Carmustine, Vincristine, and Prednisone in the Treatment of Canine Lymphosarcoma
A chemotherapeutic protocol using carmustine in combination with vincristine and prednisone was tested in dogs with multicentric malignant lymphosarcoma. Of seven dogs treated, six (85.7%) achieved complete remission. A partial response occurred in one dog. Median survival time was 224 days (mean 386 days), and median duration of remission was 183 days (mean 323 days). Marked neutropenia was observed following carmustine administration. There were no significant alterations in platelets and red blood cell counts during treatment, and no abnormalities attributable to the chemotherapy were found in serum biochemical profiles. Results of this study showed that carmustine is an effective alternative option in the treatment of canine lymphosarcoma.
Introduction
Lymphosarcoma is the most common hematopoietic neoplasm of dogs, representing 83% of all hematological malignancies.1 Over the past 30 years, many articles have been published regarding the epidemiological, clinical, histopathological, immunological, and therapeutic aspects of this disease.2–8 Malignant lymphosarcoma is one of the malignancies most likely to respond to chemotherapy in dogs, and systemic chemotherapy is the treatment of choice.9 Numerous protocols utilizing sequential or simultaneous combinations of chemotherapeutic agents have been reported, and they have resulted in variable remission and survival times.10 The advantage of combination chemotherapy over single-agent chemotherapy is that the different mechanisms of action of the various drugs have greater tumoricidal efficacy and lower toxicity rates.11
Although a wide variety of drugs and protocols have been effective in inducing remission, there is no current consensus regarding the best treatment in terms of efficacy, remission induction and survival times, toxicity, and overall cost. Consequently, the optimal protocol for use in all dogs with lymphosarcoma is not known.
Carmustine (1,3-Bis [2-chloroethyl]-1-nitrosourea) or BCNU is a highly lipid-soluble, low-molecular weight chemotherapeutic agent in the nitrosourea subclass. It is lipophilic, crosses the blood-brain barrier, and acts as a nonphase-specific alkylating agent.1213 In humans, it is commonly used in the treatment of neurological tumors (e.g., malignant gliomas), multiple myeloma, lymphosarcomas, and some forms of gastrointestinal neoplasia. The recommended dosage in people is 200 to 300 mg/m2 intravenously (IV) q 6 weeks.13–16
After administration, spontaneous decomposition of the carmustine molecule occurs, yielding an isocyanate compound and a diazonium hydroxide molecule.14 The chloroethyl diazonium ion or the chloroethyl carbonium ion generated then alkylates biological molecules. Bone marrow suppression is the most common dose-limiting toxicity in humans. It usually manifests as neutropenia and thrombocytopenia, with the nadir at 3 to 4 weeks.13 Additionally, the risk of fibrosing alveolitis and interstitial pneumonia is high when the cumulative dose of carmustine is ≥1200 to 1400 mg/m2.1316
In dogs, carmustine can be used at a dosage of 50 mg/m2 IV q 6 weeks, but there has been no protocol established for the use of carmustine in the treatment of canine lymphosarcoma.1215 In a previous unpublished study, carmustine (50 mg/m2) was used as a single agent to treat three dogs with lymphosarcoma.a One dog had a complete remission, and two dogs had partial remissions. The duration of remission was <30 days. Nevertheless, the evidence of antitumor activity in these cases suggested that carmustine could be used in a combination chemotherapeutic protocol.
Cyclophosphamide, vincristine (i.e., Oncovin), and prednisone (COP) comprise a commonly used chemotherapeutic protocol for the treatment of canine lymphosarcoma. In this study, carmustine replaced cyclophosphamide in a traditional COP protocol. Both drugs act as alkylating agents.3 The purpose of the present study was to assess a combination chemotherapeutic protocol using carmustine, vincristine, and prednisone in the treatment of canine lymphosarcoma.
Materials and Methods
Case Selection
Seven dogs with malignant multicentric lymphosarcoma, stage IV to V (according to the World Health Organization [WHO] classification), were treated at the Veterinary Medical Teaching Hospital of the University of São Paulo, Brazil.17 The diagnosis of lymphosarcoma was initially based on cytological examination of a fine-needle aspirate of an enlarged lymph node and confirmed by histopathological examination of a surgically excised lymph node. For the purpose of WHO staging, each dog underwent a physical examination and had a complete blood count (CBC), thoracic radiography, abdominal ultrasonography, and a bone marrow aspirate performed. Dogs that had previously received chemotherapy or prednisone, or had concomitant life-threatening diseases were excluded from the study.
