Blood Cells, Molecules, and Diseases 47 (2011) 166–175

Contents lists available at ScienceDirect

Blood Cells, Molecules, and Diseases

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / y b c m d

Iron chelation therapy in thalassemia major: A systematic review with meta-analyses

of 1520 patients included on randomized clinical trials
Aurelio Maggio a,⁎, Aldo Filosa b, Angela Vitrano a, Giuseppina Aloj b, Antonis Kattamis c, Adriana Ceci d,
Suthat Fucharoen e, Paolo Cianciulli f, Robert W. Grady g, Luciano Prossomariti h, John B. Porter i,
Angela Iacono j, Maria Domenica Cappellini k, Fedele Bonifazi l, Filippo Cassarà a, Paul Harmatz m,
John Wood n, Christian Gluud o
a
U.O.C. “Ematologia II con Talassemia”, A.O. “V. Cervello”, Palermo, Italy
b
U.O.S. Talassemia Pediatrica, A.O. Cardarelli, Napoli, Italy
c
First Department of Pediatrics, University of Athens, Medical School, Aghia Sophia Children's; Hospital, Thivon and Levadias, Athens, Greece
d
Consorzio per Valutazioni Biologiche e Farmacologiche, Pavia, Italy
e
Thalassemia Research Center, Institute of Science and Technology for Research and Development, Mahdol University, Bangkok, Thailand
f
U.O. D. Talassemia, A.O. “S. Eugenio”, Roma, Italy
g
Department of Pediatrics, Cornell University Medical Center, New York, NY, USA
h
Centro Regionale per la Cura delle Microcitemie, A.O.Cardarelli, Napoli, Italy
i
Department of Haematology, University College London, UCL Cancer Institute, London, UK
j
Foundation Leonardo Giambrone, Italy
k
Policlinico, Mangiagalli, Regina Elena Foundation IRCCS, University of Milan, Italy
l
I.RI.D.I.A. srl, Bari, Italy and Consorzio per Valutazioni Biologiche e Farmacologiche, Pavia, Italy
m
Division of Gastroenterology, Children's Hospital & Research Center Oakland, Oakland, USA
n
Division of Pediatric Cardiology, Children's Hospital Los Angeles, Los Angeles, California, USA
o
Copenhagen Trial Unit, Centre for Clinical Intervention Research, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark

a r t i c l e i n f o a b s t r a c t

Article history: The effectiveness of deferoxamine (DFO), deferiprone (DFP), or deferasirox (DFX) in thalassemia major was
Submitted 26 June 2011 assessed. Outcomes were reported as means ± SD, mean differences with 95% CI, or standardized mean
Revised 2 July 2011 differences. Statistical heterogeneity was tested using χ2 (Q) and I2. Sources of bias and Grading of
Available online 16 August 2011
Recommendations Assessment, Development and Evaluation system (GRADE) were considered. Overall, 1520
patients were included. Only 7.4% of trials were free of bias. Overall measurements suggest low trial quality
(Communicated by G. Stamatoyannopoulos,
M.D., Dr. Sci., 3 July 2011) (GRADE). The meta-analysis suggests lower final liver iron concentrations during associated versus monotherapy
treatment (pb 0.0001), increases in serum ferritin levels during DFX 5, 10, and 20 mg/kg versus DFO-treated
groups (pb 0.00001, p b 0.00001, and p = 0.002, respectively), but no statistically significant difference during
DFX 30 mg/kg versus DFO (p= 0.70), no statistically significant variations in heart T2* signal during associated or
sequential versus mono-therapy treatment (p= 0.46 and p = 0.14, respectively), increases in urinary iron
excretion during associated or sequential versus monotherapy treatment (p= 0.008 and p = 0.02, respectively),
and improved ejection fraction during associated or sequential versus monotherapy treatment (p= 0.01 and
p b 0.00001, respectively). These findings do not support any specific chelation treatment. The literature shows
risks of bias, and additional larger and longer trials are needed.
© 2011 Elsevier Inc. All rights reserved.

Introduction
Thalassemia and hemoglobin disorders are an emerging global
health burden, [1] accounting for about 3.4% of deaths in children less
than 5 years of age [2]. The current therapeutic management of
thalassemia is based on regular blood transfusions and iron chelation
therapy [3]. Chelation aims to reduce iron stores or correct iron
⁎ Corresponding author. Fax: + 39 0916880828.
E-mail address: aureliomaggio@virgilio.it (A. Maggio).
imbalance, and constitutes a major improvement for all people with
thalassemia major [3]. Therefore, standards for the treatment of
children and adults with thalassemia have been developed [4,5].
The three main chelators used in current clinical practice
are deferoxamine (DFO) (Desferal®; Novartis), deferiprone (DFP)
(Ferriprox®; Apotex), and deferasirox (DFX) (Exjade®; Novartis) [6].
The main findings concerning the effectiveness and safety profiles of
DFO, DFP, and DFX have been described extensively elsewhere [7–11].
Moreover, there is also growing off-license usage of DFP in associated
with (administration of the two drugs during the same day) or in
sequential with (administration of the two drugs on different days) DFO

1079-9796/$ – see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.bcmd.2011.07.002
 A. Maggio et al. / Blood Cells, Molecules, and Diseases 47 (2011) 166–175 167

following recent reports of their synergistic effect, particularly with Table 1

regard to the iron burden in the heart [12,13]. We identified eight Study design.

