Hematopoietic tumors account for approximately 33%

J Vet Intern Med 00;6:9 86 Chemotherapy with Cyclophosphamide, Vincristine, and Prednisolone (COP) in Cats with Malignant Lymphoma: New Results with an Old Protocol Erik Teske, Giora van Straten, Ronald van Noort, and Gerard R. Rutteman This retrospective study in 6 cats with malignant lymphomas examined the efficacy of a well-established chemotherapy protocol (cyclophosphamide, vincristine, and prednisolone [COP]) in the Netherlands, a country with a low prevalence of feline leukemia virus (FeLV). Twenty-two cats (36.%) had mediastinal lymphoma, (8.0%) had alimentary lymphoma, (.5%) had peripheral lymphoma, 8 (3.%) had nasal lymphoma, and 3 (.3%) had miscellaneous lymphoma (including renal lymphoma in [3.3%]). Of the 54 cats that were tested, only 4 (.4%) were FeLV positive. Complete remission (CR) was achieved in 46 of the 6 cats (5.4%). The estimated - and -year disease-free periods (DFPs) in the 46 cats with CR were 5.4 and 3.8%, respectively, whereas the median duration of remission was 5 days. The overall estimated -year survival rate in all cats was 48.%, and the -year survival rate was 39.9%, with a median survival of 66 days. The median survival time and the -year survival rate for mediastinal lymphoma were 6 days and 49.4%, respectively. Siamese cats had a more favorable prognosis for survival and remission than other breeds. Response to therapy in this study was shown to be a significant prognostic indicator. CR is necessary for long-term survival. Cats that did not achieve CR had little chance of survival for longer than year. Young Siamese cats in this study had a greater tendency to develop mediastinal malignant lymphoma at a young age, and all were FeLV negative. In comparison with results reported in other studies with different combination chemotherapy protocols, these are among the highest percentages of remission and the longest survival rates for cats with malignant lymphoma. Key words: Anatomic types; Feline leukemia virus; Prognosis; Siamese cats. Hematopoietic tumors account for approximately 33% of all tumors in the cat, and almost 90% are malignant lymphomas. An age-adjusted annual incidence rate for malignant lymphoma in the cat has been estimated to be 60 per 00,000 cats. The occurrence of malignant lymphoma in the cat has been strongly associated with feline leukemia virus (FeLV) infection. Until the 990s, approximately 0% of cats with malignant lymphoma in the United States and Western Europe were FeLV positive. 3 5 However, there appears to have been a decrease in the prevalence of FeLV-positive cats during the past 0 5 years, probably because of increased routine testing and the availability of an FeLV vaccine. 6 The malignant lymphomas in cats can be classified according to their anatomic location as mediastinal, alimentary, multicentric, true leukemia, and miscellaneous, the latter including renal, nasal, and ocular lymphomas. 3, Geographical distribution and FeLV status influence the frequency of occurrence of lymphoma at these anatomic sites. In addition, the malignant lymphomas can be characterized on the basis of their histologic morphology according to one of the classification schemes used in humans, such as the Working Formulation. 8 Although large studies using feline monoclonal antibodies are lacking, the majority of ma- From the Department of Clinical Sciences of Companion Animals, Faculty of Veterinary Medicine, Utrecht University, Utrecht, The Netherlands (Teske, van Straten); the Referral Specialist Clinic Amsterdam, Amsterdam, The Netherlands (van Noort); and the Referral Specialist Clinic De Wagenrenk, Wageningen, The Netherlands (Rutteman). Reprint requests: Erik Teske, DVM, PhD, Dip ECVIM-CA, Department of Clinical Sciences of Companion Animals, Faculty Veterinary Medicine, Utrecht University, P.O. Box 80.5, Utrecht, The Netherlands; e-mail: e.teske@vet.uu.nl. Submitted July 0, 00; Revised September 4, 00; Accepted October 5, 00. Copyright 00 by the American College of Veterinary Internal Medicine 089-6640/0/60-000/$3.00/0 lignant lymphomas in cats are thought to be of T-cell origin, transformed by FeLV. 9,0 Most of the FeLV-negative lymphomas, which include many of the alimentary lymphomas and many of the feline immunodeficiency virus (FIV)-related lymphomas, are of B-cell origin., Multicyclic chemotherapy protocols have been successful in the treatment of malignant lymphomas in the cat. Remission rates of up to 9% and median remission intervals of up to 8 days have been reported, depending on anatomic type and FeLV status. 