Impact of Zoledronic Acid on Renal Function in Patients With Cancer ...

O R I G I N A L R E S E A R C H
<strong>Impact</strong> <strong>of</strong> <strong>Zoledronic</strong> <strong>Acid</strong> on Renal
Function in Patients With Cancer:
Clinical Significance and Development
<strong>of</strong> a Predictive Model
Raymond S. McDermott, MD, Dwight D. Kloth, PharmD, Hao Wang, MS,
Gary R. Hudes, MD, and Corey J. Langer, MD
Z
oledronic acid (Zometa) is a third-generation,
nitrogen-containing bisphosphonate that
showed significantly greater potency than did
pamidronate in preclinical testing. 1 It is the
first bisphosphonate to demonstrate clinical activity
in patients with osteoblastic metastases from prostate
cancer and with bone metastases from solid tumors
other than breast and prostate cancers. 2,3 In contrast
to pamidronate, for which approval was limited to
multiple myeloma and breast cancer, zoledronic acid
was approved by the US Food and Drug Administration
(FDA) for the treatment <strong>of</strong> multiple myeloma
and documented bone metastases from solid tumors
in conjunction with standard antineoplastic therapy.
Moreover, zoledronic acid <strong>of</strong>fers the convenience
<strong>of</strong> a 15-minute infusion as opposed to the 90–120
minutes required for pamidronate. This difference is
clinically significant because these treatments, once
initiated, are typically administered monthly for the
duration <strong>of</strong> the illness.
As with pamidronate, renal dysfunction is the
most serious toxicity associated with zoledronic
acid therapy and has been related to dose (more
common with 8 mg than 4 mg), infusion duration
(more pronounced with shorter infusions [eg, 5
minutes vs 15 minutes]), and total number <strong>of</strong> infusions.
4 In licensing studies <strong>of</strong> zoledronic acid, 2–5
renal deterioration—defined as a rise in serum
creatinine level <strong>of</strong> 0.5 mg/dL in patients with a
normal baseline creatinine level or 1 mg/dL in
From Fox Chase Cancer Center, Philadelphia, Pennsylvania.
Manuscript submitted September 8, 2005; accepted May 23, 2006.
Correspondence to: Corey J. Langer, MD, Director <strong>of</strong> Thoracic
Oncology, Fox Chase Cancer Center, 333 Cottman Avenue,
Philadelphia, PA 19111; telephone: (215) 728-2985; fax:
(215) 728-3639; e-mail: cj_langer@fccc.edu
J Support Oncol 2006;4:524–529
© 2006 Elsevier Inc. All rights reserved.
Abstract <strong>Zoledronic</strong> acid is a potent bisphosphonate licensed for the
treatment <strong>of</strong> myeloma and bone metastases from solid tumors. Renal deterioration
is the most significant toxicity associated with zoledronic acid.
We attempted to define the incidence and clinical significance <strong>of</strong> renal
deterioration in patients receiving zoledronic acid and to develop a riskfactor
pr<strong>of</strong>ile for this treatment sequela. This study is a retrospective analysis
<strong>of</strong> all patients who received zoledronic acid at Fox Chase Cancer Center,
Philadelphia, Pa, between 1/10/02 and 1/30/04. Data recorded included
patient demographics, tumor characteristics, comorbid illnesses, concomitant
medications, cancer therapy, number <strong>of</strong> zoledronic acid doses administered,
and serial creatinine measurements. In total, 3,115 evaluable doses
<strong>of</strong> zoledronic acid were administered to 446 patients (median, 4 doses;
mean, 6.98 doses; range, 1–28 doses) at a dose <strong>of</strong> 4 mg over 15 minutes
every 3–4 weeks. Of these 446 patients, 42 experienced renal deterioration
(median rise in creatinine level, 1.0 mg/dL; range, 0.5–4.4 mg/dL), requiring
discontinuation <strong>of</strong> zoledronic acid therapy in 8 cases. No patient required
dialysis and no patient died as a result <strong>of</strong> zoledronic acid-induced renal
dysfunction. On multivariable analysis, predictive factors for the development
<strong>of</strong> renal deterioration were patient age, a diagnosis <strong>of</strong> myeloma or
renal cell cancer, cumulative number <strong>of</strong> doses, concomitant therapy with
a nonsteroidal anti-inflammatory drug, and current or prior therapy with
cisplatin. Using these factors, we constructed a predictive model with an
area under the receiver operating characteristic curve <strong>of</strong> 0.75. The incidence
<strong>of</strong> clinically significant renal deterioration in patients treated with
zoledronic acid is low. We present a predictive model for decision support
when estimating this risk.
