University of Southern California Dissertations and Theses

The effect of renal function on toxicity of E7389 (eribulin) among patients with bladder cancer

(USC Thesis Other)

The effect of renal function on toxicity of E7389 (eribulin) among patients with bladder cancer

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
The effect of renal function on toxicity of E7389 (eribulin) among patients with
bladder cancer

by

Weinan Wang

A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(APPLIED BIOSTATISTICS AND EPIDEMIOLOGY)

August 2016

2
Table of Contents
Abstract ............................................................................................................................... 3
Chapter 1. Background ....................................................................................................... 5
I. Introduction .......................................................................................................................... 5
II. Phase I portion of the study ............................................................................................... 7
III. Phase I portion of the study ........................................................................................... 11
IV. Purpose of this thesis ..................................................................................................... 12
Chapter 2. Methods ........................................................................................................... 13
I. Data collection .................................................................................................................... 13
II. Study population ............................................................................................................... 14
III. Statistical analysis ............................................................................................................ 15
A. Basic summary and descriptions .............................................................................. 15
B. Preliminary analysis of the serum creatinine value (CV) and the calculated
creatinine clearance (ClCr) ........................................................................................... 16
C. Assessing variables that were associated with toxicity ............................................ 16
Chapter 3. Results ............................................................................................................. 18
I. Demographic and clinical characteristics ........................................................................... 18
II. Analysis of serum creatinine value and the calculated creatinine clearance .................... 21
III. Exact logistic regression analyses of treatment toxicity .................................................. 26
A. Baseline associations with hematologic toxicity ...................................................... 26
B. Baseline associations with non-hematologic toxicity .............................................. 33
Chapter 4. Discussion ....................................................................................................... 39
Conclusion ........................................................................................................................ 42
References ......................................................................................................................... 43

3
ABSTRACT
Background Many patients affected with advanced bladder cancer are elderly and have
injured kidneys or kidneys with some degree of dysfunction, which precludes them from
receiving a standard cisplatin-based therapy. Thus, there is an unmet medical need for
treatment in patients who have renal dysfunction. The halichondrin B analog, E7389 (and
now called eribulin), is a candidate to fulfill this role, given its favorable toxicity profile
and promising activity in early clinical trials.
Objectives The main objective of this thesis was to assess the effect of renal function on
the toxicity of eribulin. We also tested the associations of demographic variables
including age, gender, and race and clinical characteristics such as serum creatinine and
calculated creatinine clearance with toxicity of eribulin.
Methods Data were available for 48 patients with bladder cancer. The association
between renal function (measured by calculated creatinine clearance) and the occurrence
of grade 3 or 4 toxicity of eribulin (categorized as hematologic and non-hematologic
toxicity) were evaluated using exact logistic regression. Covariates including age, race,

4
sex, serum creatinine value, calculated creatinine clearance, prior chemotherapy status
and treatment dose were also analyzed for inclusion as potential confounders in renal
function-hematologic toxicity or renal function-non-hematologic toxicity associations.
We also assessed the differences in serum creatinine value and calculated creatinine
clearance among three renal function groups using analysis of variance (ANOVA).
Results Patients’ renal function was not statistically significantly associated with
hematologic or non-hematologic toxicity (p>0.05). Demographic characteristics
including age, gender and race, and clinical characteristics such as serum creatinine and
prior chemotherapy were also not related with toxicity of eribulin (all p>0.05). Calculated
creatinine clearance was negatively associated with hematologic toxicity (p=0.05).
Conclusion Renal function seemed to have no effect on the toxicity of eribulin. Only
calculated creatinine clearance was associated with hematologic toxicity. Based on the
null findings of the study, a larger study will need to be conducted to test the associations
among patients’ renal function, calculated creatinine clearance and toxicity of eribulin.

5
CHAPTER 1. BACKGROUND
I. Introduction
Bladder cancer is the second most common cancer in the genitourinary system
1
.
More than 100,000 cases of muscle invasive or advanced disease are diagnosed per year
worldwide
2
. Over 90% of the cases are transitional cell carcinomas (TCC)
3,4
. Muscle
invasive disease that recurs or metastasizes after local therapy is considered a fatal
disease and nearly all deaths caused by bladder cancer result from metastatic disease
5
.
Patients with advanced disease have a very poor prognosis.
In recent years, bladder cancer has come to be called “cancer of the urothelium’.
Although this latter term also includes tumors that arise in the ureter and urethra, the
majority of cancers of the urothelium arise in the bladder. That said, all of the urothelial
cancers are often grouped together for clinical trials. Figure 1 shows the structures of the
urinary system where tumors arise
6
.

6
Fig. 1. Urinary bladder cancer

Cisplatin-based chemotherapy is regarded as a standard treatment for recurrent or
metastatic transitional cell carcinoma of the urothelium. However, many patients affected
with this disease are elderly and have injured kidneys or kidneys with some degree of
dysfunction, which precludes them from receiving a standard cisplatin-based therapy.
Thus, there is an unmet medical need for treatment in patients who have renal
dysfunction.

