MEDICAL PROGRESS

Volume 341 Number 2

·

99

Review Article

Medical Progress

T

HE

b

-T

HALASSEMIAS

N

ANCY

F. O

LIVIERI

, M.D.

From the University of Toronto, Toronto. Address reprint requests toDr. Olivieri at the Hospital for Sick Children, 555 University Ave., Toronto,ON M5G 1X8, Canada, or at noliv@sickkids.on.ca.

©1999, Massachusetts Medical Society.

N 1925, Thomas Cooley and Pearl Lee describeda form of severe anemia, occurring in childrenof Italian origin and associated with splenomeg-

aly and characteristic bone changes.

1

Over the nextdecade, a milder form was described independently byseveral Italian investigators.

2-4

Because all early caseswere reported in children of Mediterranean origin,the disease was later termed thalassemia, from theGreek word for sea,

thalassa

.

5

Over the next 20 years,it became apparent that Cooley and Lee had de-scribed the homozygous or compound heterozygousstate for a recessive mendelian disorder not confinedto the Mediterranean, but occurring widely through-out tropical countries. In the past 20 years, the twoimportant forms of this disorder,

a

- and

b

-thalas-semia, resulting from the defective synthesis of the

a

- and

b

-globin chains of hemoglobin, respectively,have become recognized as the most common mon-ogenic diseases in humans.

6

This article focuses on the

b

-thalassemias, the se-vere forms of which are by far the most importantof all the thalassemias. The molecular and clinical as-pects of the severe

a

-thalassemia syndromes havebeen reviewed elsewhere.

7,8

DISTRIBUTION AND POPULATION

AT RISK

The

b

-thalassemias are widespread throughout theMediterranean region, Africa, the Middle East, theIndian subcontinent and Burma, Southeast Asia in-cluding southern China, the Malay Peninsula, andIndonesia. Estimates of gene frequencies range from3 to 10 percent in some areas.

9

Within each popula-tion at risk for

b

-thalassemia a small number ofcommon mutations are found, as well as rarer ones;each mutation is in strong linkage disequilibriumwith specific arrangements of restriction-fragment–

I

length polymorphisms, or haplotypes, within the

b

-globin cluster. A limited number of haplotypes arefound in each population, so that 80 percent of themutations are associated with only 20 different hap-lotypes. This observation has helped demonstrate theindependent origin of

b

-thalassemia in several pop-ulations.

10

There is evidence that the high frequencyof

b

-thalassemia throughout the tropics reflects anadvantage of heterozygotes against

Plasmodium fal-ciparum

malaria,

11

as has already been demonstratedin

a

-thalassemia.

12

MOLECULAR PATHOLOGY

Structure and Synthesis of Hemoglobin

The structure and regulation of the human globingenes have been reviewed elsewhere

13

; only aspectswith direct relevance to an understanding of themolecular pathology of the

b

-thalassemias are out-lined here.

The

b

-like globin genes, a linked cluster on chro-mosome 11, are arranged over approximately 60,000nucleotide bases (Fig. 1). Promoter elements up-stream from the initiation codon of each active geneare involved in the initiation of transcription. Thecluster also contains other regulatory elements that in-teract to promote erythroid-specific gene expressionand to coordinate the developmental regulation ofeach gene.

Hemoglobin Switching

As an adaptation to changing oxygen requirements,different hemoglobins, all composed of two differ-ent pairs of globin chains each attached to a hememoiety, are synthesized in the embryo, fetus, andadult.

14

Severe

b

-thalassemia usually becomes manifestas a result of the decline in the synthesis of fetalhemoglobin (

a

2

g

2

) during the first year of life (Fig.1). The precise mechanisms that control the switchfrom the production of fetal hemoglobin to that ofadult hemoglobin (

a

2

b

2

) (Fig. 1) are not fully un-derstood.

13-16

Mutations Causing

b

-Thalassemia

Nearly 200 different mutations have been describedin patients with

b

-thalassemia and related disorders.Although most are small nucleotide substitutionswithin the cluster, deletions may also cause

b

-thalas-semia.

9

All the mutations result in either the absenceof the synthesis of

b

-globin chains (

b

0

-thalassemia)or a reduction in synthesis (

b

+

-thalassemia) (Fig. 2).Mutations in or close to the conserved promoter

sequences and in the 5' untranslated region down-

100

·

July 8, 1999

The New England Journal of Medicine

regulate transcription, usually resulting in mild

b

+

-thalassemia. Transcription is also affected by dele-tions in the 5' region, which completely inactivatetranscription and result in

b

0

-thalassemia.Both splicing of the messenger RNA (mRNA) pre-

cursor and ineffective cleavage of the mRNA tran-

script result in

b

-thalassemia. In some mutations, nonormal message is produced, whereas other muta-tions only slightly reduce the amount of normallyspliced mRNA. Mutations within invariant dinucle-otides at intron–exon junctions, critical to the remov-al of intervening sequences and the splicing of exons

Figure 1.

The

b

-Globin Gene Cluster on the Short Arm of Chromosome 11.In Panel A, the

b

-globin–like genes are arranged in the order in which they are expressed during development. The

G

g

and

A

g

genesare both active genes that produce

g

-globin chains that differ only at position 136 (glycine is the product of the

G

g

gene, and alanineis the product of the

A

g

gene). The

cb

gene is a pseudogene, an evolutionary remnant of a previously active

b

-globin–like gene.Areas of nucleotide homology upstream from the initiation codon of each active gene, termed promoter elements, are involved inthe initiation of transcription and hence play a vital part in gene regulation. The cluster also contains regulatory elements that in-teract to promote erythroid-specific gene expression and to coordinate the developmental regulation of each gene, including he-moglobin switching. These include enhancers, distant regulatory elements that increase gene expression, and a master sequence,the cluster’s essential distal regulatory element, the

b

-globin locus-control region. This is a region that lies 20 kb upstream fromthe

e

-globin gene.

13

It encompasses five erythroid-lineage–specific nuclease hypersensitive sites (shown in red) that permit expres-sion of the downstream genes. In addition to these elements for up-regulation, several suppressor regions, or silencers, have beendefined in the

b

-globin gene cluster. Panel B shows the timing of the normal developmental switching of human hemoglobin. Early in fetal life the synthesis of theembryonic

a

-globin–like (

z

) chains switches to that of

a

-globin, which is produced thereafter. At the same time, the synthesis ofembryonic beta-like (

e

) chains switches to that of

g

-globin chains. The

a

-globin and

g

-globin chains combine to form fetal hemo-globin (

a

2

g

2

), the main

b

-globin–like globin during the remainder of fetal life and throughout early postnatal life. As a result of thedecline in the synthesis of

g

-globin chains in patients with

b

-thalassemia, fetal hemoglobin production becomes insufficient tocompensate for the excess of

a

-globin chains, the production of which is unaffected in

b

-thalassemia.

