Guideline:Iron overload

Iron overload occurs when iron intake is increased over a sustained period of time, either as a result of red blood cell transfusions or increased absorption of iron through the gastrointestinal tract (GI). Both of these occur in thalassaemia, with blood transfusion therapy being the major cause of iron overload in thalassaemia major and increased GI iron absorption being more important in thalassaemia intermedia.

''In the absence of any mechanism of the human body to excrete excess iron, chelation therapy is essential and constitutes the second important arm, besides transfusion therapy, of the clinical management of these patients.''

The Rate of Iron Loading
Blood transfusion: Knowledge of the rate of iron loading from transfusion to as high a level of accuracy as possible will contribute significantly to the formulation of chelation therapy appropriate for each patient. Simple calculations, such as those described in the Blood Transfusion Chapter of this book, can provide the treating physician with this information.

''In case organisational or other difficulties do not allow such estimations, a rough approximation can be made based on the assumption that 200 mg of iron is contained in each donor unit.''

Thus, irrespective of whether the blood used is packed, semi-packed or diluted in additive solution, if the whole unit is given, this will approximate to 200 mg of iron intake.

According to the recommended transfusion scheme for thalassaemia major, the equivalent of 100–200 ml of pure RBC per kg per year are transfused (equivalent to 116- 232 mg of iron per kg body weight per year or 0.32-0.64 mg/kg/day). Regular blood transfusion therapy therefore increases iron stores to many times the norm unless chelation treatment is given.

'''Increased gastro-intestinal absorption of iron:''' Normal intestinal iron absorption is about 1-2 mg/day. In patients with thalassaemia who do not receive any transfusion, iron absorption increases several-fold.

''It has been estimated that iron absorption exceeds iron loss when expansion of red cell precursors in the bone marrow exceeds five times that of healthy individuals.''

Transfusion regimens aimed at keeping the pre-transfusion haemoglobin above 9 g/dl have been shown to prevent such expansion (Cazzola 1997). In individuals who are poorly transfused, absorption rises to 3-5 mg/day or more representing an additional loading of 1-2 grams of iron loading per year.

Toxicity from Iron Overload
Mechanism of iron toxicity Iron is highly reactive, easily alternating between two states – iron III and iron II – in a process which results in the gain and loss of electrons, generating harmful free radicals (atoms or molecules with unpaired electrons). These can damage lipid membranes, organelles and DNA causing cell death and the generation of fibrosis. In health, iron is ‘kept safe’ by binding to molecules such as transferrin, but in iron overload their capacity to bind iron is exceeded both within cells and in the plasma compartment. The resulting ‘free iron’ damages many tissues in the body and is fatal unless treated by iron chelation therapy.

Complications of iron overload Untreated transfusional iron overload in thalassaemia major is fatal in the second decade of life, usually as a result of cardiac complications (Zurlo 1989). Iron overload also causes pituitary damage, leading to hypogonadism and poor growth. Endocrine complications, namely diabetes, hypothyroidism and hypoparathyroidism, are also seen. Liver disease with fibrosis and eventually cirrhosis, particularly if concomitant chronic hepatitis is present, is also a serious complication. (These complications are described in greater detail in the relevant chapters of this book.)

Monitoring of Iron Overload
Monitoring closely and assessing as accurately as possible iron overload is essential in establishing effective iron chelation regimes, such as those mentioned in this chapter, tailored to the individual patient’s specific needs. However, some general principles of monitoring iron overload apply to all treatments:

