Lymphoma is the most common hematopoietic neoplasia diagnosed and treated in cats.1 In the past 20 years, there has been a change in the most common age and anatomic location associated with the disease. In the 1980s, the most common anatomic locations were mediastinal and multicentric.1–3 Affected cats were typically young (mean age, 4 to 6 years), and the majority (60% to 70%) were FeLV positive.1,3,4 Currently, the gastrointestinal tract is the most common location, the population of cats with lymphoma is now more commonly middle aged to older (mean age, 10 to 12 years), and fewer cats are positive for FeLV.1,3–6
Previous studies3,7–12 have evaluated possible prognostic factors for cats with lymphoma. These studies3,7–12 have determined the following negative prognostic factors: substage b disease, mediastinal location, poor body condition score at diagnosis, and the presence of retroviral infections. Initial response to treatment was also prognostic in that cats that went into a complete remission survived significantly longer than those that did not respond.3,10,13
Reported chemotherapy protocols, response rates, and survival times for feline lymphoma vary. The most common chemotherapy protocols use a combination of COP or CHOP. With CHOP-based protocols, response rates vary from 84% to 95%, with median survival times from 210 to 296 days,13–15 whereas COP-based protocol response rates vary from 39% to 79%, with median survival times of 45 to 239 days.9,11,16–18 Some of these studies include small numbers of cats and thus may not be representative of the feline lymphoma-affected population in general. In addition, few reports include data on response rate, progression-free survival time, and overall survival time; for most studies, 1 or 2 of these endpoints are reported. The variation in the reporting of anatomic location, FeLV status, and treatment protocol of these studies makes it difficult to develop a comprehensive representation of feline lymphoma. Furthermore, the major changes in feline lymphoma over the past 2 decades make comparing the data from these studies even more challenging.
At the Matthew J. Ryan Veterinary Hospital of the University of Pennsylvania, the Oncology Service currently uses a COP-based protocol to treat cats with lymphoma. The purpose of the study reported here was to determine the response rate, progression-free survival time, and overall survival time for cats that received this protocol. A secondary purpose was to evaluate known and potential new prognostic factors in a large population of cats that received the same chemotherapy protocol.
Materials and Methods
Case selection—A search of the Matthew J. Ryan Veterinary Hospital of the University of Pennsylvania medical records database for cats with lymphoma was performed. Cats with a cytologic or histopathologic diagnosis of lymphoma that were treated with a weekly COP-based protocol19 (Appendix 1) from 1998 to 2008 were included. Cats were excluded from the study if they were treated with radiation therapy or chemotherapy (including prednisone given within 7 days prior to the start of treatment) prior to initiation of the protocol evaluated in this study and were determined to have another type of cancer at the time of diagnosis or prior to the diagnosis of lymphoma, all measurable disease was surgically removed, or the medical record was incomplete or unavailable. A medical record was considered incomplete or unavailable if the hard copy was not available or if diagnosis confirmation or the chemotherapy protocol and dosing information were not available from the information in the medical record.
Medical records review—Information recorded from the medical records included signalment, baseline weight (the cats were classified into weight groups on the basis of a previous study12), retroviral status, anatomic location and cell type and size of lymphoma, treatment delays, dose reductions, dates of treatment initiation, disease progression, response to the protocol, number and type of rescue protocols, and death or last known status. Anatomic localization was assigned according to a published scheme for feline lymphoma,3 with the addition of a bone marrow location (Appendix 2). Data were recorded on standardized data sheets by 1 author (AHKW), and referring veterinarians or owners were contacted for follow-up information when necessary.