Treatment Protocol
After clinical evaluation and staging, each dog received induction chemotherapy consisting of carmustineb (50 mg/m2 IV), vincristinec (0.75 mg/m2 IV), and prednisone (40 mg/m2 per os [PO] q 24 hours) during the first week. Subsequently, the dogs received vincristine once weekly for 3 weeks and then at 3-week intervals; prednisone (20 mg/m2) on alternate days; and carmustine at 6-week intervals until the end of the treatment period [Table 1]. A CBC was performed before every injectable treatment, and treatment was delayed if the neutrophil count was <2.5 × 103 cells/μL. The carmustine solution was diluted in 150 to 250 mL of 0.9% sodium chloride and infused IV initially over a 30- to 40-minute period.
The induction period was considered as the time from the first dose of chemotherapy (week 1) through week 4. Tumor response was assessed weekly by measuring solid tumors, and overall response was determined on day 22 (at the end of the induction period). Overall response was classified as a complete remission (CR) if there were no detectable solid tumors; a partial remission (PR) if the size of the tumors decreased by >50%; or no response (NR) if there was <50% reduction in tumor(s) size.
Remission duration was calculated from the first day the remission was identified until a relapse of clinical signs. Survival time was calculated from the first day of therapy until the dog died spontaneously or was euthanized.
Laboratory Tests
Pretreatment laboratory tests included a CBC, serum biochemical profile (urea, creatinine, protein, albumin, alanine aminotransferase [ALT], alkaline phosphatase [ALP], γ-glutamyl transferase [GGT], calcium), and urinalysis. A CBC was also performed before the administration of carmustine and vincristine. Urinalyses were performed at 6-week intervals or when the dogs showed alterations in the color, odor, or volume of their urine. Serum biochemical profiles were performed at least monthly during treatment.
Results
The median age of the dogs was 4 years (range, 3 to 7 years). Three were male and four were female. The median body weight of the dogs was 27 kg (range, 7 to 54 kg). All dogs had peripheral lymphadenopathy and splenic and/or liver disease. Six dogs presented with bone marrow involvement. Three dogs were classified by WHO criteria as having stage Va lymphosarcoma, three dogs had stage Vb, and one dog had stage IVa disease [Table 2]. The histopathological grade of the lymph node tumors was available for all dogs [Table 2]. There were three high-grade and four intermediate-grade tumors. Only one dog had hypercalcemia at the time of diagnosis.
The dogs received two to seven treatments (median 5.0) of carmustine plus vincristine and four to eight treatments (median 6.0) of only vincristine. Cumulative doses of carmustine ranged from 100 to 298 mg/m2. The dogs were treated for different periods of time. The original protocol was designed to last 1 year, but in some cases, the owners chose to stop the chemotherapy earlier. Case no. 1 was treated for only 9 months (270 days), but it remained in remission for 37.8 months. Case no. 2 was treated for 10 months and stayed in remission for 19.5 months [Table 2].
Clinical improvement occurred in all dogs within 7 days after the first dose of chemotherapy. Six (85.7%) dogs achieved CR after the induction phase (week 4). Partial remission was observed in one dog. The duration of remission ranged from 33 to 1133 days (median 183 days). The survival times ranged from 102 to 1140 days (median 224 days). Of the six dogs that achieved CR, 28.5% were in CR for >1 year, and 14.0% remained in CR for >3 years. Two dogs (case nos. 4, 5) failed to remain in remission during maintenance therapy and were reinduced using the same protocol but with a 50% increase in the carmustine dosage. These dogs failed to achieve new remissions with the incremental increase in the carmustine dosage.
Side Effects
Toxicity induced by chemotherapy was evaluated at each visit, and the criteria for grading the toxic effects are described in Table 3. The primary signs of toxicity (e.g., vomiting, fever, diarrhea, alopecia) are shown in Table 4. Most of the toxic reactions were observed during the induction phase and usually occurred after the first dose of the combined chemotherapeutic agents. There was no mortality during the induction phase, even in the dogs with advanced stages of the disease. During the maintenance phase, vomiting occurred occasionally. One animal died with oral mucositis and sepsis during the maintenance phase, after it had relapsed and received a 50% higher dose of carmustine [Table 2].