Serum ferritin This is a relatively easy test to perform, well established, generally correlating with body iron stores and prognostically relevant in thalassaemia major. Up to a value of about 3,000 μg/L serum ferritin is secreted in an iron-free form from macrophages, but above this value increasing proportions of ironladen ferritin ‘leaks’ from hepatocytes (Worwood, 1980; Davis, 2004). Day-to-day variations are particularly marked: high degrees of iron loading, inflammation, hepatitis and/or liver damage may falsely increase serum ferritin, while vitamin C deficiency may depress it. A sudden and unexpected rise in ferritin level should prompt a search for hepatitis, other infections or inflammatory conditions. In thalassaemia intermedia, serum ferritin tends to underestimate the degree of iron overloading (Pootrakul 1981). Therefore although there is a broad correlation between serum ferritin level and liver iron, the prediction of iron loading from serum ferritin can be unreliable (Olivieri 1995). Importantly, however, at least five studies have shown an association between the control of serum ferritin and prognosis (Gabutti V and Piga A. 1996; Olivieri, N. et al 1994; Telfer PTl, 2000; Davis BA, et al. 2004; Borgna-Pignatti, 2004). Studies have identified a significantly lower risk of cardiac disease and death in at least two-thirds of cases where serum ferritin levels have been maintained below 2,500 μg/L (with desferrioxamine) over a period of a decade or more (Olivieri, 1994). Observations with larger patient numbers show that maintenance of an even lower serum ferritin of 1,000 μg/L may be associated with additional advantages (Borgna-Pignatti, 2004) (see Table 2, page 36).

Liver iron concentration (LIC) Liver iron concentration is now regarded as the reference standard for estimating body iron loading and has been shown accurately to predict total body iron stores (Angelucci, 2000), using the formula:

'''Total body iron stores in mg/kg = 10.6 x the LIC (in mg/g dry wt)'''

Normal LIC values are up to 1.8 mg/g dry wt, with levels of up to 7 mg/g dry wt seen in some non-thalassaemic populations withoutSeveral studies link high Liver Iron Content (LIC) (above 15-20 mg/g dry wt) to worsening prognosis (Brittenham, 1993; Telfer, 2000), liver fibrosis progression (Angelucci, 1997) or liver function abnormalities (Jensen, 2003).

In the absence of prior iron chelation therapy, the risk of myocardial iron loading increases with the number of blood units transfused (Buja and Roberts, 1971; Jensen, 2003). However, more recent studies have identified discordance between liver and cardiac iron loading in some patients receiving iron chelation: patients with increased liver iron may have normal cardiac iron stores, while patients with normal or near normal liver iron may have increased cardiac iron. While the control of total body iron over a period of years is important to prognosis, liver iron concentrations are less important than cardiac iron in determining the immediate risk of heart failure. Thus, while the long-term control of body iron is important to prognosis, the risk for specific organ damage to the liver or heart at any given time is best assessed by measuring the iron in the organ of interest.

LIC determination should be considered, by the treating physicians for those patients whose serum ferritin levels deviate from expected trends (i.e. those with suspected co-existing hepatitis, or patients on chelation regimens with variable or uncertain responses), as this may reduce the risk of giving either inadequate or excessive doses of chelation therapy. Since the relationship of serum ferritin to iron overload and iron balance has not yet been established, assessment of LIC may be particularly useful when new chelating regimes are being used.

Measurement of LIC can be done by chemical determination on a liver biopsy sample (fresh, fixed or from dewaxing of paraffinembedded material)(see Table 3 see page 37) or by noninvasive methods such as magnetic biosusceptometry (SQUID) (Brittenham, 1993) or magnetic resonance imaging (MRI)(see Table 4 see page 38). Biopsy is an invasive procedure, but in experienced hands has a low complication rate (Angelucci 1997). Inadequate sample size (<1 mg/g dry weight, 4 mg wet wt or about a 2.5 cm core length) or uneven distribution of iron, particularly in the presence of cirrhosis (Villeneuve 1996), may give misleading results.

LIC can also be measured accurately using a method known as SQUID (supercoducting quantum interference device). However, only four such machines are currently available worldwide: they are expensive to purchase and maintain, and require dedicated trained staff. Liver iron measured by SQUID has the advantage of possessing a wide linear range but each SQUID machine has to be individually calibrated.

