1. health and fitness
2. disease
3. cancer

VOLUME
22
䡠
NUMBER
15
䡠
AUGUST
1
2004
JOURNAL OF CLINICAL ONCOLOGY
O R I G I N A L
R E P O R T
Safety, Pharmacokinetics, and Activity of ABX-EGF, a
Fully Human Anti–Epidermal Growth Factor Receptor
Monoclonal Antibody in Patients With Metastatic Renal
Cell Cancer
Eric K. Rowinsky, Garry H. Schwartz, Jared A. Gollob, John A. Thompson, Nicholas J. Vogelzang,
Robert Figlin, Ronald Bukowski, Naomi Haas, Pamela Lockbaum, Yu-Ping Li, Rosalin Arends,
Kenneth A. Foon, Gisela Schwab, and Janice Dutcher
From the Institute for Drug Development, Cancer Therapy and Research
Center and Brook Army Medical
Center, San Antonio, TX; Beth Israel
Deaconess Cancer Center, Boston, MA;
University of Washington, Seattle, WA;
University of Chicago Cancer Research
Center, Chicago, IL; Cleveland Clinic,
Cleveland, OH; Fox Chase Cancer Center, Philadelphia, PA; UCLA School of
Medicine, Los Angeles; and Abgenix
Inc, Fremont, CA; and Our Lady of
Mercy Cancer Center, New York
Medical College, Bronx, NY.
Submitted November 11, 2003; accepted
March 31, 2004.
Presented in part at the 38th annual
meeting of the American Society of
Clinical Oncology, Orlando, FL,
May 18-21, 2002.
Authors’ disclosures of potential conflicts of interest are found at the end of
this article.
Address reprint requests to Eric K.
Rowinsky, MD, Institute for Drug Development, Cancer Therapy and Research Center, 7979 Wurzbach Rd, 4th
Floor, Zeller Building, San Antonio, TX
78229; e-mail: erowinsk@idd.org.
A
B
S
T
R
A
C
T
Purpose
To determine the antitumor activity of ABX-EGF, a fully human monoclonal antibody to the epidermal
growth factor receptor (EGFr), in previously treated patients with metastatic renal cell carcinoma, and to
characterize its toxicity, immunogenicity, pharmacokinetics, and pharmacodynamics.
Patients and Methods
The antitumor activity, as well as the toxicity, pharmacokinetics, pharmacodynamics, and immunogenicity of ABX-EGF, were assessed.
Results
Eighty-eight patients were treated with ABX-EGF doses of 1.0, 1.5, 2.0, or 2.5 mg/kg weekly with no
loading dose. EGFr immunostaining was performed on 76 tumor biopsy specimens (86%), and 69 (91%)
scored positive. Major responses occurred in three patients, and two patients had minor responses.
Forty-four patients (50%) also had stable disease at their first 8-week assessment, and the median
progression-free survival (PFS) was 100 days (95% CI, 58 to 140 days). Low hemoglobin and high
alkaline phosphatase predicted for short PFS. The principal toxicity, an acneiform rash, occurred in 68%,
95%, 87%, and 100% of patients who received at least three doses of ABX-EGF at 1.0, 1.5, 2.0, and
2.5 mg/kg/wk, respectively. A trend indicated that the severity of the rash may relate to PFS. No human
antihuman antibodies were detected. ABX-EGF pharmacokinetics fit a model that incorporated both
linear and saturable EGFr-mediated clearance mechanisms, and interindividual variability was low. At
2.5 mg/kg/wk, ABX-EGF concentrations throughout treatment exceeded those estimated to saturate
nonlinear clearance and inhibit xenograft growth by 90%.
Conclusion
ABX-EGF was generally well tolerated. The objective response rate was low in previously treated
patients with metastatic renal cell carcinoma. Although skin rash may be a pharmacodynamic marker of
drug action, its potential as a surrogate marker of clinical benefit requires further evaluation.
J Clin Oncol 22:3003-3015.
0732-183X/04/2215-3003/$20.00
DOI: 10.1200/JCO.2004.11.061
INTRODUCTION
The binding of epidermal growth factor
(EGF) or transforming growth factor-alpha
(TGF␣) to the epidermal growth factor receptor (EGFr) seems to trigger autophosphorylation and internalization of the
receptor, which culminate in mitogenic signal transduction.1,2 The notion of targeting
EGFr as a therapeutic strategy against cancer
is supported by experimental evidence that
aberrations of EGFr-mediated signal transduction play critical roles in tumorogenesis
and cancer cell proliferation.1,2 The relevance of this strategy is further substantiated
by EGFr expression, overexpression, or aberrant function in renal, lung, prostate,
breast, head and neck, colorectal, and other
epithelial malignancies, as well as evidence
that the level of EGFr expression may be a
3003
Downloaded from jco.ascopubs.org on June 9, 2014. For personal use only. No other uses without permission.
Copyright © 2004 American Society of Clinical Oncology. All rights reserved.
Rowinsky et al
determinant of tumor proliferation, invasiveness, and angiogenesis, and relates adversely to prognosis.3-14 Concomitant with EGFr expression, EGFr ligands, particularly
TGF␣, are commonly upregulated, and therefore, blockade
of autocrine loops is a rational strategy for the treatment of
many epithelial malignancies.15-18
ABX-EGF (Abgenix Inc, Fremont, CA; Immunex Corp, a
subsidiary of Amgen Inc, Thousand Oaks, CA), a high-affinity
(Kd, 5 ⫻ 10⫺11M) fully human immunoglobulin G2 (IgG2)
monoclonal antibody against human EGFr, was generated
from human antibody-producing XenoMouse strains
(Abgenix Inc), which were engineered to be deficient in mouse
antibody production and contain integrated megabase-sized
fragments from the human heavy and ␬ light chain loci with
the majority of the human antibody repertoire.19 ABX-EGF
completely blocks binding of EGF and TGF␣ to the EGFr and
induces profound and rapid internalization of the receptor in
EGFr-expressing human cancers.19,20 These actions abolish
EGFr-dependent cellular responses, including EGFr tyrosine
phosphorylation, extracellular acidification, angiogenesis, and
cell proliferation.19,20 The treatment of athymic mice with
ABX-EGF 0.2 mg intraperitoneally twice weekly for 3 weeks
completely prevents the formation of EGFr-overexpressing
human epidermoid carcinoma A431 xenografts.19,20 Furthermore, administration of ABX-EGF at doses as low as 0.3 mg
intraperitoneally for 3 weeks without concomitant chemotherapy or radiation completely eradicates tumors as large as
1.2 cm3, and a total dose of 0.6 mg completely eradicates
tumors in 65% of mice.19,20 Although similar results have been
reported with other antibodies and therapeutics against EGFr,
the results with ABX-EGF are unique in that no tumor recurrences have been noted for more than 8 months after the last
antibody injection in a sizable proportion of animals.16,17,21-24
Similar therapeutic effects have been observed following treatment of a wide range of human tumor xenografts with variable
EGFr expression levels ranging from 17,000 to 1,000,000 receptors/cell using renal, breast, pancreatic, ovarian, prostate,
and colorectal cancer cell lines.19 Furthermore, ABX-EGF has
not only been demonstrated to inhibit growth of other cancers
that express high levels of EGFr, such as A431 and MDA-MB468, but also to inhibit growth of other cancers with much
lower levels of EGFr expression.19,20 However, the magnitude
of tumor growth inhibition achieved following ABX-EGF
treatment seems to relate to the EGFr number; significant
growth inhibition of xenografts expressing at least 17,000 receptors per cell has been noted, whereas xenografts expressing
11,000 or fewer receptors per cell seem not to be affected by
ABX-EGF treatment.20
Toxicology, efficacy, and pharmacologic studies of
ABX-EGF in mice and nonhuman primates have been used
to simulate the pharmacokinetic profiles in humans and
predict effective dosing regimens.25,26 Based on the pharmacokinetics of ABX-EGF in monkeys treated with ABXEGF doses ranging from 0.3 to 30 mg/kg, human clearance
3004
(CL) was predicted to be nonlinear, possibly due to progressive saturation of an EGFr sink.25 Allometric scaling was
used to determine a dose-schedule for a phase I study, in
which patients with malignancies known to express EGFr
were treated with four weekly doses of ABX-EGF.26 The
most common toxicity was a dose-dependent skin rash,
occurring in 100% of patients treated with ABX-EGF at
2.0 mg/kg/wk or 2.5 mg/kg/wk. No allergic reactions, human antihuman antibody (HAHA) formation, nor serious
adverse events were reported at doses up to 2.5 mg/kg/wk,
which encompassed the dosing range evaluated in the
present study. Furthermore, the interpatient variability in
ABX-EGF exposure was low, and trough concentrations at
doses of at least 2 mg/kg exceeded IC90 values derived from
xenograft studies.19,20,25,26 This phase I study had not yet
been completed when the present study began.
