Multicenter Randomized Controlled Clinical Trial to
`
`Evaluate Cardioprotection of Dexrazoxane Versus No
`Cardioprotection in Women Receiving Epirubicin
`Chemotherapy for Advanced Breast Cancer
`
`By Marco Venturini, Andrea Michelotti, Lucia Del Mastro, Luigi Gallo, Flavio Carnino, Ornella Garrone,
`Carmelo Tibaldi, Nicola Molea, Riccardo Calogero Bellina, Paolo Pronzato, Patricia Cyrus, Jan Vinke, Franco Testore,
`Monica Guelfi, Rita Lionetto, Paolo Bruzzi, Pier Franco Conte, and Riccardo Rosso
`
`Pur
`
`se: Dexrazoxane was found effective in reduc-
`
`ing doxorubicin cardiotoxicity when given ata dose ratio
`(dexrazoxane:doxorubicin) of 20:1. Preclinical studies
`indicated that dexrazoxane at a dose ratio of 10 to 15:1
`also protected against epirubicin-induced cardiotoxicity.
`The main obiective of this study was to investigate the
`efficacy of dexrazoxane, given at a dose ratio of 10:1
`against epirubicin cardiotoxicity.
`Patients and Methods: One hundred sixty-two ad-
`vanced breast cancer patients were randomized to re-
`ceive epirubicin-based chemotherapy with or without
`dexrazoxane. Patients who had previously received ad-
`iuvant chemotherapy that contained anthracyclines
`were treated with cyclophosphamide 600 mg/m2 intra-
`venously (IV), epirubicin 60 mg/m2 IV, and fluorouracil
`600 mg/m2 IV, on day 1 every 3 weeks. The other pa-
`tients were treated with epirubicin 120 mg/m2 IV on
`day 1 every 3 weeks. Cardiac toxicity was defined as
`clinical signs of congestive heart failure, a decrease in
`resting left ventricular ejection fraction (LVEF) to s 45%,
`
`HE USE OF DOXORUBICIN, a drug with an estab-
`lished place in cancer therapy and among the most
`active single agents in the treatment of advanced breast
`cancer,
`is particularly limited by its cumulative dose-
`related cardiotoxicity. In an effort to reduce this toxicity,
`numerous analogs of doxorubicin have been developed.
`One such analog, 4'-epidoxorubicin (epirubicin), does not
`
`From the Divisione di Oncologia Medica 1 and Servizio di Epide-
`miologia Clinica e Sperimentazioni Cliniche Controllate,
`lstituto
`Nazionale per la Ricerca sul Cancro; Servizio di Oncologia Medica,
`Ospedale Galliera. Genova; Unita Operativa di Oncologia e Medi-
`cina Nucleare, Ospedale S. Chiara, Pisa: Ginecologia, Ospedale S.
`Anna, Torino; Ospedale S. Andrea, La Spezia; Unita Operativa di
`Oncologia, Ospedale Civile, Asti, Italy; and Chiron, Amsterdam, the
`Netherlands.
`Submitted November 27, 1995; accepted May I, 1996.
`Presented in part at the Thirty—First Annual Meeting of the Ameri-
`can Society of Clinical Oncology, Los Angeles, CA, May 20-23,
`1995.
`
`Supported in part by Associazione Italiana per la Ricerca sul
`Cancro, Milano, Italy, 1996.
`Address reprint requests to Marco Venturini, MD, Divisione di
`Oncologia Medica 1, Istituto Nazionale per la Ricerca suI Cancro,
`Viale Benedetto XV, 10, 16132 Genova. Italy; Email marco.ventu
`rini@schering-pl.it.
`© 1996 by American Society of Clinical Oncology.
`0732-183X/96/1412-0012$3.00/0
`
`or a decrease from baseline resting lVEF of a 20 EF
`units.
`
`Results: One hundred sixty patients were evaluated.
`Caaiotoxicity was recorded in 18 of 78 patients (23.1%)
`in the control arm and in six of 82 (7.3%) in the dexrazox-
`one arm. The cumulative probability of developing cardio-
`toxicity was significantly lower in dexrazoxane-treated pa-
`tients than in control patients (P = .006; odds ratio, 0.29;
`95% confidence limit [Cl], 0.09 to 0.78). Noncardiac toxic-
`ity, obiective response, progression-free survival, and over-
`all survival were similar in both arms.
`
`Conclusion: Dexrazoxane given at a dexrazoxano:epi-
`rubicin dose ratio of 10:1 protects against epirubicin—in-
`duced cardiotoxicity and does not affect the clinical activity
`and the noncardiac toxicity of epirubicin. The clinical use
`of dexrazoxane should be recommended in patients whose
`risk of developing cardiotoxicity could hamper the eventual
`use and possible benefit of epirubicin.
`J Clin Oncol 14:3112-3120. © 1996 by American So-
`ciety of Clinical Oncology.
