VOLUME

23

NUMBER

12

APRIL

20

2005

JOURNAL OF CLINICAL ONCOLOGY

O R I G I N A L

R E P O R T

Phase II Clinical Trial of Ixabepilone (BMS-247550), an

Epothilone B Analog, in Metastatic and Locally
Advanced Breast Cancer
Jennifer A. Low, Suparna B. Wedam, James J. Lee, Arlene W. Berman, Adam Brufsky, Sherry X. Yang,
Marianne S. Poruchynsky, Seth M. Steinberg, Nitin Mannan, Tito Fojo, and Sandra M. Swain
From the Cancer Therapeutics Branch,
Medical Oncology Clinical Research
Unit, and Biostatistics and Data
Management Section, Center for
Cancer Research, National Cancer
Institute, National Institutes of
Health, Bethesda, MD; and the
Magee-Womens Hospital, University
of Pittsburgh Cancer Institute,
Pittsburgh, PA.
Submitted October 5, 2004; accepted
January 20, 2005.
Presented at the 40th Annual
Meeting of the American Society of
Clinical Oncology, New Orleans, LA,
June 5-8, 2004 (preliminary portions
of the study, abstract 545).
Authors disclosures of potential conflicts of interest are found at the end of
this article.
Address reprint requests to Sandra M.
Swain, MD, Cancer Therapeutics Branch,
Center for Cancer Research, National
Cancer Institute, Bldg 8, Rm 5101, 8901
Wisconsin Ave, Bethesda, MD 208895015; e-mail: swains@mail.nih.gov.
0732-183X/05/2312-2726/$20.00
DOI: 10.1200/JCO.2005.10.024

Purpose
Ixabepilone (BMS-247550) is an epothilone B analog that stabilizes microtubules and has
antitumor activity in taxane-refractory patients in phase I studies. In a phase II trial, we
evaluated the efficacy and safety of ixabepilone in women with metastatic and locally
advanced breast cancer.
Patients and Methods
Breast cancer patients with measurable disease who had paclitaxel and/or docetaxel as
prior neoadjuvant, adjuvant, or metastatic therapy were treated with ixabepilone at
6 mg/m2/d intravenously on days 1 through 5 every 3 weeks. Levels of glutamate (glu)
-terminated and acetylated -tubulin, markers of microtubule stabilization, were detected
by Western blot and by immunohistochemistry in a subset of matched pre- and
post-treatment tumor biopsies.
Results
Thirty-seven patients received 153 cycles of ixabepilone. The best responses were a
complete response in one patient (3%), partial responses in seven patients (19%), and stable
disease in 13 patients (35%). Grade 3 and 4 toxicities included neutropenia (35%), febrile
neutropenia (14%), fatigue (14%), diarrhea (11%), nausea/vomiting (5%), myalgia/arthralgia
(3%), and sensory neuropathy (3%). Two patients were removed from study because of
prolonged grade 2 or 3 neurotoxicity, and three patients were removed from study for other
grade 3 and 4 nonhematologic toxicities. Compared with baseline levels, levels of both
glu-terminated and acetylated -tubulin were increased in tumor biopsies performed after
ixabepilone therapy.
Conclusion
An objective response was seen in 22% of the patients in a population who had been
previously treated with a taxane. Sensory neuropathy was mild with grade 3 neurotoxicity rarely seen. Microtubule stabilization occurred in tumor biopsies after treatment
with ixabepilone.
J Clin Oncol 23:2726-2734.

INTRODUCTION

The microtubule-stabilizing drugs paclitaxel and docetaxel are considered among

the most effective chemotherapeutic agents
for the treatment of metastatic breast cancer.1 However, during the course of treatment of metastatic disease, even tumors that

initially respond to these drugs eventually develop resistance, and the sequential use of
other chemotherapeutics is standard practice.
For many patients receiving paclitaxel and docetaxel, sensory peripheral neuropathy can be
significant and even dose limiting. New therapies with microtubule-stabilizing activity are
being investigated for their potential activity in

2726

Downloaded from jco.ascopubs.org on January 4, 2016. For personal use only. No other uses without permission.
Copyright 2005 American Society of Clinical Oncology. All rights reserved.

