Anda di halaman 1dari 11

The n e w e ng l a n d j o u r na l of m e dic i n e

clinical therapeutics

Iron-Chelating Therapy
for Transfusional Iron Overload
Gary M. Brittenham, M.D.

This Journal feature begins with a case vignette that includes a therapeutic recommendation. A discussion
of the clinical problem and the mechanism of benefit of this form of therapy follows. Major clinical studies,
the clinical use of this therapy, and potential adverse effects are reviewed. Relevant formal guidelines,
if they exist, are presented. The article ends with the author’s clinical recommendations.

A 16-year-old boy with sickle cell anemia undergoes routine screening with transcra-
nial Doppler ultrasonography to assess the risk of stroke. This examination shows an
abnormally elevated blood-flow velocity in the middle cerebral artery. The hemoglo-
bin level is 7.2 g per deciliter, the reticulocyte count is 12.5%, and the fetal hemoglo-
bin level is 8.0%. Long-term treatment with red-cell transfusion is initiated to prevent
stroke. A hematologist recommends prophylactic iron-chelating therapy.

The Cl inic a l Probl em

From the Division of Pediatric Hematol- Long-term treatment with red-cell transfusion effectively prevents stroke and other
ogy, Department of Pediatrics, Columbia complications of sickle cell anemia1 and can sustain patients with chronic con-
University College of Physicians and Sur-
geons, New York. Address reprint re- genital and acquired refractory anemia, including thalassemia major, Diamond–
quests to Dr. Brittenham at the Division Blackfan anemia, myelodysplastic syndromes, myelofibrosis, aplastic anemia, and
of Pediatric Hematology, Department of other disorders. In the United States, 10,000 to 20,000 patients with sickling disor-
Pediatrics, Columbia University Medical
Center, Rm. CHN 10-08, 3959 Broadway, ders receive repeated transfusions. An estimated 4000 to 5000 patients with myelo-
New York, NY 10032, or at gmb31@ dysplastic syndromes and other forms of acquired refractory anemia require red- cell transfusions. The number of patients with transfusion-dependent thalassemia
N Engl J Med 2011;364:146-56. in the United States is smaller — probably less than 1000.2 However, globally, al-
Copyright © 2011 Massachusetts Medical Society. most 100,000 patients with thalassemia syndromes undergo transfusions. The ma-
jority of these patients are in low- and middle-income countries.3
Because humans lack any effective means to excrete excess iron, long-term trans-
fusion alone inexorably produces the clinical problem of iron overload. In patients
with thalassemia who undergo transfusion from infancy, iron-induced liver dis-
ease and endocrine disorders develop during childhood and are almost inevitably
followed in adolescence by death from iron-induced cardiomyopathy.4 In patients
with sickle cell anemia, although iron-induced complications appear to develop later,
eventually, liver disease with cirrhosis as well as cardiac and pancreatic iron deposi-
tion can develop.5,6 The annual per-patient costs of care for complications of iron
overload are estimated at $15,000 to $20,000.7,8


At the end of their life span, transfused red cells are phagocytosed by reticuloendo-
thelial macrophages in the liver, bone marrow, and spleen (Fig. 1). Their hemoglobin
is digested, and the iron is freed from heme and released into the cytosol. Early in
the course of long-term transfusion, most of this additional iron can be stored
within reticuloendothelial macrophages. Gradually, limits on the capacity of macro­
phages to retain iron result in the release of excess iron into plasma.9 Transferrin

146 n engl j med 364;2  january 13, 2011

The New England Journal of Medicine

Downloaded from on October 19, 2015. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.
clinical ther apeutics

