of Surgical and Radiological Sciences (Kent), School of Veterinary Medicine, University of California, Davis, CA; Department of Oncology (LeBlanc), Florida Veterinary Specialists and Cancer Treatment Center, Tampa, FL; and Department of Oncology (Sabhlok), Animal Specialty Group, Inc, Los Angeles, CA. Dr Skorupski s current address is University of California, Davis, CA. Dr Paoloni s current address is National Cancer Institute, Bethesda, MD. Dr LeBlanc s current address is University of Tennessee, Knoxville, TN. A portion of this manuscript was presented at the Veterinary Cancer Society annual meeting, Madison, WI, September 26-29, Reprint requests: Dr Katherine Skorupski, DVM, Dipl. ACVIM, Veterinary Medicine: Surgical and Radiological Sciences, Room 2112, Tupper Hall, University of California, 1 Shields Ave, Davis, CA 95616; kskorups@ucdavis.edu. Submitted February 6, 2006; Revised May 1, 2006; June 20, 2006; Accepted August 9, Copyright E 2007 by the American College of Veterinary Internal Medicine /07/ /$3.00/0 is expressed most abundantly on cells of histiocytic origin, and antibodies against this molecule can be targeted against formalin-fixed tissues. HS may present with either localized organ involvement or disseminated, multi-organ involvement. Reported anatomical sites include lung, lymph node, liver, spleen, stomach, pancreas, mediastinum, skin, skeletal muscle, central nervous system, bone and bone marrow. 3 7 Five percent of primary brain tumors in a clinical population and 4.5% of secondary brain tumors in a postmortem population were HS. 8,a In addition, 18 of 35 tumors of synovial origin previously diagnosed as synovial cell sarcomas were reclassified as HS based on immunohistochemical expression of CD18. 7 HS currently is the preferred term identifying all malignancies arising from dendritic cells, and the older term malignant histiocytosis refers only to the disseminated form of HS. 9 Shortly after canine HS was first reported in the late 1970s, a predisposition in male Bernese Mountain Dogs was noted. 6 This neoplasm has since been identified in a variety of breeds and Flat Coated Retrievers and Rottweilers also are believed to be overrepresented. Affected dogs commonly are middle-aged or older, but HS has been reported in dogs as young as 3 years of age. 4 Canine HS has an aggressive clinical course and is uniformly fatal. In previous reports of untreated dogs with unresectable or metastatic disease, death, or euthanasia usually occurred within days of diagnosis. 3 5 A few reports of responses to chemotherapy with doxorubicin, liposomal doxorubicin, or paclitaxel exist. 10,11 A case report of a dog with cutaneous disseminated HS documented temporary remissions resulting from multiple protocols, including cyclophosphamide, vincristine, prednisone, mitoxantrone, dacarbazine, and etoposide. 12 In addition, durable clinical remissions and prolonged survival have been documented in 4 dogs with HS treated systemically with immunotherapy consisting of a human cytotoxic T-cell line (TALL-104). 13

2 122 Skorupski et al 1-(2-Chloroethyl)-3-cyclohexyl-1-nitrosourea (CCNU) b is a long-acting nitrosourea alkylating agent that has been used to treat canine lymphoma, mast cell tumors, and intracranial tumors To the authors knowledge, the use and effect of CCNU for HS have not been evaluated. The purpose of this study was to determine the efficacy of CCNU in treatment of dogs with incompletely resected or metastatic histiocytic sarcoma. Secondary goals were to describe clinical characteristics and to identify factors affecting prognosis in this population. Materials and Methods A multi-institutional, retrospective, single-arm cohort study was conducted. Survey questionnaires were submitted to participating institutions, including 5 veterinary universities and 3 private referral practices. Data were collected retrospectively by participants at each institution by review of medial records and follow-up calls to owners and referring veterinarians when necessary. Information collected in questionnaires included treating institution, signalment, presenting signs, method and date of diagnosis, results of initial blood work and staging tests, CCNU dose and dosing interval, total number of doses, treatments administered in addition to CCNU, requirement for dose reduction or delay, response, response duration, reason for discontinuing CCNU therapy, survival, cause of death, and postmortem examination information if available. For inclusion into the