The results of the aforementioned studies, as well as a
high (70% to 90%) incidence of EGFr expression in patients
with renal cell carcinoma, served as the rationale for this
study. The principal objectives were to evaluate the antitumor activity of multiple weekly doses of ABX-EGF in patients with metastatic renal carcinoma who either failed to
respond to treatment with interleukin-2 (IL-2) and/or
interferon-alfa (IFN-␣) or in subjects in whom this therapy
was felt to be contraindicated from a medical standpoint, as
well as to characterize the safety, immunogenicity, and
pharmacokinetics of ABX-EGF on a multidose schedule.
PATIENTS AND METHODS
Study Objectives
The study was designed as a multicenter, dose-rising trial to
principally evaluate the antitumor activity of multiple doses of
ABX-EGF in patients with metastatic renal cell carcinoma. The
toxicity, pharmacokinetics, pharmacodynamics, and immunogenicity of ABX-EGF were also assessed. Enrollment was planned for
successive cohorts of 20 patients treated at 1.0, 1.5, 2.0, and
2.5 mg/kg weekly with no testing or loading dose. After each series
of eight weekly treatments, antitumor activity was evaluated, and
stable or responding patients were able to receive as many as five
series of eight weekly treatments.
Eligibility
Subjects in whom IL-2 or IFN-␣ were felt to be contraindicated from a medical standpoint, or patients with histologically
confirmed, metastatic renal cell carcinoma who had previously
received and failed treatment with IL-2 or IFN-␣ in the metastatic
setting, were eligible for this study. Patients who received chemotherapy and radiation in the adjuvant and/or metastatic settings
were also eligible, but subjects who had been previously treated
with EGFr-targeted therapeutics were ineligible. Although the eligibility requirements did not mandate documentation of EGFr
expression before enrollment, the study required the availability of
tumor tissue to characterize EGFr immunostaining. Eligibility
criteria also included: age ⱖ 18 years; an Eastern Cooperative
Oncology Group performance status less than 2; no prior anticancer treatment of any type or investigational agents within 30 days;
JOURNAL OF CLINICAL ONCOLOGY
Downloaded from jco.ascopubs.org on June 9, 2014. For personal use only. No other uses without permission.
Copyright © 2004 American Society of Clinical Oncology. All rights reserved.
ABX-EGF in Renal Cancer
no hypercalcemia; adequate hematopoietic (absolute neutrophil
count ⱖ 1,500/␮L, platelet count ⱖ 100,000/␮L), hepatic (total
bilirubin ⱕ 1.5 mg/dL, transaminases and alkaline phosphatase
ⱕ 3 times the institutional upper normal limit), renal (creatinine
ⱕ 2.2 mg/dL), and cardiac (left ventricular ejection fraction
ⱖ 45% as measured by multiple gated acquisition [MUGA] scan)
functions; no prior treatment with anthracyclines or related therapeutics; no myocardial infarction within 1 year; measurable disease; and no coexisting medical problem of sufficient severity to
limit compliance with the study. Patients with a history of brain
metastases were eligible as long as there was no evidence of progressive disease or peritumoral edema on a magnetic resonance or
computed tomography scan of the brain performed before treatment, symptomatic manifestations of brain metastases, nor requirement for corticosteroids. Patients with a history of other
malignancies, except basal cell carcinoma of the skin and cervical
carcinoma-in-situ, within 5 years before study, were ineligible.
All patients of childbearing potential were required to use
appropriate contraception during treatment. Patients gave
written informed consent according to Federal and institutional guidelines before enrollment.
Immunohistochemistry
Paraffin-embedded sections and fresh tumor tissue from tumor biopsies were assessed in a central laboratory by IMPATH
(Los Angeles, CA) to evaluate EGFr expression. The immunostaining procedure used an EGFr immunohistochemistry kit developed by Dako Cytomation Inc (Carpinteria, CA) in
collaboration with Abgenix Inc for investigational purposes, and a
two-step immunostaining procedure was routinely used to process paraffin-embedded tissues. Following incubation with the
primary rabbit antibody to human EGFr protein, a ready-to-use
visualization reagent based on dextran technology was used. This
reagent consisted of both secondary goat antirabbit antibody and
horseradish peroxidase molecules linked to a common dextran
polymer backbone, thereby eliminating the need for sequential
application of link antibody and peroxidase-conjugated antibody.
The enzymatic conversion of the subsequently added chromogen
resulted in the formation of a visible reaction product at the
antigen site. The specimen was then counterstained, and a coverslip was applied. Control slides containing formalin-fixed,
paraffin-embedded human cell lines with various levels of EGFr
expression were provided to validate staining runs. EGFr expression was scored using light microscopy as follows: 0 (no staining),
1⫹ (weak staining of tumor cells), 2⫹ (moderate staining of
tumor cells), and 3⫹ (strong staining of tumor cells). Staining was
scored as positive if the stain was visualized within the cell membrane and/or cytoplasmic rim. In addition, an EGFr H score,
which reflected both the intensity of staining and the percentage of
tumor cells with the various immunostaining intensity scores, was
calculated for each biopsy sample by adding the products of
percentages of cells with weak, moderate, and strong immunostaining and weighting factors of 1, 2, and 3, respectively (ie, H
scores ⫽ [% tumor cells with weak intensity ⫻ 1] ⫹ [% tumor
cells with moderate intensity ⫻ 2] ⫹ [% tumor cells with strong
intensity ⫻ 3]), so that the minimum and maximum H scores
were 0 and 300, respectively.
Drug Administration
ABX-EGF was supplied by Abgenix Inc in 5-mL glass vials
containing 50 mg of antibody at a concentration of 10 mg/mL in
pyrogen-free 0.9% sodium chloride solution US Pharmacopeia
(USP). The vials were stored at 2 to 8°C. The appropriate volume
of the contents of the ABX-EGF vial, which was calculated according to dose level assignment, was diluted with 0.9% saline solution
USP in a polyvinyl chloride infusion bag to yield an infusion
volume ranging from 100 to 150 mL. The solution was infused
intravenously over 60 minutes using a syringe or infusion pump
with a 0.22 ␮m filter in the intravenous line. Vital signs were
recorded from 30 minutes before treatment, to 60 minutes posttreatment.
Pretreatment and Follow-Up Evaluations
Histories that included recording of performance status and
concurrent medications, physical examination, MUGA scan,
HAHA studies, an ECG, archived tumor tissues, and routine laboratory studies, were obtained before treatment. An interval history with recording of adverse events and a physical examination
were repeated weekly. The physical examination included a dermatologic assessment, which consisted of descriptions of the type
and location of lesions, related symptoms, and specific treatment
rendered, as well as a quantitative assessment of proportion of skin
affected, all of which was recorded on a designated form. Cutaneous manifestations were also photographed. Routine laboratory
studies were performed every 4 weeks, and MUGA scans and
HAHA studies were performed after each 8-week course. Routine
laboratory studies included a CBC and differential WBC count,
chemistries, coagulation studies, and cardiac enzymes. Toxicity
was graded according to the National Cancer Institute Common
Toxicity Criteria (NCI CTC; version 2.0). Although a prior phase I
study had confirmed the safety of ABX-EGF in the dosing range
evaluated in the present study, definitions for dose-limiting toxicity (DLT) and maximum-tolerated dose were established a priori.
Dose-escalation was to cease if a maximum-tolerated dose, which
was defined as the dose at which DLT occurred in at least 30% of
patients (with at least two events), was observed in the first 8-week
series of treatment. DLT was defined as any NCI CTC grade 3 or 4
adverse experience (except skin rash), any severe or lifethreatening complication not encompassed in the NCI CTC
guidelines; severe skin toxicity defined as desquamation, generalized urticaria, or symptomatic skin-related toxicity requiring narcotics or systemic corticosteroids; or event felt to be intolerable by
the patient. Dose reduction to 1.5 mg/kg/wk was permitted for
patients who developed severe skin-related toxicity following resolution of the toxic event within 3 weeks after dose interruption.