those with pre-existing renal impairment—was
observed in 9%–15% <strong>of</strong> patients treated with zoledronic
acid (4 mg over 15 minutes) and was not
significantly different from that seen with pamidronate
in patients with breast cancer or myeloma
(8.8% vs 8.2%). 5 Most cases <strong>of</strong> renal impairment
were mild and reversible, and episodes <strong>of</strong> acute
renal failure were rare. In total, 30 <strong>of</strong> 2,873 patients
(1%) enrolled on the three large phase III
524 www.SupportiveOncology.net THE JOURNAL OF SUPPORTIVE ONCOLOGY

McDermott, Kloth, Wang, et al
Also recorded were pretreatment serum creatinine levels
(SCr 1 ), maximum serum creatinine levels while the patient was
on therapy with zoledronic acid (SCr max ), and serum creatinine
levels within 30 days <strong>of</strong> the last zoledronic acid dose administered
(SCr 2 ) when available (or the next creatinine level recorded if
there were no intervening hospitalizations or clinically significant
events). Maximum change in serum creatinine was defined using
the formula ∆Cr max = SCr max –SCr 1 , and overall change in serum
creatinine was calculated as ∆Cr = SCr 2 –SCr 1 .
Renal deterioration was defined as a rise in serum creatinine
level <strong>of</strong> ≥ 0.5 mg/dL in patients with a normal baseline creatilicensing
studies required dialysis; most <strong>of</strong> these individuals
had either myeloma or prostate cancer. 2,3,5
However, with marketed use <strong>of</strong> the drug, renal deterioration
progressing to renal failure and dialysis has been reported,
and acute tubular necrosis has been described as one potential
mechanism. 6 Chang et al 7 characterized 72 cases in which physicians
reported renal failure associated with zoledronic acid
therapy to the FDA; 42 <strong>of</strong> these patients had multiple myeloma,
and 39 had baseline serum creatinine levels ≥ 1.4 mg/dL.
Patients developed renal failure after a median 2.4 doses, 27
patients required dialysis, and 18 died. As a result, Chang et
al 7 advocated that caregivers monitor renal function before
each dose <strong>of</strong> zoledronic acid, ensure adequate hydration, and
discontinue therapy in the presence <strong>of</strong> renal deterioration. 7
Nonetheless, there are many potential reasons for renal deterioration
in patients with cancer, including progressive malignant
disease, nephrotoxic chemotherapy, or other medications
and comorbid illnesses such as chronic renal impairment,
hypertension, or diabetes. 8 The aim <strong>of</strong> this study was to better
define the incidence and clinical significance <strong>of</strong> renal deterioration
in a broad spectrum <strong>of</strong> patients treated with zoledronic acid
and to identify potential risk factors for renal deterioration.
Patients and Methods
A retrospective analysis was performed with approval <strong>of</strong> the
institutional review board, which determined that this research
was exempt under the Code <strong>of</strong> Federal Regulations (45 CFR
46.101.b). Investigators identified and retrieved data on all
patients who received zoledronic acid therapy using pharmacy
records. In addition to the number <strong>of</strong> zoledronic acid doses
administered and serial serum creatinine measurements, data
collected included patient age, sex, cancer diagnosis, comorbid
illnesses (diabetes or hypertension), pre-existing renal impairment,
concomitant medications (nonsteroidal anti-inflammatory
drugs [NSAIDs], diuretics, or aspirin), and current or prior
therapy with either thalidomide (Thalomid) or cisplatin.