7
The halichondrin B analog, E7389 (and now called eribulin), is a candidate to fulfill
this role, given its favorable toxicity profile and promising activity in early clinical trials.
A group of institutions, including USC, conducted a clinical trial that evaluated the safety,
efficacy and side effects of E7389 (eribulin). This trial,“A Phase I / II Study of E7389
Halichondrin B analog (NSC # 707389; IND # 64395) in Metastatic Urothelial Tract
Cancer and Renal Insufficiency”, was a phase II clinical trial of E7389 with a safety
lead-in phase I study for patients with moderate and severe renal dysfunction. The study
had two portions, a Phase I and Phase II study.
II. Phase I portion of the study
The primary objective of the lead-in phase I portion was to establish the safety of
eribulin, when given to patients with moderate or severe renal dysfunction. Patients with
renal insufficiency (defined by their calculated creatinine clearance level) were classified
as having moderate renal dysfunction (calculated creatinine clearance between 40 and 59
mL/min) and severe renal dysfunction (calculated creatinine clearance <40 mL/min).
Eligible patients were administered eribulin on an outpatient basis as an intravenous

8
bolus over 1-2 minutes, once a week for two weeks in a row (on days 1 and 8) of a
21-day cycle. Patients with renal dysfunction were included in the phase II portion of the
study, once the phase I portion demonstrated that it was safe to treat them with 1.4
mg/m
2
/week on days 1 and 8 of a 21-day cycle. Since eribulin is metabolized primarily in
the liver
7
and no renal toxicity had been observed in phase I studies performed with this
agent prior to this Phase I portion, it was expected to be well tolerated in patients with
renal dysfunction at the maximum tolerated dose (MTD).
In the phase I portion of the study, the “standard” 3+3 design was modified to
eliminate waiting periods between dose levels since the clinical stability of patients was
limited
8
. Three dose levels (0.7 mg/m
2
/week, 1.0 mg/m
2
/week and 1.4 mg/m
2
/week) were
tested independently for each of two renal dysfunction groups. The 1.4 mg/m
2
/week dose
level was found to be well tolerated in patients with moderate renal dysfunction and in
patients with severe renal dysfunction.
In the moderate renal dysfunction group, 2 patients were treated with 0.7 mg/m
2
, and
then 3 patients were treated with 1.0 mg/m
2
; none of these patients experienced

9
unacceptable toxicity. Finally 6 patients with moderate renal dysfunction were given 1.4
mg/m
2
/week; since none of these patients experienced dose limiting toxicity (DLT), 1.4
mg/m
2
/week was declared to be safe in this group and the eligibility criteria for the phase
II trial was expanded to include patients with moderate renal dysfunction.
In the group of patients with severe renal dysfunction, 3 patients were treated at dose
0.7 mg/m
2
/week at the beginning of the study. Since none of these patients experienced
DLT, the dose was escalated to 1.0 mg/m
2
/week and 3 new patients were treated at
1.0mg/m
2
/week. Since none of these patients experienced DLT, dose was escalated to 1.4
mg/m
2
/week. Only 1 of 6 patients with severe renal dysfunction experienced a DLT, and
therefore 1.4 mg/m
2
/week was declared to be safe in this group and the eligibility criteria
for the phase II portion was expanded to include patients with severe renal dysfunction
See Table 1 for details.

10
Table 1. Phase I Summary
Renal
Insufficiency
Group
Dose Level
Patients
Entered
Patients
Eligible
Evaluable for
toxicity
Patients with
DLT
Moderate
0.7 mg/m2/week 2 2 2 0
1.0 mg/m2/week 3 3 3 0
1.4 mg/m2/week 6 6 6 0
Severe
0.7 mg/m2/week 3 3 3 0
1.0 mg/m2/week 4* 4* 3 0
1.4 mg/m2/week 6 6 6 1

*USC-020 never began treatment / did not receive any study drug and is not counted in
any of the other lists or tables in this document.

11
III. Phase II portion of the study
The objectives of the phase II portion of the study were: (1) determination of the
response rate and progression-free survival of patients with advanced urothelial
carcinomas treated with eribulin, and (2) documentation of the toxicity associated with
the administration of eribulin to patients with varying degrees of renal function. In this
thesis, we were concerned with the effects of eribulin on the bone marrow (hematologic
system) as well as on non-hematologic tissues and organs during the first course of
treatment. Toxicities were graded according to the NCI Common Terminology Criteria
for Adverse Events (CTCAE) v3.0. Hematologic toxicity and non-hematologic toxicity
mentioned in this thesis indicated grade 3 or grade 4 toxicity. In the Phase II portion of
this trial, only those patients who completed the first course of treatment with eribulin or
who experienced unacceptable toxicity, which was defined as any toxicity possibly
related to treatment that resulted in a > 2 week delay in treatment, were included in our
analysis (USC-020 never received any treatment and was deleted from the analysis set).