Relevant radiological studies for evaluation of all measurable
or assessable sites of malignancy were performed pretreatment,
after each 8-week course, and as needed to confirm response.
Patients were able to continue treatment if they did not develop
progressive disease. The response to treatment was assessed using
the Response Evaluation Criteria in Solid Tumors criteria. All
objective tumor responses were confirmed 4 weeks after response
diagnosis and centrally verified by an independent radiologist.
Statistical Analysis
Descriptive statistics of relevant demographic and clinical
features were compiled. Summary statistics were provided for
tumor response assessment at the end of the initial 8-week cycle of
treatment. The relationship between time to disease progression
and each of the pertinent factors was analyzed using univariate
survival analysis (the log-rank test for categoric variables and a
score test based on the Cox proportional hazards model for continuous variables). Kaplan-Meier estimates, median values, 95%
CIs, and both 25% and 75% quartile values were calculated. The
3005
www.jco.org
Downloaded from jco.ascopubs.org on June 9, 2014. For personal use only. No other uses without permission.
Copyright © 2004 American Society of Clinical Oncology. All rights reserved.
Rowinsky et al
rates of adverse events as a function of dose were determined. All
analyses and summaries were based on the modified intent-totreat population, defined as patients who received at least one dose
of ABX-EGF.
The relationships between ABX-EGF dose and the incidence of cutaneous toxicity were described using the sigmoidal
maximum effect (Emax) model of drug action (ie, incidence of
cutaneous toxicity ⫽ Emax ⫻ Dose␥/(ED50␥ ⫹ Dose␥), where
the Emax was fixed at 100%, and ED50 was the dose at which the
effect is 50% of the maximal effect. The exponent ␥ is a constant
that describes the sigmoidicity or steepness of the curve.27
From the ED50, the ABX-EGF dose predicted to yield a 90%
incidence of skin rash (ED90) was derived as a secondary parameter.
Pharmacokinetics and Immunogenicity
To measure the relevant parameters indicative of exposure to
ABX-EGF, blood samples were collected within 30 minutes before
and 30 minutes following each of the first eight weekly infusions of
ABX-EGF, and every 4 weeks thereafter. The samples were allowed
to clot for 30 to 120 minutes followed by centrifugation at 3,000
rpm for 15 minutes. The serum was transferred to a sample tube,
which was frozen at ⫺80°C until assayed. Serum ABX-EGF concentrations were measured using a validated electrochemiluminescence assay based on Origen technology (IGEN International
Inc, Gaithersburg, MD) with a lower limit of quantitation (LLOQ)
of 0.079 ␮g/mL. Samples with ABX-EGF concentrations below the
LLOQ of the electrochemiluminescence assay were considered
zero in calculating mean concentration values. An enzyme-linked
immunosorbent assay was used to detect the formation of HAHA
in serum sampled immediately before each 8-week course, and 4
weeks after the eighth dose of the last course.
Noncompartmental analysis of the ABX-EGF serum concentration data was accomplished using WINNonlin Professional,
Version 4.0 (Pharsight Corp, Mountain View, CA) to estimate the
steady-state area under the serum concentration-time curve for
the weekly dosing interval (AUC0-7 days) after the seventh dose. If
the data for concentration (C; Cmax and Ctrough) were incomplete
for the seventh dose, the steady-state AUC0-7 days of an earlier dose,
as close as possible to the seventh dose, was calculated. To ensure
steady-state condition, this alternate dose interval had to be after
the third dose at the 1.0-mg/kg dose level, after the fourth dose at
the 1.5-mg/kg dose level, and after the fifth dose at the 2.0- and
2.5-mg/kg dose levels. AUC0-7 days values were calculated from
time 0 to ␶ using the linear trapezoidal rule, where ␶ is the time (7
days) at the end of the dosing interval. The systemic CL was
calculated by dividing the weekly dose by AUC0-7 days.
The steady-state ABX-EGF Cmax ⫾ SEM values were calculated based on the average values for doses three to eight in the
1.0-mg/kg dose group, doses four to eight in the 1.5-mg/kg dose
group, and doses five to eight for the 2.0-mg/kg and 2.5-mg/kg
dose groups. The steady-state ABX-EGF Ctrough ⫾ SEM values
were calculated based on the average values for doses three to
seven in the 1.0-mg/kg dose group, doses four to seven in the
1.5-mg/kg dose group, and doses five to seven for the 2.0-mg/kg
and 2.5-mg/kg dose groups.
The serum ABX-EGF concentration profile was described by
a two-compartment model. Drug was introduced into the central
compartment from which elimination occurred in parallel by
first-order elimination (linear kinetics) and capacity-limited elimination (nonlinear kinetics). Distribution of ABX-EGF was described by interchange of drug between the central compartment
3006
and the second or peripheral compartment according to the intercompartmental rates k12 (input from central to peripheral) and
k21 (input from peripheral back into central). The nonlinear pathway was set according to Michaelis-Menten kinetics characterized
by the parameters Vmax and Km, where Vmax represents the maximum elimination rate for the nonlinear pathway and Km is the
serum ABX-EGF concentration at which ABX-EGF is eliminated
at 50% of Vmax. From the Km, the IC90 was derived as a secondary
parameter representing the serum ABX-EGF concentration at
which ABX-EGF is eliminated at 10% of Vmax by the nonlinear
elimination pathway.
Nonlinear regression analysis of skin rash data and serum
ABX-EGF concentration data was accomplished with SAAM II
version 1.2 (SAAM Institute Inc, Seattle, WA). The goodness of
model fit for the pharmacokinetic or the pharmacodynamic
model was guided by inspection of the SEs of the mean parameter
estimates, the minimization of the objective function (SAAM II
User Guide 1997-1998; SAAM Institute Inc) and the minimization
of the Akaike information criterion.28 Reduction of the model was
deemed appropriate if the SEs of the relevant parameters were
either improved or were not adversely affected, the visual fit of the
data improved or remained acceptable, and the objective function
and/or the Akaike information criterion did not increase.
RESULTS
General
Ninety-five total patients, whose characteristics are
presented in Table 1, were enrolled; 88 patients received at
least one dose of ABX-EGF and are considered a modified
intent-to-treat population for assessment of antitumor response and toxicity. Seven patients provided informed consent but were withdrawn from study before receiving drug.
Four patients withdrew consent, two patients were found to
be ineligible, and one patient experienced an adverse event
(femoral fracture). Seventy-nine patients (90%) underwent
nephrectomy before enrollment. Only eight patients
(9%) had not previously received systemic biologic
and/or chemotherapy, whereas 75 (85%), 19 (22%), and
27 patients (31%) received prior treatment with IL-2,
chemotherapy, and/or radiation, respectively. The median number of prior biologic and/or chemotherapy regimens per patient was two, and 32 patients (36%)
received at least three prior regimens.
EGFr Immunostaining
EGFr immunostaining was performed on the tumor
biopsy specimens of 76 patients (86%), and 69 patients
(91%) scored positive (2⫹ to 3⫹ immunostaining in
ⱖ 10% of cells). Immunostaining was positive in 10 (77%)
of 13 patients with non– clear-cell cancer whose malignant
tissues were assessable, in 56 (93%) of 60 patients with
clear-cell cancer, and in all three (100%) patients in whom
no histologic subtype information was available. H scores
were at least 200 in 42 patients (55%), and below 200 in 34
patients (45%).
JOURNAL OF CLINICAL ONCOLOGY
Downloaded from jco.ascopubs.org on June 9, 2014. For personal use only. No other uses without permission.
Copyright © 2004 American Society of Clinical Oncology. All rights reserved.
