Oral deferiprone for iron chelation in people with thalassaemia

  • Review
  • Intervention

Authors


Abstract

Background

Thalassaemia major is a genetic disease characterised by a reduced ability to produce haemoglobin. Management of the resulting anaemia is through red blood cell transfusions.

Repeated transfusions result in an excessive accumulation of iron in the body (iron overload), removal of which is achieved through iron chelation therapy. A commonly used iron chelator, deferiprone, has been found to be pharmacologically efficacious. However, important questions exist about the efficacy and safety of deferiprone compared to another iron chelator, desferrioxamine.

Objectives

To summarise data from trials on the clinical efficacy and safety of deferiprone and to compare the clinical efficacy and safety of deferiprone with desferrioxamine for thalassaemia.

Search methods

We searched the Cochrane Cystic fibrosis and Genetic Disorders Group's Haemoglobinopathies trials Register and MEDLINE, EMBASE, CENTRAL (The Cochrane Library), LILACS and other international medical databases, plus registers of ongoing trials and the Transfusion Evidence Library (www.transfusionevidencelibrary.com). We also contacted the manufacturers of deferiprone and desferrioxamine.

All searches were updated to 05 March 2013.

Selection criteria

Randomised controlled trials comparing deferiprone with another iron chelator; or comparing two schedules or doses of deferiprone, in people with transfusion-dependent thalassaemia.

Data collection and analysis

Two authors independently assessed trials for risk of bias and extracted data. Missing data were requested from the original investigators.

Main results

A total of 17 trials involving 1061 participants (range 13 to 213 participants per trial) were included. Of these, 16 trials compared either deferiprone alone with desferrioxamine alone, or a combined therapy of deferiprone and desferrioxamine with either deferiprone alone or desferrioxamine alone; one compared different schedules of deferiprone. There was little consistency between outcomes and limited information to fully assess the risk of bias of most of the included trials.

Four trials reported mortality; each reported the death of one individual receiving deferiprone with or without desferrioxamine. One trial reported five further deaths in patients who withdrew from randomised treatment (deferiprone with or without desferrioxamine) and switched to desferrioxamine alone. Seven trials reported cardiac function or liver fibrosis as measures of end organ damage.

Earlier trials measuring the cardiac iron load indirectly by magnetic resonance imaging (MRI) T2* signal had suggested deferiprone may reduce cardiac iron more quickly than desferrioxamine. However, a meta-analysis of two trials suggested that left ventricular ejection fraction was significantly reduced in patients who received desferrioxamine alone compared with combination therapy.  

One trial, which planned five years of follow up, was stopped early due to the beneficial effects of combined treatment compared with deferiprone alone in terms of serum ferritin levels reduction.

The results of this and three other trials suggest an advantage of combined therapy over monotherapy to reduce iron stores as measured by serum ferritin. There is, however, no conclusive or consistent evidence for the improved efficacy of combined deferiprone and desferrioxamine therapy over monotherapy from direct or indirect measures of liver iron. Both deferiprone and desferrioxamine produce a significant reduction in iron stores in transfusion-dependent, iron-overloaded people. There is no evidence from randomised controlled trials to suggest that either has a greater reduction of clinically significant end organ damage.

Evidence of adverse events were observed in all treatment groups. Occurrence of any adverse event was significantly more likely with deferiprone than desferrioxamine in one trial, RR 2.24 (95% CI 1.19 to 4.23). Meta-analysis of a further two trials showed a significant increased risk of adverse events associated with combined deferiprone and desferrioxamine compared with desferrioxamine alone, RR 3.04 (95% CI 1.18 to 7.83). The most commonly reported adverse event was joint pain, which occurred significantly more frequently in patients receiving deferiprone than desferrioxamine, RR 2.64 (95% CI 1.21 to 5.77). Other common adverse events included gastrointestinal disturbances as well as neutropenia or leucopenia, or both.

Authors' conclusions

In the absence of data from randomised controlled trials, there is no evidence to suggest the need for a change in current treatment recommendations; namely that deferiprone is indicated for treating iron overload in people with thalassaemia major when desferrioxamine is contraindicated or inadequate. Intensified desferrioxamine treatment (by either subcutaneous or intravenous route) or use of other oral iron chelators, or both, remains the established treatment to reverse cardiac dysfunction due to iron overload. Indeed, the US Food and Drug Administration (FDA) recently only gave support for deferiprone to be used as a last resort for treating iron overload in thalassaemia, myelodysplasia and sickle cell disease. However, there is evidence that adverse events are increased in patients treated with deferiprone compared with desferrioxamine and in patients treated with combined deferiprone and desferrioxamine compared with desferrioxamine alone. There is an urgent need for adequately-powered, high-quality trials comparing the overall clinical efficacy and long-term outcome of deferiprone with desferrioxamine.

Plain language summary

The use of the iron chelator deferiprone in people with thalassaemia who are dependent on blood transfusions

In thalassaemia sometimes the body cannot produce enough haemoglobin. Regular blood transfusions can manage this, but may lead to excess iron in the body, which, if not removed, may damage major organs. Iron chelation therapy removes excess iron; one common chelator is deferiprone. Questions exist about whether deferiprone is as good and safe as the most widely-used iron chelator, desferrioxamine. Desferrioxamine is administered by a needle under the skin or into a vein, and was the standard monotherapy for 20 years.

Seventeen randomised controlled trials compared deferiprone with desferrioxamine. They report little data on death or end organ damage, so we report the effects of therapy using mainly other markers. Removing excess iron was assessed by iron concentration in the blood and liver; heart function; and the amount of iron passed in urine. However, the amount of iron removed with either deferiprone or desferrioxamine was not consistent; one reason being that outcomes were measured differently. This makes it difficult to compare results between trials.

Adverse events included joint pain, nausea, stomach upsets and low white blood cell count with deferiprone and pain or skin reactions at the injection site and joint pain with desferrioxamine. In one trial, the risk of an adverse event with deferiprone was twice that of the risk with desferrioxamine. Two further trials showed a three-fold increased risk of an adverse event with combined deferiprone and desferrioxamine therapy compared with desferrioxamine alone.

We found no evidence to change current recommendations to treat iron overload in thalassaemia with deferiprone when desferrioxamine cannot be used or is inadequate. Intensified desferrioxamine treatment or use of other oral iron chelators (or both of these) remains the established treatment to reverse heart problems due to iron overload. Indeed, the US Food and Drug Administration (FDA) recently gave support for deferiprone only to be used as a last resort treatment in patients with thalassaemia, myelodysplasia and sickle cell disease. The danger of raised liver enzymes or a very low white blood cell count with deferiprone means that this treatment should not be used unless close monitoring of full blood counts and liver function is possible. Large trials of chelation therapy with standardised measures of iron stores and end organ damage are needed so valuable trial data can be compared and analysed.

Background

Description of the condition

Iron overload constitutes a major health problem for all people who require regular blood transfusions, such as people with thalassaemia. Effective iron removal, known as chelation therapy, is essential to the survival and quality of life of iron overloaded people. Before iron chelation therapy was available, the prognosis for transfusion-dependent individuals was poor and less than 5% of children survived beyond 15 years of age. Control of iron load using desferrioxamine (DFO) showed that if the children's total body iron was controlled adequately, survival was increased and the frequency of cardiac deaths fell (Brittenham 1994; Olivieri 1994). The clinical effectiveness of DFO was confirmed in long-term follow-up studies, where the cohort born after 1970 shows that 68% of the individuals are alive at the age of 35 years (Borgna-Pignatti 1998; Borgna-Pignatti 2004).

In people with transfusion-dependent thalassaemia major, failure to control the iron load has been demonstrated to be the main cause of death (Olivieri 1994). In these people, where serum ferritin was maintained at less than 2500 mg/L, there was a survival rate without cardiac disease of 91% at 15 years, whereas in those whose serum ferritin levels were greater than 2500 mg/L there was a survival rate without cardiac disease of only 20% over the same period (Olivieri 1994).

Description of the intervention

Currently, the iron chelator of first choice for clinical use is desferrioxamine (DFO). To be clinically effective, DFO must be administered as a subcutaneous infusion over 8 to 12 hours, 5 to 7 days a week. This regimen has been demonstrated to regulate iron balance, reduce the body iron load, prevent the onset of iron-induced complications, reverse some of the induced organ damage and increase survival as a result of iron excretion (Aldouri 1990; Borgna-Pignatti 1998; Brittenham 1994; Ehlers 1991; Gabutti 1996; Modell 2000; Olivieri 1999; Pippard 1978; Propper 1976; Propper 1977; Wolfe 1985).

Unfortunately, chronic therapy with DFO has had a number of problems, particularly the adherence to an arduous daily regimen of infusions (Olivieri 1997b; Weatherall 2002) and adverse events due to DFO may cause renal impairment (Bacon 1983; Richardson 1993), local skin reactions (Kushner 2001), growth retardation (Olivieri 1992; Piga 1988), increased susceptibility to Yersinia infections (Gallant 1986), high frequency sensorineural hearing loss (Kontzogolou 1996; Olivieri 1986; Porter 1989b) and retinal damage (Davies 1983; Olivieri 1986). However, reports of these adverse events, which appear frequently in the literature from the early years of DFO use, have declined in recent years as the relationship between toxicity and drug dose has been more clearly understood and managed (Porter 1989a; Porter 2002). The principle problem with desferrioxamine remains the low rate of compliance - only 70% in some series of participants (Olivieri 1995b). A Cochrane Review has examined the efficacy of different regimens of desferrioxamine in people with transfusion-dependent thalassaemia (Fisher 2013).

The introduction of a pharmacologically efficacious and well-tolerated oral iron chelator, which would negate some of the problems associated with DFO, has been widely seen as highly desirable. The drug 1,2-dimethyl-3-hydroxypyroid-4-one, or deferiprone (also known as L1, Kelfer or DMHP) has been the first oral iron chelator to be clinically evaluated. It was first synthesised in 1984 (Hider 1984) and was soon shown to be pharmacologically efficacious in achieving iron excretion (Agarwal 1992; Kontoghiorghes 1990; Tondury 1990). Since these early studies, a large number of centres have reported their experience with deferiprone (Cohen 2003; Del Vecchio 2000; Fischer 2003; Lucas 2002; Olivieri 1995a; Olivieri 1997a; Rombos 2000).

As with desferrioxamine, adverse events have been reported in people with thalassaemia taking deferiprone. These include gastrointestinal disturbances (Ceci 2002; Cohen 2003; Taher 2001), arthropathy (Cohen 2003; Hoffbrand 1998; Lucas 2002; Mazza 1998; Taher 2001), raised liver enzymes (Ceci 2002; Cohen 2003), neutropenia and agranulocytosis (Cohen 2003; Pati 1999). Progression of liver fibrosis during treatment with deferiprone has been cited by some studies (Berdoukas 2000; Olivieri 1998; Tondury 1998), but not others (Wanless 2002); and ensuing correspondence (Brittenham 2003). Variation in the length of treatment (Berdoukas 2000) and the failure to record baseline values of liver fibrosis (Hoffbrand 1998; Tondury 1998) has made precise evaluation of the progression and significance of liver fibrosis on treatment difficult.

Why it is important to do this review

Different iron chelators may have variable ability to chelate iron from specific tissues although the side effects of the drugs in combination may also be, in some way, additive. Therefore, the benefits and safety profiles of the combination of iron chelators compared with monotherapy cannot be predicted but can only be determined in high-quality, adequately-powered, long-term, randomised controlled trials. Indeed any new treatment must be evaluated against the best available treatment by randomised clinical trials.

Comparisons of iron chelation regimens face a number of methodological problems, including accurately assessing the clinical efficacy and effectiveness of therapy, determining the consequences of underlying disease, attributing side effects to therapy and monitoring participants for a sufficient period of time to reach clinically useful endpoints. It is therefore essential for trials to carefully report baseline values, choose appropriate outcome measures and consider observed changes in outcome measures in the context of underlying conditions. Moreover, the relatively short period of clinical use of deferiprone to date limits long-term comparison of not only safety but also effectiveness with desferrioxamine, which has been available for clinical use since the 1970s.

The need for long-term trials cannot be overemphasised. Iron chelators in thalassaemia are rather different to many drug trials simply because of the time-scale involved. Iron accumulates slowly and the onset of complications again is very variable in its timing. Thus the only reliable way to learn more about the relative value of the two iron-chelating drugs would be a long-term, i.e. more than a 5-year and preferably a 10-year prospective trial, starting in early life before there is really significant iron loading and carried through to the critical period of adolescence and beyond.

The importance of long-term considerations validate the need for a systematic review of the clinical efficacy and safety profile of the iron chelator deferiprone to determine and analyse the available evidence regarding such treatment.

One early systematic review examining the use of deferiprone in thalassaemia was published prior to the completion of 12 of the 17 relevant trials included in the current review and considered only one outcome, namely reduction of hepatic iron overload (Caro 2002). Two reviews have examined the limited number of studies looking at the psychosocial aspects of chelation therapy. Poor adherence to iron chelation therapy was documented to negatively impact survival (Abetz 2006), although older age was consistently associated with lower levels of chelation adherence (Evangeli 2010).

The role of iron chelation on cardiac function has been of general interest after reports that indirect measures of cardiac iron loading, the myocardial T2*, showed a statistically significant difference, favouring deferiprone over DFO in one trial (Pennell 2006) and favouring deferiprone combined with DFO over DFO alone in another trial (Tanner 2007). However, a meta-analysis of the influence of iron chelators on myocardial iron and cardiac function in thalassaemia reported that while DFO and deferiprone both reduced myocardial iron by similar amounts, there was no significant difference between the two chelators nor was there any improvement in the left ventricular ejection fraction (LVEF) as a measure of cardiac function (Mamtani 2008). They were able to analyse data on over 290 patients and also suggest there was a publication bias in smaller studies, favouring reporting of improvements in LVEF but not myocardial iron (Mamtani 2008).

The Italian Society of Haematology produced a systematic review of the available data surrounding the wider questions of iron chelation therapy in thalassaemia patients but did not undertake formal meta-analysis of the identified studies (Angelucci 2008). A later systematic review and meta-analysis of all iron chelation therapies in thalassaemia, focusing on the role of combined or sequential therapy versus monotherapy and examining 16 trials with 1500 patients, concluded that lower final liver iron concentrations were associated with combined chelator therapy compared with monotherapy. There were no significant differences in heart T2* signal during combined or sequential therapy versus monotherapy, although there was an improvement in the LVEF after combined or sequential therapy compared to monotherapy (P = 0.01 and P < 0.00001 respectively) (Maggio 2011).

These somewhat disparate findings in generally small series of patients highlight the need for continued systematic analysis of chelation therapy for thalassaemia.

Objectives

To summarise data from trials on the effects and safety of deferiprone as an iron chelating agent in people with transfusion-dependent thalassaemia and to compare the safety and effectiveness of deferiprone for thalassaemia with desferrioxamine.

Methods

Criteria for considering studies for this review

Types of studies

Randomised controlled trials (RCTs).

Types of participants

People of any age, with thalassaemia, who are transfusion-dependent, from any setting worldwide.

Types of interventions

For deferiprone (all schedules and doses), the following comparisons were considered:

  1. deferiprone compared with placebo;

  2. deferiprone compared with another iron chelating treatment schedule;

  3. deferiprone dose A compared with deferiprone dose B.

No trials comparing deferiprone with placebo were identified.

Types of outcome measures

Primary outcomes
  1. Mortality

Secondary outcomes
  1. Evidence of reduced end organ damage due to iron deposition:

    1. cardiac failure

    2. endocrine disease

    3. surrogate markers of end organ damage (e.g. liver enzymes, fasting glucose, cardiac dysfunction)

    4. histological evidence of liver fibrosis

  2. Measures of iron overload - including serum ferritin, assessment of liver and other tissue iron levels by biopsy with biochemical measurement by SQUID (superconducting quantum interference device) or by MRI (magnetic resonance imaging) and iron excretion (urine and faecal) over 24 hours.

  3. Adverse events*, including

    1. neutropenia and agranulocytosis

    2. arthralgia, joint effusions

    3. gastrointestinal disturbances

    4. liver fibrosis

    5. other

  4. Participant compliance with iron chelation treatment.

  5. Cost of interventions

* The review of adverse events was confined solely to those reported in the RCTs included in this review. No adverse event data were collected from non-randomised trials or observational studies.

Search methods for identification of studies

Electronic searches

Relevant trials were identified from the Cystic Fibrosis and Genetic Disorders Group's Haemoglobinopathies Trials Register using the terms: thalassaemia AND (deferiprone OR 1,2-dimethyl-3-hydroxyproid-4-one).
The Haemoglobinopathies Register is compiled from electronic searches of the Cochrane Central Register of Controlled Trials (CENTRAL) (updated each new issue of The Cochrane Library) and quarterly searches of MEDLINE. Unpublished work is identified by searching the abstract books of five major conferences: the European Haematology Association annual conference; the American Society of Hematology annual conference; the British Society for Haematology Annual Scientific Meeting; the Caribbean Health Research Council Meetings; and the National Sickle Cell Disease Program Annual Meeting. For full details of all searching activities for the Haemoglobinopathies Trials Register, please see the relevant section of the Cochrane Cystic Fibrosis and Genetic Disorders Review Group Module.

Date of the last search of the Haemoglobinopathies Trials Register: 1 March 2013.

In addition, the following databases were searched for the review update in March 2013:

  • CENTRAL (The Cochrane Library 2013, Issue 1) (Appendix 1)

  • PubMed (epublications only) (5 March 2013) (Appendix 1)

  • MEDLINE (1948 to 5 March 2013) (Appendix 1)

  • Embase (1974 to 5 March 2013) (Appendix 1)

  • UKBTS/SRI Transfusion Evidence Library (www.transfusionevidencelibrary.com) (1980 to 5 March 2013) (Appendix 1)

  • LILACS (1982 to 5 March 2013) (Appendix 1)

  • KoreaMed (1997 to 5 March 2013) (Appendix 1)

  • IndMed (1986 to 5 March 2013) (Appendix 1)

  • PakMediNet (1995 to 5 March 2013) (Appendix 1)

  • Databases of ongoing trials (all years to 5 March 2013): Novartis Clinical Trial Results database (www.novartisclinicaltrials.com); ClinicalTrials.gov; WHO International Clinical Trials Registry Platform (ICTRP); ISRCTN Register; Hong Kong Clinical Trials Register (Appendix 1).

Search strategies were designed to search for all iron chelators and no language restrictions were placed upon any of the searches. Search strategies used for the original searches in June 2006 can be found in Appendix 2.

Searching other resources

Handsearching of reference lists

The review authors checked the reference lists of all identified trials, relevant review articles and current treatment guidelines for further literature, but limited these searches to the 'first generation' reference lists.

Contact was made with the manufacturer of deferiprone (Apotex) and other iron chelators (Novartis, Biomedical Frontiers, CIPLA, Lipomed) and selected experts requesting details of unpublished trials that involve deferiprone for the original review (2006), but not for the 2013 update.

Data collection and analysis

Selection of studies

An information specialist (CD) undertook electronic searches for potentially relevant papers. The Cystic Fibrosis and Genetic Disorders Group carried out additional searches. From the papers identified, the information specialist (CD) removed references that were duplicates, clearly irrelevant or previously screened. One author (SJB or SAF) screened all titles and abstracts of papers identified for relevancy. Two authors working independently, for both the original review (Senani Williams and CD), and the 2013 update (SJB or SAF and CD) excluded only trials that were clearly irrelevant and assessed all other trials using the criteria indicated above. Agreement between the authors was good and they resolved any discrepancies entirely by themselves: any disagreements related to the interpretation of clinical characteristics.

They recorded reasons for the exclusion of trials that did not meet the review's eligibility criteria. They have included trials where important screening information is lacking in the 'Studies awaiting classification' section of the review.

Data extraction and management

Aside from details relating to the risk of bias of the included trials, the authors extracted two groups of data (as below).

  1. Trial characteristics: place and date of publication; population characteristics; setting; detailed nature of intervention, of comparator and of outcomes. Key purposes of these data were to define unexpected clinical heterogeneity of included trials independently from analysis of results.

  2. Results of included trials, in respect of each of the main outcomes indicated in the review question and failure to report data for a particular outcome, were recorded. For dichotomous outcomes the numbers of events in treatment and control groups were recorded. For continuous outcomes, mean and standard deviation (SD) at baseline, end of treatment and change from baseline were recorded where possible. The 'denominators' for all outcomes were intended to be the numbers randomly allocated to treatment and control group. However, several of the included trials did not report outcome data by numbers randomised. For these trials the 'denominators' are the number of participants for whom outcome data were reported.

For both the original review (CD, Senani Williams) and the update (SJB and either SAF, OC or SG), two review authors independently extracted data onto trial-specific data extraction forms (created and previously piloted by two authors (DR, SJB)). At update stage, the authors made minor adjustments to the layout as required. The two authors undertaking data extraction resolved disagreements by consensus. Disagreements related to interpretation of clinical outcomes. Once they had resolved disagreements, the authors recorded the consensus data onto a third data extraction form. Two review authors (SJB, SAF) transcribed this into the computer software Review Manager 5.1 (Review Manager 2011).

Assessment of risk of bias in included studies

The authors undertook assessment of the risk of bias for all included trials (original and trials identified in the update) using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011), as summarised below. They resolved any disagreements by discussion.

  1. Generation of random sequence.

  2. Concealment of treatment allocation schedule.

  3. Blinding of clinician (person delivering the treatment), participant and outcome assessors to treatment allocation.

  4. Completeness of the outcome data, checking for possible attrition bias through withdrawals, loss to follow up and protocol violations.

  5. Selective reporting bias, checking that all of a trial's pre-specified outcomes and all expected outcomes of interest to the review have been reported.

  6. Other sources of bias in the included trials.

  7. An overall risk of bias assessment was made based on items 1 to 6 above. An explicit judgement about whether trials are at high risk of bias was made according to criteria given in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).The likely magnitude and direction of the bias was assessed with reference to items 1 to 6, with particular emphasis on the likely impact of bias on the findings.

The authors rated the above criteria were rated according to criteria identified in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). This assigns ratings of low, high or unclear risk of bias. The authors recorded these ratings in each trial's 'Risk of Bias' table. In addition, the authors provided a narrative summary of the findings of this assessment alongside the individual ratings. They reported the overall risk of bias assessment in the results section of this review.

The authors anticipated that the issue of blinding would be a challenge in the identified trials, given the different routes of administration of the iron chelators currently available (deferiprone: oral; desferrioxamine: predominantly subcutaneous infusion over 8 hours to 12 hours). It would have been very difficult to blind either the clinician or the participant to the trial treatments. However, blinding of the outcome assessor to the treatment allocation would have been possible for these trials and has been given particular attention.

Measures of treatment effect

The main method of analysis was quantitative but the authors made an overall interpretation from a balanced assessment of the patterns of results identified across the included trials. Due to the disparity in methods of reporting results between trials, in order to formally assess as many trials as possible, the authors analysed outcomes using both end of trial data and mean change from baseline data where appropriate. Where the papers did not report a SD for the mean change from baseline and the trial did not report sufficient data to enable SD calculation (i.e. a correlation coefficient), the authors did not want to make the necessary assumptions about unknown statistical distributions (Deeks 2011) and in this case, they used end of trial data to analyse reported outcomes.

The authors presented results as risk ratio (RR) for binary data and mean difference (MD) for continuous data. The authors provided 95% confidence intervals (CI) for all measures of effect size. The authors analysed all participants in the treatment groups to which they had been randomised, with the exception of one trial (Gomber 2004). In this trial, the authors analysed end of trial data according to the treatment received rather than the treatment group to which the participants were randomised.

The authors took care to record the method of data analysis used in the two eligible cross-over trials. Although neither of the two cross-over trials presented outcome data in a way that would have helped in the undertaking of a meta-analysis (paired-samples analysis) (Elbourne 2002), this was not the reason why the authors analysed data from these two cross-over trials qualitatively. Differences in the interventions being compared and the clinical setting of the trials precluded data from the cross-over trials being pooled in a meta-analysis with data from the parallel group trials.

To facilitate comparison of results between trials, where SDs were not explicitly stated, the authors converted the standard error (SE) of the mean to the SD. Within this review, unless otherwise stated, the authors present data as mean and SD. Further, the authors have changed the units of serum ferritin concentration where necessary from µg/L or pmol/L to ng/ml and the units of liver iron concentration changed from mg/g to µg/g, to facilitate the pooling, analysis and plotting of outcome data. Where results of individual trials were displayed graphically and estimation was considered reasonable, the authors estimated values visually from the graphs and stated this within tabulated results.

In one trial, the ratio of geometric means was used to describe the difference in change between treatment arms and exact P values were reported (Pennell 2006). The authors calculated the SE from the exact P value using methods described in the Cochrane Handbook for Systematic Reviews of Interventions (Deeks 2011). In a second paper, the ratio of geometric means was reported for baseline and end of trial measurements; however, exact P values were not available for all treatment arms and therefore the authors could not calculate the difference in mean change (Tanner 2007).

The authors had intended to group the outcomes into those measured at six-monthly intervals, but this proved impossible due to the limited amount of outcome data reported in the included trials. Please refer to the results section to determine how they grouped outcome measures. The authors analysed extracted data using Review Manager 5.1 (Review Manager 2011).

The authors presented individual trial results for each outcome in Table 1.

Table 1. Individual trial data (mean (SD)): measures of iron overload
  1. d/w = dry weight liver; w/w = wet weight liver.
    aHepatic iron only, measured as [excretion = iron transfused + [(LIC t=o -LIC t=54 weeks) x 10.6 x wt/kg)]/no of days between biopsies]; faecal iron excretion not accounted for in calculations.
    bSerum ferritin measures were given annually for up to 5 years of follow up. Only values for the first 12 months (for consistency with reporting from other trials) and for 5 years are reported in this table.
    cThe mean (SD) duration from entry into trial to time of T2* MRI basal assessment was 17 (18) months in the combined treatment group and 18 (17) in the deferiprone alone treatment group. Final evaluation was performed after a further 16 (5) months and 14 (6) months respectively. Global left ventricle heart T2* are reported here; paper also reports septum heart T2* values at baseline and endpoint.
    dGeometric mean (CV; coefficient of variation).

Trial details Endpoint (treatment duration) Intervention Serum ferritin: start Serum ferritin: end Urinary iron: start Urinary iron: end Liver iron: start Liver iron: end Myocardial T2*: start Myocardial T2*: end
Abdelrazik 200712 monthsDeferiprone + DFO4500 (1250) ng/ml1250 (750) ng/ml0.41 (0.27) mg/kg/24h0.76 (0.49) mg/kg/24hnot reportednot reportednot measurednot measured
DFO4250 (1500) ng/ml1200 (850) ng/ml0.35 (0.12) mg/kg/24h0.53 (0.21) mg/kg/24hnot reportednot reportednot measurednot measured
Aydinok 2005not reportedDeferipronenot measurednot measuredSee Table 3See Table 3not measurednot measurednot measurednot measured
DFOnot measurednot measuredSee Table 3See Table 3not measurednot measurednot measurednot measured
Deferiprone + DFOnot measurednot measuredSee Table 3See Table 3not reportednot reportednot measurednot measured
Aydinok 200712 monthsDeferiprone4070 (3223) µg/L3209 (2279) µg/L0.38 (0.22) mg/kg/day (mean on days of therapy)not measured30.7 (10.6) mg/g d/w28.6 (12.8) mg/g d/wnot measurednot measured
Deferiprone + DFO

4350 (3342) µg/L

 

2954 (2765) µg/L0.88 (0.32) mg/kg/day (mean on days of therapy)not measured26.6 (5.4) mg/g d/w18.1 (11.6) mg/g d/wnot measurednot measured
Choudhry 200412 monthsDeferiprone (50 mg)6232 (3379) ng/ml4008 (1946) ng/mlnot measurednot measurednot measurednot measurednot measurednot measured
Deferiprone (75mg)7214 (4426) ng/ml3785 (2050) ng/mlnot measurednot measurednot measurednot measurednot measurednot measured
Placebo3681 (2015) ng/ml7705 (1418) ng/mlnot measurednot measurednot measurednot measurednot measurednot measured
El-Beshlawy 200854 weeksDeferiprone2926 (1107) µg/Lapprox. 900 µg/L (estimated from graph)not reported0.38 (0.23)a mg/24h15.8 (7.1) mg/g d/wapprox. 7.5 (3.6) mg/g d/w (estimated from graph)not measurednot measured
DFO2838 (967) µg/Lapprox. 1700 µg/L (estimated from graph)not reported0.18 (0.10)a mg/24h22.5 (10.1) mg/g d/wapprox. 11.5 (6.3) mg/g d/w (estimated from graph)not measurednot measured
Deferiprone + DFO2865 (983) µg/Lapprox. 1050 µg/L (estimated from graph)not reported0.45 (0.26)a mg/24h17.1 (9.1) mg/g d/wapprox. 6.5 (3.2) mg/g d/w (estimated from graph)not measurednot measured
Galanello 200612 monthsDeferiprone + DFO2048 (685) µg/Lapprox. 1750 µg/L (estimated from graph)not measurednot measured1629 (744) µg/g w/wnot reportednot measurednot measured
DFO2257 (748) µg/Lapprox. 1850 µg/L (estimated from graph)not measurednot measured1629 (642) µg/g w/wnot reportednot measurednot measured
Gomber 20046 monthsDeferiprone2673 (886) ng/ml3423 (1581) ng/mlnot measured4.78 (2.12) mg/daynot measurednot measurednot measurednot measured
DFO5077 (1715) ng/ml3718 (738) ng/mlnot measured7.96 (5.05) mg/daynot measurednot measurednot measurednot measured
Deferiprone + DFO3348 (1526) ng/ml3377 (1222) ng/mlnot measured7.14 (4.99) mg/daynot measurednot measurednot measurednot measured
Ha (i) 2006up to 20 months (trial terminated early)Deferiprone + DFOnot reportednot reportednot measurednot measurednot reportednot reportednot measurednot measured
DFOnot reportednot reportednot measurednot measurednot reportednot reportednot measurednot measured
Ha (ii) 2006up to 20 months (trial terminated early)Deferipronenot reportednot reportednot measurednot measurednot reportednot reportednot measurednot measured
DFOnot reportednot reportednot measurednot measurednot reportednot reportednot measurednot measured
Maggio 200212 monthsDeferiprone2283 (754) ng/ml2061 (853) ng/ml11.4 (8.5) mg/24h15.8 (10.9) mg/24h3363 (5490) µg/g d/w2341 (2197) µg/g d/wnot measurednot measured
DFO2019 (678) ng/ml1787 (893) ng/ml15.7 (12.8) mg/24h19.9 (13.6) mg/24h3516 (2974) µg/g d/w3166 (2519) µg/g d/wnot measurednot measured
Maggio 2009up to 5 years (trial terminated early)Deferiprone + DFO1787 (735) ng/ml

12 mths:

1400 (770) ng/ml

5 years:

1369 (816) ng/mlb

not measurednot measuredLiver T2* 4.0 (2.9) mscLiver T2* 4.4 (3.4) msc20.1 (11.9) msc21.8 (12.6) msc
Deferiprone1890 (816) ng/ml

12 mths:

1633 (841) ng/ml

5 years:

1588 (1217) ng/mlb

not measurednot measuredLiver T2* 4.0 (5.9) mscLiver T2* 3.5 (4.3) msc25.0 (11.3) msc26.0 (11.8) msc
Mourad 200312 monthsDeferiprone + DFO4153 (1715) µg/L

6 mths:

3005 (1303) µg/L

12 mths:

2805 (1084) µg/L

not measured22.9 (19.2) mg/kg/24hrnot measurednot measurednot measurednot measured
DFO5506 (2375) µg/L

6 mths:

4856 (2615) µg/L

12 mths:

3998 (2409) µg/L

not measurednot measurednot measurednot measurednot measurednot measured
Olivieri 19906 daysDeferipronenot measurednot measurednot measured12.3 (6.7) mg/kg/daynot measurednot measurednot measurednot measured
DFOnot measurednot measurednot measured18.2 (15.3) mg/kg/daynot measurednot measurednot measurednot measured
Olivieri 199724 monthsDeferipronenot measurednot measurednot reportednot reported8.9 (5.23) mg/g d/w13.7 (5.23) mg/g d/wnot measurednot measured
DFOnot measurednot measurednot reportednot reported6.9 (3.82) mg/g d/w7.9 (5.52) mg/g d/wnot measurednot measured

Pennell 2006

 

12 monthsDeferiprone

1791 (1029) µg/L

 

1610 (980.91) µg/L

 

not measurednot measured6.16 (6.0) mg/g d/wnot reported13.0 ms (32%)d16.5 ms (+27%, CV 38%)d
DFO2795 (2441) µg/L

2329 (2078.7) µg/L

 

not measurednot measured6.32 (5.8) mg/g d/wnot reported13.3 ms (30%)d15.0 ms (+13%, CV 39%)d
Tamaddoni 201012 monthsDeferiprone + DFO2986 (612) ng/ml

6 mths: 2453 (318) ng/ml

12 mths: 2082 (221) ng/ml

not measurednot measurednot reportednot reportednot measurednot measured
DFO2945 (591) ng/ml

6 mths: 2702 (242) ng/ml

12 mths: 2451 (352) ng/ml

not measurednot measurednot reportednot reportednot measurednot measured
Tanner 200712 monthsDeferiprone + DFO1574 (CV 11) µg/Ld598 (CV 14) µg/Ldnot measurednot measuredLiver T2* 4.9 (CV 0.52) ms [>19]d10.7 msd11.7 (CV 0.08) ms [>20]d17.7 msd
DFO + placebo1379 (CV 10) µg/Ld1146 (CV 11) µg/Ldnot measurednot measuredLiver T2* 4.2 (CV 0.62) ms [>19]d5.0 msd12.4 (CV 0.11) ms [>20]d15.7 msd

Unit of analysis issues

For the original review, the authors extracted data from one cross-over trial using the same data extraction form as for the parallel-design trials. They took care to extract and record baseline, intervention and outcome data by cross-over arm. However, this trial reported baseline data across all participants, and outcome data were reported by intervention (as per a parallel design trial) and not by the sequence of the intervention.