binds the released iron, with an increase in the In contrast to deferoxamine, the synthetic che-
plasma iron concentration and transferrin satu- lator deferasirox is well absorbed from the gastro-
ration. As the transferrin saturation increases, intestinal tract and is cleared from the circulation
hepatocytes are recruited to serve as storage sites slowly.13,15 Two molecules of deferasirox are need-
for the excess iron. ed to bind a single atom of iron. Like deferox-
With continued transfusion, macrophages and amine, deferasirox forms complexes with plasma
hepatocytes can no longer retain all the surplus iron, but deferasirox–iron complexes are elimi-
iron. Iron then enters plasma in amounts that nated predominantly through a hepatobiliary
exceed the transport capacity of circulating trans- route. Hepatocytes readily take up deferasirox,
ferrin. As a consequence, non–transferrin-bound which chelates hepatocellular iron. The defera-
iron appears in the plasma (Fig. 1) as a hetero- sirox–iron complexes are then excreted in the
geneous assortment of iron complexes that ap- bile.15 Within cells, deferasirox chelates cyto-
pear to be the major mediators of extrahepatic solic iron, leading to ferritin degradation by the
tissue damage in transfusional iron overload.10 proteasome.14
Non–transferrin-bound plasma iron enters spe- A third iron chelator, the synthetic oral agent
cific cells, particularly hepatocytes, cardiomyo- deferiprone (Ferriprox, Apotex; Kelfer, Cipla), is
cytes, anterior pituitary cells, and pancreatic beta- not approved for use in the United States or
cells. In these cells, iron accumulation leads to Canada. In the European Union and some other
the generation of reactive oxygen species, result- countries, it is approved specifically for patients
ing in damage to lipids, proteins, DNA, and with thalassemia major when deferoxamine is
subcellular organelles, including lysosomes and contraindicated or inadequate (Table 1).
mito­chondria. This injury may result in cellular
dysfunction, apoptosis, and necrosis. CL INIC A L E V IDENCE
Therapy with chelating agents that form a
complex with iron and promote its excretion can The use of deferoxamine therapy antedates the
clear plasma non–transferrin-bound iron, remove common use of randomized, controlled trials to
excess iron from cells, and maintain or return establish the efficacy of medical treatments. Only
body iron to safe levels (Fig. 1). (An interactive one small, randomized trial has compared chela-
An interactive
graphic depicting iron supply and storage in tion plus deferoxamine with no therapy; this trial graphic is
sickle cell anemia with long-term red-cell trans- enrolled 20 children with β-thalassemia. After a available at
fusion and iron-chelating therapy is available mean of 5.8 years of treatment with intramuscu-
with the full text of this article at
Two iron-chelating agents are approved for use Figure 1 (see next two pages). Iron Supply and Storage in Sickle Cell
in North America (Table 1): parenteral deferox- Anemia with Long-Term Red-Cell Transfusion and Iron-Chelating Therapy.
amine mesylate (Desferal, Novartis) and oral In all four panels, the area of the red and blue circles is roughly proportional
to the amount of iron in each pool, and the width of the arrows is roughly
deferasirox (Exjade, Novartis). proportional to the daily magnitude of the iron flux. Panel A shows normal
Deferoxamine is a siderophore (an iron-bind- iron supply and storage in a healthy person without sickle cell disease (i.e.,
ing compound) produced by the bacterium Strep- with hemoglobin A). The major pathway of internal iron exchange is a unidi-
tomyces pilosus. It is poorly absorbed after oral ad- rectional flow from plasma transferrin to the erythroid marrow to circulating
ministration and is rapidly cleared; consequently, red cells to reticuloendothelial macrophages and back to plasma transferrin
(orange arrows). In the circulating plasma, virtually all iron is bound to trans-
subcutaneous or intravenous administration is ferrin. Panel B shows that in sickle cell anemia, hemolysis shortens the aver-
necessary. One molecule of deferoxamine binds a age life span of the red cell from about 4 months to 5 or 6 weeks, increasing
single atom of iron, forming a feroxamine com- red-cell catabolism by reticuloendothelial macrophages and increasing iron
plex that is virtually inert metabolically. Plasma delivery to the erythroid marrow by 6 to 8 times the normal rate. There is lit-
iron chelated with deferoxamine is eliminated pre- tle ineffective erythropoiesis, and iron absorption from the gastrointestinal
tract is not increased. Panel C shows that long-term red-cell transfusion de-
dominantly by the kidneys. Hepatocytes efficiently creases erythroid marrow activity to 2 to 3 times the normal rate but results
take up deferoxamine, which then chelates hepato- in accumulation of iron in reticuloendothelial macrophages and hepatocytes.
cellular iron, with the feroxamine excreted in the Eventually, the capacity for safe storage is exceeded, with the appearance of
bile. Within cells, deferoxamine is localized to ly- plasma non–transferrin-bound iron (dashed arrow) and its progressive depo-
sosomes, where it induces autophagy of cytosolic sition in the heart and endocrine organs. In Panel D, the green arrows show
that iron-chelating therapy with deferasirox can clear plasma non–transferrin-
ferritin. Lysosomal degradation of cytosolic ferritin bound iron and remove excess iron from the liver, heart, and other organs,
releases iron that is bound by deferoxamine, and with subsequent excretion through the bile into the stool.
the chelated iron is then cleared from the cell.14

n engl j med 364;2  january 13, 2011 147

The New England Journal of Medicine
Downloaded from on October 19, 2015. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e

A Normal Iron Supply and Storage

Functional iron
Storage iron Heart and Muscle and other
endocrine organs parenchymal cells

bound iron
red cells tract


B Sickle Cell Anemia

Heart and Muscle and other
endocrine organs parenchymal cells

bound iron
red cells
tract C

Author D
Fig # 1
Erythroid DE Ja
marrow Artist D
Figure has been r

Issue date 01

148 n engl j med 364;2 january 13, 2011

The New England Journal of Medicine

Downloaded from on October 19, 2015. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.
clinical ther apeutics

C Sickle Cell Anemia with Long-Term Non–transferrin-bound iron

Red-Cell Transfusion
Heart and Muscle and other
Transfusion parenchymal cells
endocrine organs

bound iron
Circulating Gastrointestinal
red cells tract


D Sickle Cell Anemia with Transfusion Chelator-bound iron

and Iron-Chelating Therapy
macrophages Muscle and other
Heart and parenchymal cells
endocrine organs

bound iron
Circulating Gastrointestinal
red cells tract COLOR FIGURE

Rev8 1
Author Dr. Bittenha
Fig # 1C-D
Erythroid DE Jarcho
marrow Artist Daniel Mulle
Figure has been redrawn and type
Please check carefull

Issue date 01/13/2011

n engl j med 364;2 january 13, 2011 149

The New England Journal of Medicine
Downloaded from on October 19, 2015. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e

lar deferoxamine, the mean hepatic iron concen-

plasma zinc level, progression of hepatic fibrosis asso-

tration was 25.9 mg per gram of liver tissue (dry

Agranulocytosis and neutropenia; gastrointestinal distur-

major, when deferoxamine therapy is contraindicated
Transfusional iron overload in patients with thalassemia

bances, arthropathy, increased liver-enzyme levels, low

ciated with increase in iron overload or hepatitis C11

Not approved in United States or Canada; approved in
weight) in the deferoxamine group and 42.2 mg
per gram in the control group.16 At 14 years, one
death had occurred in the deferoxamine group and
six deaths had occurred in the control group.17
In lieu of randomized trials, observational
studies have investigated the effects of deferox-