Ixabepilone Phase II Trial in Breast Cancer

the treatment of cancer,2 and have generated considerable

interest for the treatment of metastatic breast cancer.
Microtubules are filaments consisting of - and
-tubulin. Microtubule homeostasis is characterized by dynamic polymerization and depolymerization of the tubulin
subunits. Disruption of microtubule function through increased stability can lead to cell cycle arrest and apoptosis.
Microtubule stabilization results in a number of posttranslational changes to -tubulin,3 including the removal
of the carboxy-terminal tyrosine to reveal a terminal glutamate (glu-terminated -tubulin),4 and acetylation of the
lysine 40 (acetylated -tubulin).5 Detyrosination, a potential marker for tumor aggressiveness,6 occurs only on polymerized microtubules, whereas its restoration occurs
exclusively on soluble -tubulin heterodimers. Tubulin
acetylation occurs only after polymerization; depolymerized tubulin is rapidly deacetylated in vivo.7 Microtubulestabilizing agents reduce microtubule dynamics and
promote tubulin polymerization.8
The epothilones are macrolide fermentation products
of the myxobacterium Sorangium cellulosum. Epothilones A
and B competitively displace paclitaxel from microtubules
in vitro, and similar to paclitaxel, stabilize microtubules and
cause cell cycle arrest and cytotoxicity.9 Ixabepilone (BMS247550) is a semisynthetic analog of epothilone B, in which
the lactone oxygen of epothilone B is replaced by nitrogen,
increasing the stability of the drug.10 In preclinical models,
ixabepilone was active in the paclitaxel-resistant breast cancer
cell line Pat-21 and in other paclitaxel-insensitive models.11
A phase I study established that ixabepilone administered at 6 mg/m2/d on 5 consecutive days every 3 weeks was
the maximum-tolerated dose; neutropenia was the doselimiting toxicity.12 This regimen was considered particularly attractive because the longer cycle administration time
had less neurotoxicity than administration schedules used
in other phase I trials of ixabepilone.
We conducted a phase II trial of ixabepilone for patients with metastatic and locally advanced breast cancer,
with the primary objectives of determining the efficacy and
safety of ixabepilone. Our secondary objectives included
assessing ixabepilone activity at the tumor site by evaluating
microtubule stabilization within the target tissue. Using
tumor biopsies collected at baseline and during treatment
(day 2 of cycle 2), we quantified -tubulin acetylation and
glu-termination to determine whether baseline levels predicted response to ixabepilone and whether changes during
treatment correlated with response.
PATIENTS AND METHODS
Eligibility
Eligible patients had a diagnosis of metastatic or locally advanced breast adenocarcinoma confirmed by the pathology department of the enrolling institution; an Eastern Cooperative

Oncology Group performance status of 0, 1, or 2; and measurable

disease by Response Criteria in Solid Tumors (RECIST) criteria.13
Laboratory values were required to be within the specified ranges
within 1 week of study enrollment, including an absolute neutrophil count of 1.5 109/L and thrombocyte count of 100
109/L. All patients had received at least one prior neoadjuvant,
adjuvant, or metastatic regimen that contained docetaxel or paclitaxel; otherwise, there were no other restrictions on prior number of therapies, including anthracycline-containing or high-dose
chemotherapy regimens. Patients with nonmetastatic locally advanced breast cancer had received and not responded to prior
anthracycline and taxane neoadjuvant chemotherapy. Because of
concerns that ixabepilone metabolism may be inhibited by potent
cytochrome P450 3A4 inhibitors, patients were also excluded from
receiving the following medications at enrollment and while
enrolled onto the study: amiodarone, clarithromycin, erythromycin, fluconazole, itraconazole, ketoconazole, indinavir, nelfinavir,
ritonavir, and saquinavir. Patients with grade 2 or greater neuropathy at baseline were excluded.
Study Design and Treatment Modifications
This phase II study was approved by the institutional review
board (IRB) for the Center for Cancer Research, National Cancer
Institute (NCI), and by the IRB of the Magee-Womens Hospital
(Pittsburgh, PA). All patients signed the IRB-approved informed
consent before participating.
Ixabepilone (provided by the Cancer Therapy Evaluation
Program, NCI) was administered during 1 hour intravenously
(IV) on 5 consecutive days every 3 weeks. All patients received a
starting dose of 6 mg/m2/d for the first cycle. All patients received
premedication with diphenhydramine at 50 mg IV or orally, or
hydroxyzine at 25 mg orally, and ranitidine at 50 mg IV, 30 to 60
minutes before each dose of ixabepilone. Prophylactic antiemetics
were not routinely administered, but were added to the regimen
for patients experiencing toxicity. An absolute neutrophil count
1.0 109/L and a thrombocyte count 75 109/L was
required on the first day of treatment of each cycle.
Toxicities were assessed using NCI Common Toxicity Criteria version 2.014 at baseline and at the conclusion of each cycle of
treatment. Dose reductions to 5 mg/m2/d and then to 4 mg/m2/d
were implemented for patients experiencing grade 3 and 4 nonhematologic toxicities during the prior cycle, febrile neutropenia, or
for toxicities requiring a delay in treatment. Toxicities (including
neurotoxicity) required resolution to grade 1 or to baseline before
the next cycle of treatment was administered. Granulocyte colonystimulating factors were not given initially at cycle 1, but could be
added to subsequent cycles of treatment for patients with febrile
neutropenia or delayed neutrophil recovery requiring dose delay.
Treatment intervals could be extended up to 5 weeks between the
first days of treatment for patients with optimal responses who had
completed six cycles of treatment. Patients were removed from
study for the following reasons: disease progression, persistent
grade 2 neuropathy or grade 3 neuropathy lasting more than 7
days, grade 3 or 4 toxicities requiring more than two dose reductions or delay of treatment for more than 5 weeks from the beginning of the last cycle, or per patient request.
Response Assessment
Measurable disease was assessed by imaging using the
RECIST criteria.13 Patients underwent baseline imaging within 4
weeks of enrollment, and were scanned before every other cycle.
Stable disease was determined before cycle 3 at 6 weeks. All partial
2727