There were no further unit of analysis issues in the update.

Dealing with missing data

For the original review, authors requested missing data from the investigators of two trials,who supplied these missing data. In addition, during the analysis of results, authors contacted three trial authors by e-mail, requesting the individual patient data (IPD) from their trial. All three responded to the initial e-mail but were unwilling or unable to provide their IPD. The authors did not contact investigators of six trials to request IPD. A tenth trial provided IPD in the report of their trial.

In the update, the authors did not contact any trial investigators requesting missing data due to time constraints and resource limitations.

Assessment of heterogeneity

The authors assessed statistical heterogeneity of treatment effects between trials by using a chi-squared test with a significant level at P < 0.1. They used the I2 statistic to quantify the amount of possible heterogeneity (I2 greater than 30% moderate heterogeneity; and I2 greater than 75% considerable heterogeneity) (Higgins 2002; Higgins 2003). The number of trials included did not allow further exploration of heterogeneity by sensitivity and subgroup analysis in any of the meta-analyses undertaken.

Assessment of reporting biases

Although the authors did not undertake any quantitative assessment of publication bias, they made attempts to minimise the likelihood of publication bias by the use of a comprehensive search strategy, the handsearching of relevant conference abstract books and contacting the manufacturers of deferiprone and other iron chelators. They plan to undertake an assessment of publication bias in future updates of this review, if they identify a sufficient number of studies.

Data synthesis

Authors undertook meta-analyses using RevMan 5.1 (Review Manager 2011). They used a fixed-effect model for combining data in the first instance. Where they identified considerable heterogeneity in a fixed-effect meta-analysis, they repeated the analysis using a random-effects model. For many outcomes, meta-analysis was prohibited due to diversity in the method or timing of the outcome measurement across trials. Authors also did not undertake any meta-analysis where differences in clinical outcome baseline levels occurred and they deemed the risk of selection bias to be unclear or high. They based conclusions on inferences drawn from clearly tabulated results from included trials, as well as on qualitative and quantitative summary measures. They considered both direction and magnitude of effects. They present individual trial results in additional tables.

Analyses based on means are not appropriate for heavily skewed data. Where authors knew data were skewed, they performed analyses on a log scale and presented these data as the ratio of geometric means.

For analyses of adverse events, the authors pooled trials reporting the same adverse event in a meta-analysis only if both arms of the trial reported occurrences of adverse events or if the absence of an adverse event in either (but not both) treatment arm was specifically reported or could be clearly inferred without ambiguity. They only undertook meta-analysis of dose reduction or temporary or permanent withdrawal when the total number of adverse events was clearly stated.

They analysed outcome data from different time points separately in this review.

Subgroup analysis and investigation of heterogeneity

Although intended, due to an insufficient number of trials, the authors were unable to perform subgroup analysis for any of the outcome measures. If sufficient trials become available in future updates of this review, the authors plan to undertake a subgroup analysis of pre-defined measures of iron overload or end organ damage.

Sensitivity analysis

The authors did not undertake any sensitivity analyses due to the paucity of trials included in meta-analyses. They will undertake sensitivity analyses in future updates of this review if they identify sufficient trials.

Results

Description of studies

Results of the search

In total, the searches identified 2974 references updated to March 2013. Initial screening and de-duplication of the citations and trials for relevance by one author (CD) excluded 2081 references. The titles and abstracts of 893 references were then screened by two authors independently and 793 were excluded for not meeting the review's eligibility criteria. The remaining 100 references were assessed on the basis of their full text for inclusion or exclusion using the criteria indicated above. Of these, 36 trials, comprising 40 references, were excluded and 13 trials comprising 20 references were included as 'Studies awaiting classification'. This left 17 trials (40 references) for inclusion in the review. See Figure 1 for PRISMA study flow diagram.

Figure 1.

PRISMA study flow diagram.

Included studies

Seventeen trials described in 40 papers (including secondary publications) with 1061 participants (range 13 to 213) were included in this review. One paper presented data for two separate trials: the trials will be presented separately in this review (Ha (i) 2006; Ha (ii) 2006). Sixteen trials were parallel in design and one was a cross-over trial (Olivieri 1990). Four trials were three-arm comparisons (Aydinok 2005; Choudhry 2004; El-Beshlawy 2008; Gomber 2004). Three of these trials compared deferiprone alone, deferiprone and DFO in combination and DFO alone (Aydinok 2005; El-Beshlawy 2008; Gomber 2004). These three trials were subdivided by intervention arms for outcome analysis within this review.

Sixteen trials involved a comparison of deferiprone alone with DFO alone, or either DFO or deferiprone as monotherapy compared with combined deferiprone and DFO. From these 16 individual trials there were eight comparisons between deferiprone alone and DFO alone (Aydinok 2005; El-Beshlawy 2008; Gomber 2004; Ha (ii) 2006; Maggio 2002; Olivieri 1990; Olivieri 1997; Pennell 2006); five comparisons between deferiprone alone and deferiprone in combination with DFO (Aydinok 2005; Aydinok 2007; El-Beshlawy 2008; Gomber 2004; Maggio 2009) and nine comparisons between deferiprone in combination with DFO and DFO alone (Abdelrazik 2007; Aydinok 2005; El-Beshlawy 2008; Galanello 2006; Gomber 2004; Ha (i) 2006; Mourad 2003; Tamaddoni 2010; Tanner 2007). One trial compared two doses of deferiprone (Choudhry 2004). No trials comparing deferiprone with placebo were identified.

These trials were published between 1990 and 2010 and conducted internationally: three in Italy; two in Canada; two in Egypt; two in Italy and Greece, two in Hong Kong; two in India; two in Turkey and one each in Iran and the Lebanon. Two trials were presented as an abstract (Aydinok 2005; Olivieri 1997). One of these trials has never been published as a full journal article, although the primary author of this study indicated that additional information was published on the FDA website by the trial sponsors and these data have been included in this review (Olivieri 1997). There is no record on any bibliographic database of a full journal article for the second trial (Aydinok 2005). The other 15 trials were published as full journal articles. All included trials were published in English.

Trial design characteristics - interventions
(A) Deferiprone alone compared with DFO alone

There were eight comparisons of deferiprone alone with DFO alone (Aydinok 2005; El-Beshlawy 2008; Gomber 2004; Ha (ii) 2006; Maggio 2002; Olivieri 1990; Olivieri 1997; Pennell 2006).

In seven of these trials deferiprone was given daily at 50 to 100 mg/kg/day in two or three oral divided doses (Aydinok 2005; El-Beshlawy 2008; Gomber 2004; Ha (ii) 2006; Maggio 2002; Olivieri 1990; Pennell 2006). Treatment with DFO was given as a subcutaneous injection at a dose of 20 to 60 mg/kg/day for 8 to 12 hours and for 5 to 7 days a week. In an eighth trial, the intended dose of chelators was not reported, however, the mean (and SD) dose received was reported as 36.7 (2.8) mg/kg/night.. This trial was stopped prematurely by the sponsoring pharmaceutical company (Apotex) (Olivieri 1997).

In six of the eight trials, DFO was administered subcutaneously at a dose of between 23 and 60 mg/kg/day over 5 and 7 days per week (Aydinok 2005; El-Beshlawy 2008; Gomber 2004; Ha (ii) 2006; Maggio 2002; Pennell 2006). One cross-over trial designed to compare excretion of iron in urine and stool, administered DFO subcutaneously over 12 hours per day for three days (Olivieri 1990). The details of the dose or schedule of administration for DFO was not reported in the eighth trial (Olivieri 1997).

The duration of trial treatment lasted six months (Gomber 2004), one year (El-Beshlawy 2008; Maggio 2002; Pennell 2006) and two years (Olivieri 1997). The intended duration of treatment in one trial was 18 months (Ha (ii) 2006). However, this trial was stopped early: treatment duration ranged from 1 to 20 months (median duration 18 months). In the cross-over trial, the duration of treatment was three days per treatment arm with each participant receiving treatment over six days. There was a gap of three to four weeks between each treatment arm for participants (Olivieri 1990). Duration of treatment was not stated in one trial (Aydinok 2005).

Participants in four trials had previously been exposed to DFO (Aydinok 2005; Maggio 2002; Olivieri 1990; Pennell 2006). In one trial, administration of DFO (dose unreported) had ceased 72 hours preceding the start of the trial (Olivieri 1990). In two trials, prior to enrolment, participants had received DFO at a dose of 50 mg/kg as a subcutaneous infusion over 12 hours per night, five times a week (Maggio 2002) or at a mean (SD) dose of 39 (8) mg/kg/day for 5 to 7 days per week (Pennell 2006). The dose of DFO treatment prior to enrolment was not reported in the fourth trial (Aydinok 2005).

(B) Deferiprone alone compared with deferiprone and DFO in combination

There were five comparisons between deferiprone alone with deferiprone and DFO in combination (Aydinok 2005; Aydinok 2007; El-Beshlawy 2008; Gomber 2004; Maggio 2009). In these trials, the schedule for deferiprone only was 75 mg/kg/day (Aydinok 2005; Aydinok 2007; Gomber 2004; Maggio 2009) or between 60 and 83 mg/kg/day (El-Beshlawy 2008), administered orally in divided doses. In all five trials, DFO was administered as a subcutaneous infusion at a dose of 20 to 50 mg/kg/day for two or three days a week.

The duration of trial treatment was six months in one trial (Gomber 2004) and one year in two trials (Aydinok 2007; El-Beshlawy 2008) but was not stated in the fourth trial (Aydinok 2005). In the fifth trial, the planned duration was five years but the trial was stopped early because of beneficial effects in terms of serum ferritin reduction (Maggio 2009). In this trial, the mean (SD) duration of treatment was 2.5 (2.2) years for deferiprone only and 2.9 (2.1) years for deferiprone in combination with DFO.

Three trials reported prior iron chelation therapy. Prior exposure to DFO was reported in two trials, although details of the dose and schedule were not reported (Aydinok 2005; Aydinok 2007) . A third trial reported prior treatment with either DFO (50 mg/kg/day, 8 to 12 hours for 5 days per week) or deferiprone (75 mg/kg/day for 7 days per week) although the duration of prior treatment was not reported (Maggio 2009).

(C) Deferiprone and DFO in combination compared with DFO alone

There were nine comparisons between deferiprone and DFO in combination with DFO alone (Abdelrazik 2007; Aydinok 2005; El-Beshlawy 2008; Galanello 2006; Gomber 2004; Ha (i) 2006; Mourad 2003; Tamaddoni 2010; Tanner 2007).

Deferiprone was given orally at a dose of 25 mg/kg/day (Galanello 2006), 75 mg/kg/day (Aydinok 2005; Abdelrazik 2007; Gomber 2004; Ha (i) 2006; Mourad 2003; Tamaddoni 2010; Tanner 2007), or 60 to 83 mg/kg/day (El-Beshlawy 2008). These oral divided doses were given daily in six trials (Aydinok 2005; El-Beshlawy 2008; Gomber 2004; Ha (i) 2006; Mourad 2003; Tanner 2007), five days per week in two trials (Galanello 2006; Tamaddoni 2010) or for four days per week in one trial (Abdelrazik 2007).

When given in combination with deferiprone, DFO was given twice weekly in all nine trials, at a daily dose of between 23 to 60 mg/kg/day (Abdelrazik 2007; Aydinok 2005; El-Beshlawy 2008; Galanello 2006; Gomber 2004; Ha (i) 2006; Tamaddoni 2010; Tanner 2007) or 2 g (Mourad 2003). The schedule for DFO alone ranged from 20 to 60 mg/kg administered 5 to 7 days a week for all nine trials. One trial also administered an oral placebo; no further details are reported (Tanner 2007).

The duration of treatment was six months (Gomber 2004), 12 months (Abdelrazik 2007; El-Beshlawy 2008; Galanello 2006; Mourad 2003; Tamaddoni 2010; Tanner 2007) and between one and 20 months (Ha (i) 2006). Although in the latter trial, only participants who had received at least six months of treatment were included in the assessment of efficacy (Ha (i) 2006). One trial did not report treatment duration (Aydinok 2005).

Six of the nine trials reported prior exposure to iron chelation therapy (Abdelrazik 2007; Aydinok 2005; Galanello 2006; Mourad 2003; Tamaddoni 2010; Tanner 2007); all six trials reported prior DFO treatment although only two reported the dose and schedule of prior treatment received. In one trial, patients received DFO at a dose of 40 mg/kg/day for 5 to 7 nights per week for "several years" (Abdelrazik 2007). A further trial reported prior exposure to DFO at a mean (SD) dose of 36.4 (11.1) mg/kg/day for 5.5 days per week but treatment duration was not reported (Tanner 2007). In this trial, participants were excluded from the trial if they had previously received deferiprone. A third trial reported prior exposure to DFO less than four times per week but the dose and duration was not reported (Mourad 2003).

(D) Deferiprone dose A versus deferiprone dose B

One trial compared doses of deferiprone (Choudhry 2004). The treatment schedules for this trial were either 50 mg/kg/day or 75 mg/kg/day. The duration of trial treatment was one year. No participants had received iron chelation therapy prior to trial enrolment.

Excluded studies

Following full text eligibility assessment, 36 trials were excluded from the review. See the 'Characteristics of excluded studies' table for further details. In summary, 25 were not RCTs, six trials did not include an eligible comparator arm, in three trials all participants received all interventions and two were commentaries on other studies (one excluded and one included). One of the trials included in the original review was excluded in this update as participants in this cross-over trial were not randomised to their first treatment arm (Fassos 1994).

Studies awaiting classification

Three trials included in this section in the original review have now been moved to included studies (Choudhry 2004; Galanello 2006; Tanner 2007). A further two have been excluded: both were from the same trial (Grady 2002).

Thirteen trials described in 25 papers are currently listed as studies awaiting classification, see 'Characteristics of studies awaiting classification' for further details. The ongoing trial from the original review (NCT00115349a) and one additional trial (N0277104959) identified from ongoing trials databases have been classified under 'Studies awaiting classification'. We have been unable to identify any publications resulting from either trial. An additional two trials published in abstract form only have been included as 'Studies awaiting classification'; no further publications relating to these trials have been identified and the trial details contained in these abstracts were deemed insufficient for inclusion (Badawy 2010; Unal 2009). A further study could not be conclusively identified as randomised; this study requires translation to further investigate the validity of the methods used before inclusion in the review (Kompany 2009). Eight trials identified from the final search period (June 2011 to March 2013) are also potentially relevant and have been included as 'Studies awaiting classification' although have not been fully evaluated for eligibility (Alpendurada 2012; Aydinok 2012; Evans 2011; Jain 2011; Maggio 2012; Mirbehbahani 2012; Pantalone 2011;Pepe 2013).

Risk of bias in included studies

The risk of bias across all included trials is summarised in Figure 2 and detailed below.

Figure 2.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation

Six trials provided details regarding the methods used to generate the randomisation sequence. Two trials were deemed as having a low risk of bias; one used sealed envelopes (Choudhry 2004) and another trial used randomisation codes generated and maintained at a site external to the trial site (Aydinok 2007). Four trials reported permuted block randomisation and the risk of bias was classified as unclear (Ha (i) 2006; Ha (ii) 2006; Maggio 2002; Maggio 2009). In the remaining 11 trials no description was given as to which methods were used to generate the random sequence, hence these were also classified as having an unclear risk of bias (Abdelrazik 2007; Aydinok 2005; El-Beshlawy 2008; Galanello 2006; Gomber 2004; Mourad 2003; Olivieri 1990; Olivieri 1997; Pennell 2006; Tamaddoni 2010; Tanner 2007).

The method of randomisation (as described above) was deemed adequate to conceal treatment allocation with a low risk of bias in three trials (Choudhry 2004; Maggio 2002; Maggio 2009). Two trials reported that the randomisation sequence was not concealed prior to allocation and the concealment of allocation was therefore deemed inadequate with a high risk of bias (Aydinok 2007; Tanner 2007).

The risk of bias associated with concealment of treatment allocation was defined as unclear in the remaining 12 trials, as no description was given of the methods used to conceal the allocation of treatment from the clinicians (Abdelrazik 2007; Aydinok 2005; El-Beshlawy 2008; Galanello 2006; Gomber 2004; Ha (i) 2006; Ha (ii) 2006; Mourad 2003; Olivieri 1990; Olivieri 1997; Pennell 2006; Tamaddoni 2010).

Blinding

In 12 trials the blinding of participants, clinicians and outcome assessors was not reported and the risk of bias in these trials was unclear (Aydinok 2005; Aydinok 2007; Choudhry 2004; El-Beshlawy 2008; Galanello 2006; Gomber 2004; Ha (i) 2006; Ha (ii) 2006; Mourad 2003; Olivieri 1990; Olivieri 1997; Tanner 2007). Blinding of participants and clinicians was also unclear in a further trial, although the authors of this trial did state that cardiac treatment was undertaken by a clinician blinded to treatment allocation (Abdelrazik 2007).

The blinding of participants and clinicians was not undertaken in four trials, but this is unsurprising given the difference in the methods of administration between treatment arms (Maggio 2002; Maggio 2009; Pennell 2006; Tamaddoni 2010). However, in all four of these trials, outcome assessors were blinded to treatment allocation which was deemed adequate and the these trials were therefore considered to have a low risk of bias.

Incomplete outcome data

All but one of the included trials lost fewer than 20% of the randomised participants by the time of statistical analysis of the trial data (Olivieri 1997). All participants were analysed in the treatment groups to which they had been randomised, with the exception of one trial (Gomber 2004).

Nine trials were considered to have a low risk of bias (Abdelrazik 2007; Aydinok 2005; Galanello 2006; Ha (ii) 2006; Maggio 2002; Mourad 2003; Olivieri 1990; Pennell 2006; Tamaddoni 2010).

Seven trials were considered to have a high risk of bias due to incomplete data (Aydinok 2007; Choudhry 2004; El-Beshlawy 2008; Gomber 2004; Ha (i) 2006; Maggio 2009; Olivieri 1997).

An imbalance in missing data across the treatment arms was noted in three trials (Aydinok 2007; Choudhry 2004; Gomber 2004). In the first of these, four withdrawals were reported; these all occurred in the DFO group (Aydinok 2007). A second trial reported three withdrawals associated with the outcome of mortality only in participants receiving deferiprone, although the authors did not specify in which of the two deferiprone treatment groups these withdrawals occurred (Choudhry 2004). All randomised participants were analysed for all other outcomes in this trial. One trial lost three (30%) of the 10 randomised participants in the DFO alone group: two participants were excluded and one participant changed treatment groups (Gomber 2004). No reasons were given as to why two participants were excluded from the trial. Follow-up data were not presented for the two excluded participants. End of trial data were analysed according to the treatment received rather than the treatment group to which the participants were randomised. Data from this trial were included in a meta-analysis within this review as it was clear to the authors which data were missing and what data analysis had been undertaken within the trial. However, given these noted concerns, the results from the meta-analysis need to be read with caution (Gomber 2004).

A further trial with a high risk of bias reported variable numbers of participants included in the analysis for each outcome (El-Beshlawy 2008). Whilst the number of withdrawals and reasons were reported, the number of participants withdrawn from the trial conflicts with the number included in each analysis.

Three trials were stopped prematurely (Ha (i) 2006; Maggio 2009; Olivieri 1997). One trial lost 46% of randomised participants (Olivieri 1997). This trial collected and reported outcome data for participants who had completed two years of treatment. Of 71 participants entering the trial, 13 (18%) had withdrawn and 20 (28%) had not completed two years on follow up. Outcome data were presented for 54% of randomised participants. This trial was stopped prematurely by the sponsoring pharmaceutical company (Apotex) when concerns were raised about the safety and effectiveness of deferiprone and the company made, as yet unproven allegations, over the conduct of the trial (Nathan 2002; Viens 2004). Two trials were prematurely stopped after 18 months. In one trial, an unexpected sudden death of a participant in the deferiprone arm prompted the termination of the trial (Ha (i) 2006). The second trial sought to demonstrate an equivalence of treatment, with 80% power and a 0.2 chance of committing a Type 1 error. The trial investigators stated that to achieve this, 26 participants were needed, but to account for a large dropout rate, a total of 60 participants would be required (Ha (ii) 2006). A further trial was stopped early before the planned five years of treatment were completed for all patients due to observed beneficial effects in terms of serum ferritin levels reduction in one of the treatment arms (Maggio 2009). In this trial 33 participants had withdrawn prior to termination of the trial.

The risk of bias was deemed to be unclear in one further trial (Tanner 2007). Tanner reported seven participants (four in the treatment group and three in the comparator group) who withdrew from the trial but the number of participants included in the final outcome assessment was not reported and the risk of bias due to incomplete outcome data was therefore unclear (Tanner 2007).

Selective reporting

Seven of the included trials were considered to have some risk of bias due to selective reporting, as they failed to report pre-specified outcomes (Abdelrazik 2007; Aydinok 2005; Galanello 2006; Gomber 2004; Mourad 2003; Pennell 2006; Tamaddoni 2010). Outcome measures which were pre-specified in individual trials but not subsequently reported included liver iron concentration (Abdelrazik 2007; Aydinok 2005), total iron binding capacity (Abdelrazik 2007), ferritin concentrations (Abdelrazik 2007), iron excretion (Mourad 2003), liver function tests (Gomber 2004; Tamaddoni 2010), hepatic markers (Gomber 2004), monthly blood counts (Gomber 2004), renal function (Mourad 2003; Tamaddoni 2010), absolute neutrophil count (Pennell 2006; Tamaddoni 2010), alanine aminotransferase (ALT) (Pennell 2006), serum zinc levels (Pennell 2006) and serum creatinine levels (Pennell 2006). In addition, compliance was pre-specified as an outcome in two trials (Galanello 2006; Tamaddoni 2010) but no compliance data were reported.

The remaining ten trials had a low risk of bias for this domain (Aydinok 2007; Choudhry 2004; El-Beshlawy 2008; Ha (i) 2006; Ha (ii) 2006; Maggio 2002; Maggio 2009; Olivieri 1990; Olivieri 1997; Tanner 2007).

Other potential sources of bias

Support and sponsorship

Eleven trials documented sources of funding (Aydinok 2007; Choudhry 2004; Galanello 2006; Ha (i) 2006; Ha (ii) 2006; Maggio 2002; Mourad 2003; Olivieri 1990; Pennell 2006; Tamaddoni 2010; Tanner 2007). Two trials received funding from a governmental agency: Medical Research Council of Canada (Olivieri 1990); and a European Community grant (Maggio 2002). One trial was supported by university funds (Tamaddoni 2010) and four received funding from a thalassaemia society: Children's Thalassaemia Foundation in Hong Kong (Ha (i) 2006; Ha (ii) 2006); Sicilian Thalassaemic Association (Maggio 2002); Cooley's Anemia Foundation (Tanner 2007) and UK Thalassaemia Society (Tanner 2007). The Tanner trial also acknowledged funding from CORDA as well as other charitable organisations.

Four trials documented sponsorship by the manufacturer of deferiprone: Apotex (Galanello 2006; Pennell 2006; Tanner 2007) or CIPLA Ltd (Choudhry 2004). Four trials reported the supplier of the trial treatment deferiprone: Lipomed, Switzerland (Aydinok 2007; Ha (i) 2006; Ha (ii) 2006; Mourad 2003). Two trials reported the supplier of DFO: Novartis, Basel, Switzerland (Ha (i) 2006; Ha (ii) 2006).

In one trial, although sources of funding were not reported, the authors declared that the trial was conducted "without the influence of the non-commercial sponsor" (Maggio 2009).

Power calculations

Six trials documented power calculations (Ha (i) 2006; Ha (ii) 2006; Maggio 2002; Maggio 2009; Pennell 2006; Tanner 2007). All of these trials calculated sample sizes required to obtain 80% power to detect an effect. One trial calculated statistical power based on a 30 ng/ml difference in mean serum ferritin concentration after one year of DFO therapy (Maggio 2002) and two trials based power calculations on a difference in myocardial T2* of 5% (Pennell 2006) or a change of 0.75 standard deviations (Tanner 2007); all of these trials achieved their target sample sizes. One trial calculated sample sizes based on repeated measure analyses; this trial achieved the target recruitment sample sizes but not all participants completed five years of treatment due to early termination of the trial (Maggio 2009). Two trials calculated power based on liver iron content of participants in earlier pilot studies (Ha (i) 2006; Ha (ii) 2006). Neither of these studies achieved their target recruitment sample sizes due to the trials being stopped early.

The remaining 11 trials did not describe any power calculations (Abdelrazik 2007; Aydinok 2005; Aydinok 2007; Choudhry 2004; El-Beshlawy 2008; Galanello 2006; Gomber 2004; Mourad 2003; Olivieri 1990; Olivieri 1997; Tamaddoni 2010).

Effects of interventions

Details of the baseline and end of trial values for serum ferritin concentration, urinary iron excretion, liver iron concentration and myocardial T2* reported by the included trials are presented in Table 1.

Results are presented for each comparison. Where there is disparity between trials in the method of outcome data reporting between trials which prohibits the calculation of mean change from baseline in all trials, outcome results are reported as both end of trial data and mean change from baseline. Outcome data in the three multi-arm trials (three arms) are presented in the results sections (A to C) due to the nature of the interventions in these trials (Aydinok 2005; El-Beshlawy 2008; Gomber 2004).

Most biochemical measures have a set of values known as 'normal ranges' defined with reference to a population sample of healthy people. Current normal ranges are 17 ng/ml to 30 ng/ml (or µg/L) for serum ferritin, up to 1.8 mg Fe/g dry weight for liver iron concentration and > 20 ms for myocardial T2* as a measure of cardiac iron.

A. Deferiprone alone compared with DFO alone

There were eight comparisons of deferiprone alone with DFO alone (Aydinok 2005; El-Beshlawy 2008; Gomber 2004; Ha (ii) 2006; Maggio 2002; Olivieri 1990; Olivieri 1997; Pennell 2006).

Primary outcome
1. Mortality

Only one trial reported mortality as an outcome (Ha (ii) 2006). One death occurred in the deferiprone treatment arm after six months of treatment; this death was attributed to cardiac complications and thought not to be related to deferiprone treatment.

Secondary outcomes
1. Evidence of reduced end organ damage

Four trials reported cardiac function as an outcome (El-Beshlawy 2008; Maggio 2002; Olivieri 1997; Pennell 2006). One trial did not report data, only that "there was no significant difference in cardiac function" between treatment arms (El-Beshlawy 2008). In the Maggio trial, cardiac function was measured by sonography to determine LVEF (Maggio 2002). In the Pennell trial, cardiac function was assessed using cardiovascular magnetic resonance to measure left ventricular (LV) and right ventricular (RV) end-systolic and end-diastolic volume (ESV, EDV) and ejection fraction (EF) (Pennell 2006). In the remaining trial, the method of measurement was not reported, however, unpublished data for LVEF were obtained from the original investigators (Olivieri 1997).

At six months, the Pennell trial showed no significant difference in mean change of LVEF from baseline between treatment arms, MD 1.48% (95% CI -0.08 to 3.04) (Pennell 2006) (Analysis 1.1). However, at 12 months, meta-analysis of three trials showed a significant difference in mean change from baseline for LVEF between treatment arms in favour of deferiprone, MD 1.60% (95% CI 0.24 to 2.97) (Maggio 2002; Olivieri 1997; Pennell 2006) (Analysis 1.1). Considerable heterogeneity was observed between these three trials (I2 = 75%) and the MD using a random-effects model was no longer significant between treatment arms, MD 1.76% (95% CI -1.42 to 4.93) (Analysis 1.1). No clear clinical differences between the trials were identified which could account for this heterogeneity, although baseline LVEF values were lower (62% and 63%) in the Maggio 2002 trial than those in the Pennell 2006 trial (68.4%, 69.7%). No significant difference in mean change in LVEF was observed between treatment groups at 24 months as reported by one trial (Olivieri 1997): MD 7.60% (95% CI -0.65 to 15.85) (Analysis 1.1)

Data to calculate mean change from baseline were not available for right ventricular ejection fraction (RVEF); however no significant differences were observed in mean RVEF at six months, MD 0.60% (95% CI -1.88 to 3.08), or 12 months, MD 2.30% (95% CI -0.22 to 4.82) (Pennell 2006) (Analysis 1.2).

Two trials reported liver fibrosis as an outcome although the Ha trial did not report results separately for each treatment arm (Ha (ii) 2006; Maggio 2002). In the Maggio trial, liver fibrosis was scored according to the Ishak scoring system (Maggio 2002). There was no significant difference in the mean of fibrosis Ishak scores between treatment arms after 12 months, MD -0.10 (95% CI -0.98 to 0.78) (Maggio 2002) (Analysis 1.3).

Evidence of reduced end organ damage as an outcome was not reported by any of the remaining trials.

2. Measures of iron overload

The aim of iron chelation therapy is to reduce serum ferritin and liver iron concentration and increase urinary iron excretion and myocardial T2* measures.

a. Serum ferritin concentration

Six trials reported serum ferritin concentration as an outcome (El-Beshlawy 2008; Gomber 2004; Ha (ii) 2006; Maggio 2002; Olivieri 1997; Pennell 2006). Data to calculate mean change in serum ferritin concentration from baseline to end of trial were available in five trials (Gomber 2004; Ha (ii) 2006; Maggio 2002; Olivieri 1997; Pennell 2006). A further trial presented data graphically and the mean change has been estimated from the graph; SDs were not reported in this trial (El-Beshlawy 2008). Individual trial data for serum ferritin concentration are presented in an additional table (Table 1); mean change from baseline data are shown in a further table (Table 2). The results from these trials were not pooled overall due to the pronounced baseline differences between the treatment arms in two trials (Gomber 2004; Pennell 2006) and the variation in treatment duration: six months (Gomber 2004; Ha (ii) 2006; Pennell 2006); 12 months (Maggio 2002; Pennell 2006); and 24 months (Olivieri 1997).

