Clinical Signs, Treatment, and Outcome in Cats with Myeloma-Related Disorder Receiving Systemic Therapy
Myeloma-related disorder (MRD) is an uncommon disease in cats, for which there is no established standard of care. In this retrospective study, we evaluated presentation, response to treatment, and toxicity in cats with MRD receiving systemic treatment. Previously reported prognostic factors were evaluated for their impact on survival in cats receiving chemotherapy. Of fifteen cases identified, thirteen received melphalan or cyclophosphamide +/- corticosteroids as first-line therapy. Chlorambucil was commonly used as rescue therapy in cats with progressive disease, or in cases of chemotherapy-related toxicity with first line agents. Overall response rates were 71% and 83% for melphalan- and cyclophosphamide-treated cats, respectively. Discontinuation of melphalan due to toxicity was common. Survival times for cats initially treated with melphalan or cyclophosphamide were not significantly different (median 252 and 394 days, respectively), and no statistically significant prognostic factors were identified. This study suggests that the combination of cyclophosphamide and corticosteroids is well tolerated and may be considered as first-line therapy for cats with systemic MRD.
Introduction
Myeloma-related disorder (MRD) is a term that describes neoplasias of plasma cells or immunoglobulin-secreting, B-lymphocyte precursors.1,2 It includes “classical” multiple myeloma as well as other manifestations of plasma cell neoplasia common in cats (e.g., liver and spleen infiltration without bone marrow involvement). MRD also includes solitary plasmacytoma, which may or may not be associated with systemic signs.1 Different presentations of, and treatments for, feline MRD have been described.1,3–15 It is important to distinguish between MRD that may be amenable to local therapy (i.e., solitary extramedullary plasmacytoma without systemic signs) and MRD that may require systemic treatment (e.g., abdominal or bone marrow MRD or any other site with concurrent systemic signs). In this study, cats with bone marrow or abdominal MRD and cats with MRD affecting other sites (e.g. skin) with concurrent systemic signs were considered together because the clinical behavior of MRD is similar in these presentations—i.e. the prognosis is poor compared to solitary extramedullary plasmacytoma without systemic signs.1,2
A combination of melphalan and prednisone is the treatment of choice for canine multiple myeloma, and a melphalan/prednisolone combination has been used in feline MRD though other combination and single agent protocols have also been reported.1–3,5,6,10,11,14–16 To date, large-scale studies evaluating chemotherapy for feline MRD are lacking.
Ancillary treatments, such as radiation therapy for palliation of osteolytic lesions and plasmapheresis for management of hyperviscosity syndrome (HVS) have been described.5,6,9 For cutaneous extramedullary plasmacytoma without systemic signs, surgical excision is recommended; however, progression to systemic MRD is possible.2,10
To the authors' knowledge, there are no previous publications directly comparing chemotherapy treatments and outcomes in feline MRD. The purpose of this retrospective study is to describe presentation, treatment, and outcomes in a group of cats with MRD treated with chemotherapy and/or corticosteroids and to compare the effects of different treatment protocols in this population.
Materials and Methods
Electronic medical records were searched for cats with a diagnosis of multiple myeloma, plasma cell tumour, or plasmacytoma presenting to the University of Minnesota Veterinary Medical Center between 2007 and 2012. Cats were included in the current study if they had a diagnosis of MRD and if systemic therapy was indicated. Indications for systemic therapy were bone marrow/abdominal MRD (with or without systemic signs) or involvement of another site with systemic signs of illness. Diagnosis of MRD was based on cytological or histological evidence of plasma cell neoplasia. Diagnosis of plasma cell neoplasia was based on increased numbers and atypical morphology of plasma cells or the presence of plasma cell aggregates. Samples where plasma cell numbers were increased but atypia was mild or where low numbers of atypical cells were present were considered suspicious but not definitive for plasma cell neoplasia.
In total, 20 medical records were identified. Two cats were excluded due to incomplete medical records, 1 due to a final diagnosis of lymphoma without recorded hyperglobulinemia, 1 because of lack of definitive diagnosis, and 1 because of the presence of a solitary cutaneous plasma cell tumour without evidence of other organ involvement, hyperproteinemia, or systemic signs. The remaining 15 cats had sufficient information in the medical record for evaluation of clinical presentation and staging. One cat was lost to follow-up immediately after starting treatment, leaving 14 cats for evaluation of response to treatment and outcome.