www.jco.org

Downloaded from jco.ascopubs.org on January 4, 2016. For personal use only. No other uses without permission.
Copyright 2005 American Society of Clinical Oncology. All rights reserved.

Low et al

and complete responses were confirmed at least 4 weeks later with

repeat imaging.
Statistical Design and Methods
This protocol was designed to evaluate the efficacy of ixabepilone. The study used an optimal two-stage design to rule out a low
5% response rate in favor of a 20% response rate.15 With a 10%
probability of accepting a poor agent and rejecting a good
agent, if at least one of the first 12 patients had a response,
enrollment would continue until 37 patients were enrolled.
Four or more responses were considered consistent with an
active agent worthy of additional development. Under the null
hypothesis (5% response probability), the study had a 60%
probability of early termination.
Time to progression was calculated for all patients enrolled
from the first day of treatment to the day of evidence of progressive
disease, with censoring at last follow-up (without progression), or
removal from study for toxicity. The one ineligible patient was
censored at the time of enrollment. Progression-free survival
probabilities as a function of time were calculated using the
Kaplan-Meier method. Duration of response was calculated for
the eight patients who had confirmed responses, and was calculated from the day of first radiologic partial response until the date
of progression. The one remaining patient enrolled onto the study
was censored at the date of her last visit, August 25, 2004.
Correlative Studies
Patients with tumors that could be biopsied under local anesthesia had core or punch biopsies obtained at baseline (before
receiving the first dose of ixabepilone), and again on cycle 2 day 2,
approximately 18 to 24 hours after the first dose of cycle 2. The
biopsy timing was chosen to allow enough time (after a full cycle,
but less than 24 hours after a treatment infusion) for ixabepilone
to theoretically have a discernible target effect.
For Western blot analysis, frozen tissues were thawed, macerated, and dounce-homogenized in less than 1 mL of a hypotonic
lysis buffer containing 1 mmol/L MgCl2, 2 mmol/L ethyleneglycoltetraacetic acid, 0.5% Nonidet P-40 (Amersham Biosciences,
Piscataway, NJ), 20 mmol/L Tris-HCl, pH 6.8, aprotinin (200
units/mL), and an additional EDTA-free protease inhibitor-mix
tablet (Roche Diagnostics, Mannheim, Germany). Tissue lysates
were vortexed and centrifuged to remove unlysed cells and tissue
debris. Equal amounts of protein (10 to 25 g) were run on 7.5%
sodium dodecyl sulfatepolyacrylamide gel electrophoresis gels
and transferred to Immobilon (Millipore, Billerica, MA) membranes. The membranes were blocked, and sequentially probed with a
mouse monoclonal antibody to -actin (Sigma Chemical Co, St
Louis, MO) as a loading control, mouse monoclonal anti--tubulin
clone DM1A (Sigma), mouse monoclonal antiacetylated -tubulin,
clone 6-11B-1 (Sigma), and rabbit antidetyrosinated tubulin polyclonal antibody (Chemicon International Inc, Temecula, CA). The
membranes were then incubated with horseradish peroxidase
conjugated secondary antibodies, with appropriate species reactivity,
and the protein-antibody complexes were detected by using SuperSignal Chemiluminescent reagents (Pierce Biochemicals, Rockford,
IL) and quantitated by densitometry.
For immunohistochemistry, formalin-fixed, paraffin-embedded
sections were processed as described previously.16 Acetylated and
glu-terminated mouse monoclonal -tubulin antibodies (described above), and mouse monoclonal anti-Ki67 (MIB-1) antibody (Dako, Carpinteria, CA) were applied to tissue sections at a
dilution of 1:1000, 1:1000, and 1:50, respectively. The sections
2728

were scored quantitatively using the Automated Cellular Imaging

System (ACIS; Chromavision, San Juan Capistrano, CA). Six areas
of each tumor were scored using a free-scoring or 40 magnification tool to generate an averaged percentage and intensity of
stained tumor cells. The staining index was calculated by multiplying the percentage of positively stained cells by the average
staining intensity after subtracting the machine readouts of the
corresponding negative control for each marker. For Ki67, a
labeling percentage was reported.
RESULTS

Patients
In this phase II study, 37 patients were enrolled at the
NCI (29 patients) or at the University of Pittsburgh Cancer
Institute (eight patients) between June 2002 and March
2004. All patients received at least one cycle and all patients
are included in the analysis.
The baseline characteristics of the patients are summarized in Table 1. All patients had at least two cycles of
a paclitaxel- or docetaxel-containing regimen (as neoadjuvant, adjuvant, or metastatic therapy) at some point in
their therapeutic history. Twenty-two (59%) patients had progressive disease while receiving paclitaxel- and/or docetaxelcontaining regimens.