ZOLEDRONIC ACID ADMINISTRATION
Between 1/10/02 and 1/30/04, 3,246 doses <strong>of</strong> zoledronic
acid (4 mg in 100 mL <strong>of</strong> normal saline) were administered
to 475 individual patients. No dose adjustments were made
according to baseline serum creatinine levels, as this was
not standard practice at the time. All patients were treated
at Fox Chase Cancer Center, Philadelphia, Pa. Adequate
follow-up was not available for 29 patients, and they were
excluded from the study population for the purposes <strong>of</strong>
analysis. Of the 3,211 doses <strong>of</strong> zoledronic acid administered
to the remaining 446 patients, 3,115 doses were evaluable
for statistical purposes (serum creatinine measurement
recorded subsequent to the last dose administered).
Patient and treatment characteristics are outlined in Table
1. There was a wide distribution <strong>of</strong> tumor types, and the
most common were breast, prostate, and lung cancers. Of
the 446 patients, 39 patients (8.7%) had evidence <strong>of</strong> renal
impairment prior to initial therapy with zoledronic acid,
Table 1
Patient and Treatment Characteristics (n = 446)
CHARACTERISTIC n (%)
Sex
Male 222 (50)
Female 224 (50)
Age, years
Median 64
Range 20–89
Tumor type
Breast 136 (30)
Prostate 115 (26)
Lung 79 (18)
Myeloma 35 (8)
Renal 19 (4)
Other 62 (14)
Comorbidity
Renal impairment a 39 (9)
Hypertension 169 (38)
Diabetes 58 (13)
Concomitant medication
NSAIDs 108 (24)
Diuretics 77 (17)
Aspirin 60 (13)
Current/prior therapy
Cisplatin 15 (3)
Thalidomide 13 (3)
Doses <strong>of</strong> zoledronic acid, n
Total 3,115
Median 4
Mean 6.98
Range 1–28
a
Baseline serum creatinine level ≥ 1.4 mg/dL
Abbreviation: NSAIDs = nonsteroidal anti-inflammatory drugs
with baseline serum creatinine levels ≥ 1.4 mg/dL (range,
1.4–3.6 mg/dL). In the majority <strong>of</strong> cases, zoledronic acid
was administered on a 3- to 4-week schedule once initiated,
although a minority <strong>of</strong> patients received less frequent
dosing. In all cases, the dose administered was 4 mg over
15 minutes. All patients were included in the analysis irrespective
<strong>of</strong> other treatments received.
RENAL DETERIORATION
VOLUME 4, NUMBER 10 ■ NOVEMBER/DECEMBER 2006
www.SupportiveOncology.net
525

<strong>Impact</strong> <strong>of</strong> <strong>Zoledronic</strong> <strong>Acid</strong> on Renal Function in Patients With Cancer
Table 2
Overall Change in Serum Creatinine Levels (n = 446)
SERUM CREATININE LEVEL (mg/dL)
VALUE BASELINE (SCR 1 ) FINAL (SCR 2 ) CHANGE (SCR 2 –SCR 1 ) P VALUE
Median 0.9 0.9 0 –
Mean 0.9 1.0 0.06 0.1730
Range 0.3–3.6 0.3–8.0 –1.1–4.4 –
nine level (< 1.4 mg/dL), or ≥ 1 mg/dL in those with pre-existing
renal impairment (≥ 1.4 mg/dL). Research records, clinical
charts, and laboratory records were reviewed so that all patients
who developed renal deterioration were identified. Also noted
was whether this deterioration required a change in management
or acute intervention (clinically significant) or did not
lead to a change in therapy (not clinically significant).
STATISTICAL ANALYSIS
The Wilcoxon signed-ranks test was used to compare the
SCr 1 and SCr 2 values observed for each patient and to determine
whether the cohort as a whole demonstrated a significant
decline in serum creatinine level over the course <strong>of</strong>
treatment. Binary logistic regression was used to evaluate the
association between renal deterioration, as defined previously,
and the patient-specific characteristics summarized in Table 1.
For this analysis, the indicator variable that identified subjects
manifesting renal deterioration constituted the dependent
variable. The patient characteristics were individually assessed
for an association with renal deterioration (univariate
analysis) and were subsequently included in a stepwise variable
selection procedure (multivariate analysis) to identify a
subset <strong>of</strong> statistically significant and independent predictors <strong>of</strong>
renal deterioration.
A receiver operating characteristic (ROC) curve analysis
Table 3
Characteristics <strong>of</strong> Patients Who Experienced
Renal Deterioration (n = 42)
CHARACTERISTIC
Sex
Male 22
Female 20
Age, years
Median 68.5
Range 43–87
Doses received, n
Median 4
Range 1–9
Rise in creatinine level, mg/dL
Median 1.0
Range 0.5–4.4
Renal deterioration is defined as a rise in serum creatinine level <strong>of</strong> ≥ 0.5 mg/dL in patients
with a normal baseline creatinine level (< 1.4 mg/dL) or ≥ 1 mg/dL in those with pre-existing
renal impairment (≥ 1.4 mg/dL).
n
was used to evaluate the performance <strong>of</strong> the logistic regression
model fit to the selected subset <strong>of</strong> factors. Specifically, the logistic
model estimated the conditional probability that a patient
exhibited renal deterioration given the patient’s observed values
for the selected set <strong>of</strong> characteristics. Each realized value
<strong>of</strong> the estimated conditional probability was then used in the
ROC analysis as a threshold for the identification <strong>of</strong> renal dysfunction;
that is, a patient was identified as having experienced
renal deterioration if the conditional probability estimated for
that patient exceeded a given threshold. Thus, the sensitivity
and specificity for the detection <strong>of</strong> renal deterioration could be
estimated for each value <strong>of</strong> the conditional probability.
The ROC analysis assesses the overall performance <strong>of</strong> the logistic
regression model through the area under the ROC curve
(AUC), representing the area under the graph <strong>of</strong> sensitivity
(ie, the true positive faction) versus one minus specificity (false
positive fraction). All statistical computations were carried out
using SAS s<strong>of</strong>tware, version 9.0 (SAS Institute Inc., Cary, NC,
2002). Results were declared statistically significant only if associated
with a comparisonwise two-sided P value < 0.05.
Results
OVERALL CHANGE IN RENAL FUNCTION
The results for overall change in renal function are listed in
Table 2. In the overall cohort, there was no significant change in
median serum creatinine levels with zoledronic acid therapy. Using
a signed-rank test, the change from SCr 1 to SCr 2 (calculated
as SCr 2 –SCr 1 ) was not statistically significant (P = 0.1730).
RENAL DETERIORATION
In total, 42 patients (9.4%) had a significant deterioration
in serum creatinine levels. Patient characteristics for this cohort
are outlined in Table 3. These patients received a median
<strong>of</strong> 4 doses <strong>of</strong> zoledronic acid (range, 1–9 doses) and the median
rise in serum creatinine level was 1.0 mg/dL (range, 0.5–4.4
mg/dL). No patient required dialysis, and no patient died as a
result <strong>of</strong> zoledronic acid-induced renal dysfunction.
UNIVARIATE ANALYSIS
Table 4 summarizes the results obtained when the patient
and treatment-related factors were tested individually for an
association with renal deterioration. Factors identified by the
binary logistic regression analysis as statistically significant were
age, pre-existing renal impairment, number <strong>of</strong> zoledronic acid
526 www.SupportiveOncology.net THE JOURNAL OF SUPPORTIVE ONCOLOGY

McDermott, Kloth, Wang, et al
Table 4
Univariate Analysis for Predictors <strong>of</strong> Deterioration in Renal Function
FACTOR VARIABLE PATIENTS (n) EVENTS (n) P VALUE a ODDS RATIO [95% CI]
Gender Male 222 22 0.7229 1.12 [0.59, 2.12]
Female 224 20
Age a 446 42 0.0160 1.04 [1.01, 1.07]
Baseline creatinine level < 1.4 mg/dL 407 34 0.0167 0.35 [0.15, 0.83]
≥ 1.4 mg/dL 39 8
Doses a 446 42 0.0455 1.05 [1.00, 1.09]
Diabetes No 388 33 0.0933 0.51 [0.23, 1.12]
Yes 58 9
Hypertension No 277 19 0.020 0.47 [0.25, 0.89]
Yes 169 23
NSAID, current use No 338 24 0.0040 0.38 [0.20, 0.74]
Yes 108 18
Diuretic, current use No 369 30 0.0454 0.48 [0.23, 0.98]
Yes 77 12
Aspirin, current use No 386 36 0.8680 0.93 [0.37, 2.30]
Yes 60 6
Cisplatin, current or prior use No 431 39 0.1670 0.40 [0.11, 1.47]
Yes 15 3
Thalidomide, current or prior use No 433 38 0.0142 0.22 [0.06, 0.74]
Yes 13 4
Cancer diagnosis Myeloma 35 5 0.0276 1.88 [0.68, 5.17]
(myeloma vs other)
Renal 19 5 4.02 [1.36, 11.87]
(renal vs other)
Other 392 32 0.47 [0.12, 1.88]
(myeloma vs renal)
Bold numbers represent statistical significance.
a
Treated as a continuous variable. Abbreviations: CI = confidence interval; NSAID = nonsteroidal anti-inflammatory drug
doses received, a diagnosis <strong>of</strong> renal cell carcinoma or myeloma,
hypertension, concomitant therapy with NSAIDs or diuretics,
and current or prior therapy with thalidomide.
MULTIVARIATE ANALYSIS
Stepwise variable selection in the context <strong>of</strong> binary logistic
regression identified the following subset <strong>of</strong> factors significantly
and independently associated with renal deterioration: patient
age, a diagnosis <strong>of</strong> myeloma or renal cell cancer, the number
<strong>of</strong> doses received, concomitant therapy with an NSAID, and
current or prior therapy with cisplatin (Table 5).
PREDICTIVE MODEL
The subset <strong>of</strong> factors found to be significant on multivariable
analysis was used to construct a logistic regression model
to predict patient risk <strong>of</strong> developing renal deterioration while
on therapy with zoledronic acid. According to this model, a
patient is classified as having experienced renal deterioration
only if his or her predicted risk (ie, estimated conditional
probability) is greater than a selected threshold. In particular,
if the threshold is set at 10%, as is consistent with the available
literature on the incidence <strong>of</strong> renal deterioration with the
use <strong>of</strong> zoledronic acid, the logistic regression prediction model
to identify patients with renal deterioration was observed to
have a sensitivity <strong>of</strong> 64.3% and a specificity <strong>of</strong> 69.8%. Figure
1 shows the ROC curve associated with the multivariable logistic
regression model. The area under the ROC curve was
estimated to be 0.75.
CLINICAL SIGNIFICANCE
The patient care pathway subsequent to the development <strong>of</strong>
renal deterioration is outlined in Figure 2. In 19 cases, zoledronic
acid therapy was continued. Two <strong>of</strong> these patients had developed
obstructive uropathy due to progressive disease in the pelvis
(one with prostate cancer, one with renal cancer). Following
urologic intervention, serum creatinine levels normalized, and
zoledronic acid therapy was resumed. In 17 cases, zoledronic
acid was continued as the treating physician thought that the
rise in serum creatinine levels was not clinically significant. In
nine <strong>of</strong> these cases, serum creatinine levels subsequently fell
with continued therapy so they no longer fulfilled the definition
<strong>of</strong> renal deterioration. In seven cases, serum creatinine levels
remained stable with continued therapy, fulfilling the criteria
for renal deterioration, and in one case, although further zoledronic
acid was planned, it had not yet been administered by
study end.
In 23 cases, no further zoledronic acid therapy was administered
after the observation <strong>of</strong> renal deterioration. Of this total,
VOLUME 4, NUMBER 10 ■ NOVEMBER/DECEMBER 2006
www.SupportiveOncology.net
527

<strong>Impact</strong> <strong>of</strong> <strong>Zoledronic</strong> <strong>Acid</strong> on Renal Function in Patients With Cancer
Table 5
Multivariate Analysis for Predictors <strong>of</strong>
Deterioration in Renal Function
DEGREES OF PARAMETER
FACTOR FREEDOM ESTIMATE a P VALUE
Cancer diagnosis 2 11.24 0.0036
(renal or myeloma vs other)
Age 1 10.14 0.0015
Number <strong>of</strong> doses 1 7.82 0.0052
Current NSAID use 1 7.10 0.0077
Current or prior cisplatin use 1 4.16 0.0414
a
Wald chi-square
Abbreviation: NSAID = nonsteroidal anti-inflammatory drug
15 <strong>of</strong> these patients experienced a rise in serum creatinine levels
at a time <strong>of</strong> significant disease progression. These patients
received no further active therapy for their advanced cancer
and died a median <strong>of</strong> 17 days (range, 1–120 days) after their
last zoledronic acid treatment. In all but three cases, the patient
died before the next dose <strong>of</strong> zoledronic acid would have been
scheduled. None <strong>of</strong> these patients required dialysis, and in no
case was renal dysfunction believed to be the cause <strong>of</strong> death.
Finally, in eight cases, zoledronic acid was stopped due to what
was thought to be a clinically significant rise in serum creatinine
levels. In each case, creatinine levels stabilized or improved following
discontinuation <strong>of</strong> therapy, indicating a probable contribution
<strong>of</strong> zoledronic acid to the deterioration in renal function.
Discussion
The potential for intravenous bisphosphonate therapy to
provoke renal insufficiency is well documented. 7,9,10 In the era
<strong>of</strong> the more potent bisphosphonate zoledronic acid, what is less
clear is how common and significant a problem this is in clinical
Sensitivity
1.0
0.8
0.6
0.4
0.2
0.0
Figure 1
0.0
AUC ROC = 0.75
0.2 0.4 0.6 0.8 1.0
Specificity
ROC Curve to Evaluate the Logistic Regression
Model for Predicting Renal Deterioration
in Patients Treated With <strong>Zoledronic</strong> <strong>Acid</strong>
Abbreviations: ROC = receiver operating characteristic; AUC = area under
the curve
Continue zoledronic
acid (n = 19)
Treated
obstructive
uropathy; no
further rise
in creatinine
level (n = 2)
Figure 2
Creatinine
level improved
(n = 9)
Renal deterioration
(n = 42)
Not
clinically
significant
(n = 17)
Creatinine
level stabilized
(n = 7)
No further zoledronic
acid (n = 23)
Creatinine
level
stable or
improved
(n = 8)
Not
evaluable
(n = 1)
Subsequent Care Pathway in Patients
Who Developed Renal Deterioration
Died <strong>of</strong>
progressive
disease with
further active
treatment
(n = 15)
practice and whether renal deterioration can be predicted using
known risk factors. In this review <strong>of</strong> patients treated with zoledronic
acid (4 mg over 15 minutes every 4 weeks) at Fox Chase
Cancer Center, the incidence <strong>of</strong> renal deterioration (9.4%
[42/446 patients] after a median <strong>of</strong> four doses and a mean <strong>of</strong> 7
doses) is broadly consistent with published data. 2–5,11
On multivariate analysis, predictive factors for the development
<strong>of</strong> renal deterioration included patient age, the number
<strong>of</strong> doses administered, current use <strong>of</strong> NSAIDs, current or prior
therapy with cisplatin, and a diagnosis <strong>of</strong> myeloma or renal
cell carcinoma. Some <strong>of</strong> these factors have been previously described,
2,3,5,7 whereas others represent new findings The FDA
has publicized episodes <strong>of</strong> acute renal failure with zoledronic
acid in patients with myeloma and has recommended strict
monitoring <strong>of</strong> serum creatinine levels. 7 Given the potential
impact <strong>of</strong> the myelomatous disease process on renal function, 12
this recommendation appears warranted. Similarly, patients
with metastatic renal cell carcinoma typically have only one
kidney, as resection <strong>of</strong> the primary tumor is warranted, even
in the presence <strong>of</strong> metastatic disease. 13 This condition may
predispose them to adverse sequelae from nephrotoxic drugs.
Although its nephrotoxic potential is well described, 14
cisplatin is not commonly used in patients with metastatic
breast or prostate cancers, who form the majority <strong>of</strong> patients
receiving therapy with zoledronic acid. Therefore, any potential
impact would be limited primarily to patients with lung
cancer. On the other hand, due to their combined anti-inflammatory
and analgesic properties, NSAIDs are frequently
prescribed for patients with bony metastases from any source,
even though their use can be associated with the development
<strong>of</strong> both acute 15 and chronic 16 renal failure.
528 www.SupportiveOncology.net THE JOURNAL OF SUPPORTIVE ONCOLOGY

McDermott, Kloth, Wang, et al
Because these patients have advanced or metastatic cancer,
it is possible that the number <strong>of</strong> doses received may be a surrogate,
at least in part, for progression <strong>of</strong> disease and overall
deterioration in multiple systems. In licensing studies, the incidence
<strong>of</strong> renal deterioration appeared to be related to the time
on study, whether the patients were receiving zoledronic acid,
pamidronate, or placebo. 2,3,5,11 This relation was also reflected in
the analysis <strong>of</strong> our own patient population with renal deterioration,
in many <strong>of</strong> whom it was a preterminal event. Finally, age
is an accepted risk factor for the development <strong>of</strong> renal failure,
most likely due to the accumulation <strong>of</strong> comorbid illnesses. 17
On multivariate analysis, we did not find a significant relationship
between an elevated baseline serum creatinine level
and renal deterioration, as described in the package insert. 18 This
finding may be attributable to the fact that the degree <strong>of</strong> baseline
renal function impairment was not severe in these patients. In a
small pharmacokinetic study <strong>of</strong> 19 cancer patients stratified by
creatinine clearance, 19 dosage adjustment <strong>of</strong> zoledronic acid did
not appear to be necessary in patients with mild to moderate renal
impairment. Nonetheless, zoledronic acid is not recommended in
patients with serum creatinine levels > 3 mg/dL, and caution is
advisable in all patients with impaired renal function. 18
Using these risk factors, we have constructed a model that
can be used to predict the likelihood <strong>of</strong> an individual patient
developing renal deterioration with the use <strong>of</strong> zoledronic acid.
We chose a cut<strong>of</strong>f point <strong>of</strong> 10%, as it is similar to the previously
published incidence <strong>of</strong> renal deterioration with zoledronic
acid 2–5 and represents a level <strong>of</strong> risk above which clinicians may
wish to proceed more cautiously. The risk-factor pr<strong>of</strong>ile and the
accompanying model potentially could be used by physicians involved
in the treatment <strong>of</strong> patients with metastatic cancer when
calculating the risk-benefit ratio associated with zoledronic acid
therapy as an adjunct to other anticancer therapies.
It is important to note, however, that the review <strong>of</strong> the cases
in this series in which there was evidence <strong>of</strong> renal deterioration
demonstrated that the rise in creatinine level was <strong>of</strong>ten
mild and not clinically significant in the eyes <strong>of</strong> the treating
physician. In many instances, zoledronic acid therapy was continued
without sequelae. Moreover, in other patients, renal
deterioration was a preterminal event, with evidence <strong>of</strong> widely
progressive disease, a general deterioration in performance
status, and a switch to symptom-based care. This finding is
consistent with the fact that once bisphosphonate therapy is
begun for treatment <strong>of</strong> bone metastasis, it is <strong>of</strong>ten continued
until the decision is made to end active cancer therapy. Nonetheless,
there are patients in whom use <strong>of</strong> zoledronic acid will
induce a rise in serum creatinine level that may drop upon
discontinuation <strong>of</strong> the drug, and our model is unable to differentiate
between these differing patient types.
Conclusion
<strong>Zoledronic</strong> acid must be regarded as a potentially nephrotoxic
agent, similar to many other medications used in the
treatment <strong>of</strong> patients with cancer. However, the impact <strong>of</strong>
monitoring renal function before each dose <strong>of</strong> zoledronic
acid effectively negates the time advantage <strong>of</strong> the infusion,
as patients typically must wait for laboratory results before
commencing therapy. Regular monitoring <strong>of</strong> serum creatinine
levels in these patients represents good clinical practice;
there are many reasons why renal function might deteriorate
in patients with advanced disease. Based on the available
data, monitoring <strong>of</strong> serum creatinine levels before each dose
<strong>of</strong> zoledronic acid is given appears warranted in patients at
high risk for renal deterioration; however, compulsive creatinine
level monitoring is burdensome and expensive and may
not be necessary in all patients.
References
1. Green JR, Muller K, Jaeggi KA. Preclinical pharmacology
<strong>of</strong> CGP 42’446, a new, potent, heterocyclic bisphosphonate
compound. J Bone Miner Res 1994;9:745–751.
2. Rosen LS, Gordon D, Tchekmedyian S, et al.
<strong>Zoledronic</strong> acid versus placebo in the treatment <strong>of</strong>
skeletal metastases in patients with lung cancer and
other solid tumors: a phase III, double-blind, randomized
trial—the <strong>Zoledronic</strong> <strong>Acid</strong> Lung Cancer and Other Solid
Tumors Study Group. J Clin Oncol 2003;21:3150–3157.