3,4 However, in contrast to the extensive literature on the treatment of malignant lymphoma in dogs, there are only few reports on treatment in cats. One of the reasons for this discrepancy could be that, in the cat, the majority of malignant lymphomas are caused by retroviral infection, and positive FeLV status has a negative prognostic influence on the outcome of chemotherapy. 4,5,6 With the decreasing prevalence of FeLV in both healthy cats and cats with malignant lymphomas 4 and possibly, as a consequence, a greater proportion of B-cell lymphomas response to chemotherapy might now be different from that reported more than 0 5 years ago. In the present study, the cyclophosphamide, vincristine, and prednisone (COP) protocol originally described by Cotter 3 was evaluated in cats with malignant lymphomas in the Netherlands, where the prevalence of FeLV in healthy cats has been much lower (less than 0.3%) since the 980s than in the other Western European countries, because of systematic elimination of carriers (Egberink, personal communication). In addition, patient characteristics were used to identify prognostic factors for treatment outcome. Materials and Methods Animals Medical records of all cats with malignant lymphoma treated at the Utrecht University Clinic for Companion Animals between 984 and 000 (n 53) and the Referral Specialist Clinic Amsterdam between 99 and 000 (n 8) were reviewed. Only cats treated according to

80 Teske et al a standard chemotherapy protocol were included in the study. No specific inclusion or exclusion criteria were used. In all cats, the diagnosis of malignant lymphoma was confirmed by cytologic or histologic examination of a biopsy or of an excised lymph node or organ. Data collected from medical records included age, gender, breed, medical treatment before chemotherapy, and laboratory findings, including hematocrit and FeLV status. Screening for FeLV was performed with an enzyme-linked immunosorbent assay (ELISA) test, and all positive results were confirmed with an immunofluorescence test. FIV status had not been examined in the early cases and was found to be negative in all later cases; therefore, it was not evaluated in this study. The lymphomas were classified according to their anatomic location as mediastinal, alimentary, renal, peripheral, nasal, or miscellaneous. Clinical staging was performed according to Mooney and Hayes. Treatment All cats were treated by the COP chemotherapy protocol described by Cotter. 3 Remission was induced with vincristine given weekly at weeks,, 3, and 4 at a dosage of 0.5 mg/m IV, cyclophosphamide weekly at weeks and 4 at a dosage of 300 mg/m PO, and prednisolone daily at a dosage of 50 mg/m PO. Remission was maintained after week 4 with vincristine and cyclophosphamide given on successive days in contrast to the original protocol, in which these drugs were given on the same day, at 3-week intervals until relapse, or for a maximum of year. Prednisolone was continued daily until relapse or for year in remission, at which time the dose was decreased gradually and stopped over a 3-week period. Response Criteria All cats were evaluated at each treatment. Disappearance of all measurable tumors was considered to be complete remission (CR). A decrease in total tumor volume by 50% was defined as a partial remission (PR), provided no new lesions developed and no progression of any existing lesion occurred. A decrease of 50% in tumor volume and an increase of 5% in any of the measurable lesions were considered to be no change. Progressive disease was defined as a 5% increase in the size of or more measurable lesions or the appearance of new lesions. Response rate was defined as the percentage of cats with CR and PR. In cats with mediastinal lymphoma, results of repeated thoracic radiographs were used to evaluate the response. For other types of lymphoma, results of physical examination before each treatment were combined as appropriate with abdominal ultrasonography in renal and alimentary lymphoma and with rhinoscopy in nasal lymphoma to evaluate response. Alleviation of clinical signs gave additional indication of response. Survival time was calculated as the interval from start of treatment