Table 2. Serum ferritin concentration (mean (SD)): deferiprone versus DFO
Trial No Deferiprone/DFO Deferiprone:baseline Def: end trial Absolute change % change DFO: baseline DFO: end trial Absolute change % change
El-Beshlawy 200818/202926 (1107) µg/Lapprox. 900 µg/L (estimated from graph)

reduction of approx. 2026 µg/L

 

not available2838 (967) µg/Lapprox. 1650 µg/L (estimated from graph)

reduction of approx. 1188 µg/L

 

not available
Gomber 200411/72672.9 (886.4) ng/ml3422.7 (1581.0) ng/mlincrease of 749.8 (1155.8) ng/mlincrease of 28%5077.2 (1715.0) ng/ml3718.3 (738.4) ng/mlreduction of 1358.9 (1374.14) ng/mlreduction of 27%
Ha (ii) 20066/7not reportednot reportedincrease of 398.4 (1073) mmol/Lnot availablenot reportednot reportedreduction of 74.2 (1629) mmol/Lnot available
Maggio 200271/ 732283 (754) ng/ml2061 (853) ng/mlreduction of 222 (783) ng/mlreduction of 9.7%2019 (678) ng/ml1787 (893) ng/mlreduction of 232 (619) ng/mlreduction of 11.4%
Pennell 200629 /321791 (1029) µg/L1610 (980.91) µg/Lreduction of 181 (826) µg/Lreduction of 10.1%2795 (2441) µg/L2329 (2078.7) µg/Lreduction of 466 (739) µg/Lreduction of 16.7%

At six months, two trials showed a statistically significant difference in mean change in serum ferritin concentration from baseline in favour of DFO, MD 2108.62 ng/ml (95% CI 882.76 to 3334.48) (Gomber 2004); MD 465.00 ng/ml (95% CI 53.70 to 876.30) (Pennell 2006) (Analysis 1.4). In a third trial, the mean change in serum ferritin concentration was not significantly different between the two treatment arms, MD -324.20 ng/ml (95% CI -1805.21 to 1156.81) (Ha (ii) 2006) (Analysis 1.4).

At 12 months, neither of two trials showed a significant difference in mean change in serum ferritin concentration between treatment arms: MD 10.00 ng/ml (95% CI -220.94 to 240.94) (Maggio 2002); MD 285.00 ng/ml (95% CI -109.9 to 679.89) (Pennell 2006) (Analysis 1.4).

At 24 months, there was no significant difference in mean change in serum ferritin concentration between treatment arms, MD 185.00 ng/ml (95% CI -270.52 to 640.52) (Olivieri 1997).

b. Urinary iron excretion

Six trials measured urinary iron excretion (Aydinok 2005; El-Beshlawy 2008; Gomber 2004; Maggio 2002; Olivieri 1990; Olivieri 1997), but only one of these trials presented data to calculate mean change in urinary iron excretion (Maggio 2002). Of the remaining five trials, three reported mean urinary iron excretion after treatment but did not report baseline values (El-Beshlawy 2008; Gomber 2004; Olivieri 1990). One trial reported mean percentage urinary iron excretion over the period of study, calculated as mean urinary iron excretion divided by total iron excretion (Aydinok 2005). The fifth trial reported only that "the change in urinary iron excretion from baseline to month 24 was not different between DFO and deferiprone subjects" (Olivieri 1997). Individual trial data are presented in additional tables (Table 1; Table 3).

Table 3. Mean (SD) measures of iron overload: Aydinok 2005
  1. DFO: desferrioxamine
    HPLC: high performance liquid chromatography
    LIC: liver iron concentration
    mM: millimolar (denotes a concentration of 1 micromole per litre)
    NTBI: non-transferrin bound iron

Treatment regimen Total iron excretion Plasma NTBI change Chelation efficiency Urinary iron excretion
Deferiprone0.56 (0.09) mg/kg/day3.00 (0.53) mM6.65 (0.99) %98 (14.8) %
DFO0.77 (0.25) mg/kg/day0.72 (0.45) mM23.1 (10.7) %28 (10.3) %
Deferiprone + DFO0.69 (0.09) mg/kg/day0.84 (0.87) mM7.34 (0.91) %84 (10.6) %
CommentMean value from across 5 measurement points: weeks 1, 12, 26, 38, 54. Calculated as [iron transfused/year (mg) + (LIC at time 0 - LIC at time 1 year) x 10.6 x body weight in kg]/ number of days of treatment.Plasma NTBI was measured by HPLC at baseline and at weeks 1, 12, 26 and 54. Data were not reported for the outcome measurements at these time points.Chelation efficiency calculated as [iron excretion / chelator dose] multiplied by [molecular weight of the respective chelator /56] x n x 100, where 56 is the molecular weight of iron and n is 3 or 1 with deferiprone and DFO respectively.Urine iron (%) mean over study calculated from mean urine iron excretion / total iron excretion.

Six trials measured urinary iron excretion (Aydinok 2005; El-Beshlawy 2008; Gomber 2004; Maggio 2002; Olivieri 1990; Olivieri 1997), but only one of these trials presented data to calculate mean change in urinary iron excretion (Maggio 2002). Of the remaining five trials, three reported mean urinary iron excretion after treatment but did not report baseline values (El-Beshlawy 2008; Gomber 2004; Olivieri 1990), and one trial reported mean percentage urinary iron excretion over the period of study (defined as mean urinary iron excretion divided by the total iron excretion) (Aydinok 2005). The fifth trial (Olivieri 1997) reported only that "the change in urinary iron excretion from baseline to month 24 was not different between DFO and deferiprone subjects". Individual trial data are presented in additional tables (Table 1; Table 3).

i. At end of trial

In two trials there were statistically significant differences in mean urinary iron excretion between the treatment arms, favouring deferiprone in one trial, MD 0.20 mg/24h (95% CI 0.08 to 0.32) (El-Beshlawy 2008) and favouring DFO in the other trial, MD -4.10 mg/24h (95% CI -8.12 to -0.08) (Maggio 2002) (Analysis 1.5). There was no statistically significant difference in mean urinary iron excretion between the treatment arms in the other two trials, MD -3.18 mg/24h (95% CI -7.13 to 0.77) (Gomber 2004) and MD -5.90 mg/24h (95% CI -13.22 to 1.42) (Olivieri 1990) (Analysis 1.5). In one trial there was a significant difference in mean % urinary iron excretion over the study period, MD 70.00% (95% CI 57.69 to 82.31) (Aydinok 2005) (Analysis 1.6).

The data for this outcome were not pooled overall in a meta-analysis because the time points for the measurement of urinary iron excretion varied (12 months (Maggio 2002), early after starting treatment (Gomber 2004), 24 hours after starting treatment (Olivieri 1990)) and because of variation in means of measurement (mg/24 hours (Gomber 2004; Olivieri 1990), as a mean of quarterly readings (El-Beshlawy 2008) and mean urinary iron excretion (%) (Aydinok 2005)).

ii. Change from baseline

In the trial reporting data to analyse mean change from baseline there was no statistically significant difference in mean change between the treatment arms, MD 0.20 mg/24h (95% CI -4.00 to 4.40) (Maggio 2002) (Analysis 1.7).

c. Liver iron concentration

Liver iron concentration was reported as an outcome in five trials (El-Beshlawy 2008; Ha (ii) 2006; Maggio 2002; Olivieri 1997; Pennell 2006). In one trial, data were presented graphically; baseline and endpoint values are estimated from the graph (El-Beshlawy 2008). The individual trial data are presented in an additional table (Table 1); mean change from baseline data are also presented in an additional table (Table 4). Liver iron concentration was also measured in one additional trial but individual outcome data were not reported (Aydinok 2005). Liver iron concentration was measured by magnetic spectrometry (SQUID) in one trial (Pennell 2006), by a combination of SQUID and biopsy analysis in one trial (Olivieri 1997) and by atomic spectrophotometry in two trials (Ha (ii) 2006; Maggio 2002). In one trial the method used to assess liver biopsy was not reported (El-Beshlawy 2008). One trial measured liver iron concentration in a subset of participants (Maggio 2002). The subset comprised participants from both treatment groups who agreed to undergo a liver biopsy; the baseline characteristics of the subset were very similar to the baseline characteristics of the trial participants overall.

Table 4. Liver iron concentration (mean (SD)): deferiprone versus DFO
  1. d/w = dry weight liver
    a Pennell 2006 reports a reduction of 10.1%, although note that this figure is not equal to % of absolute change over baseline value.

Trial No. Deferiprone/ DFO Deferiprone: baseline Def: end of study Absolute change % change DFO: baseline DFO: end of study Absolute change % change
El-Beshlawy 200817/1515.8 (7.1) mg/g d/wapprox. 7.5 (3.6) mg/g d/w (estimated from graph)reduction of approx. 8.5 mg/g d/wnot reported22.5 (10.1) mg/g d/wapprox. 11.5 (6.3) mg/g d/w (estimated from graph)reduction of approx. 11 mg/g d/wnot reported
Ha (ii) 20066/7not reportednot reportedincrease of 5.63 (4.24) mg/g d/wnot availablenot reportednot reportedincrease of 2.90 (1.27) mg/g d/wnot available
Maggio 200221/153363 (5490) µg/g d/w2341 (2197) µg/g d/w (end of trial = median 30 months)reduction of 1022 (3511) µg/g/ d/wreduction of 30.4%3516 (2974) µg/g d/w3166 (2519) µg/g d/w (end of trial = median 30 months)reduction of 350 (524) µg/g d/wreduction of 9.95%
Olivieri 199719/188.90 (5.23) mg/g d/w13.70 (5.23) mg/g d/w (end of trial = mean 33 months)increase of 4.8 mg/g d/wincrease of 53.9%6.90 (3.82) mg/g d/w (end of trial = mean 33 months)7.90 (5.52) mg/g d/wincrease of 1.0 mg/g d/wincrease of 14.5%
Pennell 200627/306.16 (6.0) µg/Lnot reportedreduction of 0.93 (2.9) mg/g d/w (-10% from baseline)reduction of 10.1%a6.32 (5.8) µg/Lnot reportedreduction of 1.54 (2.5) mg/g d/wreduction of 24.4%

i. At end of trial

Liver iron concentration at the end of the trial was reported in three trials (El-Beshlawy 2008; Maggio 2002; Olivieri 1997). Analysis of these three trials was carried out on a log scale due to the apparent skewing of the data in one of the trials (Maggio 2002).

One trial reported liver iron concentration (mg/kg dry liver weight (d/w)) at 12 months (El-Beshlawy 2008). At the end of the trial, the mean (SD) liver iron concentration was approximately 11.5 (6.3) mg/g d/w for DFO and 7.5 (3.6) mg/g d/w for deferiprone (Table 1); values of liver iron concentration in the deferiprone treated patients were 0.67 times those in patients treated with DFO: ratio of geometric means, 0.67 (95% CI 0.48 to 0.94), favouring deferiprone (Analysis 1.8).

One trial reported liver iron concentration (mg/g d/w) at 24 months (Olivieri 1997). The mean (SD) liver iron concentration values at 24 months were 8.90 (2.83) for deferiprone and 7.78 (4.68) for DFO, ratio of geometric means, 1.27 (95% CI 0.90 to 1.80).

Two trials reported end of trial measurements for this outcome at between 30 and 34 months from the start of the trial; the mean (SD) time to end of trial was 34.6 (6.7) months for DFO and 30 (2.4) months for deferiprone (Maggio 2002) and 33 ( 6.1) months (Olivieri 1997). At the end of the trial, mean liver iron concentration for deferiprone was almost two-fold higher than for DFO in one trial (Olivieri 1997), favouring DFO in this trial, ratio of geometric means 1.98 (95% CI 1.41 to 2.79) (Analysis 1.8). In the second trial, the mean value in patients who received deferiprone was 0.69 times that in patients who received DFO although this did not reach statistical significance, ratio of geometric means 0.69 (95% CI 0.42 to 1.12) (Maggio 2002) (Analysis 1.8). The results from these two trials were not pooled due to the presence of hepatitis C in 86% of participants in one trial (Maggio 2002) with no details of hepatitis C levels in participants in other trials and the different techniques used to assess liver iron concentration between the trials.

ii. Change from baseline

Five trials reported change from baseline in liver iron concentration (El-Beshlawy 2008; Ha (ii) 2006; Maggio 2002; Olivieri 1997; Pennell 2006). Liver iron concentration decreased from baseline to the end of the trial in both treatment groups in three trials (El-Beshlawy 2008; Maggio 2002; Pennell 2006) (Table 4). In these trials, the greatest decrease was observed in the DFO group in two trials (El-Beshlawy 2008; Pennell 2006) and in the deferiprone group in one trial (Maggio 2002). In two other trials, liver iron concentration had increased at the end of the trial in both treatment groups (Ha (ii) 2006; Olivieri 1997); the increase was greatest in the deferiprone treated group, favouring DFO.

Four trials reported SDs for mean change from baseline (Ha (ii) 2006; Maggio 2002; Olivieri 1997; Pennell 2006), although the data were skewed in one of these trials and data from this trial were therefore excluded from analysis of mean change from baseline (Maggio 2002). For the remaining three trials, the difference in mean change from baseline between treatment groups was calculated. There was no statistically significant difference in mean change in liver iron concentration in any trial: MD 2.73 mg/g d/w at six months (95% CI -0.79 to 6.25) (Ha (ii) 2006); MD 0.61 mg/g d/w at 12 months (95% CI -0.80 to 2.02) (Pennell 2006), and MD -0.33 mg/g d/w at 24 months (95% CI -3.56 to 2.90) (Olivieri 1997) (Analysis 1.9). Data from these trials were not pooled due to variation in time points for outcome measurements.

d. Myocardial T2*

Myocardial T2* was reported as an outcome measure in one trial (Pennell 2006). Results were reported on a log scale as geometric means; baseline and endpoint data are reported in an additional table (Table 1). Low T2* values were an entry criteria in this trial. T2* values increased at six and 12 months in both treatment arms; the increase was two-fold higher at 12 months in patients who received deferiprone (26.9% increase from baseline) compared with those who received DFO (12.8%). There was a small statistically significant difference between treatment arms in myocardial T2* levels, favouring deferiprone. The geometric mean value of myocardial T2* was approximately 10% higher in patients who received deferiprone than those who received DFO after six months, ratio of geometric means 1.09 (95% CI 1.01 to 1.18) and at the end of the trial (12 months), ratio of geometric means 1.12 (95% CI 1.02 to 1.23), favouring deferiprone (Analysis 1.10).

e. Chelation efficiency

Mean chelation efficiency over the trial was reported in one trial (Aydinok 2005) (Table 3). Chelation efficiency was calculated as [Iron excretion (mg/kg/day) / chelator dose (mg/kg/day)] x [molecular weight of the respective chelator /56] x n x 100, where 56 is the molecular weight of iron and n = 3 with deferiprone and n = 1 with DFO. There was a statistically significant difference in mean chelation efficiency in favour of DFO for this treatment comparison, MD -16.45% (95% CI -25.85% to -7.05%) (Analysis 1.11).

f. Plasma non-transferrin bound iron (NTBI)

Plasma NTBI (mM) was measured by one trial (Aydinok 2005) (Table 3). Mean results at the end of treatment (time point not defined) were reported. There was a statistically significant difference in mean plasma NTBI at the end of treatment in favour of the DFO treatment arm, MD 2.28 mM (95%CI 1.78 to 2.78) (Analysis 1.12).

g. Total Iron excretion

Total iron excretion (mg/kg/day) at the end of the trial was reported in one trial (Aydinok 2005) (Table 3). Total iron excretion per day was calculated as [iron transfused/year (mg) + (liver iron concentration at time 0 - liver iron concentration at time 1 year) x 10.6 x body weight in kg] / number of days of treatment. There was no statistically significant difference in mean total iron excretion between the treatment arms, MD -0.21 mg/kg/day (95% CI -0.43 to 0.01) (Analysis 1.13).

3. Adverse events due to deferiprone

Adverse events were reported as an outcome in six trials (El-Beshlawy 2008; Gomber 2004; Ha (ii) 2006; Maggio 2002; Olivieri 1997; Pennell 2006) although one trial (Ha (ii) 2006) did not report adverse events incidences per treatment arm; adverse event data for this trial has not been reported in the review. Another trial (Gomber 2004) did not differentiate adverse events between deferiprone alone and deferiprone in combination with DFO and therefore adverse events in this trial could not be incorporated into meta-analyses. See an additional table for details of adverse events reported in individual trials (Table 5).

One trial reported data that enabled a comparison of the risk of experiencing any adverse event (Maggio 2002). In this trial, there was a statistically significant increased risk of experiencing an adverse event in participants receiving deferiprone compared with those receiving DFO, RR 2.24 (95% CI 1.19 to 4.23) (Analysis 1.14; Figure 3). The remaining trials did not provide enough data to allow for an analysis of the risk of experiencing cumulative adverse events between the treatment arms.

Figure 3.

Forest plot of comparison: 1 Deferiprone alone versus DFO alone, outcome: 1.14 Adverse events.

Table 5. Adverse events
  1. DFO: desferrioxamine
    sc: subcutaneous

    a Abdelrazik 2007: 2 participants reported taste disorders and one patient reported transient dizziness and fatigue but type of therapy received by these participants was not reported in these patients.
    b Aydinok 2007: 5 participants developed nausea but type of therapy received by these participants was not reported.
    c Gomber 2004: results for all participants receiving deferiprone alone or in combination were reported together in the paper.
    d Maggio 2009: the period of observation for adverse effects was 307.98 person-years for combined deferiprone and DFO and 274.52 person-years for deferiprone only.
    e Mourad 2003: other reported adverse events were fatigue (1); loss of appetite (1); headache (3); transient skin rash (2); abdominal discomfort (2). Treatment group was not reported for these adverse events which did not result in discontinuation of therapy.
    f Tanner 2007: 28 participants completed combination treatment but actual sample sizes in which adverse events reported is unclear.

Trial Name Prior iron chelation therapy Intervention Total Adverse Events Adverse Event Details Dose Reduction Temporary Withdrawal Permanent Withdrawal

Abdelrazik 2007

 

Prior DFO received at a daily dose of 40 mg/kg/day by sc infusion pump over 8 - 10h, 5 - 7 nights a week for "several years". Deferiprone + DFO8/301/30 (mild arthropathy); 1/30 (vomiting); 2/30 (abdominal pain); 1/30 (diarrhoea)anone reportednone reported0
DFO 3/30none notedanone reportednone reported0

Aydinok 2007

 

Previous DFO therapy. All patients had washout period of two weeks with no iron chelation therapy prior to initiating study treatment.

 

Deferipronenot reported1/12 (neutropenia); 1/12 (aseptic meningitis at week 45); 1/12 (acute cerebellar syndrome with dizziness, tinitus, truncal ataxia)bnone reportednone reported1 (acute cerebellar syndrome)
Deferiprone + DFOnot reported 1/9 (neutopenia weeks 4/10 with subsequent agranulocytosis week 14); 1/9 (arthralgia grade 2); “mild, local reactions observed in several patients treated with DFO”bnone reportednone reported1 (agranulocytosis at week 14)

Choudhry 2004

 

not reported

 

Deferiprone (50 mg)not reported 15/30 (joint pain: 6 mild, 8 moderate, 1 severe); 7/30 (leucopenia/neutropenia)05 (leucopenia/neutropenia: drug immediately withdrawn and managed conservatively)3 (2 neutropenia/leucopenia; 1 severe multiple joints involvement)
Deferiprone (75 mg)not reported 6/21 (joint pain: 3 mild, 3 moderate); 5/21 (leucopenia/neutropenia)1 (neutropenia/leucopenia)4 (leucopenia/neutropenia: drug immediately withdrawn and managed conservatively)1 (neutropenia/leucopenia)

El-Beshlawy 2008

 

 

not reported

 

 

Deferipronenot reported 1/21 (agranulocytosis); 1/21 (jaundice/very high liver enzymes); 8/21 (arthropathy); 7/21 (nausea/vomiting; 2/21 (anorexia)2 (to 50 mg/kg due to arthropathy)none reported2 (1 agranulocytosis; 1 jaundice/very high liver enzymes)
DFONot reported  1/23 (neutropenia); 1/23 (jaundice/very high liver enzymes); 5/23 (skin reactions/allergy, swelling); 1/23 (systemic allergy); 1/23 (arthropathy)0none reported0
Deferiprone + DFOnot reported  1 /22 (neutropenia); 2/22 (jaundice/very high liver enzymes); 3/22 (skin reactions/allergy, swelling); 6/22 (arthropathy); 3/22 (musculoskeletal pain); 4/22 (nausea/vomiting); 2/22 (weakness, fever); 1/22 (insomnia; 5/22 (anorexia)2 (to 50 mg/kg due to arthropathy)none reported1 (jaundice/very high liver enzymes)

Galanello 2006

 

Prior iron chelation therapy received: DFO treatment inferred but not explicitly stated.

 

Deferiprone + DFO7/295/29 (vomiting); 3/29 (abdominal pain); 1/29 (diarrhoea); 5/12 (transient increase in ALT three times of upper limit).none reportednone reported1 (neutropenia on day 2 after receiving DFO therapy only)
DFO 2/301/30 (abscess at site of infusion); 1/30 (allergic reaction); 1/30 (neutropenia); 1/14 (transient increase in ALT three times of upper limit).none reportednone reported0

Gomber 2004 c

 

not reported

 

Deferiprone (with or without DFO in combination)

not reported

 

2/21 (arthropathy of large joints)02 (arthropathy of large joints)0
DFOnot reportedno adverse effects observed000

Maggio 2002

 

All participants had received DFO prior to trial enrollment.

 

Deferiprone24/7116/71 (hypertransaminasemia); 3/71 (nausea); 2/71 (leucopenia); 2/71 (mild joint pain); 1/71 (infection)3 (nausea). Dose and duration of reduction were not reported.4 (3 transient hypertransaminasemia; 1 infection). Dose and duration of withdrawal were not reported.5 (3 recurrence of hypertransaminaesima; 2 leukocytopenia). Timepoint for withdrawal was not reported.
DFO11/736/73 (pain/erythema at injection site); 2/73 (ototoxicity); 2/73 (infection); 1/73 (hypertransaminasemia)7 (6 pain and erythema at the injection site; 1 transient hypertransaminasemia). Dose and duration of reduction were not reported.4 (2 infection (Yersinia enterocolitiica); 2 ototoxicity). Dose and duration of withdrawal were not reported.0
Maggio 2009 dAll patients treated with either DFO or deferiprone prior to study with one week washout period.Deferiprone + DFOnot reported15/65 (neutropenia); 5/65 (arthralgia); 7/65 (gastrointestinal problems); 22/65 (> 2-fold increase in ALT)Dose reduction reported in 25/51 (49%) patients. Dose and duration of reduction were not reported.Number of temporary withdrawals not reported, but "no statistically significant difference in temporary and definitive discontinuation of treatment between the groups"12
Deferipronenot reported3/88 (agranulocytosis); 11/88 (neutropenia); 6/88 (arthralgia); 16/88 (gastrointestinal problems); 23/88 (>2-fold increase in ALTDose reduction reported in 26/46 (56.5%) patients. Dose and duration of reduction were not reported.Number of temporary withdrawals not reported, but "no statistically significant difference in temporary and definitive discontinuation of treatment between the groups"21

Mourad 2003

 

All participants had received DFO prior to trial enrolment

 

Deferiprone + DFOnot reported5/11 (nausea); 3/11 (joint pain/swelling/stiffness)none reported00
DFOnot reported12/14 (problems at site of injection: pain, itching, erythema, swelling and induration)none reported00
Olivieri 1997not reportedDeferipronenot reported2/27 (agranulocytosis)none reportednone reported2 (2 x agranulocytosis)
DFOnot reported0/30 (agranulocytosis)none reportednone reported0

Pennell 2006

 

All participants had received DFO prior to trial enrolment.

 

Deferipronenot reported19/29 (gastrointestinal disturbance); 9/29 (increased appetite); 8/29 (pain/ swelling in joints); 1/29 (neutropenia)000
DFOnot reported12/31 (local reactions at infusion site); 6/31 (pain/swelling in joints)none reportednone reportednone reported
Tamaddoni 2010All participants had received DFO prior to trial enrolment.Deferiprone + DFOnot reported3/40 (nausea); 3/40 (nausea with abdominal pain); 2/40 (diarrhoea); 2/40 (arthralgia); 8/40 (transient fluctuations in serum alanine ALT levels)Temporary deferiprone dose reduction reported in "some" patients with gastrointestinal disturbances.00
DFOnot reported1/40 (abscess at site of infusion); 11/40 (skin allergic reactions)None reported00

Tanner 2007 f

 

Prior DFO received: mean dose 36.4 (11.1) mg/kg per day for 5.5 day/week (equivalent to 40.5 mg/kg for 5 d/week). No prior deferiprone received.

 

Deferiprone + DFOnot reported38% (gastrointestinal symptoms – generally mild); 3% (reactions at infusion site); 9% (joint pain/swelling); 1/28 (agranulocytosis; 2/28 (neutropenia)none reportednone reported4 (1 x agranulocytosis; 1 x reduction of myocardial T2* to < 8ms; 1 x gastrointestinal symptoms; 1 x personal reasons)
DFO + Placebonot reported24% (gastrointestinal symptoms – generally mild); 6% (reactions at infusion site); 18% (joint pain/swelling)none reportednone reported3 (1 x artrial fibrillation; 2 x personal reasons)

Permanent treatment withdrawal due to adverse events was reported in two trials (El-Beshlawy 2008; Maggio 2002). All of the permanent treatment withdrawals in both trials occurred in patients who received deferiprone. Only one trial reported temporary treatment withdrawals which occurred in four participants in each treatment arm (Maggio 2002).

Two trials reported dose reduction due to adverse events (El-Beshlawy 2008; Maggio 2002); both trials reported dose reduction in the deferiprone treatment arm, due to arthropathy (El-Beshlawy 2008) and nausea (Maggio 2002). Only one of these trials reported details of the reduction in deferiprone: the dose was reduced to 50 mg/kg (El-Beshlawy 2008). The Maggio trial also reported dose reduction in patients receiving DFO alone, due to pain or erythema or transient hypertransaminasaemia (Maggio 2002).

Joint pain or arthralgia was reported as an adverse event in four of the five trials; two of these documented joint pain in both treatment arms (El-Beshlawy 2008; Pennell 2006), whereas in a third trial joint pain was only reported in patients who received deferiprone (Maggio 2002). The fourth trial reported joint pain in patients who received deferiprone (with or without DFO in combination) but no joint pain was reported in patients receiving DFO alone (Gomber 2004). Meta-analysis of data from the first three trials showed a significant increased risk of joint pain or arthralgia associated with deferiprone, RR 2.64 (95% CI 1.21 to 5.77) (Analysis 1.14; Figure 3).

Three trials reported gastrointestinal symptoms in the form of nausea or vomiting in patients receiving deferiprone (El-Beshlawy 2008; Maggio 2002; Pennell 2006). No cases of nausea or vomiting were reported in the DFO treatment arm although in one trial an absence of gastrointestinal symptoms in participants receiving DFO could not be clearly inferred (Pennell 2006). A meta-analysis of two trials showed a significantly increased risk of nausea or vomiting, or both, associated with deferiprone, RR 11.71 (95% CI 1.57 to 87.31) (Analysis 1.14; Figure 3).

Five trials reported incidence of neutropenia or leucopenia, or both (El-Beshlawy 2008; Gomber 2004; Maggio 2002; Olivieri 1997; Pennell 2006). In two of these trials, one case of neutropenia (Maggio 2002) and two cases of leucopenia (Pennell 2006) were reported only in patients who received deferiprone. Three cases of agranulocytosis were observed in a third trial, all of which occurred in patients who received deferiprone (Olivieri 1997). The fourth trial reported one case of neutropenia which occurred in a patient treated with DFO and one case of agranulocytosis in a patient who received deferiprone (El-Beshlawy 2008). No cases of neutropenia or leucopenia were observed in the Gomber trial (Gomber 2004). When data were pooled across trials, neither treatment arm showed a significantly increased risk of neutropenia, leucopenia or agranulocytosis, RR 2.51 (95% CI 0.66 to 9.55) (Analysis 1.14).

Two trials observed increased liver transaminase in patients receiving either deferiprone or DFO, reported as hypertransaminasaemia (Maggio 2002) or jaundice and very high liver enzymes (El-Beshlawy 2008). In one trial, hypertransaminaemia was reported in over 20% of patients who received deferiprone compared with only one case of hypertransaminasaemia in patients who received DFO, RR 16.45 (95% CI 2.24 to 120.79) (Maggio 2002) (Analysis 1.14; Figure 3). The second trial reported two patients (one in each treatment group) with jaundice or very high liver enzymes, RR 1.10 (95% CI 0.07 to 16.43) (El-Beshlawy 2008) (Analysis 1.14; Figure 3). The combined risk of increased liver transaminase was significantly higher patients treated with deferiprone than DFO, RR 8.90 (95% CI 2.08 to 38.11) (Analysis 1.14; Figure 3).

Three trials reported adverse events specifically related to the administration of DFO, including skin reactions/swelling (5 out of 23, 21.7%) and systemic allergy (1 out of 23, 4.3%) (El-Beshlawy 2008); pain or erythema at site of injection (6 out of 73) (Maggio 2002) and local reactions at infusion site (12 out of 31, 38.7%) (Pennell 2006).

Other adverse events reported in patients receiving deferiprone included infection (Maggio 2002) and anorexia (El-Beshlawy 2008).

4. Participant compliance

Participant compliance with iron chelation was measured in four trials (El-Beshlawy 2008; Ha (ii) 2006; Olivieri 1997; Pennell 2006); see an additional table for individual trial data (Table 6). Two of these trials did not provide data for this outcome (El-Beshlawy 2008; Ha (ii) 2006). Although the first of these reported "excellent compliance during the study period" and documented four patients who were excluded from the trial due to lack of compliance, all of whom were randomised to the DFO treatment arm. At 12 months, there was no significant difference in participant compliance between treatment groups, MD 1.00 (95% CI -2.88 to 4.88) (Pennell 2006) (Analysis 1.5). However, in one trial at three years, there was a statistically significant difference in participant compliance in favour of deferiprone, MD 23.30% (95%CI 21.52 to 25.08) (Olivieri 1997) (Analysis 1.15).

Table 6. Participant treatment compliance
Study Intervention How measured? Compliance Rate

Abdelrazik 2007

 

Deferiprone + DFOAuditNo results reported
DFOAuditNo results reported

Aydinok 2007

 

DeferiproneDrug accounting at each visit - counting returned empty blisters of deferiprone and study specific questionnaire completed by patient or legal guardian at quaterly intervals.“Generally excellent during the entire study period”. Only one patient who missed more than one chelation dose per week due to problems with swallowing.
Deferiprone + DFODrug accounting at each visit - counting the used vials of DFO and study specific questionnaire completed by patient or legal guardian at quaterly intervals.“Generally excellent during the entire study period” – no further details provided.

El-Beshlawy 2008

 

 

Deferiprone + DFODrug accounting at each visit; counting returned empty blisters of deferiprone and used vials of DFO"Compliance excellent during study period". 80% patients complained about difficulties in the parenteral use of DFO or problems inserting the needle. One patient dropped out for reasons of non-compliance.
DeferiproneDrug accounting at each visit; counting returned empty blisters of deferiprone."Compliance excellent during study period". 76% patients complained about difficulties in the parenteral use of DFO or problems inserting the needle.
DFODrug accounting at each visit: counting returned used vials of DFO."Compliance excellent during study period". 4 patients were excluded due to lack of compliance (timepoint not stated).

Galanello 2006

 

Deferiprone + DFO

DFO: diary cards, weekly physical exam of infusion sites & by the Crono infusion pump that recorded the number of completed infusions.

Deferiprone: pill counts, diary cards & an electronic cap that recorded the time & date of each opening of the tablet container.

96.1 (5.0)% (compliance with DFO)
DFOdiary cards, weekly physical exam of infusion sites & by the Crono infusion pump that recorded the number of completed infusions.95.7 (5.7)%

Ha (i) 2006

 

Deferiprone + DFOBy reviewing the diaries maintained by the patients or their parents at each visit.No results reported
DFOBy reviewing the diaries maintained by the patients or their parents at each visit.No results reported

Ha (ii) 2006

 

DeferiproneBy reviewing the diaries maintained by the patients or their parents at each visit.No results reported
DFOBy reviewing the diaries maintained by the patients or their parents at each visit.No results reported

Maggio 2002

 

DeferiproneNot reportedCompliance was not reported as an outcome in this trial. However the authors do note that 4/71 participants (6%) took a reduce dose of deferiprone due to low compliance. The dosage, duration and outcome of this change were not reported.
DFONot reportedCompliance was not reported as an outcome in this trial. However the authors do note that 7/73 participants (10%) took a reduce dose of DFO due to low compliance. The dosage, duration and outcome of this change were not reported
Maggio 2009Deferiprone + DFOBy counting pills in each returned bag of deferiprone and assessing the number of infusions of DFO registered on the electronic pump.

Deferiprone: 92.7% (+/- 15.2%; range 27-100%);

DFO 70.6% (+/- 24.1%; range 25-100%)

DeferiproneBy counting pills in each returned bag of deferiprone.93.6% (+/-9.7%; range 56-100%)

Mourad 2003

 

Deferiprone + DFOBy counting the number of vials of DFO or tablets of deferiprone used. Trial does not report what constituted "adequate levels of adherence", but rates compliance as 'excellent' or 'good'.Excellent (taking >90% of recommended dose): 10 participants; good (taking 75% to 90% of recommended dose:) 1 participant.
DFOBy counting the number of vials of DFO used. Trial does not report what constituted "adequate levels of adherence", but rates compliance as 'excellent' or 'good'.Excellent (taking >90% of recommended dose) : 11 participants; good (taking 75% to 90% of recommended dose): 3 participants.

Olivieri 1997

 

DeferiproneComputerised bottles. Trial does not report what constituted "adequate levels of adherence".94.9 +/- 6.69%
DFOAmbulatory pumps. Trial does not report what constituted "adequate levels of adherence".71.6 +/- 22.5%

Pennell 2006

 

DeferiproneUsing the medication event monitoring system device and calculated as the percent of openings with an interval >4 hours recorded, divided by the number of doses prescribed. Trial does not report what constituted "adequate levels of adherence".94 +/- 5.3%
DFOCalculated as the percent of completed infusions, as determined by the Crono pumps, divided by the number of infusions prescribed. Trial does not report what constituted "adequate levels of adherence".93 +/- 9.7%

Tanner 2007

 

Deferiprone + DFOPill counting at bimonthly visits; % of complete DFO infusions divided by prescribed infusions.DFO: 91.4 +/- 12.7%; deferiprone: 82.4 +/-18.1%
DFO% complete infusions divided by prescribed infusions.92.6 +/- 12.7%
5. Cost of treatment

The cost of treatment was reported in one trial (Gomber 2004). Full details are presented in an additional table (Table 7).

Table 7. Relative cost of treatment (participant = 10 kg): Gomber 2004
  1. Rs = Rupees

Variables Group 1: Deferiprone Group 2: DFO Group 3: Deferiprone+DFO
Cost of drugRs. 18: Rs. 6 per 250 mg capsule, 3 capsules daily.Rs. 180: Rs.150 for vial of 500 mg +Rs. 30 for needle and syringe.Rs. 198
Cost of treatment per weekRs. 126Rs. 900Rs. 486
Cost of treatment for 6 monthsRs. 3024Rs. 21600Rs. 11664

As this trial was undertaken in India, cost is presented in rupees. In a comparison of the costs of treatment per week of deferiprone versus DFO; deferiprone was the cheaper treatment: 126 rupees with deferiprone and 900 rupees with DFO. Costs were based on a participant weighing 10 kg.

B. Deferiprone alone compared with deferiprone and DFO in combination

There were five comparisons between deferiprone alone with deferiprone and DFO in combination (Aydinok 2005; Aydinok 2007; El-Beshlawy 2008; Gomber 2004; Maggio 2009).

Primary outcome
1. Mortality

Mortality was reported in two trials (Aydinok 2007; Maggio 2009). In the first trial, one individual randomised to receive deferiprone and DFO in combination, died at the start of the trial due to arrhythmia-induced congestive heart failure (Aydinok 2007). One death due to arrhythmia whilst receiving deferiprone and DFO in combination was also reported in the second trial (Maggio 2009). In this latter trial, a further five deaths were reported in patients in whom the randomised treatment was withdrawn and treatment changed to DFO alone due to adverse events; mortality occurred 11 to 60 months after withdrawal of the randomised treatment.

Secondary outcomes
1. Evidence of reduced end organ damage

Two trials reported evidence of reduced end organ damage as an outcome, although the latter of these reported only that "there was no significant difference in cardiac function" between treatment arms (Aydinok 2007; El-Beshlawy 2008). In the Aydinok trial, cardiac function was measured by echocardiogram to determine LVEF (%) (Aydinok 2007). In this trial, the mean LVEF after 12 months was not significantly different between the two treatment arms, MD -5.20% (95% CI -12.39 to 1.99) (Analysis 2.1).

One trial reported liver fibrosis as an outcome, scored according to the Ishak scoring system (Aydinok 2007). In this trial, results were presented graphically, the graphs being of insufficient quality to enable estimation of fibrosis scores. However, the trial authors reported that the fibrosis score "did not change significantly after one year in patients in any of the treatment arms".

2. Measures of iron overload
a. Serum ferritin concentration

Four trials reported serum ferritin concentration as an outcome; individual trial data are shown in an additional table (Aydinok 2007; El-Beshlawy 2008; Gomber 2004; Maggio 2009) (Table 1); mean change from baseline are reported in a further table (Table 8). One trial presented data graphically and the mean change has been estimated from the graph; SDs were not reported in this trial (El-Beshlawy 2008).

Table 8. Serum ferritin concentration (mean (SD)): deferiprone versus deferiprone with DFO
  1. anot all patients completed follow-up due to the trial being stopped early. Baseline values are shown for all randomised patients whereas endpoint values are given only for patients who completed follow up.

Trial No Treatment Deferiprone:baseline Def: end Absolute change % change Def+DFO: baseline Def+DFO: end Absolute change % change
Aydinok 20078/124350 (3342) µg/L2954 (2765) µg/Lreduction of 1396 µg/L (SD not reported)reduction of 32%4070 (3223) µg/L3209 (2279) µg/Lreduction of 861 µg/L (SD not reported)reduction of 32%
El-Beshlawy 200818/202926 (1107) µg/Lapprox. 900 µg/L (estimated from graph)

reduction of approx. 2026 µg/L

 

 

 

not available2865 (983) µg/Lapprox. 1050 µg/L (estimated from graph)

reduction of approx. 1815 µg/L

 

not available
Gomber 200411/102672.90 (886.44) ng/ml3422.66 (1581.01) ng/mlincrease of 749.75 (1155.77) ng/mlincrease of 28%3347.78 (1526.46) ng/ml3376.57 (1222.41) ng/mlincrease of 28.79 (915.86) ng/mlincrease of 0.9%
Maggio 2009

74/78 (12 mths)

26/32 (5 years)

1890 (816) ng/ml

12 mths:

1633 (841) ng/ml

5 years:

1588 (1217) ng/mla

12 mths:

reduction of 132 (724) ng/ml

5 years:

reduction of 115 (1009) ng/mla

not availablea1787 (735) ng/ml

12 mths:

1400 (770) ng/ml

5 years:

1369 (816) ng/mla

12 mths:

reduction of 417 (589) ng/ml

5 years:

reduction of 396 (894) ng/mla

not availablea

i. At end of trial

Of the three trials which reported SDs for end of trial data (Aydinok 2007; Gomber 2004; Maggio 2009), none showed a significant difference between treatment arms in serum ferritin concentration after six months, MD 46.10 ng/ml (95% CI -1156.78 to 1248.98) (Gomber 2004) (Analysis 2.2) or after 12 months, MD -255.00 ng/ml (95% CI -2564.49 to 2054.49) (Aydinok 2007); and MD 233.00 ng/ml (95% CI -23.74 to 489.74) (Maggio 2009) (Analysis 2.2). The latter trial reported data for each year over five years of follow up; these results are described as change from baseline below.

ii. Change from baseline

Data to calculate mean change in serum ferritin from baseline to end of trial were available in two trials (Gomber 2004; Maggio 2009). In the Gomber trial, there was no statistically significant difference in mean change in serum ferritin concentration between the two treatment arms, MD 720.96 ng/ml (95% CI -167.14 to 1609.06) (Gomber 2004) (Analysis 2.3). Mean serum ferritin concentration increased from baseline to six month end of trial in both treatment arms, the greater increase was observed in the deferiprone arm.

The Maggio trial reported results for up to five years of follow up (Maggio 2009). After one year of follow up, both arms of the trial reported a reduction in serum ferritin concentration with a significantly greater reduction in patients receiving deferiprone in combination with DFO, MD 285.00 ng/ml (95% CI 74.53 to 495.47) (Analysis 2.3). A reduction in serum ferritin concentration was maintained across the five years of follow up in the combined treatment arm. In patients receiving deferiprone only, a reduction in mean serum ferritin concentration from baseline values was observed at two and five years of follow-up although an increase was reported at three and four years of follow up. A significant difference in mean change of serum ferritin concentration in patients receiving deferiprone in combination with DFO was maintained over four years of follow up. At four years, MD 579.00 ng/ml (95% CI 116.11 to 1041.89) (Analysis 2.3). At five years, there was no significant difference in mean change of serum ferritin concentration between the two treatment arms, MD 281.00 ng/ml (95% CI -215.35 to 777.35) (Analysis 2.3) although the number of patients in each treatment arm was considerably reduced after five years of follow-up due to early termination of the trial.

b. Urinary iron excretion

Urinary iron excretion was measured in three trials (Aydinok 2005; El-Beshlawy 2008; Gomber 2004). As none of these trials presented baseline data for this outcome, mean change in urinary iron excretion from baseline to end of trial could not be calculated. See additional tables for individual trial data (Table 1; Table 3).

One trial showed a statistically significant difference in mean urinary iron excretion (%) over the period of trial in favour of deferiprone alone, MD 14.00% (95% CI 0.87 to 27.13) (Aydinok 2005) (Analysis 2.4). There was no statistically significant difference in mean urinary iron excretion between treatment arms in the other two trials, MD -2.36 mg/24h (95% CI -5.70 to 0.98) (Gomber 2004) and MD -0.07 mg/24h (95% CI -0.24 to 0.10) (El-Beshlawy 2008) (Analysis 2.5). Data for mean urinary iron excretion after treatment was not pooled overall in a meta-analysis because of variation in the means of measurement: mg/24 hours (Gomber 2004); as a mean of quarterly readings (El-Beshlawy 2008) and as mean urinary iron excretion over the trial (%) (Aydinok 2005).

c. Liver iron concentration

Five trials measured liver iron concentration (Aydinok 2005; Aydinok 2007; El-Beshlawy 2008; Gomber 2004; Maggio 2009). However, in the Gomber trial, liver iron concentration, although measured, was not reported as an outcome (Gomber 2004). Liver iron concentration was measured by atomic emission spectrophotometry in one trial (Aydinok 2007). A second trial measured liver iron content by T2* MRI in a subset of patients with a mean (SD) duration from entry into the trial until the final MRI scan of 16 (5) months in the combined deferiprone and DFO group and 14 (6) months in patients receiving deferiprone alone (Maggio 2009). The method used for liver biopsy assessment was not reported in two trials (Aydinok 2005; El-Beshlawy 2008).

In the 2005 Aydinok trial, liver iron concentration was measured but data were not individually reported (Aydinok 2005). In addition, one trial presented data graphically and the mean change was estimated from the graph (El-Beshlawy 2008). Data to calculate mean change in liver iron concentration from baseline to end of trial were available in two trials, although SDs could not be calculated (Aydinok 2007; Maggio 2009 (Table 9). Data to calculate the mean liver iron concentration (mg/g dry weight) at the end of the trial was reported in all three trials (El-Beshlawy 2008; Aydinok 2007; Maggio 2009).

Table 9. Liver iron concentration (mean (SD)): deferiprone versus deferiprone with DFO
  1. d/w: dry weight liver
    aThe mean (SD) duration from entry into trial to time of liver T2* MRI basal assessment was 17 (18) months in the combined treatment group and 18 (17) in the deferiprone alone treatment group. Final evaluation was performed after a further 16 (5) months and 14 (6) months respectively.
    d/w: dry weight liver

Trial Number Deferiprone: baseline Deferiprone: end Absolute change % change Deferiprone + DFO: baseline Deferiprone + DFO: end Absolute change % change
Aydinok 200712/830.7 (10.6) mg/g d/w28.6 (12.8) mg/g d/wreduction of 2.1 mg/g d/wreduction of 6.8%26.6 (5.4) mg/g d/w18.1 (11.6) mg/g d/wreduction of 8.5 mg/g d/wreduction of 32.0%
El-Beshlawy 200817/1615.8 (7.1) mg/g d/wapprox. 7.5 (3.6) mg/g d/w (estimated from graph)reduction of approx. 8.3 mg/g d/wnot reported17.1 (9.1) mg/g d/wapprox. 6.5 (3.5) mg/g d/w (estimated from graph)reduction of approx. 10.6 mg/g d/wnot reported
Maggio 200920/344.0 (5.9) msa3.5 (4.3) msareduction of 0.4 msnot reported4.0 (2.9) msa4.4 (3.4) msaincrease of 0.4 msnot reported

In the two trials reporting liver iron concentration after 12 months of treatment, the combined mean difference from meta-analysis of these two trials was not significant: 1.45 mg/g d/w (95% CI -0.91 to 3.82) (Aydinok 2007; El-Beshlawy 2008) (Analysis 2.6). The trial measuring liver iron concentration by T2* MRI showed a reduction in liver T2* with deferiprone alone compared with an increase in liver T2* with combined chelation therapy (Maggio 2009) (Table 9). The differences in follow-up time across this subset of patients prohibited calculation of the mean change in liver iron concentration; however the trial reported no significant differences in the T2* signals of the liver between the two treatment groups (Maggio 2009).

d. Myocardial T2*

Myocardial T2* was reported as an outcome in one trial (Maggio 2009); baseline and endpoint data are shown in an additional table (Table 1). In this trial, myocardial T2* was measured by T2* MRI in a subset of patients with a mean (SD) duration from entry into the trial until the final MRI scan of 16 (5) months in the combined deferiprone and DFO group and 14 (6) months in patients receiving deferiprone alone. The differences in duration of follow up time across this subset of patients prohibited calculation of the mean change in myocardial iron concentration; however the trial reported no significant differences in the T2* signals of the heart between the two treatment groups.

e. Chelation efficiency

Chelation efficiency (%) was reported by one trial (as described in the previous comparison) (Aydinok 2005) (Table 3). There was no statistically significant difference in chelation efficiency between the treatment arms, MD -0.69% (95%CI -1.53 to 0.15) (Analysis 2.7).

f. Plasma non-transferrin bound iron (NTBI)

Plasma NTBI was measured in one trial as change in concentration from baseline (Aydinok 2005) (Table 3). Concentration was measured as millimolar (mM). Mean results at the end of treatment (time point not defined) were reported. There was a small statistically difference in mean plasma NTBI in at the end of treatment in favour of deferiprone and DFO, MD 2.16 mM (95% CI 1.49 to 2.83) (Analysis 2.8).

g. Total Iron excretion

Two trials reported total iron excretion (Aydinok 2005; Aydinok 2007). In both trials, total iron excretion per day was calculated as:

[iron transfused/year (mg) + (liver iron concentration at time 0 - liver iron concentration at time 1 year) x 106 x body weight in kg] / number of days of treatment.

In the first trial, total iron excretion was reported at the end of the trial (Aydinok 2005) (Table 3). There was a small statistically significant difference in mean total iron excretion in favour of deferiprone combined with DFO, MD -0.13 mg/kg/day (95%CI -0.21 to -0.05) (Analysis 2.9). In the second trial, mean change in total iron excretion from baseline was reported (Aydinok 2007). This trial also showed a statistically significant difference in the mean change in total iron excretion in favour of deferiprone and DFO compared with deferiprone alone, MD -0.21 mg/kg/day (95% CI -0.39 to -0.03) (Analysis 2.9).

3. Adverse events due to deferiprone

Adverse events were reported as an outcome in four trials (Aydinok 2007; El-Beshlawy 2008; Gomber 2004; Maggio 2009) although the Gomber trial did not report adverse events per treatment arm, i.e. between patients who received deferiprone alone or deferiprone with DFO, and therefore adverse events in this trial could not be incorporated into meta-analyses (Gomber 2004). See an additional table for details of adverse events reported in the individual trials (Table 5).

None of the trials reporting adverse events provided sufficient data to allow an analysis of the risk of experiencing cumulative adverse events between the treatment arms.

Permanent treatment withdrawal due to adverse events was observed in three trials (Aydinok 2007; El-Beshlawy 2008; Maggio 2009); this occurred in patients who received deferiprone alone and deferiprone with DFO in all three trials. Temporary treatment withdrawal was reported in one trial in participants receiving deferiprone (with or without DFO) due to arthropathy (Gomber 2004). Additionally, Magggio reported temporary treatment withdrawals, stating "no statistically significant difference in temporary and definitive discontinuation of treatment between the groups", but no further data were provided (Maggio 2009).

Two trials reported dose reduction (El-Beshlawy 2008; Maggio 2009). In the first of these trials, the dose of deferiprone was reduced to 50 mg/kg due to arthropathy; this was reported in four patients (two in each treatment arm) (El-Beshlawy 2008). In the latter trial, dose reduction was reported in 49% (deferiprone alone) and 56.5% (deferiprone with DFO) of participants; no further details were given (Maggio 2009).

Joint pain was observed in three trials (Aydinok 2007; El-Beshlawy 2008; Maggio 2009). In one trial, joint pain or arthralgia occurred in over 30% of all patients who received deferiprone (with or without DFO) (El-Beshlawy 2008). The combined risk of joint pain was not significantly higher for either treatment group, RR 1.03 (95% CI 0.53 to 2.01) (Analysis 2.10).

Two trials reported gastrointestinal disturbance or nausea or vomiting in both treatment arms, although neither treatment arm was associated with an increased risk, RR 1.74 (95% CI 0.90 to 3.35) (El-Beshlawy 2008; Maggio 2009) (Analysis 2.10). The third trial also reported nausea and vomiting, but for both treatment arms combined, and therefore, results could not be combined in a meta-analysis (Aydinok 2007).

Leucopenia or neutropenia, or both, were reported in three trials, in patients who received deferiprone alone and in patients who received deferiprone in combination with DFO (Aydinok 2007; El-Beshlawy 2008; Maggio 2009). One patient in the combined treatment group with neutropenia reported in the Aydinok trial subsequently developed agranulocytosis (Aydinok 2007). Agranulocytosis was also observed in a total of four patients who received deferiprone alone (El-Beshlawy 2008; Maggio 2009). No significant increased risk of leucopenia, neutropenia and/or agranulocytosis was associated with either treatment group, RR 0.71 (95% CI 0.38 to 1.32) (Analysis 2.10).

Two trials observed increased liver transaminase in patients receiving either deferiprone or the combined therapy, reported as at least two-fold increase in ALT (Maggio 2009) or jaundice or very high liver enzymes (El-Beshlawy 2008). Meta-analysis of these two trials showed that neither treatment arm was associated with a significantly increased risk of high liver transaminase, RR 0.73 (95% CI 0.45 to 1.17) (Analysis 2.10). A third trial reported "transient fluctuations" in ALT levels in patients who received DFO with deferiprone but no change in ALT levels in those who were treated with DFO alone (Aydinok 2007).

Two trials reported adverse events specifically related to the administration of DFO (Aydinok 2007; El-Beshlawy 2008). The first of these reported "mild local reactions observed in several patients treated with DFO" (Aydinok 2007), whilst the second reported the occurrence of skin reactions/swelling (3 out of 22, 13.6%) (El-Beshlawy 2008).

Other adverse events reported in both treatment arms included anorexia (El-Beshlawy 2008). Musculoskeletal pain was also reported in patients receiving the combined therapy (El-Beshlawy 2008).

4. Participant compliance

Three trials measured participant compliance (Aydinok 2007; El-Beshlawy 2008; Maggio 2009) (Table 6). Neither of the first two trials reported data for this outcome, although the first trial described overall compliance for the two treatment arms as "generally excellent during the entire study period" and noted one patient who missed more than one chelation dose per week due to problems with swallowing (Aydinok 2007). The second trial reported "excellent compliance during the study period" (El-Beshlawy 2008). In the third trial, mean compliance with deferiprone was reported as over 90% in both treatment arms, whereas mean compliance with DFO was 70.6% (Maggio 2009).

5. Cost of treatment

The cost of treatment was reported in one trial (Gomber 2004); full details are presented in an additional table (Table 7). As this trial was undertaken in India costs are presented in rupees. In a comparison of the costs of treatment per week of deferiprone versus deferiprone and DFO, deferiprone was the cheaper treatment: 126 rupees with deferiprone and 486 rupees with deferiprone and DFO. Costs were based on a participant weighing 10 kg.

C. Deferiprone and DFO in combination compared with DFO alone

There were nine comparisons between deferiprone and DFO in combination with DFO alone (Abdelrazik 2007; Aydinok 2005; El-Beshlawy 2008; Galanello 2006; Gomber 2004; Ha (i) 2006; Mourad 2003; Tamaddoni 2010; Tanner 2007).

Primary outcome
1. Mortality

Mortality as an outcome was not reported by any of these trials.

Secondary outcomes
1. Evidence of reduced end organ damage

Three trials reported evidence of reduced end organ damage as an outcome (Abdelrazik 2007; El-Beshlawy 2008; Tanner 2007). The El-Beshlawy trial reported only that "there was no significant difference in cardiac function" between treatment arms (El-Beshlawy 2008). The remaining two trials reported cardiac function by LVEF (%), measured by echocardiogram (Abdelrazik 2007) or cardiac magnetic resonance (Tanner 2007). Neither trial reported data to calculate mean change in LVEF from baseline (Abdelrazik 2007; Tanner 2007). However, both trials showed a significant difference in mean LVEF (%) at 12 months and when combined in a meta-analysis, LVEF was shown to be significantly reduced in patients who received DFO alone compared with deferiprone in combination with DFO, MD 6.22% (95% CI 4.32 to 8.12) (Analysis 3.1). Considerable heterogeneity was observed between these two trials (I2 = 89%); the difference between treatment arms remained significant under a random-effects model, MD 6.08% (95% CI 0.28 to 11.88). No clear clinical differences between these two trials were identified which could account for this heterogeneity.

One trial reported liver fibrosis as an outcome, but results were not reported separately for each treatment arm and therefore are not presented here (Ha (ii) 2006).

2. Measures of iron overload
a. Serum ferritin concentration

Serum ferritin concentration was reported as an outcome in eight trials (Abdelrazik 2007; El-Beshlawy 2008; Galanello 2006; Gomber 2004; Ha (i) 2006; Mourad 2003; Tamaddoni 2010; Tanner 2007); one trial presented data graphically and the mean change is estimated from the graph (El-Beshlawy 2008). Individual trial data are reported in an additional table (Table 1); mean change from baseline to endpoint data are reported in a further table (Table 10). One trial reported serum ferritin levels at baseline and end of trial as geometric means with coefficient of variation (Tanner 2007). In a second trial, data were observed as skewed (Mourad 2003); the skewed nature of these data was only observed because data were presented as individual patient data (IPD) (Mourad 2003); such data were not available in the remaining trials.

Table 10. Serum ferritin concentration (mean (SD)): deferiprone with DFO versus DFO
  1. SD: standard deviation
    a28/32 patients completed treatment – actual sample size values for serum ferritin not given
    bvalues given are geometric mean (CV: coefficient of variation)
    mths: months

Trial No Def and DFO/ DFO Def + DFO: baseline Def + DFO: end trial Absolute change % change DFO: baseline DFO: end trial Absolute change % change
Abdelrazik 200730/304500 (1250) ng/ml1250 (750) ng/mlreduction of 3250 ng/ml (SD not reported)reduction of 72%4250 (1500) ng/ml1200 (850) ng/mlreduction of 3050 ng/ml (SD not reported)reduction of 72%
El-Beshlawy 200818/202865 (983) µg/Lapprox. 1050 µg/L (estimated from graph)

reduction of approx. 1815 µg/L

 

not available2838 (967) µg/Lapprox. 1650 µg/L (estimated from graph)reduction of approx. 1188 µg/Lnot available
Galanello 200629/302048 (685) µg/Lapprox. 1750 µg/L (estimated from graph)reduction of 248 (791) ug/Lreduction of 12.1%2257 (748) µg/Lapprox. 1850 µg/L (estimated from graph)reduction of 349 (573) ug/Lreduction of 15.4%
Gomber 200410/ 73347.78 (+/- 1526.46) ng/ml3376.57 (1222.41) ng/mlincrease of 28.79 (915.86) ng/mlincrease of 0.9%5077.18 1714.99) ng/ml3718 (738.39) ng/mlreduction of 1358.87 (1374.14) ng/mlreduction of 26.8%
Ha (i) 200617/ 14not reportednot reportedreduction of 987.0 (2984) mmol/Lnot availablenot reportednot reportedincrease of 1059 (2285) mmol/Lnot available
Mourad 200311/ 144153 (1715) µg/L

6 mths: 3005 (1303) µg/L

12mths: 2805 (1084) µg/L

6 mths:

reduction of 1148 µg/L

12 mths:

reduction of 1348 µg/L

6 mths:

reduction of 27.6%

12 mths:

reduction of 32.5%

5506 (2375) µg/L

6 mths:

4856 (2615) µg/L

12 mths:

3998 (2409) µg/L

6 mths:

reduction of 650 µg/L

12 mths:

reduction of 1508 µg/L

6 mths:

reduction of 11.8%

12 mths:

reduction of 27.4%

Tamaddoni 201040/402986 (612) ng/ml

6 mths: 2453 (318) ng/ml

12 mths: 2082 (221) ng/ml

6 mths: reduction of 533 ng/ml

12 mths: reduction of 904 ng/ml

6 mths: reduction of 17.8%

12 mths: reduction of 30.3%

2945 (591) ng/ml

6 mths: 2702 (242) ng/ml

12 mths: 2451 (352) ng/ml

6 mths: reduction of 243 ng/ml

12 mths: 494 ng/ml

6 mths: reduction of 8.3%

12 mths: reduction of 16.8%

Tanner 200728/30a1574 (11) µg/Lb598 (CV 14) µg/Lbreduction of 976 µg/Lreduction of 62%1379 (CV 10) µg/Lb1146 (CV 11) µg/Lbreduction of 233 µg/Lreduction of 16.9%

i. At end of trial

In view of the skewed data observed in the Mourad trial and the form of reporting of results in the Tanner trial, serum ferritin concentration at endpoint was analysed on a log scale as ratio of geometric means (Mourad 2003; Tanner 2007).

Three trials reported serum ferritin at six months follow up; all three trials showed a significant difference between treatment arms in favour of the combined treatment of deferiprone with DFO (Gomber 2004; Mourad 2003; Tamaddoni 2010): ratio of geometric means 0.66 ng/ml (95% CI 0.48 to 0.91) (Gomber 2004); ratio of geometric means 0.64 ng/ml (95% CI 0.45 to 0.92) (Mourad 2003); ratio of geometric means 0.90 ng/ml (95% CI 0.86 to 0.95) (Tamaddoni 2010) (Analysis 3.2). Results at six months were not pooled in a meta-analysis due to pronounced baseline differences between the treatment arms in one trial (Gomber 2004).

At 12 months follow up, combined evidence from meta-analysis of three trials showed a significant difference between treatment arms, favouring deferiprone combined with DFO (Abdelrazik 2007; Mourad 2003; Tamaddoni 2010). The mean value in patients who received combination therapy was 0.86 times that in patients who received DFO alone, ratio of geometric means 0.86 (95% CI 0.81 to 0.91) (Analysis 3.3).

ii. Change from baseline

Data to calculate mean change in serum ferritin from baseline to end of trial were available in eight trials (Abdelrazik 2007; El-Beshlawy 2008; Galanello 2006; Gomber 2004; Ha (i) 2006; Mourad 2003; Tamaddoni 2010; Tanner 2007), although only three trials reported SDs or sufficient data to allow calculation of the SD (Galanello 2006; Gomber 2004; Ha (i) 2006). There were statistically significant differences in mean change in serum ferritin concentration at six months in two trials (Gomber 2004; Ha (i) 2006). In one trial, mean change favoured combined deferiprone and DFO, MD -2046.00 micromol/l (95%CI -3902.00 to -190.00) (Ha (i) 2006) (Analysis 3.4). In the second trial, mean change favoured DFO, MD 1387.66 ng/ml (95% CI 222.13 to 2553.19) (Gomber 2004) (Analysis 3.4). The CIs were wide suggesting considerable variation in the results for individual participants (Table 10), which may have arisen from the pronounced baseline differences in this trial (Gomber 2004). The Galanello trial, with mean change in serum ferritin concentration data at 12 months showed no significant difference in mean change, MD 101.00 ng/ml (95% CI -252.44 to 454.44) (Analysis 3.4) (Galanello 2006).

In the remaining trials with data reported for serum ferritin concentration, the mean change was a reduction in serum ferritin concentration for both treatment arms at six months (Mourad 2003; Tamaddoni 2010) and 12 months (Abdelrazik 2007; El-Beshlawy 2008; Mourad 2003; Tamaddoni 2010; Tanner 2007) (Table 10).

b. Urinary iron excretion

Urinary iron excretion was measured by five trials (Abdelrazik 2007; Aydinok 2005; El-Beshlawy 2008; Gomber 2004; Mourad 2003). Individual trial data are reported in an additional table (Table 1). In one of these trials, outcome data were only reported for one treatment arm and will not be addressed further in these results (Mourad 2003). Of the remaining four trials, three, did not present baseline data for this outcome, thus not enabling mean change from baseline to end of trial to be calculated (Aydinok 2005; El-Beshlawy 2008; Gomber 2004).

In the trial reporting data at six months, mean urinary iron excretion at the end of trial was not statistically different between the treatment arms, MD -0.82 mg/24h (95%CI -5.67 to 4.03) (Gomber 2004) (Analysis 3.5).Two trials showed a statistically significant difference in urinary iron excretion measurement after treatment, both in favour of combined treatment of deferiprone and DFO: one after 12 months, MD 0.23 mg/24h (95% CI 0.04 to 0.42) (Abdelrazik 2007); and one as a mean of quarterly readings, MD 0.27 mg/24h (95% CI 0.13 to 0.41) (El-Beshlawy 2008) (Analysis 3.5). An additional trial which reported mean % urinary iron excretion over the trial and showed a statistically significant difference between treatment arms in favour of combined deferiprone with DFO, MD 56.00% (95% CI 42.44 to 169.56) (Aydinok 2005) (Analysis 3.6). Data for mean urinary iron excretion after treatment were not pooled overall in a meta-analysis because the time points for the measurement of urinary iron excretion varied (six months (Gomber 2004) and 12 months (Abdelrazik 2007)) and because of variation in the means of measurement (mg/24h (Gomber 2004), as a mean of quarterly readings (El-Beshlawy 2008) and as a % mean urinary iron excretion (Aydinok 2005)).

c. Liver iron concentration

Liver iron concentration was reported as an outcome measure in four trials (El-Beshlawy 2008; Galanello 2006; Ha (i) 2006; Tanner 2007); mean change in liver iron concentration from baseline to end of trial was reported in two of these trials (Galanello 2006; Ha (i) 2006). The third trial reported liver iron concentration as a geometric mean of liver T2* (Tanner 2007). One trial presented data graphically and the mean change is estimated from the graph (El-Beshlawy 2008). Individual trial data are presented in an additional table (Table 1); mean change from baseline data are reported in a further table (Table 11).

Table 11. Liver iron concentration (mean (SD)): deferiprone with DFO versus DFO
  1. a values given are geometric mean of liver T2* (CV, coefficient of variation)
    d/w: dry weight liver
    w/w: wet weight liver

Trial Number Def + DFO: baseline Def + DFO: end Absolute change % change DFO: baseline DFO: end Absolute change % change
El-Beshlawy 200816/1517.1 (9.1) mg/g d/wapprox. 6.5 (3.5) mg/g d/w (estimated from graph)reduction of approx. 10.6 mg/g d/wnot reported22.5 (10.1) mg/g d/wapprox. 11.5 (6.3) mg/g d/w (estimated from graph)reduction of approx. 11 mg/g d/wnot reported
Galanello 200629/301629 (744) µg/g w/wnot reportedreduction of 65 (615) µg FE/g liverreduction of 4.0%1625 (642) µg/g w/wnot reportedreduction of 239 (474) µg FE/g liverreduction of 14.7%
Ha (i) 200617/14not reportednot reportedincrease of 0.95 (15.49) mg/g d/wnot availablenot reportednot reportedincrease of 0.82 (8.25) mg/g d/wnot available
Tanner 200728/324.9 (CV 0.52) msa10.7 msaincrease of 5.8 msaincrease of 18.4%4.2 (CV 0.62) msa5.0 msaincrease of 0.8 msaincrease of 20.0% 

Liver iron concentration was measured by atomic absorption spectrophotometry in one trial (Ha (i) 2006) and by magnetic spectrometry (SQUID) in a second trial (Galanello 2006). One trial used liver T2* measured by cardiovascular magnetic resonance (CMR) to quantify liver iron concentration (Tanner 2007). The method of liver biopsy assessment was not reported in one trial (El-Beshlawy 2008).

Two trials reporting mean change in liver iron concentration were not pooled in a meta-analysis due to variation in treatment duration: six months (Ha (i) 2006); and 12 months (Galanello 2006). Mean change in liver iron concentration was not significantly different between the treatment arms in either of these trials: MD 0.13 mg/g d/w (95% CI -8.41 to 8.67) (Ha (i) 2006); MD 0.17 mg/g w/w (95% CI -0.11 to 0.45) (Galanello 2006) (Analysis 3.7). In the Tanner trial, the geometric mean of liver T2* was reported at baseline and after treatment (Tanner 2007). Exact P values were not available for all trial arms and therefore SDs could not be calculated. However, this trial reported a between-group difference in geometric means of 39% (95% CI 20% to 61%) in favour of the combined treatment group.

d. Myocardial T2*

One trial reported myocardial T2* as an outcome measure (Tanner 2007) (Table 1). In this trial, the between group difference in geometric means of myocardial T2* was reported as significantly in favour of the combined treatment group, with an estimated 10% (95% CI 2% to 19%) increase in the combined group compared with the DFO group.

e. Chelation efficiency

Chelation efficiency (%) was reported in one trial as described above (Aydinok 2005) (Table 3). There was a statistically significant difference in mean chelation efficiency in favour of DFO, MD -15.76 (95%CI -25.16 to -6.36) (Analysis 3.8).

f. Plasma non-transferrin bound iron (NTBI)

Plasma NTBI was measured by one trial (Aydinok 2005) (Table 3). Mean results at the end of treatment (time point not defined) were reported. There was no statistically significant difference in mean plasma NTBI at the end of treatment between the treatment arms, MD 0.12 mM (95%CI -0.60 to 0.84) (Analysis 3.9).

g. Total iron excretion

In one trial, total iron excretion (mg/kg/day) was reported at the end of the trial (Aydinok 2005). Total iron excretion per day was calculated as:

[iron transfused/year (mg) + (liver iron concentration at time 0 - liver iron concentration at time 1 year) x 10.6 x body weight in kg] / number of days of treatment.

There was no statistically significant difference in total iron excretion between the treatment arms, MD -0.08 mg/kg/day (95%CI -0.31 to 0.15) (Analysis 3.10). See additional tables for actual trial data (Table 9).

3. Adverse events

Adverse events were measured as an outcome in eight trials (Abdelrazik 2007; El-Beshlawy 2008; Galanello 2006; Gomber 2004; Ha (i) 2006; Mourad 2003; Tamaddoni 2010; Tanner 2007); although one trial with three treatment arms did not differentiate adverse events between patients who received deferiprone alone or deferiprone with DFO and therefore this trial could not be included in meta-analyses (Gomber 2004). See an additional table for details of the adverse events experienced in each trial (Table 5). A further trial did not report adverse event incidences per treatment arm and adverse event data for this trial have not been reported in this review (Ha (i) 2006).

Two trials reported data that enabled a comparison of the risk of experiencing any adverse event (Abdelrazik 2007; Galanello 2006). Both trials observed a greater proportion of adverse events in patients receiving deferiprone with DFO than those receiving DFO alone. When data were pooled into a meta-analysis, there was a statistically significant increased risk of experiencing an adverse event in participants receiving deferiprone with DFO compared with those receiving DFO alone, RR 3.04 (95% CI 1.18 to 7.83) (Analysis 3.11; Figure 4). The remaining trials did not provide enough data to allow for an analysis of the risk of experiencing cumulative adverse events between the treatment arms.

Figure 4.

Forest plot of comparison: 3 Deferiprone and DFO in combination versus DFO alone, outcome: 3.11 Adverse events.

Three trials each reported permanent treatment withdrawals due to adverse events (El-Beshlawy 2008; Galanello 2006; Tanner 2007); only the latter trial reported permanent treatment withdrawals in the DFO only treatment arm. Two other trials did not observe any permanent withdrawals throughout the trial (Mourad 2003; Tamaddoni 2010).

A reduction in deferiprone dose was reported in two trials (El-Beshlawy 2008; Tamaddoni 2010). In the first of these, the deferiprone dose was reduced to 50 mg/kg in two participants due to arthropathy (El-Beshlawy 2008). The second trial reported temporary dose reduction in "some" patients with gastrointestinal disturbances; no further details were given (Tamaddoni 2010) .

Joint pain or arthralgia was reported as an adverse event in five trials (Abdelrazik 2007; El-Beshlawy 2008; Mourad 2003; Tamaddoni 2010; Tanner 2007). Two of these reported patients with joint pain in both treatment arms (El-Beshlawy 2008; Tanner 2007); a further trial observed arthralgia only in patients receiving the combined therapy (Tamaddoni 2010). Incidences of joint pain were also reported in the combined deferiprone with DFO treatment arm in the two other trials (Abdelrazik 2007; Mourad 2003), although the absence of joint pain in patients receiving DFO alone could not be clearly inferred and these trials could not be included in a meta-analysis. A meta-analysis of the first three trials showed no significant increased risk of joint pain or arthralgia for either treatment, RR 1.58 (95% CI 0.66 to 3.80) (Analysis 3.11; Figure 4).

Six trials reported gastrointestinal disturbance or nausea and vomiting (Abdelrazik 2007; El-Beshlawy 2008; Galanello 2006; Mourad 2003; Tamaddoni 2010; Tanner 2007). One trial reported gastrointestinal disturbance in both treatment arms (Tanner 2007), whilst the remaining five trials only observed gastrointestinal disturbance in patients receiving deferiprone combined with DFO (Abdelrazik 2007; El-Beshlawy 2008; Galanello 2006; Mourad 2003; Tamaddoni 2010). In three of these trials incidence of gastrointestinal disturbance in patients receiving DFO alone could not be clearly inferred without ambiguity (Abdelrazik 2007; Mourad 2003; Tamaddoni 2010). Additionally, in three trials the number of affected patients was reported separately for nausea or vomiting (or both), diarrhoea and abdominal pain; any overlap between the number of patients reported in each of these groups could not be ruled out (Abdelrazik 2007; Galanello 2006; Tamaddoni 2010). Meta-analysis of the two trials with sufficient data showed a significant increased risk of gastrointestinal disturbance associated with deferiprone combined with DFO, RR 2.11 (95% CI 1.02 to 4.37) (El-Beshlawy 2008; Tanner 2007) (Analysis 3.11; Figure 4).

Cases of neutropenia or leucopenia, or both, were reported in three trials (El-Beshlawy 2008; Galanello 2006; Tanner 2007). Two of these trials reported incidences of neutropenia in a total of three patients receiving deferiprone in combination with DFO (El-Beshlawy 2008; Tanner 2007), and one case of agranulocytosis in a patient in the combined therapy arm. One trial also reported neutropenia in one patient who received DFO alone (El-Beshlawy 2008). A further case of neutropenia in a patient receiving DFO alone was reported by a third trial (Galanello 2006). No cases of neutropenia or leucopenia were observed in the Abdelrazik trial. When data from these trials were pooled in a meta-analysis, no significant increased risk of neutropenia, leucopenia or agranulocytosis (or a combination of these) was observed for either treatment arm, RR 1.80 (95% CI 0.45 to 7.17) (El-Beshlawy 2008; Galanello 2006; Tanner 2007) (Analysis 3.11; Figure 4).

Three trials observed patients with increased liver transaminase occurring in both treatment arms, reported as: elevated ALT at least two-fold above normal values (Abdelrazik 2007); a transient increase in ALT three times the normal upper limit in HCV positive patients (Galanello 2006); and as jaundice or very high liver enzymes (El-Beshlawy 2008). When combined, no significantly increased risk of increased liver transaminase was observed in either treatment group, RR 1.73 (95% CI 0.83 to 3.61) (Analysis 3.11; Figure 4). A fourth trial reported transient fluctuations in serum alanine ALT levels in patients receiving combined deferiprone and DFO treatment but not in patients receiving DFO alone (Tamaddoni 2010). A fifth trial reported that no significant change in ALT was observed over the trial period in either treatment group (Tanner 2007).

Adverse events specifically related to the administration of DFO were reported in five trials (El-Beshlawy 2008; Galanello 2006; Mourad 2003; Tamaddoni 2010; Tanner 2007). Skin allergic reactions occurred in 8 out of 45 (17.8%) (El-Beshlawy 2008), 11 out 40 (27.5%) (Tamaddoni 2010) and 12 out 14 (85.7%) (Mourad 2003) of all patients, and in 6% and 3% in the DFO and combined treatment arms respectively (Tanner 2007). Abscess at the site of infusion was also reported by two trials, in 1 out of 30 (3.3%) patients (Galanello 2006) and 1 out of 40 (2.5%) patients (Tamaddoni 2010).

Other adverse events in patients receiving deferiprone with DFO reported in one trial included musculoskeletal pain, anorexia, and insomnia (El-Beshlawy 2008).

4. Participant compliance

Participant compliance with DFO was measured in six trials (Abdelrazik 2007; El-Beshlawy 2008; Galanello 2006; Ha (i) 2006; Mourad 2003; Tanner 2007); see an additional table for individual trial data (Table 6). Two trials reported compliance rates with DFO (Galanello 2006; Tanner 2007) which in the first trial was 96.1% in the combined treatment arm and 95.7% for DFO alone (Galanello 2006), and in the second trial was 91.4% in the combined treatment arm and 92.6% for DFO alone (Tanner 2007). The latter trial also reported a compliance rate of 82.4% for deferiprone; deferiprone compliance was not reported in the other trial.

Two further trials did not provide compliance rates, but reported compliance descriptively. In one trial, compliance was rated as "excellent" in 91% of participants in the deferiprone and DFO arm and 79% of participants in the DFO arm; and as "good" in 9% of participants in the deferiprone and DFO arm and 21% of participants in the DFO arm (Mourad 2003). The final trial reported "excellent compliance during the study period" and documented four patients who were excluded from the trial due to lack of compliance, all of whom received DFO alone (El-Beshlawy 2008).

Although reported as measured, the remaining two trials did not report the results for this outcome (Abdelrazik 2007; Ha (i) 2006).

5. Cost of treatment

The cost of treatment was reported in one trial (Gomber 2004). Full details are presented in an additional table (Table 7). As this trial was undertaken in India, cost is presented in rupees. In a comparison of the costs of treatment per week of deferiprone and DFO versus DFO, deferiprone and DFO was the cheaper treatment, 486 rupees with deferiprone and DFO and 900 rupees with DFO. Costs were based on a participant weighing 10 kg.

D. Deferiprone dose A versus deferiprone dose B

One trial compared different doses of deferiprone (50 mg/kg/day versus 75 mg/kg/day) (Choudhry 2004).

Primary outcome
1. Mortality

This trial reported one death in a patient receiving deferiprone at a dose of 50 mg/kg/day. This individual, who presented with leucopenia with neutropenia, developed severe respiratory tract infection after 11 months of therapy and following treatment with oral amoxacillin for 48 hours, was hospitalised and died within 24 hours of this diphtheria-like infection.

Secondary outcomes
1. Evidence of reduced end organ damage

Evidence of reduced end organ damage as an outcome was not reported by this trial.

2. Measures of iron overload
a. Serum ferritin concentration

Data to calculate mean change in serum ferritin from baseline to end of trial were available in this trial (Table 1). The mean change was a reduction in serum ferritin in both treatment arms; the MD was not calculated due to the absence of SDs or sufficient data (i.e. a correlation coefficient) to enable calculation of these. However, the MD between treatment arms at the end of the trial was not significant between the two doses of deferiprone, MD -223.00 ng/ml (95% CI -1342.67 to 896.67) (Analysis 4.1).

b. Urinary iron excretion

Urinary iron excretion was not reported as an outcome in this trial.

c. Liver iron concentration

Liver iron concentration was not reported as an outcome in this trial.

d. Myocardial T2*

Myocardial T2* was not reported as an outcome in this trial.

e. Chelation efficiency

Chelation efficiency (%) was not reported as an outcome in this trial.

f. Plasma non-transferrin bound iron (NTBI)

Plasma NTBI was not reported as an outcome in this trial.

g. Total Iron excretion

Total iron excretion was not reported as an outcome in this trial.

3. Adverse events

Adverse events associated with deferiprone were reported in this trial, see an additional table (Table 5). In particular, joint pain was reported in 29% of patients who received the higher dose of deferiprone (75 mg/kg/day) compared with 50% in patients who received the lower dose (50 mg/kg/day). Cases of leucopenia with neutropenia were also reported in both treatment arms with similar rates: 23% in patients who received the lower dose of deferiprone; and 24% in patients who received the higher dose. Of the 12 patients on either treatment dose who experienced neutropenia with leucopenia, treatment was withdrawn temporarily in 10 patients and withdrawn permanently in one patient, whilst one patient died from infection. A total of 19 patients reported infections as an adverse event; these were not reported separately for each treatment arm.

4. Participant compliance with deferiprone treatment

Participant compliance with deferiprone treatment was not reported as an outcome in this trial.

5. Cost of treatment

The cost of treatment was not reported as an outcome in this trial.

Discussion

The aims of this systematic review were to summarise data from trials on the efficacy and safety of deferiprone as an iron chelating agent in people with transfusion-dependent thalassaemia and to compare the efficacy and safety of deferiprone with desferrioxamine. Both aims were severely compromised by incompatible trial design or reporting of data, or both, and clinical diversity of participants between the trials.

This updated review complements a concurrent systematic review of iron chelation entitled 'Desferrioxamine mesylate for managing transfusional iron overload in people with transfusion-dependent thalassaemia', which we previously undertook and which we have now also updated (Fisher 2013). All but one of the trials presented in this current review are common to both papers (Choudhry 2004).

Seventeen RCTs were eligible for analysis in this review. Within these trials there were five different intervention comparisons: five trials compared deferiprone to DFO; two trials compared deferiprone alone to deferiprone combined with DFO; six trials compared deferiprone combined with DFO to DFO alone; three trials included a three-way comparison of deferiprone, DFO and deferiprone with DFO; and one trial compared different doses of deferiprone.

There was little opportunity for meta-analysis for the majority of outcomes; analysis of data was principally a commentary on the findings from each included trial. Few trials measured long-term outcomes, mortality was only measured in four trials and evidence of reduced end-organ damage was measured in six trials. Although measures of iron overload were reported in all but one trial, meta-analysis was not possible for most outcomes due to different methods of assessment and measurement of outcomes used, variation in time points and a lack of sufficient data reporting. Trials aimed at evaluating iron chelators may benefit from the development and application of agreed standardised sets of outcomes to be collected and reported, ideally through the Core Outcome Measures in Effectiveness Trials (COMET) initiative (www.comet-initiative.org), which will better enable results of trials to be compared, contrasted and combined as appropriate (see also 'Implications for research' below).

Mortality was reported by four trials; each trial reported the death of one individual receiving deferiprone (with or without DFO). Most of these deaths were due to cardiac complications and could not be attributed to therapy. However, one patient receiving deferiprone presented with leucopenia with neutropenia, developed a severe respiratory tract infection after 11 months and died as a result of this infection (Choudhry 2004). One trial reported five further deaths in patients who withdrew from randomised treatment (deferiprone with or without DFO) and switched to DFO alone (Maggio 2009).

Reporting of long-term outcomes was limited and inconclusive; few trials reported evidence of reduced end organ damage as an outcome. The effect of deferiprone and DFO has attracted some interest since reports that cardiac iron load may be reduced by deferiprone (Pennell 2006). The earlier trials measuring cardiac iron load indirectly by measurement of the MRI T2* signal had suggested deferiprone may reduce cardiac iron more quickly than DFO. In a direct comparison of patients with low T2* (high cardiac iron) receiving deferiprone or DFO, T2* values increased in both treatment arms; the increase was two-fold higher at 12 months in patients who received deferiprone (26.9% increase from baseline) compared with those who received DFO (12.8%) (Pennell 2006). In a trial of DFO compared with DFO and deferiprone, increases in myocardial T2* (representing a reduction in cardiac iron) was significantly in favour of the combined treatment group, with an estimated 10% (95% CI 2% to 19%) increase in the combined group compared with the DFO group (Tanner 2007).

In the comparison of deferiprone versus DFO monotherapy, meta-analysis of three trials which reported LVEF as a measure of cardiac function showed no significant differences between treatment arms (Maggio 2002; Olivieri 1997; Pennell 2006). However, in a meta-analysis of two trials which compared LVEF between patients who received combined deferiprone with DFO and those who received DFO alone, LVEF was shown to be significantly reduced in patients who received DFO alone compared with combination therapy (Abdelrazik 2007; Tanner 2007).

There have been reports of deaths due to heart disease in trial participants. A recent trial did report a sudden death, probably due to a cardiac arrhythmia, although the death was not attributed to deferiprone (Ha (ii) 2006). Two further cardiac-related deaths were reported in patients receiving deferiprone (with or without DFO) (Aydinok 2007; Maggio 2009). There are many reports of the incidence of cardiac disease in observational studies, but it is impossible to compare the rates of cardiac dysfunction or disease in such studies.

The significance of differences in improvements in LVEF and MRI T2* and how they translate into the ability to prevent cardiac disease is unclear in the absence of more data and longer follow up of randomised participants (Neufeld 2006). It is clear from retrospective and prospective clinical studies that intensified DFO treatment, by either subcutaneous or intravenous route, can reverse cardiac dysfunction because of iron overload and increase survival in thalassaemia-major patients with early or overt cardiomyopathy (Maggio 2007). Any benefit of deferiprone to prevent heart disease needs to be evaluated in appropriately-powered trials with robust measures of cardiac iron load and function.

In the absence of long-term follow up in all but one trial (Maggio 2009) and with limited data on mortality or end organ damage, data on the effects of chelation therapy rely on markers of iron overload or measurements of iron loading or excretion.

In the direct comparison of deferiprone and DFO, after six months, a statistically significant difference in mean change in serum ferritin concentration in favour of DFO was observed in two trials (Gomber 2004; Pennell 2006); a third trial found no significant difference between treatment arms (Ha (ii) 2006). However, none of the trials which reported longer-term follow up showed a significant difference in mean change in serum ferritin concentration between treatment arms, at 12 months (Maggio 2002; Pennell 2006) or 24 months (Olivieri 1997). 

In a single large trial with four years of follow up, there was a significant difference in mean change of serum ferritin concentration in patients receiving deferiprone in combination with DFO compared to deferiprone alone (Maggio 2009). Furthermore, in three trials comparing DFO to DFO and deferiprone, meta-analysis showed a significant difference between treatment arms, favouring the combined therapy (Abdelrazik 2007; Mourad 2003; Tamaddoni 2010).

These results suggest an advantage of combined therapy with DFO and deferiprone over monotherapy to reduce iron stores as measured by serum ferritin. Can such advantages for combination therapy be supported by other measurements of iron fluxes or stores, or both? Measures of urinary iron excretion are difficult to interpret as comparative measures of efficacy since these estimates do not include biliary iron excretion, and so would underestimate the total iron excretion for DFO. Hence no real conclusion can be made examining measurements of urinary iron excretion in five trials comparing deferiprone with DFO (Aydinok 2005; El-Beshlawy 2008; Gomber 2004; Maggio 2002; Olivieri 1990).

In three trials which compared deferiprone alone with combination therapy, no significant differences were observed in serum ferritin concentration between treatment groups either at the end of the trial or as mean change from baseline, with the exception of one trial with a planned follow up of five years (Maggio 2009) In this trial, a significant difference in mean change from baseline was observed in favour of patients receiving deferiprone combined with DFO over deferiprone monotherapy; this significant difference was maintained over four years of study. Of note, this large trial was terminated early due to the beneficial effects of combined treatment in terms of serum ferritin reduction compared with deferiprone alone.

In the comparison of combination therapy with DFO alone, all three trials which reported serum ferritin concentration after six months reported significantly lower serum ferritin values in patients who received DFO combined with deferiprone (Gomber 2004; Mourad 2003; Tamaddoni 2010). Furthermore, after 12 months, meta-analysis of three trials showed a significant difference between treatment arms in favour of deferiprone combined with DFO (Abdelrazik 2007; Mourad 2003; Tamaddoni 2010). Two trials which reported a significant difference between treatment arms in mean change in serum ferritin concentration from baseline favoured different treatment regimens.

Urinary iron excretion was measured in six trials comparing deferiprone with DFO (Aydinok 2005; El-Beshlawy 2008; Gomber 2004; Maggio 2002; Olivieri 1990; Olivieri 1997). However, in all but one of these trials, no baseline data were presented for this outcome and therefore mean change from baseline to end of trial could not be calculated (Maggio 2002). As already stated, these estimates of urinary iron excretion do not include biliary iron excretion, and so would underestimate the total iron excretion for DFO.

Of three trials which compared deferiprone with combination therapy, only one showed a significant difference in urinary iron concentration at the end of the trial in favour of deferiprone (Aydinok 2005). However, in the comparison of combined therapy with DFO alone, two trials showed a significant difference in urinary iron concentration at the end of the trial in favour of deferiprone combined with DFO (Abdelrazik 2007; El-Beshlawy 2008). In addition, one trial which reported mean percentage urinary iron excretion over the trial period showed a significantly greater level of urinary iron excretion in patients who received combined therapy compared with DFO alone (Aydinok 2005). Only one other trial reported urinary iron excretion for this comparison and this trial found no significant difference between treatment arms (Gomber 2004). Data from the majority of trials did not allow analysis of mean change from baseline between treatment arms. Nevertheless, the data suggest addition of deferiprone to DFO increases urinary iron excretion over DFO alone.

One trial used three other measures of iron burden and chelation, namely total iron excretion, chelation efficiency and plasma NTBI (Aydinok 2005). There was a small statistically significant difference in total iron excretion in favour of deferiprone compared with deferiprone and DFO. The results of other comparisons were not significant.

Liver iron concentration is a valuable measure of stored iron. Indeed, direct measurement of liver iron is the gold standard by which other assays are validated. Five trials comparing deferiprone with DFO reported liver iron concentration (El-Beshlawy 2008; Ha (ii) 2006; Maggio 2002; Olivieri 1997; Pennell 2006). However, conflicting results were observed between trials. Liver iron concentration at the end of the trial was significantly lower in the deferiprone treatment group in one trial (El-Beshlawy 2008), significantly lower in the DFO group in a second trial (Olivieri 1997), and showed a non-significant difference in favour of DFO in a third trial (Maggio 2002). None of three trials reporting mean change from baseline found significant differences between treatment groups (Ha (ii) 2006; Olivieri 1997; Pennell 2006).

Neither trial which reported liver iron concentration for the comparison of deferiprone alone versus combination therapy found significant differences in liver iron concentration at the end of the trial (Aydinok 2007; El-Beshlawy 2008). Similarly, in the comparison of deferiprone with DFO versus DFO alone, neither trial reporting this outcome found a significant difference in mean change from baseline between treatment arms (Galanello 2006; Ha (i) 2006).

Pooling of liver iron concentration was generally was prevented by differences in techniques used for measurement of liver iron content between trials (SQUID, atomic spectrophotometry, liver T2*) as well as the variable presence of hepatitis C and the wide variation in treatment duration between and possibly within trials. In addition, the degree of reporting of results (displayed graphically or a lack of SDs or CIs) in a number of trials was inadequate to formally assess differences between treatment groups.

Only one trial compared different doses of deferiprone (Choudhry 2004). This trial only reported serum ferritin concentration as a measure of iron overload. No difference in serum ferritin concentration at the end of the trial was observed between patients who received 50 mg/kg/day and those who received 75 mg/kg/day. Evidence of reduced end organ damage was not reported as an outcome in this trial.

There is therefore no conclusive or consistent evidence for the improved efficacy of combined deferiprone and DFO therapy against monotherapy from direct or indirect measures of liver iron.

Given the apparently similar efficacy of deferiprone and DFO, the safety, compliance and cost of these drugs are of special interest. Indeed the safety of chelation therapy has been somewhat controversial. Treatment with DFO has been in clinical usage for some 20 years longer than deferiprone and adverse events as a result of DFO therapy have been observed, with epidemiological data suggesting that adverse events may be dose-related (Porter 1989a; Porter 2002). The longer-term safety profile is much less certain for deferiprone than for DFO, although in the late 1990s liver fibrosis appeared to be associated with deferiprone during a clinical trial (Olivieri 1998).

Adverse event data were reported in 14 trials, but differences in how adverse events were measured (number of events or events per person), the time points for measurement and the lack of standard reporting of the severity of reactions precluded pooling of data across the trials in many instances.

In a direct comparison of DFO against deferiprone, meta-analysis showed an 11.7-fold increased risk of gastrointestinal side effects associated with deferiprone (El-Beshlawy 2008; Maggio 2002), an 8.9-fold increased risk of raised liver transaminase levels (El-Beshlawy 2008; Maggio 2002) and a 2.6-fold increased risk of joint pain or arthralgia (El-Beshlawy 2008; Maggio 2002; Pennell 2006). Furthermore, the FDA review of the primary data from the trial conducted by Pennell concluded that "Regarding safety, adverse events related to elevation of serum alanine aminotransferase levels were reported in 38% of the deferiprone group and in 18% of the deferoxamine group. In the context of additional concerns, this observation signals the potential for deferiprone induced liver toxicity." (Pennell 2006).

Local adverse events specifically related to the administration of DFO are well described. Local reactions at infusion sites occurred in 17.8% (n = 45) (El-Beshlawy 2008), 27.5% (n = 40) (Tamaddoni 2010), 85.7% (n = 14) (Mourad 2003) and 38.7% (n = 31) (Pennell 2006) of patients; local abscesses at the site of infusion occurred in 3.3% (n = 30) (Galanello 2006) and 2.5% (n = 40) (Tamaddoni 2010) of patients; and systemic allergy occurred in 4.3% (n = 23) of patients (El-Beshlawy 2008).

Three trials reported data that enabled a comparison of the risk of experiencing any adverse event. In a comparison of DFO versus deferiprone, there was a statistically significant 2.2-fold increased risk of experiencing an adverse event in participants receiving deferiprone compared with those receiving DFO (Maggio 2002). Furthermore, two trials observed a greater proportion of adverse events in patients receiving deferiprone with DFO than those receiving DFO alone; meta-analysis showed a statistically significant three-fold increased risk of experiencing an adverse event in participants receiving deferiprone with DFO compared with those receiving DFO alone (Abdelrazik 2007; Galanello 2006). The danger of raised liver enzymes or agranulocytosis with deferiprone means that close monitoring of full blood counts and liver function is required and precludes its use where close monitoring is not available.

However, these conclusions must be viewed with some caution, given that RCTs are not designed to measure the adverse effects of an intervention and thus data from RCTs does not represent a formal comparison of adverse events caused by iron chelation therapy. A complete review of adverse events should be the subject of a separate, formal analysis incorporating data from non-RCTs and observational studies. Nevertheless, reporting of serious adverse events from post-marketing surveillance, has shown that use of deferiprone may be associated with fatal agranulocytosis and when used at more than 100 mg/kg/day a neurological syndrome of cerebellar and psychomotor retardation that has progressively regressed after deferiprone has been discontinued (Henter 2007; Swedish Orphan Drug 2007). It is now recommended that weekly full blood counts are monitored in people receiving deferiprone (Swedish Orphan Drug 2007).

A further major concern of DFO is compliance. It has been suggested that participants given an oral iron chelator may be more compliant with treatment than those given a continuous subcutaneous infusion of DFO. Eleven trials reported compliance, with this suggestion being supported in two trials (Maggio 2009; Olivieri 1997) although reporting of compliance in other trials was often purely descriptive. The observed differences may be medically important, especially as the difference in compliance associated with these two iron chelators may become greater outside the rigours and discipline of a formal RCT setting. Thus, objective assessment of safety and compliance with these chelating agents will require studies with long-term follow up in a variety of settings.

Several factors have contributed to the disappointing absence of meta-analyses of trial results. Trials included participants with substantial clinical and demographic diversity which was not always fully defined. In particular, there were differences in the measurement of outcomes across trials, notably the different techniques used to assess liver iron concentration between trials, the presence of hepatitis C in participants in one trial (Maggio 2002), the skewed nature of the data in two trials (Maggio 2002; Mourad 2003) and the undermining of the randomisation process (Gomber 2004). Some trials failed to report baseline data, making valid comparisons impossible between trials. Moreover, where reported, baseline values for serum ferritin and liver iron concentration differed substantially between trials, and in two trials between treatment arms (Gomber 2004; Pennell 2006). The impact of these differences on any overall results is difficult to evaluate due to the limited number of trials and the small sample sizes of the included trials. The skewed nature of the data in two trials gives rise to concerns of the likely skewing of the data in the remaining trials. Without obtaining the IPD from all trials, this issue could not be confirmed.

Variation in the time points used for outcome measurement was also a factor that precluded meta-analysis. While very short-term outcome measurement was the appropriate objective of the two iron balance trials and measures of iron overload can be assessed over a six or twelve month period, much longer follow up is needed to examine the effect of iron chelation on mortality and reduced end-organ damage or other toxicity.

The validity of the data is a third issue to note. With the exception of two trials (Maggio 2002; Maggio 2009), the sample sizes of the included trials were small with 40 or fewer per treatment arm. Only six trials presented information on sample sizes required to power the trial around a main outcome, whilst several trials failed to clearly state the primary endpoint for analysis. Data were presented in abstract form for one trial, presumably with a lack of peer review and limiting the amount of data that can be reported.

The risk of bias of the included trials was difficult to assess given the general absence of information on randomisation and blinding to treatment allocation. As such, the influence of the quality of the included trials was not explored with respect to the robustness of any results. With the interventions in these trials, blinding to treatment allocation would be impossible for clinician and participant in all but one trial in which different doses of deferiprone were compared. However, blinding of outcome assessors would be desirable as a means of minimising bias, particularly where subjective measurements such as functional status and histological results are concerned.