published reviews on iron chelators with our search strategy [14–21]. Inclusion criteria Exclusion criteria
However, only four of these included meta-analyses, [19] none included
Patient populations Patients with Patients with other forms of
sequential chelation treatment, and none reported GRADE. Moreover, thalassemia major thalassemia or chronic
none was conducted systematically, i.e. based on a previously public regardless of age anemia requiring blood
protocol guiding the review process. Furthermore, none assessed the transfusion.
Interventions/comparators Deferoxamine (DFO)
risks of systematic errors (‘bias’), random errors (‘play of chance’), and
versus placebo or
design errors [22–24]. different route of
Because these questions remain unanswered, the main objective of administration
our study was to conduct a systematic review of the clinical Deferiprone (DFP)
effectiveness profile of iron chelators for patients with transfusion- versus DFO
Deferasirox versus DFO
dependent thalassemia major, according to previously published
Associated therapy
criteria [25]. (DFO plus DFP) versus
DFO or DFP alone
Design and methods Sequential therapy
(DFP–DFO) versus DFP
or DFO alone
Literature searches Types of studies Randomized clinical Editorials, letters, opinions,
trials case reports, cross-over
The following electronic databases were utilized (AG and FA) to trials, non-randomized
search for relevant published literature: Cochrane Central Register of studies, quasi-randomized
studies, non-English
Controlled Trials (CENTRAL), MEDLINE, ISI Web of Knowledge, the
language papers.
ClinicalTrials.gov register, Current Controlled Trials, World Health Outcomes Ejection fraction
Organization (WHO) International Clinical Trials Registry Platform Liver iron
(ICTRP), and the reference lists of articles. The date of the last search concentration
Magnetic resonance
was June 2010. Two authors (AG and FA) independently scanned all
T2*
titles and abstracts identified in the research to select reports that Serum ferritin as
might be relevant to the clinical review, separating the cohort studies difference between
from randomized clinical trials (RCTs). Only trials that were clearly baseline and the end of
irrelevant were excluded at this stage. Agreement between the treatment
Urinary iron excretion
authors was good. For trials where important information was
lacking, a letter for request of information was sent to the authors
of the trials.

Keywords utilized for the searches
MEDLINE mesh terms: deferoxamine, desferrioxamine, deferoxa-
mine mesilate, deferoxamine mesylate, deferrioxamine B, Desferal,
desferioximine, desferrioxamine B, desferrioxamine B mesylate,
desferroxamine, deferiprone, 1,2-dimethyl-3-hydroxy-4-pyridinone,
1,2-dimethyl-3-hydroxypyrid-4-one, 1,2-dimethyl-3-hydroxypyridin-
4-one, 3-hydroxy-1,2-dimethyl-4-pyridinone, Chiesi brand of defer-
iprone, Ferriprox, L1 oral chelate, deferasirox, Exjade, ICL670, and
ICL670 A. MEDLINE other key words: iron chelation therapy and
thalassemia, iron chelation and beta-thalassemia. CENTRAL, ISI Web of
Knowledge, ClinicalTrial.gov and WHO ICRTP mesh terms: deferox-
amine AND beta-thalassemia, deferiprone and beta-thalassemia, and
deferasirox and beta-thalassemia. CENTRAL, ISI Web of Knowledge,
ClinicalTrial.gov and WHO ICRTP other key words: iron chelation therapy
and thalassemia, iron chelation and thalassemia, deferoxamine and
thalassemia, deferiprone and thalassemia, deferasirox and thalassemia.
Current controlled trials: deferoxamine, deferiprone, deferasirox,
thalassemia, and iron chelation.
Inclusion and exclusion criteria
The inclusion/exclusion criteria are shown in Table 1. One author
(AG) transcribed the data into the systematic review computer
software RevMan 5.0 [26]. Another author (FA) checked all data entry
for discrepancies and verified the data. Extracted data were analyzed
using the most up-to-date version of RevMan (5.0) available at the
time of analysis (Review Manager 2008). Only randomized clinical
trials (RCT) were analyzed. Phase 2 cross-over trials and studies that
presented poor data or case reports were excluded. Only patients of
any age with transfusion-dependent thalassemia major, from any
setting worldwide, were considered.
The types of interventions described in these papers were (1) DFO
twice daily subcutaneous bolus injection versus DFO subcutaneous
continuous infusion; (2) DFO versus other chelators; (3) DFP versus
other chelators; (4) DFX versus other chelators; (5) DFO plus DFP
versus other chelators; and (6) sequential DFP and DFO versus other
chelators.
Types of outcomes
The following measures were considered: (1) ejection fraction;
(2) liver iron concentration (LIC) (μg/g dry weight) at the end
of treatment; (3) myocardial iron concentration (evaluated by MRI
T2*) at the end of the trial; (4) serum ferritin as the difference
between the final versus the basal value; and (5) urinary iron
excretion.
Statistical analysis
The continuous outcomes relevant to this study were presented as
means and standard deviations (SD), and summarized in terms of
weighted mean difference (WMD) with 95% confidence intervals (CI)
or the standardized mean difference (SMD) when outcomes had a
different measurement scale. The dichotomous outcomes for each
trial were expressed as a relative risk (RR) with 95% CIs, and were
combined in terms of RR for the meta-analysis using the Peto-
modified Mantel–Haenszel method.
The effect was computed across studies using either a fixed-effects
model, if there was little statistical heterogeneity, or a DerSimonian
and Laird random-effects model when there was unexplained
heterogeneity between trial results. Statistical heterogeneity was
tested using the Cochrane chi-square (Q) test of homogeneity and
with I 2 statistic, both available through RevMan 5 [26]. This test has
168 A. Maggio et al. / Blood Cells, Molecules, and Diseases 47 (2011) 166–175

low power to detect true heterogeneity, especially with a small Potentially relevant titles and abstracts identified
number of studies; therefore, a cut-off p-value of 0.10 was used.
through electronic search: 541
The I 2 statistic quantifies inconsistency across studies; a value of I 2
of 25% or less can be considered a low degree of heterogeneity, a
value around 50% as moderate heterogeneity, and a value of 75% or
more as high heterogeneity. In this review, these quantitative
interpretations were limited because few studies were available for Color studies: 504
comparisons.

Evaluation of methodological quality of included trials

Articles selected for meta-analysis: 37
The risk of bias
The internal validity of the trials was assessed considering the
sources of bias that have a major influence on the magnitude of effect
size in clinical trials; adequate sequence generation, allocation Excluded articles: 21
concealment, blinding, incomplete outcome data addressed, free or
selective reporting, and free of other biases.
Other bias is defined as bias not addressed in the other domains in
the tool. Bias is a systematic error; therefore, it only goes one way, an Articles used for meta-analysis: 16 :
overestimation of benefit and an underestimation of harm. Risk of bias
tables were performed using the “Risk of bias” tool in RevMan 5 [26]. Fig. 1. Research flow-chart.
Each of the six domains (adequate sequence generation, allocation
concealment, blinding, incomplete outcome data addressed, free of
selective reporting, free of other bias) for a study, was completed Adequate sequence generation
providing a judgment categorized as “yes”, “no”, or “unclear”. “Yes” Of the RCTs, 48.2% had a low risk of bias; a low number of studies
meant that there is a low risk of bias, and “no” meant that there is a had high risk of bias, whereas the risk for 40.75% of studies was
high risk of bias. “Unclear” was used if there was insufficient classified as “unclear” (Figs. 2–3).
information to make this judgment or if the item was not relevant
to the study [27]. Allocation concealment
Overall, 22.3% of the studies had a low risk of bias, only one study
had a high risk of bias (3.7%), and most of the studies were classified as
“unclear” (74%) (Figs. 2–3).
The GRADE system