Response to treatment was defined as the clinical response after 1 cycle of chemotherapy (recorded at the sixth week of treatment). The physical examination sheet as well as the discharge summary were used, when available, to place cats into the category of a responder or a nonresponder. If results of diagnostic tests, such as CBC, serum biochemical analysis, or imaging studies, were available, those results were also used to assess the cat's clinical response. Because the response evaluation was based on subjective history and physical examination findings for most cats, it was often difficult to objectively determine the extent of a response (eg, complete vs partial). Cats were therefore categorized simply as responders or nonresponders. A clinical response was defined as improvement of clinical signs and lymphoma-related physical examination abnormalities. Cats that had no improvement or worsening of clinical signs or physical examination findings were categorized as nonresponders. In addition, cats that died, were lost to follow-up, or received a rescue protocol before the response endpoint were considered to be nonresponders. Neutropenia occurrence and severity during the first cycle of chemotherapy were recorded according to the Veterinary Comparative Oncology Group criteria.20 Given the retrospective nature of the present study, reporting on gastrointestinal tract toxicosis would be dependent on owner reporting and nonstandardized recording, thus making it difficult to make accurate conclusions about the prevalence and severity of gastrointestinal tract toxicosis.
Statistical analysis—Fisher exact or χ2 tests were used to evaluate the association between FeLV status (positive or negative), weight at diagnosis (< 3.3, 3.3 to 5, or > 5 kg [< 7.3, 7.3 to 11, or > 11 lb),12 substage (a vs b), anatomic location (gastrointestinal tract vs other locations), and response to treatment.
Progression-free survival time was defined as the time from diagnosis to disease progression. Overall survival time was defined as the time from diagnosis to death from any cause. If cats were lost to follow-up or still alive, they were censored at the date of last contact. Disease progression was defined as clinical signs or physical examination or diagnostic test results consistent with relapse of lymphoma or that led to a switch to a more intensive chemotherapy protocol or cessation of treatment because of a lack of response. The Kaplan-Meier product limit method was used to determine progression-free survival time and overall survival time, and the log-rank method was used to compare groups according to FeLV status (positive vs negative), baseline weight (≤ 3.3, 3.3 to 5, or ≥ 5 kg), anatomic localization (gastrointestinal tract location vs all other locations), treatment delay (yes vs no), dose reduction of at least 1 chemotherapy drug excluding prednisone (yes vs no), and clinical response to treatment after 1 cycle (responders vs nonresponders). Values of P < 0.05 were considered significant.a,b
Results
Patient population—One hundred fourteen cats met the inclusion criteria, and 841 cats were excluded. There were 71 (62.3%) castrated males, 42 (36.8%) spayed females, and 1 (0.9%) sexually intact female. Mean ± SD age of cats was 11.1 ± 4 years, and mean weight was 4.5 ± 1.6 kg (9.9 ± 3.5 lb). Most were domestic shorthair cats (n = 93 [81.6%]), with 13 (11.4%) domestic longhair cats, 3 (2.6%) Maine Coon cats, 2 (1.8%) each of Siamese and unknown breed, and 1 (0.9%) Persian. The most common anatomic location was the gastrointestinal tract (n = 57 [50%]), with an additional 34 (30%) cats having nongastrointestinal tract abdominal lymphoma. Other locations included extranodal for 9 (7.9%) cats, thoracic cavity for 7 (6.1%) cats, mixed for 3 (2.6%) cats, multicentric for 3 (2.6%) cats, and bone marrow for 1 (0.9%) cat. Most cats had substage b lymphoma (n = 94 [82.4%]); 12 (10.5%) had substage a. Among 97 cats with evaluation of cytologic or biopsy specimens, 86 (89%) had large cell lymphoma, 2 (2.1%) had small cell lymphoma, and 10 (10.3%) had intermediate cell lymphoma. For 78 cats, the results of FeLV testing were available, with 4 (5.1%) positive and 74 (94.9%) negative results, whereas FIV test results were available for 69 cats, with 7 (10.1%) positive and 62 (89.9%) negative results. In 106 (93%) cats, a diagnosis was made on the basis of cytologic samples, whereas in 8 (7%) cats, a diagnosis was made on the basis of histologic samples.
Protocol variations—A protocol variation was documented in 59 (51.8%) cats of this study. Most of those cats received vinblastine instead of vincristine (n = 22 [37%]). Reasons for administering vinblastine instead of vincristine included lymphoma-related gastrointestinal tract signs at the initiation of chemotherapy (n = 5) and gastrointestinal tract intolerance of vincristine (17). Another common variation was the substitution of cytarabine arabinoside for methotrexate in cats with renal involvement (n = 12 [20%]).