`
`eliminate cardiotoxicity, but has been shown to be less
`cardiotoxic and myelotoxic than the parent compound at
`equimolar doses.I Although epirubicin has commonly
`been given at a dose of 75 to 90 mg/m2 every 3 weeks,
`it recently has been reported2 that epirubicin can be given
`safely at higher doses than previously believed. Two ran-
`domized trials performed in advanced breast cancer pa-
`tients that compared higher doses of epirubicin versus
`lower doses showed an improvement in terms of response
`rate14 and time to progression.4 A recent report5 suggested
`a possible advantage of using high-dose epirubicin in the
`adjuvant setting. Hence, the problem of cardiotoxicity
`may now become clinically important when using high
`doses of epirubicin.
`A different approach to reduce the anthracycline car—
`diotoxicity is the use of cardioprotective agents. Genera-
`tion of free radicals is believed to be among the most
`important mechanisms of anthracycline-induced cardio-
`toxicity.6 Dexrazoxane is an iron— and copper—chelating
`agent that was found effective in reducing cardiotoxicity
`of doxorubicin both in preclinical7 and clinical settings.a
`The mechanism of action for dexrazoxane is not fully
`elucidated, but it has been postulated that it exerts its
`cardioprotective effect by chelating iron, thus preventing
`the generation of oxygen free radicals by anthracyclines.
`A randomized clinical study in advanced breast cancer
`
`3ll2
`
`Journal of Clinical Oncology, Vol 14, Na 12 (December), 1996: pp 3112-3120
`
`Hospira v. Genentech
`Downloaded from asconuhsnrg hv Renrints Desk on November 27. 2017 from 216.185.156.028
`lPR201 7-0073?
`
`Genentech 2103
`
`Genentech 2103
`Hospira v. Genentech
`IPR2017-00737
`
`

`

`DEXRAZOXANE AND EPIRUBICIN CARDIOTOXICITY
`
`3113
`
`patients showed that dexrazoxane given at a ratio 20:1 of
`dexrazoxanezdoxorubicin effectively reduced doxorubi—
`cin—induced cardiotoxicity. The addition of dexrazoxane
`did not affect the activity of the chemotherapy regimen,
`but mildly increased myelotoxicity.“1
`Animal studies suggested that cardiac protection could
`be achieved with a lower dexrazoxanezanthracyclines ra-
`tio, ie, 10 to 15:1.7 Although the cardioprotective effect
`of dexrazoxane was shown against epirubicin~induced
`cardiotoxicity in preclinical studies,6 no phase III trial
`was undertaken. On this basis, we designed a randomized
`trial to investigate the efficacy of dexrazoxane, given at
`a ratio of 10:1, as a cardioprotective agent against epiru-
`bicin-induced cardiotoxicity in advanced breast cancer
`patients.
`
`PATIENTS AND METHODS
`
`Patient Eligibility
`
`Patients were eligible if the following criteria were met: clinical
`or hystologic evidence of metastatic, locally advanced (III B) or
`inflammatory breast cancer; either measurable, or assessable disease;
`Eastern Cooperative Oncology Group (ECOG) performance status
`(PS) 5 2; and WBC count 2 4,000 X 10‘IL, platelet count (PLT)
`2 100 X log/L, and hemoglobin (Hgb) level 2 10 g/dL. Patients
`who had received radiotherapy were eligible, provided that the radio«
`therapy was completed more than 2 weeks before study entry. Pa—
`tients had to have a baseline resting left ventricular ejection fraction
`
`(LVEF) 2 50%, determined by multigated radionuclide (MUGA)
`scan. Other eligibility criteria were no previous or concomitant ma-
`lignancy, and no other serious medical or psychiatric illness that
`would preclude informed consent and/or intensive treatment. Prior
`adj uvant chemotherapy was allowed. Any prior horrnonotherapy was
`acceptable, provided that these were discontinued 2 weeks before
`study start.
`Criteria for exclusion were myocardial infarction within the previ-
`ous year; history of congestive heart failure, unless full recovery
`was documented; symptomatic valvular heart disease; prior chemo-
`therapy, with the exception of adjuvant chemotherapy; and bilirubin
`level a 3 mg/dL and creatinine concentration 2 2 mg/dL.
`The study was approved by the Protocol Review Committee and
`Ethical Committee of the Istituto Nazionale per la Ricerca sul Can-
`cro, and of each collaborating center. Informed consent was obtained
`from all patients before study entry.
`
`Study Design and Treatment
`
`Patients undergoing epirubicin-containing chemotherapy were
`randomized to receive dexrazoxane at the ratio of 10:1 or no addi-
`tional treatment (Fig 1).
`Patients who had received previous adjuvant chemotherapy that
`contained anthracyclines (doxorubicin or epirubicin) were treated
`with cyclophosphamide 600 mg/m2 intravenously (IV), epirubicin
`60 mg/m2 IV, and fluorouracil 600 mg/m2 IV, on day 1 given every
`3 weeks (CEF). Patients randomized to dexrazoxane received this
`drug at a dose of 600 mgjm2 by IV infusion over 15 minutes, begin-
`ning 30 minutes before epirubicin.
`Patients who did not receive prior adjuvant anthracycline-con-
`taining therapy were treated with high—dose epirubicin (HD-Epi) 120
`mg/m2 IV, on day 1 given every 3 weeks. Patients randomized to
`
`Previous adjuvant chemotherapy
`containing anthracycliue
`
`/\.
`
`CEF
`
`Cyclophosphamide 600 mg/m2
`Epirubicin 60 mg/m2
`S-Fluorouracil 600 mg/m2
`
`Fig 1. Study design.
`
`HD-Epi
`
`Epirubicin 120 mg/m2
`
`DEXRAZOXANE (10:1)
`
`CONTROL
`
`Downloaded from asconnhs org hv Renrints Desk on November 27 2017 from 216 IRS lid 02%
`
`

`

`3114
`
`VENTURlNI ET AL
`
`dexrazoxane received it at a dose of 1,200 mg/m2 by IV infusion
`over 15 minutes, 30 minutes before HD-Epi.
`Dexramxane (Cardioxane) was provided by Chiron BV, Amster-
`dam, the Netherlands. Each vial contained 500 mg of dexrazoxane
`as lyophilized powder and was reconstituted with 25 rnL of sterile
`water. The contents of the appropriate number of vials were further
`diluted with ringer lactate solution to a volume of 500 mL.
`Cycles were repeated every 3 weeks provided the WBC count
`was 2 4,000 X lO‘IL and PLT a 100 X 10%; otherwise treatment
`was delayed until hematologic recovery. If the WBC count was 5
`2,000 X 106/L and/or PLT s 50 x 109/L, treatment was delayed
`and the doses of all drugs reduced by 25%. During therapy, any
`dose reduction of epirubicin was followed by an equivalent dose
`reduction of dexrazoxane to maintain the ratio between these two
`
`drugs at 10:1. The doses of both epirubicin and dexrazoxane were
`reduced by 50% in case of a bilirubin value between 1.5 and 3.0
`mg/dL, and by 75% for a value more than 3.0 mg/dL. For other
`toxicities, appropriate dose reductions were made. No dose escala-
`tion of any drug and no prophylactic use of antibiotics or hematopoi-
`etic growth factors was planned.
`
`Study Parameters
`At the start of therapy, all patients underwent history, cardiac
`assessment, tumor assessment, physical examination, evaluation of
`PS, hematology (Hgb, WBC count, absolute neutrophil count, PLT),
`and blood chemistries (blood urea nitrogen [BUN], creatinine, biliru—
`bin, alkaline phosphatase, lactate dehydrogenase [LDH], total pro-
`tein, albumin, glucose, AST, ALT, and serum electrolytes [sodium,
`potassium, and calcium]). Cardiac monitoring consisted of ECG,
`chest x-rays, and MUGA scan. Tumor assessment was made by
`chest x-rays, bone scan with x—rays of abnormal areas, and other x-
`rays or relevant scans if indicated. Whenever possible. a bidimensio-
`nal measurement was performed.
`Hematology, blood chemistry, and evaluation of toxicity ac-
`cording to World Health Organization (WHO) criteria9 were per—
`formed every cycle. Grading of clinical signs of cardiac toxicity was
`performed according to the New York Heart Association (NYHA)
`classification of cardiac status.lo Laboratory cardiac assessment, in-
`cluding ECG, chest x-rays, and multigated radionuclide (MUGA)
`scan, was performed every other cycle. The nuclear physicians and
`cardiologists were blinded to the treatment assignment of the pa-
`tients. Tumor response was evaluated after at least two courses of
`chemotherapy according to standard WHO criteria.9 Tumor evalua-
`tions were performed by repeating the same examinations used at
`baseline.
`
`Criteria for Stopping Therapy
`Patients with stable disease (SD) received no more than eight
`cycles. Responding patients (complete response [CR] or partial re-
`sponse [PRD continued treatment for more than eight cycles, until
`the development of disease progression or if at the discretion of the
`physician further treatment was without apparent benefit to the pa-
`tient. Patients were removed from the treatment if they developed
`progressive disease, grade IV nonhematologic toxicity, or cardiac
`toxicity. Cardiac toxicity was defined as clinical signs of congestive
`heart failure (CHF) classified as 2 grade 2 according to NYHA, a
`decrease in resting LVEF as measured by MUGA scan to s 45%,
`or a decrease from baseline resting LVEF of 2 20 BF units.
`
`Statistical Considerations
`
`This was a randomized multiinstitutional controlled clinical study.
`All patients were stratified before randomization by institution and
`
`according to previously received adjuvant chemotherapy widi an-
`thracyclines. Randomization was performed by a phone call to the
`study coordination center.
`The primary objective of the study was to compare the two treat-
`ment arms in terms of incidence of cardiotoxicity as defined earlier.
`The sample size was based on the estimation that 30% of patients
`who receive anthracycline treatment develop a cardiac event ac-
`cording to the aforementioned definitions. To have an 80% power
`against the hypothesis that dexrazoxane reduces the proportion of
`patients who develop cardiotoxicity from 30% to 10% with a sig—
`nificance level of 0.05, it was estimated that 80 patients per arm
`would be required.
`Dose-intensity, defined as the amount of drug (milligrams per
`square meter) administered per unit time (week), was calculated
`according to the method reported by Hryniuk et a1.” As the doses
`of the drugs of CEF were uniformly reduced in the case of toxicity,
`the dose-intensity of all three drugs corresponded to the dose-inten—
`sity of any one drug. Hence, only the dose-intensity of epirubicin
`was calculated. In patients treated with the CEF regimen, the cumula-
`tive dose of epirubicin was calculated by adding the dose of epiru-
`bicin received as adjuvant therapy to the dose of epirubicin received
`in this study. If patients received doxorubicin as adjuvant therapy,
`the dose was doubled. The ratio 2:1 of epirubicin/doxorubicin was
`chosen based on the ratio of the cumulative doses of these drug
`above which the probability of developing Cl-[F increases rapidly,
`ie, 450 to 550 mg/m2 for doxorubicin and 900 to 1,000 mg/ml for
`epirubicin.
`The X2 statistic (or Fisher’s exact test where appropriate) was
`used to analyze differences in categorical variables. The Kaplan-
`Meier method was used to estimate overall survival and progression-
`free survival. The log-rank test was used to assess differences be-
`tween curves. The cumulative probability of developing cardiac
`toxicity, while taking into account the different number of patients
`at the various dose levels of epirubicin, was computed as follows.
`According to the life-table method, for each cumulative dose level,
`the number of patients who experienced cardiac events between that
`level and the following one was divided by the number of patients
`at risk, ie, the number of patients treated at that level. The resulting
`dose-specific probabilities of cardiac events in each treatment arm
`were pooled to obtain the cumulative probability of cardiac events,
`which were compared using the Mantel—Haenszel x2 statistic, which
`takes into account the differing number of patients at each cumula-
`tive dose level. The Mantel-Haenszel odds ratio was also computed
`to provide an estimate of the protective effect of the study treatment
`while adjusting for the number of patients at each dose level. Patients
`who did not experience a cardiac event were considered censored
`after the last administration of epirubicin. Patients pretreated with
`anthracyclines in the adjuvant setting were entered onto the analysis
`at the corresponding cumulative dose (left censoring).
`According to the study protocol, both the cumulative probabilities
`of overall cardiotoxicity (clinical events plus laboratory events) and
`that of laboratory cardiotoxicity (only events observed by MUGA
`scan) were computed. In general, results were considered significant
`for P values less than .05. All P values are two-sided.
`
`RESULTS
`
`Between February 1992 and June 1994, 162 advanced
`breast cancer patients were enrolled. No statistically sig-
`nificant difference in main patients’ characteristics was
`recorded (Table 1). Only seven patients had PS 2, all in
`the dexrazoxane arm. Eleven patients in the control arm
`
`Downloaded front ascomrhs org hv Renrints Desk on November 27 20l 7 from 2| 6 185 I56 028
`
`

`

`DEXRAZOXANE AND EPIRUBICIN CARDIOTOXICITY
`
`3115
`
`and 14 in the dexrazoxane arm had received prior adju-
`vant chemotherapy that contained anthracyclines. No sig-
`nificant difference in the median cumulative dose of pre—
`viously received anthracyclines was recorded: 360 mg/
`m2 (range, 240 to 600) in the control arm, and 410 mg/m2
`(range, 180 to 800) in the dexrazoxane arm. The median
`baseline LVEF was 62.5% (range, 50 to 84) in the control
`arm and 63% (range, 51% to 84%) in the dexrazoxane
`am. No difference in cardiovascular risk factors between
`
`the control and dexrazoxane arms was recorded: diabetes,
`2.6% v 1.2%; hypertension, 14.1% v 9.5%; age 2 65
`years, 19.2% v 22.6%; and radiation therapy to the medi-
`astinum or left chest wall, 16.6% v 13.1%, respectively.
`Two patients, both randomized to the dexrazoxane arm,
`never received chemotherapy and were considered nonas—
`sessable. Two other patients were not eligible: one patient
`in the control arm because the presence of metastatic
`
`Table I. Patient Characteristics
`Control
`Dexrazaxane
`Characteristic
`No.
`%
`No.
`%
`
`
`disease was not confirmed, and the other in the dexrazox-
`ane arm because of prior chemotherapy treatment for met—
`astatic disease. However, both patients were analyzed be-
`cause they received at least one cycle of chemotherapy.
`Overall, 160 patients, 78 in the control arm and 82 in the
`dexrazoxane arm, were evaluated.