Blinding
The Grading of Recommendations Assessment, Development and
For blinding, 18.5% of the studies had a low risk of bias, 37%
Evaluation system (GRADE) was used to summarize the findings and
had a high risk of bias, and 44.5% were classified as “unclear”
rate the overall quality of evidence for each outcome [28]. The
(Figs. 2–3).
GRADE system involves specifying at least 5 main outcomes relevant
to the management of the patients. From these, the guideline
Incomplete outcome data addressed
development process begins. GRADE suggests a nine-point scale to
Overall, 7.4% of the studies had a low risk of bias, most studies
judge importance. Ratings of 7 to 9 equal outcomes of critical
(85.2%) had high risk of bias, and 7.4% of studies were classified as
importance for decision making. Ratings of 4 to 6 indicate outcomes
“unclear” (Figs. 2–3).
important but not critical. Ratings of 1 to 3 equal items of limited
importance [27].
Free of selective reporting
In this category, 7.4% of the studies had a low risk of bias; in
contrast, 70.4% of studies had a high risk of bias and 22.2% were
Results classified as “unclear” (Figs. 2–3).

Search results Free of other bias

The results of our research strategies are summarized in Fig. 1.
Sixteen RCTs were included in this review with a total of 1520 patients
aged from 5 to 50 years. Five RCTs were long-term follow-ups of the
trials already included in the analysis. Within these 16 trials, 11
different comparisons were detected. In these, the chelators were
compared with each other as monotherapies, in combination with
each other, or as sequentially administered chelators. Table 2 reports
the main findings of the RTCs included in this systematic review and
used in the meta-analysis.
The risk of bias
Overall, 7.4% of the studies had a low risk of bias; in contrast, most
studies (88.9%) had a high risk of bias, and 3.7% were classified as
“unclear” (Figs. 2–3).
The GRADE system
Randomized trials are considered the highest quality evidence
unless limited by methodological issues, inconsistency, indirectness
(generalizability), or imprecision [28]. In this review, all outcomes
were limited by weaknesses related to trial quality, such as lack of
blinding, small sample size, and imprecise estimates of treatment
effects. This resulted in a low quality for most outcomes. The 5 most
important outcomes chosen for quality assessment are described
below in decreasing order of importance:

The risk of bias is reported as a summary of the studies for each 1) Ejection fraction: GRADE = 9
judgment (proportion) in Fig. 2, and as single evaluation for each RCT 2) Liver iron concentration (LIC) (μg/g dry weight) at the end of
in Fig. 3. treatment: GRADE = 8
 A. Maggio et al. / Blood Cells, Molecules, and Diseases 47 (2011) 166–175 169

Table 2
Type of comparison among interventions reported in randomized clinical trials included in the systematic review with meta-analyses.

Comparison Exp Ctrl Exp Ctrl N. pts Author

N. pts

Monotherapy 1 DFO bolus/day s.c.i.a 10 10 Yarali et al. [49]

2 DFP DFO 10 10 Gomber et al. [33]
DFP DFO 6 7 Ha et al. [45]
DFP DFO 71 73 Maggio et al. [47]
DFP DFO 21 23 El Beshlawy et al. [50]
DFP DFO 29 32 Pennell et al. [41]
3 DFX (10 mg) Placebo 5 5 Nibset-Brown et al. [46]
4 DFX (20 mg) Placebo 6 5 Nibset-Brown et al. [46]
5 DFX (40 mg) Placebo 7 5 Nibset-Brown et al. [46]
6 DFX (5 mg) DFO 15 14 Cappellini et al. [32]
7 DFX (10 mg) DFO 78 79 Cappellini et al. [32]
8 DFX (20 mg) DFO 84 91 Cappellini et al. [32]
9 DFX (30 mg) DFO 119 106 Cappellini et al. [32]
Associated 10 10.0 versus DFP DFP + DFO DFP 12 12 Aydinok et al. [29]
10.1 versus DFO DFP + DFO DFO 10 10 Gomber et al. [33]
DFP + DFO DFO + placebo 32 33 Tanner et al. [13]
DFP + DFO DFO 22 23 El-Beshlawy et al. [50]
DFP + DFO DFO 20 16 Ha et al. [45]
DFP + DFO DFO 11 14 Mourad et al. [48]
Sequential 11 11.0 versus DFP DFP − DFO DFP 105 108 Maggio et al. [31]
11.1 versus DFO DFP − DFO DFO 29 30 Galanello et al. [44]
DFP − DFO DFO 30 30 Abdelrazik et al. [30]
a
s.c.i. = subcutaneous continuous infusion; Exp: experimental interventions, Ctrl: control interventions =; =; =.

3) Myocardial iron concentration (evaluated by MR T2*) at the end of Comparison of sequential DFP and DFO versus DFO. Ejection fraction at
the trial: GRADE = 8 the end of the intervention was significantly higher in the sequential
4) Serum ferritin levels such as the difference between the final DFP and DFO groups than in the DFO control group [30] (Fig. 4b)
versus the basal value: GRADE = 7 (WMD 9.02, 95% CI 6.4 to 11.64, p b 0.00001). GRADE quality of
5) Urinary iron excretion: GRADE = 6. evidence was low.