ABX-EGF in Renal Cancer
Table 1. Patient Demographics According to ABX-EGF Dose Level
Dose (mg/kg)
1.0
(n ⫽ 22)
Characteristic
No. of
Patients
Mean age, years
Sex
Female
Male
Performance status (ECOG)
0
1 or 2
Prior nephrectomy
Yes
No
No. of prior biological and/or chemotherapy regimens
0
1-2
ⱖ3
Prior additional cancer surgery
Yes
No
Prior biological therapy
Yes
No
Prior radiation therapy
Yes
No
Prior chemotherapy
Yes
No
Prior hormonal therapy
Yes
No
1.5
(n ⫽ 22)
%
No. of
Patients
56
2.0
(n ⫽ 23)
%
No. of
Patients
57
2.5
(n ⫽ 21)
%
No. of
Patients
58
Total
(N ⫽ 88)
%
No. of
Patients
60
%
58
9
13
41
59
8
14
36
64
4
19
17
83
3
18
14
86
24
64
27
73
16
6
73
27
9
13
41
59
15
8
65
35
11
10
52
48
51
37
58
42
20
2
91
9
19
3
86
14
21
2
91
19
19
2
90
10
79
9
90
10
2
13
7
9
59
32
1
11
10
5
50
45
3
14
6
13
61
26
2
10
9
10
48
43
8
48
32
9
55
36
9
13
41
59
14
8
64
36
11
12
48
52
9
12
43
57
43
45
49
51
16
6
73
27
20
2
91
9
20
3
87
13
19
2
90
10
75
13
85
15
5
17
23
77
8
14
36
64
6
17
26
74
8
13
38
62
27
61
31
69
6
16
27
73
7
15
32
68
3
20
13
87
3
18
14
86
19
69
22
78
0
22
100
1
21
95
0
23
100
0
21
100
0
88
100
Abbreviation: ECOG, Eastern Cooperative Oncology Group.
Safety Overview
ABX-EGF was generally well tolerated with no hypersensitivity reactions. The most common toxicity was skin
rash, which was qualitatively similar in most affected patients and typically involved the face in a periorificial distribution, as well as the upper trunk. It usually consisted of
clusters of monomorphous pustular lesions that resembled
an acneiform-type drug eruption; however, a maculopapular eruption predominated in some individuals. The rash
was usually evident by the second or third week, and the
intensity was maximal by weeks 3 to 5. In most patients, the
intensity of the eruption gradually decreased despite uninterrupted treatment. The cutaneous manifestations were
usually associated with minimal or no symptoms. The incidence of rash as a function of ABX-EGF dose in patients
receiving a minimum of three doses is shown in Figure 1A,
and the distribution of the worst grades of rash as a function
of dose level is displayed in Figure 1B. The incidences of any
grade of cutaneous toxicity in patients treated with a mini-
mum of three doses of ABX-EGF were 68%, 95%, 87%, and
100% at the 1.0-, 1.5-, 2.0-, and 2.5-mg/kg/wk dose levels,
respectively. At the 2.5-mg/kg dose level, 100% of patients
developed skin toxicity, and 75% of patients developed
either grade 2 or 3 toxicity compared with 32% to 52% of
patients treated with ABX-EGF at the three lower doses. The
relationship between the incidence of skin rash and ABXEGF dose fits a sigmoidal model, as shown in Figure 1A.
Based on this model, the ABX-EGF dose predicted to yield a
50% incidence in skin rash (ED50 ⫾ SEM) was 0.78 ⫾ 0.13
mg/kg (95% CI, 0.51 to 1.0 mg/kg), whereas ␥ was estimated
as 3.3 ⫾ 1.4 mg/kg (95%, CI 0.42 to 6.2 mg/kg). This
latter parameter ␥ (ie, the factor that accommodates the
shape of the curve, being estimated at a value greater than
1) reflects the narrow dose range within which the incidence of skin rash increases from 0% to 100%. The
ABX-EGF dose predicted to result in a 90% incidence of
skin rash (ED90) was estimated to be 1.5 ⫾ 0.26 mg/kg
(95% CI, 1.0 to 2.0 mg/kg).
3007
www.jco.org
Downloaded from jco.ascopubs.org on June 9, 2014. For personal use only. No other uses without permission.
Copyright © 2004 American Society of Clinical Oncology. All rights reserved.
Rowinsky et al
Fig 1. (A) Scatter graph of the incidence of rash at different ABX-EGF doses (F), with the line representing the percentage of rash predicted using a sigmoidal
Emax model. This model predicted that 90% of patients would develop rash at 1.5 ⫾ 0.26 mg/kg/wk (95%CI, 1.0 to 2.0 mg/kg/wk). (B) The percentages of
subjects with a particular grade of rash at different doses.
Patients treated with ABX-EGF also complained of
asthenia, diarrhea, stomatitis, and nausea, but these effects
were generally mild to moderate in severity and not doserelated in the dosing range evaluated. The incidences of
adverse effects that were graded as at least grade 2 in severity in
at least 5% of patients are listed in Table 2. Five serious adverse
events that were possibly related to ABX-EGF included dys-
pnea and diarrhea at 1.0 mg/kg, deep venous thrombosis at
1.5 mg/kg, and vomiting and rigors at 1.5 mg/kg.
Antitumor Activity
Of the 88 patients who received at least one dose of ABXEGF, three whose pertinent disease- and treatment-related
details are presented in Table 3 had partial or complete
Table 2. Incidence of Moderate and Severe Adverse Events As a Function of ABX-EGF Dose
Dose (mg/kg)
1
(n ⫽ 22)
1.5
(n ⫽ 22)
2.0
(n ⫽ 23)
2.5
(n ⫽ 21)
Total
(N ⫽ 88)
Adverse Event
No. of Patients
%
No. of Patients
%
No. of Patients
%
No. of Patients
%
No. of Patients
%
Asthenia
Pain
Abdominal pain
Back pain
Anorexia
Constipation
Nausea
Peripheral edema
Hyperglycemia
Cough
Dyspnea
Diarrhea
Bone pain
Thrombophlebitis
2
2
1
2
0
1
1
2
1
0
1
2
0
2
9
9
5
9
6
5
2
6
2
2
2
0
1
2
2
1
1
2
27
23
9
27
9
9
9
4
4
2
1
1
2
1
1
1
1
2
0
1
0
17
17
9
4
4
9
4
4
4
4
9
1
0
0
2
1
0
0
1
1
5
4
2
2
0
5
13
11
5
11
4
5
4
4
4
8
9
5
4
4
15
13
6
13
5
6
5
5
5
9
10
6
5
5
5
5
9
5
5
9
9
5
9
9
5
5
9
4
10
5
5
5
24
31
10
10
NOTE. Includes moderate and severe adverse events with ⱖ 5% incidence level.
3008
JOURNAL OF CLINICAL ONCOLOGY
Downloaded from jco.ascopubs.org on June 9, 2014. For personal use only. No other uses without permission.
Copyright © 2004 American Society of Clinical Oncology. All rights reserved.
ABX-EGF in Renal Cancer
Table 3. Pertinent Characteristics of Patients Experiencing Objective Disease Regression
Patient No.
ABX-EGF
Dose
(mg/kg)
Disease
Response
Time to Disease
Progression
(days)ⴱ
EGFr
Expression
(H score)
No. of Prior
Treatment
Regimens
Hemoglobin
Nephrectomy
Skin Rash
1
2
3
4
5
1.0
1.0
1.5
2.5
2.5
PR
MR
CR
MR
PR
128
631 ⫹
540 ⫹
78
221
ⱖ 200
⬍ 200
⬍ 200
⬍ 200
NA
4
0
3
2
1
⬍ LLN
Normal
Normal
Normal
Normal
No
Yes
Yes
Yes
Yes
Moderate
Moderate
Mild
Moderate
Moderate
Abbreviations: EGFr, epidermal growth factor receptor; PR, partial response; LLN, lower limit of normal; MR, minor response; CR, complete response; NA,
not available.
ⴱ
Tabulated as of August 18, 2003.
responses, and two other individuals had objective evidence
of tumor regression that did not meet the criteria for a
partial response. In these patients, tumor regression was
noted at the planned assessment following the first 8-week
course and was confirmed 4 weeks later. The complete
response was documented in a 60-year-old woman who
developed multiple large biopsy-confirmed pulmonary metastases 3 months following nephrectomy, that did not respond to treatment with high-dose IL-2, ending 2 months
before ABX-EGF treatment. As shown in the computerized
tomographic scans in Figure 2, a partial response was documented following eight weekly treatments with ABX-EGF
at the 1.5-mg/kg dose level, and further disease regression
occurred during additional therapy. The patient elected to
discontinue ABX-EGF after 10 months of treatment and
experienced progressive regression of the multiple pulmonary metastases over the next year until complete disappearance of all disease was documented on computed
tomography scan 23 months following the start of ABXEGF treatment. Forty-four other patients (50%) had stable
disease as their best response at their first 8-week assessment, and received additional treatment. The median
progression-free survival (PFS) was 100 days (95% CI, 58 to
140 days). At this juncture, follow-up is too short to estimate parameters pertaining to overall survival. As presented in Table 4, neither PFS nor response was related to
ABX-EGF dose. Furthermore, neither the intensity of EGFr
immunostaining by H score, Eastern Cooperative Oncology Group performance status, or number of prior systemic
therapies correlated with PFS. However, low hemoglobin
and high alkaline phosphatase values were related adversely
to PFS. PFS was significantly longer for the 60 patients who
entered the study with normal hemoglobin levels than for
the 27 patients with low hemoglobin levels (median, 140
days; [95% CI, 88 to 165 days] v 52 days [95% CI, 46 to 57
days]). In addition, the PFS of 74 patients with normal
serum alkaline phosphatase was significantly longer than
that of the 14 patients with elevated serum alkaline phosphatase (median, 108 days [95% CI, 80 to 161 days] v 51
days [95% CI, 46 to 56 days]; log-rank P ⫽ .0023). Although
the number of patients with non– clear-cell histology was
relatively low, patients with this histology seemed to have a
better overall outcome than those with clear-cell histology.