One hundred twenty-four treatment delays occurred for 46 cats, with a median of 1 treatment delay/cat (range, 1 to 25 treatment delays/cat) and the most common reason being neutropenia (n = 27 [58.7%]). The next most common reason for a treatment delay was gastrointestinal tract upset (6 [13%]).
Whether dose reductions occurred could be determined from the records of 82 cats. Cats that received only 1 dose of a given chemotherapy drug could not be evaluated for dose reductions, so cats that died (n = 13), were lost to follow-up (4), or had disease progression and started a new protocol (14) during the first cycle of chemotherapy could not be evaluated. One cat's record was not available for dose reduction review, but all other data were collected. Twenty-six of the 82 (31.7%) cats had dose reductions. The most common drug for which dose was reduced was vincristine (n = 16 [61.5%], with 1 cat having the vincristine dose reduced twice), then vinblastine (6 [23%], with 1 cat having the vinblastine dose reduced 4 times), and lastly cyclophosphamide (2 [7.7%]). One cat had both the vincristine and vinblastine dose reduced (4%), and 1 cat had the vincristine, vinblastine (dose reduced twice), cyclosphosphamide, and cytosar arabinoside dose reduced (4%). The most common reasons for dose reduction were neutropenia (n = 13 [50%]), high bilirubin concentration (3 [11.5%]), weight loss (3 [11.5%]), vomiting (2 [7.7%]), and nonspecified gastrointestinal tract signs (2 [7.7%]); 1 cat each had a dose reduction for renal insufficiency (4%) or increased liver enzyme activities (4%). An additional 3 (11.5%) cats had dose reductions for a nonspecified reason.
Hematologic toxicosis—Overall, there were 23 episodes of neutropenia (grades 1 to 4; Table 1). There were no episodes of septicemia or other hematologic abnormalities.
Occurrence (No. of affected cats) of toxicosis (neutropenia) of various grades during the first cycle of chemotherapy for treatment of cats with lymphoma and for which CBC results were available 1 week after administration of various chemotherapy drugs.
| Toxicosis grade | ||||
|---|---|---|---|---|
| Drug | 1 | 2 | 3 | 4 |
| Vincristine (93*) | 7 | 3 | 3 | 0 |
| Vinblastine (19*) | 4 | 1 | 1 | 1 |
| Cyclophosphamide (85*) | 1 | 0 | 0 | 0 |
| Methotrexate (31*) | 1 | 0 | 1 | 0 |
| Cytosine arabinosidase (2*) | 0 | 0 | 0 | 0 |
Severity was graded according to published guidelines.19
No. of treated cats.
Outcome—The clinical response rate was 47.4%. None of the factors evaluated for association with response rate were significantly associated with response. The median progression-free survival time was 65.5 days, and the median overall survival time was 108 days.
Fifty-three cats received rescue chemotherapy after developing progressive disease or not responding to the weekly sequential protocol. The most commonly used rescue protocols were a more intensive form of the weekly COP-based protocol (n = 38 [72%]), lomustine (15 [28%]), and doxorubicin (13 [25%]). Of the cats that were considered nonresponders, 31 did not receive any further chemotherapy and had a median overall survival time of 28.5 days (95% CI, 9 to 391 days), compared with 28 nonresponder cats that did receive rescue chemotherapy and had a median overall survival time of 74 days (95% CI, 17 to 912 days).
Seventy-one (62%) cats died of disease progression, 38 (33%) were lost to follow-up, 3 (3%) died of other causes determined at necropsy (1 each of heart disease, lipid pneumonia, and fungal disease), and 2 (2%) were alive at the time of analysis. Of the 2 cats alive, 1 had completed the protocol, with 716 days of follow-up.
Of the factors evaluated for association with progression-free survival time and overall survival time, only response to treatment was significantly positively associated with both. Cats that responded to the chemotherapy protocol had a significantly (P < 0.001) longer progression-free survival time (median, 364 days [95% CI, 47 to 681 days]), compared with nonresponders (median, 31 days [95% CI, 27 to 35 days]; Figure 1). Responders had a significantly (P < 0.001) longer overall survival time (median, 581 days [95% CI, 157 to 1,005 days]), compared with nonresponders (median, 73 days [95% CI, 51 to 95 days]; Figure 2). The Kaplan-Meier survival analysis for progression-free survival time was significantly (P = 0.033) different in cats with substage a versus substage b disease. The median progression-free survival time (time at which the proportion of survivors reaches 0.50) was not attained for the substage a cats. The 95% CI of the median was incalculable because the median progression-free survival time was unattainable. Significant differences in overall survival time were not found between substage a cats and substage b cats.