Table 1. Baseline Characteristics of Patients With Metastatic and

Locally Advanced Breast Cancer Enrolled Onto a Phase II
Trial of Ixabepilone
No. of
Patients

Characteristic
Total
Age, years
Median
Range
ECOG performance status
0
1
2
ER or PR positive
Her2 3/FISH positive
Receiving warfarin anticoagulation
Stage of disease
IIIB (all IBC)
IV
Sites of metastatic disease
Visceral
Not visceral
Both
No. of prior metastatic chemotherapy regimens
0
1
2
3-9

37
51
29-63
10
26
1
19
6
5

27
70
3
51
16
14

3
34

8
92

25
32
20

68
86
54

5
10
6
16

14
27
16
43

Abbreviations: ECOG, Eastern Cooperative Oncology Group; ER, estrogen receptor; PR, progesterone receptor; FISH, fluorescence in situ
hybridization; IBC, inflammatory breast cancer.

JOURNAL OF CLINICAL ONCOLOGY

Downloaded from jco.ascopubs.org on January 4, 2016. For personal use only. No other uses without permission.
Copyright 2005 American Society of Clinical Oncology. All rights reserved.

Ixabepilone Phase II Trial in Breast Cancer

Patients received a median of four cycles (range, one to

11 cycles) of ixabepilone chemotherapy. Of the 37 patients
enrolled, 29 patients were removed from study for disease
progression, two patients were removed after maximum
benefit was achieved (one for neoadjuvant therapy and one
for resolution of nonmeasurable disease), five patients were
removed for prolonged toxicity, and one patient remains
enrolled onto the study.
Efficacy
The objective response was 22% (95% CI, 9.8% to
38.2%), with one complete response and seven partial responses. Stable disease for at least 6 weeks was the best
response for 35% of patients (13 patients). The median time
to progression for all patients was 80 days (Fig 1). Patients
classified as having progressive disease included one patient
enrolled and treated with good clinical response but determined to be ineligible retrospectively when found to be
without measurable disease by RECIST criteria, one patient
with severe toxicity during the first cycle who refused additional evaluation, and two patients who were considered to
have progressive disease 8 and 10 days after starting therapy
because of life-threatening tumor complications.
For eight patients with confirmed partial responses, the
median response duration was 118 days. One patient with
a partial response at 6 weeks improved to a confirmed
complete response after 8 months (250 days). Measurable
responses were seen in lung and liver lesions, and supraclavicular, mediastinal, and axillary lymph nodes. Improvements were also seen in nontarget lesions, such as breast and
bone lesions and pleural effusions. Responses were seen in
patients who had experienced disease progression after receiving a prior taxane and well as patients previously exposed to an anthracycline, capecitabine, and a taxane (Table
2). One of two patients enrolled who had experienced dis-

Fig 1. Kaplan-Meier curves showing progression-free duration for patients

receiving ixabepilone on the phase II study. Patients were censored for
removal from study for toxicity (five patients), ineligibility (one patient), or
continuing to be progression-free (one patient). ( ) 95% CIs.