Among the 14 patients with non– clear-cell carcinoma,
there were two partial responders, one minor responder,
and six patients with stable disease as their best response.
Patients with non– clear-cell carcinoma had a longer,
though not statistically significant PFS than those with
clear-cell carcinoma (median, 92 days [95% CI, 57 to 140
days] v 56 days [95% CI, 46 to 272 days]). As shown in
Figure 3, an exploratory analysis of PFS as a function of skin
toxicity demonstrated that the PFS of 43 patients who had
moderate or severe skin toxicity was significantly longer
than that of 34 patients with mild skin toxicity, and the PFS
of each of the aforementioned groups was significantly
longer than that of the 11 patients who did not experience
skin toxicity (median, 128 days [95% CI, 80 to 168 days] v
76 days [95% CI, 53 to 161 days] v 56 days [95% CI, 41 to
115 days], respectively; log-rank P ⫽ .0139); however, these
results should be interpreted with caution because of the
small numbers of patients and broadly overlapping CIs,
particularly in the latter group.
Pharmacokinetics
The mean ABX-EGF Cmax and Ctrough concentrations
during the first course of treatment are shown in Figure 4.
The steady-state ABX-EGF Cmax ⫾ SEM values, which were
calculated based on at least four average values, were 22.0 ⫾
0.350, 42.2 ⫾ 0.555, 70.1 ⫾ 3.10, and 130 ⫾ 10.2 ␮g/mL
for patients treated with ABX-EGF doses of 1.0, 1.5, 2.0, and
2.5 mg/kg/wk, respectively. The steady-state ABX-EGF
Ctrough ⫾ SEM values were calculated based on three or
more average values (see Patients and Methods) and were
0.473 ⫾ 0.0300, 9.69 ⫾ 0.611, 27.4 ⫾ 0.195, and 48.4 ⫾ 1.37
␮g/mL for patients treated with ABX-EGF doses of 1.0, 1.5,
2.0, and 2.5 mg/kg/wk, respectively. ABX-EGF concentrations increased nonlinearly with dose, which was most
likely due to progressive saturation of a fixed EGFr sink.
EGFr-mediated CL requires occupancy of the EGFr by
ABX-EGF before internalization. A pharmacokinetic
3009
www.jco.org
Downloaded from jco.ascopubs.org on June 9, 2014. For personal use only. No other uses without permission.
Copyright © 2004 American Society of Clinical Oncology. All rights reserved.
Rowinsky et al
Fig 2. Computed tomography scans in a patient who experienced a CR. (A and D) Two pulmonary metastases before treatment with ABX-EGF (1.5 mg/kg),
and (B and E) following the first 8-week course. All lesions decreased until the end of 10 months of therapy, at which time the patient discontinued treatment.
(C and F) Lesions decreased on observation only until a complete remission was documented 1 year after discontinuing ABX-EGF.
into the systemic circulation at 0.908 ⫾ 0.208 day⫺1 (95%
CI, 0.490 to 1.33 day⫺1). The mean ⫾ SEM estimates for
linear CL and volume of distribution were 2.59 ⫾ 0.196
mL/d/kg (95% CI, 2.19 to 2.98 mL/d/kg), and 41.8 ⫾ 0.675
mL/kg (95% CI, 40.4 to 43.1 mL/d/kg), respectively. The
parameters characterizing the nonlinear CL, Vmax, and Km,
model, which incorporates both linear CL and nonlinear
Michaelis-Menten saturable CL, was fit to the ABX-EGF
peak and trough data, as shown in Figure 4. The distribution of ABX-EGF was reflected by the rate constant out of
the systemic circulation, k12 at 0.366 ⫾ 0.0798 day⫺1 (95%
CI, 0.206 to 0.527 day⫺1), and k21, the rate constant back
Table 4. Best Objective Antitumor Response and Progression-Free Survival in the Modified Intent-to-Treat Population As a Function of
ABX-EGF Dose Level
Dose (mg/kg)
1.0
(n ⫽ 22)
CR, PR/MR
Stable disease
PD
Not available
Kaplan-Meier estimate of time to disease progression, days
Median
95% CI
1.5
(n ⫽ 22)
2.0
(n ⫽ 23)
2.5
(n ⫽ 21)
Total
(N ⫽ 88)
No. of
Patients
%
No. of
Patients
%
No. of
Patients
%
No. of
Patients
%
No. of
Patients
%
1/1
11
8
1
9
50
36
5
1/0
12
8
1
5
55
36
5
0/0
9
11
3
0
39
48
13
1/1
12
6
1
10
57
29
5
5
44
33
6
6
50
38
7
108
56 to 40
165
54 to 246
53
46 to 100
103
60 to 162
100
8 to 140
Abbreviations: CR, complete response; PR/MR, partial response/minor response; PD, progressive disease.
3010
JOURNAL OF CLINICAL ONCOLOGY
Downloaded from jco.ascopubs.org on June 9, 2014. For personal use only. No other uses without permission.
Copyright © 2004 American Society of Clinical Oncology. All rights reserved.
ABX-EGF in Renal Cancer
Fig 3. Kaplan and Meier curves showing progression-free survival as a
function of skin rash intensity classified as none, mild, or moderate to severe.
were estimated at 165 ⫾ 7.16 ␮g/d/kg (95% CI, 151 to 179
␮g/d/kg) and 1.24 ⫾ 0.186 ␮g/mL (95% CI, 0.777 to 1.70
␮g/d/kg), respectively. The half-life, which was derived as a
secondary parameter expected when the nonlinear CL pathway is fully saturated, averaged 15.9 ⫾ 1.57 days (95% CI,
12.7 to 19.1 days). Also derived as a secondary parameter,
was the ABX-EGF concentration at which 90% saturation
of EGFr-mediated CL occurs and was estimated at 11.2 ⫾
1.86 ␮g/mL (95% CI, 6.70 to 15.3 ␮g/mL).
Mean CL ⫾ SEM values, derived by noncompartmental methods as a function of dose are plotted in Figure 5.
Mean ⫾ SEM ABX-EGF CL values decreased from 14 ⫾ 1.3,
11 ⫾ 1.5, 8.5 ⫾ 1.5, to 4.8 ⫾ 0.49 mL/d/kg as the dose of
ABX-EGF increased from 1.0 (n ⫽ 20), to 1.5 (n ⫽ 20), to
Fig 4. Mean (⫾SEM) ABX-EGF concentration-time data fit to the model
described in the Patients and Methods section. Graph shows the observed
ABX-EGF concentrations represented as symbols with (F), (E), (), and (ƒ)
for the 1.0-, 1.5-, 2.0-, and 2.5-mg/kg doses, respectively. The different lines
represent the model fits for each dose. The horizontal line represents the
IC90 for saturation of the nonlinear clearance.
Fig 5. Plot graph of the mean (⫾ SEM) clearance values of ABX-EGF as a
function of dose.
2.0 (n ⫽ 19), to 2.5 (n ⫽ 14) mg/kg. At the 2.5-mg/kg dose,
the CL of ABX-EGF (4.8 mL/d/kg) was close to the typical
CL range (1 to 4 mL/d/kg) of human IgG antibodies that are
not subject to an antigen sink (Adalimumab prescribing
information; North Chicago, IL; Abbott Laboratories 2002;
ABX-IL8, Abgenix data on file), but are cleared via the
reticuloendothelial system.29,30 This observation supports
that the majority of the EGFr sink was saturated at the ABXEGF exposure associated with 2.5 mg/kg administered weekly.