Outcome 2: Liver iron concentration (LIC) (μg/g dry weight) at the end
Outcome results of treatment

Outcome 1: Ejection fraction at the end of treatment
Comparisons of DFP plus DFO versus other chelators. Ejection fraction
at the end of the intervention was reported in two trials [13–29]
(Fig. 4a) comparing 36 participants in the deferiprone plus DFO
group with 42 participants in the control group, which included
12 participants receiving DFP [29] and 30 receiving DFO [13].
Ejection fraction at the end of intervention was significantly
higher in the DFP plus DFO groups (WMD 3.37, 95% CI 0.79 to
5.95, p = 0.01). GRADE quality of evidence was low. Heterogene-
ity for this outcome was not statistically significant (CHI 2 = 0.29,
df = 1, p = 0.59, I 2 0%).
Comparisons of DFP plus DFO versus other chelators. Liver iron
concentration at the end of the intervention was measured in two
studies as liver iron concentration by biopsy (mg/g dw) [29] and liver
iron concentration by magnetic resonance imaging as (T2*) (ms) [13]
using different techniques (Fig. 4c). The results were pooled with
standardized mean difference. Thirty-six participants in the defer-
iprone plus deferoxamine group were compared with 42 participants
in the control group, which included 12 participants who received
DFP [29] and 30 who received DFO [13]. Liver iron concentration at the
end of the intervention was significantly lower in the DFP plus DFO
groups compared with the control group (WMD − 1.06, 95% CI −1.54
to − 0.58, p b 0.0001). GRADE quality of evidence was very low.

Fig. 2. Risk of bias summary.

170 A. Maggio et al. / Blood Cells, Molecules, and Diseases 47 (2011) 166–175
Fig. 3. Risk of bias graph for each study.
Heterogeneity was not statistically significant (CHI 2 = 0.35, df = 1,
p = 0.55, I 2 0%).
Outcome 3: Myocardial iron concentration (evaluated by MRI T2*) at
the end of the treatment
Comparison of deferiprone plus DFO versus DFO. Myocardial iron
concentration (T2*) at the end of the intervention was not
significantly lower in the DFP plus DFO groups in comparison with
the control group (WMD = 1.7, 95% CI −2.78 to 6.18, p = 0.46)
(Fig. 5a) [13]. GRADE quality of evidence was low.
Comparison of sequential DFP and DFO versus DFP. Myocardial iron
concentration (T2*) at the end of the intervention was not
significantly different between participants taking sequential DFP
and DFO compared with those taking DFP in the control group
(Fig. 5b)[31]. However, the myocardial iron concentration (T2*) at the
end of the intervention was lower in the control group compared with
the sequential DFP and DFP groups (WMD = − 5.8, 95% CI −13.52 to
1.92, p = 0.14). GRADE quality of evidence was low.
Outcome 4: Serum ferritin as the difference between final versus basal
values
Comparison of DFX (5 mg/kg) and DFO (30 mg/kg). A statistically
significant increase in serum ferritin levels, determined as the
difference between baseline values and values at the end of
intervention, was shown in the group of patients treated with DFX
5 mg/kg/day compared with the DFO control group (WMD 978, 95% CI
544.71 to 1411.29, p b 0.00001) (Fig. 6a) [32]. GRADE quality of
evidence was low.
Comparison of DFX (10 mg/kg) versus DFO (35 mg/kg). A statistically
significant increase in serum ferritin levels, determined as the
difference between baseline values and value at the end of
intervention, was shown in the group of patients treated with DFX
10 mg/kg/day compared with the DFO control group (WMD 801, 95%
CI 565.98 to 1036.02, p b 0.00001) (Fig. 6b) [32]. GRADE quality of
evidence was low.
Comparison of DFX 20 mg/kg with DFO 40 mg/kg. A statistically
significant increase in serum ferritin levels, determined as the difference
between baseline values and the value at the end of intervention, was
shown in the group of patients treated with DFX 20 mg/kg/day
compared with the DFO control group [32] (WMD 328, 95% CI 121.65
to 534.35, p=0.002) (Fig. 6c) (Table 3). GRADE quality of evidence was
low.
Comparison of DFX 30 mg with DFO 51 mg/kg. No statistically
significant difference was observed in the group of patients treated
with DFX 30 mg/kg/day compared with the DFO control group [32]
(WMD 77.00, 95% CI −311.91 to 465.91, p = 0.70) (Fig. 6d) (Table 3).
Outcome 5: Urinary iron excretion
Comparison of DFP plus DFO versus other chelators. Urinary iron
excretion was measured in two trials [29,33] (Aydinok 2007
[urinary iron excretion (mg/kg/day)]; Gomber 2004 [urinary iron
excretion (mg/day)]) in 18 participants who received DFP plus DFO,
and 19 in the control group, which included 12 participants who
received DFP [29] and 7 who received DFO [33] (Fig. 7a). Urinary
iron excretion was significantly higher in the DFP plus DFO groups
(WMD 1.28, 95% CI 0.53 to 2.02, p = 0.0008). GRADE quality of
evidence was low. Heterogeneity for this outcome was not
statistically significant (CHI 2 = 1.74, df = 1, p = 0.19, I 2 43%).
Comparison of sequential deferiprone and deferoxamine versus defer-
oxamine. Urinary iron excretion was significantly higher in the
sequential DFP and DFO groups compared with the DFO control
group [30] (WMD 0.23, 95% CI 0.04 to 0.42, p = 0.02) (Fig. 7b). GRADE
quality of evidence was very low.
Discussion
The aim of this systematic review was to evaluate the effectiveness
of DFO, DFP, and DFX alone or in associated or as sequential therapies
in transfusion-dependent patients with thalassemia major. This meta-
analysis suggests that ejection fraction at the end of the intervention
 A. Maggio et al. / Blood Cells, Molecules, and Diseases 47 (2011) 166–175 171

Outcome 1: Ejection Fraction at the end of intervention

(a) Associated deferiprone plus deferoxamine versus other chelators

(b) Sequential deferiprone and deferoxamine versus deferoxamine

Outcome 2: Liver Iron concentration at the end of treatment

(c) Associated Deferiprone plus deferoxamine versus deferiprone

Fig. 4. Foster plot analysis of the outcomes 1 and 2.

was statistically significantly higher after combined (Fig. 4a) or
sequential DFP and DFO treatments (Fig. 4b). Liver iron concentration
at the end of the intervention was statistically significantly decreased
after associated DFP plus DFO treatments (Fig. 4c). Myocardial iron
concentration (T2*) at the end of the intervention did not show any
statistically significant change after associated (Fig. 5a) or sequential
chelation treatment (Fig. 5b). Serum ferritin, measured as the
difference between the final versus the basal value, was statistically
significantly increased after treatment with DFX 5 mg, 10 mg, and
20 mg (Fig. 6a,b,c), and urinary iron excretion was statistically
significantly increased after associated (Fig. 7a) or sequential DFP
and DFO treatments (Fig. 7b).
172 A. Maggio et al. / Blood Cells, Molecules, and Diseases 47 (2011) 166–175

Outcome 3: Myocardial Iron Concentration (evaluated by MR T2*) at the end of the intervention
(a) Associated deferiprone plus deferoxamine versus deferoxamine

(b) Sequential deferiprone and deferoxamine versus deferoxamine

Fig. 5. Foster plot analysis of the outcome 3.

