Hematologic Toxicosis Associated with Doxorubicin Administration in Cats Deborah A. O’Keefe, DVM, MS, and David J. Schaeffer, PhD
The hematologic toxicity of doxorubicin, 30 mg/m2 body surface area (BSA) every 21 days to a cumulative dose of 300 mg/m2, was evaluated in six cats. Complete blood and platelet counts were performed daily during the first treatment cycle. They were monitored before treatment for all remaining cycles, and at the average neutrophil nadir (day 8) starting with cycle 4. Significant poikilocytosis developed after the first treatment and remained throughout the study, although anemia did not occur. No other red blood cell abnormalities were seen. Platelet counts remained within the reference range throughout the first treatment cycle, but mild thrombocytopenia (88,000-288,000/uL) was found in 11.3%of subsequent complete blood counts (CBCs). Thrombocytosis was seen in 30.9%of CBCs. Neutropenia did not occur during the first treatment cycle although neutrophil counts did decrease, with the nadir occurring between days 8 and 11. All neutrophil counts returned to pretreatment values by day 14. Neutropenia was documented after 14 of 46 (30.4%)doxorubicin treatments, and was associated with fever in 5 cats (10.9%).All fevers responded to oral antibiotic therapy. Neutropenia that lasted more than 14 days developed in two cats, necessitating dosage reduction to 25 mg/m2 BSA. At the dose used in this study, doxorubicin administration was associated with acceptable hematologic toxicosis in most cats. (Journal of Veterinary Internal Medicine 1992; 6:276-282)
DOXORUBICIN IS an anthracycline antibiotic with proven antineoplastic activity against a broad range of tumors in both dogs and people.’-4 Its use can be associated with a variety of toxicoses in both species, including myelosuppression,enterocolitis,alopecia, and cardiomy~pathy.~-’Although doxorubicin has also been shown to have antineoplastic activity against a variety of feline tumor^,^,^ there are few reports of its toxicity in this Reported toxicities have included anorexia, myelosuppression, cardiac arrhythmias, and renal disease. Although myelosuppression is reported to occur, little information regarding the degree or timing of doxorubicin-induced cytopenias exists. This information would be valuable in the development and use ~
~
~~~~
~
From the Departments of Veterinary Clinical Medicine (OKeefe) and Veterinary Biosciences (Schaeffer), University of Illinois, Urbana, Illinois. Presented as an abstract at the American College of Veterinary Internal Medicine Eighth Annual Veterinary Medical Forum, May 10-13, 1990. The authors thank Adria Laboratories, Columbus, Ohio, for donating the doxorubicin used in this study. Accepted for publication February 15, I99 I . Reprint requests: Deborah A. OKeefe, DVM, Department of Veterinary Clinical Medicine, University of Illinois, 1008 W. Hazelwood Dr., Urbana, IL 6 180 I .
of doxorubicin-containing chemotherapy protocols in the cat. This study determined the acute and chronic hematologic toxicoses associated with doxorubicin when administered to cats at a dose used in veterinary practice.8-11Systemic toxicoses will be reported in a separate article. Materials and Methods Cats
Six healthy, young adult cats (5 female, 1 male) were obtained for this study. Baseline data evaluations included: complete blood count (hematocrit, red blood cell count, hemoglobin concentration, white blood cell and white blood cell differentialcount), platelet count, urinalysis, serum chemistry profile (urea nitrogen, creatinine, sodium, potassium, calcium, chloride, phosphorus, alanine transaminase, aspartate transaminase, alkaline phosphatase, glucose, bilirubin, total protein, albumin, cholesterol), FeLV, ELISA, and fecal examination for parasitic ova. All results were normal. All cats had percutaneous kidney biopsies done 2 weeks before the first doxorubicin treatment.
276
VOl. 6
. NO. 5,1992
277
HEMATOLOGIC TOXlCOSlS AND DOXORUBICIN ADMINISTRATION
Doxorubicin Administration Doxorubicin* was administered initially at a dose of 30 mg/m2 body surface area (BSA) every 21 days. The doxorubicin was diluted to 0.5 mg/mL and was given through a cephalic catheter over a 5-minute period. Due to prolonged neutropenia, dosage reduction to 25 mg/ m2 BSA was necessary in cats 3 and 5 starting with treatments 5 and 6, respectively. The cumulative doxorubicin dose for five cats (cats 1-4, 6) was 300 mg/m2 BSA. Cat number 5 died during the study after receiving 225 mg/m2 BSA.
Hematologic Monitoring Complete blood and platelet counts were monitored daily for the first 21-day cycle. These data were used to determine the average neutrophil nadir, as has been done for other chemotherapeutic agents.12J3Complete blood counts and platelet counts were monitored before treatment for all remaining cycles, and at the average neutrophil nadir (day 8) starting with cycle 4. Neutropenia was defined as an absolute neutrophil count of less than 250O/uL.l4 Thrombocytopenia and thrombocytosis were defined as platelet counts less than 300,OOO/uL and greater than 800,OOO/uL, respe~tive1y.l~Blood smears taken 2 1 days after each treatment and at the end of the study were stained with a modified Wright's stain? and evaluated for abnormalities of erythrocyte morphology, i.e., poikilocytosis. A poikilocyte was defined as any erythrocyte with abnormal morph~logy.'~ The slides were randomized and evaluated for the number of poikilocytes per 1000 red blood cells without knowledge of treatment or cat number.
Statistics After logarithmic transformation, the poikilocyte data were analyzed as a nested analysis of variance in a repeated measures design.16 Linear contrasts were constructed to test for differences between treatment times. The Greenhouse-Geiser-theta-statistic was used to test the correctness of the assumptions of the repeated measures analysis and to effect a small correction on the Fstat is ti^.'^ Because cat 5 was lost shortly after treatment 8, three analyses were carried out. The first used all six cats out to treatment 7; the second dropped cat 5 and analyzed the remaining cats out to treatment 7; the third analyzed the five cats out to the end of the study. Significant differenceswere present when P < 0.05 was determined. Neutrophil counts at the nadir of the first treatment cycle were compared with pretreatment values using the paired sample t-test." Linear regression was used to eval-
* Adriamycin, Adna Laboratories, Columbus, OH.
t HEMA-TEK, Miles Laboratories, Inc., Elkhart, IN.
uate the effect of repeated treatments on both the pretreatment and nadir neutrophil counts." Significant differences were considered present when P < 0.05 was determined.
Results
Red Blood Cells Table 1 summarizes the red blood cell parameters at the beginning and end of the study. There were no significant changes in packed cell volumes (PCVs) or red blood cell indices throughout the study. However, as shown in Table 2, the number of poikilocytes did increase in all cats except cat 1. Ovalocytes and echinocytes were the most frequently observed poikilocytes, although keratocytes, acanthocytes, and schistocytes were also seen (Figs. 1, 2). Poikilocytosis was not associated with anemia. The three statistical analyses of the poikilocyte data gave similar results. Out to treatment 7, with either 5 or 6 cats, the analyses showed a linear (increasing) trend in the number of poikilocytes per 1000 RBCs. This increase was statistically significant even after the first treatment. When data from the 5 cats were analyzed to the end of the study, polynomial contrasts included a quadratic term. These latter results indicate that the number of poikilocytes increased through treatment 7 and then became either asymptotic or began to slightly decrease. From plots of the data we favor an asymptote but cannot rule out the onset of a decline.