DISCUSSION
Nonspecific cytotoxic therapeutics have had negligible impact, if any, on the treatment of advanced renal cell cancer.31-33
Although some biologic therapies, particularly IL-2, can induce durable responses, the response rate is low and the
toxicity of IL-2 limits the number of patients who can
receive this treatment.31-33 The transduction of proliferative signals mediated by extracellular growth factor receptors such as the EGFr, is an attractive target for therapeutic
development against renal cell cancer and a wide range of
other malignancies.3 The notion that EGFr may be a strategic target is supported by studies that have demonstrated
that EGFr is expressed, overexpressed, or mutated in most
epithelial malignancies, and receptor expression is an adverse prognostic determinant in some settings.1-10 Since
antibodies against the extracellular domain of EGFr and
small molecule inhibitors of the EGFr tyrosine kinase have
been demonstrated to inhibit EGFr-dependent tumor proliferation, angiogenesis, invasion, and metastases in preclinical studies, these therapeutic strategies are being
3011
www.jco.org
Downloaded from jco.ascopubs.org on June 9, 2014. For personal use only. No other uses without permission.
Copyright © 2004 American Society of Clinical Oncology. All rights reserved.
Rowinsky et al
evaluated in the clinic, and the results of these efforts in
many disease settings have been encouraging.15-18,34-39
In the present study, the antitumor activity of four dose
levels of ABX-EGF, a fully human monoclonal antibody
against EGFr, was evaluated in previously treated patients
with metastatic renal carcinoma. Although EGFr receptor
number in experimental models of renal cancer has been
related to sensitivity to ABX-EGF, receptor number is not
readily quantifiable on archival biopsy specimens. In addition, since most renal cell carcinomas are known to express
EGFr as determined immunohistochemically, and EGFr
immunostaining has not been related to antitumor activity
in clinical trials of EGFr inhibitors to date, EGFr immunostaining was not used as a criterion for study eligibility.10,34
In fact, EGFr immunostaining was performed on 76 tumor
biopsy specimens (86%) in this study, and most (69 specimens [91%]) demonstrated 2⫹ to 3⫹ EGFr immunostaining. Nevertheless, because of the lack of reliable patient
enrichment strategies to facilitate the establishment of
proof of principle in disease-directed studies of EGFrtargeting therapeutics, the low overall response rates noted
in studies of EGFr-targeted therapeutics to date, and the
evaluation of a wide dosing range in the present study, the
trial was designed so that approximately 80 patients (20
patients at each dose level—is larger than traditional phase
II studies of nonspecific cytotoxic therapies) could be
treated to provide adequate power to detect a low, albeit
potentially relevant, level of overall antitumor activity and
to assess dose-dependent effects.
ABX-EGF produced a low rate of objective antitumor
responses, with one complete response, two partial responses, and two minor responses observed in the patients
in this study, the overwhelming majority of whom had been
treated with at least one prior biotherapy regimen (85%).
Furthermore, 22% and 31% of the patients had received
prior chemotherapy and/or radiotherapy, respectively. In
addition, 50% of the patients had stable disease as their best
response, and the median PFS was 100 days (95% CI, 8 to
140 days). No relationships between the dose and parameters indicative of clinical benefit were evident between doses
ranging from 1.0 to 2.5 mg/kg/wk. In addition, EGFr expression did not relate to objective activity or PFS, but these
results must be interpreted with caution since EGFr immunostaining was negative in only a few patients, thereby
limiting the statistical power of this analysis. Interestingly,
both low serum hemoglobin and elevated alkaline phosphatase strongly correlated with short PFS. Similar predictive
variables have been reported by Motzer et al in their derivation of a multivariate model that was predictive of survival in 670 patients with metastatic renal cancer.32 In this
model, low serum hemoglobin, low performance status,
high serum lactate dehydrogenase, high serum calcium, and
short time from diagnosis to treatment, predicted for short
survival; alkaline phosphatase, however, was not included
3012
in the analysis. Median survival times were 20, 10, and 4
months for patients with 0, 1 to 2, and ⱖ 3 of these risk
factors, respectively, and patients in these respective groups
were classified as low-, intermediate-, and high-risk. To
roughly gauge the categorical distribution of the patients in
the present study in terms of risk, these predictive factors
were modified with alkaline phosphatase, and time from the
last treatment until ABX-EGF, substituting for lactate dehydrogenase and time from diagnosis to initial treatment,
respectively. Using this modified risk scoring scheme, the
distribution of scores in the present study veered toward the
poorer prognostic groups, with 2%, 72%, and 26% of patients fitting into low-, intermediate-, and high-risk groups,
respectively. Furthermore, PFS was significantly related to
this modified risk group distribution (median PFS, 274 days
[95% CI, NA] v 110 days [95% CI, 82 to 162 days] v 50 days
[95% CI, 42 to 56]; log-rank P ⬍ .0002).
Relevant differences between EGFr-targeting therapeutics with respect to their antitumor spectra have been
emerging. Both cetuximab, a chimeric monoclonal antibody that has demonstrated notable activity in advanced
colorectal cancer, and gefitinib, a small-molecule inhibitor
of EGFr tyrosine kinase that has received regulatory approval for treatment of patients with recurrent non–smallcell lung cancer, were devoid of objective activity in phase II
studies in renal cell cancer.40,41 No antitumor activity was
reported in 55 previously untreated patients with metastatic
renal carcinoma who received cetuximab.40 The median
PFS was 57 days, which was similar to that reported with
IFN-␣.31-33 Additionally, no objective activity was noted
with gefinitib in a phase II study of patients with advanced
renal cell carcinoma.41 Although the potential for a drug to
induce tumor response and a relevant degree of clinical
benefit seems to be discordant in patients with renal cell
carcinoma, tumor response may serve as a rough screen for
therapeutics of ultimate impact.
The principal toxicity of ABX-EGF exposure at the
dosing rate of 1.0 to 2.5 mg/kg/wk, skin rash, was generally
tolerable and predictable. Skin toxicity was generally well
tolerated, even by patients in whom the rash was graded as
severe by NCI CTC, and typically improved, reaching a steadychronic state, during treatment. Although rash was a doserelated effect, it was not dose limiting. Of note, diarrhea has
been reported rarely in patients treated with ABX-EGF, and
has been uncommon in patients treated with both chimeric
and humanized antibodies against EGFr.14-18,26,34-46 In contrast, diarrhea has been the dose-limiting toxicity for most
small-molecule inhibitors of EGFr tyrosine kinase.15-18,26,34-46
Furthermore, neither hypersensitivity reactions, nor anaphylactic reactions, qualitatively similar to those that have
been reported with other EGFr-targeting small molecules
and antibodies have been noted following treatment with
ABX-EGF.15-18,26,34-46 These differences compared with
other EGFr-targeting antibodies may be due to the fact that
JOURNAL OF CLINICAL ONCOLOGY
Downloaded from jco.ascopubs.org on June 9, 2014. For personal use only. No other uses without permission.
Copyright © 2004 American Society of Clinical Oncology. All rights reserved.
ABX-EGF in Renal Cancer
ABX-EGF is fully human, whereas cetuximab and other
antibodies such as EMD 72,000 are comprised of various
degrees of murine-derived amino acid sequences. These
antibodies seem to require premedication with antihistamines and corticosteroids, which is apparently obviated by
using a fully human antibody.16-18,26, 34-36
The most notable aspect of the dermatologic toxicity
was its remarkably low interindividual variability in both
severity and temporal onset as exemplified by the development of skin-related toxicity in 90% of patients treated with
at least three doses of ABX-EGF at the 1.5- or 2.0-mg/kg/wk
dosing rates and 100% of patients treated with 2.5 mg/kg/
wk. The potential importance of this finding is illustrated by
several retrospective analyses of phase II studies of both
EGFr-targeting small molecules and antibodies, in which
the propensity to develop skin-related toxicity, as well as the
severity of skin toxicity, seemed to correlate with the magnitude of therapeutic benefit.42,43 In a retrospective review
of various phase I and II studies with the chimeric EGFr
antibody cetuximab, the duration of survival was related to
the development and severity of skin rash.42 Similarly, a strong
relationship between the occurrence and severity of drugrelated skin rash and overall survival has been reported in
patients with advanced non–small-cell lung cancer treated
with the EGFr tyrosine kinase inhibitor erlotinib.43 An exploratory analysis of PFS as a function of skin toxicity
demonstrated that the PFS of 43 patients who had moderate
or severe skin toxicity was significantly longer than that of
34 patients with mild skin toxicity, and the PFS of each of
the aforementioned groups was significantly longer than
that of the 11 patients who did not experience skin toxicity
(median, 128 days [95% CI, 80 to 168 days] v 76 days [95%
CI, 53 to 161 days] v 56 days [95% CI, 41 to 115]; log-rank
P ⫽ .0139); however, these results should be interpreted
with caution due to the small numbers of patients and
broadly overlapping CIs, particularly in the latter group.