3. Saad F, Gleason DM, Murray R, et al. A randomized,
placebo-controlled trial <strong>of</strong> zoledronic acid in patients
with hormone-refractory metastatic prostate carcinoma.
JNCI Cancer Spectrum 2002;94:1458–1468.
4. Ibrahim A, Scher N, Williams G, et al. Approval
summary for zoledronic acid for treatment <strong>of</strong> multiple
myeloma and cancer bone metastases. Clin Cancer Res
2003;9:2394–2399.
5. Rosen LS, Gordon D, Kaminski M, et al. <strong>Zoledronic</strong>
acid versus pamidronate in the treatment <strong>of</strong> skeletal
metastases in patients with breast cancer or osteolytic
lesions <strong>of</strong> multiple myeloma: a phase III, double-blind,
comparative trial. Cancer J 2001;7:377–387.
6. Markowitz GS, Fine PL, Stack JI, et al. Toxic acute
tubular necrosis following treatment with zoledronate
(Zometa). Kidney Int 2003;64:281–289.
7. Chang JT, Green L, Beitz J. Renal failure with the use
<strong>of</strong> zoledronic acid. N Engl J Med 2003;349:1676–1679.
8. Weinman EJ, Patak RV. Acute renal failure in
cancer patients. Oncology (Huntingt) 1992;6:47–52;
discussion 52–54.
9. Banerjee D, Asif A, Striker L, et al. Short-term, highdose
pamidronate-induced acute tubular necrosis: the
postulated mechanisms <strong>of</strong> bisphosphonate nephrotoxicity.
Am J Kidney Dis 2003;41:E18.
10. Markowitz GS, Appel GB, Fine PL, et al. Collapsing
focal segmental glomerulosclerosis following treatment
with high-dose pamidronate. J Am Soc Nephrol
2001;12:1164–1172.
11. Rosen LS, Gordon D, Kaminski M, et al. Longterm
efficacy and safety <strong>of</strong> zoledronic acid compared
with pamidronate disodium in the treatment <strong>of</strong>
skeletal complications in patients with advanced
multiple myeloma or breast carcinoma: a randomized,
double-blind, multicenter, comparative trial. Cancer
2003;98:1735–1744.
12. Blade J, Fernandez-Llama P, Bosch F, et al. Renal
failure in multiple myeloma: presenting features and
predictors <strong>of</strong> outcome in 94 patients from a single
institution. Arch Intern Med 1998;158:1889–1893.
13. Flanigan RC, Salmon SE, Blumenstein BA, et al.
Nephrectomy followed by interferon alfa-2b compared
with interferon alfa-2b alone for metastatic renal-cell
cancer. N Engl J Med 2001;345:1655–1659.
14. Ries F, Klastersky J. Nephrotoxicity induced by
cancer chemotherapy with special emphasis on cisplatin
toxicity. Am J Kidney Dis 1986;8:368–379.
15. Oates JA, FitzGerald GA, Branch RA, et al. Clinical
implications <strong>of</strong> prostaglandin and thromboxane A2
formation. N Engl J Med 1988;319:761–767.
16. Segasothy M, Samad SA, Zulfigar A, et al. Chronic
renal disease and papillary necrosis associated with the
long-term use <strong>of</strong> nonsteroidal anti-inflammatory drugs
as the sole or predominant analgesic. Am J Kidney Dis
1994;24:17–24.
17. Berthoux F, Gellert R, Jones E, et al. Epidemiology
and demography <strong>of</strong> treated end-stage renal failure
in the elderly: from the European Renal Association
(ERA-EDTA) Registry. Nephrol Dial Transplant 1998;13
(suppl):65–68.
18. Zometa [package insert]. East Hanover, NJ:
Novartis Pharmaceuticals Corporation; 2003.
19. Skerjanec A, Berenson J, Hsu C, et al. The pharmacokinetics
and pharmacodynamics <strong>of</strong> zoledronic acid in
cancer patients with varying degrees <strong>of</strong> renal function.
J Clin Pharmacol 2003;43:154–162.
VOLUME 4, NUMBER 10 ■ NOVEMBER/DECEMBER 2006
www.SupportiveOncology.net
529