574
`
`T H E NEW ENGLAND JOURNAL OF MEDIC I NE
`
`Sept. 1, 1994
`
`SURVIVAL IN MEDICALLY TREATED PATIENTS WITH HOMOZYGOUS JJ-THALASSEMIA
`
`NANCY F. OLIVIERI, M .D ., DAVID G. NATHAN, M .D ., j AMES H . M AcM1LLAN, M.Sc., ALAN S. WAYNE, M.D.,
`PETER P. Lru, M .D., ALLISON M cG EE, M .D ., M ARIE M ARTIN, R.N., GmEON K OREN, M.D.,
`AND ALAN R. C OHEN, M.D .
`
`Abstract Background. The prognosis of patients with
`homozygous f3-thalassemia (thalassemia major) has been
`improved by transfusion and iron-chelation therapy. We
`analyzed outcome and prognostic factors among patients
`receiving transfusions and chelation therapy who had
`reached the age at which iron-induced cardiac disease,
`the most common cause of death, usually occurs.
`Methods. Using the duration of life without the need
`for either inotropic or antiarrhythmic drugs as a measure of
`survival without cardiac disease, we studied 97 patients
`born before 1976 who were treated with regular transfu(cid:173)
`sions and chelation therapy. We used Cox proportional(cid:173)
`hazards analysis to assess the effect of prognostic factors
`and life-table analysis to estimate freedom from cardiac
`disease over time.
`Results. Of the 97 patients, 59 (61 percent) had no
`cardiac disease; 36 (37 percent) had cardiac disease, and
`18 of them had died. Univariate analysis demonstrated
`
`T HE prognosis of patients with transfu sion(cid:173)
`
`dependent homozygous /3-thalassemia (thalasse(cid:173)
`mia maj or) has been improved by regular transfusion
`and iron-chelation therapy .1 Before the introduction
`of therapy with deferoxamine, an iron-chela ting
`agen t, in the late l970s,2 iron overload from transfu(cid:173)
`sions was a frequent cause of morbidity and mortality
`in these patien ts. Death was often due to cardiac fail(cid:173)
`ure, which typically began before the patient reached
`20 years of age.3 Previous studies have suggested that
`deferoxamine therapy, begun early in life, prolongs
`survival without cardiac disease,._9 but follow-up was
`too short for unequivocal conclusions. The need for
`definitive information about the long-term prognosis
`of patients with thalassemia maj or has increased since
`allogeneic bone marrow transplantation emerged as
`an alternative treatment. 1
`0- 12
`In this report, we present the results of transfusion
`and chelation therapy in 97 patients with thalassemia
`major who were followed at three North American
`centers. The mean age of the group at the close of the
`study was 23 years, an age at which cardiac disease
`would be expected in most patients treated only with
`transfusions. 3 T he results demonstrate a markedly im-
`
`From the Hospital for Sick Children (N.F.O., J.H.M., A.M., G.K.) and To(cid:173)
`ronto Hospital (N. F.O., P.P.L.), University of Toronto, Toronto; the Division of
`Hematology- Oncology, Children's Hospital and the Dana- Farber Cancer Insti(cid:173)
`tute, and the Department of Pediatrics, Harvard Medical School, Boston
`(D.G. N., A.S.W.); and the Division of Hematology, Children's Hospital of
`Philadelphia, and lhe Department of Pediatrics, University of Pennsylvania
`School of Medicine, Philadelphia (M.M., A.R.C.). Address reprint requests to
`Dr. Olivieri at the Hacmoglobinopalhy Program, Division of Haematology/On(cid:173)
`cology, Hospital for Sick Children, 555 University Ave., Toronto, ON MSG
`IX8, Canada.
`Supported in part by lhe Medical Research Council of Canada, the Ontario
`Hean and Stroke foundation, General Clinical Research Center grants from the
`National Institutes of Hcallh (2 MOI RR02172-12 and MOI RR00240), and a
`contract from the Conunonweahh of Pennsylvania. Dr. Olivieri and Dr. Koren
`are Career Scientists of the Ontario Ministry of Health.
`
`that factors affecting cardiac disease-free survival were
`age at the start of chelation therapy (P< 0.001 ), the natural
`log of the serum ferritin concentration before chelation
`therapy began (P = 0.01 ). the mean ferritin concentration
`(P< 0.001 ), and the proportion of ferritin measurements
`exceeding 2500 ng per milliliter (P< 0.001). With stepwise
`Cox modeling, only the proportion of ferritin measure(cid:173)
`ments exceeding 2500 ng per milliliter affected cardiac
`disease-free survival (P< 0.001 ). Patients in whom less
`than 33 percent of the serum ferritin values exceeded
`2500 ng per milliliter had estimated rates of survival with(cid:173)
`out cardiac disease of 100 percent after 1 O years of chela(cid:173)
`tion therapy and 91 percent after 15 years.
`Conclusions. The prognosis for survival without cardi·
`ac disease is excellent for patients with thalassemia major
`who receive regular transfusions and whose serum ferritin
`concentrations remain below 2500 ng per milliliter with
`chelation therapy. (N Engl J Med 1994;331 :574-8.)
`
`proved prognosis for survival without cardiac disease
`in patients with thalassemia major who begin chela(cid:173)
`tion therapy before iron loading is severe and who
`comply with this treatment regimen.
`
`Study Design
`
`M ETHODS
`
`We examfoed su rvival without cardiac d isease, which was de(cid:173)
`fined as survival without the need for cardiac inotropic or antiar(cid:173)
`rhythmic med ication, in all 97 patients with thalassemia major born
`between 1954 and 1975 a nd treated at three centers: the Hospital for
`Sick Children in Toronto, C hildren 's Hospital in Boston, and the
`C hildren 's Hospital of Philadelphia. Pericarditis was not consid (cid:173)
`ered a n outcome measure; no patient in the stud y had this complica(cid:173)
`tion. The ethnic background of the patients was as follows: 53 of
`Ita lian origin, 30 of G reek origin, 5 of Chinese origin, 4 of Ind ian
`origin, 3 of Saudi Arabian origi n, and 1 each of Lebanese and
`Turkish o rigin. Mutations in the /3-globin gene had previously been
`cha racterized in only half th e patients. Begin ning in 1970, most
`patients received approximately 15 ml of packed red cells per kilo(cid:173)
`gram of body weight at each transfusion to maintain h emoglobin
`levels above 9.0 g per deciliter. They underwent splenectomy if th e
`volume of packed red cells exceeded 250 ml per kilogram per year.
`Begin ning in 1989, one patient in Boston received red cells separat(cid:173)
`ed according to density to enhance the collection of young erythro(cid:173)
`cytes (n eocytes) with prolonged survival in vivo. 13 Most patients
`administered dcferoxamine (50 to 75 mg per kilogram) to them(cid:173)
`selves as a nightly 10-to-12-hour subcutaneous infusion, using
`standard a mbulatory pumps. I n Philadelphia, most patients re(cid:173)
`ceived 2 g of deferoxamine each nigh t regardless of weight; some
`patients also received intermittent intravenous infusions of 4 to 12 g
`of deferoxamine. Beginning in 1989, 10 patients in Toronto received
`deferoxamine by continuous intravenous in fusion (50 to 80 mg per
`kilogram per 24 hours). 1• No patient had clinical manifestations of
`iron-related d iabetes mellitu s, h ypothyroidism, or hypoparath y(cid:173)
`roidism a t the start of deferoxa mine therapy. Patien ts were followed
`for a median of 12 years of deferoxamine treatment.
`At each center, all the patients were interviewed and examined by
`a staff physician a t intervals of o ne to six months. Serum ferritin
`concentrations were measured at similar intervals by a standard
`method .15 Cardiac evaluation included a n annual h istory a nd phys(cid:173)
`ical examination by a cardiologist and annu al chest radiography,
`electrocardiography, and resting echocardiography. Cardiac dis-
`
`The New England Journal of Medicine
`Downloaded from nejm.org at INFOTRIEVE on September 28, 2016. For personal use only. No other uses without permission.
` Copyright © 1994 Massachusetts Medical Society. All rights reserved.
`
`1 of 5
`
`Taro Pharmaceuticals, Ltd.
`Exhibit 1014
`
`

`

`Vol. 331 No. 9
`
`SURVIVAL WITH MEDICAL TREATMENT OF THALASSEMIA MAJOR
`
`575
`
`ease was considered to be present if a patient required inotropic
`drugs (including digoxin) for signs or symptoms of cardiac failure,
`therapy for substantial arrhythmias, or ooth.
`
`Statlatlcal Analyala
`
`For survival without cardiac disease, we considered the following
`prognostic factors: clinical center, sex, age at the start of treatment
`with deferoxamine, serum ferritin concentration before chelation
`therapy, mean serum ferrit.in concentration, proportion of ferritin
`measurements exceeding certain threshold values, and degree of
`reduction in the scrum ferritin concentration approximately one
`and two years after therapy began. The mean serum ferritin concen(cid:173)
`tration and the proportion of serum ferritin values exceeding a given
`threshold were used because the timing of serum ferritin measure(cid:173)
`ments was irregular. A window of IO to 18 months was used in
`estimating the degree of reduction in the serum ferritin concentra(cid:173)
`tion at approximately one and two years. The mean serum ferritin
`concentrations and the concentrations before chelation therapy
`were analyzed on a log scale to reduce the influence of very high
`values. Cox proportional-hazards analysis was used to investigate
`the effect of each prognostic factor on survival without cardiac dis(cid:173)
`ease. The Cox model with stepwise seleetion was then used to inves(cid:173)
`tigate the joint effect of the prognostic factors. A significance level
`of 5 percent was used for the entry of terms, and a level of 10
`percent was used for the removal of terms. P values are given for
`both univariate and stepwise analyses. The results of Cox modeling
`have been summarized with risk ratios for each variable, 95 per(cid:173)
`cent confidence intervals, and associated P values. Proportions of
`events are reported relative to the number of patients who could be
`evaluated. Life-table plots were used to assess disease-free survival
`over rime.
`BMDP-PC90 software programs were used to calculate life(cid:173)
`table estimates of survival without cardiac disease and for Cox
`modeling (University of California Press, Berkeley) . Data manip(cid:173)
`ulation was performed and descriptive statistics were calculated
`with the use of SAS software (version 6.07.01 , SAS Institute,
`Cary, N.C.).
`
`CHA1tACT£a1$TIC
`
`RE'sULTS
`Of 10 l patients initially enrolled in the study, 97
`could be evaluated. Two patients began treatment at
`or after the onset of cardiac disease, and data on se(cid:173)
`rum ferritin concentrations were unavailable for two
`other patients. Patients were followed for a median of
`12 years after beginning chelation
`therapy. The mea11 (±SO) age of
`the. patie.nts at the close of the study
`in June 1991 was 23±5 years.
`Table l summarizes the patients'
`clinical characteristics and out(cid:173)
`comes. Fifty-nine patients (61 per-
`cent} have no evidence of cardiac
`disease at this writing. Thirty-six
`(37 percent} had cardiac disease,
`and 18 of them (50 percent} had
`died. Two patients died of causes
`unrelated to cardiac disease; data
`on them were included as censored
`observaticms in the analysis. Pa(cid:173)
`tients without cardiac disease be(cid:173)
`gan treatment at an earlier age than
`those with cardiac disease, had
`lower mean serum fertitin concen(cid:173)
`trations before beginning deferox(cid:173)
`amine therapy, maintained lower
`mean ferritin concentrations during
`
`treatment, and had lower proportions of ferritin meas(cid:173)
`urements that exceeded the threshold value of 2500 ng
`per milliliter during treatment. This proportion be(cid:173)
`came the most important factor predicting disease-free
`survival, as discussed below. Transfusion require(cid:173)
`ments were similar in patients with and without cardi(cid:173)
`ac disease. The differences in cardiac disease-free sur(cid:173)
`vival among ethnic groups were attributable to the
`degree of iron loading.
`For the full cohort, the estimated survival without
`cardiac disease was 80 percent after 5 years of chela(cid:173)
`tion therapy, 65 percent after 10 years, and 55 percent
`after 15 years (Fig. l}. As shown in Table 2, factors
`influencing cardiac disease-free survival were age
`at the start of deferoxamine therapy, serum ferritin
`concentration before treatment, mean serum ferritin
`concentration during treatment, and proportion of
`ferritin measurements that exceeded 2500 ng per mil(cid:173)
`liliter during treatment. With stepwise Cox model(cid:173)
`ing, only a higher proportion of ferritin measure(cid:173)
`ments exceeding this threshold value was associated
`with poorer cardiac disease-free survival (risk ratio,
`19.l; 95 percent confidence interval, 6.3 to 58.1;
`P<0.001).
`Because some patients had more serum ferritin
`measurements than others, we also evaluated the pro(cid:173)
`portion of measurements exceeding 2500 ng per milli(cid:173)
`liter, using only the first serum ferritin determination
`each year. The relation between this variable and car(cid:173)
`diac disease-free survival was unchanged.
`For patients in whom less than 33 percent of ferritin
`measurements exceeded 2500 ng per milliliter, the es(cid:173)
`timated survival without cardiac disease was 100 per(cid:173)
`cent after 10 years of deferoxamine therapy and 91
`percent after 15 years (Fig. 2). In contrast, for patients
`in whom 33 to 67 percent of ferritin measurements
`exceeded 2500 ng per milliliter, the estimated disease(cid:173)
`free survival was 48 percent after 10 and 15 years of
`
`Table 1. Clinical Characteristics and Outcome.*
`
`NO. OF PATIENTS
`WHO COULD BE
`EVALUATED
`
`97
`97
`
`89
`
`93
`
`93
`
`CWllCAL STATUS
`
`DISEASE•
`fUE
`(N ~ S9)
`
`ALIVE WYTH
`CAkDlAC
`DlSEASE
`(N = 18)
`
`0£AD OF
`CAlDIAC
`DISEASE
`( N 3 18)
`
`OE.AO OF
`UNkELATED
`CAUSES
`(N = 2)
`
`51 /49
`10.5±5.6
`
`33/67
`14.4± 5.7
`
`61/39
`14.8±3.6
`
`100/0
`9.0± 1.4
`
`3416± 2225
`2740
`
`4829± 3772 4556 ±2167
`3450
`3960
`
`2695±2439
`1880
`
`5497±3367 4592± 1658 4504±4252
`4414
`5572
`
`0.32±0.35
`0.20
`
`0.77±0.34 0.83±0.23 0.53±0.62
`0.99
`0.93
`
`Female/male (%)
`Mean age at stan of chela(cid:173)
`tion therapy (yr)
`Serum ferritin level before
`chelation (ng/ml)
`Mean
`Median
`Serum ferritin level during
`chelation (ngtml)
`Mean
`Median
`Proportion of measurements
`>2500 ng/ml during
`chelation
`Mean
`Median
`
`•Plus- minus values arc means :!:SD.
`
`The New England Journal of Medicine
`Downloaded from nejm.org at INFOTRIEVE on September 28, 2016. For personal use only. No other uses without permission.
` Copyright © 1994 Massachusetts Medical Society. All rights reserved.
`
`2 of 5
`
`Taro Pharmaceuticals, Ltd.
`Exhibit 1014
`
`

`

`576
`
`THE NEW ENGLAND JOURNAL OF MEDICINE
`
`Sept. I , 199~
`
`therapy. For patients in whom more than 67 percent
`of ferritin measurements exceeded 2500 ng per milli(cid:173)
`liter, the estimated disease-free survival was 38 per(cid:173)
`cent after 10 years of therapy and 18 percent after 15
`years.
`
`Table 2. Effect of Prognostic Factors on Survival without Cardiac
`Disease.
`
`PROGNOSTIC FACTOR
`
`No. OF
`PATIEN'TS
`
`RtsX RATIO
`(95 ... Cl)*
`
`P VALUE
`
`UNtVARlATB STEPWISE
`
`DISCUSSION
`During the past 10 years, strong evidence of im(cid:173)
`proved survival without cardiac disease in patients
`with thalassemia major has accumulated,.._9 and the
`prognosis appears particularly good for children with
`thalassemia born since the current treatment became
`widely available.6 Nonetheless, there are still cases of
`iron-related illness and death, even in patients who
`apparently complied with deferoxamine therapy. 16
`This study describes the outcome of the Jong-term
`treatment of 97 patients with homozygous /3-thalasse(cid:173)
`mia. Using the end point of survival without cardiac
`disease, which we defined as survival without inotrop(cid:173)
`ic or antiarrhythmic therapy, we identified factors re(cid:173)
`lated to the success or failure of an iron-chelation pro(cid:173)
`gram. Some of our patients, although they did not
`need cardiac therapy, may have had early cardiac dys(cid:173)
`function that could have been detected by methods
`such as radionuclide angiography. The end points in
`this study may therefore have overestimated disease(cid:173)
`free survival. Evaluation of the effect of transfusion
`and chelation therapy on subclinical cardiac dysfunc(cid:173)
`tion must await the development of tests with suffi(cid:173)
`cient sensitivity and specificity to predict accurately
`the later development of clinical cardi~c disease. We
`could not ethically compare treated and untreated pa(cid:173)
`tients during the same period. However, one of the
`last reports of survival in patients with thalassemia
`before the era of iron-chelation therapy described the
`onset of iron-related cardiac failure in 26 of 41 pa(cid:173)
`tients (63 percent) at a mean age of 16 years; more
`than half of the affected patients died of cardiac dis(cid:173)
`ease within one year of its onset.3 Pericarditis was also
`frequent, although the relation of this complication to
`iron overload remains uncertain .3 Our results indicate
`a markedly improved outlook for patients who receive
`
`Q) en
`~1.00~
`i:S
`0 i 0.75
`
`----..___ ........
`
`--.
`
`0 a. e 0.00 ..Lr---.----.----r---..---..---r- --r--1
`a..
`
`4
`
`8
`6
`10
`Years of Chelation Therapy
`
`0.50
`
`0.25
`
`(.)
`5
`0
`.£:.
`·~
`c
`0
`t:
`
`0
`
`2
`
`12
`
`14
`
`16
`
`0.47
`
`I .SO (0.6-4.0)
`2.00 (0.6-6.5)
`0 .80 (0.4-1.6)
`0.51
`1.10 ( 1.07-1.2) <0.001
`
`2.00 (1.2-3.8)
`
`0.01
`
`1.00 (0.8-1.S)
`0.68
`0.06
`1.40 (l.0- 2.0)
`3.20 (2.0-5. 1) <0.001
`
`0.87
`0.89
`0.56
`0.65
`
`0.08
`
`0.80
`0.51
`0.46
`
`19.10 (6.3-58 .1) <0.001
`
`<0.001
`
`Centert
`
`97
`
`97
`97
`
`89
`
`89
`86
`93
`
`Sex
`Age at stan of chelation
`therapy
`Seium ferritinf
`Natural log of level before
`chelation
`Reduction ai I yr
`Reduction at 2 yr
`Natural log of mean concen-
`tration
`Proportion of measurements 93
`>2500 ng/ml
`-a deoo~s confidence interval.
`tTwo dummy variables were used for analysis of d4ta from throe centers.
`*5erom ferritin concenirations befon: chewion thenlpy were not available for eight patien1s.
`Patients were excloded if cardiac disease developed before a measurement was obtained.
`
`chelation therapy sufficient to maintain a reduced se(cid:173)
`rum ferritin concentration over a long period. The
`favorable outcome of patients who began chelation
`therapy in early childhood emphasizes the benefits of
`treatment with deferoxamine.
`A sustained reduction in iron, as measured by the
`proportion of serum ferritin measurements that did
`not exceed 2500 ng per milliliter, emerged as the most
`important factor in the survival without cardiac dis(cid:173)
`ease among the patients with homozygous /3-thalasse(cid:173)
`mia we studied. The estimated disease-free survival 15
`years after the beginning of ~helation therapy was 91
`percent among patients in whom fewer than one third
`of ferritin measurements exceeded 2500 ng per millili(cid:173)
`ter. Even patients who began chelation therapy later
`in life but maintained a reduction in their iron stores
`had a good prognosis. Conversely, failure to prevent
`the accumulation of excess iron or to remove large
`stores of tissue iron was associated with a poor prog(cid:173)
`nosis at any age. For example, the probability of sur(cid:173)
`vival after 15 years of chelation therapy was less than
`20 percent if more than 67 percent of a patient's ferri(cid:173)
`tin measurements exceeded 2500 ng per milliliter.
`Although a single measurement of serum ferritin
`may be an imprecise asses~ment of total iron stores, 17
`serial measurements may be useful in predicting the
`likelihood of iron-induced cardiac disease. Hepatitis C
`infection, the most common viral complication of
`long-term transfusion therapy, may further elevate the
`serum ferritin concentration, complicating its inter(cid:173)
`pretation. We were unable to analyze the effect of
`hepatitis C infection on serum ferritin concentrations
`because assays to detect hepatitis C antibodies were
`unavailable during most of the period of investigation.
`A similar effect of hepatitis C would be expected in
`patients with and without cardiac disease, all of whom
`began transfusion therapy before donor blood was
`screened for this infectious agent.
`Our study could not determine the influence of var-
`
`Figure 1. Survival without Cardiac Disease during Chelation Ther(cid:173)
`apy in 97 Patients with Thalassemia Major.
`
`The New England Journal of Medicine
`Downloaded from nejm.org at INFOTRIEVE on September 28, 2016. For personal use only. No other uses without permission.
` Copyright © 1994 Massachusetts Medical Society. All rights reserved.
`
`3 of 5
`
`Taro Pharmaceuticals, Ltd.
`Exhibit 1014
`
`

`

`Vol. 331 No. 9
`
`SURVIVAL WITH MEDICAL TREATMENT OF THALASSEMlA MAJOR
`
`577
`
`<1> en
`<1> en
`i5
`(.)
`al 0.75
`~
`co
`C..)
`:; 0.50
`0
`
`al 1.00 "L::K. • • ·-·-~
`""'!
`-·----·
`'-,.._.~
`
`0.25
`
`= ·~
`c
`0 t:
`8. 0.00
`e a..
`
`Factors other than survival without cardiac disease
`(for transfusion and chelation therapy) and survival
`without thalassemia (for bone marrow transplanta(cid:173)
`tion) may influence the choice between these treat(cid:173)
`ment options. Cost may be a factor. The estimated
`cost (in 1990 dollars) of bone marrow transplantation
`for hemoglobinopathy was $173,250 in 1991.18 Addi(cid:173)
`tional costs can be anticipated during at least the first
`five years after transplantation. 19 The cost of transfu(cid:173)
`sion and chelation therapy in a patient who weighs 30
`kg was about $32,000 per year in 1991, with more than
`60 percent of the cost attributable to chelation thera(cid:173)
`py.18 The cost of medical therapy would be expected
`to increase as the patient grew and required more
`blood and more deferoxamine. The cost of transfusion
`therapy in some centers is now substantially higher
`than this estimate and can exceed $30,000 per year in
`patients who receive two units of packed red cells ev(cid:173)
`ery three weeks (unpublished data). Although the
`short-term costs of bone marrow transplantation and
`transfusion and chelation therapy may be similar, the
`higher continuing costs of transfusion and chelation
`may make them the more expensive option.
`This study was not designed to establish the optimal
`serum ferritin concentration required to prevent iron(cid:173)
`related cardiac disease or other complications of iron
`overload. In practice, physicians stress the importance
`to their patients of reducing body iron to the lowest
`possible level. It is possible that maintaining the se(cid:173)
`rum ferritin concentration considerably below 2500 ng
`per milliliter may be optimal on a long-term basis,
`but physicians must weigh the benefit of treatment
`against the toxic effects of deferoxamine in patients
`with reduced iron burdens.20·23 Measurements of body
`iron stores, in addition to the serum ferritin concentra(cid:173)
`tion, may be useful in determining the duration and
`•25
`intensity of chelation therapy.24
`
`REFERENCES
`
`l. Cohen A. Management of iron overload in the pediatric patie nt. Hematol
`Oocol Clio North Am 1!>87;1:521-44.
`2. Propper RD, Cooper B, Rufo RR, ct al . Continuous subcutaneous adminis(cid:173)
`tration of deferoxamine in patients with iron overload. N Engl J Med
`1977;297:418-23.
`3 . Engle MA, Erlandson M, Smith CH. Late cardiac complications of chronic,
`severe, refractory anemia with bcmocbromatosis. Circulation I 964;30:698-
`705.
`4 . Wolfe L, Olivieri NF, Sallan D, ct al. Prevention of cardiac disease by
`subcutaneous defcroxamine in patients with thalassemia major. N Engl
`J Med 1985;312:1600-3.
`5. Brittenham G, Nienhuis AW. Dcsferrioxamine use protects against hcan
`disease and death from transfusional iron overload in thalassemia major.
`Blood 1988;72:Suppl:56a. abstract.
`6. Zurlo MG, De Stefano P, Borgna-Pignatti C , et al. Survival and causes of
`death in thalassaemia major. Lancet 1989;2 :27· 30.
`7. Aldouri MA, Wonkc B, Hoffbrand AV, ct al. High incidence ofcardiomy·
`opathy in beta·thalassemia patients receiving regular transfusion and iron
`chelation: reversal by intensified chelation. Acta Haematol 1990;84: 113· 7 .
`8. Olivieri NF, McGee A, Liu P, Koren G , Freedman MH, Benson L. Cardiac
`disease- free survival in patients with thalassemia major treated with subcu(cid:173)
`taneous defcroxamine: an update on the Toronto cohort. Ann N Y Acad Sci
`.
`1990;612:585-6.
`9. Ebler:s KH, Giardina PJ, Lesser ML, Engle MA, Hilgartner MW. Prolonged
`survival in patients with beta·thalassemia major treated with defcroxamine.
`J Pediatr 1991;118:540-5.
`IO. Thomas ED, Buckner CD, Sanders JE, ct al. Marrow transplantation for
`thalassaemia. Lancet 1982;2:227·9 .
`
`0
`
`2
`
`6
`
`8
`4
`Years of Chelation Therapy
`
`10
`
`12
`
`14
`
`16
`
`Figure 2. Survival without Cardiac Disease According to the Pro(cid:173)
`portion of Serum Ferritin Measurements Greater Than 2500 ng
`per Milliliter.
`The circles show cardiac disease-free survival among patients in
`whom less than 33 percent of ferritin measurements exceeded
`2500 ng per milliliter; squares show survival among patients In
`whom 33 to 67 percent of ferritin measurements exceeded 2500
`ng per milliliter; and triangles show survival among patients in
`whom more than 67 percent of ferritin measurements exceeded
`2500 ng per milliliter.
`
`ious mutations of the J3-globin gene on survival with(cid:173)
`out cardiac disease, because the mutations had been
`characterized in only half the patients. However, any
`effect of genotype on disease-free survival would pre(cid:173)
`sumably be mediated through transfusion require(cid:173)
`ments, which did not differ in patients with and those
`without cardiac disease.
`Chelation therapy has improved survival without
`cardiac disease in patients with thalassemia major,
`but bone marrow transplantation from an HLA(cid:173)
`identical donor has resulted in thalassemia-free sur(cid:173)
`vival in many patients. Although we did not com(cid:173)
`pare patients treated with transfusion and chelation
`therapy with patients who underwent bone marrow
`transplantation, our results provide valuable infor(cid:173)
`mation for patients, families, and clinicians faced
`with a choice between the two forms of treatment.
`Patients with thalassemia major who begin chelation
`therapy in early childhood and most of whose se(cid:173)
`rum ferritin measurements are below 2500 ng per
`milliliter are clinically similar to patients with class I
`disease as defined by the Pesaro bone marrow-trans(cid:173)
`plantation group. 11
`12 Both groups of patients have
`•
`benefited from good medical care from an early age.
`The rate of survival without cardiac disease in our
`study (91 percent after 15 years of chelation therapy)
`compares favorably with the 3-year disease-free sur(cid:173)
`vival of 85 to 93 percent for patients with class I
`disease treated with bone marrow transplantation. 12
`The poor long-term prognosis of patients who have
`large iron stores because they cannot get or do not
`comply with deferoxamine therapy may argue for
`bone marrow transplantation, even in the face of
`hepatic enlargement or fibrosis , which negatively
`affect the outcome of transplantation.11
`
`The New England Journal of Medicine
`Downloaded from nejm.org at INFOTRIEVE on September 28, 2016. For personal use only. No other uses without permission.
` Copyright © 1994 Massachusetts Medical Society. All rights reserved.
`
`4 of 5
`
`Taro Pharmaceuticals, Ltd.
`Exhibit 1014
`
`

`

`578
`
`THE :'\E\V ENGLA:-<D JOUR~!\L Of :\IEDICINE
`
`Sept. I, 1994
`
`11. Lucarelli G, Galimberti M. Polchi P, ct al . Bone marrow transplantation in
`patients with thalassemia. I' Engl J Med 1990;322:417-21.
`12. Lucarelli G, Galimberti M. Polchi P. ct al. Marrow transplantation in pa(cid:173)
`tients with thalassemia responsive to iron chelation therapy. N Engl J Med
`1993;329:840-4 ,
`13. Simon TL. Sohmcr P, Nelson El. Extended survival of ococytes produced
`by a new system. Transfusion 1989:29:221 -5 .
`14. Olivieri NF. Bcrriman AM. Tyler BJ , Davis SA. Francombe WH. Liu PP.
`Reduction in tissue iron stores with a new regimen of continuous ambulatory
`intravenous dcfcroxaminc. Am J Hcmatol 1992:41 :61-3 .
`15. Addison GM. Beamish :VIR. Hales CN. Hodgkins M. Jacobs A , Llewellin
`P. An immunoradiometric assay for fcrritin in the scrum of normal subjects
`and patients with iron deficiency and iron overload. J Clin Pathol 1972;25:
`326-9.
`16. Lerner N. Rici I'. Ricnnan !'.Johnson L. Piomclli S. Chelation therapy and
`cardiac status in older patients with thalasscmia major. Am J Pcdiatr Hcma(cid:173)
`tol Oncol 1990:12:56·60.
`17. Brittenham GM. Cohen AR. McLaren CE, ct al. Hepatic iron stores and
`plasma ferritin concentration in patients with sickle cell anemia and thalassc·
`mia major. Am J Hcmatol 1993;42:81· 5 .
`
`18. Kirkpatrick DV. Barrios NJ. Humbert JH . Bone marrow transplantation for
`sickle cell anemia. Semin Hcmatol 1991 :28:240-3.
`19. Welch HG, Larson EB. Cost effectiveness of bone marrow transplantation
`in acute nonlymphocytic leukemia. N Engl J Med 1989:321:807-12.
`20. Porter JB. Jaswon MS. Huehns cR. East CA. Hazell JWP. Desferrioxamine
`mot0xicity: evaluation of risk factors in thalassaemic patients and guidelines
`for safe dosage. Br J Hacmatol 1989:73:403·9.
`21 . O li vieri NF. Buncic JR. Chew E. et al. Visual and auditory neurotoxicity in
`patients receiving subcutaneous deferoxamine infusions. N Engl J Med
`1986:314:869-73.
`22 . De Virgi liis S. Congia M. Frau F. et al. Dcferoxaminc-induced growth
`retardation in patients with thah1ssemia major. J Pediatr 1988:113:661-
`9.
`23. Hankamp MJ. Babyn PS . Olivieri NF. Spinal dcforrnities in dcfcroxamine(cid:173)
`treatctl homozygous beta-thalassemia major patienK Pediatr Radial 1993:
`23:525-8 .
`fosburg MT. Nathan DG. Treatment of Cooley·s anemia. Blood 1990:76:
`435-44.
`25 . Brittenham GM. Farrell DE. Harris JW. et al. Magnetic-susceptibility meas(cid:173)
`urement or human iron stores. N Engl J Med 1982:307: 1671-5.
`
`24.
`
`Antelope Cart)'on, Aril;o1w
`
`ScoTT VALENT, :\1.D.
`
`The New England Journal of Medicine
`Downloaded from nejm.org at INFOTRIEVE on September 28, 2016. For personal use only. No other uses without permission.
` Copyright © 1994 Massachusetts Medical Society. All rights reserved.
`
`5 of 5
`
`Taro Pharmaceuticals, Ltd.
`Exhibit 1014
`
`