Therefore, the therapeutic significance of skin toxicity following treatment with EGFr-targeting agents will require
large prospective studies. At this juncture, these exploratory
analyses do not necessarily indicate true biologic concordance between the drug-related effects in skin and malignant tissues, or that skin toxicity is a true surrogate for
therapeutic benefit. Although additional clinical studies, particularly larger evaluations in which specific grades of toxicity
are prospectively targeted, would be required to precisely define the relationship between skin toxicity and therapeutic
benefit, the incidence of skin rash likely represents a pharmacodynamic index relevant to antitumor activity. If this is the
case, with a consistent pharmacokinetic and toxicity profile,
ABX-EGF may confer a higher therapeutic index than other
EGFr-targeting therapeutics, especially small molecules. In
general, small-molecule inhibitors of EGFr tyrosine kinase exhibit highly variable exposure across individuals, which is
likely due in part to variability in gastrointestinal absorption,
first-pass hepatic processing, and/or hepatic metabolism.44-46
These intersubject differences may be further compounded by
the variability in tumor penetration and intracellular drug
transport, all of which may cumulatively produce profound
inter-individual variability in drug behavior, not to mention all
of the, as of yet undefined, determinants of biologic responsiveness in both toxicity and antitumor benefit.
ABX-EGF was administered at dosing rates projected
to yield antibody concentrations that exceed relevant concentrations in xenograft models. Indeed, at all dosing rates
evaluated, at some time during treatment, serum ABX-EGF
concentrations exceeded such levels. At steady-state, trough
concentrations at the 2.0- and 2.5-mg/kg/wk dose rates
exceeded the xenograft IC90. The estimated IC90 of the
nonlinear CL pathway (11.2 ␮g/mL) based on the clinical
pharmacokinetic data was exceeded at steady-state at
trough by both the 2.0- and 2.5-mg/kg/wk ABX-EGF doses,
and approached by the 1.5 mg/kg/wk dose (Fig 4). Interestingly, the ABX-EGF dosing rate is at least 50% less than the
dose of the chimerized anti-EGFr antibody cetuximab selected for clinical trials.17,18,34,35 The disparity may reflect
the four-fold higher affinity of ABX-EGF (5 ⫻ 10⫺11M)
compared with cetuximab (2 ⫻ 10⫺10M), as well as differences in their propensity to generate human antihuman immune responses. The potential implications of these estimates
are based to the reasoning that treatment of patients with
ABX-EGF in the high end of its feasible dosing range will
maximize the occupancy of EGFr, thereby maximizing therapeutic efficacy. Although relationships between dose and either objective response or PFS were not apparent, the
pharmacokinetic and pharmacodynamic results support the
selection of a dosing rate for subsequent evaluations that
achieves serum trough ABX-EGF concentrations at least as
high as what has been observed in this study at the 2.5-mg/
kg/wk dose rate. At this dose rate, repetitive treatment was also
well tolerated, and HAHA was not detected despite chronic
long-term treatment. The lack of HAHA may be due to the
inherent absence of immunogenicity of this fully human antibody. It also suggests an absence of alterations in pharmacokinetics over time, which was indeed observed in this study.
ABX-EGF demonstrated a low rate of objective antitumor activity in previously treated patients with metastatic
renal cell carcinoma, most of whom were of intermediateor high-risk and heavily-pretreated. The highly consistent
toxicologic, pharmacokinetic, and pharmacodynamic profiles of ABX-EGF are encouraging. At 2.5 mg/kg/wk, 100%
of patients had skin rash and ABX-EGF concentrations
were maintained above the IC90 values estimated from
xenograft studies and from the clinical nonlinear CL pathway. Although skin rash may be a pharmacodynamic
marker of drug action, its potential as a surrogate marker of
clinical benefit requires further evaluation.
■ ■ ■
3013
www.jco.org
Downloaded from jco.ascopubs.org on June 9, 2014. For personal use only. No other uses without permission.
Copyright © 2004 American Society of Clinical Oncology. All rights reserved.
Rowinsky et al
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. Owns stock (not including
shares held through a public mutual fund): Gisela Schwab,
Abgenix. Acted as a consultant within the last 2 years: Eric
K. Rowinsky, Abgenix; Robert Figlin, Abgenix; Ronald
Bukowski, Abgenix. Performed contract work within the last
REFERENCES
1. Yarden Y, Sliwkowski MX: Untangling the
ErbB signalling network. Nat Rev Mol Cell Biol
2:127-137, 2001
2. Walker RA: The erbB/HER type 1 tyrosine
kinase receptor family. J Pathol 185:234-235,
1998
3. Sargent ER, Gomella EG, Belldegrun A, et
al: Epidermal growth factor gene expression in
normal human kidney and renal cell carcinoma.
J Urol 142:1364-1368, 1989
4. Yoshida K, Tosaka A, Takeuchi S, et al:
Epidermal growth factor receptor content in human renal cell carcinomas. Cancer 73:19131918, 1994
5. Weidner U, Peter S, Strohmeyer T, et al:
Inverse relationship of epidermal growth factor
receptor and Her2/neu gene expression in human renal cell carcinoma. Cancer Res 50:45044509, 1990
6. Weber KL, Doucet M, Price JE, et al:
Blockade of epidermal growth factor receptor
signaling leads to inhibition of renal cell carcinoma growth in the bone of nude mice. Cancer
Res 63:2940-2947, 2003
7. Yao M, Shuin T, Misaki H, et al: Enhanced
expression of c-myc and epidermal growth factor
receptor (C-erbB-1) genes in primary human renal cancer. Cancer Res 48:6753-6757, 1988
8. Stumm G, Eberwein S, Rostock-Wolf S, et
al: Concomitant overexpression of the EGFR and
erbB-2 genes in renal cell carcinoma (RCC) is
correlated with dedifferentiation and metastasis.
Int J Cancer 69:17-22, 1996
9. Scher HI, Sarkis A, Reuter V, et al: Changing pattern of the epidermal growth factor receptor and transforming growth factor alpha in the
progression of prostatic neoplasms. Clin Cancer
Res 1:545-550, 1995
10. Shirai H, Ueno E, Osaki M, et al: Expression of growth factors and their receptors in
human early colorectal carcinomas: Immunohistochemical study. Anticancer Res 15:2889-2894,
1995
11. Porebska I, Harlozinska A, Bojarowski T:
Expression of the tyrosine kinase activity growth
factor receptors (EGFR, ERB B2, ERB B3) in
colorectal adenocarcinoma and adenomas. Tumour Biol 21:105-115, 2000
12. Radinsky R, Risin, Fan, et al: Level and
function of epidermal growth factor receptor
predict the metastatic potential of human colon
carcinoma cells. Clin Cancer Res 1:19-31, 1995
13. Itakura Y, Sasano H, Shiga C, et al: Epidermal growth factor receptor overexpression in
3014
2 years: Eric K. Rowinsky, Abgenix; Robert Figlin, Abgenix;
Ronald Bukowski, Abgenix; Jared A. Gollob, Abgenix; John A.