Europe11 and other countries

amine in the management of transfusion-related
iron overload. One study involved 977 children
with transfusion-dependent thalassemia major
Bidentate, 3:1 complex

Predominantly urinary
Oral, three times daily

who survived beyond the first decade of life.18

75–100 mg/kg/day

or inadequate
Subsequent survival was examined according to
5-year birth cohorts beginning in 1960; deferox-
amine was introduced in 1975. The survival rate
2–3 hr

increased progressively in each 5-year cohort

(see Fig. 1 in the Supplementary Appendix, avail-
serum creatinine level; potentially fatal renal able at The survival rate was signifi-
Gastrointestinal disturbances, rash, increase in

cantly higher among children born after 1975

and hepatic impairment or failure, gastro­
Approved in United States, Canada, Europe,

than among those in previous cohorts.

Deferasirox has been compared with deferox-
amine in a few short-term trials sponsored by
Novartis.19-23 In the largest of these trials, 586

children with β-thalassemia were randomly as-

intestinal hemorrhage13
Transfusional iron overload

signed to either agent, with dosing according to

Tridentate, 2:1 complex

and other countries

Predominantly biliary

the baseline hepatic iron concentration.19 The pri-

20–40 mg/kg/day

mary end point was the percentage of subjects

Oral, once daily

with either a maintained or reduced hepatic iron

concentration at 1 year; this end point was
8–16 hr

reached in 52.9% of patients assigned to de-

ferasirox and in 66.4% of patients assigned to
deferoxamine. This result, which did not meet a
changes, allergy, respiratory distress syndrome
disturbances, growth retardation and skeletal
Irritation at the infusion site, ocular and auditory

prespecified noninferiority target, was attributed

to the relative underdosing of deferasirox. The
Approved in United States, Canada, Europe,

with higher-than-recommended doses12

Subcutaneous or intravenous, 8–10 hr/day,

total hepatic iron concentration decreased by a

mean of 2.4 mg per gram (dry weight) in the
deferasirox group and by 2.9 mg per gram in the

deferoxamine group. No trial has established the

Transfusional iron overload

long-term effectiveness of deferasirox in prevent-

Hexadentate, 1:1 complex

and other countries

ing organ toxicity or improving survival.

Table 1. Iron-Chelating Agents in Clinical Use.

Deferiprone has also been compared with de­

Biliary and urinary
25–50 mg/kg/day

5–7 days/wk

feroxamine in several small, randomized trials.24

As is the case with deferasirox, no long-term
20–30 min

trials have been performed to evaluate the effect

of deferiprone on organ function or survival.
Chelator-iron complex

Route of elimination
Regulatory approval

Iron-chelating therapy should be considered in

Plasma half-life

Adverse effects

all patients who require long-term red-cell trans-

Usual dose


fusion. Such patients include those with sickle

cell disease, myelodysplastic syndromes, thalas-
semia major, Diamond–Blackfan anemia, aplastic

150 n engl j med 364;2  january 13, 2011

The New England Journal of Medicine

Downloaded from on October 19, 2015. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.
clinical ther apeutics

anemia, and other congenital and acquired forms In the United States and Canada, the choice
of refractory anemia. of an iron chelator for transfusional iron over-
There are alternatives to chelation in some load is either parenteral deferoxamine or oral
patients. Some of the underlying disorders re- deferasirox. The decision is best made with the
quiring transfusion may be cured by hematopoi- patient and, if the patient is a child, with his or
etic stem-cell transplantation. In some patients her parents. Despite the lack of data on long-
with sickle cell disease, exchange transfusion may term effectiveness, most patients now opt for
reduce or obviate the need for iron chelation. In- deferasirox because of the ease of oral adminis-
frequently, phlebotomy may be an option for the tration. Deferasirox is preferred for prophylactic
removal of excess iron in the event of cure or re- or maintenance therapy. Deferoxamine, which
mission of a refractory anemia. Iron-chelating has been proved to reverse iron-induced heart
therapy itself may sometimes decrease or elimi- disease and increase long-term survival,28 may
nate the need for transfusion in patients with my- be indicated if deferasirox is ineffective in a par-
elodysplasia25 or myelofibrosis. Chelation therapy ticular patient, and it may be favored for severe
may not be needed in patients with myelodyspla- iron overload, especially with cardiac involvement.
sia or other acquired refractory anemias who have Conversely, deferasirox may be the better choice
an estimated survival of less than 1 year.26 in patients who are unable to tolerate subcutane-
With the exception of these groups, iron- ous infusions of deferoxamine. Deferasirox also
chelating therapy is indicated in almost all pa- may be substituted for deferoxamine after suc-
tients requiring long-term red-cell transfusion. cessful clearance of cardiac iron. Deferiprone is
Iron chelation is contraindicated in patients who available in the United States on a compassion-
are hypersensitive to the chelating agent or ex- ate-use basis, usually in combination with defer-
cipients in the chelator formulation, and it re- oxamine, in patients in whom iron-induced
quires specialized management in patients with heart failure has developed or who are at high
several renal disease or anuria. Chelators should risk for the development of heart failure.29,30
be avoided or used with great caution in patients Deferoxamine is administered subcutaneously
who are pregnant or breast-feeding. or intravenously, usually with a portable pump, for
Ideally, iron-chelating therapy should be initi- 8 to 10 hours each day, 5 to 7 days per week.
ated prophylactically, before clinically significant Subcutaneous administration is preferred except
iron accumulation has occurred. Treatment should in patients with severe cardiac iron deposition, for
begin when patients have received between 10 and whom continuous intravenous deferoxamine ther-
20 red-cell transfusions. Patients who have already apy is recommended.28,31 Deferasirox is adminis-
undergone repeated transfusion without sufficient tered orally once daily, and deferiprone is admin-
chelation can also be successfully treated, but they istered orally three times daily.
may require more intensive regimens (see below). The dose of an iron-chelating agent is deter-
Evaluation of the patient before the initiation mined by three principal factors: the presence or
or adjustment of iron-chelating therapy includes a absence of cardiac iron overload, the rate of
detailed characterization of the underlying dis- transfusional iron loading, and the body iron
order, with thorough documentation of the his- burden (see the Supplementary Appendix and
tory of transfusion and chelation; determination Table 2). In brief, if cardiac iron overload is pres-
of the body iron load by measurement of the ent, ridding the heart of the excess iron becomes
hepatic iron and serum ferritin concentrations; the critical therapeutic goal. In the absence of
estimation of the rate of transfusional iron load- cardiac iron overload, the long-term objective is
ing; and assessment of cardiac iron deposition. to maintain the body iron at a level that permits
The techniques for assessing cardiac iron over- safe storage while avoiding chelator toxicity. The
load, transfusional iron loading, and body iron greater the rate of transfusional iron loading,
burden are described in the Supplementary Ap- the greater the dose of an iron chelator that will
pendix. The extent of any existing iron-induced be needed to control the accumulation of iron.
hepatic, cardiac, or endocrine dysfunction should During treatment, tests to monitor chelator-
be established, and in children and adolescents, associated toxicity should be performed, depend-
growth and maturation should be assessed. Nu- ing on the potential adverse effects of the specific
tritional evaluation with correction of deficien- agent to be used (see Table 1 and the Adverse
cies is recommended.27 Effects section below). In patients who receive

n engl j med 364;2  january 13, 2011 151

The New England Journal of Medicine
Downloaded from on October 19, 2015. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.
Table 2. Usual Doses of Deferoxamine or Deferasirox for Transfusional Iron Overload.*

Hepatic Iron
Concentration No Cardiac Iron Overload (T2*, ≥20 msec) Cardiac Iron Overload (T2*, <20 msec)

Daily Transfusional Iron Intake Mild to Moderate Severe

<0.3 mg/kg of body weight 0.3 to 0.5 mg/kg of body weight >0.5 mg/kg of body weight T2*, 10 to <20 msec T2*, <10 msec

≥15 mg/g, dry weight Deferoxamine: 40–50 mg/kg/ Deferoxamine: 40–50 mg/kg/ Deferoxamine: 40–50 mg/kg/ Deferoxamine: 50 mg/kg/day Deferoxamine: 50 mg/kg/day
day, 8 to 10 hr/day, 6 or 7 day, 8 to 10 hr/day, 6 or 7 day, 8 to 10 hr/day, 6 or 7 by continuous intravenous by continuous intravenous
days/wk, by subcutaneous days/wk, by subcutaneous days/wk, by subcutaneous infusion28; deferasirox: oral infusion28; deferasirox: oral
infusion; deferasirox: oral infusion; deferasirox: oral infusion; deferasirox: oral dose of 40 mg/kg daily, but dose of 40 mg/kg daily, but
dose of 30–40 mg/kg daily dose of 30–40 mg/kg daily dose of 30–40 mg/kg daily uncertain efficacy in reduc- uncertain efficacy in reduc-
ing cardiac iron ing cardiac iron
7 to <15 mg/g, dry Deferoxamine: 30–40 mg/kg/ Deferoxamine: 40–50 mg/kg/ Deferoxamine: 40–50 mg/kg/ Deferoxamine: 40–50 mg/kg/ Deferoxamine: 40–50 mg/kg/
weight day, 8 to 10 hr/day, 5 days/ day, 8 to 10 hr/day, 6 or 7 day, 8 to 10 hr/day, 6 or 7 day, 8 to 10 hr/day, 6 or 7 day by continuous infu-
wk, by subcutaneous infu- days/wk, by subcutaneous days/wk, by subcutaneous days/wk, by subcutaneous sion28; deferasirox: oral
sion; deferasirox: oral dose infusion; deferasirox: oral infusion; deferasirox: oral infusion; deferasirox: oral dose of 30–40 mg/kg daily
of 20–30 mg/kg daily dose of 30–40 mg/kg daily dose of 30–40 mg/kg daily dose of 30–40 mg/kg daily
3 to <7 mg/g, dry Deferoxamine: 30–40 mg/kg/ Deferoxamine: 30–40 mg/kg/ Deferoxamine: 30–40 mg/kg/ Deferoxamine: 40–50 mg/kg/ Deferoxamine: 40–50 mg/kg/
weight day, 8 to 10 hr/day, 5 days/ day, 8 to 10 hr/day, 5 days/ day, 8 to 10 hr/day, 5 days/ day, 8 to 10 hr/day, 6 or 7 day by continuous infu-
n e w e ng l a n d j o u r na l

The New England Journal of Medicine

wk, by subcutaneous infu- wk, by subcutaneous infu- wk, by subcutaneous infu- days/wk, by subcutaneous sion28; deferasirox: oral

sion; deferasirox: oral dose sion; deferasirox: oral dose sion; deferasirox: oral dose infusion; deferasirox: oral dose of 30–40 mg/kg daily
of 20–30 mg/kg daily of 20–30 mg/kg daily of 20–30 mg/kg daily dose of 30–40 mg/kg daily
<3 mg/g, dry weight Deferoxamine: suspend thera- Deferoxamine: suspend thera- Deferoxamine: suspend thera- Deferoxamine: adjust intrave- Deferoxamine: adjust intrave-

n engl j med 364;2  january 13, 2011

py; deferasirox: suspend py; deferasirox: suspend py; deferasirox: suspend nous or subcutaneous dose nous or subcutaneous dose
therapy therapy therapy according to therapeutic according to therapeutic

Copyright © 2011 Massachusetts Medical Society. All rights reserved.

m e dic i n e

index32; deferasirox: adjust ­index32; deferasirox: adjust

oral dose, monitoring renal oral dose, monitoring renal
and hepatic function and and hepatic function and
blood count blood count

* To minimize interference with growth and skeletal development, the dose of deferoxamine in young children should not exceed 25 to 30 mg per kilogram of body weight. The dose

Downloaded from on October 19, 2015. For personal use only. No other uses without permission.
should be adjusted according to the therapeutic index.32 The bioavailability of deferasirox may affect the response. T2* denotes the cardiac effective transverse relaxation time on mag-
netic resonance imaging.
clinical ther apeutics

deferoxamine, these tests include annual assess- the dose as the hepatic iron concentration ap-
ments of auditory function and vision. In patients proaches normal levels. Although treatment with
who receive deferasirox, serum creatinine, serum deferoxamine may reduce endocrine complica-
aminotransferases, and bilirubin levels and com- tions of iron overload, such as a delay of puberty,
plete blood counts should be assessed monthly. the chelator itself can interfere with growth,35
In patients who receive deferiprone, weekly as- apparently as a result of skeletal dysplasia.36 To
sessment of complete blood counts and monthly minimize this effect, the dose of deferoxamine in
assessments of serum aminotransferases should young children should not exceed 25 to 30 mg per
be performed. kilogram.37
The effectiveness of iron-chelating therapy is In the registration trial of deferasirox de-
best monitored by periodic measurements of car- scribed above,19 gastrointestinal disturbances
diac iron concentrations by magnetic resonance occurred in approximately 15% of patients, rash
imaging (MRI), of cardiac function, and of he- in 11%, and increases in serum creatinine levels
patic iron concentrations and by review of the in 38%. Similar rates have been observed in
actual rate of transfusional iron loading once subsequent trials.20-22 In January 2010, on the
or twice a year, depending on the severity of the basis of postmarketing studies, the Food and
iron overload (see the Supplementary Appendix). Drug Administration required a change in the
Dose adjustments can be made according to the prescribing information for deferasirox. The new
guidelines in Table 2. Serum ferritin concentra- information states that the drug could cause
tions are usually measured at least quarterly. potentially fatal renal and hepatic impairment or
As at baseline, hepatic, cardiac, and endocrine failure as well as gastrointestinal hemorrhage.13
function should be assessed periodically along These adverse effects were reported to occur
with nutritional status, and, in children and more frequently in older patients and in patients
adolescents, growth and maturation should be with high-risk myelodysplastic syndromes, throm-
monitored. bocytopenia, or underlying renal or hepatic im-
In the United States, on the basis of 2006 pairment.
wholesale acquisition prices, the annual costs of The most common adverse effects of deferi-
deferoxamine for iron-chelating therapy have been prone are diarrhea and gastrointestinal effects,
estimated to range from $6,824 to $29,209, plus arthropathy (including severe arthritis with clin-
$9,286 for infusion, and the estimated annual ically significant disability), increased levels of
costs of deferasirox range from $24,404 to $53,095, serum liver enzymes, and progression of hepatic
with the actual cost depending on dose and body fibrosis associated with an increase in iron over-
weight.7 Administration of a chelator will be load or hepatitis C.11 The most serious adverse
needed as long as transfusion is continued and effects are agranulocytosis (incidence, 1.1%) and
will be lifelong in most patients. neutropenia (incidence, 4.9%); weekly monitor-
ing of the neutrophil count is recommended.
A DV ER SE EFFEC T S Neurologic abnormalities have been reported
with higher-than-recommended doses of deferi-
Discomfort or pain at the site of subcutaneous prone.38
infusion develops in almost all patients treated
with subcutaneous deferoxamine, and induration A r e a s of Uncer ta in t y
or erythema develops in some patients. These
symptoms can often be mitigated with topical Several areas of uncertainty exist with regard to
anesthetic or glucocorticoid creams. Visual and the optimal approach to iron-chelating therapy.
auditory toxicity associated with deferoxamine First, a variety of binary combinations of chelat-
has been reported, and in one series involving 89 ing agents are being examined in off-label uses,
patients treated with this agent, 13 presented with with either synchronous or sequential adminis-
vision loss, deafness, or both.33 Subsequent stud- tration. The clinical usefulness of such combina-
ies suggest a much lower incidence of toxicity,34 tion therapies is unclear at present, in the ab-
and these risks can be minimized by not exceed- sence of unequivocal evidence of the superiority
ing doses of 50 mg per kilogram of body weight of any specific combination over treatment with
in patients with iron overload and by decreasing a single agent.24

n engl j med 364;2  january 13, 2011 153

The New England Journal of Medicine
Downloaded from on October 19, 2015. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.
The n e w e ng l a n d j o u r na l of m e dic i n e