Response was primarily assessed by evaluation of blood globulin or total protein levels. Complete response (CR) was defined as normalization of these values. Partial response (PR) was defined as >30% decrease in globulin or total protein levels without complete normalization, progressive disease (PD) was defined as ≥20% increase in globulin or total protein levels, and stable disease (SD) was defined as insufficient change to qualify for either a PR or PD. In humans and dogs with multiple myeloma, repeated evaluation of M-component protein and/or bone marrow is used for assessment of response with PR defined as a ≥50% reduction in M-component protein.17,18 In this study, a 30% reduction in total protein or globulin was considered a fair indicator of response because proportional change in those parameters would be less than in the M-component protein, and repeated M-component protein or bone marrow evaluation were not performed. In cats with measurable skin masses, a CR was defined as the complete resolution of lesions, PR as a ≥30% reduction in size, PD as ≥20% growth, and SD as an insufficient change to qualify for PR or PD based on RECIST guidelines.19 Hematologic toxicity from chemotherapy was graded according to Veterinary Cooperative Oncology Group criteria.20 Other toxicities could not be retrospectively graded due to inconsistent reporting.
Time to response was defined as time from commencement of chemotherapy to first documented CR or greatest PR. Recheck schedules varied between cases; however, for all cats receiving melphalana or cyclophosphamideb, rechecks were performed q 1–2 wk in the 1st mo then at least q 1 mo until documented CR, PD, or chemotherapy toxicity. For patients that achieved CR without toxicity, subsequent rechecks were recommended q 2–3 mo.
Survival time was defined as the time from the date of diagnosis to death. Patients alive at follow-up were censored from survival analysis. Patients were assumed to have died of MRD-related causes if they had PD or SD and clinical signs consistent with MRD at the time of death.
Median survival times (MSTs) and duration of treatment with cyclophosphamide or melphalan were determined using Kaplan-Meier survival analysis. Survival curves were compared using the log-rank test. Effects of chemotherapy protocol and prognostic factors on overall survival or duration of treatment were compared using Cox proportional hazards. A statistical software packagec was used for these calculations, and statistical significance was set at P < .05. Spearman rank correlation was used to assess the correlation between starting dose and treatment duration with melphalan. Time to progression was not evaluated because many patients either changed or discontinued treatment or were euthanized for reasons other than disease progression.
Results
Of the 15 included cats, 9 cats (60%) were castrated males and 6 cats (40%) were spayed females. Domestic shorthair was the most common breed (n = 10), with 1 each of domestic longhair, Siamese, Maine Coon, Persian, and Somali. Median patient age was 13 yr (range, 9–17 yr). Median body weight was 5.01 kg (range, 3.28–6.86 kg). Historical signs and physical examination abnormalities were not specific for MRD (Table 1), with weight loss and reduced appetite being most common.
Laboratory abnormalities were common (Table 1). The median globulin level was 107 g/L (range, 50–140 g/L; reference range, 25–53 g/L). One cat had globulins within the reference range. In 3 cases, asymptomatic hyperglobulinemia was first noted 4–19 mo prior to diagnosis of MRD. Serum protein electrophoresis was performed in 10 out of 14 hyperglobulinemic cats, and a monoclonal gammopathy was detected in all.
The median albumin level was 23 g/L (range, 18–31 g/L; reference range, 24–41 g/L), and 9 cats were hypoalbuminemic. Cholesterol was normal in 12 cats (and below the reference range, greater than the reference range, or not evaluated in 1 cat each). Twelve cats were anemic. The median haematocrit was 27.8% (range, 12-40%; reference range, 29.5–47%). Total leukocyte counts were within the reference range in all cats. One cat was thrombocytopenic (platelet count, 22 × 109/L; reference range, 110–413 × 109/L). In other cats, platelet counts were either within the reference range or deemed adequate on blood smear.
Two cats were hypercalcemic. One cat had elevations in both total (3.18 mmol/L) and ionized Ca (1.65 mmol/L; reference ranges, 2.23-2.73 and 1.28-1.48 mmol/L). The other cat had elevated total Ca at 2.98 mmol/L at the time of diagnosis (ionized Ca was not evaluated). That cat was incidentally diagnosed with MRD during staging for mammary gland and mast cell tumors. Total Ca was normal at the time of starting therapy, when there was progression of MRD based on the development of skin masses and progressive hyperglobulinemia.
Six cats were azotemic. Three of those cats had a prior history of chronic kidney disease (CKD). Of 11 cats that had urinalysis performed, three cats had trace, two had 1+, three had 2+, and two had 3+ proteinuria. The urine protein/creatinine ratio was not assessed in any case. Urine protein electrophoresis was performed in one cat and showed a monoclonal gammopathy.
Prothrombin and partial thromboplastin times were evaluated in two cats. Values were within reference ranges in both cats.
Staging tests varied in the included cases (Table 2). Splenic cytology was performed in 12 cases and was either positive or suspicious for plasma cell neoplasia in 10 cases (92%). Liver cytology was performed in 10 cases and was either positive or suspicious for plasma cell neoplasia in 8 cases (80%). No cat with definitive or suspicious cytological hepatic involvement had elevated liver enzymes. One cat (case 9) with confirmed hepatic and splenic involvement had no abnormal findings on abdominal ultrasonography. None of the cats with increased liver enzymes had liver cytology performed.
Complete survey radiographs were not performed in any case. Thoracic radiographs were performed in 11 cats. In 1 cat, a pathological humeral fracture was identified. Two cases had radiographic changes consistent with heart failure, including cardiomegaly and pulmonary interstitial infiltrate (suspected pulmonary edema). No other case had radiographic changes assessed as MRD-related. Three cases had abdominal radiography (2 in conjunction with thoracic radiography), and no lesions suspected to be related to MRD were identified.
Treatment
The median time from diagnosis to institution of therapy was 0 days (range, 0–51 days). Only two cats had delays in commencement of therapy. Thirteen cats received chemotherapy and two received prednisoloned alone. Eight cats received melphalan as initial chemotherapy. Of those, six cats received concurrent prednisolone, one received dexamethasonee, and one received no corticosteroid. In two cats, melphalan was instituted after starting prednisolone (at 27 and 56 days, respectively). Five cats were treated with cyclophosphamide and prednisolone as first-line therapy.
Five cats had two chemotherapy protocols, all receiving chlorambucilf following cessation of their initial protocol due to either toxicity or PD. One cat had three protocols (melphalan/prednisolone, then chlorambucil, then cyclophosphamide).
When evaluating chemotherapy toxicity, it is possible that some cats identified as having chemotherapy-related myelotoxicity may have had PD causing myelosuppression. Where there was no clinical evidence of progression, myelosuppression was presumed to be chemotherapy-related. In cases with either increasing globulins or other signs of progression (e.g., enlargement of skin masses), myelosuppression was attributed to MRD.
Dosing and Toxicity
The starting dose of melphalan was 2 mg/cat per os (PO) q 2 days (n = 1), q 3 days (n = 4), q 5 days (n = 1), or q 7 days (n = 2). The median starting dose was equivalent to 0.103 mg/kg/day (1.89 mg/m2/day; range, 0.075–0.24 mg/kg/day or 1.16–3.867 mg/m2/day). Seven cats had adequate follow-up information to assess toxicity (Table 3). Hematologic toxicity was common. One cat had PD (confirmed with bone marrow aspirate) with severe neutropenia and thrombocytopenia. Melphalan may have contributed to myelosuppression in that case because cytopenias improved following cessation of melphalan and prior to introduction of chlorambucil.
Melphalan was discontinued in five cats due to grade 3–4 myelosuppression. In one cat, melphalan was temporarily discontinued due to the development of an abscess then permanently discontinued when the patient became infected with Mycoplasma hemofelis. The median duration of melphalan treatment was 51 days (range, 17–303 days), which negatively correlated with starting dose (Spearman rank correlation, −1).
Five cats were treated with cyclophosphamide as their first protocol, and one cat as its third protocol. The starting dose was 25 mg/cat PO twice weekly (n = 4), q 3 days (n = 1), or q 5 days (n = 1). The median dose intensity was 168 mg/m2/wk (range, 149–224 mg/m2/wk). No cats received concurrent diuretic with cyclophosphamide. Toxicity was common, but generally mild (Table 3). Two cats had dose reductions due to neutropenia. In no case was cyclophosphamide discontinued due to hematologic toxicity. In one cat, grade 4 anemia developed early in the course of treatment and improved following medical management with gastroprotectants and corticosteroid dose reduction. The other cat with grade 4 anemia concurrently developed neutropenia and PD.
One cat developed hematuria after 467 days of cyclophosphamide treatment, which was discontinued due to suspected cyclophosphamide-induced sterile hemorrhagic cystitis. Abdominal ultrasonography revealed a urinary bladder mass that was cytologically diagnosed as a carcinoma, and which was not identified at initial staging prior to chemotherapy.
Although not significantly different, the median duration of cyclophosphamide as initial therapy was 352 days (range, 151–528 days, with two cats still alive and receiving cyclophosphamide at follow up) compared with 51 days for melphalan (Figure 1).
Citation: Journal of the American Animal Hospital Association 51, 4; 10.5326/JAAHA-MS-6216
Of the cats that received chlorambucil, one had previously received cyclophosphamide (discontinued due to hematuria). Five had previously received melphalan, which was discontinued due to myelosuppression in four cats and PD in one cat. The starting dose was 2 mg/cat PO q 2 days (n = 2), q 3 days (n = 1), q 4 days (n = 2), or q 5 days (n = 1). The median dose intensity was 14.2 mg/m2/wk (range, 10.7–21.9 mg/m2/wk). Chlorambucil was well tolerated (Table 3). One cat received two dose reductions due to grade 1 neutropenia. Median duration of chlorambucil therapy was 82.5 days (range, 19–351 days). Three cats had either suspected or confirmed PD, and three cats died or were euthanized without obvious PD while receiving chlorambucil.
For the 13 cats that received prednisolone, median starting dose was 5 mg/cat/day (1.22 mg/kg/day; range, 5–10 mg/cat/day or 0.77–2.7 mg/kg/day). The dexamethasone-treated cat received 1 mg/day (0.26 mg/kg/day). In 5 cats, the corticosteroid was discontinued prior to death (due to PD in 2 cats, heart failure in 1 cat, institution of piroxicam in 1 cat, and taper and discontinuation of prednisolone with ongoing CR in 1 cat). The median duration of therapy was 226 days. The other cats continued to receive a corticosteroid (tapering schedules varied) until death or euthanasia.
Response
Melphalan ± Corticosteroid
Two of seven evaluable cats (29%) cats achieved CR and three cats achieved a PR (43%). The overall response rate was 71%. One cat had SD for 51 days, following progression with prednisolone alone. In the remaining cat, response could not be determined because globulin normalized following amputation (despite abdominal organ involvement) and repeat imaging and cytology were not performed. Median time to response was 133 days (range, 11–222 days).
Cyclophosphamide and Prednisolone
Three of six cats achieved CR and two achieved PR. The overall response rate was 83%. One cat receiving cyclophosphamide as its third chemotherapy protocol had SD for 34 days. Median time to response was 45 days (range, 10–66 days).
Chlorambucil
Two cats achieved a CR after an initial PR on melphalan (discontinued due to toxicity). One cat had a CR after previous PD. Times to CR in those three cats were 48, 84, and 266 days, respectively; however, in the third cat, no globulin or total protein values were recorded until 266 days after starting chlorambucil. Two cats maintained a CR after discontinuation of a previous protocol due to toxicity, and one cat had progressive disease 6 wk after starting chlorambucil (response to prior melphalan therapy in that cat could not be determined).
The variability in the recheck schedule meant that actual time to response may have been shorter than that calculated. Because rechecks were performed frequently in nearly all cases until documented CR, it was likely that actual and calculated times to response were very close.
Of the cats initially receiving prednisolone alone, one had an initial PR but experienced PD after 26 days of therapy and then started melphalan, one had a PR and maintained PR following initiation of melphalan at 57 days, one had SD for 10 wk, and one cat died 3 days after diagnosis and response could not be evaluated.
Outcome
Of 14 cats with adequate follow up information, 11 cats (79%) were deceased. Of those that died, death was attributed to MRD-related causes in 7 cats (64%), “other” causes in 2 cats (i.e., renal failure with anemia and recurrent pyelonephritis, respectively), and possible treatment-related causes or unknown cause in 1 cat each. Three cats (21%) were alive at 151, 528 days, and 615 days after diagnosis, 2 with CR and 1 with PD.
The MST of seven evaluable cats initially treated with melphalan was 252 days (range, 107 to >615 days). The five cats treated with cyclophosphamide and prednisolone as their first protocol had a MST of 394 days (range, 174 to >528 days) as shown in Figure 2. Choice of first treatment protocol was not significantly associated with survival [Cox proportional hazard for melphalan versus cyclophosphamide, 1.43; 95% confidence interval (CI), 0.34–6; P = .62]. The survival times for the two cats treated with prednisolone alone were 3 and 86 days, respectively.
Citation: Journal of the American Animal Hospital Association 51, 4; 10.5326/JAAHA-MS-6216
Cats that were azotemic at diagnosis had a MST of 174 days, versus 434 days for nonazotemic cats. Azotemia was not associated with an increased risk of death (Cox proportional hazard, 1.71; 95% CI, 0.42–6.9; P = .454). Cats that were anemic at the time of diagnosis had a MST of 252 days versus 289 days for nonanemic cats. Anemia was also not associated with survival (Cox proportional hazard, 2.31; 95% CI, 0.28–18.9; P = .435).
Cats that were hypoalbuminemic at diagnosis had a MST of 684 days, versus 214 days for those with normal albumin. Risk of death for cats with normal albumin versus hypoalbuminemia approached statistical significance (Cox proportional hazard, 5.11; 95% CI, 0.96–22.3; P = .0564).
Cats achieving a CR had a MST of 870 days versus 289 days for cats achieving PR as best response to therapy. Response was not statistically significant (Cox proportional hazard, 3.61 for PR versus CR; 95% CI, 0.59–22; P = .164).
Discussion
In the authors' institution, MRD requiring systemic therapy is more common than solitary plasmacytoma in cats, consistent with previous studies.1,3–15 This may be influenced by referral bias because suspected solitary plasmacytoma may be treated surgically by the primary veterinarian. Because of the high incidence of abdominal organ involvement in MRD in this and other studies, abdominal ultrasonography with cytology of liver and spleen should be considered in the evaluation of cats with suspected MRD.1–4
Clinical signs and historical abnormalities in this study population were nonspecific, with weight loss and reduced appetite being the most common. In other studies, vomiting, diarrhea, polyuria, and polydipsia were also commonly noted.1,4,15 In the current study, one cat had lameness and a radiographic lytic bone lesion. Reported incidence of lameness or bone lesions ranges from ~7 to 67% in other studies.1,4,15 The frequency of osteolytic lesions in the current study may be underestimated because complete survey radiographs were not performed. None of the cats in this study had clinical evidence of bleeding tendencies (e.g., petechiation, epistaxis). In contrast, up to 13% of cases reported in other studies had evidence of bleeding tendencies.1,4,15 However, only two cats had coagulation panels performed, so subclinical coagulopathy may have been present. Other physical examination abnormalities, including heart murmur, dehydration, abdominal masses, organomegaly, ocular abnormalities, and skin masses were consistent with previous reports.1,4,15
Although not specifically evaluated for HVS, two cats exhibited signs consistent with heart failure shortly before their MRD diagnoses and one cat had tortuous retinal vessels. Taken together, those findings suggest that HVS may have been present in this patient population.
Hyperglobulinemia was the most common laboratory abnormality, consistent with previous publications.1,4,15 In three cases, asymptomatic hyperglobulinemia was noted months prior to diagnosis of MRD. Monoclonal gammopathy of undetermined significance (MGUS) is considered a premalignant disorder in humans with increased risk of progression to MRD.21 It has been suggested that MGUS occurs in cats, with one case having documented paraproteinemia for 9 yr before clinical progression to MRD.4 Without full investigation at the time of first-documented hyperglobulinemia, it is unknown if the cats in this study had MGUS or were simply asymptomatic for MRD.
Similar to previous reports, proteinuria, anemia, and azotemia were common in this study.1,4,15 Biochemical evidence of hepatopathy has been described in cats with MRD, although it is not specific for plasma cell infiltration in the liver.1 The results of this study also suggest that lack of liver enzyme abnormalities and ultrasonographic changes cannot rule out liver involvement in MRD and cytological evaluation should be considered.