12
Toxicity data collected at the end of the first course of treatment are included in this
analysis.
IV. Purpose of this thesis
The main objective of this thesis was to assess the effect of renal function on the
toxicity of eribulin. In addition, we also tested the associations of demographics including
age, gender, race and clinical characteristics such as serum creatinine value, calculated
creatinine clearance etc. with the occurrence of eribulin-related toxicity.

13
CHAPTER 2. METHODS
I. Data collection
The data used in this study were obtained from the clinical phase I / II study of
eribulin. Our study included 48 patients enrolled in the clinical phase I / II study. Patient
demographic characteristics and clinical information during the first course of treatment
were obtained from the study database. Serum creatinine was measured at the time of
registration and used to calculate creatinine clearance by the modified Cockcroft and
Gault Formula. Three renal function categories were based on calculated creatinine
clearance (approximation of glomerular filtration rate, GFR). Information on toxicities
was collected at the end of the first course of treatment. Pharmacokinetic parameters of
eribulin were also obtained in our study; since only 13 patients had pharmacokinetic data,
we did not use these data for analysis. The relevant data from the clinical trial database
were transferred to an EXCEL file and a SAS database was created for analysis. Missing
values and outliers were confirmed with principal investigators before analysis.

14
II. Study population
Patients were recruited from four medical centers including: City of Hope National
Medical Center, Duarte CA (COH) and affiliates; USC/Norris Comprehensive Cancer
Center, Los Angeles (USC); University of California, Davis Cancer Center, Sacramento
(UCD) and affiliates; and University of Pittsburgh, Pittsburgh, PA (UOP). Eligible
participants were patients who developed locally measurable advanced or metastatic
urothelial cancer that is not amenable to surgical treatment. Patients in this study had
normal hepatic and marrow function with ECOG performance status 0-2 and Karnofsky ≥
60%. Patients were excluded if they received any other investigational agents or prior
therapy with E7389 Halichondrin analog (eribulin), or had brain metastases that were
unstable or untreated. Pregnant women and HIV-positive patients on combination
antiretroviral therapy were also ineligible.

15
III. Statistical analysis
A. Basic summary and descriptions
Baseline evaluations were conducted within 15 days prior to start of protocol
therapy and variables included age, gender, race, renal function, serum creatinine value,
and prior chemotherapy status were assessed at baseline. Patient race was dichotomized
as Caucasian and others. Renal function was categorized as normal renal function
(calculated creatinine clearance >60 mL/min), moderate renal dysfunction (calculated
creatinine clearance between 40 and 59 mL/min) and severe renal dysfunction (calculated
creatinine clearance <40 mL/min). Toxicity related to eribulin was dichotomized as
hematologic toxicity (toxicities occurred on blood/bone marrow system) and
non-hematologic toxicity (any other toxicities). Hematologic and non-hematologic
toxicities were graded according to CTCAE v3.0 and coded as 0 if maximum toxicity
grades were less than or equal to 2, and 1 if the grades were greater than 2 (i.e. 3 or 4).
Prior chemotherapy status was dichotomized as yes (had received prior chemotherapy) or
no (no prior chemotherapy).

16
B. Preliminary analysis of the serum creatinine value (CV) and the calculated
creatinine clearance (ClCr)
Prior to analysis, data were examined to (1) determine whether the data were
approximately normally distributed and (2) if not normally distributed, to find a
transformation that would result in data that were approximately normally distributed.
This was done so that we could use parametric statistical methods. We also assessed the
differences in serum creatinine value and calculated creatinine clearance among three
renal function groups (normal renal function group, moderate renal dysfunction group
and severe renal dysfunction group) using analysis of variance (ANOVA).
C. Assessing variables that were associated with toxicity
We used exact logistic regression to test whether the occurrence of grade 3 or 4
toxicity of eribulin differed by patients’ renal function. According to whether the toxicity
occurred on blood/bone marrow system or not, toxicity was categorized as hematologic
toxicity and non-hematologic toxicity. We ran two separate exact logistic regressions,
with two different outcomes (hematologic toxicity – yes/no, and non-hematologic

17
toxicity – yes/no). Univariate associations of covariates including age, race, sex, serum
creatinine value, calculated creatinine clearance, prior chemotherapy status and treatment
dose with toxicity were analyzed for inclusion as potential confounders in renal
function-hematologic toxicity or renal function-non-hematologic toxicity associations,
respectively. We used probability plots to describe the patterns of the univariate
associations of each continuous independent variable with hematologic and
non-hematologic toxicity.
Statistical analyses were conducted using SAS 9.4 (SAS Inc., Cary, NC). Two-sided
P value and significance level of 0.05 were used for all analyses.