Although no quantitative assessment of publication bias was undertaken, attempts were made to minimise the likelihood of publication bias by the use of a comprehensive search strategy, the handsearching of relevant conference abstract books and contacting the manufacturers of deferiprone and other iron chelators.

This review has highlighted the difficulty of undertaking an analysis in such intervention comparisons, not least the limited ability to pool and meta-analyse crucial aspects of measures of iron stores and chelation across the trials. These difficulties seem to be inherent in the population being studied. Children with thalassaemia may enter trials at different points in their clinical course, after variable amounts of blood transfusion, with variable endogenous iron loading, with different prior treatment and with differing biochemical levels as a result of past treatment and their thalassaemia. It seems unlikely that any one particular methodology could overcome these problems in future trials. It would therefore be important that each RCT comparing iron chelators or different schedules, doses or methods of administration of iron chelators should provide extensive details of baseline measures, collect follow-up data at times appropriate to the outcomes of interest and be sufficiently powered to provide a significant comparison within the trial.

These difficulties highlight the importance not only in designing trials to optimise subsequent meta-analysis, but of ensuring IPD are made available to the scientific community. Perhaps deposition of the IPD for an RCT will ultimately become a requirement for publication in the same way as molecular data for experimental data is deposited in public databases now. Moreover, consensus on the appropriate methods and sampling to measure iron overload would greatly help data from future trials to be combined. Finally, accurate recording of compliance and adverse events for different schedules of iron chelation should be an essential feature of future trials.

There is clearly an absence of adequate RCTs for understanding the relative benefit of DFO and deferiprone in long-term studies. Nevertheless, there has been some considerable use of combined DFO and deferiprone therapy. There are now many reports of a decline in the incidence of cardiac disease and mortality in observational studies since the implementation of regular measurement of myocardial T2* to detect those patients at risk of developing cardiac disease and subsequent intensification of chelation therapy in those with MRI evidence of myocardial iron loading. It is unclear whether the observed declines in mortality have followed improved detection of those at risk of cardiac disease or specific chelation regimens.

The absence of high quality data from RCTs to support specific recommendations for the use of deferiprone is not just the conclusion of this review but also the considered view of the FDA in the United States of America. The FDA reviewed an application to license deferiprone in 2011, having previously rejected an application in 2009. At that time Apotex, the manufacturer of deferiprone, was advised that the FDA would require at least one new RCT as well as a full audit of the original trial led by Dr Olivieri in Toronto. However, neither further RCTs nor other requested audits were submitted by Apotex, and Apotex abandoned its application for full approval despite previous and continued claims of efficacy and safety. Downgraded approval of deferiprone under the lower 'accelerated' standards, as "last resort treatment of iron overload in thalassemia, myelodysplasia and sickle cell disease" was later provided. This decision arose from the lack of new RCT evidence and the failure to provide answers to the FDA's questions on efficacy and safety.

In summary, there were insufficient data available to fulfil the aim of this review; namely to determine the effectiveness of the iron chelating agent deferiprone in people with transfusion-dependent thalassaemia and to compare the efficacy and safety of deferiprone with other iron chelating agents. Deferiprone is indicated for the treatment of iron overload in patients with thalassaemia major when DFO is contraindicated or inadequate. Intensified DFO treatment by either subcutaneous or intravenous route and/or use of other oral iron chelators remains the established treatment to reverse cardiac dysfunction because of iron overload. Indeed, the FDA have approved deferiprone only as "last resort treatment of iron overload in thalassemia" and combination therapy with DFO and deferiprone has been used in spite of the lack of RCT data in specific clinical situations, particularly where there is evidence of myocardial iron deposition or where compliance with continuous DFO is a severe problem, or both. Factors are suggested that should be considered in further trials to improve the efficacy and applicability of deferiprone in people with iron overload.

Authors' conclusions

Implications for practice

Both deferiprone and DFO produce a significant reduction in iron stores in transfusion-dependent, iron-overloaded people. There is no evidence to suggest that either treatment is more clinically efficacious over the length of these trials or has a greater reduction of clinically significant end organ damage.

The significance of differences in improvements in LVEF and MRI T2* and how these are translated into the ability to prevent cardiac disease is unclear in the absence of more data and longer follow up of randomised participants. Intensified DFO treatment (subcutaneously or intravenously) or use of other oral iron chelators, or both, remains the established treatment to reverse cardiac dysfunction because of iron overload.

The results of this review do suggest an advantage of combined therapy with DFO and deferiprone over monotherapy to reduce iron stores as measured by serum ferritin. However, there is no conclusive or consistent evidence for the improved efficacy of combined deferiprone and DFO therapy against monotherapy from direct or indirect measures of liver iron.

Direct comparison of DFO versus deferiprone showed an increased risk of gastrointestinal side effects, increased liver transaminase levels and joint pain or arthralgia associated with deferiprone. Local adverse events specifically related to the administration of DFO are well-described. Deferiprone, used alone or in combination with DFO, is associated with a statistically significant increased risk of side effects compared to DFO.

There is no evidence to change current treatment recommendations, namely that deferiprone is indicated for the treatment of iron overload in patients with thalassaemia major when DFO is contraindicated or inadequate. The FDA have approved deferiprone as "last resort treatment of iron overload in thalassemia". The spectrum and incidence of adverse events vary between the two drugs and people treated with both drugs must be kept under close medical supervision. In particular, the danger of raised liver enzymes or agranulocytosis with deferiprone means that close monitoring of full blood counts and liver function is required and this precludes its use where close monitoring is not available. The major difference in the route of administration between oral deferiprone and subcutaneous infusion of DFO may lead to differences in compliance, but the majority of the included trials in which compliance was reported achieved good to excellent compliance with deferiprone and DFO. Therefore, practical issues, patient factors and cost will continue to be the determinants of choice.

Implications for research

It is a major concern that adequate RCTs to demonstrate the clinical efficacy and safety of deferiprone for such an important clinical indication have not been completed some 15 years after the drug was initially available for clinical use.

There are four implications for further research, derived from this review.

First, the most urgent need is for long-term (i.e. minimum of five years) prospective trials starting in early life, before there is a significant iron loading and carried through to adolescence and beyond.

Secondly, future trials should include outcomes that enable a comparison of the long-term clinical efficacy and safety of these agents. Trials comparing the two drugs should use comparable and combinable measures of pharmacological and clinical outcomes that have been validated or may be validated as associated with outcome. Liver and cardiac iron burden, liver histology and cardiac function are of special interest. Measures and outcomes should therefore include serum ferritin, liver histology, direct measures of hepatic iron stores and indirect measures of hepatic and cardiac iron using magnetic resonance or similar techniques, and measures of cardiac function including LVEF. To combine data it is necessary to have full data including the SD for baseline and end of trial data and the change between start and finish for all participant groups. Ideally, agreed standardised sets of outcomes should be collected and reported, ideally though the COMET initiative (www.comet-initiative.org), which will better enable results of trials to be compared, contrasted and combined as appropriate. Where necessary, agreement to use IPD should be forthcoming.

Thirdly, information is required about minimal clinically important differences for people with transfusion-dependent thalassaemia receiving iron chelation therapy, i.e. what level of difference in tissue (liver and cardiac iron) concentration would lead to a clinically important difference in survival. This information need not be derived from RCTs, but from observational studies specifically designed for this task. The reporting of adverse events would require comparable indices and outcomes to be used across trials to allow more comprehensive pooling of data.

Finally, there are few data on the relative cost of these chelating agents, but this will clearly be of major interest in many countries where iron chelators are widely used.

Acknowledgements

This project (Ref. NIHR-RP-PG-0310-1004) is supported by National Institute of Heath Research (NIHR), UK.

For the update (2013):

We thank Dr Sally Hopewell for providing methodological advice for this review.

For the original review (2006):

We thank Dr Senani Williams (eligibility assessment, data extraction and analysis, quality assessment and content expert), Dr Jo Howard (development of protocol and content expert) and Professor Chris Hyde (development of protocol, data analysis and methodological expert) for their contribution as authors to the original review.

We thank Neelam Thapar for her thoughts and comments on the protocol and Simon Stanworth for his help with data verification and comments on the review.

We thank Jon Deeks and Phil Alderson for their advice and comments on the methods section of this review.

We thank Carol Lefebvre for providing advice on the search strategy.

Dr Antoni Addis prepared an initial draft for this review and we acknowledge his contribution. However, Dr Addis had no involvement with the preparation of the protocol and was not involved in any stage of the review process.

Data and analyses

Download statistical data

Comparison 1. Deferiprone alone versus DFO alone
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Left ventricular ejection fraction: mean change from baseline (%)3 Mean Difference (IV, Random, 95% CI)Subtotals only
1.1 At 6 months161Mean Difference (IV, Random, 95% CI)1.48 [-0.08, 3.04]
1.2 At 12 months3245Mean Difference (IV, Random, 95% CI)1.76 [-1.42, 4.93]
1.3 At 24 months123Mean Difference (IV, Random, 95% CI)7.6 [-0.65, 15.85]
2 Right ventricular ejection fraction: mean at endpoint (%)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
2.1 At 6 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
2.2 At 12 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
3 Liver fibrosis Ishak score: mean at endpoint1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
3.1 At 12 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
4 Serum ferritin concentration: mean change from baseline (ng/ml)5 Mean Difference (IV, Fixed, 95% CI)Totals not selected
4.1 At 6 months3 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
4.2 At 12 months2 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
4.3 At 24 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
5 Urinary iron excretion: mean at endpoint (mg/24h)4 Mean Difference (IV, Fixed, 95% CI)Totals not selected
5.1 As a mean of quarterly readings1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
5.2 At 12 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
5.3 Early after starting treatment1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
5.4 24 hours after starting treatment1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
6 Urinary iron excretion: mean over study (%)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
7 Urinary iron excretion: mean change from baseline (mg/24h)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
7.1 At 12 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
8 Liver iron concentration: ratio of geometric means at endpoint (mg/g dry weight)3 Ratio of GM (Fixed, 95% CI)Totals not selected
8.1 At 12 months1 Ratio of GM (Fixed, 95% CI)0.0 [0.0, 0.0]
8.2 At 24 months1 Ratio of GM (Fixed, 95% CI)0.0 [0.0, 0.0]
8.3 At 31-36 months2 Ratio of GM (Fixed, 95% CI)0.0 [0.0, 0.0]
9 Liver iron concentration: mean change from baseline (mg/g dry weight)3 Mean Difference (IV, Fixed, 95% CI)Totals not selected
9.1 At 6 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
9.2 At 12 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
9.3 At 24 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
10 Myocardial T2*: ratio of geometric means of change from baseline1 Ratio of GM (Fixed, 95% CI)Totals not selected
10.1 At 6 months1 Ratio of GM (Fixed, 95% CI)0.0 [0.0, 0.0]
10.2 At 12 months1 Ratio of GM (Fixed, 95% CI)0.0 [0.0, 0.0]
11 Chelation efficiency (%)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
11.1 At end of trial1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
12 Plasma NTBI: mean change from baseline (mM)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
12.1 At end of trial1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
13 Total iron excretion: mean at endpoint (mg/kg/day)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
13.1 At end of trial1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
14 Adverse events5 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
14.1 Number of participants experiencing an adverse event1144Risk Ratio (M-H, Fixed, 95% CI)2.24 [1.19, 4.23]
14.2 Risk of pain or swelling in joints3248Risk Ratio (M-H, Fixed, 95% CI)2.64 [1.21, 5.77]
14.3 Risk of gastrointestinal disturbances2188Risk Ratio (M-H, Fixed, 95% CI)11.71 [1.57, 87.31]
14.4 Risk of leucopenia, neutropenia and/or agranulocytosis5323Risk Ratio (M-H, Fixed, 95% CI)2.51 [0.66, 9.55]
14.5 Risk of increased liver transaminase2188Risk Ratio (M-H, Fixed, 95% CI)8.90 [2.08, 38.11]
15 Participant compliance (%)2 Mean Difference (IV, Fixed, 95% CI)Totals not selected
15.1 At 12 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
15.2 At 3 years1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
Analysis 1.1.

Comparison 1 Deferiprone alone versus DFO alone, Outcome 1 Left ventricular ejection fraction: mean change from baseline (%).

Analysis 1.2.

Comparison 1 Deferiprone alone versus DFO alone, Outcome 2 Right ventricular ejection fraction: mean at endpoint (%).

Analysis 1.3.

Comparison 1 Deferiprone alone versus DFO alone, Outcome 3 Liver fibrosis Ishak score: mean at endpoint.

Analysis 1.4.

Comparison 1 Deferiprone alone versus DFO alone, Outcome 4 Serum ferritin concentration: mean change from baseline (ng/ml).

Analysis 1.5.

Comparison 1 Deferiprone alone versus DFO alone, Outcome 5 Urinary iron excretion: mean at endpoint (mg/24h).

Analysis 1.6.

Comparison 1 Deferiprone alone versus DFO alone, Outcome 6 Urinary iron excretion: mean over study (%).

Analysis 1.7.

Comparison 1 Deferiprone alone versus DFO alone, Outcome 7 Urinary iron excretion: mean change from baseline (mg/24h).

Analysis 1.8.

Comparison 1 Deferiprone alone versus DFO alone, Outcome 8 Liver iron concentration: ratio of geometric means at endpoint (mg/g dry weight).

Analysis 1.9.

Comparison 1 Deferiprone alone versus DFO alone, Outcome 9 Liver iron concentration: mean change from baseline (mg/g dry weight).

Analysis 1.10.

Comparison 1 Deferiprone alone versus DFO alone, Outcome 10 Myocardial T2*: ratio of geometric means of change from baseline.

Analysis 1.11.

Comparison 1 Deferiprone alone versus DFO alone, Outcome 11 Chelation efficiency (%).

Analysis 1.12.

Comparison 1 Deferiprone alone versus DFO alone, Outcome 12 Plasma NTBI: mean change from baseline (mM).

Analysis 1.13.

Comparison 1 Deferiprone alone versus DFO alone, Outcome 13 Total iron excretion: mean at endpoint (mg/kg/day).

Analysis 1.14.

Comparison 1 Deferiprone alone versus DFO alone, Outcome 14 Adverse events.

Analysis 1.15.

Comparison 1 Deferiprone alone versus DFO alone, Outcome 15 Participant compliance (%).

Comparison 2. Deferiprone alone versus deferiprone and DFO in combination
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Left ventricular ejection fraction: mean at endpoint (%)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
1.1 At 12 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
2 Serum ferritin concentration: mean at endpoint (ng/ml)3 Mean Difference (IV, Fixed, 95% CI)Totals not selected
2.1 At 6 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
2.2 At 12 months2 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
3 Serum ferritin concentration: mean change from baseline (ng/ml)2 Mean Difference (IV, Fixed, 95% CI)Totals not selected
3.1 At 6 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
3.2 At 12 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
3.3 At 2 years1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
3.4 At 3 years1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
3.5 At 4 years1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
3.6 At 5 years1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
4 Urinary iron excretion: mean over study (%)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
5 Urinary iron excretion: mean at endpoint (mg/24h)2 Mean Difference (IV, Fixed, 95% CI)Totals not selected
5.1 Early after starting treatment1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
5.2 As a mean of quarterly readings (mg/kg/day)1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
6 Liver iron concentration: mean at endpoint (mg/g dry weight)2 Mean Difference (IV, Fixed, 95% CI)Subtotals only
6.1 At 12 months253Mean Difference (IV, Fixed, 95% CI)1.45 [-0.91, 3.82]
7 Chelation efficiency (%)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
7.1 At end of trial1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
8 Plasma NTBI: mean change from baseline (mM)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
8.1 At end of trial1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
9 Total Iron excretion: mean at endpoint (mg/kg/day)2 Mean Difference (IV, Fixed, 95% CI)Totals not selected
9.1 At end of trial1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
9.2 Mean change from baseline1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
10 Adverse Events3 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
10.1 Risk of pain or swelling in joints3217Risk Ratio (M-H, Fixed, 95% CI)1.03 [0.53, 2.01]
10.2 Risk of gastrointestinal disturbances2196Risk Ratio (M-H, Fixed, 95% CI)1.74 [0.90, 3.35]
10.3 Risk of leucopenia, neutropenia and/or agranulocytosis3217Risk Ratio (M-H, Fixed, 95% CI)0.71 [0.38, 1.32]
10.4 Risk of increased liver transaminase2196Risk Ratio (M-H, Fixed, 95% CI)0.73 [0.45, 1.17]
Analysis 2.1.

Comparison 2 Deferiprone alone versus deferiprone and DFO in combination, Outcome 1 Left ventricular ejection fraction: mean at endpoint (%).

Analysis 2.2.

Comparison 2 Deferiprone alone versus deferiprone and DFO in combination, Outcome 2 Serum ferritin concentration: mean at endpoint (ng/ml).

Analysis 2.3.

Comparison 2 Deferiprone alone versus deferiprone and DFO in combination, Outcome 3 Serum ferritin concentration: mean change from baseline (ng/ml).

Analysis 2.4.

Comparison 2 Deferiprone alone versus deferiprone and DFO in combination, Outcome 4 Urinary iron excretion: mean over study (%).

Analysis 2.5.

Comparison 2 Deferiprone alone versus deferiprone and DFO in combination, Outcome 5 Urinary iron excretion: mean at endpoint (mg/24h).

Analysis 2.6.

Comparison 2 Deferiprone alone versus deferiprone and DFO in combination, Outcome 6 Liver iron concentration: mean at endpoint (mg/g dry weight).

Analysis 2.7.

Comparison 2 Deferiprone alone versus deferiprone and DFO in combination, Outcome 7 Chelation efficiency (%).

Analysis 2.8.

Comparison 2 Deferiprone alone versus deferiprone and DFO in combination, Outcome 8 Plasma NTBI: mean change from baseline (mM).

Analysis 2.9.

Comparison 2 Deferiprone alone versus deferiprone and DFO in combination, Outcome 9 Total Iron excretion: mean at endpoint (mg/kg/day).

Analysis 2.10.

Comparison 2 Deferiprone alone versus deferiprone and DFO in combination, Outcome 10 Adverse Events.

Comparison 3. Deferiprone and DFO in combination versus DFO alone
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Left ventricular ejection fraction: mean at endpoint (%)2 Mean Difference (IV, Fixed, 95% CI)Subtotals only
1.1 At 12 months2118Mean Difference (IV, Fixed, 95% CI)6.22 [4.32, 8.12]
2 Serum ferritin concentration: ratio of geometric means at endpoint (ng/ml)3 Ratio of GM (Fixed, 95% CI)Totals not selected
2.1 At 6 months3 Ratio of GM (Fixed, 95% CI)0.0 [0.0, 0.0]
3 Serum ferritin concentration: ratio of geometric means at endpoint (ng/ml)3 Ratio of GM (Fixed, 95% CI)Subtotals only
3.1 At 12 months3 Ratio of GM (Fixed, 95% CI)0.86 [0.81, 0.91]
4 Serum ferritin concentration: mean change from baseline (ng/ml)3 Mean Difference (IV, Fixed, 95% CI)Totals not selected
4.1 At 6 months2 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
4.2 At 12 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
5 Urinary iron excretion: mean at endpoint (mg/24h)3 Mean Difference (IV, Fixed, 95% CI)Totals not selected
5.1 At 6 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
5.2 At 12 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
5.3 As a mean of quarterly readings1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
6 Urinary iron excretion: mean over study (%)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
7 Liver iron concentration: mean change from baseline2 Mean Difference (IV, Fixed, 95% CI)Totals not selected
7.1 At 6 months (mg/g dry weight)1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
7.2 At 12 months (mg/g wet weight)1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
8 Chelation efficiency (%)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
8.1 At end of trial1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
9 Plasma NBTI: mean change from baseline (mM)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
9.1 At end of trial1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
10 Total iron excretion: mean at endpoint (mg/kg/day)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
10.1 At end of trial1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
11 Adverse events5 Risk Ratio (M-H, Fixed, 95% CI)Subtotals only
11.1 Number of participants experiencing an adverse event2119Risk Ratio (M-H, Fixed, 95% CI)3.04 [1.18, 7.83]
11.2 Risk of pain or swelling in joints3190Risk Ratio (M-H, Fixed, 95% CI)1.58 [0.66, 3.80]
11.3 Risk of gastrointestinal disturbances2106Risk Ratio (M-H, Fixed, 95% CI)2.11 [1.02, 4.37]
11.4 Risk of leucopenia, neutropenia and/or agranulocytosis4226Risk Ratio (M-H, Fixed, 95% CI)1.80 [0.45, 7.17]
11.5 Risk of increased liver transaminase3131Risk Ratio (M-H, Fixed, 95% CI)1.73 [0.83, 3.61]
Analysis 3.1.

Comparison 3 Deferiprone and DFO in combination versus DFO alone, Outcome 1 Left ventricular ejection fraction: mean at endpoint (%).

Analysis 3.2.

Comparison 3 Deferiprone and DFO in combination versus DFO alone, Outcome 2 Serum ferritin concentration: ratio of geometric means at endpoint (ng/ml).

Analysis 3.3.

Comparison 3 Deferiprone and DFO in combination versus DFO alone, Outcome 3 Serum ferritin concentration: ratio of geometric means at endpoint (ng/ml).

Analysis 3.4.

Comparison 3 Deferiprone and DFO in combination versus DFO alone, Outcome 4 Serum ferritin concentration: mean change from baseline (ng/ml).

Analysis 3.5.

Comparison 3 Deferiprone and DFO in combination versus DFO alone, Outcome 5 Urinary iron excretion: mean at endpoint (mg/24h).

Analysis 3.6.

Comparison 3 Deferiprone and DFO in combination versus DFO alone, Outcome 6 Urinary iron excretion: mean over study (%).

Analysis 3.7.

Comparison 3 Deferiprone and DFO in combination versus DFO alone, Outcome 7 Liver iron concentration: mean change from baseline.

Analysis 3.8.

Comparison 3 Deferiprone and DFO in combination versus DFO alone, Outcome 8 Chelation efficiency (%).

Analysis 3.9.

Comparison 3 Deferiprone and DFO in combination versus DFO alone, Outcome 9 Plasma NBTI: mean change from baseline (mM).

Analysis 3.10.

Comparison 3 Deferiprone and DFO in combination versus DFO alone, Outcome 10 Total iron excretion: mean at endpoint (mg/kg/day).

Analysis 3.11.

Comparison 3 Deferiprone and DFO in combination versus DFO alone, Outcome 11 Adverse events.

Comparison 4. Deferiprone dose A versus deferiprone dose B
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Serum ferritin concentration: mean at endpoint (ng/ml)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
1.1 At 12 months1 Mean Difference (IV, Fixed, 95% CI)0.0 [0.0, 0.0]
Analysis 4.1.

Comparison 4 Deferiprone dose A versus deferiprone dose B, Outcome 1 Serum ferritin concentration: mean at endpoint (ng/ml).

Appendices

Appendix 1. Search strategies for review update (March 2013)

CENTRAL (The Cochrane Library)
#1       IRON CHELATING AGENTS explode all trees (MeSH)
#2       CHELATION THERAPY single term (MeSH)
#3       deferiprone OR L1* OR kelfer* OR DMHP* or ferriprox* OR cp20 OR dmohpo OR hdmpp NEXT cpd OR hdpp
#4       exjade* or deferasirox* or (icl NEXT 670) or icl670* or (cgp NEXT 72670) or cgp72670
#5       deferoxamine* or deferoximine* or deferrioxamine* or desferioximine* or desferrioxamine* or desferroxamine* or desferal* or desferral* or DFO or  desferin* or desferol* or dfom
#6       (iron NEAR/5 (chelat* or reduc*))
#7       #1 OR #2 OR #3 OR #4 OR #5 OR #6
#8       THALASSEMIA explode all trees (MeSH)
#9       IRON OVERLOAD explode all trees (MeSH)
#10     thalassemi* OR thalassaemi*
#11     cooley* anemia OR cooley* anaemia
#12     hemoglobin NEAR disease OR haemoglobin NEAR disease
#13     mediterranean anemia* OR mediterranean anaemia*
#14     erythroblastic anemia* OR erythroblastic anaemia*
#15     iron NEAR overload*
#16     hemochromatosis or haemochromatosis OR hemosiderosis OR haemosiderosis
#17     #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16
#18     #7 AND #17

PubMed (epublications only)
(deferiprone[ti] OR L1[ti] OR ferriprox[ti] OR deferoxamine[ti] OR deferoximine[ti] OR deferrioxamine[ti] OR desferioxamine[ti] OR desferrioxamine[ti] OR iron chelat*[ti] OR iron overload[ti]) AND (random* or trial or study or group or groups or cohort* or allocat* or assign*) AND (publisher[sb] NOT pubstatusnihms)

MEDLINE (Ovid)
1.      exp IRON CHELATING AGENTS/
2.      CHELATION THERAPY/
3.      (deferoxamine* or deferoximine* or deferrioxamine* or desferioximine* or desferrioxamine* or desferroxamine* or desferal* or desferral* or desferin* or desferol* or DFO or dfom).mp.
4.      (deferiprone or L1* or kelfer or DMHP or ferriprox or cp20 or dmohpo or (hdmpp adj cpd) or hdpp).mp.
5.      (exjade* or deferasirox* or (icl adj 670*) or icl670* or (cgp adj "72670")).mp.
6.      (iron adj5 (chelat* or reduc*)).tw.
7.      or/1-6
8.      exp THALASSEMIA/
9.      exp IRON OVERLOAD/
10.    (thalassemi* or thalassaemi*).tw.
11.    (cooley* and (anemi* or anaemi*)).tw.
12.    ((hemoglobin or haemoglobin) adj3 disease).tw.
13.    (iron adj3 overload*).tw.
14.    ((mediterranean or erythroblastic) adj (anemi* or anaemi*)).tw.
15.    (hemochromatosis or haemochromatosis or hemosiderosis or haemosiderosis).tw.
16.    or/7-15
17.    6 and 16

Embase (Ovid)
1.      IRON CHELATING AGENT/
2.      CHELATION THERAPY/    
3.      DEFEROXAMINE/
4.      DEFEROXAMINE MESYLATE/ 
5.      DEFERIPRONE/   
6.      DEFERASIROX/
7.      (deferoxamine* or deferoximine* or deferrioxamine* or desferioximine* or desferrioxamine* or desferroxamine* or desferal* or desferral* or desferin* or desferol* or DFO or dfom).mp.
8.      (exjade* or deferasirox* or (icl adj 670*) or icl670* or (cgp adj 72670).mp.
9.      (deferiprone OR L1 OR kelfer OR DMHP OR ferriprox* OR cp20 OR dmohpo OR (hdmpp adj cpd) OR hdpp OR deferrum OR deferum).mp.
10.    (iron adj5 (chelat* or reduc*)).tw.
11.    or/1-10
12.    exp THALASSEMIA/
13.    IRON OVERLOAD/
14.    (thalassemi* or thalassaemi*).tw.
15     (cooley* and (anemi* or anaemi*)).tw.
16.    ((hemoglobin or haemoglobin) adj3 disease).tw.
17.    (iron adj3 overload*).tw.
18.    ((mediterranean or erythroblastic) adj (anemi* or anaemi*)).tw.
19.    (hemochromatosis or haemochromatosis or hemosiderosis or haemosiderosis).tw.
20.    or/11-19
21.    11 and 20

Transfusion Evidence Library, KoreaMed, IndMed, PakMediNet, WHO ICTRP, Hong Kong Trials Register & ClinicalTrials.gov
defer* OR desfer* OR DFO OR ferriprox OR exjade OR icl 670 or icl670 OR chelat*

LILACS & ISRCTN REGISTER
(deferiprone OR L1 OR kelfer OR DMHP or ferriprox or deferoxamine or deferoximine or deferrioxamine or desferioximine or desferrioxamine) AND (thalassemia OR thalassaemia OR thalassemic OR thalassaemic OR anemia OR anaemia) OR (desferroxamine or desferal or desferral or DFO or dfom or desferol or desferin or exjade or deferasirox or “icl 670” or icl670 or “cgp 72670”) AND (thalassemia OR thalassaemia OR thalassemic OR thalassaemic OR anemia OR anaemia)

Appendix 2. Search strategies for original review (June 2006)

CENTRAL (The Cochrane Library) & UK National Research Register
#1      IRON CHELATING AGENTS explode all trees (MeSH)
#2      CHELATION THERAPY single term (MeSH)
#3      deferiprone
#4      L1* OR kelfer* OR DMHP* or ferriprox* OR cp20 OR dmohpo OR hdmpp  
          NEXT cpd OR hdpp
#5      #1 OR #2 OR #3 OR #4
#6      THALASSEMIA explode all trees (MeSH)
#7      thalassemia* OR thalassaemia* OR thalassemic* OR thalassaemic*
#8      cooley* anemia OR cooley* anaemia
#9      hemoglobin NEAR disease OR haemoglobin NEAR disease
#10    mediterranean anemia* OR mediterranean anaemia*
#11    erythroblastic anemia* OR erythroblastic anaemia*
#12    #6 OR #7 OR #8 OR #9 OR #10 OR #11
#13    #5 AND #12 

MEDLINE (Ovid)
1.      exp IRON CHELATING AGENTS/
2.      CHELATION THERAPY/
3.      DEFERIPRONE.mp.
4.      30652-11-0.RN.
5.       (L1 OR kelfer OR DMHP OR ferrirprox OR cp20 OR dmohpo OR (hdmpp adj cpd) OR hdpp).ti,ab.       
6.      or/1-5
7.      exp THALASSEMIA/
8.      (THALASSEMIA$ OR THALASSAEMIA$ OR THALASSEMIC$ OR THALASSAEMIC$).ti,ab.
9.      (COOLEY$ ANEMIA$ OR COOLEY$ ANAEMIA$1).ti,ab.
10.     (HEMOGLOBIN H DISEASE OR HAEMOGLOBIN H DISEASE).ti,ab.
11.     (HEMOGLOBIN F DISEASE OR HAEMOGLOBIN F DISEASE).ti,ab.
12.     (MEDITERRANEAN ANEMIA$ OR MEDITERRANEAN ANAEMIA$).ti,ab.
13.     (ERYTHROBLASTIC ANEMIA$ OR ERYTHROBLASTIC ANAEMIA$).ti,ab.
14.     or/7-13
15.     6 AND 14

Embase (Ovid)
1.      DEFERIPRONE/
2.      CHELATION THERAPY/
3.      DEFERIPRONE.mp.
4.      30652-11-0.RN.
5.       (L1 OR kelfer OR DMHP OR ferriprox$ OR cp20 OR dmohpo OR (hdmpp adj cpd) OR hdpp).ti,ab.
6.      or/1-5
7.      exp THALASSEMIA/
8.      (THALASSEMIA$ OR THALASSAEMIA$ OR THALASSEMIC$ OR THALASSAEMIC$).ti,ab.
9.      (COOLEY$ ANEMIA$ OR COOLEY$ ANAEMIA$1).ti,ab.
10.     (HEMOGLOBIN H DISEASE OR HAEMOGLOBIN H DISEASE).ti,ab.
11.     (HEMOGLOBIN F DISEASE OR HAEMOGLOBIN F DISEASE).ti,ab.
12.     (MEDITERRANEAN ANEMIA$ OR MEDITERRANEAN ANAEMIA$).ti,ab.
13.     (ERYTHROBLASTIC ANEMIA$ OR ERYTHROBLASTIC ANAEMIA$).ti,ab.
14.     or/7-13
15.     6 AND 14

Biological Abstracts, Cambridge Scientific Abstracts, ISI Web of Knowledge, Zetoc, ClinicalTrials.gov & Current Controlled Trials
(deferiprone OR L1 OR kelfer OR DMHP or ferriprox) AND (thalassemia OR thalassaemia OR thalassemic OR thalassaemic OR anemia OR anaemia OR hemoglobin disease OR haemoglobin disease)

NHSBT SRI Handsearched RCT Procite Database
#4=defer OR #4=L1 OR #4=kelfer OR #4=DMHP or #4=ferriprox or #4=dfo or #4=exjade* or #4=icl670 or #4="icl 670" or #4="cgp 72670" or #4=dfom

What's new

DateEventDescription
22 August 2013AmendedContact person changed and funding sources (internal and external) corrected.

History

Protocol first published: Issue 3, 2004
Review first published: Issue 3, 2007

DateEventDescription
7 August 2013New search has been performed

Eight new trials have been included in the review update (Abdelrazik 2007; Aydinok 2007; Choudhry 2004; El-Beshlawy 2008; Galanello 2006; Maggio 2009; Tamaddoni 2010; Tanner 2007).

One trial by Fassos previously included in the original review has now been confirmed as excluded in this update (Fassos 1994).

All studies identified from the most recent search (March 2013) which appear to meet the eligibility criteria have been listed as 'Studies awaiting classification' for inclusion in the next update.

7 August 2013New citation required and conclusions have changed

The adverse event analyses from the original review have been changed to calculate the risk associated with treatment (i.e. in all individuals) rather than only in individuals experiencing an adverse event.

Evidence of reduced end organ damage data have been added for all trials.

Minor data errors throughout the original review have been amended and data have been re-analysed where necessary.

The conclusion of the US Food and Drug Administration (FDA) who reviewed the evidence surrounding the efficacy and safety for an application for the licensing of deferiprone in the USA has been added to the Background and Discussion sections. The FDA only gave support for deferiprone to be used as a last resort treatment of iron overload in thalassemia, myelodysplasia and sickle cell disease.  

10 September 2008AmendedConverted to new review format.

Contributions of authors

Sheila Fisher (for the 2013 update): searching and selection of trials; eligibility assessment; data extraction and analysis; input of trial results; quality assessment; methodological expert; wrote the abstract, results and discussion sections for the update.

Susan Brunskill: development of protocol; searching and selection of trials; eligibility assessment; data extraction and analysis; input of trial results; quality assessment; methodological expert; wrote the abstract, results and discussion sections for the original review.

Carolyn Doree: eligibility assessment; searching and selection of trials; data extraction and analysis; quality assessment and verification of data entry.

Onima Chowdhury: (for the 2013 update): data extraction.

Sarah Gooding: (for the 2013 update): data extraction.

David Roberts: development of protocol and content expert.

All authors were involved in drawing conclusions and making specific recommendations for future research.

Declarations of interest

None known.

Sources of support

Internal sources

  • NHS Blood and Transplant, Research and Development, UK.

External sources

  • National Institute for Health Research (NIHR) Oxford Biomedical Research Centre Programme (SF, CD) and the NIHR under its Programme Grant Scheme (NIHR-RP-PG-0310-1004, SF), UK.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Abdelrazik 2007

Methods

2-arm parallel randomised controlled trial.
Number of centres: 1.
Duration of treatment:12 months.
Follow up: not stated.

Number (N) of participants:
N randomised: 60 (30 to each arm).
N analysed: 60 (30 to each arm).

ParticipantsChildren with thalassaemia.
Age: treatment group: mean (SD) 12 (5.1) years; comparator group: mean (SD) 14 (2.1) years.
Sex: treatment group: 60% male; comparator group: 50% male.
Ethnicity: not stated.
Trial undertaken: outpatient clinic of a children's hospital in Egypt.
Interventions

Treatment group: deferiprone: 75 mg/kg/day for 4 days/week taken 1 hour before meals with DFO: 40 mg/kg/day subcutaneously for 8 - 10 hours on 2 days/week; (DFO mean (SD) dose at baseline: 37 (4.5) & end of trial: 37.9 (7.8) mg/kg/day].

Comparator group: DFO 40 mg/kg/day subcutaneously for 8 - 10 hours on 5 - 7 days/week (mean (SD) dose at baseline: 35.5 (7.5) & end of trial: 34.4 (6.5) mg/kg/day).

All participants had 1 day off chelation therapy at the weekend.

Outcomes
  1. "Efficacy" (primary outcome)

  2. Compliance

  3. Safety

  4. Serum ferritin

  5. Total iron binding capacity and ferritin concentrations

  6. Urinary iron excretion

  7. Liver & renal function

  8. Cardiac assessment (LVEF & shortening function)

  9. Iron balance

NotesPrior exposure to iron chelators: DFO at a daily dose of 40 mg/kg/day by subcutaneous infusion pump over 8 - 10 hours, 5 - 7 nights a week for "several years".
Source of funding: not stated.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe authors did not report any information about how randomisation was undertaken.
Allocation concealment (selection bias)Unclear riskThe authors did not report any information about how treatment allocation was concealed.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskCardiac assessment was undertaken by a clinician blinded to treatment allocation. It was unclear as to whether there was blinding to treatment allocation for assessment of the other outcomes. It was unclear whether participants and clinicians administering treatment were blinded to treatment allocation.
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll randomised participants were included in the analysis for all outcomes: there were no missing outcome data.
Selective reporting (reporting bias)High riskThe authors reported that whilst liver iron concentration was measured, this could only be performed in a small number of participants and the data were not presented in the manuscript. The manuscript also states that total iron binding capacity and ferritin concentrations would be measured, but data for this outcome were not reported.
Other biasLow riskThe trial appears to be free of other sources of bias.

Aydinok 2005

Methods

3-arm parallel RCT.
Number of centres: not stated, but assumed 2.
Duration of treatment: not stated.
Follow up: not stated.

Number (N) of participants:
N randomised: 25 (treatment group: 12; comparator group 1: 5; comparator group 2: 8).
N analysed: 25 (treatment group: 12; comparator group 1: 5; comparator group 2: 8).

ParticipantsThalassaemia participants.
Age: not stated.
Sex: not stated.
Ethnicity: not stated.
Trial undertaken: not stated.
Interventions

Treatment group: deferiprone: 75 mg/kg/day orally, 7 days a week.

Comparator group 1: deferiprone: 75 mg/kg/day and DFO 40 - 50 mg/kg/day as a subcutaneous infusion, 2 days a week.

Comparator group 2: DFO by subcutaneous injection, 40 - 50 mg/kg/day over 5 days/week.

Outcomes
  1. Measures of iron overload: liver iron concentration; total iron excretion; chelation efficiency (%); average urinary iron excretion; plasma NTBI

Primary outcome: not identified.

NotesPrior exposure to iron chelators: all participants had previously received DFO.
Source of funding: not stated.
Abstract publication.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe authors did not report any information about how randomisation was undertaken.
Allocation concealment (selection bias)Unclear riskThe authors did not report any information about how treatment allocation was concealed.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskThe authors did not report any information as to whether participants, personnel or outcome assessors were blinded to treatment allocation.
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll randomised participants were included in the analysis for all outcomes: there were no missing outcome data.
Selective reporting (reporting bias)High riskLiver iron concentration was identified as an outcome to be measured in this trial but whilst it was reported to be measured, no actual data were presented rather the data for this outcome were converted into 'iron excretion' in mg/kg/day and only this is recorded.
Other biasUnclear riskThe trial was reported in an abstract thus there is little data available to make an assessment of whether the trial was free of other bias.

Aydinok 2007

Methods

2-arm parallel RCT.
Number of centres: 1
Duration of treatment: 12 months.
Follow up: not stated.

Number (N) of participants:
N randomised: 24 (12 to each arm).
N analysed: 20 (treatment group: 12; comparator group: 8). Trial reported details of why data from 4 participants were not analysed.

ParticipantsParticipants aged 4 years or older with thalassaemia major.
Age: treatment group: mean (SD) 15.9 (4.2) years; comparator group: mean (SD) 16.6 (4.8) years. Age range for both: 9 - 23 years.
Sex: not stated.
Ethnicity: not stated.
Trial undertaken: Turkey.
InterventionsTreatment arm: deferiprone: 75 mg/kg/day, daily (mean (SD) dose: 78.2 (2.6) mg/kg/day).

Comparator group: deferiprone: 75 mg/kg/day, daily (mean (SD) dose: 78.2 (1.4) mg/kg/day) with DFO 50 mg/kg/day subcutaneously twice weekly (mean (SD) dose: 43.8 (2.8) mg/kg).
Outcomes
  1. Changes in liver iron concentration and serum ferritin (primary)

  2. Total iron excretion

  3. Urinary iron excretion

  4. Iron balance

  5. Cardiac function (Echo)

  6. Toxicity

  7. Assessment of tolerance to treatment and quality of life

NotesAll participants had had prior exposure to DFO (dose, schedule and duration were not reported) and all had a washout period of 2 weeks with no iron chelation before initiating trial treatment.
Source of funding: none stated although the drugs were supplied by Lipomed AG, Switzerland.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"The randomization sequence was generated by the Department of Mathematical Statistics at the University of Berne, Switzerland according to local policy". Following central registration of a subject by the investigator, the trial co-ordinator assigned the intervention according to the randomisation sequence.
Allocation concealment (selection bias)High riskThe trial report states that the intervention was assigned according to the randomisation sequence "without concealing the sequence prior to allocation".
Blinding (performance bias and detection bias)
All outcomes
Unclear riskThe authors did not report any information as to whether participants, personnel or outcome assessors were blinded to treatment allocation.
Incomplete outcome data (attrition bias)
All outcomes
High riskThere was an imbalance in missing data across the treatment arms. 4 participants from the comparator group (DFO) were not included in the outcome analysis: 2 withdrew consent due to refusal to take DFO; 1 died from arrhythmia-induced congestive heart failure at the start of the trial and 1 developed agranulocytosis at week 14.
Selective reporting (reporting bias)Low riskAll outcomes pre-specified were reported in the manuscript.
Other biasLow riskThe trial appears to be free of other sources of bias.

Choudhry 2004

Methods

3-arm parallel RCT.
Number of centres: 1
Duration of treatment: 12 months.
Follow up: not stated.

Number (N) of participants:
N randomised: 75 (treatment group: 30; comparator group 1: 21; comparator group 2: 24).
N analysed: 75 (treatment group: 30; comparator group 1: 21; comparator group 2: 24) for all outcomes except mortality where the number analysed was deferiprone: 48 (details of the 'missing' 3 were reported); comparator group 2: 24

ParticipantsPaediatirc participants with thalassaemia major.
Age range for both: 4 - 14 years.
Sex: 55% male participants across the whole trial.
Ethnicity: not stated.
Trial undertaken: India
Interventions

Treatment arm: deferiprone: 50 mg/kg/day, daily.

Comparator group 1: deferiprone: 75 mg/kg/day, daily.

Comparator group 2: control: no treatment.

Outcomes
  1. Ferritin

  2. Leucopenia/agranulocytosis

  3. Arthropathy

Primary outcome: not identified.

NotesNo participant had received iron chelation therapy before trial enrolment.
Source of funding: trial supported by CIPLA Ltd, makers of deferiprone.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskEmail correspondence between a review author and the corresponding author identified the following information regarding randomisation methods. "We make slips indicating the dose of deferiprone in a box. Patients were asked to pick up one slip from the box. Child was given the dose as mentioned on the slip".
Allocation concealment (selection bias)Low riskEmail correspondence between a review author and the corresponding author identified the following information regarding randomisation methods. "We make slips indicating the dose of deferiprone in a box. Patients were asked to pick up one slip from the box. Child was given the dose as mentioned on the slip".
Blinding (performance bias and detection bias)
All outcomes
Unclear riskThe authors did not report any information as to whether participants, personnel or outcome assessors were blinded to treatment allocation.
Incomplete outcome data (attrition bias)
All outcomes
High riskFor all pre-specified outcomes, there were no missing data. However, for the non-pre-specified outcome, there were missing data in the deferiprone treated arms and whilst the reason why the participants (n = 3) were not included in the analysis was reported, the authors did not state which of the 2 deferiprone treated groups, the missing data were derived from.
Selective reporting (reporting bias)Low riskAll outcomes pre-specified were reported in the manuscript.
Other biasHigh riskThe trial documented sponsorship by the manufacturer of deferiprone: CIPLA Ltd.

El-Beshlawy 2008

Methods

3-arm parallel RCT.
Number of centres: 1.
Duration of treatment: 54 weeks.
Follow up: not stated.

Number (N) of participants:
N randomised: 66 (treatment group: 21; comparator group 1: 22; comparator group 2: 23).
N analysed: variable across outcomes. Minimum and maximum numbers analysed were: treatment: 16 - 22; comparator group 1: 17 - 21; comparator group 2: 15 - 23). Trial reported details of why data from 4 participants in the treatment group, 3 in comparator group 1 and 3 in comparator group 2 were withdrawn from treatment.

ParticipantsTransfusion-dependent participants aged 4 years and older with thalassaemia major.
Age: treatment group: mean (SD) 11 (4.9) years (data only given for 20/23 participants); comparator group 1: mean (SD) 10.8 (5.1) years (data only given for 18/21); comparator group 2: 13.1 (5.9) years (data only given for 18/22).
Sex: treatment group: 44.4% male; comparator group 1: 66.67% male; comparator group 2: 55% male.
Ethnicity: not stated.
Trial undertaken: Haematology outpatients clinic, Cairo University Children Hospital, Egypt.
Interventions

Treatment arm: deferiprone: 60 - 83 mg/kg/day orally for 7 days a week (mean (SD) dose = 71.1 (7.1) mg/kg/day) with DFO: 23 - 50 mg/kg/day subcutaneously over 8 hours for 2 days a week (mean (SD) dose 42.4 (5.5) mg/kg/day).

Comparator group 1: deferiprone: 60 - 83 mg/kg/day, daily for 7 days a week (mean (SD) dose: 71.1 (7.1) mg/kg/day).

Comparator group 2: DFO: 23 - 50 mg/kg/day subcutaneously over 8 hours for 5 days a week (mean (SD) dose: 42.4 (5.5) mg/kg/day).

Outcomes
  1. Liver function

  2. Liver iron concentration

  3. Serum ferritin

  4. Urinary iron excretion

  5. Net iron balance

  6. Compliance

  7. Complications of treatment

Primary outcome: not identified.

NotesThe trial did not report details as to whether participants had received prior iron chelation therapy.
Source of funding: none stated.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe authors did not report any information about how randomisation was undertaken.
Allocation concealment (selection bias)Unclear riskThe authors did not report any information about how treatment allocation was concealed.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskThe authors did not report any information as to whether participants, personnel or outcome assessors were blinded to treatment allocation.
Incomplete outcome data (attrition bias)
All outcomes
High riskVariable numbers of participants were included in the analysis for each outcome. The number of participants withdrawn from the trial was reported and reasons detailed, but this number does not tally with the numbers included in each analysis.
Selective reporting (reporting bias)Low riskAll outcomes pre-specified were reported in the manuscript.
Other biasLow riskThe trial appears to be free of other sources of bias.

Galanello 2006

Methods

2-arm parallel RCT.
Number of centres: multicentre (3 centres).
Duration of treatment: 12 months.
Follow up: not stated.

Number (N) of participants:
N randomised: 60 (treatment group: 30; comparator group: 30).
N analysed: 59 (treatment group: 29; comparator group: 30. The reason for the withdrawal (intolerance to deferiprone) was reported by the trial.

ParticipantsParticipants were 10 years or older with a diagnosis of thalassaemia major undergoing iron chelation therapy with subcutaneous DFO with a serum ferritin value between 1000 and 4000 µg/l over the previous year.

Age: treatment group: mean (SD) 18.7 (4.8) years; comparator group: mean (SD) 19.8 (6.1) years.
Sex: treatment group: 55% male; comparator group: 40% male.
Ethnicity: not stated.
Trial undertaken: Italy and Greece.
InterventionsTreatment: deferiprone: 25 mg/kg/ body weight 3 x daily for 5 days a week with DFO: 20 - 60 mg/kg/day subcutaneously on 2 days a week. (mean (SD) dose DFO for the 29 participants who completed the trial at baseline: 36.0 (5.8) mg/kg/day and end of trial: 33.3 (6.64) mg/kg/day).

Comparator group: DFO: 20 - 60 mg/kg/day subcutaneously on 5 - 7 days a week (mean (SD) dose baseline: 34.8 (8.9) mg/kg/day and at end of trial: 37.8 (8.9) mg/kg/day).
Outcomes
  1. Serum ferritin change at 1 year

  2. Liver iron concentration (measured by SQUID) change at 1 year

  3. ALT

  4. Full blood count

  5. Zinc levels

  6. Adverse events

  7. Participant compliance

Primary outcome: not identified.

NotesThe trial inferred that participants had previously received DFO treatment but no details as to dose, schedule or duration were reported.
Source of funding: Apotex Research Inc, Toronto, Canada.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe authors did not report any information about how randomisation was undertaken.
Allocation concealment (selection bias)Unclear riskThe authors did not report any information about how treatment allocation was concealed.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskThe authors did not report any information as to whether participants, personnel or outcome assessors were blinded to treatment allocation
Incomplete outcome data (attrition bias)
All outcomes
Low riskAlthough 1 participant in the treatment group was withdrawn due to intolerance to deferiprone, this is unlikely to effect the findings of the trial.
Selective reporting (reporting bias)High riskCompliance to deferiprone was pre-specified as an outcome but was not measured or reported in the manuscript.
Other biasHigh riskThe last author of the paper is an Apotex Research Inc. employee: Apotex Research Inc. funded this trial.

Gomber 2004

Methods

3-arm parallel RCT.
Number of centres: 1.
Trial dates: not stated.
Duration of treatment: 6 months.
Follow up: none.

Number (N) of participants:

N randomised: 30 (10 per treatment arm).
N analysed: 28 (treatment group: 11; comparator group 1: 10; comparator group 2: 7).

ParticipantsChildren with thalassaemia.
Age: not stated.
Sex: not stated.
Ethnicity: assumed Indian.
Trial undertaken: Thalassaemia Day Care Centre, Department of Paediatrics, Delhi, India.
InterventionsTreatment: deferiprone: 75 mg/kg/day orally in 2 - 3 divided doses, 7 days a week.
Comparator group 1: deferiprone: 75 mg/kg/day in 2 - 3 divided doses and DFO 40 mg/kg/day as a subcutaneous infusion over 8 - 10 hours 2 days a week.
Comparator group 2: DFO by subcutaneous injection, 40 mg/kg/day over 8 - 10 hours a day, 5 days/week.
Outcomes
  1. Serum ferritin levels at baseline and end of trial (primary outcome)

  2. 24 hour urinary iron excretion

  3. Monthly blood counts

  4. Liver function tests (measured 3 monthly)

  5. Kidney function tests (measured 3 monthly)

  6. Hepatic markers (hbsag)

  7. Adverse events

NotesPrior exposure to iron chelators: not reported.
Source of funding: not stated.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe authors did not report any information about how randomisation was undertaken.
Allocation concealment (selection bias)Unclear riskThe authors did not report any information about how treatment allocation was concealed.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskThe authors did not report any information as to whether participants, personnel or outcome assessors were blinded to treatment allocation.
Incomplete outcome data (attrition bias)
All outcomes
High risk

No reasons for the missing data from 2 participants were reported.

1 participant in the DFO treatment arm opted to received deferiprone. Outcome data for this participant were analysed by received treatment.

Selective reporting (reporting bias)High riskThree pre-specified outcomes: monthly blood counts, liver function tests (measured 3 monthly) and hepatic markers were not reported on. Of these, liver function tests are critical outcomes for this participant population.
Other biasUnclear riskThere was insufficient information to assess whether an important risk of bias exists.

Ha (i) 2006

Methods

2-arm parallel RCT.
Number of centres: 5.
Trial dates: not stated.
Duration of treatment: median (range) duration: 18 months (1 to 20 months). Only patients receiving at least 6 months of treatment were included for efficacy assessment.
Follow up: none.

Number (N) of participants:

N randomised: 36 (treatment group: 20; comparator group: 16).
N analysed: 31 (treatment group: 17; comparator group: 14). The trial reported details of why 5 participants were excluded from the analysis.

ParticipantsThalassaemia major participants.
Age: range 8 - 40 years (median: 20 years).
Sex: not stated for this trial alone.
Ethnicity: Chinese.
Trial undertaken: 5 paediatric units in Hong Kong.
Interventions

Treatment group: deferiprone: 75 mg/kg/day orally in 3 divided doses, 1 hour before food, 7 days a week and DFO 30 - 60 mg/kg/day as a subcutaneous infusion over 8 hours 2 days a week.

Comparator group: DFO by subcutaneous injection, 30 - 60 mg/kg/day, 5 - 7 days/week.

Outcomes
  1. Reduction of serum ferritin (primary outcome)

  2. Liver iron concentration (measured through liver biopsies & atomic absorption spectophotometry)

  3. Liver fibrosis & grading (using the Knodell scoring system) through pre & post biopsies

  4. Adverse events: biweekly blood counts, liver & renal functioning & rheumatoid factor; monthly determinations of neutropenia & agranulocytosis, gastrointestinal and joint symptoms

  5. Participant compliance

Notes

Prior exposure to iron chelators: not reported.
Source of funding: Children's Thalassaemia Foundation, Hong Kong.
Only individuals deemed to be poorly chelated (liver iron content > 7 mg/g dry weight) were eligible for randomisation in this trial.

Trial was stopped after 18 months recruitment due to an unexpected sudden death in the deferiprone treatment arm in Ha (ii) 2006.

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskPermutted block randomisation.
Allocation concealment (selection bias)Unclear riskPermutted block randomisation.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskThe authors did not report any information as to whether participants, personnel or outcome assessors were blinded to treatment allocation.
Incomplete outcome data (attrition bias)
All outcomes
High riskOverall 5 participants were excluded from outcome analysis as the trial was stopped early due to an unexpected death in the deferiprone arm in Ha (ii) 2006. Only participants who had completed a minimum of 6 months of treatment were included in the final analysis.
Selective reporting (reporting bias)Low riskAll outcomes pre-specified were reported in the manuscript.
Other biasHigh riskTrial was stopped after 18 months recruitment due to an unexpected sudden death in the deferiprone treatment arm in Ha (ii) 2006.

Ha (ii) 2006

Methods

2-arm parallel RCT.
Number of centres: 5.
Trial dates: not stated.
Duration of treatment: median (range) duration: 18 months (1 to 20 months). Only individuals receiving at least 6 months of treatment were included for efficacy assessment.
Follow up: none.

Number (N) of participants:

N randomised: 13 (treatment group: 6; comparator group: 7).
N analysed: 13 (treatment group: 6; comparator group: 7).

ParticipantsThalassaemia major patients.
Age: median (range): 20 years (8 - 40 years).
Sex: not stated for this trial alone.
Ethnicity: 12 Chinese, 1 Pakistani.
Trial undertaken: 5 paediatric units in Hong Kong.
Interventions

Treatment: deferiprone: 75 mg/kg/day orally in 3 divided doses, 1 hour before food, 7 days a week.

Comparator group: DFO by subcutaneous injection, 30 - 60 mg/kg/day, 5 - 7 days/week.

Outcomes
  1. Reduction of serum ferritin (primary outcome)

  2. Liver iron concentration (measured through liver biopsies & atomic absorption spectophotometry)

  3. Liver fibrosis & grading (using the Knodell scoring system) through pre & post biopsies

  4. Adverse events: biweekly blood counts, liver & renal functioning & rheumatoid factor; monthly determinations of neutropenia & agranulocytosis, gastrointestinal and joint symptoms

  5. Participant compliance

Notes

Prior exposure to iron chelators: not reported.
Source of funding: Children's Thalassaemia Foundation, Hong Kong.
Only individuals deemed to be well chelated (liver iron content ≦ 7 mg/g dry weight) were eligible for randomisation in this trial.

Trial was stopped after 18 months recruitment due to an unexpected sudden death in the deferiprone treatment arm.

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskPermutted block randomisation.
Allocation concealment (selection bias)Unclear riskPermutted block randomisation.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskThe authors did not report any information as to whether participants, personnel or outcome assessors were blinded to treatment allocation.
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll randomised participants were included in the analysis for all outcomes: there were no missing outcome data.
Selective reporting (reporting bias)Low riskAll outcomes pre-specified were reported in the manuscript.
Other biasHigh riskTrial was stopped after 18 months recruitment due to an unexpected sudden death in the deferiprone treatment arm in Ha (ii) 2006.

Maggio 2002

Methods

2-arm parallel RCT.
Number of centres: 15.

Trial dates: September 1994 - October 1997.
Duration of treatment: 1 year.
Follow up: none (36 patients underwent liver biopsy before and after treatment with a mean (SD) follow up until second biopsy 30 (2.4) months and 34 (6.7) months for deferiprone and DFO respectively).

Number (N) of participants:

N randomised: 144 (treatment group: 71; comparator group: 73).
N analysed: 144 (treatment group: 71; comparator group: 73).

ParticipantsThalassaemia major participants.
Age: treatment group: mean (SD) 20 (5.3) years; comparator group: mean (SD) 21 (4.2) years.
Sex: treatment group: 52%; comparator group: 47%.
Ethnicity: not stated.
Trial undertaken: "participating centres" in Italy.
Interventions

Treatment: deferiprone: 75 mg/kg/day orally in 3 divided doses, 7 days a week.

Comparator: DFO by subcutaneous injection, 50 mg/kg for 12 hours a day, 5 days a week.

Outcomes
  1. Change in serum ferritin following 1 year of treatment (primary outcome)

  2. Liver iron concentration (g/gram of dry liver), measured by biopsy

  3. Liver fibrosis & inflammation: measured monthly (Ishak)

  4. Liver iron concentration measured monthly by NMR

  5. Heart iron content measured monthly by NMR

  6. Heart function measured monthly by ultrasound

  7. 24 hr urinary iron excretion measured monthly during treatment

  8. RBC units in year before and during trial

  9. Adverse events

NotesPrior exposure to iron chelators: DFO: 50 mg/kg, subcutaneously over 12 hours, 5-times weekly, duration not reported.
Source of funding: European Community grant and Sicilian Thalassaemic Association.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe authors report "computer generated random list with permuted blocks of 10" were used.
Allocation concealment (selection bias)Low riskThe authors report "to ensure...., treatment was assigned by telephone contact of each participating centre with a physician of the co-ordinating centre who kept the randomisation sequence but was not otherwise involved in the study".
Blinding (performance bias and detection bias)
All outcomes
Low riskOutcome assessors were blinded to treatment allocation: "under code by physicians blinded to the trial treatment"
Statistical analysis was also performed under code by a biostatistician blinded to trial treatment.
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll randomised participants were included in the analysis for all outcomes: there were no missing outcome data.
Selective reporting (reporting bias)Low riskAll outcomes pre-specified were reported in the manuscript.
Other biasLow riskThe trial appears to be free of other sources of bias.

Maggio 2009

Methods

2-arm parallel RCT.
Number of centres: 25

Trial dates: September 2000 - January 2008.
Duration of treatment: the planned duration of treatment was 5 years. However, because of beneficial effects observed after the interim analysis at January 2008 the trial was stopped before the planned 5 years of treatment were completed for all patients.
Follow up: outcome data recorded for duration of treatment.

Number (N) of participants:
N randomised: 213 (treatment group: 105; comparator group: 108).
N analysed: 58 after 5 years (treatment group: 32; comparator group: 26). The number of patients analysed decreased with the number of years of follow up studied due to early termination of the trial (for follow-up period of 1/2/3/4/5 years, treatment group: 78/65/55/38/32; comparator group: 74/53/44/33/26); 33 patients withdrew from the trial with no loss to follow up.

ParticipantsThalassaemia major patients over 13 years old with a serum ferritin concentration of 800 - 3000 ug/l.
Age: treatment group: mean (SD) 23 (8.0) years; comparator group: mean (SD) 23 (7.8) years.
Sex: treatment group: 49.1% male; comparator group: 38.9% male.
Ethnicity: not stated.
Trial undertaken: 25 centres of the Italian Society for the Study of Thalassaemia and Haemoglobinopathies (SoSTE).
InterventionsTreatment: deferiprone: 75 mg/kg/day orally in 3 divided doses for 4 days/week and DFO: 50 mg/kg subcutaneously over 8 - 12 hours for remaining 3 days/week.

Comparator: deferiprone: 75 mg/kg/day orally in 3 divided doses daily.
Outcomes
  1. Differences between multiple observations of serum ferritin concentrations (primary outcome)

  2. Survival analysis

  3. Adverse events

  4. Costs

  5. T2* (heart and liver) in a subset of patients from January 2004

NotesPrior exposure to iron chelators: deferiprone 75 mg/kg/day orally for 7 days/week or DFO 50 mg/kg/day 8 - 12 hours for 5 days/week. Duration not reported.
Source of funding: the trial was reported as "conducted without the influence of the non-commercial sponsor" with no other details of sponsorship.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe authors report "computer generated random list with permuted blocks of 10" were used.
Allocation concealment (selection bias)Low riskThe authors report "to ensure allocation concealment, treatment was assigned by telephone contact from the coordinating centre".
Blinding (performance bias and detection bias)
All outcomes
Low riskOutcome assessors were blinded to treatment allocation: "under code by physicians blinded to the trial treatment". Statistical analysis was also performed under code by a biostatistician blinded to trial treatment.
Incomplete outcome data (attrition bias)
All outcomes
High riskStatistical analysis was performed based on the "intention-to-treat" principle. However, the trial was stopped before the planned 5 years of treatment were completed for all patients due to success of treatment. A total of 33 patients (12 who received combined therapy and 21 who received deferiprone only) withdrew prior to termination of the trial.
Selective reporting (reporting bias)Low riskAll outcomes pre-specified were reported in the manuscript.
Other biasHigh riskThe planned duration of treatment was 5 years. However, because of the beneficial effects, in terms of serum ferritin levels reduction in the sequential deferiprone-DFO group, observed after the interim analysis performed at January 2008, the trial was stopped before the planned 5 years of treatment were completed for all patients.

Mourad 2003

Methods

2-arm parallel RCT.
Number of centres: 1.
Trial dates: not stated.
Duration of treatment: 1 year.
Follow up: none.

Number (N) of participants:

N randomised: 25 (treatment group: 11; comparator group: 14).
N analysed: 25 (treatment group: 11; comparator group: 14).

ParticipantsBeta thalassaemia participants, severely iron overloaded and previously poorly chelated.
Age range: 12 - 40 years.
Sex: treatment group: 64% male; comparator group: 43% male.
Ethnicity: not stated.
Trial undertaken: Chronic Care Centre, Beirut, Lebanon
InterventionsTreatment group: deferiprone: 75 mg/kg/day orally in 3 divided doses, 7 days a week, DFO by subcutaneous injection, daily dose of 2 g over 8 - 12 hours, 2 days a week.

Comparator group: DFO by subcutaneous injection, 40 - 50 mg/kg over 8 - 12 hours a day, 5 - 7 days/week.
Outcomes
  1. Mean serum iron concentration at baseline, 6 & 12 months (primary outcome)

  2. Number RBC units during the trial

  3. Iron excretion at 1 & 12 months

  4. Haemoglobin level measured weekly for 3 months then monthly for 9 months

  5. Liver function measured weekly for 3 months then monthly for 9 months

  6. Renal function measured weekly for 3 months then monthly for 9 months

  7. Side effects

  8. Participant compliance

NotesPrior exposure to iron chelators: DFO, less than 4 times a week, dose and duration not reported.
Source of funding: not stated.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe authors did not report any information about how randomisation was undertaken.
Allocation concealment (selection bias)Unclear riskThe authors did not report any information about how treatment allocation was concealed.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskThe authors did not report any information as to whether participants, personnel or outcome assessors were blinded to treatment allocation.
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll randomised participants were included in the analysis for all outcomes: there were no missing outcome data.
Selective reporting (reporting bias)High riskData for two pre-specified outcomes were not reported in the paper: iron excretion at 1 and 12 months and renal function. Both are important clinical markers of the efficacy of iron chelation therapy.
Other biasLow riskThe trial appears to be free of other sources of bias.

Olivieri 1990

Methods

Cross-over trial.
Number of centres: 1.
Trial dates: not stated.
Duration of treatment: 6 days (3 days each arm), treatment arms separated by 3 - 4 weeks.
Follow up: none.

