A Phase II study of paclitaxel by 24-hour infusion and ifosfamide in anthracycline-resistant metastatic breast carcinoma




A Phase II study was performed to investigate the efficacy and tolerability of paclitaxel and ifosfamide chemotherapy for the treatment of anthracycline-resistant metastatic breast carcinoma (MBC).


Recurrent or progressed MBC within 12 months after anthracycline-based chemotherapy was defined as anthracycline-resistant. A 24-hour infusion of paclitaxel (175 mg/m2) on Day 1 and subsequent infusions of ifosfamide (1.8 g/m2/day) with mesna (360 mg/m2/day) on Days 2– 4, were performed every 3 weeks. Twenty-one patients were eligible for toxicity analysis. Response rate and survival duration were evaluated in 21 patients. Frontline chemotherapy was the FAC (5-fluorouracil, doxorubicin, cyclophosphamide) regimen in all patients.


Objective response was found in 9 patients (42.9%), including complete response in 3 (13.4%). Median response duration and median survival duration were 10 months (range, 2–24+) and 19+ months (range, 2–32+), respectively. Sixteen (76%) experienced Grade 3/4 leukopenia controllable with granulocyte macrophage colony-stimulating factor. Other significant toxicities were peripheral neuropathy (n = 3), mucositis (n = 2), and liver dysfunction (n = 1). However, there was no chemotherapy-related death.


Paclitaxel by 24-hour infusion combined with ifosfamide is efficacious in the treatment of anthracycline-resistant MBC with tolerable toxicity. Further trials verifying the result of the authors' study are warranted. Cancer 2002;94:1925–30. © 2002 American Cancer Society.

DOI 10.1002/cncr.10431

The main goal of management in anthracycline-resistant metastatic breast carcinoma (MBC) had been palliation until comparatively lately. New drugs have achieved meaningful response rates without cross-resistance to anthracyclines, and autologous peripheral blood stem cell transplantation (PBSCT) has allowed the possibility of overcoming irreversible bone marrow suppression, an important dose-limiting toxicity. Autologous PBSCT with high-dose chemotherapy has limited applications in the treatment of patients with MBC because of relatively high costs and treatment-related mortality.1, 2 Furthermore, the usefulness of autologous PBSCT with high-dose chemotherapy to the patients with anthracycline-resistant MBC has not yet been verified.2 Among the new drugs recently developed for breast carcinoma, paclitaxel, a novel antimicrotubule agent derived from the bark of the western yew Taxus brevifolia, is one of the most exciting anticancer drugs.3 Paclitaxel facilitates the formation of tubule dimers and stabilizes microtubules against depolymerization, resulting in growth inhibition, a mechanism of action completely different from that of anthracyclines.4 Paclitaxel therefore may be effective in the treatment of anthracycline-resistant breast carcinoma.5 Various drugs have been tested in paclitaxel-based combination regimens, and one of these is ifosfamide.6 Ifosfamide is an axazaphosphorine analog of cyclophosphamide, in which one of the two 2-chloroethyl groups is shifted to the cyclic oxazaphosphorine nitrogen atom. This molecular modification provides a more effective DNA cross-linking distance between two independent functional alkylating moieties, producing a different mechanism of action from that of paclitaxel.7 Moreover, the main toxicities of ifosfamide do not overlap with those of paclitaxel. Therefore, we designed a Phase II trial of paclitaxel and ifosfamide combination chemotherapy for the treatment of anthracycline-resistant MBC patients.


Eligibility Criteria

Patients were enrolled in this study with histologically confirmed MBC progressed or recurred after the treatment of anthracycline-containing chemotherapy. Other inclusion criteria are as follows. 1) age 18–65 years; 2) Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 0–2; 3) life expectancy more than 3 months; 4) bidimensionally measurable metastatic lesion; 5) no central nervous system involvement or carcinomatous meningitis; 6) no active infection; 7) no history of prior malignancy or significant medical disease; 8) no history of taxoid and ifosfamide treatment; 9) normal hematologic function: neutrophil count greater than 2.0 × 109/L, platelet count greater than 150 × 109/L, hemoglobin level greater than 10 g/dL; 10) adequate renal and hepatic function: serum total bilirubin, alanine aminotransferase, and aspartate aminotransferase level less than 2-fold the upper limit of normal, serum creatinine level less than 1.5 mg/dL; 11) normal heart function with left ventricular ejection fraction greater than 60%. Prior radiation completed at least 4 weeks from the study registration date was permitted, insofar as it encompassed less than 30% of the total bone marrow–bearing skeleton. Any hormonal therapy was discontinued at least 3 weeks before the study entry. Anthracycline resistance was classified as primary resistance if the disease recurred during chemotherapy or progressed without response or stabilization and secondary resistance if the disease recurred within 12 months of adjuvant therapy or progressed after complete response (CR), partial response (PR), or stabilization.