LIC can also now be measured using MRI techniques, previously limited to a relatively narrow linear range. One recently described approach, is the R2 or Ferriscan technique which appears to have acceptable linearity and reproducibility over the range of clinical interest (St Pierre TG, et al, 2005). The technique demonstrates an average sensitivity of >85% and specificity of >92% up to an LIC of 15 mg/g dry wt, and has been registered in the EU and US. For calibration, the MRI machine must use a Phantom supplied by the company, while the data acquired is sent via internet for analysis by dedicated Ferriscan software (payment per scan analysed). A particular advantage of this technique is that it can be applied with little training, at any centre with a reasonably up-to-date MRI machine (see Table 4 see page 38).

Heart function Regular monitoring of left ventricular ejection fraction (LVEF) has allowed identification of a group of patients with poor prognosis at high risk of subsequent heart failure and death who responded well to intensification of desferrioxamine (Davis et al, 2004). Patients with a fall in ejection fraction below reference values for the method used have a 35-fold increased risk of cardiac failure and death, with a median interval to progression of 3.5 years allowing time for intensification of chelation treatment. Left ventricular function can be quantified using MRI, MUGA or echocardiography. The first two methods have advantages over echocardiography in that they are less operator-dependent and therefore more easily adapted to longitudinal monitoring.

Myocardial iron estimation (T2* or other measures) Estimation of myocardial iron using MRI is becoming increasingly available but requires expertise in its use and standardisation. The T2* value in tissues shortens as the iron concentration increases. A shortening of myocardial T2* to <20 ms (implying increased myocardial iron) is associated with an increased chance of decreased LV function (Anderson et al, 2001). For example, patients with T2* values >20 ms have a very low chance of decreased LVEF. T2* values of 10-20 ms indicate up to a 10% chance of decreased LVEF; 8-10 ms indicates an 18% chance; 6 ms indicates a 38% chance; and T2* values of just 4 ms indicate a 70% chance of decreased LVEF (Westwood, 2007). In centres where such methodology is available, the T2* value may identify patients at high risk of developing a fall in LVEF before it occurs permitting a more informed choice regarding patients whose chelation treatment should be intensified.

The ability to estimate heart iron offers an additional way to stratify risk, opening up a new diagnostic window. However, factors affecting the risk of developing heart failure from myocardial iron overload are complex, while T2* measures storage iron – not in itself directly toxic to cells. Factors that may increase the availability of labile intracellular iron to cause intracellular damage such as myocarditis, or lack of continuous exposure to intracellular chelation, may influence therisk posed by excess heart iron, and explain why only a proportion of people with short T2* values show abnormal heart function at any moment in time. Prospective data on the relationship between myocardial T2* and survival are still required. However, the relationship between short T2* values (<10ms) and the risk of heart dysfunction is clear (see Table 5).

Urinary iron estimation Measurement of the urinary iron excretion can assist in assessing the effect on iron excretion of desferrioxamine (about half of total iron excreted in urine) or deferiprone (over 80% of iron excreted in urine). However, the inherent variability in daily iron excretion necessitates repeated determinations. Faecal iron excretion contributes an additional, but variable (30- 100%) to the amount of urinary iron excretion, depending on the level of iron stores, the dose of desferrioxamine and the level of haemoglobin (Pippard 1982).

Plasma non-transferrin bound iron (NTBI) In iron overload, transferrin, the normal carrier of iron in plasma becomes saturated leaving iron unbound i.e. Non-Transferrin Bound Iron (or NTBI). NTBI is cleared by different cells than transferrin iron, and is mainly responsible for the abnormal pattern of iron distribution in transfusional iron overload. Because these forms of iron rapidly reappear once iron chelators are cleared from the blood, experts suggest that the optimal treatment is 24- hour chelation (Porter, 1996).

''NTBI consists of several chemical entities, only some of which are readily chelatable and only some of which can redox cycle. One way of measuring the NTBI fraction that is labile and can redox cycle is the labile plasma iron’ assay (LPI assay). However, although the measurement of NTBI (or LPI) has proved a useful tool for examining how chelators interact with plasma iron pools, its value as a guide to routine treatment or prognosis has yet to be demonstrated.''

Other markers of oxidative damage A wide variety of markers for oxidative damage have been investigated. Malondialdehyde (MDA) is increased in iron overload, while a wide range of antioxidants are depleted.