18
CHAPTER 3. RESULTS
I. Demographic and clinical characteristics
Demographic and clinical characteristics of the 48 participants included in the
analysis are summarized in Table 2. The median and interquartile range (IQR) for age
were 69.5 (58.5-74.5) years; 75% of the patients were male (36). Patients were primarily
Caucasian (36, 75%); only 12 patients were other races including Asian, Black and others.
Of the 48 patients, 36 (75%) had normal renal function, 6 (12.5%) had moderate renal
dysfunction and 6 (12.5%) had severe renal dysfunction. The median (IQR) serum
creatinine value and calculated creatinine clearance of the 48 patients were 1.2 mg/dL
(0.9-1.6) and 52 mL/min (42-85), respectively. 31 patients had received prior
chemotherapy before the treatment with eribulin and 17 patients had not.

19
Table 2. Patient demographic and clinical characteristics (n=48)
Variables Category N (%) Median IQR Range
Age

48 69.5 (58.5-74.5) (37-86)
Gender Male 36 (75)

Female 12 (25)

Race Caucasian 36 (75)

Others
a
12 (25)

Renal Function
b
Normal 36 (75)

Moderate dysfunction 6 (12.5)

Severe dysfunction 6 (12.5)

Serum Creatinine
Value

48 1.2 (0.9-1.6) (0.5-2.6)

20
Calculated Creatinine
Clearance

48 52 (42-85) (22-123)
Prior Chemotherapy Yes 31 (35.4)

No 17 (64.6)

Abbreviations: IQR=interquartile range.
Note:
a. Others include Asian, Black and other races.
b. NCI stands for the renal function of patients.

21
II. Analysis of serum creatinine value and calculated creatinine clearance
The distribution of serum creatinine value was skewed to higher values ( Figure 2).
The mean (1.3 mg/dL) was greater than the median (1.2mg/dL). The result of
Shapiro-Wilk test was statistically significant (p=0.001). indicating that . The distribution
of calculated creatinine clearance was also skewed to higher values (Figure 3). The mean
(61.8 mL/min) was much larger than the median (52.5 mL/min). The result of
Shapiro-Wilk test was statistically significant (p=0.02). Therefore, variable
transformations were attempted to make the data normally distributed.
Fig. 2. The distribution and probability/ box plot for creatinine values

Abbreviation: CV=creatinine value

22
Fig. 3. The distribution and probability/ box plot for calculated creatinine clearance

Abbreviation: ClCr= calculated creatinine clearance
We attempted to find a Box-Cox transformation to normalize the univariate
histogram for both serum creatinine value and calculated creatinine clearance. For serum
creatinine, the result showed that convenient lambda was equal to 0, which indicated that
it was appropriate to make a log transformation to the original serum creatinine value.
For calculated creatinine clearance, the convenient lambda was equal to 0.25, which
indicated that we should transform the data with a fourth root. After transformation, the
distribution and probability plots and box plots for serum creatinine values and calculated
creatinine clearance are shown in Figures 4 and 5.

23
Fig. 4. The distribution and probability/ box plot for creatinine values after
transformation

Abbreviation: logCV=creatinine value in log scale
Although the log-transformed serum creatinine value was still not normally
distributed, the center of the distribution moved towards the mean and the distribution
became more symmetric and bell-shaped after transformation. The reason for the non
normal distribution could be due to certain patients with extreme serum creatinine values.

24
Fig. 5. The distribution and probability/ box plot for calculated creatinine clearance
after transformation

Abbreviation: tClCr=fourth root-transformed calculated creatinine clearance
After transformation, the distribution of fourth root-transformed creatinine clearance
became much more symmetric and bell-shaped. Even if the distribution seemed to be
bimodal, the mean (2.8) and the median (2.7) were very close. The values of skewness
and kurtosis were close to 0 (0.07, -0.98 respectively). These all indicated that the
distribution of creatinine clearance became much more normal after transformation.
However, certain patients with extreme values need to be confirmed with principal
investigators.

25
The results of ANOVA indicated that there was a statistically significant difference
in serum creatinine value (p<0.001) and calculated creatinine clearance (p<0.001) among
the three renal function groups. The results were confirmed by Kruskal-Wallis tests and
are shown in Table 3.
Table 3. Serum creatinine and calculated creatinine clearance value by renal
function

Normal renal
function (n=36)
Moderate renal
dysfunction (n=6)
Severe renal
dysfunction (n=6)
p value
Serum creatinine
value (mg/dL)
1.05 1.60 2.33 <0.001
Calculated
creatinine
clearance
(mL/min)
65.10 51.80 27.15 <0.001