Platelets Platelet counts remained within the reference range (300-800,000/uL) throughout the first cycle of doxorubicin therapy. Mild thrombocytopenia (range 88,000288,OOO/uL, mean 192,80O/uL)was detected on 12 of 97 ( 1 1.3%) subsequent CBCs. Six of these were samples taken before a doxorubicin treatment (three before treatment 3, one before treatment 4, and two before treatment 5 ) and six were samples taken 8 days after treatment (one each during treatments 5 through 10). Cat number I was thrombocytopenic before treatment 5 , but had a normal platelet count 8 days after this treatment. Thrombocytosis (range 8 17,000-1,300,000/uL, mean
TABLEI . Red Blood Cell Parameters in Doxorubicin-treated Cats ~~
PCV 70 RBC X106 Hb (g) MCV fl. MCHC 70
Prior to Treatment 1 Mean (Range)
Post Treatment 10 Mean (Range)
Reference RangeI4
32. I (27-36.4) 5.93 (5.01-6.74) 10.4 (9.2-12.2) 54.0 (52.0-56.0) 32.6 (30.0-36.0)
27.6 (24-32) 5.31 (4.46-6.26) 9.2 (8.1-10.5) 52.3 (48.5-56.0) 33.2 (32.7-34.0)
24.0-45.0 5.0-10.0 8.0-15.0 39.0-55.0 3 1.0-35.0
278
Journal of Veterinary Internal Medicine
O’KEEFE AND SCHAEFFER
TABLE 2. Poikilocytosis in Doxorubicin-treated Cats (Poikilocytes per 1000 RBCs) Treatment Number Cat No.
0
I
2
3
4
5
6
7
8
9
10
1
2 3 4 5 6
12 20 14 5 10 6
26 32 53 22 37 39
6 I25 47 9 22 23
41 57 I34 68 32 34
13 I94 I69 21 95 Ill
8 374 I09 71 161 I22
16 284 I59 25 I 63 I20
23 224 232 380 26 I 170
31 265 235 215 ND I22
33 159 247 307 ND 137
22 I 252 158 ND 143
Mean
I1
35*
39
IOI*
141*
I49*
215*
174*
177*
157*
61*
I1
* Statisticallydifferent from pretreatment (treatment 0 ) value, P < 0.05. ND, no data points due to cat’s death.
999,OOO/uL) was seen in 30 of 97 (30.9%)CBCs, with 19 cases occurring before a treatment and 11 cases occurring 8 days after a treatment.
The neutrophil was the only white blood cell affected by doxorubicin administration. The average daily neutrophi1 counts during the first treatment cycle are shown in Figure 3. During this cycle, doxorubicin caused a significant decrease in the neutrophil count with the average nadir occurring on day 8. Neutrophil nadirs in individual cats ranged from 2640-7 106 neutrophils/uL (mean 4095/uL, median 363 l/uL) and occurred on days 8-1 1.
Neutrophil counts in all cats returned to pretreatment values by day 14. A small increase in band neutrophils (300- 1200/uL) was noted in five of six cats between days 13 and 18. Figure 4 shows the average neutrophil counts before and at the average neutrophil nadir of each doxorubicin treatment. The pretreatment and nadir average neutrophil counts did not change significantly as treatment progressed. Although the average nadir neutrophil counts did not decrease as treatments progressed, neutropenia did occur in individual cats. Neutropenia was documented after 14 of 46 treatments (30.4%) and all neutropenic episodes occurred in cats 3-6 (Table 3). Fever occurred during five neutropenic episodes, which represented
FIG. I . Peripheral blood smear from cat 3 after a cumulative doxorubicin dose of 245 mg/mz BSA. Ovalocytes(closed arrows) and ovaloechinocytes (open arrows) are present. Two blister cells, which are vacuolated red blood cells in the process of keratocyte formation, are also present (arrowheads) (original magnification X 160, modified Wright’s stain).
FIG.2. Peripheral blood smear from cat 4 after a cumulative doxorubicin dose of 2 10 mg/m2 BSA. Ovalocytes (closed arrow) and ovaloechinocytes (open arrow) can be seen. One blister cell (small arrowhead) and one keratocyte (large arrowhead)are present, as are other unclassified poikilocytes (original magnification X 160, modified Wright’s stain).
White Blood Cells
Vol. 6
. NO. 5, 1992
279
HEMATOLOGIC TOXICOSIS AND DOXORUBICIN ADMINISTRATION
25
T
35
I
t
T
1
4
5
6
7
8
9
1
0
Treatment Number FIG.3. Average (+/- SD) daily neutrophil counts before and after the initial administration of doxorubicin, 30 mg/m2 BSA IV (n = 6). The neutrophilcount on day 1 was taken before treatment. The day 8 value is significantly different than the day 1 value, P = 0.029.
10.9%of treatments given. All fevers were responsive to oral amoxicillinS or cefadroxils therapy. Fevers developed during four additional treatments but unfortunately CBCs were not done to determine if neutropenia was present. These fevers lasted for at least 2 days, ranged from 103.2- 104.7"F, and were responsive to amoxicillin therapy. They occurred on days 6 and 8 of treatment 2, and days 7 and 9 of treatment 3, which are times compatible with the neutrophil nadir. If these fevers represented neutropenic episodes, then febrile neutropenia occurred in 9 of 58 (15.5%)treatments. Cats 3 and 5 had prolonged neutropenia after treatments 4 and 5 , respectively. Cat 3 had neutrophil counts of 1 152/uL and 2074/uL on days 8 and 2 1, respectively, of treatment cycle 4. Cat 5 had neutrophil counts of 164/ UL and 700/uL on days 8 and 21 of treatment cycle 5. This necessitated discontinuing treatment for one treatment cycle in each cat. Subsequent doxorubicin doses were decreased to 25 mg/m2 BSA in both cats; these doses eliminated the prolonged neutropenia. However, neutropenia still developed after 3 of 8 subsequent treatments. Cat number 5 was found dead in her cage 3 days after the eighth doxorubicin treatment. Her neutrophil count before that treatment had been 14,552/uL with 1155 bands/uL. At necropsy, a 0.5-cm abscess was found in the subcutis overlying the left abdominal wall. It did not appear to communicate with the abdominal cavity. However, 20-30 mL of sanguinous, purulent fluid was found
$ Amoxi-tabs, Beecham Laboratories,Bristol, TN. Cefa-tabs, Aveco Co., Inc., Fort Dodge, IA.
FIG.4. Average (+/- SD) neutrophil counts in cats treated with doxorubicin, 30 mg/m2 BSA every 2 1 days (n = 6). Solid squares are average pretreatment values, open squares are average nadir values.
in the abdominal cavity. Cytology of this fluid was consistent with septic peritonitis and a group G Streptococcus species was cultured.
Discussion Doxorubicin administration has been associated with dose-related decreases in hematocrit, hemoglobin, and red blood cell numbers in dogs, rabbits, and p e ~ p l e . ~ , ~ " - * ~ When dogs were given doses similar to those used in this study, hematocrits and hemoglobin concentrations were transiently decreased 7 days after treatment but returned to pretreatment values before the next dose.23Higher weekly doses have produced sustained regenerative and
TABLE 3. Neutropenic Episodes in Doxorubicin-treated Cats
Cat Number
Treatment Number ~
3 3 3 4 4 4 4 5 5 5 6 6 6 6
Day 8 Neutrophil Count (Cells/uL)
Fever*
~
4
8t 9t 7 8 9 10 4 5 6t 5 8 9 10
1152 1536 2444 1058 2494 264 I972 I248 164 2278 I764 1505
736 2379
* Temperature > 102.5"F for at least 2 consecutive days.
t Dose reduced to 25 mg/m2 BSA.