''There has been interest in the use of antioxidants or naturally occurring products that contain antioxidant properties, such as Curcumin. However, until controlled data are available caution is advised in the use of these, as the effects of antioxidants in the presence of iron can be unpredictable due to redox cycling of iron between the iron (II) and iron (III) states.''

Other markers of organ dysfunction. These are discussed more fully in other chapters. However, iron overloaded patients should be monitored for evidence of hypoganadotrophic hypogonadism (growth and sexual development and biochemical 39 markers of HH), diabetes mellitus (yearly GTT), hypothyroidism and hypoparathyroidism.

Treatment of Iron Overload
Goals of iron chelation therapy ''The primary goal of chelation therapy is to maintain safe levels of body iron at all times. Unfortunately, once iron overload has accumulated, removal of storage iron is slow and inefficient, because only a small proportion of body iron is available for chelation at any given time.''

Consequently, when an iron chelator is given, only a small proportion of the drug binds iron, before it is excreted or metabolised. Once a patient is iron overloaded, it may take months or years to reduce body storage iron to safe levels, even with the most intensive treatment. Chelation must therefore begin soon after (2-3 years) the initiation of transfusion therapy.

Iron appears to be removed more quickly from some tissues, such as the liver, than others – for example, the heart.

Increasing the dose of chelators given in an attempt to speed up iron removal runs the risk of increasing the toxicity of an iron chelator, by chelating the iron needed for normal tissue metabolism. The twin goals of iron chelation in iron overloaded patients is therefore to decrease tissue iron to safe levels, while simultaneously making the iron as safe as possible by binding the toxic iron pools responsible for causing tissue damage. Iron is constantly being turned over, either as a result of the breakdown of red cells in macrophages or the breakdown of ferritin within cells. These same fractions are redoxactive and potentially harmful; the plasma component of this iron (NTBI) is mainly responsible for the iron loading of tissues. As mentioned above, NTBI appears within minutes of a chelator being cleared from the body. Thus, in order to achieve the second goal of chelation – the minimisation of toxic (labile) iron pools – 24-hour chelation coverage is the ideal, especially in heavily iron loaded patients. Once low levels of iron have been achieved, it is theoretically more appropriate to reduce the dose of chelator than to interrupt or decrease the frequency of chelation.

Desferrioxamine (Desferal® or deferoxamine)
Desferrioxamine has been in clinical use since the 1970s and widely used as subcutaneous infusions since about 1980. Provided that treatment is 1) begun within 2-3 years of beginning transfusion therapy, 2) administered regularly and 3) administered in adequate doses, desferrioxamine has a wellestablished impact on survival and on cardiac and other complications of iron overload described above (Brittenham, 1993; Gabutti and Piga, 1996; Borgna-Pignatti, 2004).

''The main disadvantages of the treatment are that it is costly and that it must be administered parenterally.''

Mechanism of action and pharmacology

Due to its molecular size, desferrioxamine is poorly absorbed from the gut. The higher the dose, the higher the proportion of iron excreted in the faeces rather than the urine. Iron excreted in the urine is derived from the breakdown of red cells in macrophages, whereas faecal iron is derived from iron chelated within the liver (Hershko, 1979; Pippard, 1982). Desferrioxamine has a short plasma half-life (initial half-life 0.3h), being eliminated rapidly in urine and bile. The process of iron chelation ceases soon after an infusion of desferrioxamine is complete. The efficiency of desferrioxamine (measured in terms of percent of dose excreted in the iron bound form) administered at standard 8- 12 hour intervals 5-7 days a week is approximately 14%. Iron excretion with desferrioxamine increases with dose, with body iron stores and in vitamin C deficient patients with the addition of vitamin C.

Evidence for the effectiveness of desferrioxamine Effects on serum ferritin Clinical experience over a period of three decades indicates that ferritin can be controlled with desferrioxamine monotherapy, and that maintaining serum ferritin <2500 Ìg/L with this drug is closely linked to protection from heart disease and to improved survival (Olivieri, 1994).