Values=Median
p values=ANOVA

26
III. Exact logistic regression analyses of treatment toxicity
A. Baseline associations with hematologic toxicity (n=48 patients)
The associations between baseline demographic and clinical characteristics and the
occurrence of grade 3 or higher hematologic toxicity are shown in Table 4a Patients’
demographic characteristics including age, gender and race were all not statistically
significantly associated with hematologic toxicity (all p>0.05). Renal function was also
not related to hematologic toxicity (p=0.20) and we failed to reject the null hypothesis of
the Cochran-Armitage trend test over the three renal function groups (one sided exact
test, p=0.08, not shown in the table). However, since the association between creatinine
clearance and renal function was statistically significant (p<0.001) and the creatinine
clearance was negatively associated with hematologic toxicity (p=0.05), there was a
probability that patients with higher creatinine clearance (better renal function) tended to
have fewer or milder hematologic toxicity compared to patients with lower creatinine
clearance (OR: 0.98, 95% CI: 0.95-1.00). For one mL/min increase in calculated
creatinine clearance, the odds of developing grade 3 or 4 hematologic toxicity decrease

27
by 2%. Serum creatinine value and prior chemotherapy status were not significantly
associated with hematologic toxicity (p>0.05). For one year increase in age, the odds of
having grade 3 or 4 hematologic toxicity increased by 1% (OR: 1.01, 95% CI: 0.96, 1.07)
and for one mg/dL increase in serum creatinine, the odds of developing grade 3 plus
hematologic toxicity increased by 77% (OR: 1.77, 95% CI: 0.57, 5.50). To allow
comparability of beta estimates on the same scale, we transformed the continuous
variables of age, serum creatinine value and calculated creatinine clearance to standard
units, using the formula Y=(X-min)/ (max-min) and conducted univariate exact logistic
regression to get the standardized odds ratios. The results are shown in Table 4b. After
transformation, the odds ratios for age and serum creatinine value became larger (OR:
1.86, 95% CI: 0.14, 28.0; OR: 3.22, 95% CI: 0.30, 36.6, respectively); the odds ratio for
calculated creatinine clearance became smaller (OR: 0.09 95% CI: 0, 1.0). For one unit
increase in age and serum creatinine, the probability of developing grade 3 or higher
hematologic toxicity increase 86% and 222% respectively, while the odds of
experiencing grade 3 plus hematologic toxicity decrease by 1%. It seemed that age and

28
serum creatinine have more influence on hematologic toxicity compared with calculated
creatinine clearance, based on these standardized exact logistic regression coefficients,
but the associations were not statistically significant.

29
Table 4a. Baseline associations with hematologic toxicity (n=48)

Note:
a. Number of cases with grade 3+ hematologic toxicity. Toxicities were graded
Variables Category N
Median
IQR
d

Range
No. of
cases
a

β
coefficient
Odds ratio
(95% CI)
p value
Age 48 69.5 0.01 1.01 (0.96, 1.07) 0.66
(58.5-74.5)
(37.0-86.0)
Gender Male 36 14 1.0 1.0
Female 12 4 0.79 (0.15, 3.66)
Race Caucasian 36 11 1.0 0.17
Others
b
12 7 3.10 (0.68, 15.48)
Renal
Function
c

Normal 36 11 1.0 0.20

Moderate
dysfunction
6 4 4.37 (0.54, 55.08)

Severe
dysfunction
6 3 2.23 (0.26, 19.41)
Serum
Creatinine
Value
48 1.2 0.56 1.75 (0.57, 5.50) 0.33
(0.9-1.6)
(0.5-2.6)
Calculated
Creatinine
Clearance
48 52 -0.02 0.98 (0.95, 1.00) 0.05
(42-85)
(22-123)
Prior
Chemotherapy
Yes 31 11 1.0 0.93
No 17 7 1.27 (0.31, 5.0)

30
according to the NCI Common Terminology Criteria for Adverse Events (CTCAE)
v3.0.
b. Others include Asian, Black and other races.
c. NCI stands for the renal function of patients.
d. Abbreviation: IQR=interquartile range.

Table 4b. Baseline (transformed) associations with hematologic toxicity (n=48)
Variables N
Median
IQR
Range
β
coefficient
Odds Ratio
(95% CI)
p value
(Wald)
Age
48 0.66 0.62 1.86 (0.14, 28.0) 0.66

(0.44-0.77)

(0-1)

Serum
Creatinine Value
48 0.33 1.17 3.22 (0.30, 36.6) 0.33

(0.19-0.52)

(0-0.98)

Calculated
Creatinine
Clearance
48 0.30 -2.47 0.09 (0, 1.0) 0.05

(0.20-0.62)

(0-1)

Abbreviations: IQR=interquartile range.

31
Probability plots of age, serum creatinine value and calculated creatinine clearance
against hematologic toxicity are shown in Figures 6 to 8. The general reference ranges for
serum creatinine are 0.5-1.2mg/dL for adult males, and 0.4-1.1mg/dL for adult females.
We can see from Figure 7 that, as serum creatinine value went higher, the probability of
experiencing grade 3 or higher hematologic toxicity also increased. Therefore, when
serum creatinine values surpass the normal range, patients were more likely to experience
hematologic toxicities. From Figure 8, we see that as the creatinine clearance increased,
the probability of developing hematologic toxicity decreased, which indicated that
patients who had higher creatinine clearance were not likely to develop hematologic
toxicities compared with patients with lower creatinine clearance (p=0.05).