Iron-induced heart failure is the most common cause of death in
patients with thalassemia major, accounting for about 70% of deaths
[3]. Therefore, maintaining left ventricular ejection fraction over the
recommended threshold of 55% [34] may play a critical role in
preventing heart failure in thalassemia major patients. The detection
of a higher ejection fraction at the end of the intervention after
associated (Fig. 4a) or sequential DFP and DFO treatments (Fig. 4b) is
consistent with the reported reduction in mortality due to heart
damage during retrospective and prospective studies carried out in
thalassemia major patients treated with DFP [35–38].
The statistically significantly lower liver iron concentration at the
end of intervention in the DFP plus DFO groups may be explained by
the previously described synergistic action between these two
chelators [39,40]. The greater effectiveness of associated DFP plus
DFO treatments is explained as the result of a transfer of chelated iron
from DFP to DFO; i.e. the “shuttle” hypothesis [39,40].
Myocardial iron concentration (T2*) at the end of the intervention
was not statistically significantly different in the DFP plus DFO versus
the DFO-treated groups [13] (Fig. 5a). However, a different result was
reported by Tanner et al. [13] who used a different statistical analysis

Outcome 4: Serum Ferritin as difference between final versus basal values

(a) Deferasirox 5 mg versus deferoxamine

(b) Deferasirox 10 mg versus deferoxamine

(c) Deferasirox 20 mg versus deferoxamine

(d) Deferasirox 30 mg versus deferoxamine

Fig. 6. Foster plot analysis of the outcome 4.

 A. Maggio et al. / Blood Cells, Molecules, and Diseases 47 (2011) 166–175 173

Table 3
Combined chelation treatment: different schedules of administration.

Interventions Control group

Dosage of DFP Dosage of DFO Versus other chelators

(mg/kg/day) (mg/kg)

Aydinok [29] 75 40–50 for Deferiprone 75 mg/kg/day

2 days/week
Gomber [33] 75 40 for 2 days/week Deferoxamine 40 mg/kg/day for 5 days/week
Tanner [13] 75 40–50 for many days Deferoxamine 40–50 mg/kg/day for
minimum 5 days minimum 5 days or more
El_Beshlawy [50] 60–83 23–50 for 2 days/week Deferoxamine 25–50 mg/kg/day for 5 nights
Ha [45] 75 30–60 for 2 days/week Deferoxamine 30–60 mg/kg/day for 5–7 days
Mourad [48] 75 Total of 2 gr for 2 days Deferoxamine 40–50 mg/kg/day for
5–7 days/week

methodology. Actually, the myocardial T2* was log-transformed
before analysis for normalization. The statistical analysis was
performed over time within the groups from baseline to 12 months.
Moreover, the analysis was carried out with a mixed-effects model
with time as a fixed effect. In contrast, the statistical analysis
recommended for meta-analysis is based on the weighted mean
difference (WMD) between the two groups at only one time point
[41]. This difference in the statistical analysis methodology may
explain the discrepancy in the results. However, these findings may
also suggest that larger and more robust multicenter randomized
clinical trials are necessary to determine whether or not clinically
significant cardiac T2* signal variations can be demonstrated among
chelators.
The comparisons between DFX (5 mg), DFX (10 mg), and DFX
(20 mg) versus DFO (Table 3) showed an increase in serum ferritin
levels at the end of the study [32]. The three doses of DFX
examined may be too low to effectively decrease serum ferritin
levels in the intervention groups. Furthermore, even the reduction
in serum ferritin levels by DFX 30 mg/day was not statistically
significantly different compared with DFO treatment [32]. More-
over, Cohen et al. suggested an effect of dose and transfusion in
DFX- versus DFO-treated patients [42]. A recently published,

Outcome 5 Urinary Iron Excretion

(a) Associated deferiprone plus deferoxamine versus other chelators

(b) Sequential deferiprone and deferoxamine versus deferoxamine

Fig. 7. Foster plot analysis of the outcome 5.