Number (N) of participants:
N randomised: 20 - cross-over trial.
N analysed: 20 - cross-over trial.

ParticipantsTransfusion-dependent, homozygous thalassaemia participants (24), transfusion-dependent Diamond-Blackfan anaemia (2).
Age: mean (range) 22 years (8 - 49 years).
Sex: not stated.
Ethnicity: not stated.
Trial undertaken: Hospital for Sick Children, Toronto.
InterventionsTreatment group: deferiprone 50 mg/kg/day orally in 3 divided doses (at 08.00, 16.00 and 24.00 hours) for trial days 2, 3 and 4.

Comparator group: DFO by infusion 50 mg/kg, 12 hours a day, on trial days 2, 3 and 4.
Outcomes
  1. Mean daily urinary iron excretion (mg/day) during treatment

NotesPrior exposure to iron chelators: DFO ceased 72 hours preceding trial, mean (SD) total of dose taken 89.4 (31.9) g/kg. Duration not reported.
Source of funding: Medical Research Council, Canada.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe authors did not report any information about how randomisation was undertaken.
Allocation concealment (selection bias)Unclear riskThe authors did not report any information about how treatment allocation was concealed.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskThe authors did not report any information as to whether participants, personnel or outcome assessors were blinded to treatment allocation.
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll randomised participants were included in the analysis for all outcomes: there were no missing outcome data.
Selective reporting (reporting bias)Low riskAll outcomes pre-specified were reported in the manuscript.
Other biasLow riskThe trial appears to be free of other sources of bias.

Olivieri 1997

Methods

2-arm parallel RCT.
Number of centres: 2.
Trial dates: November 1993 - September 1995.
Duration of treatment: analysis undertaken after 24 months (mean (SD) duration 33 (1.0) months, range 24 - 43 months).
Follow up: none.

Number (N) of participants:

N randomised: 64 (treatment group: 32; comparator group: 32)

N analysed: 37 (treatment group: 19; comparator group: 18). The trial reports details for why 7 and 6 participants respectively were not included in the analysis. The remaining participants had not completed 24 months treatment at the time of analysis for this trial report.

ParticipantsThalassaemia participants.
Age range: not stated.
Sex: not stated.
Ethnicity: not stated.
Trial undertaken: Hospital Centres in Toronto and Montreal, Canada. These data are from the Toronto participants only.
InterventionsTreatment group: deferiprone mean dose: not stated. No further details given.

Comparator: DFO, by subcutaneous injection, mean (SD) dose: 36.7 (2.8) mg/kg/night. No further details given.
Outcomes
  1. Change in liver iron concentration (measured by SQUID or biopsy) between 12 months prior to randomisation & 24 months duration on trial treatment.

NotesPrior exposure to iron chelators: not reported.
Source of funding: not stated.
Abstract publication; additional trial data reported on the US Food and Drug Administration website.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe authors did not report any information about how randomisation was undertaken.
Allocation concealment (selection bias)Unclear riskThe authors did not report any information about how treatment allocation was concealed.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskThe authors did not report any information as to whether participants, personnel or outcome assessors were blinded to treatment allocation.
Incomplete outcome data (attrition bias)
All outcomes
High risk

The trial analysed data from 58% of randomised participants. Of the 42% randomised participants who were not available for outcome analysis:

  • 22% randomised participants had not completed the required 24 months treatment at the time of analysis for this trial report;

  • 16% deferiprone treated participants and 5% DFO treated participants were withdrawn due to treatment induced side effects.

This missing data may inappropriately effect the statistical findings of the trial.

Selective reporting (reporting bias)Low riskAll outcomes pre-specified were reported in the manuscript.
Other biasUnclear riskThe trial was reported in an abstract thus there is little data available to make an assessment of whether the trial was free of other bias.

Pennell 2006

Methods

2-arm parallel RCT.
Number of centres: 4.

Trial dates: December 2002 - March 2005.
Duration of treatment: 1 year.
Follow up: outcome data recorded for duration of treatment.

Number (N) of participants:

N randomised: 61 (treatment group: 29; comparator group: 32).
N analysed: variable across outcomes. Minimum and maximum numbers analysed were: treatment group: 27 - 29; comparator group: 30 - 32. Trial reported details of why data from 2 participants in the treatment group and 1 in the comparator group were withdrawn from treatment.

ParticipantsTransfusion-dependent homozygous patients with beta-thalassaemia major.
Age: mean (SD) treatment group: 25.1 (5.8) years; mean (SD) comparator group: 26.2 (4.7) years.
Sex: treatment group: 52% male; comparator group: 50% male.
Ethnicity: Greek/Italian: treatment group: 16/13; comparator group: 18/14.
Trial undertaken: 4 participating centres in Italy and Greece.
InterventionsTreatment group: deferiprone: initial dose 75 mg/kg/day increasing to 100 mg/kg/day. Mean actual dose: 92 mg/kg/day.

Comparator group: DFO by subcutaneous injection, 50 mg/kg for 5 or more days a week.
Outcomes
  1. Change over 1 year in myocardial T2* (primary outcome)

  2. Cardiac volumes & function

  3. Liver iron concentration

  4. Serum ferritin

  5. Absolute neutrophil count

  6. Adverse events

  7. ALT

  8. Serum zinc levels

  9. Serum creatinine levels

NotesPrior exposure to iron chelators: DFO at a mean (SD) dose of 39 (8) mg/kg/day for 5 - 7days/week.
Trial sponsor: Apotex (manufacturer of deferiprone).
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe authors did not report any information about how randomisation was undertaken.
Allocation concealment (selection bias)Unclear riskThe authors did not report any information about whether treatment allocation was concealed.
Blinding (performance bias and detection bias)
All outcomes
Low riskThe primary outcome was independently measured in a different country (UK) to where the trial took place and the findings were not communicated back to the clinicians during the course of the trial. No one else (participants, personnel, assessors of other outcomes were blinded to treatment allocation).
Incomplete outcome data (attrition bias)
All outcomes
Low risk

All participants were included in the analysis of the outcomes serum ferritin and adverse events.

Data from 1 participant in the comparator (DFO) group was not included in the analysis of the cardiac outcomes (primary outcome) and last observation carried forward method was used to accommodate the missing data from 3 other participants (2 treatment group and 1 from the comparator group) in the cardiac outcomes (primary outcome).

Two participants in each treatment group did not have a liver iron concentration assessment at 12 months and the data from these participants were missing from the analysis.

Selective reporting (reporting bias)High riskThe following pre-specified outcomes were not reported in the manuscript: absolute neutrophil count; ALT; serum zinc levels and serum creatinine levels.
Other biasHigh riskThe trial documented sponsorship by the manufacturer of deferiprone: Apotex

Tamaddoni 2010

Methods

2-arm parallel RCT.
Number of centres: 1.
Duration of treatment: 12 months.

Trial dates: non stated.
Follow up: not stated.

Number (N) of participants:
N randomised: 80 (treatment group: 40; comparator group: 40).
N analysed: 80.

Participants

Thalassaemia major patients.

Age: at least 10 years old; mean (SD) treatment group: 18.7 (4.8) years; mean (SD) comparator group: 17.8 (6.1) years.
Sex: treatment group: 55% male; comparator group: 52.5% male.
Ethnicity: not stated.
Trial undertaken: Iran.

InterventionsTreatment: deferiprone: 75 mg/kg/daily orally and DFO by subcutaneous injection, daily dose of 40 - 50 mg/kg/day, 2 days a week.

Comparator: DFO by subcutaneous injection, 40 - 50 mg/kg/day, 5 days/week.
Outcomes
  1. Haemoglobin, hematocrit, whole blood count, absolute neutrophil and platelet count measurements taken every 1 - 3 weeks

  2. Liver and renal function checked monthly

  3. Serum ferritin (controlled every three months)

  4. Adverse effects and compliance at each transfusion visit

Primary outcome: not identified.

NotesPrior exposure to iron chelators: DFO received by all patients, dose and duration not reported.
Source of funding: Babol University of Medical Sciences.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe authors did not report any information about how randomisation was undertaken.
Allocation concealment (selection bias)Unclear riskThe authors did not report any information about how treatment allocation was concealed.
Blinding (performance bias and detection bias)
All outcomes
Low riskThe authors reported by separate communication that outcome assessors but not clinicians were blinded to treatment allocation.
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll randomised participants were included in the analysis for all outcomes: there were no missing outcome data.
Selective reporting (reporting bias)High riskThe following pre-specified outcomes were not reported in the manuscript: haemoglobin, hematocrit, whole blood count, absolute neutrophil and platelet count measurements, liver and renal function and compliance.
Other biasLow riskThe trial appears to be free of other sources of bias.

Tanner 2007

  1. a

    ALT: alanine aminotransferase
    BNP: Brain natriuretic peptide
    DFO: desferrioxamine
    hbsag: hepatitis b surface antigen
    Hr: hour
    N: number
    LVEF: left ventricular ejection fraction
    NTBI: non-transferrin bound iron
    RCT: randomised controlled trial
    RBC: red blood cell
    SQUID: superconducting quantum interference device

Methods

2-arm parallel RCT.
Number of centres: multicentre (12 centres).
Duration of treatment: 12 months.
Follow up: not stated.

Number (N) of participants:
N randomised: 65 (treatment group: 32; comparator group: 33).
N analysed: not reported.
N completing treatment: 60 (treatment group: 28; comparator group: 30). The reason for the withdrawal was not fully reported by the trial.

ParticipantsParticipants aged 18 years or older with a diagnosis of beta thalassaemia, currently maintained on subcutaneous DFO and with a myocardial T2* between 8 - 20 ms.
Age: treatment group: mean (SD) 28.8 (4.2) years; comparator group: mean (SD) 28.7 (5.3) years. Age range for both arms was 18 - 42 years.
Sex: treatment group: 44% male; comparator group: 39% male.
Ethnicity: not stated.
Trial undertaken: thalassaemia out-patients clinics in Sardinia.
InterventionsTreatment group: deferiprone: 75 mg/kg daily for 7 days a week with DFO: 40 - 50mg/kg subcutaneously for 5 days a week (DFO actual dose: 34.9 mg/kg for 5 days).

Comparator: DFO: 40 - 50 mg/kg subcutaneously for 5 days a week (DFO actual dose: 43.4 mg/kg for 5 days) with an oral placebo (no further details reported).
Outcomes
  1. Change over 1 year in myocardial T2* (primary outcome)

  2. Change in liver T2* at 12 months

  3. Serum ferritin

  4. Left ventricular volume & function

  5. Brachial artery reactivity as a marker of heart failure

  6. Participant compliance with chelation treatments

  7. Adverse events

  8. BNP test

NotesPrior exposure to iron chelation: DFO mean (SD) dose 36.4 (11.1) mg/kg per day for 5.5 day/week (equivalent to 40.5 mg/kg for 5 d/week). Participants were excluded if they had previously received deferiprone.
Source of funding: CORDA, Royal Brompton & Harefield Hospitals Charitable funds, Cooley's Anemia Foundation, Apotex, UK Thalassaemia Society, University College London Special trustees Charity.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe authors did not report any information about how randomisation was undertaken.
Allocation concealment (selection bias)High riskTrial reports that the patients and clinicians were aware of how treatment was to be allocated.
Blinding (performance bias and detection bias)
All outcomes
Unclear riskThe authors did not report any information as to whether participants, personnel or outcome assessors were blinded to treatment allocation.
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskAs the trial does not report the number of participants included in each outcome assessment. The trial reports the number completing treatment and the reasons why 4 participants in the treatment group (3 adverse events & 1 participant request) and 3 (1 adverse event & 2 participant requests) in the comparator group were withdrawn from the trial.
Selective reporting (reporting bias)Low riskAll outcomes pre-specified were reported in the manuscript.
Other biasHigh riskThe trial documented sponsorship by the manufacturer of deferiprone: Apotex.

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
  1. a

    DFO: desferrioxamine
    LVEF: left ventricular ejection fraction
    NTBI: non-transferrin bound iron
    RCT: randomised controlled trial
    SCD: sickle cell disease

Agouzal 2010In the observational study reported in this abstract, patients were not randomised to their interventions (deferiprone or deferiprone with DFO), rather the patient's previous treatment was continued following enrolment into the study.
al Refaie 1992This is a study exploring the serum NTBI in people with beta-thalassaemia major. Serum samples were taken from people treated with either deferiprone of DFO and healthy volunteers. The aim of the study was not to compare clinical outcomes between the interventions. Included participants were not randomised to an intervention.
Anderson 2002This was a matched-patient study comparing myocardial iron content and cardiac function in people with thalassaemia major receiving long-term deferiprone and DFO. Individuals were not randomised between study arms, rather the DFO treated group were matched to the deferiprone treated group in a 2:1 ratio on the basis of age, sex and current ferritin concentration.
Athanassiou 2003This was a study of the immune status of thalassaemia patients receiving deferiprone alone or in combination with DFO. The study does not report data on any clinical outcome measures included in the 'Types of outcome measures' for this review and patients were not randomised to their treatment groups.
Athanassiou-Metaxa 2004This study was not a RCT, rather patients were assigned by the investigators to their treatment group (deferiprone monotherapy or deferiprone and DFO combined).
Bartfay 1999This study compared the cytotoxic aldehyde generation in people with beta-thalassaemia chelated with deferiprone or DFO compared to non-chelated controls. The aim of the study was not to compare clinical outcomes between the interventions. Included participants were not randomised to an intervention.
Cassinerio 2012A retrospective observational study of the efficacy of different iron chelation therapies on removal of myocardial iron content and LVEF assessed by cardiac magnetic resonance.
Christoforidis 2007Deferiprone and DFO were given as combined treatment in this study, thus all patients received both interventions for the duration of this study.
Cohen 1997This trial was a multicentre safety trial of deferiprone in people with transfusion-dependent thalassaemia major. There was no comparator treatment arm in this trial.
Diav-Citrin 1997This trial investigated variability in response to deferiprone in people with beta-thalassaemia major. There was no comparator treatment arm in this trial.
Drakonaki 2010Patients were not randomised to their interventions (DFO monotherapy or deferiprone with DFO), rather the patient's previous treatment was continued following enrolment into the study.
Elalfy 2006Patients were not randomised to their interventions (DFO or deferiprone with DFO), rather the patient's previous treatment was continued following enrolment into the study.
Fassos 1994Participants were not randomised to their first treatment arm in this cross-over study.
Fragatou 2007Patients were not randomised to the interventions, rather pre-existing clinical characteristics defined which interventions (deferiprone and DFO or DFO monotherapy) each patient received.
Galanello 2001This trial looks at the efficacy of iron chelation in people with thalassaemia major with poor compliance to DFO. All participants received combined therapy of deferiprone and DFO. There was no comparator treatment arm in this trial.
Goulas 2012The objective of this non-randomised study was to compare the quality of life, self-esteem, satisfaction and adherence to treatment of patients already receiving one of three treatments: deferiprone with DFO, DFO or deferasirox.
Grady 2002Although the study was of a cross-over design, patients were not randomised to their treatment group (deferiprone or DFO). The study does not include outcome measures that are relevant to this review.
Hershko 1996This abstract reports the long-term data from the International Study Group on Oral Iron Chelators. The aim of the study was to provide information on the benefits and risks associated with long-term deferiprone in people with transfusional iron overload. The study did not include a comparison group.
Kattamis 2003This was not a RCT, rather each patients received each phase of treatment (DFO monotherapy, deferiprone monotherapy or DFO and deferiprone combined) in the same consecutive pattern.
Kontoghiorghes 1987This trial was a multicentre safety trial of deferiprone in people with transfusion-dependent thalassaemia major. There was no comparator treatment arm in this trial. This trial was a small trial exploring urinary iron excretion following receipt of deferiprone alone or in combination with ascorbic acid in people with thalassaemia or myelodysplasia. Outcome data for the thalassaemia participants were not presented separately from the myelodysplasia participants.
Lai 2010This trial was not randomised; all patients were offered combined treatment of DFO and deferiprone but some either refused or experienced adverse effects due to deferiprone and were given DFO alone.
Loebstein 1997This trial is neither randomised nor includes outcome measures that are relevant to the review. This trial studied immune function in people with thalassaemia: 36 received deferiprone and 21 received DFO and were treated in parallel.
Marshall 2003This was a matched patient cross-over study investigating the chromosomal aberration frequencies in people with thalassaemia major. Individuals treated long-term with deferiprone were matched for age, sex and iron overload with those treated long-term with DFO. This was not a RCT.
Nielsen 1995This was a study of liver iron stores in patients with beta-thalassaemia major or other post-transfusional siderosis receiving treatment with deferiprone or parenteral DFO, or both. Individuals were not randomised to their respective treatment arms.
Olivieiri 1993This abstract reports data from the first 3 years of the Canadian trial which investigated the safety and effectiveness of deferiprone in people with transfusion-dependent thalassaemia and SCD. There was no comparator treatment arm in this trial.
Peng 2003This was a 3-year prospective clinical trial comparing deferiprone with DFO with regards to efficacy and side effects of these interventions on cardiac or hepatic systems, or both. Participants were not randomised to their treatment group; instead those who were unwilling or unable to use DFO were assigned to treatment with deferiprone: all other participants stayed on the DFO regimen.
Peng 2006Email communication with the authors confirmed that the two group of participants in this study were not randomised to their treatment arm (deferiprone or DFO), rather they were assigned to their treatment arm by clinicians.
Pepe 2006This was a matched patient study seeking to evaluate the efficacy of oral deferiprone in comparison to subcutaneous DFO by multislice multiecho T2*. Participants were not randomised between study arms, rather the DFO treated group were matched to the deferiprone treated group on the basis of age and sex.
Ricchi 2010In this retrospective study, patients were allocated to 1 of 4 treatment groups (differing doses or schedules of deferiprone with DFO) based on severity of iron overload and previous adverse effects to deferiprone therapy.
St Pierre 2003This is a commentary on the Anderson 2002 trial; in particular the misleading claims that the use of advanced magnetic resonance techniques has made possible the accurate assessment of both liver and cardiac iron. St Pierre discusses the appropriate use of T2* measurements and the interplay of other factors on T2* findings.
Taher 2001This trial aimed to compare the effectiveness, safety and participant compliance of deferiprone and DFO in people with thalassaemia major. Participants were not randomised to the treatment arms, rather those who were non-compliant with or unable to tolerate DFO received deferiprone: all other participants received DFO.
Tanner 2008All patients in this study received both deferiprone and DFO in a non-randomised method of treatment allocation.
Tsakok 2004This was a commentary on some of the statistical decisions made by Maggio 2002 RCT.
Vlachaki 2007Patients had already been assigned to their treatment (DFO or deferiprone) prior to the start of the study.
Wang 2006Patients were not randomised to the treatment (deferiprone or DFO) they received in this study.
Zareifar 2009This trial aimed to evaluate the efficacy and safety of combined deferiprone with DFO compared with DFO monotherapy. Participants were not randomised to the treatment arms; treatment was allocated according to prior compliance with DFO treatment.

Characteristics of studies awaiting assessment [ordered by study ID]

Alpendurada 2012

MethodsRCT.
Participants65 adult thalassaemia major patients with mild-moderate cardiac iron loading.
Interventions

Group 1 (n = 32): DFO combined with deferiprone.

Group 2 (n = 33): DFO with placebo.

OutcomesRight ventricular function using cardiac MRI.
NotesThis trial was identified during the final search period (March 2013) and has not been fully evaluated to determine eligibility for inclusion, although patients are reported to be a subset of those included in the Tanner 2007 study.

Aydinok 2012

MethodsRCT.
Participants20 thalassaemia major patients.
InterventionsPatients received one year of deferiprone monotherapy 75 mg/kg in 3 divided doses daily, either alone or in combination with DFO, 40 - 50 mg/kg sc, twice weekly on consecutive nights.
Outcomes
  1. Urinary iron excretion

  2. Transfusional iron loading rates

  3. LIC

  4. Plasma concentrations of NTBI and labile plasma iron

NotesThis appears to be a substudy of Aydinok 2007 which has been included in this review.

Badawy 2010

MethodsA retrospective-prospective randomised cohort study.
Participants150 transfusion-dependent children with thalassaemia aged 8 years and over in east delta of Egypt who are on irregular DFO chelation therapy prior to study.
Interventions

Group 1 received deferiprone for 5 days/week alternating with DFO for 2 days/week.

Group 2 received daily deferiprone only.

Group 3 received DFO only for 5 days/week. Doses were the same for all groups: deferiprone 75 mg/kg/day; DFO 40 mg/kg/day.

Outcomes
  1. Non-compliance

  2. Adverse effects

  3. Serum ferritin levels

  4. Liver and kidney functions

  5. Blood glucose level

  6. Serum calcium and phosphorus/3 months

  7. T3, T4, TSH, LH, FSH, echocardiography, bone density, auditory and visual examination

NotesWe have not identified any further publications to these conference abstracts.

Evans 2011

MethodsRCT.
Participants21 thalassaemia major patients.
Interventions

Group 1 received deferiprone monotherapy, 25 mg/kg tds (12 patients).

Group 2 received deferiprone at the same dose with DFO 40 - 50 mg/kg for 2 nights per week (12 patients).

Outcomes
  1. LIC

  2. transfusional iron loading rates

  3. Urinary iron excretion

  4. Plasma NTBI and LPI concentrations

NotesFollow up at 52 weeks.

Jain 2011

MethodsProspective comparative study.
Participants98 children with transfusion-dependent thalassaemia.
InterventionsCombined chelation therapy (DFO and oral chelation) versus oral chelation monotherapy versus DFO monotherapy.
OutcomesRed cell transfusion requirement.
NotesThis trial, presented as a conference abstract, was identified during the final search period (March 2013) and has not been fully evaluated to determine eligibility for inclusion. In particular, it is unclear whether treatment allocation was randomised.

Kompany 2009

MethodsRCT.
Participants40 thalassaemia patients.
Interventions

Intervention group: DFO sc for 3 days/week and deferiprone orally for 4 days/week.

Control group: sc DFO for 6 days/week.

OutcomesCardiac indices including: right ventricle diameter; LVEDD; LVESD; aortic root diameter, LVEF.
NotesThis Iranian trial requires translation from Persian in order to establish whether the trial is appropriately randomised; only the abstract is published in English and does not provide sufficient information to determine whether this trial is eligible for inclusion.

Maggio 2012

MethodsMulticentre RCT.
Participants99 patients with thalassaemia major.
InterventionsLong-term sequential deferiprone-deferoxamine versus deferiprone alone.
OutcomesLVEF.
NotesThis trial was identified during the final search period (March 2013) and has not been fully evaluated to determine eligibility for inclusion, although the background to this study as presented in this abstract suggests that this is a retrospective follow up of patients included in the Maggio 2009 study.

Mirbehbahani 2012

MethodsControlled clinical trial.
Participants26 patients with beta-thalassaemia major.
Interventions

Intervention group: DFO sc, every other day at 30 - 50 mg/kg/day mg/kg/day with deferiprone at 75 mg/kg/day in 3 divided oral doses.

Comparator group: DFO subcutaneously for 6 - 12 hours/day, 5 - 6 days per week at 30 - 50 mg/kg/day.

Participants were randomised to 6 months of treatment.

OutcomesSerum ferritin (primary).
NotesThis trial was identified during the final search period (March 2013) and has not been fully evaluated to determine eligibility for inclusion.

N0277104959

MethodsProspective RCT.
Participants60 young thalassaemia participants with a myocardial T2* < 10 ms: 30 participants with normal ejection fraction as measured by CMR; and 30 participants with reduced ejection fraction.
Interventions

All participants to receive "the current gold standard continuous chelation therapy via implantable device such as pic line or a port-a-cath using disposable desferrioxamine infusors" (dose and schedule not reported).

Group 1: deferiprone 75 mg/kg/body weight (schedule not reported).

Group 2: placebo.

Outcomes
  1. Myocardial Iron measured by CMR T2* relaxometry.

  2. Rest cardiac function, measured by CMR measurements in left ventricle of end diastolic volume, end systolic volume, ejection fraction and left ventricular mass.

  3. Clinical events.

  4. Compliance.

NotesWe have been unable to identify any publications from this trial, registered on the National Research Register. Moreover despite repeated emails to the study contact person (now retired) and potential colleagues, we have been unable to identify whether the trial was started and completed, if participants were randomised to treatment allocation and whether a publication has arisen from this trial.

NCT00115349a

MethodsDouble-blind (participant, caregiver and investigator), placebo-controlled RCT of parallel design.
Participants

Aged 18 years and older with transfusion dependent beta-thalassaemia, with a LVEF by MRI < 56%, serum ferritin > 1000 µg/L or ferritin between 500 and 1000 µg/L and cardiac T2* < 20 ms.

Participants to be receiving DFO sc or iv and willing to chelate 7 days per week for 12 - 24 hours per day.

Interventions

Intervention group: DFO daily for 12 - 24 hours/day, 7 days a week either sc or iv at up to 50 - 60 mg/kg/day with deferiprone at 75 mg/kg/day in 3 divided oral doses.

Comparator group: DFO daily for 12 - 24 hours/day, 7 days a week either sc or iv at up to 50 - 60 mg/kg/day.

Participants were randomised to 1 year of treatment.

Outcomes
  1. Rate of change in LVEF (MRI measured) from screening to 1 year (primary outcome)

  2. Myocardial iron burden estimated by myocardial T2* at 1 year

  3. Change in left ventricular volume from screening to 1 year

  4. Change in Echo left ventricular volume, ejection fraction, shortening fraction and VCFc/Wall stress z score from screening to 1 year

  5. Change in Holter monitor scores from screening to 1 year

  6. Initiation or increase in cardiac medication during study

  7. Adverse events during study

NotesEmail correspondence with the study contact person (Professor John Porter) in February 2009 identified that the study had been terminated due to slow recruitment but that "there were interesting findings" and the study data were being held by the New England Research Institutes. No publication has yet arisen from this study.

Pantalone 2011

MethodsMulticentre open-label RCT.
ParticipantsBeta-thalassaemia major patients.
Interventions

Group 1: deferiprone 75 mg/kg 4 days/week combined with DFO 50 mg/kg/day for 3 days/week.

Group 2: deferiprone monotherapy daily, 75 mg/kg/day.

Outcomes
  1. Serum ferritin

  2. Adverse events

  3. Costs

NotesThis trial, which included a 5-year follow up, was identified during the final search period (March 2013) and has not been fully evaluated to determine eligibility for inclusion.

Pepe 2013

MethodsA prospective study of iron chelation efficacy in a large clinical setting. It is unclear whether patients were randomised to treatment arm.
Participants139 participants enrolled in the Myocardial Iron Overload in Thalassemia (MIOT) network with thalassaemia major.
Interventions

Group 1 (n = 43): deferiprone (mean (SD) 66 (23) mg/kg/day for 6 (1) day/week) and DFO 41 (7) mg/kg/day for 4 (1) day/week).

Group 2 (n = 30): deferiprone only (mean (SD) 73 (16) mg/kg/day).

Group 3 (n = 66): DFO only (mean (SD) 41 (7) mg/kg/day for 5.5 days/week).

Outcomes
  1. Myocardial and LICs measured by T2*

  2. Biventricular functions parameters

Notes 

Unal 2009

  1. a

    CMR: cardiovascular magnetic resonance
    DFO: desferrioxamine
    FSH: follicle-stimulating hormone
    iv: intravenously
    LH: luteinizing hormone
    LIC: liver iron concentration
    LPI: labile plasma iron
    LVEDD: left ventricular end diastolic dimension
    LVEF: left ventricular ejection fraction
    LVESD: left ventricular end systolic dimension
    MRI: magnetic resonance imaging
    NTBI: non-transferrin bound iron
    sc: subcutaneously
    RCT: randomised controlled trial
    TSH: thyroid-stimulating hormone
    T3 & T4: thyroid hormones
    VCFc: velocity of circumferential fibre shortening

MethodsStudy of efficacy between 2 iron chelation treatment groups. It is unclear whether patients were randomised to treatment arm.
ParticipantsBeta-thalassaemia major patients currently maintained on sc DFO monotherapy; age 10 years and over and maintaining pre-transfusion haemoglobin > 9 g/dL.
Interventions

Group 1 (n = 26): deferiprone 20 - 25 mg/kg/dose in addition to DFO 3 - 5 times/week 30 - 40 mg/kg/day.

Group 2 (n = 13): DFO monotherapy (dose not stated).

Outcomes
  1. Myocardial T2*

  2. Ejection fraction levels

  3. Serum ferritin levels

NotesWe have not identified any further publications to this conference abstract.

Characteristics of ongoing studies [ordered by study ID]

EU_2009-014666-25

Trial name or titleSequential deferasirox-deferiprone versus deferasirox or deferiprone multicentre randomised trial.
MethodsRCT.
ParticipantsBeta-thalassaemia major patients aged between 12 - 50 years and with a serum ferritin concentration > 1000g/l. Planned number of participants: 363.
InterventionsSequential or monotherapy: deferasirox 30 mg/kg; deferiprone 75 mg/kg.
Outcomes

Primary outcome

  1. Chelating efficacy assessment of sequential therapy deferasirox-deferiprone versus deferasirox or deferiprone alone

Secondary outcomes:

  1. To assess in the sequential deferasirox-deferiprone treated group a reduction at least in 50% of cases of creatininemia significant increase (> 33%) compared to the deferasirox alone group

  2. To assess in the sequential deferasirox-deferiprone treated group a reduction at least in 70% of cases of neutropenia compared to the deferiprone alone group

  3. To assess, using MRI, in a subgroup, the possible organ-specific (heart, liver, pancreas) iron overload variation during therapy.

Starting dateJuly 2009.
Contact informationFondazione Franco e Piera Cutino, Italy.
NotesNon-commercial sponsored study.

IRCT201110087677N1

Trial name or titleThe comparative study of incidence of lens opacity between Osferal and Deferoxamine in major thalassaemia.
MethodsRCT.
ParticipantsMale and female paediatric patients aged 1 to 14 years with major thalassaemia who have never received any chelator therapy prior to study.
Interventions

Group 1: Osferal (no dose or schedule details reported).

Group 2: DFO (no dose or schedule details reported).

OutcomesLens opacity (primary)
Starting dateDecember 2010.
Contact informationVahid Falahati, Imam Reza Hospital, KUMS, Kermanshah, Iran (vahidfalahati89@yahoo.com).
Notes

Estimated enrolment: 50 (25 per treatment arm).

Estimated study completion date: December 2011.

NCT01511848

  1. a

    DFO: desferrioxamine
    MRI: magnetic resonance imaging
    RCT: randomised controlled trial
    sc: subcutaneously
    SCD: sickle cell disease

Trial name or titleStudy of efficacy, safety of combined deferasirox and deferiprone versus combined deferiprone and desferal in conditions of iron overload.
MethodsA prospective open-label phase 2/phase 3 cross-over RCT.
ParticipantsMale and female participants with transfusional iron overload secondary to thalassaemia major or SCD aged 6 - 18 years with a serum ferritin > 2500 ng/mL.
Interventions

Intervention group: DFO 40 mg/kg administered sc over 8 hours for 5 days, combined with oral deferiprone 75 mg/kg in 3 divided doses for 7 days.

Comparator group: oral deferasirox 20 mg/kg once daily for 5 days, combined with deferasirox 25 mg/kg 3 times per day; dose may be increased up to 100 mg/kg/day guided by serum ferritin levels.

Outcomes
  1. To assess efficacy of combining deferiprone and deferasirox compared to combined deferiprone and DFO in decreasing the serum ferritin level in conditions with severe chronic iron overload (primary)

  2. To determine the safety upon administering the drugs in combination (deferiprone and deferasirox) compared to the combination of DFO and deferiprone

Starting dateFebruary 2012.
Contact informationAsst Prof Adly, Pediatric Hematology Clinic, Aim Shams University, Cairo, Egypt (amiradiabetes@yahoo.com).
Notes

Estimated enrolment: 60 (30 per treatment arm).

Estimated study completion date: February 2013.

Ancillary