study, patients must have had a definitive histopathologic or cytologic diagnosis of HS, gross measurable disease or documented residual microscopic disease, treatment with CCNU, and adequate follow-up to assess response. Cases were excluded if dogs were treated with CCNU as a part of a multi-agent chemotherapeutic protocol (excluding corticosteroids), but dogs may have been treated with other chemotherapeutics before or after CCNU treatment. Dogs treated with radiation therapy were excluded unless response to CCNU therapy could be assessed in tumor lesions outside the radiation field. Responses were defined as follows: complete response (CR), 100% reduction in measurable tumor volume; partial response (PR),.50% but,100% reduction; stable disease (SD),,50% reduction or,10% increase in tumor volume and no new lesions; progressive disease (PD), increase of.10% in tumor volume or the appearance of new lesions. A minimum response duration of 21 days was required for CR and PR and dogs that died before their first reevaluation were considered to have PD. Biopsy specimens were obtained when available and were reviewed by 1 pathologist (EAM). Immunohistochemistry using an antibody against CD18 was performed on these specimens for confirmation of diagnosis. For tumor samples that had previously been tested for CD18 expression at the submitting institution, immunohistochemistry was not repeated. Cases were excluded if reviewed histologic samples were not morphologically consistent with a diagnosis of HS or if they did not express CD18. The following factors were evaluated for effect on response to CCNU and survival: age, sex, anemia (PCV, 30%), thrombocytopenia (,100,000/mL), hypercalcemia, hypoalbuminemia, visceral involvement, thoracic involvement, splenic involvement, bone involvement, prednisone administration, prior chemotherapy, and requirement for dose reduction or delay. Abnormal serum calcium and albumin concentrations were defined as any result outside of the reference range at the treating institution. These factors also were evaluated for effect on remission duration in dogs who achieved CR or PR. Only dogs who received more than 1 dose of CCNU were included when comparing requirements for dose reduction or delay. Toxicities were graded according to the published Veterinary Co-operative Oncology Group criteria for adverse events. 17 Survival was calculated from the date of the first treatment with CCNU to the date of death, and response duration was calculated from the date response was noted to the date of relapse. Dogs who were still alive at the completion of the study or who were lost to follow-up were censored at the last date they were reported to be alive or at last known follow-up. Remission duration for dogs who were alive and in remission at study s completion was censored at the last date they were reported to be in remission. The Kaplan-Meier product limit method was used to estimate the proportion of dogs who were alive or had died from HS. The logrank test was used to determine differences in survival time or remission duration associated with various prognostic factors. To assess the independent contributions of the various prognostic factors, a forced entry Cox proportional hazards model was developed. Prognostic factors with a P value less than.1 were included in the hazards model. The Fisher exact test was used to compare the frequency of prognostic factors with response to therapy. Statistical significance was defined as P value less than.05. All analyses were performed by using SAS statistical software. c Results Fifty-nine dogs with HS diagnosed between 1999 and 2004 were included in data analysis. Institutions and numbers of cases per institution included the Ohio State University (14); University of Pennsylvania (13); University of Wisconsin, Madison (8); Tufts University (7); University of California, Davis (6); Red Bank Veterinary Hospital (5); Florida Veterinary Specialists (3); and Animal Specialty Group (3). The mean age of the dogs was 8.0 years (range, 2 14 years). Four dogs were intact females, 25 were spayed females, 7 were intact males, and 23 were castrated males. There were 13 Golden Retrievers, 13 Rottweilers, 11 Bernese Mountain Dogs, 10 Labrador Retrievers, 8 other purebred dogs, and 4 mixed breed dogs. The mean weight was 35.7 kg (range, kg). Clinical signs at presentation included 1 or more of the following: lethargy (39%, n 5 23), presence of a mass (36%, n 5 21), cough or dyspnea (32%, n 5 19), inappetence (31%, n 5 18), lameness (25%, n 5 15), weight loss (12%, n 5 7), fever (8%, n 5 5), vomiting (10%, n 5 6), diarrhea (7%, n 5 4), lymphadenopathy (7%, n 5 4), polyuria and polydipsia (5%, n 5 3), epistaxis (2%, n 5 1), and ocular signs (2%, n 5 1). Before treatment, all 59 dogs had a CBC and serum biochemistry profile, 57 dogs had thoracic radiographs, 52 dogs underwent abdominal ultrasonography, and 11 dogs had a bone marrow aspirate evaluated. Clinicopathologic abnormalities noted at presentation included anemia (PCV, 30%) in 29% (n 5 17), thrombocytopenia (,100,000 platelets/ml) in 22% (n 5 13), neutrophilia with or without a left shift in 24% (n 5 14), hypercalcemia in 3% (n 5 2), hypoalbuminemia in 27% (n 5 16), hyperglobulinemia in 7% (n 5 4), and mild or moderate increases in liver enzyme activity in 34% (n 5 20). The distribution of documented tumor lesions is summarized in Table 1. Six dogs had solitary tumor lesions, 22 dogs had lesions at 2 sites, and 31 dogs had lesions at 3 or more anatomic sites. Eighty-five percent of dogs had visceral involvement. Dogs were treated with a mean starting CCNU dosage of 70.8 mg/m 2 (median, 70 mg/m 2 ; range, mg/m 2 ), with dosing intervals of 3 to 4 weeks. Sixteen

3 CCNU for Histiocytic Sarcoma 123 Table 1. Distribution of documented tumor lesions. Organ Affected N % Dogs Lymph nodes Liver Lung Spleen Skin and soft tissues Bone/joint Kidney 6 10 Mediastinum 6 10 Pleura (effusion) 5 8 Bone marrow 3 5 Nasal passages 1 2 Eye 1 2 dogs (27%) were treated at 60 or 65 mg/m 2, 32 dogs (54%) at 70 or 75 mg/m 2, 5 dogs (8%) at 80 or 85 mg/m 2, and 6 dogs (10%) at 90 mg/m 2. A median of 4 doses was administered to each dog (range, 1 20 doses). Twentyfour dogs required dose reduction or delay because of one or more episodes of toxicity. There were 15 dose reductions because of neutropenia, including 1 dog with grade 1, 3 dogs with grade 2, 4 dogs with grade 3, and 7 dogs with grade 4 toxicity. Three dogs experienced thrombocytopenia, including one with grade 1 and 2 with grade 4 toxicity. Six dogs experienced increased liver enzyme activity, including 1 with grade 2, 3 with grade 3 and 2 with grade 4 increases in alanine aminotransferase activity. Two dogs experienced grade 2 vomiting, and 3 dogs experienced diarrhea: 1 had grade 1 and 2 had grade 2 diarrhea. One dog had dose reduction because of a grade 1 weight loss and 1 dog experienced grade 1 lethargy. In addition to the 24 dogs that had dose reductions, 1 dog had grade 4 neutropenia, resulting in sepsis and death after 1 dose of CCNU. Fifteen dogs underwent surgery before CCNU chemotherapy. Five dogs had debulking procedures, and 10 had attempted complete excision of their tumors. Five dogs received radiation therapy. Three had palliative fractionation, and 2 had definitive radiation therapy. Nineteen dogs had other chemotherapy, including doxorubicin, liposomal doxorubicin, dacarbazine, cyclophosphamide, vincristine, L-asparaginase, procarbazine, or etoposide. Seven of these dogs were treated with these drugs before receiving CCNU, and 12 dogs were given these drugs after discontinuing CCNU. Thirty-five dogs were treated with corticosteroids concurrent with CCNU, and 2 additional dogs received corticosteroids at relapse. Six dogs were treated with nonsteroidal anti-inflammatory drugs. Diagnosis was obtained histologically in 36 cases and cytologically in 23 cases. Biopsy samples were available for review in 13 cases, and the diagnosis was confirmed morphologically and by CD18 immunostaining and reviewed by a single pathologist (EAM). Twenty-three of the cases, including tissues stained by the treating institution, were evaluated for expression of CD18 by using immunohistochemistry or immunocytochemistry, and all were positive. Fifty-six dogs had gross measurable disease, and response to treatment could be objectively measured. The 3 remaining dogs had residual microscopic disease after tumor resection, which could not be evaluated for response to treatment. Survival was censored for 12 dogs, because 10 were alive at the time of manuscript preparation and 2 were lost to follow-up. Five dogs were alive and in remission at the time of manuscript preparation and their remission durations were censored. Of the 56 dogs with gross measurable disease, 5 achieved CR, 21 had PR, 10 had SD, and 20 had PD, resulting in an overall response rate of 46%. The response rate in dogs with tumors confirmed to be CD18 positive was 62%. The response rate in 50 dogs that