174 A. Maggio et al. / Blood Cells, Molecules, and Diseases 47 (2011) 166–175
Table 4
Sequential chelation treatment: different schedules of administration.
Interventions
Dosage of DFP
mg/kg Days
Maggio [31] 75 4 50
Galanello [44] 75 5 N.R.
Abdelrazik [30] 75 for 4 days a week
N.R. = not reported.
prospective, but non randomized escalator trial suggested an
increased effect with dosage of DFX greater than 30 mg/kg in a
large cohort of patients [43]. However, because a comparator
group was lacking, it is difficult to evaluate the significance of this
result [43].
The greater effectiveness of associated and sequential treatments
in increasing urinary iron excretion could be explained by the “shuttle”
hypothesis [39,40].
This meta-analysis has some important limitations. The quality of
included randomized trials was very low because only 50% (n = 8) of
the trials showed adequate sequence generation, whereas only 25%
showed a low risk of bias regarding allocation concealment and
blinding, and only 7.4% showed a low risk of bias for incomplete
outcome data addressed, selective reporting bias, and other bias.
Therefore, there were few trials conducted with an optimal design.
Furthermore, there were a number of studies excluded from the
analysis because they did not reach the minimal criteria to be
considered randomized controlled trials.
There were very few trials for each comparison, and the
heterogeneity of the data between the trials together with the
different treatment schedules (Tables 3–4) made comparisons and
conclusions difficult.
Implications for practice
This systematic review with meta-analyses of randomized clinical
trials does not suggest any change in chelation treatment protocols.
Evidence supporting the effectiveness of associated and sequential
DFP plus DFO treatment should be confirmed in larger randomized
clinical trials before having definitive implications for clinical practice.
Finally,these findings suggest as authorization by the national and
international regulatory agencies of the current off-license use of
associated and sequential DFP plus DFO protocols should be urgently
considered.
Implications for research
This review suggests that we need to have (1) larger and longer
randomized clinical trials for each comparison; (2) improvement of
the quality of future clinical trials; and (3) decreased heterogeneity in
the direct and indirect methods for measuring iron overload.
Support and sponsorships
Nine trials reported the source of funding [11,13,29,31,41,44–47].
The trials that received funding from a governmental agency
included co-funding from the Sicilian Thalassaemic Association and
European Community [47].
The trials that documented support from industry included
Novartis Pharma [11] and Lipomed, Switzerland, for DFP, [29,44]
and Novartis Pharmaceuticals Corporation, [46] Apotex, Novartis,
Apopharma [45].
One trial received funding from both non industry and industry;
CORDA, Royal Brompton, Harefield Hospital Charitable Funds, Cool-
Control group
Dosage of DFO
mg/kg Days
3 Deferiprone 75 mg/kg/day
2 Deferoxamine 34.8 for 5–7 days
40 for 2 days Deferoxamine 40 mg/kg/day for 5–7 days
ey's Anemia Foundation, Apotex, Apopharma, UK Thalassemia Society,
University College London Special Trustees Charity, Novartis, Siemens
Medical Solutions, Cardiovascular Imaging Solutions, and the British
Heart Foundation [13].
Authorship contributions
AF, AV, GA, AM, and CG designed the research, performed the
research, and prepared an initial draft of the paper.
AK, AC, MDC, SF, PC, RWG, LP, JBP, FB, and FC analyzed the data and
discussed the draft issue by issue.
AF and AM wrote the final draft of the paper.
PH and JW addressed scientific and language issues as interna-
tional external reviewers.
AI was a representative of the patients who examined the
document.
Declarations of interest
Conflict of interest disclosure: AF, AM, AV, GA, CG, FB, LP, FC, AI,
RWG, AC, and SF declare no competing financial interests; MDC, PC,
and JBP report having received consulting fees from Novartis
Pharmaceuticals. PH received research grants from Novartis, and
support from Novartis and Ferrokin, and participated in medical
advisory boards for Apotex; JW received research grants from
Novartis and is a consultant for Ferrokin Biosciences. AK reports
participating in study monitor committees for Novartis Pharmaceu-
ticals, and receiving lecture fees from Novartis Pharmaceuticals and
Apopharma Inc.
Acknowledgments
The research leading to these results has received funding by the
Italian Ministry of Health and the Italian Foundation for the Cure of
Thalassaemia “Leonardo Giambrone”, under the project “Inter-
regional Network for Thalassaemia: HTA for the diagnostic and
therapeutic intervention for iron overload”, coordinated by the
Basilicata Region. Prof. David Nathan provided valuable advice that
was greatly appreciated. We thank Foundation Franco and Piera
Cutino for their continuous encouragement. Finally, we thank Sarah
Louise Klingenberg and Dimitrinka Nikolova for their help in
upgrading the literature research.
References
[1] D. Weatherall, The inherited diseases of hemoglobin are an emerging global
health burden, Blood 115 (2010) 4331–4336.
[2] B. Modell, M. Darlison, Global epidemiology of haemoglobin disorders and
derived service indicators, Bull. World Health Organ. 86 (2008) 480–487.
[3] C. Borgna-Pignatti, S. Rigolotto, P. De Stefano, H. Zhao, M.D. Cappellini, G.C. Del
Vecchio, M.A. Romeo, G.L. Forni, M.R. Gamberini, R. Ghilardi, A. Piga, A. Cnaan,
Survival and complications in patients with thalassemia major treated with
transfusion and deferoxamine, Haematologica 89 (2004) 1187–1193.
[4] A. Yardumian, P. Telfer, P. Darbyshire, Standards for the Clinical Care of Children
and Adults with Thalassaemia in the UK, United Kingdom Thalassaemia