Kaplan-Meier survival curves comparing progression-free survival time according to clinical response after 1 cycle of chemotherapy in 114 cats with lymphoma.
Citation: Journal of the American Veterinary Medical Association 242, 8; 10.2460/javma.242.8.1104
Kaplan-Meier survival curves comparing overall survival time according to clinical response after 1 cycle of chemotherapy in 114 cats with lymphoma.
Citation: Journal of the American Veterinary Medical Association 242, 8; 10.2460/javma.242.8.1104
Discussion
This study represents one of the largest retrospective studies evaluating a COP-based chemotherapy protocol for feline lymphoma. The patient population characteristics were consistent with most previous studies.1,3–6 The gastrointestinal tract location was the most common, which is a temporal change that has also been noted in other publications and is thought to be associated with the decreasing prevalence of FeLV infections in the past 20 years.4 Results from the study reported here also confirmed information from other studies which have found that response to treatment and substage are prognostic for feline lymphoma; however, FeLV status was not found to be prognostic.
The reported progression-free survival time and overall survival time for cats that were responding after 1 cycle of chemotherapy were longer than previously reported. The reported median survival time for complete responders treated with COP-based protocols is 150 to 239 days,9,11,17 but the responders in the present study had a median overall survival time of 581 days. A CHOP-based protocol also had a shorter overall survival time for responders (291 to 563 days) in 3 studies,9,15,16 whereas a longer progression-free survival time and overall survival time (654 days) for responders were reported in another study.14 However, that study14 did include small cell gastrointestinal tract lymphoma cases. The response rate in the present study (47.4%) was similar to or lower than that in previous studies,9,11,14–17 and the median progression-free survival time and overall survival time of the study population were shorter.
Most cats in the population had large cell lymphoma, which was most likely attributable to the selection criteria of only including cats that were treated via a weekly sequential COP-based protocol. Prednisone and chlorambucil chemotherapy is the current standard-of-care treatment for cats with small cell lymphoma, and cats with this form of lymphoma generally have much longer survival times, compared with cats that have large cell lymphoma.20–24 A recent study25 evaluating histologic classification of feline gastrointestinal tract lymphoma determined that small cell lymphoma accounts for roughly one-third of cases and is predominantly T cell in origin, whereas the remaining two-thirds are large to intermediate B-cell lymphomas. It is possible that previous studies included cats with small cell lymphoma, given that many were conducted before the distinction between large and small cell lymphoma was established. The predominance of large cell lymphoma cases in the present study may explain the large difference in overall survival time between the cats in this study and previous studies.
Hematologic toxicosis was uncommon and mild. The single instance of grade 4 toxicosis occurred after treatment with vinblastine. However, the 40% treatment delay rate and 32% dose reduction rate indicated that this protocol is fairly intense or feline lymphoma patients have poor tolerance of chemotherapy, compared with dogs. This poor tolerance of chemotherapy is likely correlated with the high prevalence of substage b disease and may also reflect the poor outcome of many cats in the study, given that it can be difficult to distinguish clinical signs of tumor progression from chemotherapy intolerance, particularly in cats with gastrointestinal tract lymphoma.1,4,5
Limitations of this study were primarily attributable to its retrospective nature. Many cats were lost to follow-up because follow-up visits were not standardized. Assessment of response was somewhat subjective and performed retrospectively, which could have introduced bias into the classification of responders and nonresponders; however, this was the only factor significantly associated with progression-free survival time and overall survival time, which suggested that the categorization was accurate. In addition, this type of response criteria is often used by practitioners on a day-to-day basis to evaluate the response to chemotherapy because most cats have internal disease and weekly or biweekly imaging studies are impractical, expensive, and stressful to the patient. Progression-free survival time and overall survival time were both evaluated; the former may be less accurate than the latter, but it is not affected by treatments beyond the study protocol, as is overall survival time. Some anatomic tumor locations were likely unrecognized for some cats, considering that once clinicians obtained a diagnosis, they often did not continue to perform aspirates from other sites even if there was clinical suspicion that lymphoma was likely present in other internal organs. Therefore, more cats may have had gastrointestinal tract lymphoma, but if the diagnosis were made on the basis of tumors elsewhere, additional diagnostic samples were not obtained. In addition, the chemotherapy protocol was not completely standardized; some cats received vinblastine if the clinician felt the patient needed a drug that potentially was better tolerated by the gastrointestinal tract than was vincristine.