''However, the results of a formal prospective study on the dose required to stabilise or decrease serum ferritin in large populations have only recently become available.''

The study – a prospective evaluation of changes in ferritin levels and LIC as a function of dose in 290 thalassaemia major patients (Cappellini, 2006) – demonstrated that a mean daily dose of 42 mg/kg resulted in a small decrease in serum ferritin of 364 Ìg/L at one year, whereas a mean daily dose of 51 mg/kg resulted in an average ferritin decrease of approximately 1,000 Ìg/L over one year. Therefore, if the serum ferritin is >2,500 Ìg/l, a mean daily dose of at least 50 mg/kg/day is recommended (except in children – see below).

Effects on liver iron Administered at least 5 times a week and in sufficient doses, desferrioxamine is effective in controlling liver iron and hence total body iron stores (Brittenham, 1993). The relationship between dose and change in LIC was not examined systematically until establishing that a mean dose of 37 mg/kg stabilised LIC for patients with baseline LIC values of between 3 and 7 mg/g dry wt. For patients with LIC values between 7 and 14 mg/g dry wt, a mean dose of 42 mg/kg resulted in a small decrease of 1.9 mg/kg dry wt. In patients with LIC values >14 mg/g dry wt, a mean dose of 51 mg/kg resulted in LIC decreases of an average of 6.4 mg/g dry wt.

Thus a dose of 50 mg/kg at least 5 days a week is recommended if a significant decrease in LIC optimal levels (see above) is required. It should be emphasised that these are average changes and that the dose required may increase or decrease depending on transfusion requirement (Cohen, 2005).

Effects on heart function Subcutaneous therapy has long been known to improve asymptomatic cardiac disease (Freeman, 1983; Wolfe, 1985; Aldouri et al, 1990). Since the introduction of desferrioxamine, the incidence of ironinduced heart disease has fallen progressively in cohorts of patients – with a key factor being the age of starting treatment (Brittenham, 1994; Borgna-Pignatti, 2004). Symptomatic heart disease can be reversed by high dose intravenous treatment (Marcus, 1984; Cohen, 1989). The same results can be obtained with excellent long-term prognosis with lower doses (50-60 mg/kg/day – see below) and consequently less drug toxicity using continuous dosing (Davis, 2000 and 2004). Continuous intravenous doses of 50- 60 mg/kg/day typically normalised LVEF in a period of three months (Anderson LJ, et al., 2004), significantly before liver or heart iron stores had been normalised. However, if advanced heart failure has developed before treatment is intensified, the chances of successful rescue are decreased. Early intervention therefore for decreased LV function is therefore recommended. Once heart function has been improved, sustained compliance is critical to outcome especially while increase myocardial iron remains (Davis, 2004).

Effects on heart iron (T2*) ''Treatment with continuous intravenous desferrioxamine has been shown to improve myocardial iron, even in the most overloaded hearts,'' with average myocardial T2* values of <6 ms (Anderson, 2004). The average rate of improvement at this level of iron loading of the heart is about 3 ms/year in severely overloaded hearts: if improvement is linear it would take several years to normalise the T2* to >20 ms (Porter 2002).

In patients with baseline T2* values of between 8-20 ms, subcutaneous treatment at relatively low doses of 35 mg/kg showed an average improvement in T2* of 1.8 ms over one year (Pennell 2006). At a slightly higher dose of 40-50 mg/kg, five days a week, patients showed an improvement of 3 ms over one year (Porter et al, 2005). Improvement in cardiac T2*, even at low, intermittent doses, has been confirmed by two prospective randomised studies (Pennell, 2006; Tanner, 2006).