Society2nd edition, 2008.
 A. Maggio et al. / Blood Cells, Molecules, and Diseases 47 (2011) 166–175
[5] J. Porter, A. Cohen, J. Kwiatkowski, Transfusion and iron chelation, in: M.
Steinberg, B. Forget, D. Higgs, D. Weatheral (Eds.), Disorders of Hemoglobin,
Cambridge University Press, New York, 2009, pp. 689–744.
[6] A. Maggio, Light and shadows in the iron chelation treatment of haematological
diseases, Br. J. Haematol. 138 (2007) 407–421.
[7] Electronic Medicines Compendium, Desferal — Summary of Product Character-
istics. URL: http://emc.medicines.org.ukAccessed 28 September 2006.
[8] EMEA, Ferriprox EPAR. 1999. URL: www.emea.europa.eu/humandocs/Humans/
EPAR/ferriprox/ferriprox.htm.
[9] British Medical Association and Royal Pharmaceutical Society of Great Britain,
British National Formulary (BNF). No. 53, March 2007. London: BMA and RPS;
2007. URL: www.bnf.org/bnf/ URL .
[10] EMEA, Ferriprox: Annex I — Summary of Product Characteristics. 2007. URL:
www.emea.europa.eu/humandocs/Humans/EPAR/ferriprox/ferriprox.htm.
Accessed 2 February 2007.
[11] A. Piga, R. Galanello, G.L. Forni, M.D. Cappellini, R. Origa, A. Zappu, G. Donato, E.
Bordone, A. Lavagetto, I. Zanaboni, R. Sechaud, N. Hewson, J.M. Ford, H.O. Pitz,
D. Alberti, Randomized phase II trial of deferasirox (Exjade, ICL670), a once-
daily, orally-administered iron chelator, in comparison to deferoxamine in
thalassemia patients with transfusional iron overload, Haematologica 91
(2006) 873–880.
[12] R. Origa, P. Bina, A. Agus, G. Crobu, E. Defraia, C. Dessì, G. Leoni, P.P. Muroni, R.
Galanello, Combined therapy with deferiprone and desferrioxamine in thalasse-
mia major, Haematological 90 (2005) 1309–1314.
[13] M.A. Tanner, R. Galanello, C. Dessi, G.C. Smith, M.A. Westwood, A. Agus, M.
Roughton, R. Assomull, S.V. Nair, J.M. Walker, D.J. Pennell, A randomized, placebo-
controlled, double-blind trial of the effect of combined therapy with deferox-
amine and deferiprone on myocardial iron in thalassemia major using
cardiovascular magnetic resonance, Circulation 115 (2007) 1876–1884.
[14] D.J. Roberts, D. Rees, J. Howard, C. Hyde, P. Alderson, S. Brunskill, Desferrioxamine
mesylate for managing transfusional iron overload in people with transfusion-
dependent thalassaemia, Cochrane Database Syst. Rev. (2005).
[15] A. Addis, R. Loebstein, G. Koren, T.R. Einarson, Meta-analytic review of the clinical
effectiveness of oral deferiprone (L1), Eur. J. Clin. Pharmacol. 55 (1999) 1–6.
[16] M. Franchini, D. Veneri, Iron overload and hematologic malignancies, Hematol. J. 5
(2004) 381–383.
[17] H.E. Vanorden, T.M. Hagemann, Deferasirox — an oral agent for chronic iron
overload, Ann. Pharmacother. 40 (2006) 1110–1117.
[18] L. Abetz, J.F. Baladi, P. Jones, D. Rofail, The impact of iron overload and its
treatment on quality of life: results from a literature review, Health Qual. Life
Outcomes 4 (2006) 73.
[19] C. McLeod, N. Fleeman, J. Kirkham, A. Bagust, A. Boland, P. Chu, R. Dickson, Y.
Dundar, J. Greenhalgh, B. Modell, A. Olujohungbe, P. Telfer, T. Walley, Deferasirox
for the treatment of iron overload associated with regular blood transfusions
(transfusional haemosiderosis) in patients suffering with chronic anaemia: a
systematic review and economic evaluation, Health Technol. Assess. 13 (2009)
1–121.
[20] Malaysian Health Technology Assessment Unit Management of Thalassemia Kuala
Lumpur, Malaysian Health Technology Assessment Unit MHTAU, 2003.
[21] J.J. Caro, K.F. Huybrechts, T.C. Gree, Estimates of the effect on hepatic iron of oral
deferiprone compared with subcutaneous desferrioxamine for treatment of iron
overload in thalassemia major: a systematic review, BMC Blood Dis. 2 (2002) 4.
[22] L. Kjaergaard, J. Villumsen, C. Gluud, Reported methodological quality and
discrepancies between small and large randomized trials in meta-analyses,
Annals of Intern. Med. 135 (2001) 982–989.
[23] D. Moher, B. Pham, A. Jones, D.J. Cook, A.R. Jadad, M. Moher, P. Tugwell, T.P.
Klassen, Does quality of reports of randomised trials affect estimates of
intervention efficacy reported in meta-analyses, Lancet 352 (1998) 609–613.
[24] L. Wood, M. Egger, L.L. Gluud, K.F. Schulz, P. Jüni, D.G. Altman, C. Gluud, R.M.
Martin, A.J. Wood, J.A. Sterne, Empirical evidence of bias in treatment effect
estimates in controlled trials with different interventions and outcomes: meta-
epidemiological study, Br. Med. J. 336 (2008) 601–605.
[25] M. Crowther, W. Lim, M.A. Crowther, Systematic review and meta-analysis
methodology, Blood 116 (2010) 3140–3146.
[26] Review Manager, Review Manager (RevMan) [Computer program]. Version 5.0,
The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, 2008.
[27] Handbook rev man 5.0 (2008), http://ims.cochrane.org/revman/documentation.
[28] D. Atkins, M. Eccles, S. Flottorp, G.H. Guyatt, D. Henry, S. Hill, D. Henry, S. Hill,
A. Liberati, D. O'Connell, A.D. Oxman, B. Phillips, H. Schünemann, T.T. Edejer,
G.E. Vist, J.W. Williams, GRADE Working Group, Systems for grading the
quality of evidence and the strength of recommendations I: critical appraisal
of existing approaches The GRADE Working Group, BMC Health Serv. Res. 4
(2004) 38.
[29] Y. Aydinok, Z. Ulger, D. Nart, A. Terzi, N. Cetiner, G. Ellis, A. Zimmermann, C. Manz,
A randomized controlled 1-year study of daily deferiprone plus twice weekly
desferrioxamine compared with daily deferiprone monotherapy in patients with
thalassemia major, Haematologica 92 (2007) 1599–1606.
[30] N. Abdelrazik, Pattern of iron chelation therapy in Egyptian beta thalassemic
patients: Mansoura University Children's Hospital experience, Hematology 12
(2007) 577–585.
[31] A. Maggio, M. Capra, A. Pepe, L. Mancuso, E. Cracolici, S. Vitabile, P. Rigano, F.
Cassarà, M. Midiri, A critical review of non invasive procedures for the evaluation
175
