Eculizumab for treating patients with paroxysmal nocturnal hemoglobinuria
Abstract
Background
Paroxysmal nocturnal hemoglobinuria (PNH) is a chronic, not malignant, disease of the hematopoietic stem cells, associated with significant morbidity and mortality. It is a rare disease with an estimated incidence of 1.3 new cases per one million individuals per year. The treatment of PNH has been largely empirical and symptomatic, with blood transfusions, anticoagulation, and supplementation with folic acid or iron. Eculizumab, a biological agent that inhibits complement cascade, was developed for preventing hemolytic anemia and severe thrombotic episodes.
Objectives
To assess the clinical benefits and harms of eculizumab for treating patients with paroxysmal nocturnal hemoglobinuria (PNH).
Search methods
We conducted a comprehensive search strategy. We searched the Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library 2014, Issue 5), Ovid MEDLINE (from 1946 to 15 May 2014), EMBASE (from 1980 to 25 June 2014), and LILACS (from 1982 to 25 June 2014). We did not apply any language restrictions.
Selection criteria
We included randomized controlled trials (RCTs) irrespective of their publication status or language. No limits were applied with respect to period of follow‐up. We excluded quasi‐RCTs. We included trials comparing eculizumab with placebo or best available therapy. We included any patient with a confirmed diagnosis of PNH. Primary outcome was overall survival.
Data collection and analysis
We independently performed a duplicate selection of eligible trials, risk of bias assessment, and data extraction. We estimated risk ratios (RRs) and 95% confidence interval (CIs) for dichotomous outcomes, and mean differences (MDs) and 95% CIs for continuous outcomes. We used a random‐effects model for analysis.
Main results
We identified one multicenter (34 sites) phase III RCT involving 87 participants. The trial compared eculizumab versus placebo, and was conducted in the US, Canada, Europe, and Australia with 26 weeks of follow‐up. This small trial had high risk of bias in many domains (attrition and selective reporting). It was sponsored by a pharmaceutical company. No patients died during the study. By using the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (scores can range from 0 to 100, with higher scores on the global health status and functioning scales indicating improvement), the trial showed improvement in health‐related quality of life in patients treated with eculizumab (mean difference (MD) 19.4, 95% CI 8.25 to 30.55; P = 0.0007; low quality of evidence). By using the Functional Assessment of Chronic Illness Therapy Fatigue instrument (scores can range from 0 to 52, with higher scores indicating improvement in fatigue), the trial showed a reduction in fatigue (MD 10.4, 95% CI 9.97 to 10.83; P = 0.00001; moderate quality of evidence) in the eculizumab group compared with placebo. Eculizumab compared with placebo showed a greater proportion of patients with transfusion independence: 51% (22/43) versus 0% (0/44); risk ratio (RR) 46.02, 95% CI 2.88 to 735.53; P = 0.007; moderate quality of evidence; and withdrawal for any reason: 4.7% (2/43) versus 22.72% (10/44); RR 0.20, 95% CI 0.05 to 0.88; P = 0.03; moderate quality of evidence. Due to the low rate of events observed, the included trial did not show any difference between eculizumab and placebo in terms of serious adverse events: 9.3% (4/43) versus 20.4% (9/44); RR 0.15, 95% CI 0.15 to 1.37; P = 0.16; low quality of evidence. We did not observe any difference between intervention and placebo for the most frequent adverse events. One participant receiving placebo showed an episode of thrombosis. The trial did not assess overall survival, transformation to myelodysplastic syndrome and acute myelogenous leukemia, or development or recurrence of aplastic anemia on treatment.
Authors' conclusions
This review has detected an absence of evidence for eculizumab compared with placebo for treating paroxysmal nocturnal hemoglobinuria (PNH), in terms of overall survival, nonfatal thrombotic events, transformation to myelodysplastic syndrome and acute myelogenous leukemia, and development and recurrence of aplastic anemia on treatment. Current evidence indicates that compared with placebo, eculizumab increases health‐related quality of life and increases transfusion independence. During the execution of the included trial, no patients died. Furthermore, the intervention seems to reduce fatigue and withdrawals for any reason. The safety profile of eculizumab is unclear. These conclusions are based on one small trial with risk of attrition and selective reporting bias.
Therefore, prescription of eculizumab for treating patients with PNH can neither be supported nor rejected, unless new evidence from a large high quality trial alters this conclusion. Therefore, we urge the reader to interpret the trial results with much caution. Future trials on this issue should be conducted according to the SPIRIT statement and reported according to the CONSORT statement by independent investigators, and using the Foundation of Patient‐Centered Outcomes Research recommendations.
PICOs
Plain language summary
Eculizumab for treating patients with paroxysmal nocturnal hemoglobinuria
Review question
We reviewed the evidence about the effects of eculizumab for treating patients with paroxsymal nocturnal hemoglobinuria.
Background
Paroxysmal nocturnal hemoglobinuria is a disorder of the hematopoietic stem cells (a cell that can self renew and differentiate into one or more cell types). It is characterized by episodes of intravascular hemolysis (destruction of red blood cells), and chronic hemolytic anemia. The intravascular destruction of red blood cells involves clinical findings in gastrointestinal, cardiovascular, pulmonary, cerebral, and urogenital systems, as well as clotting disorders.
The treatment of paroxysmal nocturnal hemoglobinuria has been largely empirical and symptomatic, with packed red blood cell transfusions, anticoagulation, and supplementation with folic acid or iron. Many different pharmacological interventions that are used for treating this medical disorder are not standardized. Eculizumab is a newly available biological agent for preventing hemolytic anemia and severe clotting episodes.
Study characteristics
We identified one study that included a limited number of patients comparing eculizumab with placebo. The study was published in 2006, and was conducted in the US, Canada, Europe, and Australia with 26 weeks of follow‐up.
Key results
No patients died during the performance of this single study. The study showed a moderate improvement in the quality of life in patients treated with eculizumab. In addition, eculizumab reduced fatigue and the number of patients that withdrew from the study for any reason. Eculizumab showed a higher proportion of patients with transfusion independence. There was no difference between eculizumab and placebo in terms of adverse events, probably due to the low rate of events observed during the study. The trial did not assess other relevant outcomes such as overall survival, transformation to myelodysplastic syndrome and acute myelogenous leukemia, or development or recurrence of aplastic anemia on treatment.
Quality of evidence
The confidence in the results is moderate to low. The study had limitations in its design and execution, and was sponsored by the manufacturer of the drug that was assessed. Moreover, the limited number of patients included in the study led to imprecise results. Larger studies should provide more information about the effect of eculizumab in patients with paroxsymal nocturnal hemoglobinuria.
This plain language summary is current as of May 2014.
Authors' conclusions
Summary of findings
| Eculizumab compared with placebo for paroxysmal nocturnal hemoglobinuria | ||||||
| Patient or population: patients with paroxysmal nocturnal hemoglobinuria | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of Participants | Quality of the evidence | Comments | |
| Assumed risk1 | Corresponding risk | |||||
| Placebo | Eculizumab | |||||
| Overall survival ‐ not measured | See comment | See comment | Not estimable | 87 | See comment | This outcome was not measured in the included study |
| All‐cause mortality | See comment | See comment | Not estimable | 87 | See comment | No patients died during the execution of the included study. The small sample size of the included trial does not allow to make judgments about the quality of evidence |
| Health‐related quality of life | The mean change from baseline in the health‐related quality of life score in the control group was | The mean change from baseline in the health‐related quality of life score in the intervention group was | 87 | ⊕⊕⊝⊝ | ||
| Fatigue | The mean change from baseline in the fatigue score in the control group was | The mean change from baseline in the fatigue score in the intervention group was | 87 | ⊕⊕⊕⊝ | ||
| Adverse events (serious and nonserious) | 205 per 1000 | 92 per 1000 | RR 0.45 | 87 | ⊕⊕⊝⊝ | |
| Transfusion independence | 20 per 10006 | 920 per 1000 | RR 46.02 | 87 | ⊕⊕⊕⊝ | |
| Withdrawal for any reason | 227 per 1000 | 45 per 1000 | RR 0.20 | 87 | ⊕⊕⊕⊝ | |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1 Assumed risk is based on the risks for the control group in the included trial. | ||||||
Background
For a glossary of medical terms, see Appendix 1.
Description of the condition
Paroxysmal nocturnal hemoglobinuria (PNH) is a disorder of the hematopoietic stem cells, but it is not a malignant disease (Parker 2009a; Parker 2012; Pu 2011). The first two cases were reported by William Gull and Paul Strübing between 1866 and 1882, respectively (Rosse 1980; Wilmanns 1982). In 1925, Enneking introduced the term 'paroxysmal nocturnal hemoglobinuria' (Brodsky 2009b). The term 'nocturnal' is misleading, as the hemolysis can occur at any time (Brodsky 2008a).
PNH arises as a consequence of nonmalignant clonal expansion of one or more hematopoietic stem cells that have acquired a somatic mutation of the X chromosome gene called phosphatidylinositol glycan class A (PIGA) (Brodsky 2006; Brodsky 2008a; Parker 2012). The protein encoded by PIGA is essential for synthesis of the glycosyl phosphatidylinositol (GPI) moiety that serves as the membrane anchor for a functionally diverse group of cellular proteins. As a consequence of mutant PIGA, all GPI proteins are deficient on progeny of affected stem cells (Parker 2012; Risitano 2012). Decay accelerating factor (CD‐55) and membrane inhibitor of reactive lysis (CD59) are complement regulatory proteins which are anchored to blood red cells surface (Thurman 2104). These proteins interact with complement proteins, mainly C3b and C4b, and inhibit the convertase complexes thereby halting prolonged activation. The deficiency of CD55 and CD59 in PNH therefore results in prolonged and uncontrolled activation of the complement pathways, resulting in complement‐mediated intravascular hemolysis (Parker 2005).
Diagnosis of PNH is based on the following criteria (Parker 2005).
-
Evidence of a population of peripheral blood cells (erythrocytes, granulocytes, or preferably both) deficient in glycosylphosphatidylinositol–anchored proteins detected by flow cytometric analysis.
-
Complete blood count, reticulocyte count, serum concentration of lactate dehydrogenase, bilirubin (fractionated), and haptoglobin.
-
Bone marrow aspirate, biopsy, and cytogenetics.
Currently, flow cytometry to detect populations of glycosylphosphatidylinositol‐anchored protein‐deficient cells is firmly established as the method of choice for diagnosis and monitoring paroxysmal hemoglobinuria (Borowitz 2010; Richards 2000).
PNH is a rare disease with an estimated incidence of 1.3 new cases per one million individuals per year (Borowitz 2010; Brodsky 2008a; Brodsky 2008b).
According to the International Paroxysmal Nocturnal Hemoglobinuria Interest Group (I‐PIG), PNH is classified into three categories:
-
classic PNH;
-
PNH in the setting of another specified bone marrow disorder (e.g. PNH/aplastic anemia, or PNH/refractory anemia‐myelodysplastic syndrome); and
-
PNH‐subclinical, in the setting of another specified bone marrow disorder (e.g. PNH‐subclinical/aplastic anemia) (Parker 2005; Parker 2009a).
Classic PNH
Affected patients have clinical evidence of intravascular hemolysis (reticulocytosis, abnormally high concentration of serum lactate dehydrogenase and indirect bilirubin, and abnormally low concentration of serum haptoglobin), but without evidence of another defined bone marrow abnormality. A cellular marrow with erythroid hyperplasia and normal or near‐normal morphology, but without nonrandom karyotypic abnormalities, is related with classic PNH (Parker 2005).
PNH in the setting of another specified bone marrow disorder
This subcategory have clinical and laboratory evidence of hemolysis showing concomitantly, or have had a history of, a defined underlying marrow abnormality. Bone marrow analysis and cytogenetics are used to determine if PNH arose in association with aplastic anemia, myelodysplastic syndrome, or other myelopathy (e.g. myelofibrosis). Nonrandom karyotypic abnormalities that are associated with a specific bone marrow abnormality may contribute diagnostically (e.g. abnormalities of chromosomes 5q, 7, and 20q are associated with myelodysplastic syndrome) (Parker 2005).
PNH‐subclinical in the setting of another specified bone marrow disorder
The patients in this subcategory have no clinical or laboratory evidence of hemolysis. Small populations of glycosyl phosphatidylinositol–anchored protein–deficient hematopoietic cells (peripheral blood erythrocytes, granulocytes, or both) are detected by very sensitive flow cytometric analysis. It is observed in association with bone marrow failure syndromes, particularly aplastic anemia and refractory anemia‐myelodysplastic (Parker 2005).
Nonhematological clinical findings
Patients with PNH show clinical findings in gastrointestinal, cardiovascular, pulmonary, cerebral, and urogenital systems, as well as clotting disorders which are mediated by the consumption of nitric oxide (Rother 2005; Savage 2007). It is associated with a hypercoagulable state (thrombosis) (al‐Hakim 1993; Audebert 2005; Barbui 2010; Dunphy 1994; Gayer 2001; Inafuku 1993). The main nonhematological clinical findings include acute or chronic renal failure (Chow 2001; de Charry 2012; Guasch 2010; Hillmen 2010; Jackson 1992; Nair 2008; Sechi 1988), pulmonary hypertension (Heller 1992; Misztal 2011), and an increasing risk for splanchnic vein thrombosis syndrome called Budd‐Chiari (Graham 1996; Hauser 2003; Jain 2010; Jimenez 1999; Hoekstra 2009; Torres 2010; Yin 2009). Visceral thrombosis, cerebrovascular events and pulmonary embolism predict a poor outcome (Ziakas 2008). The exact reason for an increase of thrombosis risk in patients with PNH is unknown (Brodsky 2009b; van Bijnen 2012a). However, a major role of complement activation has been suggested to explain this clinical finding (van Bijnen 2012b).
Prognosis (overall survival)
PNH is a chronic disorder associated with significant morbidity and mortality (Harris 1999; Hernandez‐Campo 2008; Rachidi 2010). The overall survival at 10 years after diagnosis with PNH has been variously estimated to be 65% (Socié 1996), 77.6% (Ge 2012), and 68% (Tudela 1993).
Description of the intervention
The treatment of PNH has been largely empirical and symptomatic, with blood transfusions, anticoagulation, and supplementation with folic acid or iron (Luzzatto 2011; Röth 2011). These interventions are mainly aimed to alleviate anemia and thrombotic episodes. The interventions include pharmacological and nonpharmacological interventions.
A) Interventions for treating hemolytic anemia and diminished hematopoiesis
Pharmacological interventions
-
Glucocorticoids (prednisone) and adrenocorticotropic hormone (ACTH) (Bourantas 1994; Etienne‐Martin 1954; Firkin 1968; Funderberg 1954; Hoffman 1952; Leonardi 1955). Prednisone is effective in hemolytic anemia but does not affect the hematopoiesis and the effective doses are generally higher (Parker 2005).
-
Danazol (Harrington 1997; Murakawa 1990).
-
Erythropoietin (Astori 1997; Balleari 1996b; Balleari 1996a; Bourantas 1994; McMullin 1996). Higher doses of erythropoietin and its derivatives may be beneficial, particularly if renal impairment is also present.
-
Iron replacement therapy when iron stores are deficient, and folic acid supplementation because of the high red cell turnover in these patients.
-
Immunosuppressive therapy such as antithymocyte globulin and cyclosporine (Ebenbichler 1996; Nakasone 2008; Paquette 1997; Scheinberg 2010). The immunosuppressive agents are an alternative to hematopoietic cell transplantation for aplastic anemia. However, the hemolytic components remain unchanged.
-
Recombinant granulocyte stimulating factor alone or in combination with cyclosporine in PNH granulocytopenic patients (Jego 1997; Schubert 1997).
Nonpharmacological interventions
-
Washed red blood cell transfusion (Guasch 1969; Jackson 1992).
-
Allogeneic hematopoietic stem cell transplantation can cure classic PNH, but treatment‐related toxicity suggests caution for this management approach (Antin 1985; Graham 1996; Kawahara 1992; Lee 2003; Matos‐Fernandez 2009; Parker 2011; Raiola 2000; Röth 2011; Woodard 2001). It has been suggested as a therapeutic option in life threatening and resistant disease associated with aplastic anemia, significant neutropenia and thrombocytopenia and severe thrombotic episodes. This treatment is used in every pediatric PNH patient with bone marrow failure, since children tolerate it better than adults (van den Heuvel‐Eibrink 2005).
B) Interventions for treating thrombotic episodes
Pharmacological interventions
-
Glucocorticoids (prednisone) (Firkin 1968).
-
Anticoagulant: heparin (Emadi 2009), and warfarin (Hall 2003).
-
Thrombolytic therapy (Araten 2012; Hauser 2003; McMullin 1994; Sholar 1985; Taniguchi 2011).
Nonpharmacological interventions
-
Whole blood or packed red blood cell transfusion (Guasch 1969).
-
Allogeneic hematopoietic stem cell transplantation (Graham 1996; Vergniol 2005).
C) Pharmacological intervention for preventing hemolytic anemia and severe thrombotic episodes
Eculizumab is a new targeted and disease‐modifying treatment strategy that inhibits a section of the complement cascade (Davis 2008; Hill 2005a; Lindorfer 2010; Luzzatto 2010; Risitano 2013; Rother 2007; Schrezenmeier 2009; Schrezenmeier 2012; Thompson 2007; Woodruff 2011; Weitz 2012). This drug effectively inhibits the formation of the membrane attack complex and intravascular hemolysis (Schrezenmeier 2012; Weitz 2012; Woodruff 2011). Eculizumab has shown significant efficacy with a marked decrease in anemia, fatigue, transfusion requirements, renal impairment, pulmonary hypertension, and risk of severe thromboembolic events, ultimately resulting in improved quality of life and survival (Brodsky 2008c; Brodsky 2009b; Hill 2005a; Hill 2005b; Hill 2010a; Hill 2010b; Hillmen 2004; Hillmen 2006; Kelly 2011; Schubert 2008). There is a need to establish the precise indications for starting treatment with eculizumab, its prophylactic role in thrombotic complications and the consideration of other available choices which include allogeneic hematopoietic cell transplantation and immunosuppressive regimens.
Clinical pharmacology of eculizumab
Treatment with eculizumab consists of an infusion of 600 mg over 25 to 45 minutes once a week for four weeks, followed by 900 mg in the fifth week. After this, the dose is maintained at 900 mg, given approximately every two weeks (EMEA 2012). Adverse events like fever, headache, back pain, nasopharyngitis, urinary tract infections, respiratory tract infections, gastrointestinal infections, nausea, fatigue, syncope, accelerated hypertension, infusion reactions, and life‐threatening desquamating rash have been reported in patients receiving eculizumab (Dmytrijuk 2008; Knoll 2008).
Eculizumab was approved by the Food and Drug Administration for the treatment of patients with PNH in March 2007 (Dmytrijuk 2008; Dubois 2009; Parker 2007; Parker 2009b). This drug was recommended for approval for the treatment of patients with PNH with a history of transfusions in the European Union in April 2007 (Parker 2007).