32
Fig. 6. Probability plot for age against maximum hematologic toxicity grade

Abbreviation: max_hem_gd=maximum hematologic toxicity grade

Fig. 7. Probability plot for serum creatinine value against maximum hematologic
toxicity grade

Abbreviations: CV=serum creatinine value; max_hem_gd=maximum hematologic
toxicity grade

33
Fig. 8. Probability plot for calculated creatinine clearance against maximum
hematologic toxicity grade

Abbreviations: ClCr=calculated creatinine clearance; max_hem_gd=maximum
hematologic toxicity grade
B. Baseline associations with non-hematologic toxicity (n=48 patients)
The associations between demographic and clinical characteristics and the
occurrence of grade 3 plus non-hematologic toxicity are shown in Table 5. None of the
independent variables were statistically significantly associated with non-hematologic
toxicity (all p>0.05). Cochran-Armitage trend test for renal function-non-hematologic
association was also not statistically significant. (one sided exact test, p=0.36, not shown
in the table). One possible reason could be that the number of patients who had

34
non-hematologic toxicities was very limited (only 7 of 48 patients), and therefore we did
not have enough power to determine the associations between patients’ baseline
characteristics and non-hematologic toxicity. Another reason could be that there was no
association between baseline characteristics and non-hematologic toxicity because of the
favorable toxicity profile of eribulin. Based on the data available for this analysis, we
were not able to determine which is the reason.

35
Table 5. Baseline associations with non-hematologic toxicity (n=48)

Note:
a. Number of cases with grade 3+ non-hematologic toxicity. Toxicities were graded
Variables Category N
Median
IQR
d

Range
No. of
cases
a

β
coefficient
Odds ratio
(95% CI)
p value
Age 48 69.5 0.02 1.02 (0.95, 1.11) 0.56
(58.5-74.5)
(37.0-86.0)
Gender Male 36 4 1.0 0.46
Female 12 3 2.60 (0.32, 18.76)
Race Caucasian 36 4 1.0 0.46
Others
b
12 3 2.60 (0.32, 18.76)
Renal
Function
c

Normal 36 6 1.0 0.36

Moderate
dysfunction
6 0 NA

Severe
dysfunction
6 1 1.0 (0.02, 11.76)
Serum
Creatinine
Value
48 1.2 -0.36 0.7 (0.12, 3.30) 0.70
(0.9-1.6)
(0.5-2.6)
Calculated
Creatinine
Clearance
48 52 -0.005 1.0 (0.96, 1.03) 0.80
(42-85)
(22-123)
Prior
Chemotherapy
Yes 31 4 1.0 0.96
No 17 3 1.44 (0.18, 9.84)

36
according to the NCI Common Terminology Criteria for Adverse Events (CTCAE)
v3.0.
b. Others include Asian, Black and other races.
c. NCI stands for the renal function of patients.
d. Abbreviation: IQR=interquartile range.

Probability plots of age, serum creatinine value and calculated creatinine clearance
against non-hematologic toxicity are shown in Figure 9 to 11. We see that regardless of
the value of the independent variable, the probability of experience grade 3 or higher
non-hematologic toxicity was always less than 25 percent. There were two possible
reasons that can be used to explain the low probability. One was that only 7 out of 48
patients experienced non-hematologic toxicity, we did not have enough power to detect
the potential associations. Another reason was that eribulin is not likely to cause
non-hematologic toxicities regardless of age, serum creatinine and calculated creatinine
clearance levels.

37
Fig. 9. Probability plot for age against maximum non-hematologic toxicity grade

Abbreviation: max_hem_gd=maximum hematologic toxicity grade

Fig. 10. Probability plot for serum creatinine value against maximum
non-hematologic toxicity grade

Abbreviations: CV=serum creatinine value; max_hem_gd=maximum hematologic
toxicity grade

38
Fig. 11. Probability plot for calculated creatinine clearance against maximum
non-hematologic toxicity grade

Abbreviations: ClCr=calculated creatinine clearance; max_hem_gd=maximum
hematologic toxicity grade

39
CHAPTER 4. DISCUSSION
The halichondrin B analog, E7389 (eribulin) has been tested for use in the treatment
of bladder cancer given its promising activity and hepatic clearance, especially for
patients with severe comorbidities and renal dysfunction. Our findings contribute to the
understanding of the toxicity of eribulin on patients with different levels of renal
function.
Toxicity of eribulin was dichotomized as hematologic toxicity and non-hematologic
toxicity in our study, based on whether the toxicities occurred on blood/bone marrow
system or not. We assessed the univariate associations between baseline characteristics
and the occurrences of grade 3 or grade 4 hematologic and non-hematologic toxicity,
seperately. Demographics including age, gender, and race were not statistically
significantly associated with either the occurrence of hematologic or the occurrence of
non-hematologic toxicity, which indicated that we couldn’t claim that demographics have
confounding effects on the renal function-hematologic or renal function-non-hematologic
toxicity associations. Clinical characteristics such as prior chemotherapy status, serum