Hematological Toxicity
All dogs experienced marked neutropenia. The average decrease in white blood cell count (WBC) was 71.7%, and in neutrophils it was 81.2% (7 days after carmustine plus vincristine). These percentages were calculated from values obtained in samples collected 7 days after the chemotherapeutic cycle when compared to values obtained just immediately prior to infusion of the chemotherapeutic agent. The median and mean WBC counts 7 days after chemotherapy were 2.5 and 3.83 × 103 cells/μL, respectively (range, 0.38 to 15.3 × 103 cells/μL). The median and mean neutrophil counts were 1.3 and 2.1 × 103 cells/μL, respectively (range, 0.17 to 12.5 × 103 cells/μL). Neutrophil counts were <1.0 × 103 cells/μL following 37% of chemotherapeutic cycles. During treatment, two dogs weighing <15 kg had a sufficient decline in the WBC to require a 50% reduction in the carmustine dose. Both animals recovered with supportive therapy and antibiotics. This lower dose was continued for all future treatments. As expected, when vincristine was given alone, the average decrease in cell counts was less than what occurred following combined treatment with carmustine and vincristine (2.37% for WBC and 19.45% for neutrophil counts). The median and mean WBC counts were 5.7 and 7.3 × 103 cells/μL, respectively (range, 1.8 to 17.9 × 103 cells/μL), and the median and mean neutrophil counts were 3.3 and 4.57 × 103 cells/μL, respectively (range, 0.47 to 15.0 × 103 cells/μL).
The median platelet count 7 days after each injectable treatment was 447 × 103/μL, and the mean was 429 × 103/μL (range, 243 to 621 × 103/μL) after the induction period (week 4). During the weeks of combined carmustine and vincristine, the median platelet count was 272 × 103/μL, and the mean was 283 × 103/μL (range, 157 to 449 × 103/μL). No significant alterations were seen in red blood cell (RBC) counts, but the dogs often had at least one occurrence of circulating erythroblasts (10 to 20 nucleated RBCs/100 WBCs) after carmustine infusion. The highest number of circulating erythroblasts was observed in case no. 1 (164 erythroblasts/100 WBCs) on the 49th day of chemotherapy. The hematological data for all dogs is summarized in Table 5.
Nonhematological Toxicity
No alterations considered to be secondary to the chemotherapy were found in the serum biochemical profiles. No animal showed creatinine values above the reference range during treatment. Elevations in serum ALT and ALP were present at diagnosis and were considered to be related to the lymphosarcoma in the liver. Most dogs with ALT and ALP alterations were in stage V at the time of diagnosis or in the final phase of the disease.
Discussion
A wide variety of chemotherapeutic protocols have been effective in inducing remission for canine lymphosarcoma, but there is no current agreement regarding optimal treatment. In light of this fact, further investigation of the efficacy of carmustine was undertaken. Carmustine has historically been used in the treatment of human lymphosarcoma. In people, the advantages of using carmustine include its capacity to cross the blood-brain barrier and prevent the occurrence of cerebral metastases. Delayed and cumulative bone marrow depression is the most frequent and serious side effect of carmustine. Other side effects include pulmonary fibrosis, renal and hepatic damage, and optic neuritis. Venous irritation may follow the intravenous injection. Nausea and vomiting, beginning up to 2 hours after a dose, are also common.16
The study reported here identified several side effects associated with carmustine in dogs. Vomiting and diarrhea were observed mostly during the induction phase, following the first combined dose of the chemotherapeutic agents. During the maintenance period, vomiting only occurred occasionally and did not differ from rates reported for other chemotherapeutic treatments.81819 Pain during carmustine infusion was noted and was prevented by increasing the infusion time to approximately 60 minutes for all future treatments.15 Decreased appetite, anorexia, and lethargy were observed 1 to 2 days after drug infusion, but in the authors’ experience, these side effects occurred at rates comparable to those observed with other chemotherapeutic protocols (e.g. cyclophosphamide-vincristine-prednisone or doxorubicin as a single agent). No clinical evidence of pulmonary toxicity was observed in the dogs, probably because the cumulative dose administered was far less than that commonly used in humans.