Effects on morbidity Regular subcutaneous therapy started before the age of 10 years reduces the incidence of hypogonadism (Bronspiegel-Weintrob, 1990), as well as other endocrine disturbances, including diabetes mellitus (Brittenham, 1993; Olivieri, 1994; Borga-Pignatti, 2004) recently (Cappellini, 2006), in a study

'''Effects on survival and complications of iron overload''' As mentioned previously, desferrioxamine was first used to treat iron overload in thalassaemia in the 1970s but was only widely used by infusion after 1980. The benefits of its regular use are clearly demonstrated by the improving survival in patients born between the 1960s and the present day (see Figure 3). Note that only patients born after 1980 will have started treatment at an early age, and that age of starting treatment is a key factor in outcome (Borgna-Pignatti, 2004; Brittenham, 1993; Davis, 2004).

''Desferrioxamine needs to be taken at least five times a week in order to optimise survival (Gabutti and Piga, 1996). Fatal complications from iron overload are also decreased if body iron (as measured by liver iron) is kept below certain levels (Brittenham, 1993) (see below).''

Unwanted effects of desferrioxamine
Local skin reactions, such as itching, erythema, induration and mild to moderate discomfort are common and may be due to inadequate dilution of desferrioxamine. Ulceration at the site of a recent infusion results from an intradermal infusion of desferrioxamine and should be addressed by deeper placement of the needle in subsequent infusions.

Infection with Yersinia enterocolitica is an important risk associated with desferrioxamine treatment (described in detail in the Chapter 9: Infections in Thalassaemia Major). Such an infection may be difficult to diagnose. However, where there is reasonable clinical suspicion of infection, with Yersinia enterolitica treatment with desferrioxamine should be temporarily discontinued. Infection should be considered in any patient with a febrile illness, especially when associated with abdominal pain, diarrhoea or joint pains, and should be treated as a medical emergency.

Desferrioxamine can usually be reintroduced once symptoms have subsided and a full course of antibiotics completed. Other infections such as Klebsiella may also be exacerbated by continued treatment with desferrioxamine.

It is therefore recommended to cease administration of desferrioxamine in anyone with an unexplained fever, until the cause has been identified and effective antibiotic treatment begun. The decision as to when to recommence treatment with desferrioxamine requires clinical judgement and a careful balancing of the potential risks and benefits. For example, a patient with high cardiac iron or poor heart function may be at high risk if desferrioxamine is withheld during a septic episode, outweighing the risks of infection once antibiotics have been commenced.

Severe allergy to desferrioxamine is a rare event and can be treated by careful desensitisation, carried out under close medical supervision (Bosquet, 1983; Miller, 1981). Desensitisation is usually successful but may need to be repeated. If unsuccessful, an alternative chelator, such as Deferiprone or Deferasirox (see below), may be considered.

Dose-related complications
Administration of excessive dosage of desferrioxamine may cause the following complications in patients who are not heavily iron loaded: may occur when desferrioxamine is given in high doses, particularly to young children whose iron burden is low (Olivieri, 1986), and when the therapeutic index is exceeded (>0.025) (Porter, 1989). Minor sensory neural deficit has been reversible in some cases, but significant hearing loss is usually permanent. It is therefore advisable to monitor audiometry yearly, bearing in mind that audiometric changes due to excessive desferrioxamine are usually symmetrical; asymmetry suggests other pathology. mg/kg/day) were given (Davies, 1983). Symptoms may include night-blindness, impaired colour vision, impaired visual fields and reduced visual acuity. Severe cases may show signs of retinitis pigmentosa on fundoscopy, whereas milder cases are only demonstrable with electroretinography. The main risk factor appears to be high dose (Olivieri, 1986) but complications are also more likely in patients who have diabetes (Arden, 1984) or those receiving concomitant phenothiazine treatment (Blake, 1985). Treatment with desferrioxamine should be temporarily suspended in patients who develop complications, to be reintroduced at lower doses once investigations indicate resolution of the problem.
 * Hearing problems: High frequency sensory neural loss, tinnitus and deafness
 * Effects on the eye: These were first noted when very high doses (>100

desferrioxamine is administered at too high a dose. Another risk factor is a young age of starting treatment (<3yrs) (De Virgillis, 1988; Piga, 1988). Growth velocity resumes rapidly when the dose is reduced to <40 mg/kg day, while it does not respond to hormonal treatment. It is therefore recommended that doses do not exceed 40 mg/kg until growth has ceased. Regular monitoring of growth is essential in all children (see Chapter 4: Endocrine Complications).
 * Growth retardation: This may occur if