Locus-controlregion

10

50

6 42

30

18

a

g

d

b

e

z

30 0 6 18 30

BirthB

Before Birth (wk) After Birth (wk)

Glo

bin

-Ch

ain

Syn

thes

is(%

of

tota

l)

A

e Gg Ag cb d b

Chromosome 11

MEDICAL PROGRESS

Volume 341 Number 2

·

101

to produce functional mRNA, result in

b

0

-thalasse-mia. Mutations in highly conserved nucleotides flank-ing these sequences, or in “cryptic” splice sites, whichresemble a donor or acceptor splice site, result insevere as well as mild

b

+

-thalassemia. Substitutionsor small deletions affecting the conserved AATAAAsequence in the 3' untranslated region result in inef-fective cleavage of the mRNA transcript and causemild

b

+

-thalassemia.Mutations that interfere with translation involve the

initiation, elongation, or termination of globin-chainproduction and result in

b

0

-thalassemia. Approxi-mately half of all

b

-thalassemia mutations interferewith translation; these include frame-shift or nonsensemutations, which introduce premature terminationcodons and result in

b

0

-thalassemia. A more recent-ly identified family of mutations, usually involvingexon 3, results in the production of unstable globinchains of varying lengths that, together with a relativeexcess of

a

-globin chains, precipitate in red-cell pre-cursors and lead to ineffective erythropoiesis, even inthe heterozygous state. This is the molecular basis fordominantly inherited (

b

+

) thalassemia. In addition,

missense mutations, resulting in the synthesis of un-stable

b

-globin chains, cause

b

-thalassemia.

PATHOPHYSIOLOGY

Mechanisms of Anemia

In severe untreated

b

-thalassemia, erythropoiesismay be increased by a factor of up to 10, more than95 percent of which may be ineffective. Ineffectiveerythropoiesis, the hallmark of

b

-thalassemia, is a re-sult of the myriad deleterious effects of a relativeexcess of

a

-globin chains.

17

This relative excess in-terferes with most stages of normal erythroid matu-ration: both intramedullary death of red-cell precur-sors through arrest in the G

1

phase of the cell cycle

18

and accelerated intramedullary apoptosis of late eryth-roblasts

19,20

have been demonstrated. Studies of theconsequences of the accumulation of excess

a

-globinchains and their degradation products within the red-cell membrane and its skeleton

20-22

have also demon-strated abnormalities in the ratio of spectrin to band3 and in the function of band 4.1. This subject hasbeen thoroughly reviewed recently.20 The observa-

Figure 2. The Normal Structure of the b-Globin Gene and the Locations and Types of Mutations Resulting in b-Thalassemia.All b-globin–like genes contain three exons and two introns between codons 30 and 31 and 104 and 105, respectively. The primaryaction of all the mutations is to abolish the output of b-globin chains (b0-thalassemia; shown in red) or reduce the output (b+-thal-assemia; shown in green). The 170 different mutations that act in this way may interfere with the action of the b-globin gene at thetranscriptional level, in the processing of the primary transcript, in the translation of b-globin messenger RNA, or in the post-trans-lational stability of the b-globin gene product.

Intron

Mutation affecting initiation oftranscription

Mutation affecting splicing ofRNA from introns

Polyadenylation-signal mutation

Mutation affecting initiation oftranslation

Nonsense mutationFrame-shift mutation

IntronExon 1 Exon 2

Deletions

5' 3'

Intron

IntronExon 3

Other Mutations

102 · July 8, 1999

The New England Journal of Medicine

tion that the presence of excess membrane iron mayaggravate membrane changes22 has led to interest inthe red-cell membrane as a potential therapeutic tar-get in b-thalassemia. In a mouse model, increased cel-lular rigidity and decreased stability in connectionwith membrane-associated a-globin chains23 have re-portedly been ameliorated during exposure to agentsthat bind membrane iron.24 Further understandingof these processes may guide future therapies.

Clinical Consequences of Anemia

The severe ineffective erythropoiesis results in eryth-roid marrow expansion to as much as 30 times thenormal level. Both an increase in plasma volume asa result of shunting through expanded marrow and

progressive splenomegaly exacerbate anemia (Fig. 3).Increased erythropoietin synthesis may stimulate theformation of extramedullary erythropoietic tissue, pri-marily in the thorax and paraspinal region. Marrowexpansion also results in characteristic deformities ofthe skull and face, as well as osteopenia and focal de-fects in bone mineralization,25,26 and may aggravate apainful periarticular syndrome characterized histo-logically by microfractures and osteomalacia.27 Marrowhyperplasia leads ultimately to increased iron absorp-tion and progressive deposition of iron in tissues.

Cellular Heterogeneity and Fetal Hemoglobin Production

Although fetal hemoglobin synthesis persists afterbirth to some degree, its production is insufficient

Figure 3. Effects of Excess Production of Free a-Globin Chains.Primary disease processes are indicated in orange, and compensatory mechanisms in yellow. Excess unbound a-globin chains andtheir degradation products precipitate in red-cell precursors, causing defective maturation and ineffective erythropoiesis. Hemolysis,owing to the presence of inclusions in the red cells and damage to their membranes by a-globin chains and their degradation prod-ucts, also contributes to the anemia. Anemia stimulates the synthesis of erythropoietin, leading to an intense proliferation of theineffective marrow, which in turn causes skeletal deformities and a variety of growth and metabolic abnormalities. The anemia isfurther exacerbated by hemodilution, caused by the shunting of blood through the expanded marrow, and by splenomegaly result-ing from entrapment of abnormal red cells in the spleen. Bone marrow expansion also results in characteristic deformities of theskull and face, severe osteopenia, and increased iron absorption.

Excess freea-globin chains

Denaturation

Degradation

Formation of hemeand hemichromes

Iron-mediated toxicity

Reduced tissue

oxygenation

Increased

iron absorption

Hemolysis

Increasederythropoietin

synthesis

Skeletaldeformities,osteopenia

Ironoverload

Erythroidmarrow

expansion

Splenomegaly

Ineffectiveerythropoiesis

Removal ofdamaged red cells

Anemia

Membranebinding ofIgG and C3

MEDICAL PROGRESS

Volume 341 Number 2 · 103

to compensate for the reduced synthesis of b-globinchains and the relative excess of a-globin chains.15

The elevated concentrations of fetal hemoglobin —2 to 4 g per deciliter — observed in patients withb-thalassemia reflect a combination of the selectionof precursors that produce relatively more fetal he-moglobin and erythroid expansion, which appears tofavor the production of g-globin chains. Even higherfetal hemoglobin concentrations are associated withspecific b-thalassemia alleles28,29 or other genetic de-terminants within or linked to the b-globin com-plex.15 At least two other determinants may affect thesynthesis of fetal hemoglobin, one on chromosome630 and the other on the X chromosome.31,32

Clinical Forms

The b-thalassemias include four clinical syndromesof increasing severity: two conditions are generallyasymptomatic, the silent carrier state and b-thalas-semia trait, and usually result from the inheritance ofone mutant b-globin gene, and two require medicalmanagement, thalassemia intermedia and thalasse-mia major. The more severe forms most often resultfrom homozygosity or compound heterozygosity fora mutant b-globin allele and, occasionally, from het-erozygosity for dominant mutations.33 Homozygousor compound heterozygous b-thalassemia usually pre-sents no diagnostic problems. The early onset of ane-mia, characteristic blood changes, and elevated fetalhemoglobin concentrations are found in no othercondition. The diagnosis can be confirmed by thedemonstration of the b-thalassemia trait in both par-ents. This condition is characterized by mild anemia,reduced mean cell volumes and mean cell hemoglobinconcentrations,29 and elevated concentrations of thenormal minor adult component of hemoglobin (usu-ally exceeding 3.5 percent), hemoglobin A2 (a2d2).

Thalassemia major and thalassemia intermedia haveno specific molecular correlate but encompass a widespectrum of clinical and laboratory abnormalities.34

Patients referred to as having thalassemia major areusually those who come to medical attention in thefirst year of life and subsequently require regulartransfusions to survive. Those who present later orwho seldom need transfusions are said to have thalas-semia intermedia.35 After thalassemia is diagnosed,patients who appear not to require immediate trans-fusion may benefit from a period of observation andfolate repletion, particularly if the disease is diagnosedafter the age of one year. This approach will allow theidentification of patients in whom early growth anddevelopment are normal and whose well-compensat-ed anemia may be exacerbated only by infection, folatedeficiency, or increasing hypersplenism.34-37 With ad-vancing age, even patients with mild forms may haveserious complications, including osteopenia, iron load-ing in tissues, and ectopic marrow expansion. Theclassic changes of untreated thalassemia major are now

regularly seen only in countries without resources tosupport long-term transfusion programs.

Relation between Genotype and Phenotype

Several genetic factors may ameliorate the severityof b-thalassemia.29 First, the underlying mutationsvary widely in their effect on the synthesis of b-globinchains.38 Co-inheritance of a-thalassemia may reducethe severity of the globin-chain imbalance. Many dif-ferent interactions with structural hemoglobin vari-ants may also result in a complex series of clinicalphenotypes.29 The interactions of b-thalassemia withtwo of these variants, hemoglobin S and hemoglo-bin E, are of global importance.

The clinical consequences of the interaction withhemoglobin S depend mainly on the b-thalassemiaallele. If inherited with b0-thalassemia or severeb+-thalassemia, the resulting clinical disorder maybe indistinguishable from sickle cell anemia. By con-trast, interactions with mild b+-thalassemia allelesproduce a milder sickling disorder.

Although hemoglobin E b-thalassemia is probablythe most common serious hemoglobinopathy world-wide,39 its natural history remains poorly under-stood.29 The mutation that produces hemoglobin Eactivates a cryptic splice site in exon 1 in the b-globingene; hence, hemoglobin E is associated with mildb-thalassemia. For reasons that are not well under-stood,29 the interaction of hemoglobin E and b-thal-assemia results in a wide spectrum of clinical disorders:some are indistinguishable from thalassemia major,and some are much milder and not transfusion-dependent. Finally, a number of acquired and envi-ronmental factors, including progressive splenomeg-aly, exposure to infections, socioeconomic factors,and the availability of medical care, may also modifythe severity of the disease.

COMPLICATIONS OF DISEASE

Iron Overload

Iron overload of tissue, which is fatal with or with-out transfusion if not prevented or adequately treated,is the most important complication of b-thalassemiaand is a major focus of management.40 In patientswho are not receiving transfusions, abnormally reg-ulated iron absorption results in increases in bodyiron burden ranging from 2 to 5 g per year, depend-ing on the severity of erythroid expansion.41,42 Reg-ular transfusions may double this rate of iron accu-mulation. Although most clinical manifestations ofiron loading do not appear until the second decadeof life in patients with inadequate chelation, evi-dence from serial liver biopsies in very young pa-tients indicates that the deleterious effects of iron areinitiated much earlier than this. After approximatelyone year of transfusions, iron begins to be depositedin parenchymal tissues,43 where it may cause sub-stantial toxicity as compared with that within retic-

104 · July 8, 1999

The New England Journal of Medicine

uloendothelial cells.44,45 As iron loading progresses,the capacity of serum transferrin, the main transportprotein of iron, to bind and detoxify iron may be ex-ceeded and a non–transferrin-bound fraction ofplasma iron may promote the generation of free hy-droxyl radicals, propagators of oxygen-related dam-age.44,45 The advances in free-radical chemistry thathave clarified the toxic properties of these and otheroxygen-derived species generated by iron, whichmay cause widespread tissue damage, have recentlybeen summarized.45 Although the body maintains anumber of antioxidant mechanisms against damageinduced by free radicals, including superoxide dis-mutases, catalase, and glutathione peroxidase, in pa-tients with large iron burdens these may not preventoxidative damage.44,45

In the absence of chelating therapy the accumula-tion of iron results in progressive dysfunction of theheart, liver, and endocrine glands.40 Within the heart,changes associated with chronic anemia are usuallypresent in patients who are not receiving transfusionsand are aggravated by iron deposition. In responseto iron loading, human myocytes in vitro increasethe transport of non–transferrin-bound iron,46 pos-sibly thereby aggravating cardiac iron loading. Exten-sive iron deposits are associated with cardiac hyper-trophy and dilatation, degeneration of myocardialfibers, and in rare cases fibrosis.47 In patients who arereceiving transfusions but not chelating therapy,symptomatic cardiac disease has been reported with-in 10 years after the start of transfusions48 and maybe aggravated by myocarditis49 and pulmonary hy-pertension.50,51 The survival of patients with b-thal-assemia is determined by the magnitude of ironloading within the heart.52,53

Iron-induced liver disease is a common cause ofdeath in older patients54 and is often aggravated byinfection with hepatitis C virus. Within two years af-ter the start of transfusions, collagen formation55

and portal fibrosis56 have been reported; in the ab-sence of chelating therapy, cirrhosis may develop inthe first decade of life.43,57,58 The extent of theseprocesses may be underestimated if fewer than threecores of liver are sampled at biopsy.59 The risk of he-patic fibrosis is augmented at body iron burdens cor-responding to hepatic iron concentrations of morethan 7 mg per gram of liver, dry weight (Fig. 4).61

As in cultured heart cells, in cultured hepatocytesthe transport of non–transferrin-bound iron is in-creased,62 possibly aggravating iron loading in vivo.

The striking increases in survival in patients withb-thalassemia over the past decade have focused at-tention on abnormal endocrine function, now themost prevalent iron-induced complication in olderpatients. Iron loading within the anterior pituitary isthe primary cause of disturbed sexual maturation, re-ported in 50 percent of both boys and girls with thecondition.63 Furthermore, early secondary amenor-

rhea occurs in approximately one quarter of femalepatients over the age of 15 years.63 Even in the mod-ern era of iron-chelating therapy, diabetes mellitus isobserved in about 5 percent of adults.63 As the ironburden increases and iron-related liver dysfunctionprogresses, hyperinsulinemia occurs as a result of re-duced extraction of insulin by the liver, leading to ex-haustion of beta cells and reduced circulating insulinconcentrations.64 Studies reporting reduced serumconcentrations of trypsin and lipase65 suggest that theexocrine pancreas is also damaged by iron loading.Over the long term, iron deposition also damagesthe thyroid, parathyroid, and adrenal glands66,67 andmay provoke pulmonary hypertension, right ventric-ular dilatation, and restrictive lung disease.51,68-70

In most studies, bone density is markedly reducedin patients with b-thalassemia, particularly those withhypogonadism. Osteopenia may be related to marrowexpansion, even in patients who receive transfusions,42

or to iron-induced osteoblast dysfunction, diabetes,hypoparathyroidism, or hypogonadism.71-73

CONTROL AND MANAGEMENT

Prevention Programs and Prenatal Diagnosis

Screening programs, aimed at prevention of thedisease, and prenatal diagnosis have resulted in amarked reduction in the birth rate of affected chil-dren in Greece, Cyprus, continental Italy, and Sar-dinia.74 Widespread use of similar programs in otherareas of the world has not yet been possible. Screen-ing for carriers is performed most efficiently bymeasurement of the red-cell indexes and, in samplesfrom persons with reduced mean cell volumes andmean cell hemoglobin concentrations, estimation ofthe hemoglobin A2 concentration. The practicalproblems associated with screening for rarer formsof b-thalassemia and the effect of coexistent a-thal-assemia on the red-cell indexes have been reviewedrecently.29 Prenatal diagnosis, first carried out by fe-tal-blood sampling and assessment of globin-chainsynthesis in fetal blood, more recently has involveddirect analysis of fetal DNA obtained by chorionic-villus sampling. This approach is associated with avery slightly increased risk of fetal loss and an errorrate in experienced laboratories of less than 1 per-cent. The practical aspects of fetal-DNA analysishave also been recently reviewed.29,74

Medical Therapy

A decision to initiate regular transfusions in pa-tients with b-thalassemia may be difficult and shouldbe based on the presence and severity of the symp-toms and signs of anemia, including failure of growthand development. Only rarely is genotyping helpfulin this decision. The goals of transfusion include cor-rection of anemia, suppression of erythropoiesis, andinhibition of increased gastrointestinal absorptionof iron. “Hypertransfusion” and “supertransfusion”

MEDICAL PROGRESS

Volume 341 Number 2 · 105

regimens, which achieve these goals but are associ-ated with substantial iron loading,40,75 have beensupplanted by regimens in which the hemoglobinconcentration before transfusion does not exceed9.5 g per deciliter.76 These newer regimens are asso-ciated with both adequate marrow suppression andrelatively lower rates of iron accumulation.

The beneficial effects of iron-chelating therapy withparenteral deferoxamine, the only chelating agentwidely available for clinical use, on the complicationsof iron loading have recently been reviewed.40 As aresult of programs of deferoxamine therapy, the prog-nosis for patients in countries able to afford thistherapy has greatly improved, in contrast to the prog-nosis for patients in developing countries, where wide-spread implementation of this regimen is still awaited.

Adequate deferoxamine therapy prevents early deathfrom cardiac disease: maintenance of body iron bur-dens corresponding to hepatic iron concentrations ofless than 15 mg per gram, dry weight, greatly de-

crease the risk of clinical disease.52 Nearly normal con-centrations of hepatic iron can be maintained withmodern regimens of deferoxamine. Moreover, defer-oxamine arrests the progression of hepatic fibrosis tocirrhosis, even when administered in regimens thatstabilize, rather than reduce, the body iron burden.77

The importance of this finding in the seminal studythat ushered in the modern era of deferoxamine ther-apy is highlighted by evidence that in another formof iron overload, hereditary hemochromatosis, pro-gression of hepatic fibrosis is a critical event associ-ated with an increased risk of death.61

A favorable effect of a sustained reduction in bodyiron is also suggested by the relatively low prevalenceof thyroid, parathyroid, and adrenal abnormalities inthe modern era.78 In parallel, early and intensive de-feroxamine therapy may increase the incidence ofnormal sexual maturation,78 but it apparently does notreverse established abnormalities.40 Similarly, althoughdeferoxamine prevents diabetes mellitus,52 there is

Figure 4. Hepatic Iron Burden over Time and the Effect of Various Hepatic Iron Concentrations in Patients with Thalas-semia Major, Homozygous Hemochromatosis, and Heterozygous Hemochromatosis.Because efforts to maintain normal hepatic iron concentrations in patients with thalassemia major are frequently asso-ciated with adverse reactions to deferoxamine, an optimal range within which hepatic iron may be safely maintained,derived in part from clinical experience with the iron-loading disorder hereditary hemochromatosis, has been proposed.In a proportion of patients who are heterozygous for hemochromatosis, moderate iron loading corresponding to con-centrations of approximately 3.2 to 7 mg of iron per gram of liver, dry weight (indicated in yellow), is associated withnormal survival without complications of iron overload.60 In patients who are homozygous for this disorder, concentra-tions exceeding this range, up to approximately 15 mg of iron per gram of liver, dry weight (indicated in blue), are as-sociated with an increased risk of complications of iron overload,61 whereas maintenance of levels exceeding 15 mg ofiron per gram of liver, dry weight, greatly increases the risk of cardiac disease and early death in patients with thalas-semia major.52 In patients with thalassemia major who receive regular transfusions, the rate of iron loading is muchmore accelerated and death usually occurs before the third decade of life. The goal of treatment is the maintenance ofa body-iron burden corresponding to a hepatic iron concentration of approximately 3.2 to 7 mg of iron per gram of liver,dry weight, achievable with regular deferoxamine therapy.40 The serum ferritin concentrations corresponding to theseranges of hepatic iron are not clearly defined.

0

15,000

0 50

5,000

10,000

0

50

30

20

10

40

10 20 30 40

Age (years)

HeterozygoushemochromatosisOptimal iron chelation level

Threshold for cardiac disease and early death

Homozygoushemochromatosis

Thalassemia major

Hep

atic

Iro

n (

µg

/g o

f liv

er, w

et w

eig

ht)

Hep

atic

Iro

n (

mg

/g o

f liv

er, d

ry w

eig

ht)

Increased risk of complications

Adequate iron chelation

Normal hepatic ironconcentration

106 · July 8, 1999

The New England Journal of Medicine

no evidence that it can reverse this complication. Insummary, modern regimens of subcutaneous defer-oxamine may extend survival free of many complica-tions of iron overload, if body iron is reduced ormaintained below critical concentrations.40,52,53,79

A balance between the effectiveness of deferoxa-mine and its toxicity — the latter observed primarilyin the presence of relatively low body iron burdens80

— can be maintained through regular determina-tions of body iron burden. In clinical practice, theserum ferritin concentration is commonly used toassess the effectiveness of treatment. It is increasing-ly recognized that reliance on this test may lead toerrors in management; changes in body iron accountfor little more than half the variation in serum fer-ritin concentrations.81 By contrast, the measurementof hepatic iron stores, whose concentrations arehighly correlated with total body iron stores,82 pro-vides the most quantitative, specific, and sensitivemethod of evaluating iron burden in patients withthalassemia. Determination of hepatic iron concen-trations in liver-biopsy specimens obtained with ul-trasonographic guidance is safe and permits rationaladjustments in iron-chelating therapy.40 Magneticsusceptometry provides a direct measure of hepaticiron stores that is quantitatively equivalent to thatdetermined by biopsy of at least 0.6 mg of liver, dryweight,83 over a range of iron concentrations.84 Mag-netic susceptometry is currently available in onlytwo centers worldwide. By contrast, the more widelyavailable technique of magnetic resonance imagingfails to provide accurate quantitation of hepatic ironconcentrations in patients with severe iron overload,hepatic fibrosis, or both.85

Bone Marrow Transplantation

Bone marrow transplantation from HLA-identicaldonors has been successfully performed worldwide inover 1000 patients with severe b-thalassemia.86 Out-comes after transplantation are greatly influenced bythe presence of hepatomegaly, portal fibrosis, and in-effective chelating therapy before transplantation.87

Children without any of these risk factors have ratesof survival and disease-free survival exceeding 90 per-cent three years after transplantation. In those withall three risk factors, and in most adults, the rates areapproximately 60 percent. Lower success rates are re-ported at smaller centers.87 Complications include arate of chronic graft-versus-host-disease ranging from2 to 8 percent and a variable incidence of mixed chi-merism.86 Post-transplantation management of preex-isting hepatic iron overload, iron-induced cardiac dys-function, and viral hepatitis may prevent progressionof these processes.86 There is interest in experimentalapproaches to bone marrow replacement in patientswith thalassemia, including cord-blood transplanta-tion,88 the use of unrelated phenotypically matcheddonors,89 and in utero transplantation.90

Experimental Therapies

Chelators Other Than Deferoxamine

Difficulties associated with deferoxamine therapyhave led to a search for alternatives, including orallyactive iron chelators. The administration of oneagent, deferiprone, was reported to have a favorableshort-term effect on body iron in the one study inwhich serial systematic determinations of hepatic ironwere obtained.91 Two subsequent long-term studieshave suggested that hepatic iron may stabilize at orincrease to concentrations associated with an in-creased risk of cardiac disease and early death52 inapproximately half of patients.92,93 Previously recog-nized adverse effects of deferiprone included embry-otoxicity, teratogenicity, neutropenia, and agranulo-cytosis.94 Long-term treatment has been reported tobe associated with progression of hepatic fibrosis; theodds of progression of fibrosis were estimated to in-crease by a factor of 5.8 with each additional year ofdeferiprone therapy.95 In another study, four deathsdue to cardiac failure were reported during long-termtherapy.92 The results of these clinical trials virtuallyrecapitulate those in two animal species, in whichdeferiprone and a structurally similar compound wereshown to increase hepatic and cardiac iron loading,worsen hepatic fibrosis, and induce cardiac and mus-culoskeletal fibrosis.95,96 Taken together, these datasuggest that deferiprone does not adequately controlbody iron in a substantial proportion of patients andmay promote worsening of hepatic fibrosis. Thesestudies support cautions previously expressed aboutthe long-term administration of this agent.97

The results of long-term follow-up of the effec-tiveness of other modes of administration of defer-oxamine are awaited. These include deferoxamineattached to high-molecular starch,98 administered intwice-daily subcutaneous boluses,99 and given in a lip-id vehicle, permitting slow release.100

Augmentation of Fetal-Hemoglobin Synthesis

Several trials have attempted to augment the syn-thesis of fetal hemoglobin in an effort to amelioratethe severity of b-thalassemia.101 Administration of in-travenous 5-azacytidine was associated with increasesin the hemoglobin concentration in a few patients102;the potential toxicity of the drug later shifted interestto less toxic alternatives. Therapy with hydroxyu-rea,103,104 butyric acid compounds,105 and these agentsin combination106 has reduced or eliminated transfu-sion requirements in some patients. Other studieshave reported only small increases in fetal and totalhemoglobin concentrations during the administra-tion of hydroxyurea107,108 and both intravenous109,110

and oral111,112 butyrate compounds.How can the augmentation of fetal hemoglobin

be optimized? Studies in humans and animal modelsof b-thalassemia, including transgenic mice,113 have

MEDICAL PROGRESS

Volume 341 Number 2 · 107

suggested that increases in the production of g-glo-bin chains may be influenced by the degree of eryth-roid marrow expansion, sequential administration ofspecific combinations of agents, the degree to whichthe g-globin gene is already partially activated, or allof these.113-116 Furthermore, the striking clinical re-sponses observed in patients with mutations104-106 thatdelete specific sequences within the b-globin genecluster that may have a key role in the silencing ofadjacent genes117 indicate that delineation of some cissequences may influence the inducibility of the g-glo-bin gene.

Although they address a highly desirable, cost-effective goal in b-thalassemia, therapies to increasethe synthesis of fetal hemoglobin in the disorder have,with few exceptions, proved disappointing to date.Nonetheless, important avenues to be pursued in fur-ther studies include the identification of specific mu-tations that may respond to therapy, particularly withspecific combinations of agents.

Gene Therapy

Permanent correction of genetic deficit of thehematopoietic system requires the transfer of genesinto stem cells and long-term, high-level, lineage-specific expression of these cells after autologous trans-plantation; mature cells and committed progenitorsdo not have the proliferative capacity to reconstitutethe entire hematopoietic system. Over the past dec-ade, there has been progress in the development oftransduction methods and vectors.118 Remainingproblems include the identification of all sequencesrequired for stable, high-level expression of the genesand the development of more effective and safe vec-tors for the transfer of genes.119 Another approach,correction of the defective gene by site-directed re-combination, is feasible, but current methods lack thedegree of efficiency required.120

CONCLUSIONS

Among the first diseases to be studied at the mo-lecular level, the b-thalassemias remain a model forunderstanding the relation between the molecularpathology of a disease and its clinical diversity. Atthe same time, these disorders have become an in-creasingly important part of clinical practice in allcountries with large populations from the tropics.The marked increase in survival, to the fifth decadeof life, of patients with well-managed b-thalassemiain developed countries represents one of the mostdramatic alterations in morbidity and mortality asso-ciated with a genetic disease in this century. Still,nearly 75 years after the fascinating initial descrip-tion of “peculiar bone changes” and other signs andsymptoms of the disorder, the b-thalassemias haveemerged as a huge public health problem worldwide.They remain a therapeutic challenge for the nextmillennium.

Supported in part by research grants from the Medical Research Councilof Canada, the Ontario Heart and Stroke Foundation, and the OntarioThalassemia Foundation. Dr. Olivieri is the recipient of a Scientist Awardfrom the Medical Research Council of Canada.

I am indebted to David Nathan, David Weatherall, Gary Brit-tenham, John Porter, Alan Schechter, Elliott Vichinsky, Brenda Gal-lie, Helen Chan, Peter Durie, John Dick, Marc Giacomelli, PaulRanalli, Arthur Schafer, Michele Brill-Edwards, Lori West, MiriamKaufman, Michael Langlois, David Kern, Bob Phillips, Maria Mu-raca, Bill Graham, Rhonda Love, Doug Templeton, John Polanyi,and Michael Baker, without whose support this review would not havebeen written.

REFERENCES

1. Cooley TB, Lee P. A series of cases of splenomegaly in children with anemia and peculiar bone changes. Trans Am Pediatr Soc 1925;37:29-30.2. Rietti F. Ittero emolitico primitivo. Alti Accad Sci Med Nat Ferrara 1925;2:14-9.3. Greppi E. Ittero emolitico familiare con aumento della resistenza dei globuli. Minerva Med 1928;8:1-11.4. Micheli P, Penati P, Momigliano LG. Ulteriori richereche sulla anemia ipocromica splenomegalica con poichilocitosi. Atti Soc Ital Ematol Haema-tol (Pavia) 1935;16:Suppl 1:10-3.5. Whipple GH, Bradford WL. Mediterranean disease — thalassemia (erythroblastic anemia of Cooley): associated pigment abnormalities simu-lating hemochromatosis. J Pediatr 1936;9:279-311.6. Weatherall DJ, Clegg JB. Thalassaemia — a global public health prob-lem. Nat Med 1996;2:847-9.7. Higgs DR. a-Thalassaemia. Baillieres Clin Haematol 1993;6:117-50.8. Chui DH, Waye JS. Hydrops fetalis caused by alpha-thalassemia: an emerging health care problem. Blood 1998;91:2213-22.9. Weatherall DJ. The thalassemias. In: Stamatoyannopoulos G, Nienhuis AW, Majerus PH, Varmus H, eds. The molecular basis of blood diseases. 2nd ed. Philadelphia: W.B. Saunders, 1994:157-205.10. Flint J, Harding RM, Boyce AJ, Clegg JB. The population genetics of the haemoglobinopathies. Baillieres Clin Haematol 1993;6:215-62.11. Weatherall DJ. Common genetic disorders of the red cell and the “ma-laria hypothesis.” Ann Trop Med Parasitol 1987;81:539-48.12. Allen SJ, O’Donnell A, Alexander NDE, et al. a+-Thalassaemia pro-tects children against disease caused by other infections as well as malaria. Proc Natl Acad Sci U S A 1997;94:14736-41.13. Orkin SH. Regulation of globin gene expression in erythroid cells. Eur J Biochem 1995;231:271-81.14. Wood WG, Weatherall DJ. Developmental genetics of the human hae-moglobins. Biochem J 1983;215:1-10.15. Wood WG. Increased HbF in adult life. Baillieres Clin Haematol 1993;6:177-213.16. Grosveld F, Dillon N, Higgs D. The regulation of human globin gene expression. Clin Haematol 1993;6:31-55.17. Nathan DG, Gunn RB. Thalassemia: the consequences of unbalanced hemoglobin synthesis. Am J Med 1966;41:815-30.18. Wickramasinghe SN. The morphology and kinetics of erythropoiesis in homozygous b-thalassaemia. In: Congenital disorders of erythropoiesis. Amsterdam: Elsevier, 1976:221-37.19. Yuan J, Angelucci E, Lucarelli G, et al. Accelerated programmed cell death (apoptosis) in erythroid precursors of patients with severe beta-thal-assemia. Blood 1993;82:374-7.20. Schrier SL. Pathobiology of thalassemic erythrocytes. Curr Opin He-matol 1997;4:75-8.21. Shinar E, Rachmilewitz EA. Haemoglobinopathies and red cell mem-brane function. Baillieres Clin Haematol 1993;6:357-69.22. Grinberg LN, Rachmilewitz EA. Oxidative stress in b-thalassemic red blood cells and potential use of antioxidants. In: Beuzard Y, Lubin B, Rosa J, eds. Sickle cell disease and thalassaemia: new trends in therapy. London: Colloque INSERM/John Libby Eurotext, 1995:519-24.23. Sorensen S, Rubin E, Polster H, Mohandas N, Schrier S. The role of membrane skeletal-associated alpha-globin in the pathophysiology of beta-thalassemia. Blood 1990;75:1333-6.24. Shalev O, Repka T, Goldfarb A, et al. Deferiprone (L1) chelates patho-logic iron deposits from membranes of intact thalassemic and sickle RBC both in vitro and in vivo. Blood 1995;86:2008-13.25. Rioja L, Girot R, Garabedian M, Cournot-Witmer G. Bone disease in children with homozygous beta-thalassemia. Bone Miner 1990;8:69-86.26. Orvieto R, Leichter I, Rachmilewitz EA, Margulies JY. Bone density, mineral content, and cortical index in patients with thalassemia major and

108 · July 8, 1999

The New England Journal of Medicine

the correlation to their bone fractures, blood transfusions, and treatment with desferrioxamine. Calcif Tissue Int 1992;50:397-9.27. Gratwick GM, Bullough PG, Bohne WHO, Markenson AL, Peterson CM. Thalassemia osteoarthropathy. Ann Intern Med 1978;88:494-501.28. Thein SL. b-Thalassaemia. Baillieres Clin Haematol 1993;6:151-75.29. Weatherall DJ, Clegg JB. The thalassaemia syndromes. 4th ed. Oxford, England: Blackwell Scientific (in press).30. Thein SL, Sampietro M, Rohde K, et al. Detection of a major gene for heterocellular hereditary persistence of fetal hemoglobin after account-ing for genetic modifiers. Am J Hum Genet 1994;54:214-28.31. Dover GJ, Smith KD, Chang YC, et al. Fetal hemoglobin levels in sick-le cell disease and normal individuals are partially controlled by an X-linked gene located at Xp22.2. Blood 1992;80:816-24.32. Chang YC, Smith KD, Moore RD, Serjeant GR, Dover GJ. An analysis of fetal hemoglobin variation in sickle cell disease: the relative contribu-tions of the X-linked factor, b-globin haplotypes, a-globin gene number, gender, and age. Blood 1995;85:1111-7.33. Thein SL, Hesketh C, Taylor P, et al. Molecular basis for dominantly inherited inclusion body b thalassemia. Proc Natl Acad Sci U S A 1990;87:3924-8.34. Camaschella C, Cappellini MD. Thalassemia intermedia. Haematolog-ica 1995;80:58-68.35. Cao A. Diagnosis of b-thalassemia intermedia at presentation. In: Fu-charoen S, Rowley PT, Paul NW, eds. Thalassemia: pathophysiology and management. Part B. Vol. 23 of Birth defects: original article series. New York: Alan R. Liss, 1988:219-26.36. Rund D, Oron-Karni V, Filon D, Goldfarb A, Rachmilewitz E, Op-penheim A. Genetic analysis of b-thalassemia intermedia in Israel: diversity of mechanisms and unpredictability of phenotype. Am J Hematol 1997;54:16-22.37. Ho PJ, Hall GW, Luo LY, Weatherall DJ, Thein SL. Beta thalassaemia intermedia: is it possible consistently to predict phenotype from genotype? Br J Haematol 1998;100:70-8.38. Huisman THJ, Carver MFH, Baysal E. A syllabus of thalassemia mu-tations. Augusta, Ga.: Sickle Cell Anemia Foundation, 1997:309.39. Weatherall DJ. Hemoglobin E b-thalassemia: an increasingly common disease with some diagnostic pitfalls. J Pediatr 1998;132:765-7.40. Olivieri NF, Brittenham GM. Iron-chelating therapy and the treatment of thalassemia. Blood 1997;89:739-61. [Erratum, Blood 1997;89:2621.]41. Pippard MJ, Callender ST, Warner GT, Weatherall DJ. Iron absorption and loading in beta-thalassaemia intermedia. Lancet 1979;2:819-21.42. Pootrakul P, Kitcharoen K, Yansukon P, et al. The effect of erythroid hyperplasia on iron balance. Blood 1988;71:1124-9.43. Risdon RA, Flynn DM, Barry M. The relation between liver iron con-centration and liver damage in transfusional iron overload in thalassaemia and the effect of chelation therapy. Gut 1973;14:421.44. Hershko C, Weatherall DJ. Iron-chelating therapy. Crit Rev Clin Lab Sci 1988;26:303-45.45. Hershko C, Konijn AM, Link G. Iron chelators for thalassaemia. Br J Haematol 1998;101:399-406.46. Parkes JG, Hussain RA, Olivieri NF, Templeton DM. Effects of iron loading on uptake, speciation, and chelation of iron in cultured myocardial cells. J Lab Clin Med 1993;122:36-47.47. Buja LM, Roberts W. Iron in the heart: etiology and clinical signifi-cance. Am J Med 1971;51:209-21.48. Wolfe L, Olivieri N, Sallan D, et al. Prevention of cardiac disease by subcutaneous deferoxamine in patients with thalassemia major. N Engl J Med 1985;312:1600-3.49. Kremastinos DT, Tiniakos G, Theodorakis GN, Katritsis DG, Toutou-zas PK. Myocarditis in b-thalassemia major: a cause of heart failure. Circu-lation 1995;91:66-71.50. Aessopos A, Stamatelos G, Skoumas V, Vassilopoulos G, Mantzourani M, Loukopoulos D. Pulmonary hypertension and right heart failure in pa-tients with b-thalassemia intermedia. Chest 1995;107:50-3.51. Du ZD, Roguin N, Milgram E, Saab K, Koren A. Pulmonary hyper-tension in patients with thalassemia major. Am Heart J 1997;134:532-7.52. Brittenham GM, Griffith PM, Nienhuis AW, et al. Efficacy of deferox-amine in preventing complications of iron overload in patients with thalas-semia major. N Engl J Med 1994;331:567-73.53. Olivieri NF, Nathan DG, MacMillan JH, et al. Survival in medically treated patients with homozygous b-thalassemia. N Engl J Med 1994;331:574-8.54. Zurlo MG, De Stefano P, Borgna-Pignatti C, et al. Survival and causes of death in thalassaemia major. Lancet 1989;2:27-30.55. Iancu TC, Neustein HB, Landing BH. The liver in thalassaemia ma-jor: ultrastructural observations. In: Iron metabolism: Ciba Symposium 51. Amsterdam: Elsevier, 1977:293-309.56. Thakerngpol K, Fucharoen S, Boonyaphipat P, et al. Liver injury due to iron overload in thalassemia: histopathologic and ultrastructural studies. Biometals 1996;9:177-83.

57. Witzleben CL, Wyatt JP. The effect of long survival on the pathology of thalassaemia major. J Pathol Bacteriol 1961;82:1-12.58. Jean G, Terzoli S, Mauri R, et al. Cirrhosis associated with multiple transfusions in thalassaemia. Arch Dis Child 1984;59:67-70.59. Maharaj B, Maharaj RJ, Leary WP, et al. Sampling variability and its influence on the diagnostic yield of the percutaneous needle biopsy of the liver. Lancet 1986;1:523-5.60. Cartwright GE, Edwards CQ, Kravitz K, et al. Hereditary hemochro-matosis: phenotypic expression of the disease. N Engl J Med 1979;301:175-9.61. Niederau C, Fischer R, Purschel A, Stremmel W, Haussinger D, Stroh-meyer G. Long-term survival in patients with hereditary hemochromatosis. Gastroenterology 1996;110:1107-79.62. Parkes JG, Randell EW, Olivieri NF, Templeton DM. Modulation by iron loading and chelation of the uptake of non-transferrin-bound iron by human liver cells. Biochim Biophys Acta 1995;1243:373-80.63. Italian Working Group on Endocrine Complications in Non-endo-crine Diseases. Multicentre study on prevalence of endocrine complications in thalassaemia major. Clin Endocrinol (Oxf ) 1995;42:581-6.64. Cavallo-Perin P, Pacini G, Cerutti F, et al. Insulin resistance and hy-perinsulinemia in homozygous b-thalassemia. Metabolism 1995;44:281-6.65. Gullo L, Corcioni E, Brancati C, Bria M, Pezzilli R, Sprovieri G. Mor-phologic and functional evaluation of the exocrine pancreas in b-thalasse-mia major. Pancreas 1993;8:176-80.66. Magro S, Puzzonia P, Consarino C, et al. Hypothyroidism in patients with thalassemia syndromes. Acta Haematol 1990;84:72-6.67. Sklar CA, Lew LQ, Yoon DJ, David R. Adrenal function in thalasse-mia major following long-term treatment with multiple transfusions and chelation therapy: evidence for dissociation of cortisol and adrenal andro-gen secretion. Am J Dis Child 1987;141:327-30.68. Bacalo A, Kivity S, Heno N, Greif Z, Greif J, Topilsky M. Blood trans-fusion and lung function in children with thalassemia major. Chest 1992;101:362-70.69. Factor JM, Pottipati SR, Rappaport I, Rosner IK, Lesser ML, Giardi-na PJ. Pulmonary function abnormalities in thalassemia major and the role of iron overload. Am J Respir Crit Care Med 1994;149:1570-4.70. Tai DYH, Wang YT, Lou J, Wang WY, Mak KH, Cheng HK. Lungs in thalassaemia major patients receiving regular transfusion. Eur Respir J 1996;9:1389-94.71. Anapliotou MLG, Kastanias IT, Psara P, Evangelou EA, Liparaki M, Dimitriou P. The contribution of hypogonadism to the development of os-teoporosis in thalassaemia major: new therapeutic approaches. Clin Endo-crinol (Oxf ) 1995;42:279-87.72. Filosa A, Di Maio S, Vocca S, Saviano A, Esposito G, Pagano L. Lon-gitudinal monitoring of bone mineral density in thalassemic patients: ge-netic structure and osteoporosis. Acta Paediatr 1997;86:342-6.73. Olivieri NF. Thalassaemia: clinical management. Baillieres Clin Hae-matol 1998;11:147-62.74. Cao A, Galanello R, Rosatelli MC. Prenatal diagnosis and screening of the haemoglobinopathies. Baillieres Clin Haematol 1998;11:215-38.75. Fosburg MT, Nathan DG. Treatment of Cooley’s anemia. Blood 1990;76:435-44.76. Cazzola M, Bornga-Pignatti C, Locatelli F, Ponchio L, Beguin Y, De Stefano P. A moderate transfusion regimen may reduce iron loading in b-thalassemia major without producing excessive expansion of erythro-poiesis. Transfusion 1997;37:135-40.77. Barry M, Flynn DM, Letsky EA, Risdon RA. Long-term chelation therapy in thalassaemia major: effect on liver iron concentration, liver his-tology, and clinical progress. BMJ 1974;2:16-20.78. Bronspiegel-Weintrob N, Olivieri NF, Tyler B, Andrews DF, Freed-man MH, Holland FJ. Effect of age at the start of iron chelation therapy on gonadal function in b-thalassemia major. N Engl J Med 1990;323:713-9.79. Gabutti V, Piga A. Results of long-term iron-chelating therapy. Acta Haematol 1996;95:26-36.80. Porter JB. A risk-benefit assessment of iron-chelation therapy. Drug Saf 1997;17:407-21.81. Brittenham GM, Cohen AR, McLaren CE, et al. Hepatic iron stores and plasma ferritin concentration in patients with sickle cell anemia and thalassemia major. Am J Hematol 1993;42:81-5.82. Angelucci E, Giardini C, Brittenham GM, Lucarelli G. Hepatic iron concentration and body iron stores determined by quantitative phlebotomy in patients cured of thalassemia major by bone marrow transplantation. Blood 1997;90:Suppl 1:265a. abstract.83. Olynyk JK, O’Neill R, Britton RS, Bacon BR. Determination of he-patic iron concentration in fresh and paraffin-embedded tissue: diagnostic implications. Gastroenterology 1994;106:674-7.84. Brittenham GM, Farrell DE, Harris JW, et al. Magnetic-susceptibility measurement of human iron stores. N Engl J Med 1982;307:1671-5.85. Angelucci E, Giovagnoni A, Valeri G, et al. Limitations of magnetic