The CR rate (85.7%) and the median survival time (224 days) were similar to those obtained with other combination protocols.36182021 With respect to survival time, consensus is lacking as to the relevance of prognostic factors like age, gender, weight, and clinical stage of the disease.1721–23 Most prior studies included variations in the treatments used, reported on inadequate numbers of dogs, and provided insufficient data for an obvious conclusion to be drawn. Some authors reported that dogs with WHO stage-III lymphosarcoma survived longer than those in other stages, while other researchers concluded that patients in stage V may not have a shorter duration of remission and survival time.1722 It has been postulated that prognosis may be related to the WHO substage of the tumor (i.e., whether the animal is clinically normal [a] or affected [b]) rather than to the tumor’s distribution.17 In this latter study, all dogs had an advanced stage of the disease; three dogs were in stage Vb and survived for less time than animals in stages IV and Va. Besides presenting with more severe clinical signs, dogs in stage Vb were also the ones with the highest body weight. In several previous studies, age and weight have not been found to be prognostic factors for CR rate or survival time.2223 Because of the low number of clinical cases in the study reported here, the data on tumor stage and histopathological grade were not statistically analyzed.
With respect to WBC counts, a carmustine dose of 50 mg/m2 when combined with vincristine caused substantial myelosuppression, which was a dose-limiting toxicity. All dogs developed WBC counts <1.0 × 103 cells/μL at least once during treatment. Secondary adverse effects associated with the leukopenia, such as fever, were not severe enough to require hospitalization, although antibiotic administration was necessary in some neutropenic dogs.
Delayed myelosuppression and thrombocytopenia are the most important potential toxic effects of carmustine in humans.15 Miyashiro, et al., exposed 10 healthy dogs to a single dose of carmustine (50 mg/m2) and followed the animals for 8 weeks; WBC nadir occurred between days 5 and 9 (range, 0.85 to 3.6 × 103 cells/μL), and the average duration of neutropenia was 8.1±3.6 days.24 Around day 10, platelet numbers also decreased (range, 66 to 196 × 103/μL) but recovered rapidly, except in one dog that had severe thrombocytopenia for 2 weeks.24 Bone marrow toxicity was acute and severe. Cytology showed the bone marrow was extremely hypocellular on day 5, with a reduction of all precursor cell series.24 Recovery of hematopoiesis was seen in the peripheral blood of all dogs within 20 days following exposure to carmustine, and this data suggested that the interval of carmustine infusion could be safely reduced to less than the 6 weeks recommended for humans.
Vincristine is often used in combination protocols for lymphosarcoma, and its toxicity is usually not a serious clinical concern in the dog.25 Furthermore, this drug increases the number of platelets produced per megakaryocyte.25 As expected, thrombocytopenia was not a problem during the treatment protocol used in this study, probably because of the use of vincristine.
Despite using a relative low dosage of carmustine, two dogs weighing <15 kg had sufficient toxicity to require a dosage reduction. In addition to marked neutropenia, case no. 1 (7.0 kg) had the highest number of circulating erythroblasts after the carmustine infusion, suggesting bone marrow toxicity. This dog died >2 years after the end of chemotherapy from chronic renal failure and was still in CR at the time of death. Interestingly, these two small dogs survived longer than the other dogs. This phenomenon has also been observed with some other chemotherapeutic protocols.26 In general, toxicosis appears to be more uniform when the chemotherapeutic drug dosage is calculated on the basis of body surface area (BSA) rather than on weight; but according to some researchers, the formula used to estimate BSA may be inaccurate.27 This may explain why larger dogs tend to be under-dosed and smaller dogs tend to be overdosed. The results of this study may indicate that a relative “overdosage” was a contributing factor to the prolonged survival time of smaller dogs.
Conclusion
Based on the response of the dogs in the present study, carmustine, vincristine, and prednisone are a viable alternative option for the treatment of canine lymphosarcoma. Neutropenia associated with carmustine administration was the most serious side effect and was a dose-limiting toxicity. Further studies are needed to investigate the optimal administration interval and duration of maintenance chemotherapy in dogs with lymphosarcoma.
Dr. Sílvia R. R. Lucas, Faculdade de Medicina Veterinária e Zootecnia da Universidade de São Paulo, Brazil, 1998.
Becenum;® Bristol-Myers Squibb, São Paulo, São Paulo, Brazil
Oncovin;® Eli Lilly do Brasil Ltda, São Paulo, São Paulo, Brazil
Acknowledgments
The authors gratefully acknowledge Mr. Geraldo Natalino Thezi, Mr. Antonio Carlos Malaquias, and Mr. Gilberto Pereira da Cruz for their technical support.