102.7
-
103.4 103.8 103.3 103.8 -
-
280
O'KEEFE AND SCHAEFFER
nonregenerative anemias in dog^.^,^' The nonregenerative anemia results from the toxic effect of doxorubicin on bone marrow erythrocyte precursors,' while the regenerative anemia may be related to increased poikilocyte destruction." In contrast to the species mentioned earlier, alterations in the hematocrit and other red blood cell parameters were not seen in these cats. The reason for this is not clear, although feline erythrocyte precursors may be more resistant to the toxic effects of doxorubicin. Decreased clearance of poikilocytes may also contribute. Poikilocytosis developed in these cats and has been reported in both dogs and people after doxorubicin therapy.21*22 The exact mechanism by which doxorubicin produces alterations in red blood cell morphology is not known. Doxorubicin is known to bind to cell membranes and can affect them in a variety of way^.'^'^ Possible mechanisms for the development of poikilocytosis include the generation of reactive oxygen compounds resulting in membrane lipid p e r o x i d a t i ~ n ,inhibition ~~,~~ of Na+/K+ dependent adenosine triphosphatase activity," and alteration in calcium tran~port.'~ The number of poikilocytes in these cats increased in a linear manner through the first seven treatments indicating that initially the effects of each treatment were additive. After treatment 7, the number of poikilocytes reached a plateau, and possibly even declined. The erythrocytes may have developed protective mechanisms that made them less susceptible to the membrane effects of doxorubicin. Significant increases in poikilocyte number were seen in these cats after one doxorubicin treatment, whereas dogs did not develop poikilocytosis until 16 weeks after the initiation of weekly treatments. The reason for this difference is uncertain. Canine erythrocytes may be more resistant to doxorubicin; abnormal erythrocytes may be more readily removed by the mononuclear phagocytic system in the dog; and/or erythrocyte precursors, rather than circulating red blood cells, may be the cells most affected by doxorubicin in the dog. The magnitude of poikilocytosis was also much greater in these cats compared with that reported in dogs. The mean number of poikilocytes in dogs before treatment and after 16 weeks of treatment was 0.4 and 3.4 per 1000 RBCs, respectively.'* The mean number of poikilocytes in these cats at comparable times was 1 1 and 141 per 1000 RBCs, respectively. The number of poikilocytes produced by the administration of doxorubicin to dogs may be underestimated on a peripheral blood smear because of the rapid removal of these abnormal erythrocytes by the mononuclear phagocytic system. This rapid removal may then contribute to the anemia seen with doxorubicin therapy in this species.22The mononuclear phagocyte system of the cat may not be as efficient at removing abnormally shaped erythrocytes as
Journal of Veterinary Internal Medicine
it is in the dog. The increased number of poikilocytes in the cat compared with that in the dog even before receiving doxorubicin would support this theory as would the observation that the spleen does not appear to remove Heinz bodies as readily in cats as it does in other spec i e ~If. this ~ ~ is the case, decreased poikilocyte removal may explain, at least in part, the lack of anemia in these cats. Thrombocytopenia does not appear to be a major toxicity when doxorubicin is administered to cats at 30 mg/ m2 BSA. Thrombocytopenia was not detected when cats were treated with two doses of 10-40 mg/m2," and platelet counts did not decrease during the first treatment cycle of the present study. The episodes of thrombocytopenia during the remainder of this study were infrequent and mild. Platelet counts never decreased to a level that would be associated with clinical bleeding. Because hematologic toxicosis is dose-related, larger doxorubicin doses could result in significant thrombocytopenia. Thrombocytosis was more common in these cats than thrombocytopenia and occurred more frequently before therapy than 8 days after treatment. This may reflect a rebound thrombocytosis due to bone marrow stimulation after thrombocytopenic or neutropenic episodes. Thrombocytopenia causes the release of thrombopoietin, which stimulates platelet production in the bone marr~w.~ Granulocyte-macrophage ' colony-stimulating factor (GM-CSF), a glycoprotein released in response to neutropenia, is also known to stimulate cells ofthe megakaryocytic line.32Splenic contraction during blood sampling may also have contributed to the increased platelet counts. Thrombocytosis of the degree seen in this study is probably clinically unimportant, although platelet counts in excess of 1,OOO,OOO/uL have been associated with hemorrhage and thrombosis in people.33 At the doxorubicin dose used in this study, a decreased neutrophil count was the most consistent hematologic abnormality seen in the cat. A neutrophil nadir occurring at 8-1 1 days post-treatment is similar to what is reported in dogs. However, dogs given a slightly larger doxorubicin dose did not develop a significantly smaller number of WBC until the nadir of the third treatment cycle,23whereas the cats in this study had smaller numbers of neutrophils after the first treatment. Also, neutrophi1 counts in the dog are reported to return to normal within 17 to 20 days, which is somewhat longer than the 14-day recovery seen during the first treatment cycle in these cats.34 The incidence of neutropenia (30%)was fairly common in these cats. Also, the incidence of febrile neutropenia was greater than that reported in dogs (10.9% vs. 3%).5Cats may be more susceptible to the toxic effects of doxorubicin because of their small size. In dogs doxorubicin toxicity is inversely related to body eight.^ The apparent increase in febrile neutropenic episodes in cats
Vol. 6
. NO.5,1992
HEMATOLOGIC TOXlCOSlS AND DOXORUBICIN ADMINISTRATION
may actually be the result of increased monitoring. Temperatures were taken daily and cats were sometimes febrile with no change in behavior or attitude. The dogs studied, however, were clinical patients that were not monitored as closely and could have had undetected episodes of fever. Regardless, the frequency of febrile neutropenia seen in these cats would be clinically acceptable, and all cats responded well to oral antibiotic therapy. Two cats remained neutropenic on the day of the next scheduled treatment, which necessitated skipping that treatment. Because of this, the neutrophil count in clinical patients both before therapy and at the neutrophil nadir should be monitored. If prolonged neutropenia should occur, the dosage of subsequent doxorubicin treatments could be lowered to 25 mg/m2 BSA which alleviated the prolonged neutropenia. At 30 mg/m2 BSA, the toxic effect of doxorubicin on the bone marrow of cats is not cumulative and does not overwhelm the ability of bone marrow to regenerate. The average pretreatment neutrophil counts and the average nadir neutrophil counts did not decline throughout the study. Also, neutropenic episodes in individual cats did not increase as treatment progressed. However, there was great variability in the degree of neutropenia in individual cats from cycle to cycle. This makes it imperative to continue hematologic monitoring in clinical patients throughout therapy. There appeared to be variability in the susceptibilityof the cats of this study to doxorubicin-induced hematologic toxicosis. One cat did not develop a significant poikilocytosis; all neutropenic episodes occurred in four of six cats; and only two cats developed prolonged neutropenia. The reason for this variability was not addressed. There may be an inherent difference in the sensitivity of target cells to doxorubicin or it could reflect differences in drug metabolism. Cat 5 died of septic peritonitis 3 days after treatment 8. The peritonitis most likely resulted from extension of a subcutaneous abscess into the left abdominal area. The cause of this abscess was not determined. It is unlikely that the percutaneous kidney biopsy done in this area at the beginning of the study was responsible because the cat had experienced no problems for the previous 20 weeks of the study. It is also unlikely that myelosuppression contributed to the development of peritonitis because the cat had not been neutropenic since the nadir of treatment 6 and actually had a slightly increased neutrophi1 count 3 days before death. Immune function was not evaluated in these cats, but doxorubicin is known to be immunosuppressive in people.35Decreased immune function may have contributed to the development of peritonitis from a superficial abscess in this cat. In summary, doxorubicin, when used as a single agent at a dose of 30 mg/m2BSA, produced acceptable hemato-
28 1
logic toxicosis in these cats. Poikilocytosis occurred, but was not associated with anemia or other red blood cell changes. Thrombocytopenia was uncommon and was mild when it occurred. Neutropenia was the most commonly observed hematologic toxicoses, but was associated with fever in only a minority of cats.