MEDICAL PROGRESS

Volume 341 Number 2 · 109

resonance imaging in measurement of hepatic iron. Blood 1997;90:4736-42.86. Giardini C. Treatment of b-thalassemia. Curr Opin Hematol 1997;4:79-87.87. Lucarelli G, Giardini C, Angelucci E. Bone marrow transplantation in thalassemia. In: Winter JN, ed. Blood stem cell transplantation. Boston: Kluwer Academic, 1997:305-15.88. Issaragrisil S, Visuthisakchai S, Suvatte V, et al. Transplantation of cord-blood stem cells into a patient with severe thalassemia. N Engl J Med 1995;332:367-9.89. Contu L, La Nasa G, Arras M, et al. Successful unrelated bone marrow transplantation in b-thalassemia. Bone Marrow Transplant 1994;13:329-31.90. Westgren M, Ringden O, Eik-Nes S, et al. Lack of evidence of perma-nent engraftment after in utero fetal stem cell transplantation in congenital hemoglobinopathies. Transplantation 1996;61:1176-9.91. Olivieri NF, Brittenham GM, Matsui D, et al. Iron-chelation therapy with oral deferiprone in patients with thalassemia major. N Engl J Med 1995;332:918-22.92. Hoffbrand AV, al-Refaie F, Davis B, et al. Long-term trial of defer-iprone in 51 transfusion-dependent iron overloaded patients. Blood 1998;91:295-300.93. Olivieri NF, Brittenham GM, McLaren CE, et al. Long-term safety and effectiveness of iron-chelation therapy with deferiprone in patients with thalassemia major. N Engl J Med 1998;339:417-23.94. Hoffbrand AV. Oral iron chelation. Semin Hematol 1996;33:1-8.95. Carthew P, Smith AG, Hider RC, Dorman B, Edwards RE, Francis JE. Potentiation of iron accumulation in cardiac myocytes during the treat-ment of iron overload in gerbils with the hydroxypyridinone iron chelator CP94. Biometals 1994;7:267-71.96. Wong A, Alder V, Robertson D, et al. Liver iron depletion and toxicity of the iron chelator deferiprone (L1, CP20) in the guinea pig. Biometals 1997;10:247-56.97. Nathan DG. An orally active iron chelator. N Engl J Med 1995;332:953-4. [Erratum, N Engl J Med 1995;332:1315.]98. Olivieri NF, Nisbet-Brown E, Srichairatanakool S, et al. Studies of iron excretion and non-transferrin-bound plasma iron (NTBP1) following a single infusion of hydroxyethyl starch-deferoxamine (HES-DFO): a new approach to iron chelation therapy. Blood 1996;88:Suppl 1:310a. ab-stract.99. Borgna-Pignatti C, Cohen AR. Evaluation of a new method of admin-istration of the iron chelating agent deferoxamine. J Pediatr 1997;130:86-8.100. Porter JB, Alberti D, Hassan I, et al. Subcutaneous depot desferriox-amine (CGH 749B): relationship of pharmacokinetics to efficacy and drug metabolism. Blood 1997;90:Suppl 1:265a. abstract.101. Olivieri NF. Reactivation of fetal hemoglobin in patients with b-thal-assemia. Semin Hematol 1996;33:24-42.102. Ley TJ, DeSimone J, Anagnou NP, et al. 5-Azacytidine selectively increases g-globin synthesis in a patient with b+ thalassemia. N Engl J Med 1982;307:1469-75.

103. Arruda VR, Lima CSP, Saad STO, Costa FF. Successful use of hy-droxyurea in b-thalassemia major. N Engl J Med 1997;336:964.104. Rigano P, Manfré L, La Galla R, et al. Clinical and hematologic response to hydroxyurea in a patient with Hb Lepore/b-thalassemia. He-moglobin 1997;21:219-26.105. Perrine SP, Ginder CD, Faller DV, et al. A short-term trial of buty-rate to stimulate fetal-globin–gene expression in the b-globin disorders.N Engl J Med 1993;328:81-6.106. Olivieri NF, Rees DC, Ginder GD, et al. Treatment of thalassemia major with phenylbutyrate and hydroxyurea. Lancet 1997;350:491-2.107. Rodgers GP, Rachmilewitz EA. Novel treatment options in the severe b-globin disorders. Br J Haematol 1995;91:263-8.108. Saxon BR, Rees D, Olivieri NF. Regression of extramedullary hae-mopoiesis and augmentation of fetal haemoglobin concentration during hydroxyurea therapy in b thalassaemia. Br J Haematol 1998;101:416-9.109. Sher GD, Ginder GD, Little JA, Wang SY, Dover G, Olivieri NF. Extended therapy with intravenous arginine butyrate in patients with b-hemoglobinopathies. N Engl J Med 1995;332:1606-10.110. Atweh GF, Sutton M, Nassif I, et al. Sustained induction of fetal he-moglobin by pulse butyrate therapy in sickle cell disease. Blood 1999;93:1790-7.111. Collins AF, Pearson HA, Giardina P, McDonagh KT, Brusilow SW, Dover GJ. Oral sodium phenylbutyrate therapy in homozygous beta thal-assemia: a clinical trial. Blood 1995;85:43-9.112. Cappellini MD, Graziadei G, Ciceri L, et al. Phase II open study of oral isobutyramide in patients with thalassemia intermedia. Blood 1996;88:Suppl 1:311a. abstract.113. Constantoulakis P, Josephson B, Mangahas L, et al. Locus control re-gion-Ag transgenic mice: a new model for studying the induction of fetal hemoglobin in the adult. Blood 1991;77:1326-33.114. Stamatoyannopoulos JA, Nienhuis AW. Therapeutic approaches to hemoglobin switching in treatment of hemoglobinopathies. Ann Rev Med 1992;43:497-521.115. McDonagh KT, Dover GJ, Donahue RE, et al. Hydroxyurea-induced HbF production in anemic primates: augmentation by erythropoietin, hematopoietic growth factors, and sodium butyrate. Exp Hematol 1992;20:1156-64.116. Pace B, Li Q, Peterson K, Stamatoyannopoulos G. a-Amino butyric acid cannot reactivate the silenced gamma gene of the beta locus YAC transgenic mouse. Blood 1994;84:4344-53.117. Kitsberg D, Selig S, Keshet I, Cedar H. Replication structure of the human b-globin gene domain. Nature 1993;366:588-90.118. Verma IM, Somia N. Gene therapy — promises, problems and pros-pects. Nature 1997;389:239-42.119. Higgs DR, Sharpe JA, Wood WG. Understanding a-globin gene ex-pression: a step towards effective gene therapy. Semin Hematol 1998;35:93-104.120. Shesely EG, Kim H-S, Shehee WR, Papayannopoulou T, Smithies O, Popovich BW. Correction of a human b S-globin gene by gene targeting. Proc Natl Acad Sci U S A 1991;88:4294-8.