Treatment Plan

Treatment consisted of paclitaxel (Taxol; Bristol-Myers Squibb, Co., Princeton, NJ), 175 mg/m2 intravenously (i.v.) in a 24-hour continuous infusion diluted in 500 mL of 0.9% normal saline on Day 1. Premedication to avoid acute allergic reaction included dexamethasone 10 mg orally (p.o.) 12 and 6 hours before paclitaxel administration, diphenhydramine 50 mg i.v. 30 minutes, and hydrocortisone 100 mg i.v. 30 minutes before paclitaxel infusion. Ifosfamide was given at the dose of 1.8 g/m2 with mesna 380 mg/m2 by 24-hour continuous infusion on Says 2– 4. The two drugs were given on separate days because of the long infusion periods of each. Daily intravenous administration of dexamethasone (20 mg) and granisetron (2 mg) was performed during chemotherapy to prevent nausea and vomiting. From the day neutrophil count decreased below 1 × 109/L after chemotherapy until neutrophil count recovered to greater than 1.5 × 109/L, granulocyte macrophage colony-stimulating factor (Leucogen; Lucky-Gumsung, Seoul, Korea) 400 μg subcutaneously was injected daily. The next cycle chemotherapy was started on Day 22 for patients whose neutrophil counts exceeded 1.5 × 109/L, and whose platelet counts exceeded 100 × 109/L. If these values were not reached by the date of scheduled retreatment, therapy was delayed by weekly intervals. If these hematologic criteria were still not fulfilled after a delay of 2 weeks, the patients were removed from the study. The doses of paclitaxel and ifosfamide were reduced by 25% in the next cycle if the nadir of neutrophil and platelet counts were less than 0.5 × 109/L and 50 × 109/L, respectively. A maximum of six cycles was given for each patient.

Pretreatment and Follow-Up Studies

Before the study enrollment, all patients underwent a complete history and physical examination. Laboratory evaluation included a complete blood cell (CBC) count with differential count, biochemical profile, carcinoembryonic antigen, CA 15-3, lactate dehydrogenase, urinalysis, baseline electrocardiogram, chest X-ray, ultrasonogram or computed tomography scan of abdomen and suspicious areas, and bone scan. Physical examination, ECOG PS, CBC counts with differential count, biochemical profile, significant tumor markers, and chest X-ray were reevaluated before the start of each cycle. Complete tumor measurements were documented by the end of every second cycles to the completion of treatment.

Response and Toxicity Evaluation

Response was evaluated by two observers not less than four weeks apart. Complete response was defined as the disappearance of all clinical evidence of active tumor and absence of disease-related symptoms for a minimum of 4 weeks. Partial response was defined as a greater than or equal to 50% reduction in the sum of the products of the biperpendicular dimensions of all measurable lesions, without the appearance of new lesions for at least 4 weeks. When there were multiple sites of metastasis, the largest masses were considered as the index lesions. Stable disease (SD) was defined as no change in tumor size or a less than 25% increase or decrease for at least 4 weeks. Progressive disease (PD) was an increase of at least 25% in the size of at least 1 measurable lesion, or the appearance of a new lesion involving pleural effusion or ascites substantiated by positive cytology.

Any patient who had received at least one course of paclitaxel and ifosfamide was evaluated for toxicities according to intention to treatment analysis. The type and severity of these toxicities were determined using the World Health Organization common toxicity scale criteria.