40
creatinine value, and a categorical variable of renal function (normal, moderate
dysfunction, severe dysfunction) were also not related to hematologic or non-hematologic
toxicity (all p>0.05). Only calculated creatinine clearance was negatively associated with
hematologic toxicity (p=0.05).
The categorical variable of renal function was defined based on calculated creatinine
clearance level and was not statistically significantly associated with hematologic or
non-hematologic toxicity (p=0.20, p=0.36, respectively). However, calculated creatinine
clearance was negatively associated with hematologic toxicity (p=0.05). For one mL/min
increase in calculated creatinine clearance, the odds of developing grade 3 or 4
hematologic toxicity decrease by 2% (OR: 0.98, 95% CI: 0.95, 1.00). One of the possible
reasons was that there were too few patients with renal insufficiency who experienced
hematologic or non-hematologic toxicity, thus, we did not have enough power to detect
the potential association between renal function and toxicity. Another reason could be the
inaccuracy of data, which due to the categorization of renal function was not according to
the criteria strictly in the phase I/II study.

41
There were some limitations in our study. Firstly, the sample size of 48 patients
wassmall, severely limiting power to detect associations between baseline
characteristics and toxicity. Another limitation was that patients’ renal function was not
categorized strictly based on creatinine clearance criteria that we mentioned above. We
re-categorized renal function using the same data from the phase I/ II study according to
the same criteria. We got 5 of 21 patients with normal renal function, 8 of 16 patients
with moderate renal dysfunction and 5 of 11 patients with severe renal dysfunction
experienced hematologic toxicity, respectively, rather than 11 of 36 normal renal function
patients, 3 of 6 moderate renal dysfunction patients and 4 of 6 severe renal dysfunction
patients who developed hematologic toxicity. However, the results of exact logistic
regression remained the same (p=0.20). A final limitation was the analysis of toxicity
only after the first course of treatment. Given that some toxicity reactions may have a
longer incubation period, we should consider including patients who completed 3 to 4
courses of treatment in the toxicity analysis.

42
CONCLUSION
There was no statistically significant association between renal function and toxicity
of eribulin. Patients’ demographic characteristics including age, gender and race and
clinical characteristics such as prior chemotherapy status and serum creatinine value were
also not significantly related with toxicity of eribulin. However, although we could not
find statistically significant difference in the occurrence of grade 3 or higher hematologic
or non-hematologic toxicity among different renal function groups, calculated creatinine
clearance was negatively associated with hematologic toxicity. Based on the null findings
of the study, a larger study needs to be conducted to test the associations among patients’
renal function, calculated creatinine clearance and toxicity of eribulin.

43
REFERENCES
1. Advanced Bladder Cancer (ABC) Meta-analysis Collaboration. "Adjuvant
chemotherapy in invasive bladder cancer: a systematic review and meta-analysis of
individual patient data Advanced Bladder Cancer (ABC) Meta-analysis
Collaboration." European urology 48.2 (2005): 189.
2. Vale, C. L. "Advanced bladder cancer meta-analysis collaboration. Adjuvant
chemotherapy for invasive bladder cancer." Cochrane Database Syst. Rev (2006).
3. Boyle, H., A. Fléchon, and J. P. Droz. "Treatment of uncommon malignant tumours
of the bladder." Current opinion in urology 21.5 (2011): 309-314.
4. Ismaili, N., et al. "Small cell cancer of the bladder: pathology, diagnosis, treatment
and prognosis." Bulletin du cancer 96.6 (2009): 10030-10044.
5. Stein JP, Lieskovsky G, Cote R, Groshen S, Feng AC, Boyd S, . . . Skinner DG.
Radical cystectomy in the treatment of invasive bladder cancer: long-term results in
1,054 patients. J Clin Oncol. 2001;19:666-75.
6. http://www.epainassist.com/images/Article-Images/Urinary-Bladder-cancer.jpg
7. Synold, T. W., et al. "A phase I pharmacokinetic and target validation study of the

44
novel anti-tubulin agent E7389: a California Cancer consortium trial." ASCO Annual
Meeting Proceedings. Vol. 23. No. 16_suppl. 2005.
8. A Phase I / II Study of E7389 Halichondrin B analog (NSC # 707389; IND #
64395) in Metastatic Urothelial Tract Cancer and Renal Insufficiency

Abstract (if available)