Thompson, Abgenix; Nicholas J. Vogelzang, Abgenix; Naomi
Haas, Abgenix; Pamela Lockbaum, Abgenix; Yu-Ping Li,
Abgenix; Rosalin Arends, Abgenix; Kenneth A. Foon, Abgenix;
Gisela Schwab, Abgenix. Served as an officer or member of the
Board of a company: Gisela Schwab, Abgenix. Received more
than $2,000 a year from a company for either of the last 2 years:
Eric K. Rowinsky, Abgenix; Robert Figlin, Abgenix; Ronald
Bukowski, Abgenix.
esophageal carcinoma: An immunohistochemical study correlated with clinicopathologic findings and DNA amplification. Cancer 74:795-804,
1994
14. Fan Z, Mendelsohn J: Therapeutic application of anti-growth factor receptor antibodies.
Curr Opin Oncol 10:67-73, 1998
15. Strawn LM, Shawver LK: Tyrosine kinases
in disease: Overview of kinase inhibitors as
therapeutic agents and current drugs in clinical
trials. Exp Opin Invest Drugs 7:553-573, 1998
16. Baselga J, Mendelsohn J: Type I receptor
tyrosine kinases as targets for therapy in breast
cancer. J Mammary Gland Biol Neoplasia 2:165174, 1997
17. Mendelsohn J: Epidermal growth factor
receptor inhibition by a monoclonal antibody as
anticancer therapy. Clin Cancer Res 3:27032707, 1997
18. Arteaga CL, Baselga J: Clinical trial design
and endpoints for epidermal growth factor
receptor-targeted therapies: Implications for
drug development and practice. Clin Cancer Res
9:1579-1589, 2003
19. Yang XD, Jia XC, Corvalan JRF, et al:
Development of ABX-EGF, a fully human antiEGF receptor monoclonal antibody for cancer
therapy. Crit Rev Oncol Hematol 38:17-23, 2001
20. Yang X-D, Jia X-C, Corvalan JRF, et al:
Eradication of established tumors by a fully human monoclonal antibody to the epidermal
growth factor receptor without concomitant chemotherapy. Cancer Res 59:1236-1243, 1999
21. Ciardiello F, Caputo R, Bianco R, et al:
Antitumor effect and potentiation of cytotoxic
drugs activity in human cancer cells by ZD-1839
(Iressa), an epidermal growth factor receptorselective tyrosine kinase inhibitor. Clin Cancer
Res 6:2053-2063, 2000
22. Fan Z, Baselga J, Masui H, et al: Antitumor effect of anti-epidermal growth factor
receptor monoclonal antibodies plus cisdiamminedichloroplatinum on well established
A431 cell xenografts. Cancer Res 53:46374642, 1993
23. Moyer JD, Barbacci EG, Iwata KK, et al:
Induction of apoptosis and cell cycle arrest by
CP-358,774, an inhibitor of epidermal growth
factor receptor tyrosine kinase. Cancer Res 57:
4838-4848, 1997
24. Fry DW: Inhibition of the epidermal growth
factor receptor family of tyrosine kinases as an
approach to cancer chemotherapy: Progression
from reversible to irreversible inhibitors. Pharmacol Ther 82:207-218, 1999
25. Roskos L, Lohner M, Schwab G, et al: A
biomathematical model of neoplastic cell growth
and prediction on silico of effective doses of
ABX-EGF in cancer patients. Proc Am Assoc
Cancer Res 42:833, 2001 (abstr 4471)
26. Figlin RA, Belldegrun A, Crawford J, et al:
ABX-EGF, a fully human anti-epidermal growth
factor receptor monoclonal antibody in patients
with advanced cancer: Phase I clinical results.
Proc Am Soc Clin Oncol 21:10a, 2002 (abstr 35)
27. Lalonde RL: Pharmacodynamics, in Evans
WE, Schentag JJ, Jusko WJ (eds): Applied Pharmacokinetics: Principles of Therapeutic Drug
Monitoring. Vancouver, WA, Applied Therapeutics, 1992, pp 4-1– 4-33
28. Ludden TM, Beal SL, Sheiner LB: Comparison of the Akaike Information Criterion, the
Schwarz criterion and the F test as guides to
model selection. J Pharmacokinet Biopharm 22:
431-445, 1994
29. Ober RJ, Radu CG, Ghetie V, et al: Differences in promiscuity for antibody-FcRn interactions across species: Implications for therapeutic
antibodies. Int Immunol 13:1551-1559, 2001
30. Ghetie V, Ward ES: Transcytosis and catabolism of antibody. Immunol Res 25:97-113,
2002
31. Yang JC, Sherry RM, Steinberg SM, et al:
Randomized study of high-dose and low-dose
interleukin-2 in patients with metastatic renal
cancer. J Clin Oncol 21:3127-3132, 2003
32. Motzer RJ, Mazumdar M, Bacik J, et al:
Survival and prognostic stratification of 670 patients with advanced renal cell carcinoma. J Clin
Oncol 17:2530-2540, 1999
33. Motzer RJ: Renal cell carcinoma: A priority
malignancy for development and study of novel
therapies. J Clin Oncol 21:1193-1194, 2003
34. Mendelsohn J, Baselga J: Status of epidermal growth factor antagonists in the biology
and treatment of cancer. J Clin Oncol 21:27872799, 2003
35. Cunningham D, Humblet Y, Siena S, et al:
Cetuximab (C225) alone or in combination with
irinotecan (CPT-11) in patients with epidermal
growth factor receptor-positve, irinotecanrefractory metastatic colorectal cancer. Proc Am
Soc Clin Oncol 22:252, 2003 (abstr 1012)
36. Meropol NJ, Berlin J, Hecht JR, et al:
Multicenter study of ABX-EGF monotherapy in
patients with metastatic colorectal cancer. Proc
Am Soc Clin Oncol 22:256, 2003 (abstr 1026)
37. Fukuoka M, Yano S, Giaccone G, et al:
Multi-institutional randomized phase II trial of
gefitinib for previously treated patients with advanced non-small-cell lung cancer. J Clin Oncol
21:2237-2246, 2003
38. Perez-Soler R, Chachoua A, Huberman M,
et al: A phase II trial of the epidermal growth
JOURNAL OF CLINICAL ONCOLOGY
Downloaded from jco.ascopubs.org on June 9, 2014. For personal use only. No other uses without permission.
Copyright © 2004 American Society of Clinical Oncology. All rights reserved.
ABX-EGF in Renal Cancer
factor receptor tyrosine kinase inhibitor OSI-774,
following platinum-based chemotherapy, in patients with advanced, EGFR-expressing, nonsmall cell lung cancer. Proc Am Soc Clin Oncol
20:310, 2001 (abstr 1235)
39. Tabernero J, Rejo F, Jimenez E, et al: A
phase I, PK , and serial tumor and skin pharmacodynamic study of weekly, every 2-week, or
every 3-week 1-hour infusion EMD72000, a humanized monoclonal anti-epidermal growth factor receptor antibody in patients with advanced
tumors. Proc Am Soc Clin Oncol 22:192, 2003
(abstr 770)
40. Motzer RJ, Amato R, Todd M, et al:
Phase II trial of antiepidermal growth factor
receptor antibody C225 in patients with ad-
vanced renal carcinoma. Invest New Drugs
21:99-101, 2003
41. Drucker BJ, Schwartz L, Marion S, et al:
Phase II trial of ZD1839 (Iressa), an EGF receptor
inhibitor, in patients with advanced renal cell
carcinoma. Proc Am Soc Clin Oncol 21:181, 2002
(abstract 720)
42. Saltz L, Kies M, Abbruzzese JL, et al: The
presence and intensity of the cetuximab-induced
acne-like rash predicts for increased survival in
studies across multiple malignancies. Proc Am
Soc Clin Oncol 22:204, 2003 (abstr 817)
43. Clark GM, Perez-Soler R, Siu L, et al:
Rash is predictive of increased survival with
erlotinib. Proc Am Soc Clin Oncol 22:196, 2003
(abstr 786)
44. Ranson M, Hammond LA, Ferry D, et al:
ZD1839, a selective oral epidermal growth factor
receptor-tyrosine kinase inhibitor, is well tolerated and active in patients with solid malignant
tumors: Results of a phase I trial. J Clin Oncol
20:2240-2250, 2002
45. Nakagawa K, Yamamoto N, Kudoh S, et al:
A phase I intermittent dose-escalation trial of
ZD1839 (Iressa) in Japanese patients with solid
malignant tumors. Proc Am Soc Clin Oncol 19:
183, 2000 (abstr 711)
46. Hidalgo M, Siu L, Nemunaitis J, et al:
Phase I and pharmacologic study of OSI-774, an
epidermal growth factor receptor tyrosine kinase
inhibitor in patients with advanced solid malignancies. J Clin Oncol 19:3267-3279, 2001
3015
www.jco.org
Downloaded from jco.ascopubs.org on June 9, 2014. For personal use only. No other uses without permission.
Copyright © 2004 American Society of Clinical Oncology. All rights reserved.