ease progression after receiving both paclitaxel and docetaxel had a partial response to ixabepilone. One patient
with inflammatory breast cancer who had stable disease
with neoadjuvant docetaxel and doxorubicin was switched
to, and responded to, ixabepilone as a neoadjuvant therapy.
Toxicities
Toxicity information was collected for all patients, who
received a total of 153 cycles of therapy. For each patient, the
worst grades seen of the most common hematologic and
nonhematologic toxicities are summarized in Table 3. One
patient was removed from study because of grade 3 sensory
neuropathy and one patient was removed from study because of prolonged grade 2 sensory neuropathy. During
their first cycle of ixabepilone, two patients had multiple
grade 3 and 4 toxicities, including febrile neutropenia,
thrombocytopenia, severe mucositis, diarrhea, fatigue, and
sensory neuropathy. Neither of these patients received additional treatment with ixabepilone; one of these two patients received concurrent warfarin and clarithromycin,
whereas the other did not.
One patient developed slowly progressive patchy pulmonary infiltrates that showed an organizing, nonspecific
interstitial pneumonia and metastatic tumor cells on lung
biopsy. She was removed from the study, and after receiving
high-dose corticosteroids the infiltrate resolved symptomatically and radiologically.
Twelve patients required dose reduction while enrolled
onto the study (median, cycle 5; range, cycles 2 to 10). The
dose was reduced because of neuropathy in six patients, and
because of diarrhea, fatigue, and neutropenia in two patients each. Only one patient required a second dose reduction because of a grade 3 myalgia.
Five patients received erythropoietin or darbepoetin.
Four patients received filgrastim or pegfilgrastim for febrile
neutropenia. Three patients received prophylactic pegfilgrastim because of prolonged neutrophil recovery in a prior
cycle or for a prior episode of febrile neutropenia.
Peripheral Sensory Neuropathy
Of the 24 patients with no baseline neuropathy, 14
patients (58%) developed grade 1, six patients (25%) developed grade 2, and one patient (4%) developed grade 3
peripheral sensory neuropathy as the worst grade of neuropathy. Only two (15%) of 13 patients with baseline grade
1 neuropathy developed grade 2 neuropathy. In total, nine
patients developed at least grade 2 neuropathy during the
course of their treatment; two of these patients developed
grade 2 neuropathy during the first cycle, along with other
severe hematologic and nonhematologic toxicities. Of the
other seven patients, the median time to development of
grade 2 neuropathy was 108 days on study (range, 42 to 189
days). The median times to development and resolution of
neuropathy and the outcomes are summarized in Figure 2.
2729

www.jco.org

Downloaded from jco.ascopubs.org on January 4, 2016. For personal use only. No other uses without permission.
Copyright 2005 American Society of Clinical Oncology. All rights reserved.

Low et al

Table 2. Responses to Ixabepilone by Prior Therapy

CR/PR
(n 8)
Prior Therapy
Prior docetaxel
Experienced disease progression while
receiving prior docetaxel
Prior paclitaxel
Experienced disease progression while
receiving prior paclitaxel
Prior docetaxel and paclitaxel
Prior doxorubicin, capecitabine, and taxane
Median number of days since last taxane

SD
(n 13)

PD
(n 16)

No. of
Patients

No. of
Patients

No. of
Patients

Total
(n 37)

6
3

21
20

10
6

34
40

13
6

44
40

29
15

5
2

28
20

7
3

39
30

6
5

33
50

18
10

3
3

30
14

4
11

40
52

3
7

30
33

10
21
301

189

369

254

Abbreviations: CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease.

Correlative Studies
Biopsies were collected at baseline and during cycle 2
from five patients (one patient had partial response, two
patients had stable disease, and two patients had progressive
disease) and were confirmed by pathology to contain tumor. Four of these paired patient biopsies were analyzed by
Western blot. For one patient (patient 5), the baseline sample was not adequate. Increased acetylation of -tubulin
was seen by Western blot for all four patients (labeled
patients 1 through 4 with their responses in Fig 3). Glutermination was also increased in patients 2, 3, and 4.
Because the protein lysates run on Western blot contain both tumor and stromal components of the biopsy, we

next chose to quantify tumor changes alone by immunohistochemistry. Four of the five paired patient biopsies (labeled patients 1, 2, 3, and 5) had tissue available for
sectioning for immunohistochemistry. For patient 4, no
tumor remained from the baseline paraffin block. For patient 1, who had a partial response to ixabepilone, levels of
acetylated -tubulin were detected in both tumor and stromal cells in the baseline sample and increased in both cellular compartments with treatment (Figs 4A and 4B). In
contrast, for patient 5, who had progressive disease, levels of
acetylated -tubulin were detected only in stromal cells at
baseline and increased in the stroma but were still absent in
the tumor cells with treatment (Figs 4C and 4D). Baseline

Table 3. Observed Toxicities for Patients Receiving Ixabepilone

CTC Grade
0
Toxicity
Hematologic
Neutropenia
Hemoglobin
Thrombocytopenia
Febrile neutropenia
Nonhematologic
Fatigue
Sensory neuropathy
Myalgia/arthralgia
Nausea
Diarrhea
Constipation
Vomiting
Alopecia
Dysgeusia
Nail changes

No. of
Patients

No. of
Patients

12
10
22
32

32
27
59
86

13
17
18
17
24
27
29
17
25
26

35
46
49
46
65
73
78
46
68
70

No. of
Patients

No. of
Patients

3
14
12

8
38
32

9
13
0

24
35
0

11
11
15
17
7
2
3
7
7
6

29
29
41
46
19
5
8
19
19
16

8
8
3
1
2
8
3
13
5
5

22
22
8
3
5
22
8
35
14
14

4
%

No. of
Patients

6
0
1
5

16
0
3
14

7
0
2
0

19
0
5
0

3
1
1
2
4
0
2

8
3
3
5
11
0
5

2
0
0
0
0
0
0

5
0
0
0
0
0
0

NOTE. Frequent and relevant toxicities are shown with the percentage of patients who developed the listed grade of toxicity as their worst grade while
enrolled onto the study.
Abbreviation: CTC, National Cancer Institute Common Toxicity Criteria.

2730

JOURNAL OF CLINICAL ONCOLOGY

Downloaded from jco.ascopubs.org on January 4, 2016. For personal use only. No other uses without permission.
Copyright 2005 American Society of Clinical Oncology. All rights reserved.

Ixabepilone Phase II Trial in Breast Cancer

Fig 2. Incidence and outcomes for patients with of peripheral sensory neuropathy.
The numbers of patients in each category
are shown, with the median number of days
to occurrence shown, where applicable.
*One patient was lost to follow-up within the
first 21 days.

levels of acetylated -tubulin were present before treatment

and increased after treatment in the patients with response
and stable disease (Fig 5A), and levels of glu-terminated
-tubulin increased in all patients with treatment (Fig 5B).
Ki67 labeling percentage, a measure of tumor proliferation,
decreased in the same two patients with increased levels
of acetylated -tubulin after treatment, but the Ki67
labeling percentage remained similar in the two patients
with no change in the level of acetylated -tubulin after
treatment (Fig 5C).
DISCUSSION

Ixabepilone has been studied in a number of phase I and II

clinical trials, including one phase I and three phase II breast

Fig 3. Expression of acetylated and glu-terminated -tubulin detected in

paired biopsies collected at baseline (BL) and on cycle 2 day 2 (C2) of
ixabepilone. Western blot showing total -tubulin, acetylated -tubulin,
glu-terminated -tubulin, and actin. The patient number is listed along the
x-axis. PR, partial response; SD, stable disease; PD, progressive disease.

cancer trials (not including the current study) that have

been presented previously in abstract form. Partial response
rates for the phase II clinical trials have been 12% for a
taxane-resistant population,17 44% for a population with
prior adjuvant or neoadjuvant anthracycline therapy and
no prior therapy for metastatic disease,18 and 31% for another mixed population of breast cancer patients.19 Because
our patients received taxanes in a variety of prior regimens,
including neoadjuvant or adjuvant therapy, and the number of prior treatments for metastatic disease varied widely,
it is not surprising that our response rate of 22% falls within
the range of response rates previously reported. In metastatic breast cancer patients with docetaxel-resistant disease, weekly paclitaxel had a 32% response rate,20 whereas
in patients with paclitaxel-resistant disease, docetaxel had
an 18% response rate.21
In our study, there was a low incidence of hematologic
toxicities seen with this schedule, and the most significant
complaints were fatigue, myalgia, and peripheral sensory
neuropathy. However, for two patients, significant toxicities occurred in the first cycle requiring cessation of additional therapy. In both patients severe febrile neutropenia,
nausea, vomiting, diarrhea, mucositis, and peripheral neuropathies occurred within a week of completing the first
cycle of treatment. The reason for the comparatively severe
toxicities seen in these two patients is unknown. One
patient, who had been extensively treated with docetaxel
previously without significant complications, received clarithromycin (a cytochrome P450 3A4 inhibitor) on day 4 of
2731

www.jco.org

Downloaded from jco.ascopubs.org on January 4, 2016. For personal use only. No other uses without permission.
Copyright 2005 American Society of Clinical Oncology. All rights reserved.

Low et al

Fig 4. Immunohistochemistry of acetylated -tubulin. Lymph node metastases collected at (A) baseline and (B) cycle 2 from patient 1 with partial response
(PR), and chest wall metastases collected at (C) baseline and (D) cycle 2 from patient 5 with progressive disease (PD) are shown at 400 magnification. Open
arrows, tumor; solid arrows, stromal staining.

cycle 1, leading us to speculate whether a drug-drug interaction may have been responsible for symptoms. However,
the second patient did not receive any cytochrome P450
inhibitors. Both patients received warfarin, as did three
other patients without severe toxicities. Because blood sampling was not done with this study, pharmacokinetic information regarding drug levels was not available.
One of the most notable complications associated with
taxane and other microtubule-targeted treatments is the
development of sensory peripheral neuropathy. Although
most of our patients who received four or more cycles of
ixabepilone had some subjective symptoms of neuropathy,
these usually did not interfere with function. Only one
2732

patient was considered to have grade 3 neurotoxicity because her peripheral neuropathy was the primary reason she
felt compelled to stop working. Grade 3 neurotoxicity was
not seen in the phase I study with the regimen administered
daily for 5 days.12 The grade 3 peripheral sensory neuropathy seen with ixabepilone at this schedule (3%) appears to
be less than that reported for other ixabepilone schedules
(4% to 25%).19,22-26 The lower incidence of neuropathy
may be related to the administration schedule, or to the
population enrolled onto this study. In our study, all of our
patients had received a taxane previously, which may make
them more susceptible to either developing a neuropathy
or to having a recurrence of previous neurotoxicity.
JOURNAL OF CLINICAL ONCOLOGY

Downloaded from jco.ascopubs.org on January 4, 2016. For personal use only. No other uses without permission.
Copyright 2005 American Society of Clinical Oncology. All rights reserved.

Ixabepilone Phase II Trial in Breast Cancer

also help to predict response when examined in a pretreatment

specimen. Interestingly, we found that acetylated -tubulin
levels were higher at baseline in the tumor cells of patients
whose tumors responded or in patients with stable disease than
in the patient whose tumor did not respond. In the patient
with progressive disease, baseline levels of acetylated -tubulin
within the tumor were not detectable by immunohistochemistry. Thus, our results suggest that a tumor with inherent
microtubule stability might be more likely to respond to a
stabilizing agent with additional microtubule stabilization and
decreased proliferation. Although the small numbers of patients
biopsied in this group limits our ability to make more definitive
statements, these interesting observations are hypothesis generating and should be considered for additional study.
In summary, ixabepilone has shown activity in patients
with metastatic breast cancer who have been previously treated
with paclitaxel or docetaxel. For most patients, ixabepilone
was well tolerated with low hematologic toxicity, and minimal
or easily controlled nausea, vomiting, and diarrhea. In addition, administration of ixabepilone does not require premedication with corticosteroids, unlike paclitaxel and docetaxel,
which improves its tolerability. Neurotoxicity, a major concern for microtubule-stabilizing drugs, was relatively mild for
ixabepilone on this schedule. Its activity and safety profile
warrant additional development of this therapy.

Fig 5. Expression of acetylated and glu-terminated -tubulin and tumor

proliferation as detected by immunohistochemistry (IHC). Tumor levels of
(A) acetylated -tubulin, (B) glu-terminated -tubulin, and (C) Ki67 were
measured at baseline and cycle 2. The patient number is listed along the
x-axis. PR, partial response; SD, stable disease; PD, progressive disease.

Acknowledgment
We thank Diana Nguyen for immunohistochemistry
and A. Dimitrios Colevas, MD, for helpful discussions regarding the conduct of this trial.

However, because all patients were either considered

grade 0 or 1 at study entry, patients who already had a
significant unresolved neuropathy would not have been
enrolled. For similar reasons, a patient with a history of
significant peripheral neuropathy may have been less
willing to enroll onto this trial.
From the available paired tumor biopsies, we observed
increased levels of both glu-terminated and acetylated
-tubulin after treatment, indicating that ixabepilone stabilized microtubules in the target tissue. Measures of stabilized
microtubules may not only demonstrate a drug effect, but may

Authors Disclosures of Potential

Conflicts of Interest
The following authors or their immediate family members have indicated a financial interest. No conflict exists for
drugs or devices used in a study if they are not being evaluated
as part of the investigation. Consultant/Advisory Role: Adam
Brufsky, Bristol-Myers Squibb. Honoraria: Adam Brufsky,
Bristol-Myers Squibb. For a detailed description of these categories, or for more information about ASCOs conflict of interest policy, please refer to the Author Disclosure Declaration
and Disclosures of Potential Conflicts of Interest found in
Information for Contributors in the front of each issue.

REFERENCES
1. Bernard-Marty C, Cardoso F, Piccart MJ:
Use and abuse of taxanes in the management of
metastatic breast cancer. Eur J Cancer 39:19781989, 2003
2. Altmann KH: Microtubule-stabilizing agents:
A growing class of important anticancer drugs.
Curr Opin Chem Biol 5:424-431, 2001
3. Westermann S, Weber K: Post-translational
modifications regulate microtubule function. Nat
Rev Mol Cell Biol 4:938-947, 2003

4. Infante AS, Stein MS, Zhai Y, et al: Detyrosinated (Glu) microtubules are stabilized by an
ATP-sensitive plus-end cap. J Cell Sci 113:39073919, 2000 (pt 22)
5. Sasse R, Gull K: Tubulin post-translational
modifications and the construction of microtubular organelles in Trypanosoma brucei. J Cell Sci
90:577-589, 1988 (pt 4)
6. Mialhe A, Lafanechere L, Treilleux I, et al:
Tubulin detyrosination is a frequent occurrence
in breast cancers of poor prognosis. Cancer Res
61:5024-5027, 2001

7. Matsuyama A, Shimazu T, Sumida Y, et al:

In vivo destabilization of dynamic microtubules
by HDAC6-mediated deacetylation. Embo J 21:
6820-6831, 2002
8. Ross JL, Santangelo CD, Makrides V, et al:
Tau induces cooperative Taxol binding to microtubules. Proc Natl Acad Sci U S A 101:1291012915, 2004
9. Bollag DM, McQueney PA, Zhu J, et al:
Epothilones, a new class of microtubulestabilizing agents with a taxol-like mechanism of
action. Cancer Res 55:2325-2333, 1995

2733

www.jco.org

Downloaded from jco.ascopubs.org on January 4, 2016. For personal use only. No other uses without permission.
Copyright 2005 American Society of Clinical Oncology. All rights reserved.

Low et al

10. Altmann KH: Epothilone B and its analogs:

A new family of anticancer agents. Mini Rev
Med Chem 3:149-158, 2003
11. Lee FY, Borzilleri R, Fairchild CR, et al:
BMS-247550: A novel epothilone analog with a
mode of action similar to paclitaxel but possessing superior antitumor efficacy. Clin Cancer Res
7:1429-1437, 2001
12. Abraham J, Agrawal M, Bakke S, et al:
Phase I trial and pharmacokinetic study of BMS247550, an epothilone B analog, administered
intravenously on a daily schedule for five days.
J Clin Oncol 21:1866-1873, 2003
13. Therasse P, Arbuck SG, Eisenhauer EA, et
al: New guidelines to evaluate the response to
treatment in solid tumors: European Organization for Research and Treatment of Cancer,
National Cancer Institute of the United States,
National Cancer Institute of Canada. J Natl Cancer Inst 92:205-216, 2000
14. National Cancer Institute: Common Toxicity Criteria v 2.0 (April 30, 1999), Cancer Therapy
Evaluation Program. Bethesda, MD, National
Cancer Institute, 1999
15. Simon R: Optimal two-stage designs for
phase II clinical trials. Control Clin Trials 10:1-10,
1989
16. Tan AR, Yang X, Hewitt SM, et al: Evaluation of biologic end points and pharmacokinetics

2734

in patients with metastatic breast cancer after

treatment with erlotinib, an epidermal growth
factor receptor tyrosine kinase inhibitor. J Clin
Oncol 22:3080-3090, 2004
17. Thomas E, Taberno J, Fornier M, et al: A
phase II study of the epothilone B analog BMS247550 in patients with taxane-resistant metastatic breast cancer. Proc Am Soc Clin Oncol
22:8, 2003 (abstr 30)
18. Roche H, Cure H, Bunnell C, et al: A phase
II study of epothilone analog BMS-247550 in
patients with metastatic breast cancer previously treated with an anthracycline. Proc Am Soc
Clin Oncol 22:18, 2003 (abstr 69)
19. Chen T, Molina A, Moore S, et al: Epothilone
B analog (BMS-247550) at the recommended
phase II dose in patients with gynecologic and
breast cancers. J Clin Oncol 22:155S, 2004 (abstr
2115)
20. Sawaki M, Ito Y, Hashimoto D, et al:
Paclitaxel administered weekly in patients with
docetaxel-resistant metastatic breast cancer: A
single-center study. Tumori 90:36-39, 2004
21. Valero V, Jones SE, Von Hoff DD, et al: A
phase II study of docetaxel in patients with
paclitaxel-resistant metastatic breast cancer.
J Clin Oncol 16:3362-3368, 1998
22. Vansteenkiste J, Breton J-L, Sandler A, et
al: A randomized phase II study of epothilone

analog BMS-247550 in patients with non-small

cell lung cancer (NSCLC) who have failed firstline platinum-based chemotherapy. Proc Am Soc
Clin Oncol 22:626, 2003 (abstr 2519)
23. Mani S, McDaid H, Hamilton A, et al:
Phase I clinical and pharmacokinetic study of
BMS-247550, a novel derivative of epothilone B,
in solid tumors. Clin Cancer Res 10:1289-1298,
2004
24. Kelly WK, Galsky M, Small EJ, et al: Multiinstitutional trial of the epothilone B analogue
BMS-247550 with or without estramustine phosphate (EMP) in patients with progressive
castrate-metastatic prostate cancer (PCMPC):
Updated results. J Clin Oncol 22:384S, 2004
(abstr 4509)
25. Hussain M, Faulkner J, Vaishampayan U,
et al: Epothilone B (Epo-B) analogue BMS247550 (NSC #710428) administered every 21
days in patients (pts) with hormone refractory
prostate cancer (HRPC): A Southwest Oncology
Group Study (S0111). J Clin Oncol 22:384S, 2004
(abstr 4510)
26. Pavlick AC, Millward M, Farrell K, et al: A
phase II study of epothilone B analog (EpoB)BMS-247550 (NSC#710428) in stage IV malignant melanoma. J Clin Oncol 22:720S, 2004
(abstr 7542)

JOURNAL OF CLINICAL ONCOLOGY

Downloaded from jco.ascopubs.org on January 4, 2016. For personal use only. No other uses without permission.
Copyright 2005 American Society of Clinical Oncology. All rights reserved.