Abstract Background: Many patients affected with advanced bladder cancer are elderly and have injured kidneys or kidneys with some degree of dysfunction, which precludes them from receiving a standard cisplatin-based therapy. Thus, there is an unmet medical need for treatment in patients who have renal dysfunction. The halichondrin B analog, E7389 (and now called eribulin), is a candidate to fulfill this role, given its favorable toxicity profile and promising activity in early clinical trials. ❧ Objectives: The main objective of this thesis was to assess the effect of renal function on the toxicity of eribulin. We also tested the associations of demographic variables including age, gender, and race and clinical characteristics such as serum creatinine and calculated creatinine clearance with toxicity of eribulin. ❧ Methods: Data were available for 48 patients with bladder cancer. The association between renal function (measured by calculated creatinine clearance) and the occurrence of grade 3 or 4 toxicity of eribulin (categorized as hematologic and non-hematologic toxicity) were evaluated using exact logistic regression. Covariates including age, race, sex, serum creatinine value, calculated creatinine clearance, prior chemotherapy status and treatment dose were also analyzed for inclusion as potential confounders in renal function-hematologic toxicity or renal function-non-hematologic toxicity associations. We also assessed the differences in serum creatinine value and calculated creatinine clearance among three renal function groups using analysis of variance (ANOVA). ❧ Results: Patients’ renal function was not statistically significantly associated with hematologic or non-hematologic toxicity (p>0.05). Demographic characteristics including age, gender and race, and clinical characteristics such as serum creatinine and prior chemotherapy were also not related with toxicity of eribulin (all p>0.05). Calculated creatinine clearance was negatively associated with hematologic toxicity (p=0.05). ❧ Conclusion: Renal function seemed to have no effect on the toxicity of eribulin. Only calculated creatinine clearance was associated with hematologic toxicity. Based on the null findings of the study, a larger study will need to be conducted to test the associations among patients’ renal function, calculated creatinine clearance and toxicity of eribulin.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Eribulin in advanced bladder cancer patients: a phase I/II clinical trial

PDF

An analysis of disease-free survival and overall survival in inflammatory breast cancer

PDF

Statistical analysis of a Phase II study of AMG 386 versus AMG 386 combined with anti-VEGF therapy in patients with advanced renal cell carcinoma

PDF

Cross-sectional association of blood pressure, antihypertensive medications, MRI volumetric measures and cognitive function scores in an aging population

PDF

Origins of the gender disparity in bladder cancer risk: a SEER analysis

PDF

Evaluation of factors influencing Los Angeles Tiered-Dispatch System’s improvement on bystander CPR rate and inter reliability between electronic patient care report (ePCR) and 911 call review on...

PDF

Incidence and survival rates of the three major histologies of renal cell carcinoma

PDF

Relationship of blood pressure and antihypertensive medications to cognitive change in the BVAIT, WISH, and ELITE clinical trials

PDF

Disparities in colorectal cancer survival among Latinos in California

PDF

Association of carotid artery stiffness with measures of cognition in older adults

PDF

Predictive factors of breast cancer survival: a population-based study

PDF

Long-term effects on vertebral body height growth of dose sculpting intensity modulated radiation therapy for children with neuroblastoma

PDF

Diet quality and pancreatic cancer incidence in the multiethnic cohort

PDF

Insulin’s effect on lactate levels in extremely low birth weight neonates. a multi-center, observational study

PDF

Factors influencing the decision and timing to undergo breast reconstruction after mastectomy for breast cancer in public hospital vs. private medical center from 2007 to 2013: a retrospective co...

PDF

Instability of heart rate and rating of perceived exertion during high-intensity interval training in breast cancer patients undergoing anthracycline chemotherapy

PDF

Carboplatin and vincristine chemotherapy for progressive low grade gliomas in pediatric patients with or without neurofibromatosis type 1 (NF1)

PDF

Development and validation of survey instrument designed for cervical cancer screening in Malawi, and other low resource settings

PDF

Associations between isoflavone soy protein (ISP) supplementation and breast cancer in postmenopausal women in the Women’s Isoflavone Soy Health (WISH) clinical trial

PDF

Racial/ethnic differences in colorectal cancer patient experiences, health care utilization and their association with mortality: findings from the SEER-CAHPS data

Asset Metadata

Creator Wang, Weinan (author)

Core Title The effect of renal function on toxicity of E7389 (eribulin) among patients with bladder cancer

School Keck School of Medicine

Degree Master of Science

Degree Program Applied Biostatistics and Epidemiology

Publication Date 08/02/2016

Defense Date 07/26/2016

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag bladder cancer,E7389,eribulin,OAI-PMH Harvest,renal function,toxicity

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Groshen, Susan (committee chair), Gauderman, James (committee member), Mack, Wendy (committee member)

Creator Email wangweinanwwn@gmail.com,weinan@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-290776

Unique identifier UC11281208

Identifier etd-WangWeinan-4702.pdf (filename),usctheses-c40-290776 (legacy record id)

Legacy Identifier etd-WangWeinan-4702-1.pdf

Dmrecord 290776

Document Type Thesis

Format application/pdf (imt)

Rights Wang, Weinan

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA