Animals

Client-owned cats were used in the study. Cats were considered eligible to receive ifosfamide when they had a resected, recurrent, or metastatic sarcoma that had been confirmed histologically; an expected survival of 4 weeks without treatment; not received chemotherapy, immunotherapy, or radiation therapy for at least 4 weeks; and adequate bone marrow, cardiac, renal, and hepatic function. The study was performed at the Sprecher Institute for Comparative Cancer Research at Cornell University College of Veterinary Medicine and the Harrington Oncology Program at Tufts University School of Veterinary Medicine. The study protocol was approved by an institutional animal care and use committee at each institution, and written informed consent was obtained from all clients.

Initial evaluation

Baseline evaluation included physical examination, a CBC count with differential and platelet counts, serum biochemical analysis, and urinalysis. Gross tumors were not required, but when appropriate, tumors were measured directly with calipers and tumor volume calculated as the number of cubic millimeters by use of the following equation: length × width × height × (π/6). The CBC and platelet counts were obtained at weekly intervals until day 21 after ifosfamide administration or until recovery of neutrophil and platelet counts, whichever was longer. Serum biochemical analysis and urinalysis were repeated on day 21 after ifosfamide administration. Evidence of toxic effects of ifosfamide was monitored by evaluation of the medical history obtained from owners and results of physical examination and laboratory data. Toxic effects were graded in accordance with predetermined criteria (Appendix).^10,11 The DLT was defined as grade 2 renal toxicosis or any grade 4 hematologic, gastrointestinal, or urothelial toxicosis that developed during the first cycle of ifosfamide treatment.

Dose escalation and ifosfamide administration

The study was conducted as an open-label phase I, dose-escalation study. Each cat was administered a single dose of ifosfamide, IV. The starting dosage of ifosfamide was 400 mg/m² of body surface area. Initially, dose escalation was planned for increments of 50 mg/m². However, when no toxic effects were observed after administration of 600 mg/m², additional dose escalation used increments of 100 mg/m².

Each dosage was administered to 3 cats, provided that none of the treated cats had DLT. When 1 of the 3 cats in a group had DLT, 3 additional cats were also administered ifosfamide at that same dosage. When no DLT was observed in the additional 3 cats, the dosage was escalated. When 2 or more cats in a group had DLT, at least 3 additional cats were administered the preceding treatment dosage. The MTD was defined as the dosage at which ≥ 2 of 3 or ≥ 2 of 6 cats had DLT during their first cycle of treatment. The dosage recommended for phase II trials of ifosfamide in cats was defined as the highest dosage below the MTD that resulted in ≤ 1 of 6 cats with DLT.

The amount of ifosfamide^a was reconstituted in saline (0.9% NaCl) solution at a volume of 9.15 mL/kg. Mesna^b for urothelial protection was administered at a dosage equal to 20% of the calculated ifosfamide dosage and was reconstituted in saline solution at a concentration of 20 mg/mL. The drugs were administered, IV, through an indwelling catheter inserted in a cephalic vein; saline solution was administered for diuresis in conjunction with drug administration. Specifically, cats received a bolus of mesna, IV, followed by diuresis with saline solution (18.3 mL/kg/h, IV, for 30 minutes); the calculated dose of ifosfamide was then infused during a 30-minute period. Diuresis with saline solution was continued for another 5 hours after completion of ifosfamide administration, with additional doses of mesna injected 2 and 5 hours after completion of the ifosfamide administration. This procedure was designed on the basis of results that our laboratory group obtained in a study¹⁰ on the use of ifosfamide in dogs.

Pharmacokinetic analysis

After the MTD of ifosfamide in cats was reached and the dosage for evaluation during phase II studies was determined, samples for pharmacokinetic analysis were obtained from 9 cats. An indwelling catheter was inserted in the jugular vein of each cat, and blood samples (1 mL) were collected into heparinized tubes before and 20, 25, 30, 45, 60, 120, and 240 minutes after start of the ifosfamide infusion. Samples were immediately centrifuged, and plasma was harvested and frozen at −70°C until analyzed.

The assay to determine ifosfamide concentrations in feline plasma was a modification of the assay described elsewhere.¹² Briefly, 100 μL of feline plasma was spiked with 10 μL of a cyclophosphamide solution (1 mg/mL) and extracted by use of columns containing cyclohexyl^c (100 mg). Loaded columns were initially washed with a solution (10:90) of 1 mL of acetonitrile:10mM KH₂PO₄ (pH, 4). Column beds were then eluted with a solution (40:60) of 500 μL of acetonitrile:25mM KH₂PO₄ (pH,4), and eluents were centrifuged (13,000 × g for 1 minute) to remove any particulates. The high-performance liquid chromatography system consisted of an autosampler,^d dual pumps,^e a 250 × 4.6-mm C8 (5-μm) column,^f and a 30 × 4.6-mm RP-8 (5-μm) guard.^g A UV detector^h and chromatography workstation softwareⁱ run on a specified platform^j with a commercially available program^k were used to store the chromatograms. The mobile phase was isocratic and consisted of a solution (23:77) of acetonitrile:25mM KH₂PO₄ (pH, 4). Flow rate was 1 mL/min at 20°C. The UV detection was set at 200 nm. Eluent samples (150 μL) were introduced by use of the autosampler. Retention time for ifosfamide was 16.5 minutes and for cyclophosphamide (internal standard) was 18.4 minutes. The standard curve was linear for the range from 10 to 100 μg of ifosfamide/mL. This ifosfamide assay had a limit of detection of 10 μg/mL. Interday and intraday coefficients of variation for the range of the standard curve varied from 8.4% to 12.9% and from 1.9% to 7.5%, respectively. Accuracy of the assay ranged from 94.6% to 105.6%.

The plasma ifosfamide concentrations for each cat after a single IV administration of ifosfamide infused during a 30-minute period were inspected on a semilogarithmic plot of ifosfamide concentration versus time. Values for C_max and the time at which C_max was achieved were determined by visual examination of raw data for the plasma ifosfamide concentration–time plot. Pharmacokinetic variables were estimated by use of a computer software program.^l The ifosfamide concentration–time data were analyzed by use of an open noncompartmental method with a constant infusion by use of a computerized program.^m The value for k_e was determined by use of linear regression of the terminal 3 to 6 points of the logarithmic plasma ifosfamide concentration-versus-time plot with a weighted paradigm of 1/Y², where Y is the plasma ifosfamide concentration. The value for t_1/2 was estimated as 0.693/k_e. The AUC to the last data point was estimated by use of the linear-logarithmic trapezoidal rule, and AUC extrapolated to infinity was determined by adding the Wagner-Nelson correction (ie, ifosfamide concentration at the last data point/k_e). The value for CL was calculated as dose/AUC extrapolated to infinity. The apparent volume of distribution was estimated as CL/k_e. The MRT was estimated as AUMC/AUC. The Vd_ss was estimated as CL × MRT.

Statistical analysis

A 1-tailed Spearman rank correlation test was used to examine whether ifosfamide-associated neutropenia or gastrointestinal toxicoses increased in severity with the ifosfamide AUC or C_max For all analyses, calculations were performed by use of a computer software programⁿ and values of P ≤ 0.05 were considered significant.

Animals

Between February 2000 and September 2002, 38 cats with sarcomas were entered into the phase I trial and received ifosfamide at 9 dosages. There were 21 males and 17 females. Cats ranged from 2 to 16 years of age (mean, 10 years; median, 10 years) and weighed between 3.2 and 13.3 kg (mean, 5.3 kg; median, 5.6 kg). The primary tumor was located at a site potentially used for vaccination (ie, neck or interscapular area, thigh, gluteal region, dorsal lumbar area, flank, and dorsolateral thoracic area) in 30 cats, antebrachium in 4 cats, oral cavity in 3 cats, and face in 1 cat.

Twenty-eight of 38 (74%) cats had fibrosarcomas, 3 (8%) had undifferentiated sarcomas, 2 (5%) had spindle cell sarcomas, and 1 (3%) each had a nerve sheath tumor, osteosarcoma, chondrosarcoma, myxosarcoma, and giant cell sarcoma. Three tumors originated in the oral cavity (2 fibrosarcomas and 1 osteosarcoma), whereas the remaining 35 tumors were soft tissue sarcomas. Duration of disease before treatment with ifosfamide ranged from 12 to 1,210 days (mean, 248 days; median, 120 days). Seventeen (45%) cats had no prior treatment before administration of ifosfamide, 12 (32%) had been treated by use of surgery, 4 (10%) had received radiation therapy, 3 (8%) had received chemotherapy combined with radiation therapy, and 2 (5%) had received chemotherapy alone. Twenty-seven (71%) cats had macroscopic tumors at the time of treatment with ifosfamide; tumor volume in these cats ranged from 1 to 461 cm³ (mean, 40 cm³; median, 14 cm³).

Determination of MTD and DLT

All cats were used for assessment of toxic effects evident after 1 ifosfamide treatment. Dosages of ifosfamide were escalated for each successive group of cats. Myelosuppression was the principal toxic effect (Table 1). At a dosage of 1,000 mg/m², 2 of 5 cats had DLT, which met the MTD criterion of ≥ 2 of 6 affected cats at a dosage. Thus, we concluded that the MTD was achieved.

The dosage recommended for phase II trials of ifosfamide in cats was defined as the highest dose below the MTD that resulted in ≤ 1 of 6 cats with DLT. Once we determined that 1,000 mg/m² was the MTD, 9 cats were then administered ifosfamide at a dosage of 900 mg/m² to further evaluate toxic effects and define the pharmacokinetic pattern. Two of these 9 cats also had DLT in the form of neutropenia (grade 4). On the basis of these results, 900 mg/m² was determined to be the dosage of ifosfamide for use in future phase II trials in cats.

Hematologic toxicoses

Myelosuppression was the DLT (Table 2). The toxic effects recorded represent the maximum grade of toxicosis observed for a specific cat during the first cycle of treatment with ifosfamide. Seven of 8 episodes of neutropenia (grades 2 to 4) were detected on day 7 after treatment. One cat developed a fever and grade 4 neutropenia on day 5 after treatment. This cat had received ifosfamide at a dosage of 1,000 mg/m², but it recovered after receiving supportive treatment consisting of IV administration of fluids and antimicrobials. Six cats treated with ifosfamide had unresolved neutropenia on day 21 after treatment (4 had grade 1, and 2 had grade 2) but had neutrophil counts within the reference range when rechecked on day 28 after treatment. Five of 6 episodes of thrombocytopenia (grades 2 to 4) were detected on day 7 after treatment, and 1 episode of grade 2 thrombocytopenia was detected on day 14 after administration of ifosfamide at a dosage of 550 mg/m². All episodes of thrombocytopenia resolved by day 21 after treatment.

Nonhematologic toxicoses

Nonhematologic toxic effects were observed after treatment (Table 3). The most common nonhematologic toxic effect was signs of nausea during the ifosfamide infusion (4 cats salivated during the 30-minute infusion of ifosfamide). This clinical sign resolved after the infusion was completed. Other adverse gastrointestinal effects (eg, loss of appetite, vomiting, and diarrhea) were mild and selflimiting. None of the cats developed clinical signs of hemorrhagic cystitis, and there was no evidence of microscopic hematuria found in any of the urine sediments analyzed; however, urinalyses were only performed on day 21 after treatment. Serum biochemical analysis was repeated on day 21 after treatment in all cats, and no renal or hepatic dysfunction was observed. One cat that received ifosfamide at a dosage of 1,000 mg/m² developed signs consistent with a hypersensitivity reaction (severe facial swelling and pruritus). These signs were evident immediately after completion of the ifosfamide infusion and resolved within 1 hour after IV administration of dexamethasone and IM administration of diphenhydramine.

Pharmacokinetics

Noncompartmental pharmacokinetic variables for ifosfamide were calculated for 9 cats administered ifosfamide at a dosage of 900 mg/m² (Table 4). The logarithmic plasma ifosfamide concentration-versus-time curve was plotted (Figure 2).

View Full Size

Graph of the plasma ifosfamide concentration-versustime curve derived from samples obtained from 9 cats after IV administration of a single dose of ifosfamide (900 mg/m² of body surface area) infused during a 30-minute period. Time 0 was the start of the ifosfamide infusion.

Citation: American Journal of Veterinary Research 67, 3; 10.2460/ajvr.67.3.510

• Download Figure
• Download figure as PowerPoint slide

There were no significant relationships between the pertinent pharmacokinetic variables reflecting drug exposure (C_max and AUC) and the principal toxic effects (neutropenia and gastrointestinal toxicosis).

Antitumor activity

For inclusion in the study reported here, cats did not need to have tumors of measurable volume, and assessment of tumor response to ifosfamide treatment was not the primary goal of our study. Nonetheless, 27 cats had measurable tumors, and 6 cats had partial responses characterized by a reduction of ≥ 50% in tumor burden. Four of the 6 cats that achieved a partial response received ifosfamide at a dosage of 900 mg/m², 1 received ifosfamide at 1,000 mg/m², and 1 received ifosfamide at 550 mg/m².

Of the cats that achieved a partial remission, 5 had vaccine-associated sarcomas and 1 had an undifferentiated soft tissue sarcoma located on the head.