How the intervention might work
Eculizumab is a humanized monoclonal antibody that binds specifically to complement protein C5 with high affinity, preventing its cleavage into C5a and C5b, thereby inhibiting complement‐mediated intravascular hemolysis in patients with PNH (Hill 2008; Hill 2010a; McKeage 2011; Parker 2007; Risitano 2009; Risitano 2011; Risitano 2012; Risitano 2013; Weitz 2012). C5 being common to all pathways of complement activation, its blockade effectively halts progression of the cascade regardless of the stimuli. Prevention of C5 cleavage also blocks the generation of the potent proinflammatory C5a and cell lytic molecules C5b‐9 (Dmytrijuk 2008). While a dramatic decrease in intravascular hemolysis is found in most trials, many patients still have persistent anemia, reticulocytosis, and hemolysis which may be due to immune‐mediated extravascular hemolysis (Hill 2010a). The mechanism could be CD55‐deficient PNH red cells becoming overloaded with C3 fragments because of inhibition of the terminal complement cascade steps by eculizumab (Risitano 2009). However, this is a rare phenomenon and treatment with eculizumab has shown better hemolytic outcomes in terms of higher rates of hemoglobin stabilization, decreased need for transfusion, greater transfusion independence and an overall improvement in quality of life (Dmytrijuk 2008; Schubert 2008; TRIUMPH 2006).
Why it is important to do this review
Controversy exists as to which patients suffering from PNH should be treated with this drug (Haspel 2008). Eculizumab therapy is associated with risk of infection by Neisseria meningitidis (N. meningitidis) (McKeage 2011) and viral infections such as influenza or viral gastroenteritis (Brodsky 2008c; Brodsky 2009a; Brodsky 2009b). There is risk of Neisseria meningitidis (N. meningitidis) infection even after vaccination, and patients frequently require revaccination when on eculizumab treatment. Since eculizumab has no effect on the underlying cellular abnormality in PNH, treatment, once started, may require prolonged administration. This also raises economic concerns since eculizumab is an expensive drug (Parker 2007). Thus, there is a need for a critical appraisal of RCTs to assess eculizumab in patients with PNH (TRIUMPH 2006). This systematic review analyzing the available data might provide more definitive evidence regarding the role and safety of eculizumab in these patients.
Eventually, this Cochrane review will help clinicians to make informed decisions on the use of eculizumab for treating patients with PNH.
Objectives
To assess the clinical benefits and harms of eculizumab for treating patients with paroxysmal nocturnal hemoglobinuria (PNH), and to evaluate which patients might benefit most from its use.
Methods
Criteria for considering studies for this review
Types of studies
According to the protocol of this review (Martí‐Carvajal 2013), we included RCTs irrespective of their publication status (trials may be unpublished or published as an article, an abstract, or a letter), language and country. We did not apply limits with respect to period of follow‐up. We excluded quasi‐RCTs.
Types of participants
We included any patient with a confirmed diagnosis of paroxysmal nocturnal hemoglobinuria (PNH) according to the International PNH interest group criteria (Parker 2005). We did not apply restrictions with respect to gender or ethnicity.
Types of interventions
We planned the following two separate comparisons.
-
Eculizumab versus placebo.
-
Eculizumab versus other treatment: best available therapy.
We only found one trial comparing eculizumab with placebo. For future updates we will continue to search for RCTs of eculizumab versus other treatment: best available therapy, or any other comparison.
Types of outcome measures
Primary outcomes
-
Overall survival defined as the time from randomisation until death from any cause, and measured in the intent‐to‐treat population (FDA 2007).
Secondary outcomes
-
All‐cause mortality.
-
Health‐related quality of life and fatigue assessed by a validated scale.
-
Any fatal or nonfatal thrombotic event.
-
Transformation to myelodysplastic syndrome and acute myelogenous leukemia.
-
Adverse events (serious and nonserious). A serious adverse event, defined according to the International Conference on Harmonisation (ICH) Guidelines for Good Clinical Practice (ICH‐GCP 1997), is any untoward medical occurrence that at any dose results in death, is life‐threatening, requires inpatient hospitalization or prolongation of existing hospitalization, results in persistent or significant disability or incapacity, or is a congenital anomaly or birth defect. All other adverse events will be considered nonserious.
-
Development and recurrence of aplastic anemia on treatment.
-
Transfusion independence.
-
Withdrawal for any reason.
Search methods for identification of studies
We developed the search strategy as indicated in theCochrane Handbook for Systematic Reviews of Interventions (Lefebvre 2011). We conducted this process with the support of the Cochrane Haematological Malignancies Group (CHMG) Trials Search Co‐ordinator (TSC) and adjusted it for each database.
Electronic searches
We searched the following electronic databases.
-
The Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library 2014, Issue 05).
-
MEDLINE (Ovid) and Ovid MEDLINE(R) In‐Process & Other Non‐Indexed Citations (from 1946 to 15 May 2014).
-
PubMed (up to May 2014).
-
EMBASE (OVID) (from 1980 to 25 June 2014).
-
LILACS (from 1982 to 25 June 2014).
See Appendix 2; Appendix 3; Appendix 4 ; Appendix 5; Appendix 6 for details.
Searching other resources
-
We searched the following trial databases for ongoing and unpublished trials on June 2014.
-
We searched the following conference proceedings from 2000 to December 2012, (if they were not included in CENTRAL).
-
American Society of Hematology (ASH): Blood (Volume 104(11) 2004; 106(11) 2005; 108(11) 2006; 110(11) 2007; 112(11) 2008; 114(22) 2009; 116(21) 2010; 118(21) 2011; 120(21) 2012).
-
European Hematology Association (EHA): Haematologica (Vol 91 2006; 92 2007; 93 2008; 94 2009; 95 2010; 96 2011; 97 2012).
-
American Society for Clinical Oncology (ASCO): Journal Clinical Oncology (Vol 22(14S) 2004; 23(16S) 2005; 24(18S) 2006; 25(18S) 2007; 26 (15S) 2008; 27(15S) 2009; 27(18S) 2009; 28(15) 2010; 28(18) 2010; 29(15) and (18) 2011; 30(15) and (18) 2012).
-
European Society of Medical Oncology (ESMO): Annals of Oncology (Volumen 15 (Suppl 3) (Suppl 4) 2004; 16 (Suppl 2), 2005; 17 (Suppl 9) and (Suppl 10) 2006; 18 (Suppl 9) 2007; 19(Suppl 7) and (Suppl 8) 2008; 21 (Suppl 7) and (Suppl 8) 2010; 23 (Suppl 9) and (Suppl 10) 2012).
-
We also searched the following websites on June 2014.
We handsearched the references of all identified included trials, of relevant review articles and of current treatment guidelines. We also contacted the authors from included trials to identify unpublished trials. We did not apply any language restrictions.
Data collection and analysis
We summarized data using standard Cochrane Collaboration methodologies (Higgins 2011a).
Selection of studies
Methods for study selection followed the steps delineated by theCochrane Handbook for Systematic Reviews of Interventions (Higgins 2011a).
AMC, VA and AFC screened the titles and abstracts identified from the above sources to identify potential studies for inclusion. If this could not be done satisfactorily from the title and abstract, we sought a full text version for assessment. We presented the results of the study selection according to the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) statement (Moher 2009). See Figure 1.
Study flow diagram.
We resolved any disagreement through discussion and consensus, or if required, we consulted Ivan Solà.
Data extraction and management
We used a form to extract data. Overall, we extracted and filled in the following data: review, review author and study information, eligibility criteria, characteristics of the participants (age, gender, country), trial design and funding, related variables, intervention duration and dosage, outcomes. We extracted quality criteria according to risk of bias using the Cochrane Collaboration’s tool for assessing risk of bias: random sequence generation; allocation concealment; blinding of participants, personnel, and outcome assessors; incomplete outcome data; selective reporting; and other bias (Higgins 2011a).
For each eligible trial, AMC and VA extracted the data using the agreed form in duplicate. We resolved discrepancies through discussion or, if required, we consulted AFC and IS.
AMC and VA entered data into Review Manager software (RevMan 2014) and IS checked it for accuracy.
Assessment of risk of bias in included studies
AMC and VA and IS in pairs, independently assessed the risk of bias of each trial using a simple form, and followed the domain‐based evaluation as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011b). We resolved any discrepancies through discussion or consultation with AFC.
We assessed the following domains as 'low risk of bias', 'unclear' or 'high risk of bias'.
-
Generation of allocation sequence.
-
Allocation concealment.
-
Blinding of participants and personnel.
-
Blinding of outcome assessors.
-
Incomplete outcome data.
-
Selective reporting.
-
Other sources of bias.
Overall risk of bias
We considered low risk of bias trials to be those that adequately generated their allocation sequence; had adequate allocation concealment, adequate blinding, adequate handling of incomplete outcome data; were free of selective outcome reporting; and were free of other bias, according to the criteria given in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011b). As it was unlikely that we would find many trials at low risk of bias in all items, we chose three core domains instead of all: generation of allocation sequence, incomplete outcome data, and selective reporting bias.
Measures of treatment effect
For the following binary outcomes, we would have calculated the relative risk (RR) with 95% confidence intervals (CI):
-
All‐cause mortality.
-
Any fatal or nonfatal thrombotic event.
-
Adverse events (serious and nonserious).
-
Transformation to myelodysplastic syndrome and acute myelogenous leukemia.
-
Development, and recurrence of aplastic anemia on treatment.
-
Transfusion independence.
-
Withdrawal due to any reason.
For continuous outcomes (health‐related quality of life and fatigue assessed by a validated scale), we would have calculated the standardized mean difference with 95% CI.
For time‐to‐event outcomes (overall survival), we would have calculated the hazard ratio for each outcome with 95% CI.
Dealing with missing data
We would have used the following procedures (and will apply these for future updates, if possible). We would have noted levels of attrition and explored the impact of high levels of missing data in the overall assessment of treatment effect by using sensitivity analysis.
For all outcomes we would have carried out analysis, as far as possible, on an intention‐to‐treat basis (i.e. we would have attempted to include all participants randomized to each group in the analyses). The denominator for each outcome in each trial would have been the number randomized minus any participants whose outcomes are known to be missing (Higgins 2011b).
Assessment of heterogeneity
We did not conduct a meta‐analysis because we identified only one RCT. In future updates, we will assess statistical heterogeneity in each meta‐analysis using the T², I² and Chi² statistics. We will regard heterogeneity as substantial if I² is greater than 30% and either T² is greater than zero, or there is a low P value (less than 0.10) in the Chi² test for heterogeneity. We will investigate possible causes of heterogeneity through subgroup analysis (Deeks 2011).
Assessment of reporting biases
We would have used the following procedures for reporting bias (and will apply these for future updates, if possible). We would have attempted to assess whether the review was subjected to publication bias by using a funnel plot to illustrate variability graphically between trials. We would have assessed publication bias if at least 10 trials were available so that it was possible to make judgments about asymmetry, and if asymmetry was present, we would have explored causes other than publication bias (Sterne 2011).
Data synthesis
We would have used the following procedures (and will apply these for future updates, if possible). We would have carried out statistical analysis using Review Manager software (RevMan 2014). If the eligible trials were sufficiently comparable in their clinical characteristics, we would have summarized their findings using a random‐effects model according to the Cochrane Handbook of Systematic Reviews of Interventions section 9.4 (Deeks 2011). If the I2 statistic had been greater than 0, we would have reported the results from both random‐effects and fixed‐effect models.
'Summary of findings' table
We used the principles of the Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group (Guyatt 2011a) to assess the quality of the body of evidence associated with up to seven outcomes. We constructed a 'Summary of findings' table using the GRADE profiler software (GRADEpro 2008). The GRADE approach appraises the quality of a body of evidence based on the extent to which one can be confident that an estimate of effect or association reflects the item being assessed. Evaluation of the quality of a body of evidence considers within‐study risk of bias, the directness of the evidence, heterogeneity in the data, precision of effect estimates and risk of publication bias (Balshem 2011; Brozek 2011; Guyatt 2011b; Guyatt 2011c; Guyatt 2011d; Guyatt 2011e; Guyatt 2011f; Guyatt 2011g; Guyatt 2011h; Guyatt 2011i; Guyatt 2011j; Guyatt 2012). Due to the inclusion of a unique study in the review, we reported in the 'Summary of findings' table the outcomes that reported the included trial. We maintained the most patient important outcomes planned at the protocol stage (see Differences between protocol and review section), and added fatigue, transfusion independence, and withdrawal for any reason. Although TRIUMPH 2006 did not assess overall survival, we included this outcome because it is the primary outcome of this Cochrane review.
Subgroup analysis and investigation of heterogeneity
We will use the following procedures, and will apply these for future updates, if possible. We will devote further efforts to identifying possible causes of heterogeneity. We will explore the impact of the included trials' risk of bias and the condition of the individuals by subgroup analyses. We anticipate clinical heterogeneity for the following participant and intervention characteristics.
-
Age.
-
Duration of follow‐up.
-
Type of PNH: classical, subclinical or associated with other bone marrow disorders.
-
Aplastic anemia.
-
Thrombotic episodes.
-
PIGA mutation status at screening.
-
Previous PNH therapy including dose and duration of therapy.
These different variables will justify subgroup analyses. We will perform subgroup analysis only for primary outcomes.
Sensitivity analysis
We will use the following procedures for future updates, if possible. We will conduct sensitivity analyses according to the Cochrane Handbook of Systematic Reviews of Interventions section 9.4 (Deeks 2011).
If sufficient trials are identified, we will conduct a sensitivity analysis excluding the following.
-
Those RCTs at a high risk of bias (see Assessment of risk of bias in included studies). Trials at high risk of bias would not have been removed from the main analysis but would be analyzed separately.
-
Those RCTs with a total attrition of more than 30%, or where baseline differences between the groups exceed 10%, or both.
We will conduct a trial sequential analysis (TSA), which is a methodology that combines an information size calculation (cumulated sample sizes of included trials) for meta‐analysis with the threshold of statistical significance. TSA is a tool for quantifying the statistical reliability of data in a cumulative meta‐analysis adjusting P values for repetitive testing on accumulating data. We will conduct a TSA on binary and continuous outcomes (Brok 2009; Pogue 1997; Pogue 1998; Thorlund 2009; Wetterslev 2008; Wetterslev 2009) in future updates, if applicable. Meta‐analysis may result in type I errors due to sparse data or due to repeated significance testing when updating meta‐analysis with new trials (Brok 2009; Higgins 2011c; Wetterslev 2008). In a single trial, interim analysis increases the risk of type I errors. To avoid type I errors, group sequential monitoring boundaries are applied to decide whether a trial could be terminated early because of a sufficiently small P value, that is, the cumulative Z‐curve crosses the monitoring boundaries (Lan 1983). Sequential monitoring boundaries can be applied to meta‐analysis as well, called trial sequential monitoring boundaries (Wetterslev 2008; Wetterslev 2009). In TSA, the addition of each trial in a cumulative meta‐analysis is regarded as an interim meta‐analysis and helps to clarify whether additional trials are needed. The idea in TSA is that if the cumulative Z‐curve crosses the boundary, a sufficient level of evidence is reached and no further trials may be needed. If the Z‐curve does not cross the boundary then there is insufficient evidence to reach a conclusion. To construct the trial sequential monitoring boundaries, the required information size is needed, and is calculated as the least number of participants needed in a well‐powered single trial (Brok 2009; Pogue 1997; Pogue 1998; Wetterslev 2008). We will apply TSA since it prevents an increase of the risk of type I error (< 5%) due to potential multiple updating in a cumulative meta‐analysis, and provides us with important information in order to estimate the level of evidence of the experimental intervention.
Additionally, TSA will provide us with important information regarding the need for additional trials and the required sample size of such trials. We will apply trial sequential monitoring boundaries according to a heterogeneity‐adjusted required information size based on an a priori 10% relative risk reduction (employing α = 0.05 and ß = 0.20.
We will conduct TSA using the TSA software (CTU 2011; Thorlund 2011).
Results
Description of studies
Results of the search
We retrieved 3987 references from electronic searches, and 174 through other sources. After removing duplicates, we screened 4161 unique references for eligibility. The revision of title and abstracts led to the exclusion of 4133 references, and the obtainment of 28 publications for their review in detail. We excluded 14 of these publications. The rest of these publications were related to a unique trial (TRIUMPH 2006). Figure 1 shows a flowchart of the study selection following PRISMA guidance (Moher 2009).
Included studies
The included trial compared eculizumab (infusions of 600 mg every week for four weeks, followed one week later by 900 mg, and then by a maintenance dose of 900 mg every two weeks) with placebo. This trial included 87 participants and the median age of the participants was 41 years (range: 20 to 85). Over 50% (52/87) of the participants were female. The trial was conducted in 34 sites in Europe, Australia, Canada and the US from October 2004 to June 2005. The trial was conducted using the parallel trial design and had a follow‐up of 26 weeks (TRIUMPH 2006).
The Characteristics of included studies table shows a detailed description of the TRIUMPH 2006.
Excluded studies
Fourteen publications were excluded for the following reasons: case series studies (Arnold 2008; Brodsky 2010; Hill 2012; Hillmen 2006; Höchsmann 2012; Kelly 2011; Kim 2010; Lopez 2011; Reiss 2011; Szer 2012); controlled or noncontrolled trials (AEGIS 2008; SHEPHERD 2008);case report (John 2012); and pooled analysis of two studies (Schubert 2008a). See Characteristics of excluded studies table.
Multiple and duplicate publications
TRIUMPH 2006 is associated with 14 publications.
Ongoing studies
We found two ongoing RCTs (NCT00098280; NCT00112983). Both ongoing trials are subsets of randomized trial records such as is reported in The Metaregister of Controlled Trials and ClinicalTrials.gov (accessed on 25 June 2014). These trials have been reported as completed. See Characteristics of ongoing studies for details.
Risk of bias in included studies
The risk of bias of the included trial (TRIUMPH 2006) is summarized in Figure 2 and Figure 3.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Allocation
Randomization was performed centrally by means of an interactive voice–response system. Therefore, we considered the risk of bias arising from the method of generation of the allocation sequence and allocation concealment to be low (TRIUMPH 2006).
Blinding
TRIUMPH 2006 was reported as a double‐blind trial. However, the method of blinding was not described to guarantee an appropriate masking. Therefore, we rated the risk of bias arising from lack of blinding of participants and personnel as unclear (Wood 2008).
We rated the risk of bias arising from lack of blinding of outcome assessment as low for adverse events (serious and nonserious), patients with transfusion independence, and withdrawal for any reason, because these are objective endpoints.
We rated the risk of bias arising from lack of blinding of outcome assessment as unclear for health‐related quality of life and fatigue as the method of blinding described did not guarantee an appropriate masking (Wood 2008).
Incomplete outcome data
In TRIUMPH 2006, 14% of participants dropped out and there was an imbalance between groups of 18%. The eculizumab group lost 5% (2/43) due to pregnancy, and the research center being a long distance from participants’ homes. In contrast, the placebo group lost 23% (10/44) because the participants perceived a lack of efficacy. In conclusion, we rated the risk of attrition bias as high.
Selective reporting
We rated risk of reporting bias as high because TRIUMPH 2006 did not assess one or more clinically relevant and reasonably expected outcomes, and data on these outcomes were likely to have been recorded. This trial did not report information on transformation to myelodysplastic syndrome and acute myelogenous leukemia, and development and recurrence of aplastic anemia.
Other potential sources of bias
We did not detect any other potential sources of bias.
Effects of interventions
The results are based on one trial involving 87 participants (TRIUMPH 2006). See summary of findings Table for the main comparison.
Primary outcomes
Overall survival
The included study did not assess this outcome, but the main publication reported that no patients died during the study (TRIUMPH 2006).
Secondary outcomes
TRIUMPH 2006 reported on only five of our outcomes of interest. No data were available on the other outcomes (transformation to myelodysplastic syndrome and acute myelogenous leukemia, and development and recurrence of aplastic anemia on treatment).
All‐cause mortality
No patients died during the execution of the included trial.
Health‐related quality of life and fatigue
The assessment of health‐related quality of life was based on the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (scores can range from 0 to 100, with higher scores on the global health status and functioning scales and lower scores on the symptom scales and single‐item measures indicating improvement). Among the 87 participants, there was a statistically and clinically significant increase in the global health status scale in the eculizumab group compared with the placebo group (mean difference (MD) 19.4, 95% confidence interval (CI) 8.25 to 30.55; P = 0.0007) (Analysis 1.1).
The assessment of fatigue was based on the Functional Assessment of Chronic Illness Therapy Fatigue instrument (scores can range from 0 to 52, with higher scores indicating improvement in fatigue). A change of three or more points in scores on this instrument represents a clinically important difference. The 43 participants into eculizumab group showed a statistical and clinical significant reduction in fatigue compared with the 44 participants of the placebo group (MD 10.4, 95% CI 9.97 to 10.83; P = 0.00001) (Analysis 1.2).
Any fatal or nonfatal thrombotic event
Over the course of the trial, an episode of thrombosis occurred in the placebo group. This event was considered a single thrombosis (TRIUMPH 2006).
Adverse events
TRIUMPH 2006 reported a low rate of serious adverse effects that resulted in a nonsignificant statistical difference between the eculizumab group: 9.3% (4/43) versus the placebo group: 20.4% (9/44) (RR 0.45, 95% CI 0.15 to 1.37; P = 0.16, 87 participants). There was a nonsignificant statistical difference according to specific serious adverse events (Analysis 1.3).
There was a nonsignificant statistical difference according to the most frequent adverse events (Headache, nasopharyngitis, upper respiratory tract infection, back pain, and nausea) (Analysis 1.4).
Transfusion independence
TRIUMPH 2006 reported 22 out of 43 participants with transfusion independence in the eculizumab group (51% (22/43)) and none in the placebo group (RR 46.02, 95% CI 2.88 to 735.53; P = 0.007) (Analysis 1.5).
Withdrawal for any reason
TRIUMPH 2006 reported a statistical significant reduction of withdrawal for any reason in the eculizumab arm (4.7% (2/43) compared with the placebo group (22.7% (10/44)) (RR 0.20, 95% CI 0.05 to 0.88; P = 0.03) (Analysis 1.6).
Discussion
Summary of main results
Our aim was to assess the influence of eculizumab on the treatment of patients with paroxysmal nocturnal hemoglobinuria (PNH). This Cochrane review found one sponsored drug company trial with 26 weeks of follow‐up, comparing eculizumab against placebo, involving 87 patients (TRIUMPH 2006). Eculizumab improved health‐related quality of life, showed a higher proportion of patients with transfusion independence, and reduced fatigue and withdrawal for any reason. The administration of eculizumab did not affect the occurrence of adverse events (serious or not serious). No patients died during the trial, and one case of thrombosis was reported in the placebo group. The trial did not assess overall survival, any fatal or nonfatal thrombotic event, transformation to myelodysplastic syndrome and acute myelogenous leukemia, and development and recurrence of aplastic anemia on treatment. See summary of findings Table for the main comparison.
Overall completeness and applicability of evidence
PNH is classified into three categories (Parker 2012). The included trial involved only classic PNH (TRIUMPH 2006). It was conducted with fewer than 100 patients, and rated as high or unclear risk of bias in many domains. Therefore, three issues emerge. First, external validity is restricted to patients with classic PNH. Second, a low number of participants carry a risk of random error (play of chance). It is also known as type I error (alpha error) and type II error (beta error) (Porta 2008). Type I error (alpha error) is the error of wrongly rejecting a null hypothesis, i.e. declaring that a difference exists when it does not (Porta 2008). Type II error (beta error) is the error of failing to reject a false null hypothesis, i.e. declaring that a difference does not exist, when in fact it does (Porta 2008). Third, TRIUMPH 2006 had a high risk of systematic errors (bias, that is, overestimation of benefits ‐ positive false result ‐ and underestimation of harms ‐ negative false result) (Button 2013; Kjaergard 2001; Savović 2012; Savović 2012a; Thorlund 2011a). Thus, overall completeness and applicability of evidence is poor due to potential spurious findings.
Three additional issues can reduce the applicability of the evidence: selective outcome reporting, incomplete outcome data issues, and unblinding for the outcome assessment of quality of life and fatigue. These issues may be particularly relevant to consider as a further trial is planned. TRIUMPH 2006 did not report the main clinical outcomes. Trials should adopt an agreed upon set of core outcomes for each medical condition (Clarke 2007). This approach may reduce the impact of outcome reporting bias (Kirkham 2010). The impact of outcome reporting bias may be reduced for adopting the recommendations of The Patient‐Centered Outcomes Research Institute (PCORI) (PCORI 2012). PCORI's research is intended to give patients a better understanding of the prevention, treatment and care options available, and the science that supports those options (Basch 2012; Gabriel 2012; PCORI 2012). Eculizumab is an expensive drug (Parker 2007). Therefore, the outcomes should take into account the financial aspects. TRIUMPH 2006 was analyzed using an intention to treat analysis approach, however, the trial had a high risk of attrition bias which acts as selection bias. Finally, lack of blinding for the outcome assessment of quality of life and fatigue reduce the reliability of the evidence (Savović 2012; Wood 2008).
Quality of the evidence
We rated the quality of evidence for the relevant outcomes of the review (see summary of findings Table for the main comparison). The quality of evidence for eculizumab in patients with PNH is moderate to low due to limitations in the design and execution of the included trial, and imprecision in some effect estimates. The unblinded outcome assessment and the high attrition due to the perceived lack of efficacy within the patients that received placebo in the TRIUMPH trial could bias the differences observed in quality of life and fatigue scores. For quality of life we also downgraded the quality of evidence for the imprecision of the effect estimate, suggesting the possibility of a moderate to important effect in the EORTC QLQ‐C30 scores. Instead, the effect estimates and the values in the confidence interval for fatigue are included between those established in the literature as minimally important differences (Cella 2002; Webster 2003). We downgraded the quality of evidence for the low rate of events reported for the rest of the outcomes.
Potential biases in the review process
In the process of performing a systematic review, there is a group of biases called significance‐chasing biases (Ioannidis 2010). This group includes publication bias, selective outcome reporting bias, selective analysis reporting bias, and fabrication bias (Ioannidis 2010). Publication bias represents a major threat to the validity of systematic reviews, particularly in reviews that include small trials. This Cochrane review has a low risk of publication bias due to the thorough trial search process. Selective outcome reporting bias operates through suppression of information on specific outcomes and has similarities to study publication bias, in that 'negative’ results remain unpublished (Ioannidis 2010). This Cochrane review found that TRIUMPH 2006 has a high risk of selective outcome reporting because this trial did not report overall survival and other main clinical outcomes such any fatal or nonfatal thrombotic events, transformation to myelodysplastic syndrome and acute myelogenous leukemia, and development and recurrence of aplastic anemia on treatment.
Agreements and disagreements with other studies or reviews
There are no other systematic reviews for comparison with this Cochrane review. The AEGIS clinical trial, an open‐label, single‐arm, multicenter study in 29 Japanese patients who were 12 years of age or older with a diagnosis of PNH for at least six months, reported reduction in transfusion requirements, and reduction in fatigue (AEGIS 2008). However, this trial is limited by the methodology issue.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Comparison 1 Eculizumab versus placebo, Outcome 1 Health‐related quality of life and fatigue (Assessed by EORTC QLQ‐C30 instrument).
Comparison 1 Eculizumab versus placebo, Outcome 2 Fatigue (Assessed by FACIT‐Fatigue instrument).
Comparison 1 Eculizumab versus placebo, Outcome 3 Serious adverse events.
Comparison 1 Eculizumab versus placebo, Outcome 4 Adverse event (most frequents).
Comparison 1 Eculizumab versus placebo, Outcome 5 Transfusion independence.
Comparison 1 Eculizumab versus placebo, Outcome 6 Withdrawal for any reason.
| Eculizumab compared with placebo for paroxysmal nocturnal hemoglobinuria | ||||||
| Patient or population: patients with paroxysmal nocturnal hemoglobinuria | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of Participants | Quality of the evidence | Comments | |
| Assumed risk1 | Corresponding risk | |||||
| Placebo | Eculizumab | |||||
| Overall survival ‐ not measured | See comment | See comment | Not estimable | 87 | See comment | This outcome was not measured in the included study |
| All‐cause mortality | See comment | See comment | Not estimable | 87 | See comment | No patients died during the execution of the included study. The small sample size of the included trial does not allow to make judgments about the quality of evidence |
| Health‐related quality of life | The mean change from baseline in the health‐related quality of life score in the control group was | The mean change from baseline in the health‐related quality of life score in the intervention group was | 87 | ⊕⊕⊝⊝ | ||
| Fatigue | The mean change from baseline in the fatigue score in the control group was | The mean change from baseline in the fatigue score in the intervention group was | 87 | ⊕⊕⊕⊝ | ||
| Adverse events (serious and nonserious) | 205 per 1000 | 92 per 1000 | RR 0.45 | 87 | ⊕⊕⊝⊝ | |
| Transfusion independence | 20 per 10006 | 920 per 1000 | RR 46.02 | 87 | ⊕⊕⊕⊝ | |
| Withdrawal for any reason | 227 per 1000 | 45 per 1000 | RR 0.20 | 87 | ⊕⊕⊕⊝ | |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1 Assumed risk is based on the risks for the control group in the included trial. | ||||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Health‐related quality of life and fatigue (Assessed by EORTC QLQ‐C30 instrument) Show forest plot | 1 | Mean Difference (Random, 95% CI) | Subtotals only | |
| 1.1 Global health status scale | 1 | 87 | Mean Difference (Random, 95% CI) | 19.4 [8.25, 30.55] |
| 2 Fatigue (Assessed by FACIT‐Fatigue instrument) Show forest plot | 1 | 87 | Mean Difference (Random, 95% CI) | 10.4 [9.97, 10.83] |
| 2.1 Fatigue | 1 | 87 | Mean Difference (Random, 95% CI) | 10.4 [9.97, 10.83] |
| 3 Serious adverse events Show forest plot | 1 | Risk Ratio (M‐H, Random, 95% CI) | Subtotals only | |
| 3.1 Overall serious adverse events | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 0.45 [0.15, 1.37] |
| 3.2 Exacerbation of PNH | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 0.34 [0.04, 3.15] |
| 3.3 Renal colic | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 3.07 [0.13, 73.30] |
| 3.4 Lumbar‐ or sacral‐disk prolapse | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 3.07 [0.13, 73.30] |
| 3.5 α‐Hemolytic streptococcal bacteremia | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 3.07 [0.13, 73.30] |
| 3.6 Central‐line and urinary tract infections | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 0.34 [0.01, 8.14] |
| 3.7 Upper respiratory tract infection | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 0.34 [0.01, 8.14] |
| 3.8 Probable viral infection | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 0.34 [0.01, 8.14] |
| 3.9 Neutropenia | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 0.34 [0.01, 8.14] |
| 3.10 Cellulitis, folliculitis, and neutropenia | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 0.34 [0.01, 8.14] |
| 3.11 Anemia and pyrexia | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 0.34 [0.01, 8.14] |
| 4 Adverse event (most frequents) Show forest plot | 1 | Risk Ratio (M‐H, Random, 95% CI) | Subtotals only | |
| 4.1 Headache | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 1.62 [0.90, 2.92] |
| 4.2 Nasopharyngitis | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 1.28 [0.56, 2.93] |
| 4.3 Upper respiratory tract infection | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 0.61 [0.24, 1.54] |
| 4.4 Back pain | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 2.05 [0.66, 6.30] |
| 4.5 Nausea | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 1.43 [0.49, 4.17] |
| 5 Transfusion independence Show forest plot | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 46.02 [2.88, 735.53] |
| 6 Withdrawal for any reason Show forest plot | 1 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 0.20 [0.05, 0.88] |