to death or to the date on which the cat was last known to be alive, including all cats and counting only deaths due to lymphoma as events. For cats in which there was a CR, the disease-free period (DFP) was calculated as the interval from start of treatment to relapse or to the date on which the cat was last known to be free of disease, counting only relapses as events. Statistical Analysis Table. Patient characteristics. Variable Category n % Siamese Breed Gender Age (years) FeLV test Stage Location European shorthaired Siamese Persian Other Intact males Castrated Intact females Spayed 0 3 4 5 6 0 Negative Positive Unknown status I II III IV Miscellaneous total (renal) Nasal Peripheral FeLV, feline leukemia virus. 4 6 3 0 30 4 5 5 50 4 35 8 6 3 8 6. 6..6 4.9 6.4 49..5 3.0 4.6 8..5 36.0 9. 9.6.4 5.4 9. 3. 9.8 36..3 (3.3) 8.0 3..5 Differences between groups were evaluated by the chi-square test for ordinal or ratio data and by two-tailed nonpaired Student s t-test or one-way parametric analysis of variance for interval data. Survival curves were drawn by the Kaplan-Meier method. Univariate tests for comparison of groups of survival data were made with the log-rank test and with a proportional hazard logistic regression model. The variables age, anatomic form of lymphoma, gender, pretreatment FeLV status, breed (Siamese versus non-siamese), hematocrit, and clinical stage were introduced in a multivariate Cox proportional hazards model regression analysis of survival without the stepwise selection. Cats were censored in the survival analysis because () they were lost to follow-up, () their death was not caused by lymphoma or chemotherapy, or (3) they were alive at the end of the study period. For the DFP analysis, cats were censored because () they were lost to followup, () they died before relapse, or (3) relapse had not occurred before the end of the study period. P.05 was considered significant. Statistical analysis of data was performed with computer software. a Results Patient Characteristics Sixty-one cats were included in the study, and 6 were Siamese (Table ). The mean age was 6.8 years (median, years; range, 0.5 years). There appeared to be a clear bimodal age distribution around age 0 3 years, with a peak at year, and around 8 years, with a peak at 0 years (Fig ). Cats with mediastinal lymphomas were significantly (P.00) younger (mean age, 3.4 years) than cats with other anatomic types (5.9 0.5 years). All 4 cats that were FeLV positive had malignant lymphoma of the mediastinal type. The 6 Siamese cats formed a relatively large group (6.%) within the total of 6 cats. Data with regard to gender, stage of disease, FeLV status, anatomic location, and age of the Siamese cats are given separately in Table. All tested (3 of 6) Siamese cats were FeLV 6 8 3 3 3 9 4 3 0

Chemotherapy in Cats with Malignant Lymphoma 8 Fig. Age distribution of 6 cats with malignant lymphoma treated with cyclophosphamide, vincristine, and prednisolone (COP). Hatched bars represent Siamese cats (n 6). negative. There were significantly more males (P.036) among the Siamese cats than among the other breeds, the mediastinal form was more common (P.0), and Siamese cats with malignant lymphoma were younger (mean age, 3. years for Siamese versus 8. years for other breeds, P.00) (Fig ). The Siamese cats that had nasal lymphoma were both years old, and the cat that had peripheral lymphoma was year old. Response to Treatment and Survival Data The chemotherapy protocol was very well tolerated by the cats. There were no therapy-related deaths, and intensive care was never required. Dermatologic toxicity was limited to reversible loss of vibrissae in approximately one third of the cats. The response rate for all 6 cats was 88.5%, and 46 of these (5.4%) achieved CR (Table ). Whereas the median DFP was 5 days, the -year estimated DFP rate was 5.4% (95% CI, 39.9 64.), and the -year DFP rate was 3.8% (95% CI, 3.6 5.0) (Fig ). The overall estimated -year survival rate was 48.% (95% CI, 35.4 60.9), and the -year survival rate was 39.9% (95% CI, 6. 5.), with a median survival of 66 days (Fig 3; Table ). Among the different anatomic types, the CR rate ranged from 63.6 to 85.%, but the differences were not significant. The DFP was longest for cats with nasal and peripheral lymphomas, 358 and 38 days, respectively, and the -year survival rate also was highest for cats with these tumors, 5.0 and 5.%, respectively (Table ). Prognostic Factors for Remission and Duration of Survival Eight parameters were studied as prognostic factors for duration of CR (ie, DFP), and survival. Results of a univariate analysis stratified for institution are listed in Table 3. Although hazards ratios are listed for the factored variables, significance levels were calculated using unfactored variables or, for the variables age and hematocrit, using absolute continuous values. The only significant prognostic variable for duration of CR was whether the cat was Siamese or not (by univariate analysis, P.03 and by multivariate analysis, P.039). Siamese cats had longer DFPs than other breeds (Fig 4). Including the anatomic form in the multivariate analysis did not lead to conversion; it was therefore excluded from the multivariate analysis. The only significant prognostic variable for predicting length of survival time shown by both univariate and multivariate analysis was response to therapy. Cats that did not achieve a CR almost all died within the st year. Discussion Feline malignant lymphoma occurs at different frequencies all over the world. 3,8,9 Although in Scotland, the United Kingdom, and some states in the United States, the alimentary form is reported to be the most common, 3,6,0 in this study, the mediastinal form occurred most frequently, followed in decreasing order of frequency by the alimentary, miscellaneous, nasal, and peripheral forms. Differences in the geographic incidence of substrains of FeLV, genetic variation among cats in different geographic areas, or

8 Teske et al Summary of the treatment results achieved with the COP chemotherapy protocol for cats with malignant lym- Table. phoma. a Anatomic Form Cats Complete Remission Rate (%) Median DFP (days) Median Survival (days) Estimated -Year Survival Rate (%) Estimated -Year Survival Rate (%) Total Renal Peripheral Miscellaneous Nasal 6 8 5.4 8.8 63.6 50.0 85.. 5.0 5 5 45 38 358 66 6 9 65 40 358 48. 49.4 36.4 5. 36.4 5.0 39.9 COP, cyclophosphamide, vincristine, and prednisolone; DFP, disease-free period. a insufficient numbers to calculate reliably. differences in the prevalence of FeLV in the population of cats could be causes for this variation in frequency of anatomic lymphoma type. 3 FeLV has been reported to be the major cause of feline malignant lymphoma. 3,9, Because the mediastinal form of lymphoma occurs mostly in young cats might indicate that the cause is related to immunological stressors such as viral infections.,3 Reports of FeLV prevalence of 5 85% in cats with mediastinal lymphoma further point to FeLV as the infectious agent of this form of lymphoma. 3,, However, this presumption was based on clinical and epidemiological information concerning feline malignant lymphoma collected mostly in the United States in the 980s (FeLV infection then being common) with a reported FeLV prevalence of 0% in cats with lymphoma. In the present study, cats with mediastinal lymphoma also were mostly young (median age, years), but in contrast to most previous reports, only.4% of all cats with lymphoma and 8.8% of cats with mediastinal lymphoma were FeLV positive. This is a very surprising finding, assuming that FeLV is indeed the most common cause of malignant lymphoma in general and of the mediastinal form in particular. Furthermore, this assumption would have been very much strengthened if, in a country like the Netherlands (having the lowest FeLV prevalence among healthy cats in the world, to3%in 986, and less than 0.3% in 99 [Egberink, personal communication]), a similarly high FeLV prevalence among the cats with mediastinal lymphoma had been found. Hence, the finding that most cats with mediastinal lymphoma, and in fact most cats with other forms of lymphoma, were FeLV negative raises many questions concerning the role of FeLV in the etiology of feline malignant lymphoma in areas with a low FeLV prevalence. Genetic predisposition, other infectious agents, or masked FeLV infection may be possible explanations. Interestingly, recent studies using polymerase chain reaction (PCR) techniques failed to dem- Fig. Disease-free survival curves for 46 cats with malignant lymphoma that went into complete remission (CR). Vertical bars represent censored observations. The overall estimated - and -year disease-free period (DFP) was 5.4 and 3.8%, respectively. The median DFP was 5 days.

Chemotherapy in Cats with Malignant Lymphoma 83 Fig 3. Survival curves for 6 cats with malignant lymphoma. Vertical bars represent censored observations. The overall estimated -year survival rate was 48.%, and the -year survival rate was 39.9%, with a median survival of 66 days. onstrate the presence of a retrovirus in blood samples or tumor tissue samples of 50 FeLV-negative cats with malignant lymphoma. 3 In a previous study from Canada, out of FeLV-negative cats with lymphoma were still positive by a PCR technique with amplification of a 66-base pair region of the FeLV long terminal repeat. 4 In most cats in the present study, the diagnosis and FeLV testing were carried out at of the specialist centers; thus, referral bias could not be an explanation for the high percentage of FeLV-negative cats. The median age of cats with malignant lymphoma in the present study was years, which is slightly higher than the 4 6 years reported in previous studies. 4,,3 However, the bimodal distribution of ages observed in this study (Fig ) is comparable to that reported previously.,5 This unique age distribution could not be explained by FeLV infection, and the reason for it still is unknown. In the present study, cats with mediastinal lymphoma were markedly younger than cats with other anatomic types of lymphoma, and many were Siamese. This finding is in accordance with earlier reports. 5,6 Males were overrepresented in the Siamese cats. A predisposition to malignant lymphoma in Siamese cats previously was observed in a study in Australia. 0 In that study, Siamese cats 4 years of age or younger had a higher prevalence of lymphoma than any other age or breed of cat. However, although FeLV is regarded as the major cause of feline malignant lymphoma, the Siamese cats in the present study were all FeLV negative. It is therefore difficult to attribute mediastinal lymphoma in the Siamese cat to FeLV infection, and a genetic predisposition to mediastinal lymphoma in this breed could be a more important cause. With regard to the efficacy of the COP protocol in the present study, a CR of 5.4% is comparable to the rate originally reported by Cotter 3 in 983 (Table 4). The CR rates achieved by the COP protocol are therefore higher than those achieved by the VCM (vincristine, cyclophosphamide, and methotrexate) protocols. 4, They are also much higher than the CR rates of only 6 3% observed in studies with doxorubicin as a single induction agent. 6,6 In the present study, cats with mediastinal lymphoma, peripheral lymphoma, and nasal lymphoma had the highest CR rates (8.8, 85., and 5%, respectively). Comparable results were obtained by Cotter, 3 who found that cats with mediastinal lymphoma treated with COP also achieved the highest CR rates (9%). However, when the VCM protocol was used, the CR rate for the mediastinal form was only 45%, and the rates for the other anatomic forms were also lower than those obtained with the COP protocol. In studies with other treatment protocols, 4,4,5, CR rates were not reported separately for the different forms of lymphoma, and a comparison is therefore not possible. Although the response rates in this study were comparable to those observed in previous studies with the same chemotherapy protocol, the duration of CR (ie, DFP) and survival time were much better (Table 4). The total median CR time of 5 days achieved in this study was only exceeded by the 8 days achieved in one study 4 in which doxorubicin was used as maintenance therapy after CR was achieved with the COP protocol; in all other reported studies, the median remission time was shorter. The study of Moore et al 4 included only cats and provided no information on median survival time or data concerning treatment results in the different forms of lymphoma. It is therefore difficult to accurately assess the contribution of doxorubicin to the survival time achieved by the COP protocol without further research. DFPs and survival times are very dependent on the different anatomic forms of malignant lymphoma. It is therefore more interesting to compare results for individual anatomic forms than to compare results for combined groups. In this study, the median DFP for cats with the mediastinal