Previously reported negative prognostic factors in feline MRD include systemic clinical signs and/or abdominal organ involvement, treatment with prednisolone versus chemotherapy, poor response to chemotherapy, aggressive disease classification (i.e., anemia, hypercalcemia, bony lesions with pathological fracture, Bence-Jones proteinuria, and azotemia), and poorly differentiated tumors.1,3,15 This study specifically excluded cats with solitary plasma cell tumors without paraproteinemia or systemic signs, and tumor differentiation could not be retrospectively determined from the available records. The majority of cats in this study were anemic at presentation, which was not associated with a poorer outcome, nor was azotemia. In humans, renal impairment is a known complication of, and negative prognostic factor for, multiple myeloma.22 One study of dogs with multiple myeloma found that abnormal renal function did not significantly impact survival time.17 Because MRD and CKD are both most commonly seen in older cats, it is unclear whether pre-existing CKD versus MRD-related renal impairment may influence prognosis. Because only one cat had radiographically confirmed bone lysis, the prognostic significance of that finding could not be determined.
Hypoalbuminemia is a negative prognostic factor in some cancers in companion animals and in multiple myeloma in humans.18,23,24 The prognostic significance of hypoalbuminemia has not been previously evaluated in cats with MRD. Serum albumin level was not prognostic in this population, which may have been due to small sample size.
Response to therapy was not prognostic in this study, despite cats with CR having numerically superior survivals compared with those achieving a PR. Again, this may be due to small sample size, and larger studies may be necessary to evaluate the effect of response on survival in feline MRD. Reported MSTs of cats with MRD treated with chemotherapy range from ~4 to 9 mo, which are similar to survival times for cats treated initially with melphalan or cyclophosphamide in the current study (8 and 13 mo, respectively).1,15 Most cats initially receiving melphalan were eventually switched to chlorambucil, which may have played a role in extending survival.
Anemia in cyclophosphamide-treated cats in this study may have been due, in part, to prolonged treatment. Other possibilities include disease progression, gastrointestinal bleeding, and hematuria secondary to a urinary bladder carcinoma in one cat. It is possible that the development of the urinary bladder carcinoma could have been related to cyclophosphamide, but a causal relationship could not be proven and histopathological analysis of the tumor was not performed. Erythrophagia has been reported in feline MRD and could contribute to anemia, although this phenomenon was not identified in any case in this study.25
Infections during therapy were noted in three cases. One developed a soft-tissue abscess requiring temporary discontinuation of melphalan and was subsequently infected with Mycoplasma hemofelis after melphalan was reinstituted. That cat was in CR when both infections were diagnosed. Another cat developed signs consistent with an upper respiratory tract infection (with concurrent PD) during cyclophosphamide therapy. The third cat developed recurrent pyelonephritis without PD during treatment with chlorambucil. Chemotherapy and MRD may both increase risk of infection, and some studies suggest using prophylactic antibiotics in cats with MRD.15 The results of this study suggest that that approach may not be necessary because most cats had no evidence of infection and, of those that did, two were in a CR. In people with multiple myeloma, close monitoring for infection, rather than antibiotic prophylaxis, is recommended.18
The limitations of this study are primarily attributable to its retrospective nature. Those limitations include a small sample size, incomplete staging, limited response evaluation (total protein and globulin versus M-component), inability to assess factors such as time to progression, and incomplete follow-up information in some patients.
Conclusion
To the authors' knowledge, this is the largest reported study of cats with MRD treated with chemotherapy and the first to compare the efficacy and toxicity of different chemotherapy protocols in feline MRD. The presentation and clinicopathological abnormalities in this series are consistent with previous reports. Cyclophosphamide and corticosteroids as first-line therapy was well tolerated. It is possible that lower starting doses of melphalan could have been similarly effective while allowing longer duration of treatment. Chlorambucil also appeared to be an effective and well-tolerated alternative for the treatment of feline MRD. Further studies evaluating the efficacy of chemotherapy for the treatment feline MRD are warranted, including the prospective evaluation of cyclophosphamide and corticosteroids in a larger number of cases, which may also allow better determination of prognostic factors.
Kaplan-Meier analysis of treatment duration for cats with systemic myeloma-related disorder receiving either melphalan (dashed line) or cyclophosphamide (solid line). Treatment was discontinued due to progressive disease or toxicity. Animals still undergoing treatment at the time of follow up were censored (crosses). Median duration of treatment with melphalan was 51 days compared to 352 days for cyclophosphamide.
Kaplan-Meier survival analysis of cats with systemic myeloma-related disorder receiving either melphalan (dashed line) or cyclophosphamide (solid line) as initial chemotherapy protocol. Animals alive at the time of follow up were censored (crosses). Median survival time for cats initially treated with melphalan was 252 days versus 394 days for cats initially treated with cyclophosphamide.
Contributor Notes