References I . Blum RH, Carter SK. Adriamycin, a new anticancer drug with significant clinical activity. Ann Intern Med 1974; 80:249-259. 2. Jeglum A, Whereat A. Chemotherapy of canine thyroid carcinoma. Comp Cont Ed 1983; 96:96-98. 3. Calvert CA, Leifer CE. Doxorubicin for treatment of canine lymphosarcoma after development of resistance to combination chemotherapy. J Am Vet Med Assoc 1981; 179:1011-1012. 4. Ogilvie GK, Reynolds HA, Richardson RC, et al. Phase I1 evaluation of doxorubicin for treatment of various canine neoplasms. J Am Vet Med Assoc 1989; 195:1580-1583. 5. Ogilvie GK, Richardson RC, Curtis CR, et al. Acute and shortterm toxicoses associated with the administration of doxorubicin to dogs with malignant tumors. J Am Vet Med Assoc 1989; 195:1584-1587. 6. Van Vleet JF, Ferrans VJ. Clinical observations, cutaneous lesions, and hematologic alterations in chronic Adriamycin intoxication in dogs with and without vitamin E and selenium supplementation. Am J Vet Res 1980; 41:691-699. 7. Bertazzoli C, Chieli T, Ferni G, et al. Chronic toxicity of Adriamycin: A new antineoplastic antibiotic. Toxicol Appl Pharmacol 1972; 21:287-301. 8. Jeglum KA, deGuzman E, Young KM. Chemotherapy of advanced mammary adenocarcinoma in 14 cats. J Am Vet Med ASSOC1985; 187:157-160. 9. Mauldin GN, Matus RE, Patnaik AK, et al. Efficacy and toxicity of doxorubicin and cyclophosphamide used in the treatment of selected malignant tumors in 23 cats. J Vet Intern Med 1988; 2160-65. 10. Henness AM, Theilen GH, Lewis JP. Clinical investigation of doxorubicin, daunomycin, and 6-thioguanine in normal cats. Am J Vet Res 1977; 38521-524. 1 1. Cotter SM, Kanki PJ, Simon M. Renal disease in five tumor-bearing cats treated with Adriamycin. J Am Anim Hosp Assoc 1985; 21:405-409. 12. Ogilvie GK, Krawiac DR, Gelberg HB, et al. Evaluation ofa shortterm saline diuresis protocol for the administration of cisplatin. Amer J Vet Res 1988; 49:1076-1078. 13. Ogilvie GK, Elmslie RE, Obradovich JE, et al. Efficacy and toxicoses associated with mitoxantrone therapy in the dog. Proceed 9th Ann Conf Vet Cancer SOC,October 15-17, 1989, pp 3-4. 14. Jain NC. Schalm’s Veterinary Hematology. 4th ed. Philadelphia, Lea & Febiger, 1986; 127. 15. Jain NC. Schalm’s Veterinary Hematology. 4th ed. Philadelphia, Lea & Febiger, 1986; 542. 16. Winer BJ. Statistical Principle in Experimental Design. 2nd ed. New York, McGraw-Hill, 1971; 514-599. 17. Milliken GA, Johnson DE. Analysis of Messy Data. I. Designed Experiments. Belmont, CA: Wadsworth, Inc., 1984; 359. 18. Winer BJ. Statistical Principle in Experimental Design. 2nd ed. New York, McGraw-Hill, 1971; 35-37. 19. Winer BJ. Statistical principle in experimental design. 2nd ed. New York, McGraw-Hill, 1971; 58-145. 20. Blum RH, Carter SK. Adriamycin. A new anticancer drug with significant clinical activity. Ann Intern Med 1974; 80:249-259. 2 1. Burrow JP, Favara BE, Brenman J, et al. Effect of Adriamycin on circulating erythrocytes. Proceed Am Assoc Cancer Res 1977; 18:6. 22. Badylak SF, Van Vleet JF, Herman EH, et al. Poikilocytosis in dogs with chronic doxorubicin toxicosis. Am J Vet Res 1985; 46505-508. 23. Henderson BM, Dougherty WJ, James VC, et al. Safety assess-
282
24. 25. 26. 27. 28.
O’KEEFE AND SCHAEFFER
ment of a new anticancer compound, mitoxantrone, in Beagle dogs: Comparison with doxorubicin. I. Clinical Observations. Cancer Treat Rep 1982; 66:1139-1143. Arancia G, Molinari A, Crateri P, et al. Adriamycin-plasma membrane interaction in human erythrocytes. Eur J Cell Biol 1988; 471379-387. Garnier-Suillerot A, Gattegno L. Interaction of adriamycin with human erythrocyte membranes. Role of the negativelycharged phospholipids. Biochem Biophys Acta 1988; 936:50-60. Henderson CA, Metz EN, Balcerzak SP, et al. Adriamycin and daunomycin generate reactive oxygen compounds in erythrocytes. Blood 1978; 52:878-885. Myers CE, Gianni L, Simone CB, et al. Oxidative destruction of erythrocyte ghost membranes catalyzed by the doxorubiciniron complex. Biochemistry 1982; 2 1 :1707- I7 13. Gonsalvez M, Van Rossum GDV, Blanco MF. Inhibition of sodium potassium activated adenosine-5-triphosphataseand ion transport by adriamycin. Cancer Res 1979; 39:257-261.
Journal of Veterinaly Internal Medicine
29. Sheetz MP, Singer SJ. On the mechanism of ATP-induced shape changes in human erythrocyte membranes. J Cell Biol 1977; 731638-646. 30. Jain NC. Studieson the occurrenceand persistence of heinz bodies in erythrocytes of the cat. Folia Haematol 1973; 99:28-38. 3 1. McDonald TP. Regulation of megakaryocytopoiesis by thrombopoietin. Ann NY Acad Sci 1987; 509:l-24. 32. Metcalf D, Johnson GR, Burgess AW. Direct stimulation by purified GM-CSF of the proliferation of multipotential and erythroid precursor cells. Blood 1980; 55: 138-142. 33 Handin RI. Hemorrhagic Disorders 11. Platelets and Purpura. In: Beck WAS, ed. Hematology, 3rd ed. Cambridge, MA: Massachusetts Institute of Technology, 1983:401-424. 34. Susaneck SJ. Doxorubicin therapy in the dog. J Am Vet Med Assoc 1983; 182170-72. 35. Braun DP, Hams JE. Cancer chemotherapy and its impact on the immune system. In: Hellmann K, Carter SK, eds. Fundamentals of Cancer Chemotherapy. New York McGraw-Hill, Inc., 1987; 77-100.