Second, MRI performed to evaluate the iron Guidel ine s

content of the liver, heart, and other organs has
become the method of choice for guiding iron- Guidelines and consensus statements on the man-
chelating therapy, but calibrated methods are not agement of sickle cell disease,45,46 thalasse-
available for all patients. A 1-year study of de- mia,32,47 Diamond–Blackfan anemia,48 aplastic
ferasirox involving patients with various types of anemia,49 and myelodysplastic syndromes26,50 all
anemias used an alternative approach; the inves- include recommendations for iron-chelating
tigators based the initial dose on the rate of therapy for transfusional iron overload. All these
transfusional iron loading and subsequently ad- guidelines are generally consistent with the ap-
justed the dose according to measurements of proach outlined in this review, although there
serum ferritin levels and safety markers.23 The are variations in the individual recommenda-
long-term efficacy and safety of this strategy are tions, depending in part on the year of publica-
uncertain. tion and the specific underlying disorder. For
Third, in patients with myelodysplastic syn- example, the guidelines for iron chelation in
dromes who have undergone long-term transfu- thalassemia32,47 generally endorse measurement
sion and for whom prolonged survival is antici- of serum ferritin levels as a useful way to moni-
pated, iron-chelating therapy may be appropriate. tor iron overload, whereas the guidelines for the
In individual patients, the benefit of such treat- management of sickle cell disease45,46 emphasize
ment may vary, given the morbidity associated that ferritin levels can be altered by liver disease
with chelation, the variable prognosis for the and inflammation. The guidelines on the man-
underlying disorder, and the latency period be- agement of thalassemia by the Italian Society of
tween the onset of the transfusion and the de- Hematology47 endorse deferoxamine as first-line
velopment of clinical manifestations of iron therapy over the oral chelators, whereas most of
overload. Data are lacking from prospective, the other guidelines do not state an explicit pref-
randomized trials examining the clinical cir- erence in this regard.
cumstances in which morbidity and mortality
improve with iron-chelating therapy in these R ec om mendat ions
patients39; one such trial is currently recruiting
patients.40 After discussing the need for iron-chelating ther-
Fourth, there is evidence to suggest that iron apy with the patient and his family, I would de-
overload is associated with lower rates of cardio- scribe the advantages and disadvantages of sub-
myopathy, endocrinopathy, and other conditions cutaneous deferoxamine and oral deferasirox so
among patients with sickle cell disease than that a genuinely informed decision can be made.
among patients with thalassemia.41 The effects At present, the most frequent choice is oral de-
of the systemic inflammatory state on iron han- ferasirox. Before the initiation of treatment, I
dling in patients with sickle cell disease,42 as would obtain information about the number of
well as differences in the rate and duration of previous transfusions and the rate of ongoing
transfusion and in the age of the patient at the transfusion and would arrange for cardiac T2*
initiation of long-term transfusion, the extent of and hepatic transverse relaxation rate (R2) mea-
ineffective erythropoiesis, and gastrointestinal surements, an MRI evaluation of cardiac function,
iron absorption, may be involved. It is unclear and echocardiographic and electrocardiographic
whether these data provide sufficient grounds to studies. Auditory and ophthalmic testing, includ-
recommend a higher body iron threshold for ing slit-lamp examination and dilated-fundus ex-
chelation therapy in patients with sickle cell amination, should be performed. Laboratory tests
disease, especially given the potential risks of should include a complete blood count with a
iron-induced liver disease. differential count and measurement of serum
Finally, there is a lack of certainty with respect creatinine, serum aminotransferases, bilirubin
to the optimal hepatic iron concentrations (Table levels, and iron indexes. After transfusion of a
2) for minimizing the risk that hepatic fibrosis total of 10 to 20 units of blood, or with the he-
will progress to cirrhosis and its ultimate compli- patic iron concentration between 3 and 7 mg per
cation, hepatocellular carcinoma.43,44 gram, I would administer once-daily oral therapy

154 n engl j med 364;2  january 13, 2011

The New England Journal of Medicine

Downloaded from on October 19, 2015. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.
clinical ther apeutics

with deferasirox at a dose of 20 mg per kilogram. and Drug Administration (R01 FD003702), and the National In-
stitutes of Health (R37 DK049108; R01 DK066251).
With good support and careful monitoring, this Dr. Brittenham reports that his institution filed a patent
regimen, adjusted as needed, should provide ­application in April 2010 for a rapid MRI method. No other
long-term protection against the complications potential conflict of interest relevant to this article was re-
of transfusional iron overload in this patient. Disclosure forms provided by the author are available with the
Supported by grants from the St. Giles Foundation, the Food full text of this article at