84 Teske et al Table 3. Results of univariate analysis of factors predicting recurrence after complete remission (using DFP, n 46) and death (using overall survival time, n 6) in feline malignant lymphoma, stratified to clinic. a Patient Characteristic P DFP Hazard Ratio (95% CI) P Survival Hazard Ratio (95% CI) Breed No Siamese Siamese Age (years) Gender Male Female Anatomic form Renal Peripheral Nasal Miscellaneous Stage I II III IV Hematocrit FeLV Negative Positive Pretreatment No Yes Response PD NC PR CR.03.56.95.4.545.4.9.65.33.89..500.48.908 0.3 (0. 0.85).03 (0.95.).04 (0.48.).9 (0.8 4.) 0.4 (0..90).04 (0.66 6.30) 0.0 (0.0.0).69 (0.39.34).053 (0.44.5).06.50.36.699.698.94..894.06.63.4.58.36.583.06.03.00 0.50 (0..6).03 (0.95.).3 (0.55.33).9 (0.48.98).50 (0.9.8) 0.48 (0..88) 0.30 (0.06.35) 0.94 (0.36.43).0 (0.5 3.89) 0.4 (0..6).94 (0.64 5.8) 0.0 (0.00 5.55).08 (0.6 6.95) 0.8 (0.3.9) 0. (0.03.38) 0.5 (0.0 0.89) 0.0 (0.004 0.4) DFP, disease-free period; FeLV, fine leukemia virus. a no convergence was reached in regression analysis; response cannot be used as a prognostic variable predicting length of DFP. Fig 4. Disease-free survival curves correlating Siamese cats to remission time. Vertical bars represent censored observations. The dashed line represents the Siamese cats and the solid line the other breeds.

Chemotherapy in Cats with Malignant Lymphoma 85 Table 4. Summary of results of treatment protocols for feline malignant lymphoma. Treatment Protocol Tumor Location Cats Complete Remission Rate (%) Median Remission (months) Median Survival (months) Reference. COP Total Peripheral nodes Multicentric Miscellaneous. VCM Total Multicentric Renal 3. VCM a. Total ( L-asparaginase). Renal c 4. COP d Total ( idarubicin) 5. COP e Total ( doxorubicin) 6. Doxorubicin Total 38 5 4 0 6 3 9 6 6 03 8 8 9 9 9 86 80 00 50 5 45 50 68 6 6 6 Were in remission Were in remission 6 5.0 6.0 4.5 8 5.0 0 4.0.0 6.0 5 4.0 6.0 9. 3.0.0.5 9.6 8 5 b 5.. 3 4 5 6 4 6 COP, cyclophosphamide, vincristine, and prednisolone;, not reported; VCM, vincristine, cyclophosphamide, and methotrexate. a L-Asparaginase (400 IU/kg IM) was administered with vincristine at week. b Median survival reported only for cats that had obtained complete remission. c After complete remission, substitute cytosine arabinoside (600 mg/m ) for cyclophosphamide. d Achieved complete remission with COP induction; maintenance with idarubicin as a single agent. e Achieved complete remission with COP induction; maintenance with doxorubicin as single agent. form was 5 days, and their median survival was 6 days, which are to our knowledge the longest reported for this form. As in this study, the poorest remission and survival times in COP protocols have been observed in cats with alimentary lymphomas. 3,8 The use of other combination protocols may produce better results in these subtypes of malignant lymphoma.,9 The longest median DFPs and survival times obtained in the present study were in cats with peripheral or nasal lymphomas. Our results in cats with peripheral lymphomas are in agreement with those in earlier studies, 3, but the remission and survival times were longer in our study. There have been only a few reports of treatment of nasal lymphomas in cats. In one study, 9 cats were treated with chemotherapy alone, 6 with radiation therapy alone, and 8 with both chemotherapy and radiation therapy. 