Statistical Analysis

Duration of response and survival determinations were measured from the starting date of paclitaxel and ifosfamide chemotherapy. The 95% confidence intervals (CIs) obtained for response rate were calculated using the binomial theorem with the Epi Info 6 (version 6.04b January 1997, Center for Disease Control) program. Survival and response duration were calculated by the Kaplan–Meier method, utilizing a microcomputer-assisted program with Statistical Package for Social Science (SPSS) for Windows 10.0.7 (SPSS Inc., Chicago, IL). Differences at P value of 0.05 were considered statistically significant.


Patient Characteristics

From March 1997 to June 1999, 21 patients were enrolled in this study. Among them, one patient was excluded from the response evaluation because the patient dropped out after one cycle of chemotherapy. However, this patient was included to toxicity evaluation using intent-to-treat analysis statistical method. Therefore, 20 were eligible for response and survival analyses. The median age was 47 (range, 27–63). The median ECOG PS was 1 (range, 0–2). All patients had been given the anthracycline-containing chemotherapy regimen FAC (cyclophosphamide 500 mg/m2, doxorubicin 50 mg/m2, 5-FU 500 mg/m2, repeated every 3 weeks) before enrollment in this study. We introduced the FAC regimen as an adjuvant therapy to 16 patients and as an initial systemic chemotherapy to 5 patients. The median interval between the initial FAC chemotherapy and the salvage treatment was 6 (range, 2–12) months. The sites of metastasis, status of estrogen receptor and menopause, intention of previous chemotherapy and hormonal therapy, and classification of anthracycline resistance are listed in Table 1.

Table 1. Baseline Characteristics of Enrolled Patients
CharacteristicNo. of patients (%)
  • a

    According to Eastern Cooperative Oncology Group criteria.

Performance statusa
 07/21 (33)
 15/21 (24)
 29/21 (43)
Predominant site of metastasis
 Lung9/21 (43)
 Bone6/21 (29)
 Liver4/21 (19)
 Skin/soft tissue6/21 (29)
 Others1/21 (5)
Estrogen receptor status
 Positive11/21 (52)
 Negative10/21 (48)
Menstrual status
 Premenopausal9/21 (43)
 Postmenopausal12/21 (57)
Intention of previous chemotherapy
 Adjuvant16/21 (76)
 Palliative5/21 (24)
Intention of hormonal therapy
 Adjuvant13/21 (62)
 Palliative3/21 (14)
Resistance to anthracycline
 Primary5/21 (24)
 Secondary16/21 (76)

Response and Survival Analysis

Twenty patients were assessable for response and survival analysis. The responses of these patients were summarized in Table 2. Objective response was found in 9 patients (42.9%; 95% CI, 22.6–65.6%), including 3 (14.3%) CRs and 6 (28.6%) PRs. Two were in the group of primary anthracycline resistance, and the others belonged to the group of secondary resistance. Of the three patients who achieved CR, one had two metastatic sites (liver and paraaortic lymph nodes), and the others had one metastatic site (lung or liver). Most of PRs were observed in the liver, lung, lymph nodes, and bone. Median response duration was 10 (range, 2–24), and median survival time was 19+ (range, 2–32+) months, respectively. The overall survival curve is plotted in Figure 1.

Table 2. Response to Treatment (Evaluated Patients, n = 21)
Efficacy variablen (95% confidence interval)
Complete response3/21 (14.3%, 3.8–37.4%)
Partial response6/21 (28.6%, 12.2–52.3%)
Stable disease8/21 (42.9%, 22.6–65.6%)
Progressive disease3/21 (14.3%, 3.8–37.4%)
Figure 1.

Kaplan–Meier estimate of the cumulative probability of survival in paclitaxel/ifosfamide salvage regimen.


Toxicity was evaluated for 98 cycles of chemotherapy. Total reversible alopecia occurred in all cases. Reversible Grade 3 and 4 leukopenia was observed in 13 and 3 patients, respectively. Three patients suffered from neutropenic fever for seven cycles of chemotherapy. However, no chemotherapy was delayed because of bone marrow suppression. Other significant (Grade 3 and 4) toxicities were anemia (n = 6), thrombocytopenia (n = 4), peripheral neuropathy (n = 3), oral mucositis (n = 2), hepatic dysfunction (n = 1), and nausea/vomiting (n = 1). Other significant (Grade 3 and 4) toxicities, such as diarrhea and hypersensitivity reaction, were not observed. There were no chemotherapy-related deaths. The details of toxicity are described in Table 3.