survived a minimum of 7 days after the first dose of CCNU was 54%. The median remission duration for the 26 dogs who responded was 85 days (range, days). The median survival time (MST) for responders was 172 days versus 60 days in dogs with SD or PD. For all 59 dogs included in the study, the MST was 106 days (range, days). In the 3 dogs with nonmeasurable residual disease, relapse occurred in all 3 dogs at 164, 231, and 250 days. One dog died at 568 days, and 2 were alive at 433 and 884 days. Univariate analysis identified anemia, thrombocytopenia, hypoalbuminemia, and splenic involvement to be factors significantly associated with shorter survival times (Table 2). Anemic dogs lived a median of 28 days compared with 163 days in dogs without anemia, and dogs with thrombocytopenia lived a median of 21 days compared with 165 days in dogs without thrombocytopenia. Hypoalbuminemic dogs lived a median of 28 days compared with 163 days in dogs with normal serum albumin concentration. Dogs with splenic involvement lived a median of 58 days compared with 165 days in dogs with normal spleens. The proportional hazards model demonstrated that only thrombocytopenia and hypoalbuminemia were negatively predictive of longterm survival. Hypoalbuminemic dogs died at 2.1 times the rate of dogs with normal albumin concentrations throughout the study period (P 5.02), and thrombocytopenic dogs died at 4.1 times the rate of dogs without thrombocytopenia (P ). Dogs without anemia, thrombocytopenia, and splenic involvement also were significantly more likely to achieve CR or PR. The response rate in dogs that were not anemic at diagnosis was 56% compared with 24% in anemic dogs (P 5.04). The response rate in dogs that were not thrombocytopenic was 56% compared with 15% in thrombocytopenic dogs (P 5.01). The response rate in dogs with normal spleens was 59% compared with 27% in dogs with splenic involvement (P 5.03). The response rate was 55% in dogs with normal albumin concentrations compared with 25% in dogs with low albumin concentrations, but this result was not statistically significant (P 5.07). Age, sex, hypercalcemia, visceral involvement, thoracic involvement, bone involvement, prednisone administration, prior chemotherapy, and requirement for dose reduction or delay were not significantly associated with response to CCNU or survival. Likewise, no significant associations were found between any

4 124 Skorupski et al Table 2. Variables examined for relation to survival. Variable factors and remission duration in dogs achieving PR or CR. Discussion The clinical course of metastatic HS has previously been reported to be rapid and uniformly fatal. There are only a few case reports of responses to various chemotherapy drugs. Our results suggest that CCNU is an active agent in the treatment of canine HS, with a 46% response rate in the gross disease setting. The dogs in this cohort were of advanced stage with 54 having metastatic or multifocal disease. The remaining 5 dogs had solitary tumors that were deemed unresectable by the treating oncologist because of location or size. The poor overall survival time (106 days) is, therefore, expected in this population. The use of CCNU as a part of a multi-agent protocol may improve response rates and survival times in dogs with advanced disease, and further study is warranted. N Median Survival (days) P Value Age at diagnosis.3 #8 years years Sex.58 Female Male Anemia (hematocrit,30%).0009 No Yes Thrombocytopenia (,100,000/mL),.0001 No Yes Hypercalcemia.52 No Yes Hypoalbuminemia.0017 No Yes Visceral involvement.19 No Yes Thoracic involvement.36 No Yes Splenic involvement.048 No Yes Bone involvement.21 No Yes Prednisone use.17 No Yes Prior chemotherapy.23 No Yes Dose reduction or delay.43 No Yes Few reports documenting survival duration after surgical excision of localized HS exist. In 1 series of 18 synovial HS confirmed with CD18 staining, the overall MST was 3.6 months, the MST for dogs undergoing amputation (with or without chemotherapy) was 6 months, and the metastatic rate was 91%. 7 Although only 3 dogs in this study had microscopic disease at the time of CCNU treatment, the survival times in these dogs were much longer than expected (.14 months) when compared with outcome in dogs with synovial HS treated with surgical excision. The first of these dogs had a pulmonary mass that was resected. The tracheobronchial lymph nodes containing metastatic disease were biopsied but not excised. The second dog with a distal humeral tumor underwent amputation, but metastasis to multiple axillary lymph nodes was noted in biopsy specimens obtained at surgery. The third dog with pulmonary HS and probable splenic metastasis had incomplete excision of the pulmonary mass. Further study of adjuvant CCNU to treat residual microscopic disease is warranted. Five dogs in this study achieved a complete response to CCNU therapy. Three of these dogs were alive and in remission at the time of manuscript preparation, with censored survival times of 105, 165, and 368 days. Another dog achieved a CR that lasted 79 days and survived 103 days from the start of treatment. The remaining dog with a CR died 845 days after diagnosis from complications related to a protein-losing nephropathy. No evidence of HS was found on postmortem examination. To the authors knowledge, this is the only reported case of a long term remission of HS achieved with chemotherapy alone. Thrombocytopenia and hypoalbuminemia were significantly associated with a grave (,1 month MST) prognosis despite CCNU therapy in this report. Because of the retrospective nature of this study, urinalysis and coagulation profile data were not available in most patients, and it is difficult to determine the exact cause of these laboratory abnormalities. Survival times were similar between patients who received corticosteroids and those who did not. This finding suggests that corticosteroids may not have specific antineoplastic activity against malignant histiocytes, although the timing of steroid use and dose were not evaluated in this study. Anatomic site(s) of tumor involvement were not important predictors of survival. Splenic involvement was associated with a lack of response to CCNU and with poorer survival in univariate analysis but was not statistically significant in multivariate analysis. Bone involvement, which was noted in 11 dogs, was not associated with a more favorable prognosis, despite the 6-month MST reported by Craig et al 7 in treated dogs with synovial HS. This finding may be because all but 1 of the dogs with bone involvement in our study had documented metastatic disease to local lymph nodes, viscera, or both. Dogs with HS localized to 1 limb who underwent amputation and who received adjuvant CCNU were excluded by study design. Because of the retrospective nature of this study, many tumor samples were not available for review to

5 CCNU for Histiocytic Sarcoma 125 confirm a diagnosis of HS. In addition, CD18 expression was confirmed in only 23 cases. Reaching a definitive diagnosis of HS can be difficult in pleomorphic tumors that have morphologic characteristics similar to other sarcomas, carcinomas, or round cell tumors. Cytologic diagnosis can be particularly challenging, especially in sparsely cellular samples. 18 Although cases were excluded when the pathologist or the clinical pathologist was unable to rule out other tumor types with reasonable certainty, it is possible that some dogs included in the study did not have HS. The 62% response rate in dogs with CD18 confirmed HS, however, suggested that the results of this study were not weakened appreciably by possible inclusion of more chemosensitive tumors. Despite the median duration of CR and PR in this series of 84 days, the MST for these dogs was 172 days. Remission duration was calculated from the time response was documented by the treating oncologist, and, because most dogs were only examined by the treating oncologist on dates CCNU was to be administered, a 21- to 28-day period often elapsed between the first dose of CCNU and documentation of response. This factor accounts for some of the disparity between the median response duration and MST, but chemotherapy administered after dogs experienced PD on CCNU also may have increased MST. Seven of the 27 dogs who had PR or CR to CCNU received additional chemotherapy, and 1 of these 7 also had palliative radiation after relapse. Responses and response durations to these rescue therapies were not available on the completed study questionnaires, so it is not possible to determine the exact effect of these treatments on MST. Recently, it has been proposed that a subtype of HS exists. Termed hemophagocytic HS, these tumors are differentiated by their immunohistochemical staining pattern consistent with macrophages (ie, CD18+, CD11d+, CD11c2, MHCII+, minimal CD1 expression), rather than dendritic cells. d Clinically, these tumors have been seen in the spleen and bone marrow and may behave more aggressively because of their cellular ability to phagocytose material, including host red blood cells. It is possible that some dogs in this study had the hemophagocytic subtype, but fresh or frozen tissues would be required to determine this definitively. Additional study is needed to elucidate whether dogs with the hemophagocytic subtype of HS have poorer responses to CCNU therapy. Conclusion HS is an aggressive disease with a poor prognosis in dogs with unresectable or metastatic disease. Treatment with CCNU results in considerable improvement in approximately a half of dogs with gross disease, and dogs with microscopic metastatic disease can have long-term survivals. Clinical data, including serum albumin concentration, platelet count, and possibly anemia and splenic involvement, should be used to prognosticate for dogs with HS. Dogs with poor prognostic factors have a grave prognosis despite CCNU therapy. Footnotes a Mikszewski J, Lipitz L, Skorupski K, et al. Canine secondary intracranial neoplasia: A retrospective study. In: Proceedings of the American College of Veterinary Internal Medicine annual meeting, June 1 4, 2005, Baltimore, MD b Lomustine, Bristol Laboratories, Princeton, NJ c SAS, Version 9.1, SAS Institute, Cary, NC d Moore P. The histiocytic disease complex. In: Proceedings of the American College of Veterinary Internal Medicine annual meeting, June 9 12, 2004, Minneapolis, MN Acknowledgments The authors thank Dr Lili Duda for her assistance in manuscript preparation. This work was funded in part by the Departmental Research Resources Committee of the Matthew J. Ryan Veterinary Hospital of the University of Pennsylvania. References 1. Vail DM. Histiocytic disorders. In: Withrow S, MacEwen E, eds. Small Animal Clinical Oncology, 3rd ed. Philadelphia, PA: WB Saunders; 2001: Affolter VK, Moore PF. Localized and disseminated histiocytic sarcoma of dendritic cell origin in dogs. Vet Pathol 2002;39: Hayden DW, Waters DJ, Burke BA, Manivel JC. Disseminated malignant histiocytosis in a Golden Retriever: Clinicopathologic, ultrastructural, and immunohistochemical findings. Vet Pathol 1993;30: Kohn B, Arnold P, Kaser-Hotz B, et al. Malignant histiocytosis of the dog: 26 cases ( ). Kleintierpraxis 1993;38: Fant P, Caldin M, Furlanello T, et al. Primary gastric histiocytic sarcoma in a dog A case report. J Vet Med 2004;51: Rosin A, Moore P, Dubielzig R. Malignant histiocytosis in Bernese Mountain dogs. J Am Vet Med Assoc 1986;188: Craig LE, Julian ME, Ferracone JD. The diagnosis and prognosis of synovial tumors in dogs: 35 cases. Vet Pathol 2002;39: Vernau KM, Higgins RJ, Bollen AW, et al. Primary canine and feline nervous system tumors: Intraoperative diagnosis using the smear technique. Vet Pathol 2001;38: Jaffe ES. Histiocytic and dendritic cell neoplasms. In: Jaffe ES, Harris HL, Stein H, Vardiman J, eds. World Health Organization Classification of Tumors: Pathology and Genetics of Tumors of Haematopoietic and Lymphoid Tissues, 3rd ed. Lyon, France: IARC Press; 2001: Vail DM, Kravis LD, Cooley AJ, et al. Preclinical trial of doxorubicin entrapped in sterically stabilized liposomes in dogs with spontaneously arising malignant tumors. Cancer Chemother Pharmacol 1997;39: Poirier VJ, Hershey AE, Burgess KE, et al. Efficacy and toxicity of paclitaxel (Taxol) for the treatment of canine malignant tumors. J Vet Intern Med 2004;18:

6 126 Skorupski et al 12. Uno Y, Momio Y, Watari T, et al. Malignant histiocytosis with multiple skin lesions in a dog. J Vet Med Sci 1993;55: Visonneau S, Cesano A, Tran T, et al. Successful treatment of canine malignant histiocytosis with the human major histocompatibility complex nonrestricted cytotoxic T-cell line TALL-104. Clin Cancer Res 1997;3: Moore A, London C, Wood C, et al. Lomustine (CCNU) for the treatment of resistant lymphoma in dogs. J Vet Intern Med 1999;13: Rassnick K, Moore A, Williams L, et al. Treatment of canine mast cell tumors with CCNU (lomustine). J Vet Intern Med 1999;13: Fulton L, Steinberg H. Preliminary study of lomustine in the treatment of intracranial masses in dogs following localization by imaging techniques. Semin Vet Med Surg (Small Anim) 1990;5: Veterinary co-operative oncology group Common criteria for adverse events (VCOG-CTCAE) following chemotherapy or biologic antineoplastic therapy in dogs and cats v1.0. Veterinary Co-operative Oncology Group Consensus Document. Vet Comp Oncol 2004;2: Brown D, Thrall M, Getzy D, et al. Cytology of canine malignant histiocytosis. Vet Clin Pathol 1994;23: