`
`research paper b
`
`Comparative effects of deferiprone and deferoxamine on survival and cardiac
`disease in patients with thalassemia major: a retrospective analysis
`
`ANTONIO PIGA, CARMEN GAGLIOTI, EUGENIA FoGUACCO, FERNANDO TRJCTA
`
`Background and Objectives. Iron-induced cardiac dis(cid:173)
`ease remains the main cause of death in patients with
`thalassemia major, despite chelation therapy with defer(cid:173)
`oxamine. Deferiprone is an iron chelator that has the
`potential to be more effective than deferoxamine in remov(cid:173)
`ing intracellular iron from the heart. However, to date, no
`study has been designed to examine the frequency of car(cid:173)
`diac complications and survival as the primary outcomes
`of a comparative study between these two chelators. This
`retrospective study assessed the survival and the occur(cid:173)
`rence of cardiac disease in all patients with thalassemia
`major treated for at least 4 years with deferiprone or defer(cid:173)
`oxamine at a single center.
`Design and Methods. The patients were, on average,
`18.4 years old at the start of the review period and were
`followed up, on average, for 6 years. At baseline there was
`no significant difference in the percentage of patients
`with cardiac disease in the two therapy groups.
`Results. At the end of the study, cardiac dysfunction,
`expressed as worsening of pre-existing cardiac abnor(cid:173)
`mality or development of new cardiac disease, was diag(cid:173)
`nosed in 2 (4%) of the 54 deferiprone-treated patients
`and in 15 (20%) of the 75 deferoxamine-treated patients,
`from the first to the last measurement (p = 0.007). The
`Kaplan Meier analysis of cardiac disease-free survival
`over the 5-year period was significantly more favorable in
`the deferiprone group (p = 0.003).
`Interpretation and Conclusions. None of the patients
`treated with deferiprone died, while 3 of the patients treat(cid:173)
`ed with deferoxamine died because of irreversible wors(cid:173)
`ening of their cardiac condition during the study period.
`Findings from this study suggest that long-term therapy
`with deferiprone provides a greater cardio-protective effect
`against the toxicity of iron overload than does subcuta(cid:173)
`neous deferoxamine. Formal prospective studies are war(cid:173)
`ranted to confirm this effect.
`
`Key words: thalassemia, chelation, iron, deferiprone,
`deferoxamine.
`
`Haematologica 2003; 88:489-496
`http://www.haematologica.org/2003_05/ 489.htm
`
`©2003, Ferrata Storti Foundation
`
`From the Department of Pediatric Hematology, University of Turin, Italy
`(AP, CG, EF), Apotex Research Inc, Canada (FT).
`
`Correspondence: Antonio Piga, MD, Department of Pediatric Hematology,
`Universily of Turin, Italy.E-mail:antonio.piga@unito.it
`
`The iron chelator deferoxamine has been available
`
`for the treatment of iron overload for more than 3
`decades. Deferoxamine use can prolong survival
`and has been shown to ameliorate hepatic, endocrine
`and cardiac dysfunction in iron-overloaded patients
`with thalassemia major. 1-6 However, cardiac disease
`remains a frequent cause of morbidity among tha(cid:173)
`lassemia patients'·' and is still responsible for 700/o of
`deaths.' Compliance with regular chelation therapy, as
`determined by the ability to perform more than 265
`deferoxamine infusions per year' or maintenance of
`most serum ferritin values below 2500 µg/L, has been
`identified as an important factor in limiting the inci(cid:173)
`dence of cardiac disease and prolonging the survival
`among patients with thalassemia major.'·' The age at
`the start of chelation therapy and the hepatic iron con(cid:173)
`centration have also been suggested as factors which
`could affect the development of cardiac disease."'
`Although optimization of subcutaneous deferoxamine
`infusions (needle, pump, adjustment of daily life) has
`been achieved in many centers, the cumbersome nature
`of slow daily subcutaneous infusions of deferoxamine,
`and the resulting discomfort and adverse effects, have
`prevented optimal compliance in approximately 500/o
`of patients.1·10
`Deferiprone (L 1, 1,2-dimethyl-3-hydroxypyrid-4-one)
`is an orally active chelator, which has been recently
`approved in Europe for the treatment of iron overload.
`As a small, lipophilic bidentate molecule, deferiprone
`would theoretically have the potential to be more effec(cid:173)
`tive than deferoxamine in removing intracellular iron
`from some tissues, as has been demonstrated in the
`case of red cell membranes. 11 •12 Link eta/. reported that
`deferiprone, at concentrations achieved in humans with
`standard doses, was more effective than deferoxamine
`in removal of iron from iron-loaded neonatal rat myo(cid:173)
`cytes.13
`These data from in vitra and animal studies are con(cid:173)
`sistent with those from clinical studies that report a
`low rate of iron-induced clinical complications during
`deferiprone therapy. 11 - 19 However, to date, no study has
`been designed to examine the frequency of cardiac
`complications and survival as the primary outcomes of
`a comparative study of deferoxamine and deferiprone.
`These observations constituted the rationale for
`designing a study to compare mortality and cardiac
`morbidity in patients with thalassemia treated long(cid:173)
`term with either deferoxamine or deferiprone.
`
`haematologica/journal of hematology vol. 88(05):may 2003
`
`489
`
`TAR0008746
`
`1 of 8
`
`Taro Pharmaceuticals, Ltd.
`Exhibit 1061
`
`
`
`A. Piga et al.
`
`Design and Methods
`
`Study design
`A retrospective analysis was conducted on sur(cid:173)
`vival and occurrence of cardiac disease in all
`patients with thalassemia major treated for at least
`4 years with deferiprone or deferoxamine at a sin(cid:173)
`gle center between 31 January 1995 and 29 March
`2001. The medical records of patients aged 2' 5
`years at the time of the start of the review period,
`with a diagnosis of thalassemia major confirmed by
`laboratory tests (electrophoresis and/or DNA analy(cid:173)
`sis) and appropriate clinical enrollment criteria
`(hemoglobin and transfusion dependency) were
`evaluated. Patients with congenital anemias other
`than thalassemia major, who were HIV-positive,
`who had a history of malignancy or who required
`radiation or chemotherapy were not included in
`the analysis.
`Patients were excluded from the data analysis
`for the following reasons: chelation therapy with
`one of the chelators for less than 4 years during the
`review period; no information available on chela(cid:173)
`tion therapy or cardiac status; only 1 cardiac
`assessment; age <5 years at the start of the review
`period (6 patients); HIV-positive status.
`The Regional Ethical Review Board approved the
`study. All eligible patients or their guardians gave
`authorization for the review of their medical
`records.
`
`Patients
`All patients received the same transfusion regi(cid:173)
`men, aimed at maintaining pre-transfusion hemo(cid:173)
`globin levels at 9.5 -10.0 g/dl and the mean hemo(cid:173)
`globin at 12.0 g/dl. A software program, previous(cid:173)
`ly designed for thalassemia-oriented clinical
`records, calculated the annual transfusional iron
`input on the basis of the net weight and hematocrit
`of the blood transfused.20
`All patients were treated with deferoxamine
`until January 1995, when some were switched to
`deferiprone because they were enrolled in clinical
`trials, or their clinical condition required the drug.
`For those patients who were treated with
`deferiprone as part of clinical studies, the major
`criteria of their inclusion into the studies were age
`(10 years of age or older), the severity of the body
`iron load (serum ferritin level >2000 ng/µL or liv(cid:173)
`er iron concentration >4 mg/g dry weight) and
`inability to use deferoxamine or unwillingness to
`continue its use despite medical advice. Exclusion
`criteria for participating in clinical trials included
`cardiac disease requiring medication, hepatic
`insufficiency or renal failure.
`
`Chelation therapy
`Deferoxamine (Desferal®, Novartis Inc.) was pre(cid:173)
`scribed at the dose of 20 to 50 mg/kg/day, as an 8-
`
`12-hour subcutaneous infusion, 4-7 days a week.
`Three patients in the deferoxamine group had their
`chelation intensified with intravenous chelation
`during the study period. Despite the intensive
`chelation with deferoxamine they were not exclud(cid:173)
`ed from the analysis. Compliance with subcuta(cid:173)
`neous deferoxamine therapy was assessed at each
`transfusional event by the following:
`• comparison of number of infusions reported by
`the patient with the number prescribed;
`•examination of infusion sites;
`• number of completed infusions recorded by a
`special electronic pump (Crono®, Cane S.r.I,
`Italy);
`• pharmacy records of syringes and needles dis(cid:173)
`pensed.
`Deferiprone (Ferriprox®, Apotex Inc.) was pre(cid:173)
`scribed at the dose of 25-100 mg/kg/day, divided
`in three doses. Compliance was assessed at each
`transfusional event by the electronic MEMS® cap
`(Medication Event Monitoring System, Ardex Ltd,
`Switzerland), which records the time and date of
`each opening of the deferiprone container. Com(cid:173)
`pliance for patients participating in clinical trials
`was also measured by monthly counts of the num(cid:173)
`ber of deferiprone tablets dispensed and returned.
`Regular determinations of serum ferritin level
`were used to monitor basic iron overload in all
`patients. The treating hematologist used all the
`above information during regular interviews with
`each patient as part of the clinical management of
`thalassemia and to promote compliance.
`
`Cardiac assessments
`A single cardiologist, experienced in hemoglo(cid:173)
`binopathies but blind to the therapy that the
`patients were receiving, assessed the cardiac sta(cid:173)
`tus of all patients. The standard assessment includ(cid:173)
`ed physical examination, ECG, echocardiogram, and
`cardiac status based on the criteria of the New York
`Heart Association (NYHA). 21 When indicated, a 24-
`hour electrocardiographic Holter and/or stress test
`were also performed. The treating physician sys(cid:173)
`tematically screened the clinical records of each
`patient followed at the center. All patients who
`had had at least two cardiac assessments were
`evaluated, even if their first assessment was com(cid:173)
`pleted after the initiation of the study period. The
`first assessment was considered as the baseline
`value for each patient. The records of patients with
`any abnormal cardiac finding were then reviewed
`by the cardiologist. A change in shortening fraction
`(SF) or ejection fraction (EF) values was considered
`as important if the result at the last echocardio(cid:173)
`gram changed from normal to abnormal or vice
`versa compared to the first echocardiogram. The
`thresholds for normality of systolic function were
`defined as 300/o for SF and 550/o for EF.
`Worsening and improvement of cardiac function
`
`490
`
`haematologica/joumal of hematology vol. 88{05):may 2003
`
`TAR0008747
`
`2 of 8
`
`Taro Pharmaceuticals, Ltd.
`Exhibit 1061
`
`
`
`Comparative effects of deferiprone and deferoxamine on survival
`
`was defined, a priori, by one or more of the fol(cid:173)
`lowing criteria:
`Worsening
`•shifting of congestive heart failure from acute
`to chronic;
`•increase in NYHA class;
`• aggravation of arrhythmia requiring medication
`or a significant change in existing medication, or
`indication for ablation, or need for a pacemaker;
`• shift from normal to abnormal echocardio-
`graphic values of systolic function.
`
`Improvement
`• shifting from congestive heart failure to
`asymptomatic;
`• decrease in NYHA class;
`• reduction of arrhythmia with reversal of med(cid:173)
`ication dependency;
`• shift from abnormal to normal echocardio(cid:173)
`graphic values of systolic function or reversal of
`medication dependency.
`In addition to the cardiac assessments, the fol(cid:173)
`lowing clinical and laboratory parameters were
`included in the evaluation of the two treatment
`groups:
`•gender;
`•age at start of the review period;
`• age at start of first chelation therapy with
`deferoxamine;
`•serum ferritin;
`• transfusional iron input;
`•compliance with chelation therapy.
`
`Statistical analysis
`A 2-sample t-test or X' test, as appropriate, was
`performed to compare differences in the study para(cid:173)
`meters in the two groups of patients. The Kaplan(cid:173)
`Meier analysis of heart disease-free survival for
`patients who were disease-free (NYHA class= 0) at
`the beginning of the review period was performed
`using SAS (version 6.12) and the disease-free sur(cid:173)
`vival curves were plotted using Statistica 5.5 (Stat(cid:173)
`Soft Inc. Tulsa, USA). The incidence of patients with
`cardiac disease diagnosed at the first cardiac assess(cid:173)
`ment who showed an improvement of their NYHA
`class during the study was determined and com(cid:173)
`pared between the 2 treatment groups using Fish(cid:173)
`er's exact test. To examine any potential effect of
`age on the prevalence and/or progression of iron(cid:173)
`induced cardiac disease, a further analysis was con(cid:173)
`ducted, with patients from the 2 therapy arms being
`matched for age at the start of chelation therapy.
`All statistical tests were 2-sided with a type 1
`error (ex) of 0.05. SAS (version 6.12) was used to
`conduct all the statistical tests. In all 2-sample t(cid:173)
`tests, when the test for equality of variances was
`significant (p<0.05). the test result based on
`unequal variances was used to determine the sta(cid:173)
`tistical significance of the comparison. Furthermore,
`
`Table 1. Comparison of deferiprone- and deferoxamine-treat(cid:173)
`ed patients at the start of the study. Mean values ± stan(cid:173)
`dard deviation are shown. The number of patients for whom
`the value was calculated is given in brackets if this differed
`from the original number of subjects.
`
`Deferiprone
`(n·54)
`
`Deferoxamine
`(n•75)
`
`p
`
`Percentage female
`
`Mean age {)'ears)
`
`Mean age (years) at start
`of chelation therapy
`with deferoxamine
`
`Mean serum ferritin+
`(µg/L)
`
`Pe1Centage of patients with
`> 50% of serum fenitin
`results> 2,500• {µg/L)
`
`44%
`
`17.1±4.1
`
`4.5t2,7
`(54)
`
`49%
`
`19.4±6.9
`
`6.8±4.7
`(72)
`
`2033±919
`{51)
`
`1809±1464
`(63)
`
`24%
`(51)
`
`15%
`(6-0)
`
`Mean transfusional iron
`input• (mg Fe/kg bodyweight/day)
`
`0.464±0.085
`(49)
`
`0.432±0.110
`(61)
`
`Number {percentage) of
`patients with abnonnal
`cardiac function
`
`7
`(13%)
`
`12
`(16%)
`
`•during !he 2 years preceding the sllldy period.
`
`0.583
`
`0.018
`
`0.002
`
`0.328
`
`0.253
`
`0.102
`
`0.631
`
`when at least 500/o of the cells in the 2x2 contin(cid:173)
`gency table had expected counts of< 5 in the X'
`test, Fisher's exact test was used to determine the
`statistical significance.
`
`Results
`
`The clinical records of 168 patients with tha(cid:173)
`lassemia major were screened, and 129 met all of the
`inclusion and exclusion criteria. Sixteen patients had
`not had chelation therapy with one of the chelators
`for at least 4 years during the review period, for 10
`patients no information was available on chelation
`therapy or cardiac assessment, 6 patients had only
`1 cardiac assessment, 6 patients were <5 years of
`age at the start of the review period and one patient
`was HIV-positive. None of the excluded patients was
`diagnosed with a change in their cardiac status or
`died during the review period.
`All patients were regularly chelated with deferox(cid:173)
`amine prior to the initiation of the study period. At
`the time of the study initiation both treatment
`groups were similar for gender distribution and the
`transfusional iron input received during the preced(cid:173)
`ing 2 years. At baseline patients whose therapy was
`switched to deferiprone were younger (p=0.018) and
`had initiated chelation therapy with deferoxamine at
`an earlier age than patients who were maintained on
`deferoxamine (p= 0.001) (Table 1).
`During the study period the average dose of
`
`haematologica/journaJ of hematology vol. 88(05):may 2003
`
`491
`
`TAR0008748
`
`3 of 8
`
`Taro Pharmaceuticals, Ltd.
`Exhibit 1061
`
`
`
`A. Piga et a!.
`
`Table 2. Comparison of deferiprone- and deferoxamine-treat-
`ed patients during the study period. Mean values± standard
`deviation are shown and the number of patients for whom
`the value was calculated is given in brackets if this differed
`from the original number of subjects.
`
`m<
`-< 2(4Yo)
`Oeforoumlne -< 15(20!'.j
`
`Wcrsenlngcl
`C•rdl•e Olu•••
`
`··-....
`
`p•0.007
`
`/
`
`"'
`""
`,.,
`
`52(96%)
`
`"'
`
`60{6tl%)
`
`F•~onto
`
`Oo'°rlFrone
`
`"""
`
`n=75
`
`35%
`
`20%
`
`0.053
`
`Figure 1. Cardiac function in patients with thalassemla
`major treated with deferiprone or deferoxamine.
`
`Deferipmne
`(n·54)
`
`Deferoxamine
`(n•75)
`
`89%±7%
`(53)
`
`85%±11%
`(73)
`
`0.432±0,076
`(53)
`
`0.408±0.085
`(64)
`
`2142±957
`(49)
`
`2143±1481
`(64)
`
`'
`
`0.011
`
`0.111
`
`0.994
`
`3/7 (43%)
`
`3/12(25%)
`
`0.617
`
`2/47 (4%)
`
`13/63(21%)
`
`0.013
`
`2/54 (4%)
`
`15/75 (20%)
`
`0.007
`
`Percentage of compliance
`with chelatlon therapy
`
`Mean overall transfusiona!
`iron input (mg Fe)/k~day
`
`Mean serum ferTitin {µg/L)
`at year 5 of the review
`period
`
`% patients > 503 of serum
`ferri!in results> 2500 µgll
`
`Ratio of patients with
`improvement of cardiac
`disease
`
`Ratio of patients with
`onset of cardiac disease
`
`Overall ratio of patients
`v.ith worsening cardiac
`status
`
`deferiprone was 73.7±11.2 mg/kg/day (range 25-
`100 mg/kg/day) and the average dose of deferox(cid:173)
`amine was 39.2±4.7 mg/kg per infusion, adminis(cid:173)
`tered an average of 6±1 times per week. The
`weighted mean compliance with deferiprone was
`890/o (range 66-980/o) whereas it was 850/o (range
`54-990/o) with deferoxamine (Table 2).
`The mean(± SD) transfusional iron load in those
`patients switched to deferiprone, prior to the
`switch (0.464±0.085 mg Fe/kg body weight/day)
`was higher than the baseline values of those who
`remained on deferoxamine (0.432±0.110 mg Fe/kg
`body weight/day), but the difference failed to reach
`statistical significance (p=0.102). The deferiprone(cid:173)
`treated patients continued to receive more iron
`throughout the study period (an average of
`0.432±0.076 mg Fe/kg body weight/day) than the
`deferoxamine-treated patients [0.408±0.085 mg
`Fe/kg body weight/day), but this trend did not
`reach statistical significance (p=0.111).
`The percentage of patients with more than 500/o
`of their serum ferritin values greater than 2500
`µg/L during the study rose from 240/o at baseline to
`350/o by the last year of the study in the deferi(cid:173)
`prone-treated group, compared with a rise from
`150/o at baseline to 200/o in the deferoxamine(cid:173)
`treated group. The difference in percentage at the
`end of the study between the two groups was not
`statistically significant (p = 0.053). No significant
`difference was observed in the mean serum fer-
`
`ritin levels between the 2 treatment groups by the
`end of the study (p = 0.994).
`
`Cardiac disease
`Seven patients in the deferiprone group and 12
`patients in the deferoxamine group had abnormal
`cardiac function at the first cardiac assessment.
`The cardiac abnormality was evaluated as NYHA
`class I in 13 patients (six in the deferiprone group
`and seven in the deferoxamine group), class II in 3
`patients (all in the deferoxamine group), class Ill in
`two patients (one in each therapy group), and class
`IV in one deferoxamine-treated patient. The over(cid:173)
`all prevalence of cardiac disease at the first assess(cid:173)
`ment was similar for both groups (p=0.606). Wors(cid:173)
`ening of cardiac function was diagnosed in none of
`the patients treated with deferiprone and in 4
`(330/o) of the patients treated with deferoxamine.
`An improvement of the NYHA cardiac disease class
`was observed in 3 (430/o) of the 7 deferiprone(cid:173)
`treated patients and in 3 (250/o) of the 12 deferox(cid:173)
`amine-treated patients with cardiac disease diag(cid:173)
`nosed at the first assessment (p= 0.617).
`Newly diagnosed cardiac disease occurred in 2
`(40/o) of the 47 deferiprone-treated patients and in
`13 (210/o) of the 63 deferoxamine-treated patients
`who were free of cardiac disease at their first car(cid:173)
`diac assessment. All but 2 of the 15 patients were
`classified as having NYHA class I disease. One
`remaining patient was classified as having NYHA
`class II, while the other one was classified as hav(cid:173)
`ing NYHA class I and then worsened to class Ill;
`both patients were treated with deferoxamine.
`Overall, cardiac dysfunction, expressed as wors(cid:173)
`ening of pre-existing cardiac abnormality or devel(cid:173)
`opment of new cardiac disease, was diagnosed in
`40/o of deferiprone-treated patients and in 200/o of
`deferoxamine-treated patients from the first to the
`last measurement (p=0.007; Figure 1).
`
`492
`
`haematologica/joumar of hematology vol. 88(05):may 2003
`
`TAR0008749
`
`4 of 8
`
`Taro Pharmaceuticals, Ltd.
`Exhibit 1061
`
`
`
`Comparative effects of deferiprone and deferoxamfne on sutvival
`
`1.0
`
`!''''"''"''""'"""""'""''"""''""""';
`
`deferiprone-treated patients and in 9 (190/o) defer(cid:173)
`oxamine-treated patients from the first ta the last
`measurement (p= 0.023). Kaplan-Meier analysis
`indicated a significant difference (p= 0.017) in the
`cardiac disease-free survival between the 2 groups
`in favor of those treated with deferiprone.
`
`Survival
`None of the 54 patients treated with deferiprone
`died, while 4 of the 750 patients treated with
`deferoxamine died during the study period. Three
`af these patients had cardiac disease at the first
`assessment of the study period and died because of
`irreversible worsening of their cardiac condition
`(Table 3). One death occurred during the second
`year, whereas the other two.occurred during the
`last year of the review period. The fourth death
`occurred in a patient with a history of drug addic(cid:173)
`tion 'but no signs of cardiac disease. This patient
`died within a few hours of being admitted into a
`provincial hospital for acute abdominal pain. No
`cause of death was determined. This death was not
`included among the deaths in the survival analy(cid:173)
`sis.
`
`Discussion
`
`This is the first published study to compare the
`development of cardiac disease and survival
`between subjects with thalassemia major treated
`with deferoxamine or deferiprone. Patients partic(cid:173)
`ipating in the study were young, with a mean age
`of 18.4 years at time of the start of the review peri(cid:173)
`od, and were followed up an average for 6 years, a
`period which corresponds to approximately 250/o
`of their life span at the end of the study. The data
`revealed that patients who had had their chelation
`therapy switched to deferiprone were less likely to
`have developed cardiac disease and if they had had
`pre-existing cardiac disease, were less likely to
`have experienced a worsening of that disease.
`
`t
`
`0.5
`
`0.9
`~ o.s
`~ 0.7
`~ 0.6
`~ § 0.4
`i 0.2
`0.1
`o.o 0.0 0.5
`
`~ 0.3
`
`l
`
`Oeferoxamine patients
`Oeferlprone patients
`
`Complete +Censored
`
`1.0 1.5 2.0 2.5
`
`3.0 3.5 4.0
`
`4.5 5.0 5.5'
`
`Time (years)
`
`Figure 2. Kaplan·Meier analysis of heart disease.free sur·
`viva! in patients with thalassemia major treated with
`deferiprone or deferoxamine. Four percent of
`the
`deferiprone..treated patients developed cardiac disease com·
`pared to 21% of deferoxamine·treated patients over the
`same period of time (p = 0.003).
`
`The Kaplan Meier analysis of cardiac disease-free
`survival over the 5-year period was significantly
`more favorable in the deferiprone group (p= 0.003)
`(Figure 2).
`To determine whether the observed difference in
`the occurrence of cardiac disease between the two
`treatment groups could have been influenced by
`differences in age or age at the start of chelation,
`we conducted a cohort analysis. Forty-seven
`patients treated with deferiprone were found to
`have a match from the deferoxamine-arm for the
`same age at start of chelation therapy. Both
`cohorts were similar for baseline data, including
`age at the start of the study (data not shown). As
`in the main analysis, after 5 years of therapy there
`was no significant difference between the 2 treat(cid:173)
`ment groups in the mean serum ferritin level. Over(cid:173)
`all. worsening of cardiac disease/development of
`new cardiac disease was diagnosed in 2 (40/o)
`
`Table 3. Demographics and clinical information on the three patients who died of cardiac disease.
`
`Age at start
`revielV period
`
`Gender
`
`Age at s!arl
`of first chelation
`
`Chelation therapy
`during sludy
`
`Cardiac disease
`al baseline
`
`26years
`
`23 years
`
`Male
`
`Male
`
`13 years
`
`8years
`
`23years
`
`Female
`
`NA
`
`DFO
`
`DFD
`
`OFO
`
`Yes
`(NYHA Class II)
`
`Yes
`(NYHA Class II)
`
`Yes
`(NYHA Class IV)
`
`Compliance
`with chelation
`during review
`
`54%
`
`73%
`
`not
`available
`
`Hepatic iron
`concentration
`closest to time
`of death
`
`Not
`available
`
`3.3 mg/g
`diyweight
`
`31.2 mg/g
`dry weight
`
`% of
`serum ferri!in
`> 2500 µg/L
`
`89%
`
`25%
`
`Not
`available
`
`haematologica/journa! of hematology vol. 88(05):may 2003
`
`493
`
`TAR0008750
`
`5 of 8
`
`Taro Pharmaceuticals, Ltd.
`Exhibit 1061
`
`
`
`A. Piga et al.
`
`Iron-induced cardiac disease remains the main
`cause of death in patients with thalassemia major.8
`In a previous study, we reported that 300/o of
`patients with thalassemia developed cardiac dis(cid:173)
`ease by the age of 20 years.' In the current study,
`19 (150/o) patients had cardiac disease at the first
`assessment of the review period and there was no
`significant difference in the percentage of patients
`with cardiac disease in the two study groups. At
`baseline, all patients were being treated with sub(cid:173)
`cutaneous deferoxamine infusions, an average of
`6.2 days per week and, based on the serum ferritin
`levels over the 2 years that preceded the baseline
`assessment, the overall patient cohort was consid(cid:173)
`ered to be relatively well chelated and at low risk
`of iron-induced cardiac disease.' However, over the
`6-year observation period, there was a significant(cid:173)
`ly greater prevalence of worsening of heart disease
`among patients maintained with deferoxamine
`than in patients switched to deferiprone. Newly
`diagnosed heart disease occurred in 15 (120/o)
`patients, which corresponds to an incidence of2.0
`per 100 patient-years. The incidence was greater in
`the deferoxamine group. Overall, the probability of
`worsening cardiac status was about 5-fold higher
`during deferoxamine therapy than during deferi(cid:173)
`prone therapy. At the completion of the study, three
`patients in the deferoxamine-treatment group had
`died of cardiac disease.
`The two groups of patients were similar in most
`aspects but differed regarding the age at baseline
`and the age at which chelation therapy had been
`started. To assess the impact of age on the study
`results, the analysis was repeated after matching
`the patients in each arm of therapy for age at start
`of chelation. Both cohorts were similar for baseline
`data, including age at the start of the study. The
`analysis, after this correction, still demonstrated a
`significantly more favorable outcome in the
`deferiprone-treated cohort in terms of cardiac dis(cid:173)
`ease-free survival and for worsening of pre-exist(cid:173)
`ing cardiac disease and indeed the difference was
`even more pronounced.
`According to the literature,' a major limiting
`factor in patients treated with deferoxamine is a
`lack of compliance. However, in this study, the clin(cid:173)
`ically favorable results in the deferiprone-treated
`group do not appear to be related to a lack of com(cid:173)
`pliance
`in
`the deferoxamine-treated group.
`Although compliance with oral chelation was high(cid:173)
`er than that with the subcutaneous infusions, the
`overall compliance with deferoxamine was only 40/o
`less than with deferiprone, and far above the lev(cid:173)
`el normally reported.10 The observed high compli(cid:173)
`ance in the deferoxamine group is probably due to
`the regular and intensive attention given to this
`aspect of management of thalassemia over the
`
`years in this center.22
`Only one patient had an extensive interruption of
`chelation therapy; a deferiprone-treated patient
`stopped chelation for approximately one year while
`being treated for hepatitis C. He was not excluded
`from the analysis. The three patients who died of
`cardiac disease were among those with the lowest
`compliance. There was no difference in compliance
`between the 2 therapies among patients with a
`worsening of cardiac disease or new cardiac dis(cid:173)
`ease (850/o vs 840/o for deferiprone vs deferoxam(cid:173)
`ine, p=0.967).
`The two treatment groups were similar for body
`iron load, as expressed by mean serum ferritin lev(cid:173)
`els at baseline and at the end of the study. The data
`from this study illustrate that serum ferriti n values
`were not predictive of development of cardiac dis(cid:173)
`ease. Baser! on reported threshold levels, the defer(cid:173)
`oxamine therapy group would have been judged to
`be at lower risk of cardiac disease.' as expressed by
`sequential ferritin assessments (O/o of serum fer(cid:173)
`ritin values lower 2500 µg/L) (Table 2). Hepatic iron
`concentration (HIC), which was measured in a frac(cid:173)
`tion of patients in both treatment groups (data not
`shown), was also not predictive of cardiac disease.
`All HIC values were below 15 mg/g dry weight, a
`suggested threshold for increased risk of cardiac
`disease in patients with iron load.4
`The results of the present analysis are consistent
`with the prevalence of cardiac disease during long(cid:173)
`term treatment with deferoxamine3•7-'3 and with
`the paucity of reports of cardiac disease in patients
`receiving long-term treatment with deferiprone
`even though the early use of deferiprone was gen(cid:173)
`erally restricted to the most severely iron over(cid:173)
`loaded and non-compliant patients.14- 19
`This retrospective study suggests that defe(cid:173)
`riprone may have a cardioprotective effect. If that
`is the case, it may be due to the drug's physico(cid:173)
`chemical characteristics: because of its lipophilic(cid:173)
`ity, neutral charge at pH 7.4, and low molecular
`weight, deferiprone can readily cross cell-mem(cid:173)
`branes and bind intracellular iron.'"""' Although
`the most commonly used dose of deferiprone (75
`mg/kg/day) is approximately twice that of defer(cid:173)
`oxamine (40 mg/kg/day), peak serum concentra(cid:173)
`tions (about 100 µM) are more than 10 times those
`seen with the injectable chelator (<10 µM). At clin(cid:173)
`ically relevant concentrations, deferiprone exhibit(cid:173)
`ed greater iron mobilization than deferoxamine
`from cultured, iron-loaded heart cells." These find(cid:173)
`ings are consistent with the results of clinical stud(cid:173)
`ies using quantitative magnetic resonance imaging
`(qMRI). Previous reports showed that deferiprone
`treatment in patients with thalassemia led to an
`increase of heart signal, compatible with a reduc(cid:173)
`tion in heart iron load.15"' Comparative qMRI stud-
`
`494
`
`haematologica/joumal of hematology vol. 88(05):may 2003
`
`TAR0008751
`
`6 of 8
`
`Taro Pharmaceuticals, Ltd.
`Exhibit 1061
`
`
`
`Comparative effects of deferiprone and deferoxamlne on SUNival
`
`ies indicated that deferiprone may also be more
`effective than deferoxamine in increasing the heart
`signal and improving heart function in patients
`with iron overload.27,28
`Other factors, such as a different sensitivity to
`iron-induced oxidative damage may play a role in
`the development of cardiac disease in patients with
`iron overload.29 Autopsy findings in patients with
`thalassemia who died of cardiac disease showed a
`heart iron concentration 10-fold lower than the
`liver iron concentration.30 Non-transferrin-bound
`iron (NTBI) has been suggested to be a contributor
`to iron damage.29 Since deferiprone acts 10 times
`faster than deferoxamine in mobilizing citrate(cid:173)
`bound iron, one major component of the NTBI
`pool.3 1 it is possible that its greater access to NTBI
`may have contributed to the preferential results. In
`addition, the longer half-life of deferiprone, asso(cid:173)
`ciated with its more frequent dosing (thrice daily)
`and uninterrupted therapy (7 days a week), may
`also serve as a factor leading to greater protection
`than that afforded by the standard regimen of
`deferoxamine (8-12 hours infusion 5 to 7 days a
`week) by providing a longer duration of chelation.
`This retrospective analysis suggests that, in
`patients with thalassemia major, long-term thera(cid:173)
`py with deferiprone results in a greater cardiopro(cid:173)
`tective effect against the toxicity of iron o