of body iron burden in thalassemia major patients, Pediatr. Endocrinol. Rev. 6
(2008) 193–203.
[32] M.D. Cappellini, A. Cohen, A. Piga, M. Bejaoui, S. Perrotta, L. Agaoglu, Y. Aydinok,
A. Kattamis, Y. Kilinc, J. Porter, M. Capra, R. Galanello, S. Fattoum, G. Drelichman,
C. Magnano, M. Verissimo, M. Athanassiou-Metaxa, P. Giardina, A. Kourakli-
Symeonidis, G. Janka-Schaub, T. Coates, C. Vermylen, N. Olivieri, I. Thuret, H.
Opitz, C. Ressayre-Djaffer, P. Marks, D. Alberti, A phase 3 study of deferasirox
(ICL670), a once-daily iron chelator; in patients with beta-thalassemia, Blood 107
(2006) 3455–3462.
[33] S. Gomber, R. Saxena, N. Madan, Comparative efficacy of desferrioxamine,
deferiprone and in combination on iron chelation in thalassemic children, Indian
Pediatr. 41 (2004) 21–27.
[34] T. Cogliandro, G. Derchi, L. Mancuso, M.C. Mayer, B. Pannone, A. Pepe, M. Pili, P.
Bina, P. Cianciulli, V. De Sanctis, A. Maggio, Guideline recommendations for heart
complications in thalassemia major, J. Cardiovasc. Med (Hagerstown) 9 (2008)
515–525.
[35] C. Borgna-Pignatti, M.D. Cappellini, P. De Stefano, G.C. Del Vecchio, G.L. Forni, M.R.
Gamberini, R. Ghilardi, A. Piga, M.A. Romeo, H. Zhao, A. Cnaan, Cardiac morbidity
and mortality in deferoxamine- or deferiprone-treated patients with thalassemia
major, Blood 107 (2006) 3733–3737.
[36] A. Maggio, A. Vitrano, M. Capra, L. Cuccia, F. Gagliardotto, A. Filosa, M.A. Romeo, C.
Magnano, V. Caruso, C. Argento, C. Gerardi, S. Campisi, P. Violi, R. Malizia, P.
Cianciulli, M. Rizzo, D.G. D'Ascola, A. Quota, L. Prossomariti, C. Fidone, P. Rigano, A.
Pepe, G. D'Amico, A. Morabito, G. Gluud, Long-term sequential deferiprone-
deferoxamine versus deferiprone alone for thalassaemia major patients: a
randomized clinical trial, Br. J. Haematol. 145 (2009) 245–254.
[37] A. Maggio, A. Vitrano, M. Capra, L. Cuccia, F. Gagliardotto, A. Filosa, C. Magnano, M.
Rizzo, V. Caruso, C. Gerardi, C. Argento, S. Campisi, F. Cantella, F. Commendatore,
D.G. D'Ascola, C. Fidone, A. Ciancio, M.C. Galati, G. Giuffrida, R. Cingari, G. Giugno,
T. Lombardo, L. Prossomariti, R. Malizia, A. Meo, G. Roccamo, M.A. Romeo, P. Violi,
P. Cianciulli, P. Rigano, Improving survival with deferiprone treatment in patients
with thalassemia major: a prospective multicenter randomised clinical trial Under
the auspices of the Italian Society for Thalassemia and Hemoglobinopathies, Blood
Cells Mol. Dis. 42 (2009) 247–251.
[38] B. Modell, M. Khan, M. Darlison, M.A. Westwood, D. Ingram, D.J. Pennell, Improved
survival of thalassaemia major in the UK and relation to T2* cardiovascular
magnetic resonance, J. Cardiovasc. Magn. Reson. 10 (2008) 42.
[39] G. Link, A.M. Konijn, W. Breuer, Z.I. Cabantchik, C. Hershko, Exploring the “iron
shuttle” hypothesis in chelation therapy: effects of combined deferoxamine and
deferiprone treatment in hypertransfused rats with labeled iron stores and in
iron-loaded rat heart cells in culture, J. Lab. Clin. Med. 138 (2001) 130–138.
[40] P.J. Giardina, R.W. Grady, Chelation therapy in beta-thalassemia: an optimistic
update, Semin. Hematol. 38 (2001) 360–366.
[41] D.J. Pennell, V. Berdoukas, M. Karagiorga, V. Ladis, A. Piga, A. Aessopos, E.D. Gotsis,
M.A. Tanner, G.C. Smith, M.A. Westwood, B. Wonke, R. Galanello, Randomized
controlled trial of deferiprone or deferoxamine in betathalassemia major patients
with asymptomatic myocardial siderosis, Blood 107 (2006) 3738–3744.
[42] A.R. Cohen, E. Glimm, J.B. Porter, Effect of transfusional iron intake on response to
chelation therapy in β-thalassemia major, Blood 111 (2008) 583–587.
[43] A. Taher, A. El-Beshlawy, M.S. Elalfy, K. Al Zir, S. Daar, D. Habr, U. Kriemler-Krahn,
A. Hmissi, A. Al Jefri, Efficacy and safety of deferasirox, an oral iron chelator, in
heavily iron-overloaded patients with beta-thalassaemia: the ESCALATOR study,
Eur. J. Haematol. 82 (2009) 458–465.
[44] R. Galanello, A. Kattamis, A. Piga, R. Fischer, G. Leoni, V. Ladis, V. Voi, U. Lund, F.
Tricta, A prospective randomized controlled trial on the safety and efficacy of
alternating deferoxamine and deferiprone in the treatment of iron overload in
patients with thalassemia, Haematological 91 (2006) 1241–1243.
[45] S.Y. Ha, K.W. Chik, S.C. Ling, A.C. Lee, C.W. Luk, C.W. Lam, I.O. Ng, G.C. Chan, A
randomized controlled study evaluating the safety and efficacy of deferiprone
treatment in thalassemia major patients from Hong Kong, Hemoglobin 30 (2006)
263–274.
[46] E. Nisbet-Brown, N.F. Olivieri, P.J. Giardina, R.W. Grady, E.J. Neufeld, R. Sèchaud, A.J.
Krebs-Brown, J.R. Anderson, D. Alberti, K.C. Sizer, D.G. Nathan, Effectiveness
and safety of ICL670 in iron-loaded patients with thalassaemia: a randomised,
double-blind, placebo-controlled, dose-escalation trial, Lancet 361 (2003)
1597–1602.
[47] A. Maggio, G. D'Amico, A. Morabito, M. Capra, C. Ciaccio, P. Cianciulli, F. Di
Gregorio, G. Garozzo, R. Malizia, C. Magnano, A. Mangiagli, G. Quarta, M. Rizzo,
D.G. D'Ascola, A. Rizzo, M. Midiri, Deferiprone versus deferoxamine in patients
with thalassemia major: a randomized clinical trial, Blood Cells Mol. Dis. 28
(2002) 196–208.
[48] F.H. Mourad, A.V. Hoffbrand, M. Sheikh-Taha, S. Koussa, A.L. Khoriaty, A. Taher,
Comparison between desferrioxamine and combined therapy with desferriox-
amine and deferiprone in iron overloaded thalassaemia patients, Br. J. Haematol.
121 (2003) 187–189.
[49] N. Yarali, T. Fisgin, F. Duru, A. Kara, N. Ecin, S. Fitoz, I. Erden, Subcutaneous bolus
injection of deferoxamine is an alternative method to subcutaneous continuous
infusion, J. Pediatric. Hematol. Oncol. 28 (2006) 11–16.
[50] A. El-Beshkawy, C. Manz, M. Naja, M. Eltagui, C. Tarabishi, I. Youssry, H. Sobh, M.
Hamdy, I. Sharaf, A. Mostafa, O. Shaker, A.V. Hoffbrand, A. Taher, Iron chelation in
thalassemia: combined or monotherapy ? The Egyptian experience, Ann.
Hematol. 87 (2008) 545–550.