`
`Treatment Modalities
`
`Twenty-five patients previously treated with adjuvant
`chemotherapy-containing anthracyclines were treated
`with CEF. The other 137 patients were enrolled into the
`HD-Epi group. Two patients, described earlier, were ex-
`cluded from analysis because they never received chemo-
`therapy. Of the remaining 135 patients, 68 received HD-
`Epi plus dexrazoxane and 67 HD-Epi alone.
`Overall, 476 cycles in the control arm and 489 cycles
`in the dexrazoxane arm were given. No significant differ—
`ence in the median number of chemotherapy cycles, or
`in the mean and median cumulative dose of anthracy-
`clines, was observed (Table 2). Both patients treated with
`CEF and those treated with HD-Epi received a median
`of six courses of chemotherapy. Patients treated with CEF
`received a median cumulative dose of epirubicin of 840
`mg/m2 (range, 360 to 1,340), and patients treated with
`HD-Epi 720 mg/m2 (range, 60 to 1,440).
`Treatment delay or dose reduction occurred in 79 cy-
`cles (16.6%) in the control arm and in 103 cycles (21.1%)
`in the dexrazoxane arm. The mean dose—intensity of epiru-
`bicin was 34.8 mg/mZ/wk and 33.3 mg/mzlwk in the con-
`trol and dexrazoxane arms, respectively. Patients in the
`control arm received 94% of the planned dose-intensity,
`and patients in the dexrazoxane arm 90%.
`
`Noncardiac Toxicity
`
`The incidence and severity of main non-cardiac toxici-
`ties are listed in Table 3. Seven patients in the dexrazox-
`ane arm died within 28 days after the last administration
`of chemotherapy. Five deaths were due to progressive
`disease, one to pulmonary embolism, and the last was a
`
`Table 2. Cycles and Doses of Chemotherapy Received
`
`by Treatment Arm
`Control
`Dexrazoxane
`
`Variable
`(n = 78)
`(n = 82)
`
`Anthracyclines dose
`Man
`Median
`Range
`No. of cycles received
`5.9
`6.1
`Mean
`6.0
`6.5
`Median
`
`Range 1 - 1 2 1 -10
`
`
`713
`720
`120-1200
`
`702
`720
`604,440
`
`84
`
`57
`32-73
`
`78
`
`14
`64
`
`57
`34-74
`
`17.9
`82.1
`
`15
`68
`1
`
`57
`20
`7
`
`52
`32
`
`7
`77
`
`11
`22
`44
`
`42
`14
`28
`
`54
`30
`
`17.9
`81 .0
`1.1
`
`67.9
`23.8
`8.3
`
`61.9
`38.1
`
`8.3
`91.6
`
`14.3
`28.6
`57.1
`
`50.0
`16.7
`33.3
`
`64.3
`35.7
`
`No. of patients
`Age, years
`Median
`Range
`Menopausal status
`Pre-
`Post-
`Unlrnown
`
`P5
`0
`1
`2
`Cardiac risk lactors'
`No
`Yes
`Extent of disease
`Locally advanced
`Metastatic
`Dominant metastatic sitef
`Sol? tissue
`Bone
`Viscero
`Adiuvont chemotherapy
`None
`Anthrocycllne-based
`CMF-based
`Previous radiotherapy
`No
`Yes
`Previous honnonotherapy
`54.8
`46
`50.0
`39
`Na
`44.0
`37
`50.0
`39
`Yes
`
`
`Unknown 1.2 — 1
`
`54
`24
`
`48
`30
`
`3
`75
`
`7
`21
`47
`
`45
`1 1
`22
`
`42
`36
`
`69.2
`30.8
`
`61.5
`38.5
`
`3.8
`96.2
`
`9.3
`28.0
`62.6
`
`57.7
`14.1
`28.2
`
`53.3
`46.2
`
`—
`
`—
`
`
`
`'Diabetes, hypertension, age 2 65 years, radiotherapy to the mediasti-
`num or left chest wall.
`TOnly in patients with metastatic disease.
`
`Downloaded from asconuhs org hv Renrints Desk on November 27 2017 from 216 185 156 028
`
`

`

`3116
`
`VENTURINI ET AL
`
`sudden death at home of unknown cause. Of note, four
`of seven deaths occurred in patients scored as PS 2. Clini-
`cal signs of phlebitis were observed in 3.8% and in 12.2%
`of patients in the control arm and dexrazoxane arms (P
`= .053), respectively. Grade 111 alopecia was virtually
`universal. Excluding leukopenia, no grade IV toxicity was
`recorded. Grade IV leukopenia was observed in three
`patients (3.8%) in the control arm and four (4.9%) in the
`dexrazoxane arm. Three patients (3.6%) in the dexrazox-
`ane arm suffered from mild peripheral neurotoxicity. An—
`other patient, in the dexrazoxane arm, had severe hand—
`foot syndrome. Apart from nausea and vomiting, the most
`common toxicity was stomatitis. Nearly 50% of patients
`suffered from this, but it was of severe degree (grade III)
`only in 5.2% of patients in the control arm and 8.5% of
`patients in the dexrazoxane arm.
`
`Cardiac Toxicity
`
`Two patients (2.6%) in the control group had asymp-
`tomatic supraventricular arrhythmia (grade I) and one pa-
`tient (1.2%) in the dexrazoxane group had grade II ar-
`rhythmia, without clinical or radiologic evidence of CHF.
`According to the protocol criteria, they were not consid—
`ered as cardiotoxic events.
`
`No patient treated with CEF chemotherapy had clinical
`and/or laboratory signs of cardiotoxicity. All cardiotoxic
`events occurred in the HD-Epi group. Cardiac toxicity
`was recorded in 18 of 78 patients (23.1%) in the control
`arm and in six of 82 (7.3%) in the dexrazoxane arm (Table
`4). The cumulative probability of developing clinical or
`laboratory signs of cardiotoxicity was significantly lower
`in dexrazoxane-treated patients than in control-arm pa-
`tients (P = .006 after adjustment for dose; odds ratio,
`0.29; 95% CL, 0.09 to 0.78) (Fig 2). At a 960-mg,/m2
`cumulative dose of anthracyclines, the cumulative proba-
`bility of developing cardiotoxicity was 30.9% in the con—
`trol arm and 9.5% in the dexrazoxane arm. At doses 2
`
`Table 4. Cardiotoxicily
`Control
`
`Variable
`
`No cardiotoxicity
`Cardiotoxicily
`
`CHF
`LVEF s 45%
`LVEF s 45% + decrease of LVEF
`a 20 EF units
`Decrease of LVEF 2 20 EF units
`
`No.
`
`78
`
`60
`18
`
`4
`7
`
`2
`5
`
`%
`
`100
`
`76.9
`23.1
`
`5.1
`9.0
`
`2.6
`6.4
`
`Dexrazoxane
`Na.
`%
`
`82
`
`76
`6
`
`2
`2
`
`2
`
`100
`
`92.7
`7.3
`
`2.4
`2.4
`
`2.4
`
`-
`
`1,200 mg/m2, the number of patients at risk was too low
`to provide reliable estimates. Similar results were seen
`when data were analyzed considering only MUGA-de-
`tectable signs of cardiotoxicity, as shown in Fig 3.
`Sixty-two of 160 patients had one or more cardiovascu-
`lar risk factors. Of these 62 patients, 14 (22.6%) devel-
`oped cardiac toxicity. Among the remaining 98 patients,
`without any cardiovascular risk factors, 10 (10.2%) expe—
`rienced cardiac toxicity. The cardioprotective effect of
`dexrazoxane seemed to be greater in patients with cardio-
`vascular risk factors, with a dose-adjusted odds ratio of
`0.15 (95% CL, 0.02 to 0.68), than in patients without
`cardiovascular risk factors who had an odds ratio of 0.53
`
`(95% CL, 0.10 to 2.38). However, the difference between
`the dexrazoxane effect in the two groups was not statisti-
`cally significant.
`Clinical cardiotoxicity was evaluated in all 160 patients
`and detected in six: four patients (5.1%) in the control
`arm (grade III NYHA in three patients, grade II in one),
`and two (2.4%) in the dexrazoxane arm (both grade II
`NYHA). Cumulative doses of epirubicin (mg/m2) in these
`patients were 480, 720, 720, and 840, in the control arm;
`and 220 and 720 in the dexrazoxane arm. One or more
`
`cardiovascular risk factors were present in five of six of
`these patients.
`
`Table 3. Noncardiac Toxicity
`Grade
`
`Control
`Dexrazoxane
`Toxicity
`0
`1
`II
`111
`N
`0
`|
`||
`111
`IV
`
`
`—
`4.9
`17.1
`50.0
`28.0
`—
`5.1
`23.1
`35.9
`35.9
`Nausea
`._..
`8.5
`35.4
`9.8
`46.3
`—
`7.7
`28.2
`7.7
`56.4
`Vomiting
`—
`8.5
`19.5
`20.8
`51.2
`—
`5.2
`33.3
`20.5
`41.0
`Stomatitis
`—-
`--
`7.3
`6.1
`86.6
`--
`-
`7.7
`7.7
`84.6
`Diarrhea
`-—
`—
`4.9
`7.3
`87.8
`—
`2.5
`-
`1.3
`96.2
`Phlebifis
`—
`4.9
`20.7
`12.2
`62.2
`—
`-
`20.5
`6.4
`73.1
`Fatigue
`--
`1.2
`3.6
`2.4
`92.8
`—
`--
`5.1
`5.1
`89.8
`Fever
`—'
`8.5
`7.3
`1.2
`83.0
`—
`2.6
`6.4
`3.8
`87.2
`Anemia
`—
`—
`2.4
`—
`97.6
`~
`—-
`2.6
`1.3
`96.1
`Thrombocytopenia
`
`Leukopenia 4.9 75.7 3.8 9.0 7.7 3.8 69.5 6.1 12.2 7.3
`
`
`
`
`
`
`
`
`
`
`NOTE. Values are percentages 01 patients.
`
`Downloaded from asconnhs org hv Renrints Desk on November 27 2017 from 216 185 I56 028
`
`

`

`DEXRAZOXANE AND EPIRUBICIN CARDIOTOXICITY
`
`31 17
`
`1W
`
`so
`
`60
`
`2
`=C
`
`3E
`3
`’6
`
`20 EF units).
`
`Fig 2. Cumulative probability
`of developing a cardiac event
`(clinical cur, LVEF s 45%, de-
`crease from baseline lVEF of z
`
`E 40
`3
`§
`5
`
`20
`
`o
`
`120
`
`240
`
`360
`
`430
`
`600
`
`720
`
`840
`
`960
`
`1080
`
`> 1200
`
`Cumulative doe. at anlnmcydtne
`
`Dexrazoxene 68
`Control
`67
`
`63
`65
`
`55
`B1
`
`58.5
`63
`
`56.5
`55
`
`53.5
`50.5
`
`37
`38
`
`34
`30
`
`B
`11
`
`3
`3
`
`2
`
`Four patients in the control arm and 14 patients in the
`dexrazoxane arm did not have further LVEF evaluations
`at baseline. Four patients, two in each arm, had progres-
`sive disease before planned assessment. Two patients in
`the control arm and one patient in the dexrazoxane arm
`were lost to follow-up evaluation. One patient in the dex-
`razoxane arm underwent surgery due to a second primary
`tumor. Another patient in the dexrazoxane arm required
`palliative radiotherapy to the backbone, after the first cy-
`ole of chemotherapy, and 10 days later developed a basal
`pleuropneumonia. The remaining nine patients, all in the
`dexrazoxane arm, did not undergo cardiac laboratory
`evaluation due to treatment-related toxicity (one patient),
`
`refusal
`(seven).
`
`to continue chemotherapy (one), and death
`
`Treatment Activzty
`There were no statistically significant differences be-
`tween the two treatments arms in survival and in progres-
`sion-free survival (Figs 4 and 5). No difference in objec—
`tive response rate (CRs plus PRs) was observed, at 46.2%
`(95% CL, 34.9% to 57.8%) and 47.6% (95% CL, 36.5%
`to 58.8%) in the control and dexrazoxane arms, respec~
`tively. CRs occurred in eight of 78 patients (10.3%) in
`the control arm and in nine of 82 (11.0%) in the dexrazox-
`ane arm.
`
`event
`Prohabllllyofcardiac
`
`0
`
`120
`
`240
`
`360
`
`720
`800
`480
`Culnulettvo dae- ot mthreeyellno
`
`840
`
`960
`
`“$0
`
`> 1200
`
`No. at patients at risk:
`Denmne 87
`Corml
`s7
`
`62
`65
`
`55
`58
`
`57.5
`81
`
`52.5
`50
`
`49.5
`435
`
`35
`35
`
`32
`29
`
`10
`5
`
`5
`3
`
`2
`
`Fig 3. Cumulative probability
`of developing laboratory cardio-
`toxicity measured by MUGA scan
`(lVEF s 45%, decrease from
`baseline LVEF at z 20 if units).
`
`Downloaded from asconuhs org hv Renrims Desk on November 27 2017 from Zlfi IRS H6 022
`
`

`

`3118
`
`VENTURlNl ET AL
`
`%
`Pts Obs
`Arm
`
`Demoxano
`32
`37
`
`
`Control
`
`78 35
`
`100
`
`75
`
`Exp
`35.69
`
`Obs/Exp
`1.03
`
`36.30
`
`0.96
`
`p = 0.75
`
`
`
`50
`
`25
`
`0
`
`0
`
`1 80
`
`360
`
`540
`
`720
`
`900
`
`1080
`
`Time (days)
`
`Fig 4. Overall
`treahnant arm.
`
`survival by
`
`DISCUSSION
`
`The extended use of anthracyclines is, in pan, limited
`by their cardiotoxicity. Although epirubicin has a lower
`cardiotoxic potential
`than doxorubicin, at cumulative
`doses greater than 1,000 mg/m2, it induces CHF in 16%
`to 35% of patients,12 with an incidence of related death
`of 20%.‘3 The most commonly used method to prevent
`anthracycline cardiotoxicity is to stop the administration
`of these drugs when a predeterminated empiric cumula-
`tive dose has been reached: 450 to 550 mg/m2 for doxoru-
`bicin and 900 to 1,000 mg/m2 for epirubicin. Since a
`broad variation in the cumulative dose that causes CHF
`
`has been reported,14 use of this strategy means to discon-
`tinue arbitrarily an active drug in some patients who might
`tolerate higher doses and might further benefit from ther-
`apy. Furthermore, this approach does not eliminate the
`
`risk of cardiotoxicity. In fact, anthracycline-induced car-
`diotoxicity can occur even at lower cumulative doses and
`particularly when some risk factors such as advanced age,
`hypertension, diabetes, and chest-wall radiotherapy are
`present.“"’ Early breast cancer patients, treated with less
`than 300 mg/m2 of doxorubicin, who received 1eft~breast
`irradiation, developed cardiac abnormalities more fre—
`quently (23.8%) compared with patients who did not re-
`ceive irradiation (5.4%).16 Another increasingly noted
`problem is late anthracycline cardiotoxicity. Late devel-
`opment of CHF has been reported in 10% of patients and
`depressed contractility in 29% of patients who survive
`childhood cancer,17 although the ceiling doxorubicin dose
`of 550 mg/m2 was not exceeded. Therefore, limiting the
`cumulative dose of anthracyclines may not be always
`sufficient to prevent cardiotoxicity.
`
`°/o
`
`100
`
`Arm
`
`Dexrazoxane
`
`Control
`
`Pts Obs
`
`82
`
`78
`
`61
`
`59
`
`EXP
`61.46
`
`58.53
`
`Obs/Exp
`0.99
`
`1.00
`
`Fig 5. Progression-free wr-
`vivel by Ireahnent arm.
`
`75
`
`50
`
`25
`
`0
`
`
`
`180
`
`360
`
`540
`
`720
`
`900
`
`Time (days)
`
`Downloaded from ascnnnlis org hv Renrints Desk on November 27 20] 7 from 2| 6 IRS l€6 028
`
`

`

`DEXRAZOXANE AND EPIRUBICIN CARDIOTOXICITY
`
`3119
`
`The use of dexrazoxane has been shown to protect
`against doxorubicin-induced cardiotoxicity.8 Although
`the cardioprotective effect of dexrazoxane has also been
`shown against epirubicin-induced cardiotoxicity in pre-
`clinical studies,6 no phase III trial was yet fully reported.
`Our study has demonstrated that dexrazoxane, given at
`a dexrazoxanezepirubicin dose ratio of 10:1, is able to
`reduce cardiac toxicity of epirubicin, as determined by
`MUGA scan and/or clinical signs. We used lower doses
`of dexrazoxane than those previously reported as effec-
`tive for protection from cardiac damage induced by dox-
`orubicin. Speyer et al,8 who used a ratio of 20:1, showed
`a beneficial effect of dexrazoxane against doxorubicin
`cardiotoxicity, but a mild, statistically significant, in-
`crease in myelotoxicity was observed in dexrazoxane—
`treated patients. This was probably related to the mild
`hematologic toxicity of dexrazoxane. With the ratio of
`10:1 used in the study, we did not observe any increase
`in myelotoxicity (Table 3): grade III/IV leukopenia was
`registered in 11.5% and in 12.2% of patients in the
`control and dexrazoxane arms, respectively. No grade
`III/IV thrombocytopenia was observed in either arm. A
`higher incidence of phlebitis was recorded in dexrazox-
`ane-treated patients (12.5%) compared with controls
`(3.5%). However, this toxicity was always graded as
`mild or moderate. This finding confirms previous sug-
`gestions of a certain degree of vascular toxicity of dex-
`razoxane.18 Seven patients treated with dexrazoxane
`died before any other laboratory cardiac evaluation.
`Most of these patients have poor general condition (PS
`2) and died due to progressive disease. This can be
`also explained by the unbalance in PS between the two
`treatment arms at randomization. Seven patients (8.3%)
`in the dexrazoxane arm had PS 2 versus no patient in
`the control group.
`The effect of dexrazoxane on the pharmacokinetics
`of HD-Epi has been recently reported.19 At dexrazoxane
`doses greater than 900 mg/m2 or at a ratio of 7:1 or
`greater, an increase in the systemic clearance of epiru-
`bicin with a resultant decrease in the area under the
`
`concentration-time curve (AUC), was observed. It may
`be speculated that this could result in a decrease of the
`antitumor activity of this drug. However, analyzing our
`data of 137 patients treated with 120 mg/m2 of epiru-
`bicin, we observed the same response rate. Among 70
`patients in the dexrazoxane arm, the CR rate was 11.4%
`and the PR rate 37.1%. Among 67 patients in the control
`arm, CRs occurred in 10.4% and PRs in 38.8%. There-
`
`fore, the addition of dexrazoxane to epirubicin, at a
`ratio 10:], does not appear to modify the antitumor
`activity of this drug.
`Increasing the single doses of epirubicin has been re—
`
`ported to correlate with a linear increase of maximum
`plasma concentration of epirubicin.20 Moreover, higher
`single doses of anthracyclines have been shown to be
`important predictors of late cardiotoxic effects in child-
`hood cancer pati