30 Median survival times were 5, 593, and 8 days, respectively. Most of the nasal lymphomas in our study were in the most recently treated cats, and follow-up periods were rather short. Nevertheless, it appears that the longest survival times can be achieved in this form of lymphoma. The estimated - and -year survival rates, which have been given in many reports on treatment of malignant lymphoma in dogs, are rarely found in reports of treatment in cats. In principle, they provide more accurate information about the efficacy of the treatment, because animals lost to follow-up, lost by death not due to lymphoma or treatment, or alive at the end of the study period are censored and taken into account in the analysis. Thus, we incorporated survival rate analysis into our investigation. The total estimated -year survival rate was 48.%, which is higher than the 30% -year survival rate reported by Mooney et al. 4 The total -year survival rate (39.9%) and the -year survival rates for different anatomic types of lymphoma have not been reported by previous investigators and therefore cannot be compared. One- and -year DFP rates (5.4 and 3.8%, respectively, in our study) also have not been reported by previous investigators. The - and -year remission rates were higher than the survival rates. The reason for this observation is that cats that had not achieved a remission and died soon after diagnosis had an influence on survival time but not on DFP. In this study, breed was shown to be a prognostic indicator for both CR time and survival time. Siamese cats have a more favorable prognosis for remission and survival than do other breeds, independent of other characteristics. A breed influence on remission and survival times has not been reported previously, and we have no explanation for it. The prognostic potential of response to therapy in this study was very significant. Cats that achieved CR had a much longer survival time than those that did not. CR therefore is necessary for long-term survival. However, because it can only be assessed after therapy, it cannot be used as a prognostic criterion before treatment. Some studies have shown that other variables such as FeLV status, 4,6 advanced stage of disease, 4,6,3 and anatomic site 3 also have prognostic value for remission or survival time. In this study, no variables except those previously mentioned were found to be significantly related to survival or remission times. However, because of the small number of cases, the probability of detecting significant differences, if they existed, was low. In summary, without increased dosages and without in-

86 Teske et al cluding newer cytostatic drugs, the COP protocol of Cotter 3 has achieved some of the highest rates of CR and the longest DFPs and survival times reported in the treatment of malignant lymphoma in cats, while treatment-related toxicity is minimal. Peripheral and nasal lymphomas appear to have a better prognosis than do other types. Footnote a SPSS Windows 9.0, SPSS, Chicago, IL, and EGRET for Windows.0.3, Cytel Software Corporation, Cambridge, MA References. Dorn CR, Taylor DON, Hibbard HH. Epizootiologic characteristics of canine and feline leukemia and lymphoma. Am J Vet Res 96;8:993 00.. Schneider R. Comparison of age- and sex-specific incidence rate patterns of the leukemia complex in the cat and the dog. J Natl Cancer Inst 983;0:9 9. 3. Hardy WD Jr. Hematopoietic tumors of cats. J Am Anim Hosp Assoc 98;:9 940. 4. Mooney SC, Hayes AA, MacEwen EG, et al. Treatment and prognostic factors in lymphoma in cats: 03 cases (9 98). J Am Vet Med Assoc 989;94:696 699. 5. Vahlenkamp T, Müller H. FeLV-assozierte tumoren. In: Kessler M, ed. Kleintieronkologie. Diagnose un Therapie von Tumorerkrankungen bei Hunden und Katzen. Berlin: Parey; 999:. 6. Vail DM, Moore AS, Ogilvie GK, Volk LM. Feline lymphoma (45 cases): Proliferation indices, cluster of differentiation 3 immunoreactivity, and their association with prognosis in 90 cats. J Vet Intern Med 998;:349 354.. Mooney SC, Hayes AA. Lymphoma in the cat: An approach to diagnosis and management. Semin Vet Med Surg (Small Anim) 986; :5 5. 