T h e ne w e ngl a nd jou r na l o f m e dic i ne
`
`original article
`
`The Complement Inhibitor Eculizumab
`in Paroxysmal Nocturnal Hemoglobinuria
`
`Peter Hillmen, M.B., Ch.B., Ph.D., Neal S. Young, M.D., Jörg Schubert, M.D.,
`Robert A. Brodsky, M.D., Gerard Socié, M.D., Ph.D., Petra Muus, M.D., Ph.D.,
`Alexander Röth, M.D., Jeffrey Szer, M.B., B.S., Modupe O. Elebute, M.D.,
`Ryotaro Nakamura, M.D., Paul Browne, M.B., Antonio M. Risitano, M.D., Ph.D.,
`Anita Hill, M.B., Ch.B., Hubert Schrezenmeier, M.D., Chieh-Lin Fu, M.D.,
`Jaroslaw Maciejewski, M.D., Ph.D., Scott A. Rollins, Ph.D.,
`Christopher F. Mojcik, M.D., Ph.D., Russell P. Rother, Ph.D., and Lucio Luzzatto, M.D.
`
`A BS TR AC T
`
`BACKGROUND
`We tested the safety and efficacy of eculizumab, a humanized monoclonal antibody
`against terminal complement protein C5 that inhibits terminal complement activa-
`tion, in patients with paroxysmal nocturnal hemoglobinuria (PNH).
`
`METHODS
`We conducted a double-blind, randomized, placebo-controlled, multicenter, phase 3
`trial. Patients received either placebo or eculizumab intravenously; eculizumab was
`given at a dose of 600 mg weekly for 4 weeks, followed 1 week later by a 900-mg
`dose and then 900 mg every other week through week 26. The two primary end
`points were the stabilization of hemoglobin levels and the number of units of
`packed red cells transfused. Biochemical indicators of intravascular hemolysis and
`the patients’ quality of life were also assessed.
`
`RESULTS
`Eighty-seven patients underwent randomization. Stabilization of hemoglobin levels
`in the absence of transfusions was achieved in 49% (21 of 43) of the patients as-
`signed to eculizumab and none (0 of 44) of those assigned to placebo (P<0.001).
`During the study, a median of 0 units of packed red cells was administered in the
`eculizumab group, as compared with 10 units in the placebo group (P<0.001). Ecu-
`lizumab reduced intravascular hemolysis, as shown by the 85.8% lower median area
`under the curve for lactate dehydrogenase plotted against time (in days) in the eculi-
`zumab group, as compared with the placebo group (58,587 vs. 411,822 U per liter;
`P<0.001). Clinically significant improvements were also found in the quality of life,
`as measured by scores on the Functional Assessment of Chronic Illness Therapy-
`Fatigue instrument (P<0.001) and the European Organization for Research and
`Treatment of Cancer Quality of Life Questionnaire. Of the 87 patients, 4 in the ecu-
`lizumab group and 9 in the placebo group had serious adverse events, none of which
`were considered to be treatment-related; all these patients recovered without se-
`quelae.
`
`CONCLUSIONS
`Eculizumab is an effective therapy for PNH. (ClinicalTrials.gov number, NCT00122330.)
`
`From Leeds General Infirmary, Leeds,
`United Kingdom (P.H., A.H.); National
`Heart, Lung, and Blood Institute, Bethes-
`da, MD (N.S.Y.); Saarland University Med-
`ical School, Homburg-Saarland, Germany
`(J. Schubert); Sidney Kimmel Comprehen-
`sive Cancer Center at Johns Hopkins, Bal-
`timore (R.A.B.); Hôpital Saint-Louis and
`INSERM, Paris (G.S.); Radboud Univer-
`sity Medical Center, Nijmegen, the Neth-
`erlands (P.M.); University Hospital of
` Essen, Essen, Germany (A.R.); Royal Mel-
`bourne Hospital, Parkville, Melbourne,
`Australia (J. Szer); St. George Hospital,
`London (M.O.E.); City of Hope National
`Medical Center and Beckman Research
`Institute, Duarte, CA (R.N.); St. James’
`Hospital, Trinity College Dublin, Dublin
`(P.B.); Federico II University, Naples
`(A.M.R.); the Institute of Transfusion Med-
`icine, University Hospital, Ulm, Germany
`(H.S.); Cleveland Clinic Florida, Weston,
`FL (C.-L.F.); Taussig Cancer Center, Cleve-
`land Clinic Foundation, Cleveland (J.M.);
`Alexion Pharmaceuticals, Cheshire, CT
`(S.A.R., C.F.M., R.P.R.); and Istituto To-
`scano Tumori, Florence, Italy (L.L.). Ad-
`dress reprint requests to Dr. Hillmen at
`the Department of Haematology, Leeds
`General Infirmary, Great George St., Leeds
`LS1 3EX, United Kingdom, or at peter.
`hillmen@nhs.net.
`
`N Engl J Med 2006;355:1233-43.
`Copyright © 2006 Massachusetts Medical Society.
`
`n engl j med 355;12 www.nejm.org september 21, 2006
`
`1233
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on January 13, 2017. For personal use only. No other uses without permission.
`
` Copyright © 2006 Massachusetts Medical Society. All rights reserved.
`
`1
`
`AMG1012
`
`

`

`T h e ne w e ngl a nd jou r na l o f m e dic i ne
`
`Paroxysmal nocturnal hemoglobin-
`
`uria (PNH), an uncommon form of hemo-
`lytic anemia, results from the clonal expan-
`sion of hematopoietic stem cells that have somatic
`mutations in the X-linked gene PIG-A.1,2 PIG-A mu-
`tations cause an early block in the synthesis of
`glycosylphosphatidylinositol (GPI) anchors, which
`tether many proteins to the cell surface. Conse-
`quently, the blood cells in patients with PNH have
`a partial deficiency (type II) or a complete defi-
`ciency (type III) of GPI-linked proteins.
`Intravascular hemolysis is a prominent feature
`of PNH and is the consequence of the absence of
`the GPI-linked complement regulatory protein
`CD59.3,4 CD59 blocks the formation of the termi-
`nal complement complex (also called the mem-
`brane-attack complex) on the cell surface, thereby
`preventing erythrocyte lysis and in vitro platelet
`activation.5-8 Excessive or persistent intravascular
`hemolysis in patients with PNH causes anemia,
`hemoglobinuria, and complications related to the
`presence of plasma free hemoglobin, including
`thrombosis, abdominal pain, dysphagia, erectile
`dysfunction, and pulmonary hypertension.9-12 In-
`deed, the symptoms in PNH are often dispropor-
`tionate to the degree of anemia. Many patients
`with this disease are dependent on transfusions.
`Currently, there is no therapy that effectively re-
`duces intravascular hemolysis or improves the
`symptoms in patients with PNH.
`Eculizumab (Soliris, Alexion Pharmaceuticals)
`is a humanized monoclonal antibody directed
`against the terminal complement protein C5.13 In
`a preliminary, 12-week, open-label clinical study
`involving 11 patients with PNH, eculizumab re-
`duced intravascular hemolysis and the patients’
`transfusion requirements.14 However, this two-
`center, uncontrolled study did not have a control
`group or predefined criteria for the administra-
`tion of a transfusion, such as a predefined hemo-
`globin level at which transfusions were adminis-
`tered or a prespecified number of units of packed
`red cells for a given hemoglobin level.
`We report the results of the phase 3 Transfu-
`sion Reduction Efficacy and Safety Clinical Inves-
`tigation, a Randomized, Multicenter, Double-Blind,
`Placebo-Controlled, Using Eculizumab in Parox-
`ysmal Nocturnal Hemoglobinuria (TRIUMPH)
`study, which investigated whether eculizumab
`stabilized hemoglobin levels and reduced trans-
`fusion requirements in 87 transfusion-dependent
`patients with PNH during 6 months of treatment.
`
`Intravascular hemolysis and the quality of life
`were also assessed.
`
`Me thods
`
`Patients
`The trial consisted of a 2-week screening period,
`an observation period of up to 3 months, and a
`26-week treatment period. Patients 18 years of age
`or older who had received at least four transfu-
`sions during the previous 12 months were eligi-
`ble. A PNH type III erythrocyte proportion of 10%
`or more, platelet counts of at least 100,000 per
`cubic millimeter, and lactate dehydrogenase levels
`that were at least 1.5 times the upper limit of the
`normal range were also required. Concomitant
`administration of erythropoietin, immunosup-
`pressive drugs, corticosteroids, coumarins, low-
`molecular-weight heparins, iron supplements, and
`folic acid were permitted, provided that the doses
`were constant before and throughout the study.
`Because persons who have a genetic deficiency of
`terminal complement proteins have an increased
`risk of neisserial infections, all patients were vac-
`cinated against Neisseria meningitidis with the use
`of locally approved vaccines. The protocol was
`approved by the institutional review board at
`each center, and all patients gave written informed
`consent.
`Patients receiving transfusions who had a
`mean hemoglobin level greater than 10.5 g per
`deciliter before transfusion during the 12 months
`before entry into the study or who had received
`another investigational drug within 30 days be-
`fore the first visit were excluded. Patients who
`had a complement deficiency, an active bacterial
`infection, or a history of meningococcal disease
`and those who had undergone bone marrow
`transplantation were also excluded.
`An individualized transfusion algorithm was
`calculated for each patient on the basis of the
`history of transfusions during the previous 12
`months; the written algorithm documented the
`number of units of packed red cells to be trans-
`fused for given hemoglobin values and served as
`a prospectively determined guide for transfusion
`during the observation and treatment periods.
`Before randomization, eligible patients were ob-
`served for up to 13 weeks. Patients who did not
`require a transfusion during the observation pe-
`riod were considered ineligible. A transfusion ad-
`ministered to a patient who had a hemoglobin
`
`1234
`
`n engl j med 355;12 www.nejm.org september 21, 2006
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on January 13, 2017. For personal use only. No other uses without permission.
`
` Copyright © 2006 Massachusetts Medical Society. All rights reserved.
`
`2
`
`

`

`eculizumab in paroxysmal nocturnal hemoglobinuria
`
`level of 9 g per deciliter or less with symptoms
`or 7 g per deciliter or less with or without symp-
`toms qualified the patient for the study (qualify-
`ing transfusion) and established the hemoglobin
`set point. This set point was required for the pri-
`mary efficacy variable and was individualized for
`each patient.
`
`Study Design
`Randomization was performed centrally in a 1:1
`ratio without blocking and with stratification ac-
`cording to the number of units of packed red cells
`transfused during the past year; patients were as-
`signed, by means of an interactive voice–response
`system, to receive either placebo or eculizumab
`within 10 days after the administration of the
`qualifying transfusion. Patients received infusions
`of 600 mg of eculizumab or placebo every week
`(±2 days) for 4 weeks, followed 1 week (±2 days)
`later by 900 mg of eculizumab or placebo, and
`then by a maintenance dose of 900 mg of eculi-
`zumab or placebo every 2 weeks (±2 days) through
`week 26.
`
`global health status and functioning scales and
`lower scores on the symptom scales and single-
`improvement)16;
`item measures
`indicating
`changes in lactate dehydrogenase levels from
`baseline through week 26; and the presence of
`thrombosis. Other prespecified measurements
`included the pharmacokinetics, pharmacody-
`namics, and immunogenicity of eculizumab. The
`time to the first transfusion during the treat-
`ment period and the proportion of PNH type III
`blood cells were assessed.
`
`Safety
`Adverse events related to study infusions and vi-
`tal signs (assessed at each of the 17 study visits
`during treatment), the results of biochemical analy-
`ses and blood counts (assessed at 9 visits), and find-
`ings on electrocardiograms (assessed at 3 visits)
`were documented. Adverse events were coded with
`the use of preferred terms from the Medical Dic-
`tionary for Regulatory Activities (MedDRA) (www.
`msso.org/MSSOWeb/index.htm) and tabulated as
`incidence rates in the two study groups.
`
`Clinical Efficacy
`The two primary end points were the stabiliza-
`tion of hemoglobin levels, defined as a hemoglo-
`bin value that was maintained above the level at
`which the qualifying transfusion was adminis-
`tered, in the absence of transfusions during the
`26-week treatment period, and the number of
`units of packed red cells transfused during that
`period. The trigger for the administration of trans-
`fusions during the study remained unchanged:
`patients received transfusions when they had symp-
`toms resulting from anemia and their hemoglo-
`bin levels reached the individualized, predeter-
`mined set point. Prespecified secondary end points
`included transfusion independence; hemolysis, as
`measured by the lactate dehydrogenase value for
`the area under the curve from baseline to 26 weeks;
`and changes in the level of fatigue, as assessed
`from baseline to 26 weeks with the use of the
`Functional Assessment of Chronic Illness Therapy-
`Fatigue (FACIT-Fatigue) instrument (scores can
`range from 0 to 52, with higher scores indicating
`improvement in fatigue).15 Prespecified explor-
`atory analyses included assessment of the quality
`of life with the use of the European Organization
`for Research and Treatment of Cancer Quality of
`Life Questionnaire (EORTC QLQ-C30) (scores can
`range from 0 to 100, with higher scores on the
`
`Statistical Analysis
`The planned sample size of 75 patients provided
`the study with a statistical power of 82%, at an
`alpha level of 0.05, to detect an increase of 35 per-
`centage points (i.e., a change from 20% to 55%)
`in the rate of the stabilization of hemoglobin lev-
`els and a reduction in the median number of units
`of packed red cells transfused from 6 to 2 (±2).
`For the two primary end points, the analyses
`were performed according to the intention-to-
`treat principle with the use of data on all 87 pa-
`tients who underwent randomization; stabiliza-
`tion of hemoglobin levels was analyzed with the
`use of Fisher’s exact test, and the total number of
`units of packed red cells transfused was analyzed
`with the use of the Wilcoxon rank-sum test. To
`assess the effect of treatment on whether or not
`transfusions were required, Fisher’s exact test
`was used. The log-rank test was used to compare
`the time to the first transfusion in the two groups.
`The area under the curve for lactate dehydroge-
`nase was compared between the two groups with
`the use of the Wilcoxon rank-sum test.
`Fatigue was assessed according to the scoring
`guidelines for the FACIT-Fatigue instrument.17
`The assessment of the quality of life was based
`on the EORTC QLQ-C30 scores and was conduct-
`ed as described previously.18 Changes in scores
`
`n engl j med 355;12 www.nejm.org september 21, 2006
`
`1235
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on January 13, 2017. For personal use only. No other uses without permission.
`
` Copyright © 2006 Massachusetts Medical Society. All rights reserved.
`
`3
`
`

`

`T h e ne w e ngl a nd jou r na l o f m e dic i ne
`
`on the FACIT-Fatigue instrument and the EORTC
`QLQ-C30 instrument from baseline through week
`26 were analyzed with the use of a mixed model,
`with baseline scores as the covariate, treatment
`and time as fixed effects, and the patient identi-
`fier as a random effect. Changes in the levels of
`lactate dehydrogenase, PNH type III erythrocytes,
`and hemoglobin from baseline through week 26
`were analyzed with the use of the same mixed
`model. All reported P values are two-sided and
`were not adjusted for multiple analyses. The inci-
`dence rates of adverse events were compared with
`the use of Fisher’s exact test. No interim analyses
`were performed, and blinding regarding the re-
`sults was maintained until the end of the study.
`The authors and the sponsor were jointly re-
`sponsible for the trial design and the development
`of the protocol. Data were collected by an elec-
`tronic case-report form with the use of InForm
`software (version 4.0, Phase Forward) and were
`analyzed by the sponsor. The decision to publish
`
`Table 1. Baseline Characteristics of the Patients.
`
`Characteristic
`
`Sex — no.
`
`Male
`
`Female
`
`Age — yr
`
`Median
`
`Range
`
`Duration of PNH — yr
`
`Median
`
`Range
`Reticulocyte counts — per mm3
`Median
`
`Range
`
`Placebo Group
`(N = 44)
`
`Eculizumab Group
`(N = 43)
`
`15
`
`29
`
`35
`
`20
`
`23
`
`41
`
`18–78
`
`20–85
`
`9.2
`
`0.5–38.5
`
`4.3
`
`0.9–29.8
`
`204,400
`
`206,600
`
`45,400–556,200
`
`40,200–570,400
`
`History of aplastic anemia — no. (%)
`
`12 (27)
`
`History of myelodysplastic syndrome
`— no. (%)
`
`0
`
`History of thrombosis — no. (%)
`
`8 (18)
`
`Total no. of thrombotic events
`
`Use of erythropoietin — no. (%)
`
`Use of cyclosporine — no. (%)
`
`Use of anticoagulant agents (couma-
`rins or heparins) — no. (%)
`
`Use of corticosteroids or androgenic
`steroids — no. (%)
`
`11
`
`0
`
`1 (2)
`
`11 (25)
`
`12 (27)
`
`6 (14)
`
`2 (5)
`
`9 (21)
`
`16
`
`3 (7)
`
`1 (2)
`
`21 (49)
`
`12 (28)
`
`the trial data and final decisions on the content
`of the manuscript rested with Dr. Hillmen in
`consultation with the other authors. The manu-
`script was prepared by Dr. Hillmen, with sub-
`stantial review and comments by the other au-
`thors. All authors had access to the primary data
`and take responsibility for the veracity and com-
`pleteness of the data reported.
`
`R esults
`
`Patients’ Characteristics
`Of a total of 115 patients with PNH who under-
`went screening, 87 (35 men and 52 women) at 34
`sites in the United States, Canada, Europe, and
`Australia who received a qualifying transfusion,
`met the inclusion criteria and did not meet any of
`the exclusion criteria were randomly assigned to
`eculizumab (43 patients) or placebo (44 patients)
`between October 2004 and June 2005. At each of
`16 sites one patient underwent randomization, at
`each of 6 sites two patients underwent random-
`ization, and at each of 12 sites 3 or more patients
`underwent randomization. There were no signifi-
`cant differences in the baseline characteristics of
`the patients in the two groups (Table 1).
`Of 87 patients who underwent randomization,
`85 completed the trial (see the Supplementary
`Appendix, available with the full text of this ar-
`ticle at www.nejm.org). Two patients in the ecu-
`lizumab group did not complete the trial, one
`because traveling to the study center was incon-
`venient and the other because of pregnancy; these
`patients were included in the analyses. Ten pa-
`tients in the placebo group discontinued infu-
`sions because of a perceived lack of efficacy but
`remained in the study for monitoring, as pre-
`specified in the protocol, and were included in
`the analyses.
`
`Pharmacokinetics and Pharmacodynamics
`In 42 of 43 patients in the eculizumab group, a
`900-mg dose of eculizumab every 2 weeks (±2
`days) completely blocked serum hemolytic activ-
`ity, as assessed by a presensitized erythrocyte
`hemolytic assay,14 throughout the study period.
`In one patient, therapeutic trough levels of eculi-
`zumab were not maintained.
`
`Effect on Hemolysis
`The effect of eculizumab on chronic intravascu-
`lar hemolysis was demonstrated by an immediate
`
`
`
`1236
`
`n engl j med 355;12 www.nejm.org september 21, 2006
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on January 13, 2017. For personal use only. No other uses without permission.
`
` Copyright © 2006 Massachusetts Medical Society. All rights reserved.
`
`4
`
`

`

`eculizumab in paroxysmal nocturnal hemoglobinuria
`
`(1 week) and sustained decrease in lactate dehy-
`drogenase levels (Fig. 1A). In the eculizumab
`group, the mean (±SE) lactate dehydrogenase
`level decreased from 2199.7±157.7 U per liter at
`
`baseline to 327.3±67.6 U per liter at 26 weeks,
`whereas in the placebo group the levels remained
`elevated, with values of 2258.0±154.8 U per liter
`at baseline and 2418.9±140.3 U per liter at 26
`
`3000
`
`2500
`
`2000
`
`1500
`
`1000
`
`500
`
`0
`
`Level (U/liter)
`
`Lactate Dehydrogenase
`
`Screening
`period
`
`No. of Patients
`Placebo group
`Eculizumab group
`
`A
`
`B
`
`Placebo
`
`P<0.001
`
`Eculizumab
`
`0
`
`2
`
`4
`
`6
`
`8
`
`10
`
`12
`14
`Week
`
`16
`
`18
`
`20
`
`22
`
`24
`
`26
`
`44
`43
`
`43
`43
`
`44
`43
`
`44
`43
`
`43
`43
`
`43
`43
`
`43
`43
`
`42
`41
`
`43
`41
`
`44
`40
`
`42
`40
`
`44
`41
`
`Eculizumab
`
`P<0.001
`
`Placebo
`
`60
`
`50
`
`40
`
`30
`
`20
`
`10
`
`0
`
`PNH Type III Erythrocytes (%)
`
`Screening
`period
`
`No. of Patients
`Placebo group
`Eculizumab group
`
`0
`
`2
`
`4
`
`6
`
`8
`
`10
`
`12
`14
`Week
`
`16
`
`18
`
`20
`
`22
`
`24
`
`26
`
`44
`43
`
`44
`42
`
`42
`43
`
`42
`41
`
`Figure 1. Levels of Lactate Dehydrogenase and PNH Type III Erythrocytes during Treatment with Eculizumab.
`Panel A shows the degree of intravascular hemolysis according to the mean levels of lactate dehydrogenase from
`baseline (week 0) to week 26 in the two study groups. The dashed line indicates the upper limit of the normal range
`for lactate dehydrogenase (normal range, 103 to 223 U per liter). In the eculizumab group the mean level of lactate
`dehydrogenase was reduced to just above the upper limit of the normal range at week 26; of 41 patients in this
`group who completed the study, 15 had levels within the normal range. In the placebo group, all patients had levels
`at least five times above the upper limit of normal at week 26. Panel B shows the mean proportion of PNH type III
`erythrocytes in patients in the two groups. Screening occurred up to 3 months before week 0. P values are from
`a mixed analysis-of-covariance model from baseline through week 26. I bars indicate the standard error.
`
`n engl j med 355;12 www.nejm.org september 21, 2006
`
`1237
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on January 13, 2017. For personal use only. No other uses without permission.
`
` Copyright © 2006 Massachusetts Medical Society. All rights reserved.
`
`5
`
`

`

`T h e ne w e ngl a nd jou r na l o f m e dic i ne
`
`weeks (P<0.001). The median value of the areas
`under the curve for lactate dehydrogenase plot-
`ted against time (in days) was 85.8% lower in the
`eculizumab group than in the placebo group
`(58,587 vs. 411,822 U per liter; P<0.001). A sec-
`ond biochemical measure of hemolysis, the serum
`level of aspartate aminotransferase, also showed
`significant improvement with eculizumab, as com-
`pared with placebo (data not shown).
`The reduction in intravascular hemolysis in the
`eculizumab group resulted in an increase in PNH
`type III erythrocytes (Fig. 1B) from a mean of
`28.1±2.0% at baseline to 56.9±3.6% at week 26.
`The proportion of PNH type III erythrocytes in
`patients in the placebo group remained constant
`(35.7±2.8% before treatment and 35.5±2.8% at
`26 weeks, P<0.001 for the comparison with the
`eculizumab group and the placebo group). The
`proportion of PNH type III granulocytes and
`monocytes did not change significantly between
`the two groups (data not shown).
`
`Clinical Efficacy
`Primary End Points
`The two primary efficacy end points were the
`stabilization of hemoglobin levels and the num-
`ber of units of packed red cells transfused. At the
`end of the treatment period, 49% of patients in
`the eculizumab group (21 of 43) had levels of he-
`moglobin that remained above the prespecified
`set point (median, 7.7 g per deciliter for both
`groups) in the absence of transfusions, whereas
`
`stabilization of hemoglobin levels did not occur in
`any patient in the placebo group (P<0.001) (Table
`2). By week 26, the median number of units of
`packed red cells transfused per patient was 0 in
`the eculizumab group and 10 in the placebo
`group (P<0.001), whereas the mean number of
`units of packed red cells transfused was 3.0±0.7
`and 11.0±0.8, respectively. In the 6-month period
`before the study, the median number of units of
`packed red cells transfused per patient was 9.0 in
`the eculizumab cohort and 8.5 in the placebo
`cohort, and the mean number of units of packed
`red cells transfused was 9.6±0.6 and 9.7±0.7, re-
`spectively. The mean hemoglobin levels changed
`from 10.0±0.2 g per deciliter and 9.7±0.2 g per
`deciliter in the eculizumab group and the placebo
`group, respectively, at baseline to 10.1±0.2 g per
`deciliter and 8.9±0.2 g per deciliter, respectively,
`at week 26 (P<0.001, by mixed-model analysis).
`The median time to the first transfusion was
`significantly longer in eculizumab-treated patients
`than in patients who received placebo (P<0.001)
`(Fig. 2). Transfusion independence was achieved
`in 51% of patients in the eculizumab group (22
`of 43) and 0% of those in the placebo group
`(0 of 44, P<0.001). By week 26, the total number
`of units of packed red cells transfused was 131
`in the eculizumab group and 482 in the placebo
`group (Table 2). By contrast, during the 6 months
`before the study, the total number of units trans-
`fused was 413 in the eculizumab group and 417
`in the placebo group.
`
`Table 2. Stabilization of Hemoglobin Levels and the Number of Units of Packed Red Cells Transfused
`during Treatment.*
`
`Primary End Point
`
`Before Treatment†
`
`During Treatment
`
`P Value
`
`Placebo
`Group
`
`Eculizumab
`Group
`
`Placebo
`Group
`
`Eculizumab
`Group
`
`Patients with stabilized hemoglobin levels (%)
`
`NA
`
`NA
`
`0
`
`49
`
`<0.001‡
`
`Packed red cells transfused
`(units/patient)
`
`Median
`
`Interquartile range
`
`Mean
`
`Total
`
`8.5
`
`7–12.5
`
`9.7±0.7
`
`417
`
`9.0
`
`6–12
`
`9.6±0.6
`
`413
`
`10
`
`6–16
`
`0
`
`0–6
`
`11.0±0.8
`
`3.0±0.7
`
`482
`
`131
`
`<0.001§
`
`* Plus–minus values are means ±SE. NA denotes not applicable.
`† Transfusion data obtained during 12 months before treatment were normalized to a value equivalent to the value for
`a 6-month period.
`‡ The P value is for the comparison between groups during treatment, calculated with the use of a two-tailed Fisher’s ex-
`act test.
`§ The P value is for the comparison between groups during treatment, calculated with the use of the Wilcoxon rank-sum
`test.
`
`1238
`
`n engl j med 355;12 www.nejm.org september 21, 2006
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on January 13, 2017. For personal use only. No other uses without permission.
`
` Copyright © 2006 Massachusetts Medical Society. All rights reserved.
`
`6
`
`

`

`eculizumab in paroxysmal nocturnal hemoglobinuria
`
`Eculizumab
`
`Placebo
`
`P<0.001
`
`100
`
`90
`
`80
`
`70
`
`60
`
`50
`
`40
`
`30
`
`20
`
`10
`
`Patients Remaining Transfusion Independent (%)
`
`0
`
`0
`
`2
`
`4
`
`6
`
`8
`
`10
`
`12
`14
`Week
`
`16
`
`18
`
`20
`
`22
`
`24
`
`26
`
`No. at Risk
`Placebo group
`Eculizumab group
`
`44
`43
`
`44
`41
`
`36
`41
`
`30
`39
`
`23
`37
`
`13
`36
`
`7
`36
`
`6
`32
`
`3
`32
`
`1
`31
`
`0
`27
`
`0
`27
`
`0
`26
`
`0
`26
`
`0
`24
`
`0
`22
`
`Figure 2. Kaplan–Meier Curves for the Time to the First Transfusion during Treatment.
`The P value for the comparison of times to the first transfusion between the two groups was calculated by the log-
`rank test.
`
`Quality of Life
`Assessments of the quality of life were performed
`with the use of two instruments, the FACIT-Fatigue
`instrument and the EORTC QLQ-C30 instrument.
`Patients in the eculizumab group had a mean
`increase (improvement) in scores on the FACIT-
`Fatigue instrument of 6.4±1.2 points from baseline
`to week 26, whereas in the placebo group the mean
`score decreased by 4.0±1.7 points during this pe-
`riod, for a total difference between the two groups
`of 10.4 points (Fig. 3). A mixed-model analysis of
`covariance was performed that showed a signifi-
`cant difference between the two groups (P<0.001).
`With respect to the EORTC QLQ-C30 instru-
`ment, the eculizumab group had significant
`improvements in scores on the scale for global
`health status, on all five scales for functioning,
`on two of three symptom scales, and on three of
`six single-item measures, as compared with the
`placebo group (P≤0.01 for each scale and mea-
`sure) (Table 3).
`
`Safety
`No patients died during the study. Serious ad-
`verse events were reported in 13 patients: 4 in the
`
`eculizumab group and 9 in the placebo group
`(Table 4). No serious adverse events were consid-
`ered to be treatment-related; all these patients re-
`covered without sequelae. The most common ad-
`verse events reported in the eculizumab group were
`headache, nasopharyngitis, back pain, and nausea.
`Headache and back pain occurred more frequent-
`ly in the eculizumab group than in the placebo
`group. The number of headaches that occurred was
`similar in the two groups after the first 2 weeks
`of therapy. There were no significant differences
`in the incidence rates between the two groups for
`any reported adverse event. A single thrombosis
`occurred in a patient in the placebo group.
`One patient in each of the two groups had de-
`tectable levels of antibodies against eculizumab.
`The levels were low, were detected at a single
`visit, and in the patient receiving eculizumab,
`the antibodies did not affect complement inhi-
`bition.
`
`Discussion
`
`Patients with PNH have chronic intravascular he-
`molysis with acute exacerbations. Anemia and the
`
`n engl j med 355;12 www.nejm.org september 21, 2006
`
`1239
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on January 13, 2017. For personal use only. No other uses without permission.
`
` Copyright © 2006 Massachusetts Medical Society. All rights reserved.
`
`7
`
`

`

`T h e ne w e ngl a nd jou r na l o f m e dic i ne
`
`Eculizumab
`
`Placebo
`
`0
`
`2
`
`4
`
`6
`
`8
`
`10
`
`12
`
`14
`
`16
`
`18
`
`20
`
`22
`
`24
`
`26
`
`8
`
`46
`
`2 0
`
`
`
`–2
`
`–4
`
`–6
`
`Change from Baseline in Fatigue Score
`
`No. of Patients
`Placebo group
`Eculizumab group
`
`41
`43
`
`39
`41
`
`41
`43
`
`41
`43
`
`40
`43
`
`Week
`
`40
`42
`
`36
`36
`
`39
`41
`
`Figure 3. Change in Fatigue Scores from Baseline to Week 26.
`Week 0 (baseline) represents the end of the observation period, within 10 days after the patient’s receipt of the
`qualifying blood transfusion. Fatigue scores are therefore higher (indicating less fatigue) at week 0 than at the initial
`screening visits (data not shown). Values for the change from baseline (dashed line) represent means. A positive
`change from baseline indicates an improvement in fatigue and a negative change from baseline indicates a worsen-
`ing in fatigue. I bars indicate the standard error.
`
`need for transfusions to sustain hemoglobin lev-
`els occur frequently, as does deterioration of the
`patient’s quality of life. In this study, in approxi-
`mately half the patients treated with eculizumab,
`the end points of stabilization of hemoglobin lev-
`els and transfusion independence were reached,
`whereas none of the patients in the placebo group
`reached either of these end points. The median
`time to the first transfusion was 4 weeks in the
`placebo group and more than 6 months in the
`eculizumab group. The overall rate of transfu-
`sion was reduced by 73% in the eculizumab group.
`Even among patients receiving eculizumab in
`whom transfusion independence was not reached,
`the number of units of packed red cells transfused
`was reduced by 44%, as compared with patients
`in the placebo group (data not shown).
`Intravascular hemolysis is central to the oc-
`currence of serious coexisting conditions in
`patients with PNH and contributes to the risk of
`death among these patients.9,12 Lactate dehydro-
`genase, a biochemical marker of hemolysis, was
`immediately reduced from approximately 10 times
`the upper limit of the normal range to normal
`
`levels or to just above normal levels in all patients
`in the eculizumab group. Residual low-level he-
`molysis in some patients despite terminal-com-
`plement blockade may be caused by an inherent
`decrease in the survival of PNH type III erythro-
`cytes19 or may be due to the fact that these cells
`are opsonized with C3b, which mediates extra-
`vascular clearance through the reticuloendothe-
`lial system.
`Before treatment with eculizumab, the hemo-
`globin levels were maintained by transfusion.
`Therefore, the stabilization of hemoglobin levels
`with a concomitant cessation of or reduction in
`the number of transfusions indicates an increase
`in endogenous erythrocyte mass. The reduction in
`hemolysis with eculizumab results in a new
`steady-state hemoglobin level, as determined by a
`balance of the underlying bone marrow dysfunc-
`tion, the increased half-life of PNH erythrocytes
`because of eculizumab therapy, and the new level
`of transfusions (if any) required.
`For most patients with PNH, the quality of life
`is impaired, and the impairment has been attrib-
`uted not only to anemia but also to excessive in-
`
`1240
`
`n engl j med 355;12 www.nejm.org september 21, 2006
`
`The New England Journal of Medicine
`
`Downloaded from nejm.org on January 13, 2017. For personal use only. No other uses without permission.
`
` Copyright © 2006 Massachusetts Medical Society. All rights reserved.
`
`8
`
`

`

`eculizumab in paroxysmal nocturnal hemoglobinuria
`
`Table 3. Change in the Quality of Life during Treatment.*
`
`Scale
`
`Global health status scale
`
`Functioning scales
`
`Role
`
`Social
`
`Cognitive
`
`Physical
`
`Emotional
`
`Symptom scales
`
`Fatigue
`
`Pain
`
`Nausea and vomiting
`
`Single-item measures
`
`Dyspnea
`
`Loss of appetite
`
`Insomnia
`
`Financial difficulties
`
`Constipation
`
`Diarrhea
`
`Mean Change in Score from Baseline
`to Week 26†
`
`Placebo Group
`
`Eculizumab Group
`
`Absolute Difference
`
`P Value‡
`
`−8.5
`
`−6.9
`
`2.0
`
`−6.1
`
`−3.5
`
`−3.7
`
`10.0
`
`5.3
`
`2.8
`
`8.9
`
`3.3
`
`4.9
`
`0.0
`
`0.0
`
`5.7
`
`10.9
`
`17.9
`
`16.7
`
`7.9
`
`9.4
`
`7.5
`
`−16.9
`
`−12.3
`
`−0.4
`
`−7.9
`
`−10.3
`
`−7.9
`
`−10.3
`
`−6.3
`
`4.8
`
`19.4
`
`24.8
`
`14.7
`
`14.0
`
`12.9
`
`11.2
`
`26.9
`
`17.6
`
`3.2
`
`16.8
`
`13.6
`
`12.8
`
`10.3
`
`6.3
`
`0.9
`
`<0.001
`
`<0.001
`
`0.003
`
`0.002
`
`<0.001
`
`0.008
`
`<0.001
`
`0.002
`
`0.06
`
`<0.001
`
`<0.001
`
`0.01
`
`0.19
`
`0.20
`
`0.15
`
`* The quality of life was assessed with the EORTC QL