Results of the present study indicated that COP-based chemotherapy for feline lymphoma can be associated with long progression-free survival time and overall survival time in cats that respond early in the treatment protocol. Although response to treatment had prognostic importance, none of the factors evaluated were significantly associated with the likelihood of response. Additional studies that focus on identifying factors associated with an early response and treatments that may increase this response rate are needed because a large number of cats are poor responders.
We conclude that an early response to COP-based chemotherapy is prognostic for a substantial survival benefit in cats with lymphoma, compared with cats that do not respond. This indicates that it may be beneficial to consider changing to a different chemotherapy protocol early in the course of treatment if a clinical response has not been achieved.
ABBREVIATIONS
| COP | Cyclophosphamide, vincristine, and prednisone |
| CHOP | Cyclophosphamide, doxorubicin, vincristine, and prednisone |
IBM SPSS Statistics, release 19.0.0, IBM Corp, Armonk, NY.
Sigmaplot, version 12.0 for Windows, Systat Software Inc, Chicago, Ill.
References
1. Vail D. Feline lymphoma and leukemia. In: Withrow SVail D, eds. Withrow and MacEwen's small animal clinical oncology 4th ed. St Louis: Saunders Elsevier, 2007 733–756.
2. Court EWatson APeaston A. Retrospective study of 60 cases of feline lymphosarcoma. Aust Vet J 1997 75 424–427.
3. Ettinger S. Principles of treatment for feline lymphoma. Clin Tech Small Anim Pract 2003 18 98–102.
4. Louwerens MLondon CPedersen Net al. Feline lymphoma in the post-feline leukemia virus era. J Vet Intern Med 2005 19 329–335.
5. Gabor LMalik RCanfield P. Clinical and anatomical features of lymphosarcoma in 118 cats. Aust Vet J 1998 76 725–732.
6. Vail DMoore AOgilvie Get al. Feline lymphoma (145 cases): proliferation indices, cluster of differentiation 3 immunoreactivity and their association with prognosis in 90 cats. J Vet Intern Med 1998 12 349–354.
7. Teske Evan Straten Gvan Noort Ret al. Chemotherapy with cyclophosphamide, vincristine, and prednisolone (COP) in cats with malignant lymphoma: new results with an old protocol. J Vet Intern Med 2002 16 179–186.
8. Haney SBeaver LTurrel Cet al. Survival analysis of 97 cats with nasal lymphoma: a multi-institutional retrospective study (1986–2006). J Vet Intern Med 2009 23 287–294.
9. Taylor SGoodfellow MBrowne Wet al. Feline extranodal lymphoma: response to chemotherapy and survival in 110 cats. J Small Anim Pract 2009 50 584–592.
10. Malik RGabor LFoster Set al. Therapy for Australian cats with lymphosarcoma. Aust Vet J 2001 79 808–817.
11. Mooney SCHayes AAMacEwen EGet al. Treatment and prognostic factors in lymphoma in cats: 103 cases (1977–1981). J Am Vet Med Assoc 1989 194 696–699.
12. Baez JMichel KSorenmo Ket al. A prospective investigation of the prevalence and prognostic significance of weight loss and changes in body condition in feline cancer patients. J Feline Med Surg 2007 9 411–417.
13. Milner RJPeyton JCooke Ket al. Response rates and survival times for cats with lymphoma treated with the University of Wisconsin-Madison chemotherapy protocol: 38 cases (1996–2003). J Am Vet Med Assoc 2005 227 1118–1122.
14. Zwahlen CHLucroy MDKraegel SAet al. Results of chemotherapy for cats with alimentary malignant lymphoma: 21 cases (1993–1997). J Am Vet Med Assoc 1998 213 1144–1149.
15. Simon DEberle NLaacke-Singer Let al. Combination chemotherapy in feline lymphoma: treatment, outcome, tolerability, and duration in 23 cats. J Vet Intern Med 2008 22 394–400.
16. Mahony OMMoore ASCotter SMet al. Alimentary lymphoma in cats: 28 cases (1988–1993). J Am Vet Med Assoc 1995 12 1593–1598.
17. Cotter S. Treatment of lymphoma and leukemia with cyclophosphamide, vincristine, and prednisone: II. Treatment of cats. J Am Anim Hosp Assoc 1983 19 166–172.
18. Oberthaler KTMauldin EMcManus PMet al. Rescue therapy with doxorubicin-based chemotherapy for relapsing or refractory feline lymphoma: a retrospective study of 23 cases. J Feline Med Surg 2009 11 259–265.
19. Vail D. Veterinary Co-operative Oncology Group—common terminology criteria for adverse events (VCOG-CTCAE) following chemotherapy or biological antineoplastic therapy in dogs and cats v1.0. Vet Comp Oncol 2004 2 194–213.
20. Fondacaro JRichter KCarpenter Jet al. Feline gastrointestinal lymphoma: 67 cases (1988–1996). Eur J Comp Gastroenterol 1999 4 5–11.
21. Richter KP. Feline gastrointestinal lymphoma. Vet Clin North Am Small Anim Pract 2003 33 1083–1098.
22. Kiselow MARassnick KMMcDonough SPet al. Outcome of cats with low-grade lymphocytic lymphoma: 41 cases (1995–2005). J Am Vet Med Assoc 2008 232 405–410.
23. Lingard ABriscoe KBeatty Jet al. Low-grade alimentary lymphoma: clinicopathological findings and response to treatment in 17 cases. J Feline Med Surg 2009 11 692–700.
24. Stein TPellin MSteinberg Het al. Lymphoma with chlorambucil and glucocorticoids. J Am Anim Hosp Assoc 2010 46 413–417.
25. Pohlman LHigginbotham MWelles Eet al. Immunophenotypic and histologic classification of 50 cases of feline gastrointestinal lymphoma. Vet Pathol 2009 46 259–256.
Appendix 1
Cyclophosphamide-, vincristine-, and prednisone-based protocol used in a study of 114 cats with lymphoma.
| Drug | Week 1 | Week 2 | Week 3 | Week 4 | Week 5 |
|---|---|---|---|---|---|
| l-asparaginase (400 U/kg [182 U/lb], SC) | X | ||||
| Vincristine (0.5 mg/m2, IV) | X | X | |||
| Cyclophosphamide (25 mg, PO, q 48 h, 2 doses) | X | ||||
| Methotrexate (2.5 mg, PO) | X | ||||
| Prednisone (1–2 mg/kg/d [0.45–0.9 mg/lb/d], PO) | X | X | X | X | X |
l-asparaginase administration was discontinued after 2 cycles. For cats in remission, treatments were given weekly for 6 months (weeks 2 to 5 repeated), every other week for 6 months (weeks 2 to 5 repeated), then every 3 weeks for 6 months (weeks 2 to 5 repeated) for a total of 18 months of chemotherapy. Prednisone was given for the duration of the protocol. Week of administration is indicated by X.
Appendix 2
Scheme for anatomic localization of feline lymphoma.3
| Anatomic location | Description |
|---|---|
| Gastrointestinal | Gastrointestinal tract with or without abdominal lymph node involvement |
| Abdominal cavity, nongastrointestinal | Abdominal organs without gastrointestinal involvement |
| Extranodal | Nonlymphoid locations, such as nasal, CNS, and cutaneous |
| Thoracic cavity | Intrathoracic lymph nodes, pulmonary, or pleural effusion |
| Multicentric | Peripheral lymph nodes including liver involvement |
| Bone marrow | Bone marrow or peripheral blood |
| Mixed | At least 2 of the above locations |