1. Adams RJ, McKie VC, Hsu L, et al. ferasirox) prescribing information. 2010. Williams S, Howard J, Hyde CJ. Oral de-
Prevention of a first stroke by transfu- ( feriprone for iron chelation in people with
sions in children with sickle cell anemia 14. De Domenico I, Ward DM, Kaplan J. thalassaemia. Cochrane Database Syst Rev
and abnormal results on transcranial Specific iron chelators determine the 2007;3:CD004839.
Doppler ultrasonography. N Engl J Med route of ferritin degradation. Blood 2009; 25. Jensen PD, Heickendorff L, Pedersen
1998;339:5-11. 114:4546-51. B, et al. The effect of iron chelation on
2. Vichinsky EP, MacKlin EA, Waye JS, 15. Waldmeier F, Bruin GJ, Glaenzel U, et haemopoiesis in MDS patients with trans-
Lorey F, Olivieri NF. Changes in the epide- al. Pharmacokinetics, metabolism, and fusional iron overload. Br J Haematol
miology of thalassemia in North Ameri- disposition of deferasirox in beta-thalas- 1996;94:288-99.
ca: a new minority disease. Pediatrics semic patients with transfusion-depen- 26. Gattermann N. Overview of guide-
2005;116(6):e818-e825. dent iron overload who are at pharmaco- lines on iron chelation therapy in patients
3. Weatherall DJ. The inherited diseases kinetic steady state. Drug Metab Dispos with myelodysplastic syndromes and
of hemoglobin are an emerging global 2010;38:808-16. transfusional iron overload. Int J Hema-
health burden. Blood 2010;115:4331-6. 16. Barry M, Flynn DM, Letsky EA, Ris- tol 2008;88:24-9.
4. Weatherall DJ, Clegg JB. The thalas- don RA. Long-term chelation therapy in 27. Claster S, Wood JC, Noetzli L, et al.
semia syndromes. 4th ed. Oxford, Eng- thalassaemia major: effect on liver iron Nutritional deficiencies in iron overload-
land: Blackwell Science, 2001. concentration, liver histology, and clini- ed patients with hemoglobinopathies. Am
5. Darbari DS, Kple-Faget P, Kwagyan J, cal progress. BMJ 1974;2:16-20. J Hematol 2009;84:344-8.
Rana S, Gordeuk VR, Castro O. Circum- 17. Modell B, Letsky EA, Flynn DM, Peto 28. Davis BA, Porter JB. Long-term out-
stances of death in adult sickle cell disease R, Weatherall DJ. Survival and desferriox- come of continuous 24-hour deferoxamine
patients. Am J Hematol 2006;81:858-63. amine in thalassaemia major. Br Med J infusion via indwelling intravenous cathe-
6. Finch CA, Lee MY, Leonard JM. Con- (Clin Res Ed) 1982;284:1081-4. ters in high-risk beta-thalassemia. Blood
tinuous RBC transfusions in a patient 18. Borgna-Pignatti C, Rugolotto S, De 2000;95:1229-36.
with sickle cell disease. Arch Intern Med Stefano P, et al. Survival and complica- 29. Cohen AR. Compassionate use of de-
1982;142:279-82. tions in patients with thalassemia major feriprone for patients with thalassemia and
7. Delea TE, Sofrygin O, Thomas SK, treated with transfusion and deferox- iron-induced heart disease. (ClinicalTrials
Baladi JF, Phatak PD, Coates TD. Cost ef- amine. Haematologica 2004;89:1187-93. .gov identifier no. NCT00293098.) (http://
fectiveness of once-daily oral chelation 19. Cappellini MD, Cohen A, Piga A, et al.
therapy with deferasirox versus infusional A phase 3 study of deferasirox (ICL670), a NCT00293098?term=deferiprone+
deferoxamine in transfusion-dependent once-daily oral iron chelator, in patients thalassemia&rank=2.)
thalassaemia patients: US healthcare sys- with beta-thalassemia. Blood 2006;107: 30. Borgna-Pignatti C, Cappellini MD, De
tem perspective. Pharmacoeconomics 3455-62. Stefano P, et al. Cardiac morbidity and
2007;25:329-42. 20. Piga A, Galanello R, Forni GL, et al. mortality in deferoxamine- or deferiprone-
8. Delea TE, Edelsberg J, Sofrygin O, et Randomized phase II trial of deferasirox treated patients with thalassemia major.
al. Consequences and costs of noncompli- (Exjade, ICL670), a once-daily, orally- Blood 2006;107:3733-7.
ance with iron chelation therapy in pa- administered iron chelator, in comparison 31. Anderson LJ, Westwood MA, Holden
tients with transfusion-dependent thalas- to deferoxamine in thalassemia patients S, et al. Myocardial iron clearance during
semia: a literature review. Transfusion with transfusional iron overload. Haema- reversal of siderotic cardiomyopathy with
2007;47:1919-29. tologica 2006;91:873-80. intravenous desferrioxamine: a prospec-
9. Fillet G, Beguin Y, Baldelli L. Model of 21. Porter J, Galanello R, Saglio G, et al. tive study using T2* cardiovascular mag-
reticuloendothelial iron metabolism in Relative response of patients with myelo- netic resonance. Br J Haematol 2004;127:
humans: abnormal behavior in idiopathic dysplastic syndromes and other transfu- 348-55.
hemochromatosis and in inflammation. sion-dependent anaemias to deferasirox 32. Iron overload. In: Cappellini MD, Co-
Blood 1989;74:844-51. (ICL670): a 1-yr prospective study. Eur J hen A, Eleftheriou A, Piga A, Porter J,
10. Breuer W, Hershko C, Cabantchik ZI. Haematol 2008;80:168-76. Taher A. Guidelines for the clinical man-
The importance of non-transferrin bound 22. Vichinsky E, Onyekwere O, Porter J, et agement of thalassemia. 2nd ed. rev.
iron in disorders of iron metabolism. al. A randomised comparison of defera- Nicosia, Cyprus: Thalassaemia Interna-
Transfus Sci 2000;23:185-92. sirox versus deferoxamine for the treat- tional Federation, November 2008:33-63.
11. European Medicines Agency. Ferriprox ment of transfusional iron overload in 33. Olivieri NF, Buncic JR, Chew E, et al.
(deferiprone): summary of product charac- sickle cell disease. Br J Haematol 2007; Visual and auditory neurotoxicity in pa-
teristics. European public assessment 136:501-8. tients receiving subcutaneous deferox-
report 2009, rev. ed. 14. (http://www.ema 23. Cappellini MD, Porter J, El-Beshlawy amine infusions. N Engl J Med 1986;314: A, et al. Tailoring iron chelation by iron 869-73.
ferriprox/ferriprox.htm.) intake and serum ferritin: the prospective 34. Cohen A, Martin M, Mizanin J, Kon-
12. Novartis Pharmaceuticals. Desferal EPIC study of deferasirox in 1744 patients kle DF, Schwartz E. Vision and hearing
(deferoxamine mesylate) prescribing infor- with transfusion-dependent anemias. Hae- during deferoxamine therapy. J Pediatr
mation. 2010. ( matologica 2010;95:557-66. 1990;117:326-30.
13. Novartis Pharmaceuticals. Exjade (de- 24. Roberts DJ, Brunskill SJ, Doree C, 35. De Sanctis V, Roos M, Gasser T, et al.

n engl j med 364;2  january 13, 2011 155

The New England Journal of Medicine
Downloaded from on October 19, 2015. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.
clinical ther apeutics

Impact of long-term iron chelation thera- (TELESTO). ( identifier no. Heart, Lung, and Blood Institute, 2002.
py on growth and endocrine functions in NCT00940602.) (http://www.clinicaltrials (
thalassemia. J Pediatr Endocrinol Metab .gov/ct2/show/NCT00940602?term=telesto blood/sickle/sc_mngt.pdf.)
2006;19:471-80. &rank=1.) 46. Standards for the clinical care of adults
36. Chan YL, Pang LM, Chik KW, et al. 41. Vichinsky E, Butensky E, Fung E, et al. with sickle cell disease in the UK. London:
Patterns of bone diseases in transfusion- Comparison of organ dysfunction in Sickle Cell Society, 2008. (http://www
dependent homozygous thalassemia ma- transfused patients with SCD or beta
jor: predominance of osteoporosis and thalassemia. Am J Hematol 2005;80:70-4. 47. Angelucci E, Barosi G, Camaschella
desferrioxamine-induced bone dysplasia. 42. Walter PB, Fung EB, Killilea DW, et al. C, et al. Italian Society of Hematology
Pediatr Radiol 2002;32:492-7. Oxidative stress and inflammation in practice guidelines for the management
37. Olivieri NF, Brittenham GM. Iron- iron-overloaded patients with beta-thal- of iron overload in thalassemia major and
chelating therapy and the treatment of assaemia or sickle cell disease. Br J Hae- related disorders. Haematologica 2008;
thalassemia. Blood 1997;89:739-61. [Er- matol 2006;135:254-63. 93:741-52.
ratum, Blood 1997;89:2621.] 43. Borgna-Pignatti C, Vergine G, Lom- 48. Vlachos A, Ball S, Dahl N, et al. Diag-
38. Beau-Salinas F, Guitteny MA, Dona- bardo T, et al. Hepatocellular carcinoma nosing and treating Diamond Blackfan
dieu J, Jonville-Bera AP, Autret-Leca E. in the thalassaemia syndromes. Br J Hae- anaemia: results of an international clini-
High doses of deferiprone may be associ- matol 2004;124:114-7. cal consensus conference. Br J Haematol
ated with cerebellar syndrome. BMJ 2009; 44. Ko C, Siddaiah N, Berger J, et al. Prev- 2008;142:859-76.
338:a2319. alence of hepatic iron overload and asso- 49. Marsh JC, Ball SE, Cavenagh J, et al.
39. Leitch HA. Controversies surround- ciation with hepatocellular cancer in end- Guidelines for the diagnosis and manage-
ing iron chelation therapy for MDS. Blood stage liver disease: results from the ment of aplastic anaemia. Br J Haematol
Rev 2010 October 26 (Epub ahead of National Hemochromatosis Transplant 2009;147:43-70.
print). Registry. Liver Int 2007;27:1394-401. 50. Bennett JM. Consensus statement on
40. Novartis Pharmaceuticals. Myelodys- 45. Division of Blood Diseases and Re- iron overload in myelodysplastic syn-
plastic syndromes (MDS) event free sur- sources NHLBI. The management of sick- dromes. Am J Hematol 2008;83:858-61.
vival with iron chelation therapy study le cell disease. Bethesda, MD: National Copyright © 2011 Massachusetts Medical Society.

specialties and topics at

Specialty pages at the Journal’s Web site ( feature articles in cardiology,
endocrinology, genetics, infectious disease, nephrology, pediatrics, and many other
medical specialties. These pages, along with collections of articles on clinical and
nonclinical topics, offer links to interactive and multimedia content and feature
recently published articles as well as material from the NEJM archive (1812–1989).

156 n engl j med 364;2  january 13, 2011

The New England Journal of Medicine

Downloaded from on October 19, 2015. For personal use only. No other uses without permission.
Copyright © 2011 Massachusetts Medical Society. All rights reserved.