The hematologic toxicity of doxorubicin, 30 mg/m2 body surface area (BSA) every 21 days to a cumulative dose of 300 mg/m2, was evaluated in six cats. Complete blood and platelet counts were performed daily during the first treatment cycle. They were monitored before treatment for all remaining cycles, and at the average neutrophil nadir (day 8) starting with cycle 4. Significant poikilocytosis developed after the first treatment and remained throughout the study, although anemia did not occur. No other red blood cell abnormalities were seen. Platelet counts remained within the reference range throughout the first treatment cycle, but mild thrombocytopenia (88,000-288,000/uL) was found in 11.3%of subsequent complete blood counts (CBCs). Thrombocytosis was seen in 30.9%of CBCs. Neutropenia did not occur during the first treatment cycle although neutrophil counts did decrease, with the nadir occurring between days 8 and 11. All neutrophil counts returned to pretreatment values by day 14. Neutropenia was documented after 14 of 46 (30.4%)doxorubicin treatments, and was associated with fever in 5 cats (10.9%).All fevers responded to oral antibiotic therapy. Neutropenia that lasted more than 14 days developed in two cats, necessitating dosage reduction to 25 mg/m2 BSA. At the dose used in this study, doxorubicin administration was associated with acceptable hematologic toxicosis in most cats. (Journal of Veterinary Internal Medicine 1992; 6:276-282)
DOXORUBICIN IS an anthracycline antibiotic with proven antineoplastic activity against a broad range of tumors in both dogs and people.’-4 Its use can be associated with a variety of toxicoses in both species, including myelosuppression,enterocolitis,alopecia, and cardiomy~pathy.~-’Although doxorubicin has also been shown to have antineoplastic activity against a variety of feline tumor^,^,^ there are few reports of its toxicity in this Reported toxicities have included anorexia, myelosuppression, cardiac arrhythmias, and renal disease. Although myelosuppression is reported to occur, little information regarding the degree or timing of doxorubicin-induced cytopenias exists. This information would be valuable in the development and use ~
~
~~~~
~
From the Departments of Veterinary Clinical Medicine (OKeefe) and Veterinary Biosciences (Schaeffer), University of Illinois, Urbana, Illinois. Presented as an abstract at the American College of Veterinary Internal Medicine Eighth Annual Veterinary Medical Forum, May 10-13, 1990. The authors thank Adria Laboratories, Columbus, Ohio, for donating the doxorubicin used in this study. Accepted for publication February 15, I99 I . Reprint requests: Deborah A. OKeefe, DVM, Department of Veterinary Clinical Medicine, University of Illinois, 1008 W. Hazelwood Dr., Urbana, IL 6 180 I .
of doxorubicin-containing chemotherapy protocols in the cat. This study determined the acute and chronic hematologic toxicoses associated with doxorubicin when administered to cats at a dose used in veterinary practice.8-11Systemic toxicoses will be reported in a separate article. Materials and Methods Cats
Six healthy, young adult cats (5 female, 1 male) were obtained for this study. Baseline data evaluations included: complete blood count (hematocrit, red blood cell count, hemoglobin concentration, white blood cell and white blood cell differentialcount), platelet count, urinalysis, serum chemistry profile (urea nitrogen, creatinine, sodium, potassium, calcium, chloride, phosphorus, alanine transaminase, aspartate transaminase, alkaline phosphatase, glucose, bilirubin, total protein, albumin, cholesterol), FeLV, ELISA, and fecal examination for parasitic ova. All results were normal. All cats had percutaneous kidney biopsies done 2 weeks before the first doxorubicin treatment.
276
VOl. 6
. NO. 5,1992
277
HEMATOLOGIC TOXlCOSlS AND DOXORUBICIN ADMINISTRATION
Doxorubicin Administration Doxorubicin* was administered initially at a dose of 30 mg/m2 body surface area (BSA) every 21 days. The doxorubicin was diluted to 0.5 mg/mL and was given through a cephalic catheter over a 5-minute period. Due to prolonged neutropenia, dosage reduction to 25 mg/ m2 BSA was necessary in cats 3 and 5 starting with treatments 5 and 6, respectively. The cumulative doxorubicin dose for five cats (cats 1-4, 6) was 300 mg/m2 BSA. Cat number 5 died during the study after receiving 225 mg/m2 BSA.
Hematologic Monitoring Complete blood and platelet counts were monitored daily for the first 21-day cycle. These data were used to determine the average neutrophil nadir, as has been done for other chemotherapeutic agents.12J3Complete blood counts and platelet counts were monitored before treatment for all remaining cycles, and at the average neutrophil nadir (day 8) starting with cycle 4. Neutropenia was defined as an absolute neutrophil count of less than 250O/uL.l4 Thrombocytopenia and thrombocytosis were defined as platelet counts less than 300,OOO/uL and greater than 800,OOO/uL, respe~tive1y.l~Blood smears taken 2 1 days after each treatment and at the end of the study were stained with a modified Wright's stain? and evaluated for abnormalities of erythrocyte morphology, i.e., poikilocytosis. A poikilocyte was defined as any erythrocyte with abnormal morph~logy.'~ The slides were randomized and evaluated for the number of poikilocytes per 1000 red blood cells without knowledge of treatment or cat number.
Statistics After logarithmic transformation, the poikilocyte data were analyzed as a nested analysis of variance in a repeated measures design.16 Linear contrasts were constructed to test for differences between treatment times. The Greenhouse-Geiser-theta-statistic was used to test the correctness of the assumptions of the repeated measures analysis and to effect a small correction on the Fstat is ti^.'^ Because cat 5 was lost shortly after treatment 8, three analyses were carried out. The first used all six cats out to treatment 7; the second dropped cat 5 and analyzed the remaining cats out to treatment 7; the third analyzed the five cats out to the end of the study. Significant differenceswere present when P < 0.05 was determined. Neutrophil counts at the nadir of the first treatment cycle were compared with pretreatment values using the paired sample t-test." Linear regression was used to eval-
* Adriamycin, Adna Laboratories, Columbus, OH.
t HEMA-TEK, Miles Laboratories, Inc., Elkhart, IN.
uate the effect of repeated treatments on both the pretreatment and nadir neutrophil counts." Significant differences were considered present when P < 0.05 was determined.
Results
Red Blood Cells Table 1 summarizes the red blood cell parameters at the beginning and end of the study. There were no significant changes in packed cell volumes (PCVs) or red blood cell indices throughout the study. However, as shown in Table 2, the number of poikilocytes did increase in all cats except cat 1. Ovalocytes and echinocytes were the most frequently observed poikilocytes, although keratocytes, acanthocytes, and schistocytes were also seen (Figs. 1, 2). Poikilocytosis was not associated with anemia. The three statistical analyses of the poikilocyte data gave similar results. Out to treatment 7, with either 5 or 6 cats, the analyses showed a linear (increasing) trend in the number of poikilocytes per 1000 RBCs. This increase was statistically significant even after the first treatment. When data from the 5 cats were analyzed to the end of the study, polynomial contrasts included a quadratic term. These latter results indicate that the number of poikilocytes increased through treatment 7 and then became either asymptotic or began to slightly decrease. From plots of the data we favor an asymptote but cannot rule out the onset of a decline.
Platelets Platelet counts remained within the reference range (300-800,000/uL) throughout the first cycle of doxorubicin therapy. Mild thrombocytopenia (range 88,000288,OOO/uL, mean 192,80O/uL)was detected on 12 of 97 ( 1 1.3%) subsequent CBCs. Six of these were samples taken before a doxorubicin treatment (three before treatment 3, one before treatment 4, and two before treatment 5 ) and six were samples taken 8 days after treatment (one each during treatments 5 through 10). Cat number I was thrombocytopenic before treatment 5 , but had a normal platelet count 8 days after this treatment. Thrombocytosis (range 8 17,000-1,300,000/uL, mean
TABLEI . Red Blood Cell Parameters in Doxorubicin-treated Cats ~~
PCV 70 RBC X106 Hb (g) MCV fl. MCHC 70
Prior to Treatment 1 Mean (Range)
Post Treatment 10 Mean (Range)
Reference RangeI4
32. I (27-36.4) 5.93 (5.01-6.74) 10.4 (9.2-12.2) 54.0 (52.0-56.0) 32.6 (30.0-36.0)
27.6 (24-32) 5.31 (4.46-6.26) 9.2 (8.1-10.5) 52.3 (48.5-56.0) 33.2 (32.7-34.0)
24.0-45.0 5.0-10.0 8.0-15.0 39.0-55.0 3 1.0-35.0
278
Journal of Veterinary Internal Medicine
O’KEEFE AND SCHAEFFER
TABLE 2. Poikilocytosis in Doxorubicin-treated Cats (Poikilocytes per 1000 RBCs) Treatment Number Cat No.
0
I
2
3
4
5
6
7
8
9
10
1
2 3 4 5 6
12 20 14 5 10 6
26 32 53 22 37 39
6 I25 47 9 22 23
41 57 I34 68 32 34
13 I94 I69 21 95 Ill
8 374 I09 71 161 I22
16 284 I59 25 I 63 I20
23 224 232 380 26 I 170
31 265 235 215 ND I22
33 159 247 307 ND 137
22 I 252 158 ND 143
Mean
I1
35*
39
IOI*
141*
I49*
215*
174*
177*
157*
61*
I1
* Statisticallydifferent from pretreatment (treatment 0 ) value, P < 0.05. ND, no data points due to cat’s death.
999,OOO/uL) was seen in 30 of 97 (30.9%)CBCs, with 19 cases occurring before a treatment and 11 cases occurring 8 days after a treatment.
The neutrophil was the only white blood cell affected by doxorubicin administration. The average daily neutrophi1 counts during the first treatment cycle are shown in Figure 3. During this cycle, doxorubicin caused a significant decrease in the neutrophil count with the average nadir occurring on day 8. Neutrophil nadirs in individual cats ranged from 2640-7 106 neutrophils/uL (mean 4095/uL, median 363 l/uL) and occurred on days 8-1 1.
Neutrophil counts in all cats returned to pretreatment values by day 14. A small increase in band neutrophils (300- 1200/uL) was noted in five of six cats between days 13 and 18. Figure 4 shows the average neutrophil counts before and at the average neutrophil nadir of each doxorubicin treatment. The pretreatment and nadir average neutrophil counts did not change significantly as treatment progressed. Although the average nadir neutrophil counts did not decrease as treatments progressed, neutropenia did occur in individual cats. Neutropenia was documented after 14 of 46 treatments (30.4%) and all neutropenic episodes occurred in cats 3-6 (Table 3). Fever occurred during five neutropenic episodes, which represented
FIG. I . Peripheral blood smear from cat 3 after a cumulative doxorubicin dose of 245 mg/mz BSA. Ovalocytes(closed arrows) and ovaloechinocytes (open arrows) are present. Two blister cells, which are vacuolated red blood cells in the process of keratocyte formation, are also present (arrowheads) (original magnification X 160, modified Wright’s stain).
FIG.2. Peripheral blood smear from cat 4 after a cumulative doxorubicin dose of 2 10 mg/m2 BSA. Ovalocytes (closed arrow) and ovaloechinocytes (open arrow) can be seen. One blister cell (small arrowhead) and one keratocyte (large arrowhead)are present, as are other unclassified poikilocytes (original magnification X 160, modified Wright’s stain).
White Blood Cells
Vol. 6
. NO. 5, 1992
279
HEMATOLOGIC TOXICOSIS AND DOXORUBICIN ADMINISTRATION
25
T
35
I
t
T
1
4
5
6
7
8
9
1
0
Treatment Number FIG.3. Average (+/- SD) daily neutrophil counts before and after the initial administration of doxorubicin, 30 mg/m2 BSA IV (n = 6). The neutrophilcount on day 1 was taken before treatment. The day 8 value is significantly different than the day 1 value, P = 0.029.
10.9%of treatments given. All fevers were responsive to oral amoxicillinS or cefadroxils therapy. Fevers developed during four additional treatments but unfortunately CBCs were not done to determine if neutropenia was present. These fevers lasted for at least 2 days, ranged from 103.2- 104.7"F, and were responsive to amoxicillin therapy. They occurred on days 6 and 8 of treatment 2, and days 7 and 9 of treatment 3, which are times compatible with the neutrophil nadir. If these fevers represented neutropenic episodes, then febrile neutropenia occurred in 9 of 58 (15.5%)treatments. Cats 3 and 5 had prolonged neutropenia after treatments 4 and 5 , respectively. Cat 3 had neutrophil counts of 1 152/uL and 2074/uL on days 8 and 2 1, respectively, of treatment cycle 4. Cat 5 had neutrophil counts of 164/ UL and 700/uL on days 8 and 21 of treatment cycle 5. This necessitated discontinuing treatment for one treatment cycle in each cat. Subsequent doxorubicin doses were decreased to 25 mg/m2 BSA in both cats; these doses eliminated the prolonged neutropenia. However, neutropenia still developed after 3 of 8 subsequent treatments. Cat number 5 was found dead in her cage 3 days after the eighth doxorubicin treatment. Her neutrophil count before that treatment had been 14,552/uL with 1155 bands/uL. At necropsy, a 0.5-cm abscess was found in the subcutis overlying the left abdominal wall. It did not appear to communicate with the abdominal cavity. However, 20-30 mL of sanguinous, purulent fluid was found
$ Amoxi-tabs, Beecham Laboratories,Bristol, TN. Cefa-tabs, Aveco Co., Inc., Fort Dodge, IA.
FIG.4. Average (+/- SD) neutrophil counts in cats treated with doxorubicin, 30 mg/m2 BSA every 2 1 days (n = 6). Solid squares are average pretreatment values, open squares are average nadir values.
in the abdominal cavity. Cytology of this fluid was consistent with septic peritonitis and a group G Streptococcus species was cultured.
Discussion Doxorubicin administration has been associated with dose-related decreases in hematocrit, hemoglobin, and red blood cell numbers in dogs, rabbits, and p e ~ p l e . ~ , ~ " - * ~ When dogs were given doses similar to those used in this study, hematocrits and hemoglobin concentrations were transiently decreased 7 days after treatment but returned to pretreatment values before the next dose.23Higher weekly doses have produced sustained regenerative and
TABLE 3. Neutropenic Episodes in Doxorubicin-treated Cats
Cat Number
Treatment Number ~
3 3 3 4 4 4 4 5 5 5 6 6 6 6
Day 8 Neutrophil Count (Cells/uL)
Fever*
~
4
8t 9t 7 8 9 10 4 5 6t 5 8 9 10
1152 1536 2444 1058 2494 264 I972 I248 164 2278 I764 1505
736 2379
* Temperature > 102.5"F for at least 2 consecutive days.
t Dose reduced to 25 mg/m2 BSA.
102.7
-
103.4 103.8 103.3 103.8 -
-
280
O'KEEFE AND SCHAEFFER
nonregenerative anemias in dog^.^,^' The nonregenerative anemia results from the toxic effect of doxorubicin on bone marrow erythrocyte precursors,' while the regenerative anemia may be related to increased poikilocyte destruction." In contrast to the species mentioned earlier, alterations in the hematocrit and other red blood cell parameters were not seen in these cats. The reason for this is not clear, although feline erythrocyte precursors may be more resistant to the toxic effects of doxorubicin. Decreased clearance of poikilocytes may also contribute. Poikilocytosis developed in these cats and has been reported in both dogs and people after doxorubicin therapy.21*22 The exact mechanism by which doxorubicin produces alterations in red blood cell morphology is not known. Doxorubicin is known to bind to cell membranes and can affect them in a variety of way^.'^'^ Possible mechanisms for the development of poikilocytosis include the generation of reactive oxygen compounds resulting in membrane lipid p e r o x i d a t i ~ n ,inhibition ~~,~~ of Na+/K+ dependent adenosine triphosphatase activity," and alteration in calcium tran~port.'~ The number of poikilocytes in these cats increased in a linear manner through the first seven treatments indicating that initially the effects of each treatment were additive. After treatment 7, the number of poikilocytes reached a plateau, and possibly even declined. The erythrocytes may have developed protective mechanisms that made them less susceptible to the membrane effects of doxorubicin. Significant increases in poikilocyte number were seen in these cats after one doxorubicin treatment, whereas dogs did not develop poikilocytosis until 16 weeks after the initiation of weekly treatments. The reason for this difference is uncertain. Canine erythrocytes may be more resistant to doxorubicin; abnormal erythrocytes may be more readily removed by the mononuclear phagocytic system in the dog; and/or erythrocyte precursors, rather than circulating red blood cells, may be the cells most affected by doxorubicin in the dog. The magnitude of poikilocytosis was also much greater in these cats compared with that reported in dogs. The mean number of poikilocytes in dogs before treatment and after 16 weeks of treatment was 0.4 and 3.4 per 1000 RBCs, respectively.'* The mean number of poikilocytes in these cats at comparable times was 1 1 and 141 per 1000 RBCs, respectively. The number of poikilocytes produced by the administration of doxorubicin to dogs may be underestimated on a peripheral blood smear because of the rapid removal of these abnormal erythrocytes by the mononuclear phagocytic system. This rapid removal may then contribute to the anemia seen with doxorubicin therapy in this species.22The mononuclear phagocyte system of the cat may not be as efficient at removing abnormally shaped erythrocytes as
Journal of Veterinary Internal Medicine
it is in the dog. The increased number of poikilocytes in the cat compared with that in the dog even before receiving doxorubicin would support this theory as would the observation that the spleen does not appear to remove Heinz bodies as readily in cats as it does in other spec i e ~If. this ~ ~ is the case, decreased poikilocyte removal may explain, at least in part, the lack of anemia in these cats. Thrombocytopenia does not appear to be a major toxicity when doxorubicin is administered to cats at 30 mg/ m2 BSA. Thrombocytopenia was not detected when cats were treated with two doses of 10-40 mg/m2," and platelet counts did not decrease during the first treatment cycle of the present study. The episodes of thrombocytopenia during the remainder of this study were infrequent and mild. Platelet counts never decreased to a level that would be associated with clinical bleeding. Because hematologic toxicosis is dose-related, larger doxorubicin doses could result in significant thrombocytopenia. Thrombocytosis was more common in these cats than thrombocytopenia and occurred more frequently before therapy than 8 days after treatment. This may reflect a rebound thrombocytosis due to bone marrow stimulation after thrombocytopenic or neutropenic episodes. Thrombocytopenia causes the release of thrombopoietin, which stimulates platelet production in the bone marr~w.~ Granulocyte-macrophage ' colony-stimulating factor (GM-CSF), a glycoprotein released in response to neutropenia, is also known to stimulate cells ofthe megakaryocytic line.32Splenic contraction during blood sampling may also have contributed to the increased platelet counts. Thrombocytosis of the degree seen in this study is probably clinically unimportant, although platelet counts in excess of 1,OOO,OOO/uL have been associated with hemorrhage and thrombosis in people.33 At the doxorubicin dose used in this study, a decreased neutrophil count was the most consistent hematologic abnormality seen in the cat. A neutrophil nadir occurring at 8-1 1 days post-treatment is similar to what is reported in dogs. However, dogs given a slightly larger doxorubicin dose did not develop a significantly smaller number of WBC until the nadir of the third treatment cycle,23whereas the cats in this study had smaller numbers of neutrophils after the first treatment. Also, neutrophi1 counts in the dog are reported to return to normal within 17 to 20 days, which is somewhat longer than the 14-day recovery seen during the first treatment cycle in these cats.34 The incidence of neutropenia (30%)was fairly common in these cats. Also, the incidence of febrile neutropenia was greater than that reported in dogs (10.9% vs. 3%).5Cats may be more susceptible to the toxic effects of doxorubicin because of their small size. In dogs doxorubicin toxicity is inversely related to body eight.^ The apparent increase in febrile neutropenic episodes in cats
Vol. 6
. NO.5,1992
HEMATOLOGIC TOXlCOSlS AND DOXORUBICIN ADMINISTRATION
may actually be the result of increased monitoring. Temperatures were taken daily and cats were sometimes febrile with no change in behavior or attitude. The dogs studied, however, were clinical patients that were not monitored as closely and could have had undetected episodes of fever. Regardless, the frequency of febrile neutropenia seen in these cats would be clinically acceptable, and all cats responded well to oral antibiotic therapy. Two cats remained neutropenic on the day of the next scheduled treatment, which necessitated skipping that treatment. Because of this, the neutrophil count in clinical patients both before therapy and at the neutrophil nadir should be monitored. If prolonged neutropenia should occur, the dosage of subsequent doxorubicin treatments could be lowered to 25 mg/m2 BSA which alleviated the prolonged neutropenia. At 30 mg/m2 BSA, the toxic effect of doxorubicin on the bone marrow of cats is not cumulative and does not overwhelm the ability of bone marrow to regenerate. The average pretreatment neutrophil counts and the average nadir neutrophil counts did not decline throughout the study. Also, neutropenic episodes in individual cats did not increase as treatment progressed. However, there was great variability in the degree of neutropenia in individual cats from cycle to cycle. This makes it imperative to continue hematologic monitoring in clinical patients throughout therapy. There appeared to be variability in the susceptibilityof the cats of this study to doxorubicin-induced hematologic toxicosis. One cat did not develop a significant poikilocytosis; all neutropenic episodes occurred in four of six cats; and only two cats developed prolonged neutropenia. The reason for this variability was not addressed. There may be an inherent difference in the sensitivity of target cells to doxorubicin or it could reflect differences in drug metabolism. Cat 5 died of septic peritonitis 3 days after treatment 8. The peritonitis most likely resulted from extension of a subcutaneous abscess into the left abdominal area. The cause of this abscess was not determined. It is unlikely that the percutaneous kidney biopsy done in this area at the beginning of the study was responsible because the cat had experienced no problems for the previous 20 weeks of the study. It is also unlikely that myelosuppression contributed to the development of peritonitis because the cat had not been neutropenic since the nadir of treatment 6 and actually had a slightly increased neutrophi1 count 3 days before death. Immune function was not evaluated in these cats, but doxorubicin is known to be immunosuppressive in people.35Decreased immune function may have contributed to the development of peritonitis from a superficial abscess in this cat. In summary, doxorubicin, when used as a single agent at a dose of 30 mg/m2BSA, produced acceptable hemato-
28 1
logic toxicosis in these cats. Poikilocytosis occurred, but was not associated with anemia or other red blood cell changes. Thrombocytopenia was uncommon and was mild when it occurred. Neutropenia was the most commonly observed hematologic toxicoses, but was associated with fever in only a minority of cats.
References I . Blum RH, Carter SK. Adriamycin, a new anticancer drug with significant clinical activity. Ann Intern Med 1974; 80:249-259. 2. Jeglum A, Whereat A. Chemotherapy of canine thyroid carcinoma. Comp Cont Ed 1983; 96:96-98. 3. Calvert CA, Leifer CE. Doxorubicin for treatment of canine lymphosarcoma after development of resistance to combination chemotherapy. J Am Vet Med Assoc 1981; 179:1011-1012. 4. Ogilvie GK, Reynolds HA, Richardson RC, et al. Phase I1 evaluation of doxorubicin for treatment of various canine neoplasms. J Am Vet Med Assoc 1989; 195:1580-1583. 5. Ogilvie GK, Richardson RC, Curtis CR, et al. Acute and shortterm toxicoses associated with the administration of doxorubicin to dogs with malignant tumors. J Am Vet Med Assoc 1989; 195:1584-1587. 6. Van Vleet JF, Ferrans VJ. Clinical observations, cutaneous lesions, and hematologic alterations in chronic Adriamycin intoxication in dogs with and without vitamin E and selenium supplementation. Am J Vet Res 1980; 41:691-699. 7. Bertazzoli C, Chieli T, Ferni G, et al. Chronic toxicity of Adriamycin: A new antineoplastic antibiotic. Toxicol Appl Pharmacol 1972; 21:287-301. 8. Jeglum KA, deGuzman E, Young KM. Chemotherapy of advanced mammary adenocarcinoma in 14 cats. J Am Vet Med ASSOC1985; 187:157-160. 9. Mauldin GN, Matus RE, Patnaik AK, et al. Efficacy and toxicity of doxorubicin and cyclophosphamide used in the treatment of selected malignant tumors in 23 cats. J Vet Intern Med 1988; 2160-65. 10. Henness AM, Theilen GH, Lewis JP. Clinical investigation of doxorubicin, daunomycin, and 6-thioguanine in normal cats. Am J Vet Res 1977; 38521-524. 1 1. Cotter SM, Kanki PJ, Simon M. Renal disease in five tumor-bearing cats treated with Adriamycin. J Am Anim Hosp Assoc 1985; 21:405-409. 12. Ogilvie GK, Krawiac DR, Gelberg HB, et al. Evaluation ofa shortterm saline diuresis protocol for the administration of cisplatin. Amer J Vet Res 1988; 49:1076-1078. 13. Ogilvie GK, Elmslie RE, Obradovich JE, et al. Efficacy and toxicoses associated with mitoxantrone therapy in the dog. Proceed 9th Ann Conf Vet Cancer SOC,October 15-17, 1989, pp 3-4. 14. Jain NC. Schalm’s Veterinary Hematology. 4th ed. Philadelphia, Lea & Febiger, 1986; 127. 15. Jain NC. Schalm’s Veterinary Hematology. 4th ed. Philadelphia, Lea & Febiger, 1986; 542. 16. Winer BJ. Statistical Principle in Experimental Design. 2nd ed. New York, McGraw-Hill, 1971; 514-599. 17. Milliken GA, Johnson DE. Analysis of Messy Data. I. Designed Experiments. Belmont, CA: Wadsworth, Inc., 1984; 359. 18. Winer BJ. Statistical Principle in Experimental Design. 2nd ed. New York, McGraw-Hill, 1971; 35-37. 19. Winer BJ. Statistical principle in experimental design. 2nd ed. New York, McGraw-Hill, 1971; 58-145. 20. Blum RH, Carter SK. Adriamycin. A new anticancer drug with significant clinical activity. Ann Intern Med 1974; 80:249-259. 2 1. Burrow JP, Favara BE, Brenman J, et al. Effect of Adriamycin on circulating erythrocytes. Proceed Am Assoc Cancer Res 1977; 18:6. 22. Badylak SF, Van Vleet JF, Herman EH, et al. Poikilocytosis in dogs with chronic doxorubicin toxicosis. Am J Vet Res 1985; 46505-508. 23. Henderson BM, Dougherty WJ, James VC, et al. Safety assess-
282
24. 25. 26. 27. 28.
O’KEEFE AND SCHAEFFER
ment of a new anticancer compound, mitoxantrone, in Beagle dogs: Comparison with doxorubicin. I. Clinical Observations. Cancer Treat Rep 1982; 66:1139-1143. Arancia G, Molinari A, Crateri P, et al. Adriamycin-plasma membrane interaction in human erythrocytes. Eur J Cell Biol 1988; 471379-387. Garnier-Suillerot A, Gattegno L. Interaction of adriamycin with human erythrocyte membranes. Role of the negativelycharged phospholipids. Biochem Biophys Acta 1988; 936:50-60. Henderson CA, Metz EN, Balcerzak SP, et al. Adriamycin and daunomycin generate reactive oxygen compounds in erythrocytes. Blood 1978; 52:878-885. Myers CE, Gianni L, Simone CB, et al. Oxidative destruction of erythrocyte ghost membranes catalyzed by the doxorubiciniron complex. Biochemistry 1982; 2 1 :1707- I7 13. Gonsalvez M, Van Rossum GDV, Blanco MF. Inhibition of sodium potassium activated adenosine-5-triphosphataseand ion transport by adriamycin. Cancer Res 1979; 39:257-261.
Journal of Veterinaly Internal Medicine
29. Sheetz MP, Singer SJ. On the mechanism of ATP-induced shape changes in human erythrocyte membranes. J Cell Biol 1977; 731638-646. 30. Jain NC. Studieson the occurrenceand persistence of heinz bodies in erythrocytes of the cat. Folia Haematol 1973; 99:28-38. 3 1. McDonald TP. Regulation of megakaryocytopoiesis by thrombopoietin. Ann NY Acad Sci 1987; 509:l-24. 32. Metcalf D, Johnson GR, Burgess AW. Direct stimulation by purified GM-CSF of the proliferation of multipotential and erythroid precursor cells. Blood 1980; 55: 138-142. 33 Handin RI. Hemorrhagic Disorders 11. Platelets and Purpura. In: Beck WAS, ed. Hematology, 3rd ed. Cambridge, MA: Massachusetts Institute of Technology, 1983:401-424. 34. Susaneck SJ. Doxorubicin therapy in the dog. J Am Vet Med Assoc 1983; 182170-72. 35. Braun DP, Hams JE. Cancer chemotherapy and its impact on the immune system. In: Hellmann K, Carter SK, eds. Fundamentals of Cancer Chemotherapy. New York McGraw-Hill, Inc., 1987; 77-100.