Table 3. Major Toxicities in 21 Patients Receiving Paclitaxel and Ifosfamide Combination Chemotherapy
ToxicityWHO grade
0–1 (%)234
  1. WHO: World Health Organization.

Leukopenia0 (0)5 (24)13 (62)3 (14)
Thrombocytopenia5 (24)12 (57)3 (14)1 (5)
Anemia7 (33)8 (38)6 (29)0 (0)
Peripheral neuropathy13 (62)5 (24)3 (14)0 (0)
Stomatitis/mucositis13 (62)6 (29)2 (9)0 (0)
Hepatic dysfunction20 (95)0 (0)1 (5)0 (0)
Nausea/vomiting3 (14)17 (81)1 (5)0 (0)


Anthracycline has been known as the most active agent for breast carcinoma. Therefore, many combination chemotherapeutic regimens containing anthracycline have been developed. Among them, the FAC regimen consisting of 5-FU, doxorubicin, and cyclophosphamide is one of the most effective regimens for breast carcinoma. The reported response rate and duration for the FAC regimen as a first-line combination chemotherapy for breast carcinoma are approximately 50–75% including approximately 10% CR and 8–15 months, respectively.5 However, it is difficult to treat breast carcinoma progressed or recurred after the treatment of anthracycline-containing combination chemotherapy. Patients with anthracycline-resistant breast carcinoma especially have a poor prognosis for second-line chemotherapy because of the subsequent low response rate. It is very important to define the concept of anthracycline resistance. One definition of anthracycline resistance requires PD during anthracycline chemotherapy. However, anthracycline resistance was less strictly defined in most studies and also has been used for disease that progressed or recurred after a chemotherapy free interval of 6–12 months or longer.8–12 Therefore, we defined disease that never responded to anthracycline-containing regimens as “primarily resistant” and disease that initially responded, but that progressed or recurred within 12 months after anthracycline-containing chemotherapy, as “secondarily resistant.”

Recently, several drugs such as paclitaxel, docetaxel, and vinorelbine have been developed and studied for the treatment efficacy of breast carcinoma patients pretreated with anthracycline-based chemotherapy.13–15 Among them, paclitaxel is the drug that does not have complete clinical cross-resistance with anthracyclines and has been revealed to have significant single-agent activity in nonpretreated MBC patients with a reported response rate of 30–71%.13, 14 Therefore, paclitaxel is ascertained to be at least as active as doxorubicin in terms of single-agent activity in many trials. In the studies of the treatment of anthracycline-resistant advanced breast carcinoma in which there was a less strict definition of resistance (pretreatment, recurrence <6 or 12 months after anthracycline-containing adjuvant therapy), the reported response rate and duration of paclitaxel single treatment with various dose schedules after prior exposure to anthracyclines was 13–48% and 5–9 months, respectively.8–12 In previous studies, the infusion time for single paclitaxel treatment was usually 3 hours. We chose a 24-hour infusion duration on the basis of the study by Seidman et al. that tried 24-hour continuous infusion of paclitaxel 250 mg/m2 delivered every 21 days to 40 breast carcinoma patients with anthracycline resistance.8 Seidman et al.'s study contained the same criteria as our study and reported a response rate and duration of 32% and 7 months. Moreover, recently, Smith et al. conducted a randomized trial of 3-hour versus 24-hour infusion of paclitaxel 250 mg/m2 delivered every 21 days and reported a significantly higher response rate with 24-hour infusion, although without significant improvement in survival over 3-hour infusion. They also observed that paclitaxel as a 24-hour infusion resulted in increased hematologic toxicity and decreased neurosensory toxicity.16

There have been many studies of combination chemotherapy containing paclitaxel for the treatment of anthracycline-resistant breast carcinoma (Table 4). Until comparatively lately, a platinum analog was widely used for the combination chemotherapy, and the response rate and duration was reported as 44–60% and 8–11.5 months, respectively.17, 18 5-Fluorouracil with or without folinic acid also was tried for the combination chemotherapy with paclitaxel, and a remission rate of approximately 60% with a median duration of 8–10 months was achieved.19, 20 We chose ifosfamide as a combination chemotherapeutic drug with paclitaxel for the treatment of MBC patients showing anthracycline resistance pretreated with FAC regimen. Ifosfamide produces objective responses in 30–40% of patients with MBC and 7–23% of patients with breast carcinoma pretreated with cyclophosphamide-containing chemotherapy as a single agent.21, 22 Walters et al. conducted a study similar to ours with regard to pretreatment chemotherapy.23 Single-agent ifosfamide 2 g/m2 was administered intravenously daily for 5 days to the patients mainly pretreated with cyclophosphamide and doxorubicin and reported a 7% (2 of 29) response rate. Recently, Bisagni et al. tried ifosfamide single-agent therapy for patients with MBC pretreated with doxorubicin or cyclophosphamide and reported a response rate of 23% without significant toxicity.24 Moreover, ifosfamide produces manageable bone marrow toxicity and has a distinct nonhematologic toxicity pattern from paclitaxel and has a different mechanism of action than paclitaxel.

Table 4. Studies of Combination Chemotherapy Containing Paclitaxel for the Treatment of Anthracycline-Resistant Breast Carcinoma
StudyNo. of patientsTreatmentObjective response rate (%)Median survival duration (mos)
  1. 5-FU: 5-fluorouracil.

Fountzilas et al.1737Paclitaxel + carboplatin4412
Liombart-Cussac et al.1823Paclitaxel + cisplatin60
Klaassen et al.2020Paclitaxel + 5-FU/folinic acid55
Current study21Paclitaxel + ifosfamide42.910

For the study of paclitaxel and ifosfamide combination chemotherapy to advanced breast carcinoma, Murad et al. performed Phase II trial of paclitaxel and ifosfamide as a salvage treatment in MBC patients.6 However, the study by Murad et al. was different from ours in some key points. They studied patients pretreated with various chemotherapeutic regimens with or without anthracycline, and the paclitaxel infusion time was 3 hours. They reported a 48% response rate with 7+ months of median response duration without lethal toxicity. In a recent study investigating clinical feasibility of paclitaxel by 24-hour infusion and ifosfamide by 72-hour infusion, Bunnell et al. conducted a Phase I trial of that regimen with granulocyte colony-stimulating factor support for patients with refractory solid tumors.25 They found that the regimen with the highest planned dose level of paclitaxel 190 mg/m2 i.v. over 24-hours and ifosfamide 3 g/m2/day i.v. for 3 days was well tolerated without significant hematologic and nonhematologic toxicities.

In the current study, the dose level of paclitaxel and ifosfamide was 175 mg/m2 i.v. for 24-hours and ifosfamide 1.8 g/m2/day i.v. for 3 days, and 45% response rate and median response duration of 10 months without significant regimen-related toxicity was observed. This result was comparable with the above-mentioned results of other salvage studies for anthracycline-resistant breast carcinoma. Our results reveal the moderate antitumor activity of the paclitaxel and ifosfamide combination as a salvage chemotherapy regimen for anthracycline-resistant breast carcinoma patients. Notably, this combined regimen was able to produce a 42.9% (95% CI, 22.6–65.9%) objective response rate with a CR in 3 of them. According to the literature, neither agent given alone would produce such a substantial response rate in anthracycline-resistant breast carcinoma patients. The main toxicity of paclitaxel by 24-hour infusion and ifosfamide combination chemotherapy in our study was bone marrow suppression. However, neutrophil and platelet count recovery was rapid, and therefore treatment delay did not occur. Furthermore, 16 patients (76%) experienced Grade 3 or 4 leukopenia, and all recovered without complication. Other toxicities were also feasible and could be tolerable. With regard to nonhematologic toxicities, nausea, vomiting, stomatitis, and diarrhea were mild. Of note, the incidence of significant peripheral neuropathy was low (3 of 21 patients, 13%), and fatal complications such as hemorrhagic cystitis and congestive heart failure did not develop. Such salvage chemotherapy regimen is the first report in anthracycline-resistant MBC patients. In conclusion, the current study shows that the combination chemotherapy of paclitaxel by 24-hour infusion and ifosfamide is effective and well tolerated in anthracycline-resistant MBC patients. Therefore, further well designed comparative studies exploiting this regimen with more patients are warranted to verify our result and the clinical feasibility of this regimen as a frontline treatment to the patients with advanced disease.