8. Valli VE, Jacobs RM, Norris A, et al. The histologic classification of 60 cases of feline lymphoproliferative disease using the National Cancer Institute working formulation. J Vet Diagn Invest 000; :95 306. 9. Cockerell GL, Krakowka S, Hoover EA, et al. Characterization of feline T- and B-lymphocytes and identification of an experimental induced T-cell neoplasm in the cat. J Natl Cancer Inst 96;5:90 93. 0. Hardy WD Jr, Zukerman EE, MacEwen EG, et al. A feline leukemia virus and sarcoma virus-induced tumor-specific antigen. Nature (London) 9;0:49 5.. Callanan JJ, Jones BA, Irvine J, et al. Histologic classification and immunophenotype of lymphosarcomas in cats with naturally and experimentally acquired feline immunodeficiency virus infections. Vet Pathol 996;33:64.. Holmberg CA, Manning JS, Osburn BI. Feline malignant lymphomas: Comparison of morphologic and immunologic characteristics. Am J Vet Res 96;3:455 460. 3. Cotter SM. Treatment of lymphoma and leukemia with cyclophosphamide, vincristine, and prednisone: II. Treatment of cats. J Am Anim Hosp Assoc 983;9:66. 4. Moore AS, Cotter SM, Frimberger AE, et al. A comparison of doxorubicin and COP for maintenance of remission in cats with lymphoma. J Vet Intern Med 996;0:3 35. 5. Mooney SC, Hayes AA, Matus RE, MacEwen EG. Renal lymphoma in cats: 8 cases (9 984). J Am Vet Med Assoc 98; 9:43 4. 6. Kristal O, Lana SE, Ogilvie GK, et al. Single agent chemotherapy with doxorubicin for feline lymphoma: A retrospective study of 9 cases (994 99). J Vet Intern Med 00;5:5 30.. Weijer K, UytdeHaag F, Osterhaus A. Control of feline leukemia virus infection by a removal programme. Vet Rec 986;9:555 556. 8. Cotter SM, Essex M. Animal model: Feline acute lymphoblastic leukemia and plastic anemia. Am J Pathol 9;8:65 68. 9. Jarrett WFH. Some illustrative systems of viral carcinogenesis: The feline leukemia virus. In: Symington T, Carter RL, ed. Scientific Foundations of Oncology. Chicago, IL: Year Book Medical Publishers Inc; 96:393 403. 0. Court EA, Watson ADJ, Peaston AE. Retrospective study of 60 cases of feline lymphosarcoma. Aust Vet J 99;5:44 4.. Cotter SM, Hardy WD Jr, Essex M. The association of the feline leukemia virus with lymphosarcoma and other disorders in the cat. J Am Vet Med Assoc 95;66:449 454.. Jeglum KA, Whereat A, Young K. Chemotherapy of lymphoma in 5 cats. J Am Vet Med Assoc 98;90:4 8. 3. Hartmann K, Gerle K, Leutenegger C, Jarret O. Feline leukemia virus. Most important oncogene in cats? Proceedings Fourth International Feline Retrovirus Research Symposium, Glasgow, May 30, 998:3g. 4. Jackson ML, Haines DM, Meric SM, et al. Feline leukemia virus detection by immunohistochemistry and polymerase chain reaction in formalin-fixed, paraffin-embedded tumor tissue from cats with lymphosarcoma. Can J Vet Res 993;5:69 6. 5. Gabor LJ, Malik R, Canfield PJ. Clinical and anatomical features of lymphosarcoma in 8 cats. Aust Vet J 998;6:5 3. 6. Peaston AE, Maddison JE. Efficacy of doxorubicin as an induction agent for cats with lymphosarcoma. Aust Vet J 999;:44 444.. Moore AS, Ruslander D, Cotter SM, et al. Efficacy of, and toxicosis associated with, oral idarubicin administration in cats with neoplasia. J Am Vet Med Assoc 995;06:550 554. 8. Mahony OM, Moore AS, Cotter SM, et al. lymphoma in cats: 8 cases (988 993). J Am Vet Med Assoc 995; 0:593 598. 9. Zwahlen CH, Lucroy MD, Kraegel SA, Madewell BR. Results of chemotherapy for cats with alimentary malignant lymphoma: cases (993 99). J Am Vet Med Assoc 998;3:44 49. 30. Klein MK, Powers BE, Calley C, et al. Feline nasal lymphoma: A retrospective analysis. Proceedings 5th Annual Conference of the Veterinary Cancer Society, Tucson, AZ, 995:9 80. 3. Mauldin GE, Mooney SC, Meleo KA, et al. Chemotherapy in 3 cats with lymphoma: 988 994. Proceedings 5th Annual Conference of the Veterinary Cancer Society, Tucson, AZ, 995.