common in cases of excessive dosage of desferrioxamine where patients have a low level of iron loading (De Virgillis, 1988; Olivieri, 1992; Gabutti, 1996). Rickets-like bony lesions and genu valgum may be seen in association with metaphyseal changes, particularly in the vertebrae, giving a disproportionately short trunk. Radiographic features include vertebral demineralisation and flatness of vertebral bodies. Patients should be regularly observed for such changes, as they are irreversible.
 * Skeletal changes: These are more

*Rare complications: Renal impairment and interstitial pneumonitis have been reported at very high doses of 10 mg/kg/h or more. In patients without iron overload, desferrioxamine has induced reversible coma when used with a phenothiazine derivative (Blake, 1985). Rapid intravenous injection, as may occur during flushing of a line containing desferrioxamine, must be avoided.

Recommended standard therapy
Standard dose and frequency ''The standard recommended method is slow subcutaneous infusion over 8-12 hours of a 10% desferrioxamine solution, using an infusion pump.'' In general, average doses should not exceed 40 mg/kg until growth has ceased. The standard dose is 20-40 mg/kg for children, and up to 50-60 mg/kg for adults, as an 8- 12-hour subcutaneous infusion for a minimum of 6 nights a week. To achieve negative iron balance in patients with average transfusion requirements, a dose of 50 mg/kg/day at least 5 days a week is required (Capellini, 2005). It is important that patients with high degrees of iron loading or at increased risk of cardiac complications receive adequate doses.

Use of desferrioxamine by subcutaneous bolus If an infusion pump is not available or if 10- hour infusions are not tolerated, bolus subcutaneous treatment may be considered if the patient is not at high risk of heart disease. A randomised study has shown that serum ferritin and liver iron can be controlled equally effectively by giving an equivalent total dose (45 mg/kg x 5 per week) either as two subcutaneous ‘boluses’ or as a nightly 10-hour subcutaneous infusion (Yarali, 2006).

Dose adjustment At low ferritin levels, the dose of desferrioxamine may need to be reduced and desferrioxamine-related toxicities monitored particularly carefully. Dose reductions can be made using the therapeutic index (see Figure 4) (Porter, 1989):

Figure 4: Therapeutic Index

Therapeutic index = mean daily dose (mg/kg)* / ferritin (Ìg/l) The aim is to keep the index < 0.025 at all times infusion x doses per day/7)
 * mean daily dose = (actual dose received on each

Although a valuable tool in protecting the patient from excess chelator, this index is not a substitute for careful clinical monitoring. Liver iron concentration (by biopsy, SQUID or MRI) has recently been advocated as a more reliable alternative to serum ferritin (see below). To avoid wasting a costly drug such as desferrioxamine, the dose can be adjusted to the nearest whole vial (500 mg or 2 g), alternating dose volumes between the higher and lower number of vials to achieve the desired mean daily dose.

When to start desferrioxamine therapy In thalassaemia major, this should start as soon as transfusions have deposited enough iron to cause tissue damage. This has not been formally determined, but current practice is to start after the first 10-20 transfusions or when the ferritin level rises above 1,000Ìg/l. If chelation therapy begins before 3 years of age, particularly careful monitoring of growth and bone development is advised, along with reduced desferrioxamine dosage. In thalassaemia intermedia, the rate of iron loading is highly variable and the relationship between serum ferritin and body iron can be different from that seen in thalassaemia major. If possible, an estimation of liver iron is advisable before starting treatment to see whether iron has exceeded safe levels (see Figure 4).

Use of vitamin C: Vitamin C increases iron excretion by increasing the availability of chelatable iron, but if given in excessive doses may increase the toxicity of iron. It is recommended not to give more than 2-3 mg/kg/day as supplements, taken at the time of the desferrioxamine infusion so that liberated iron is rapidly chelated. Where a patient has just started on desferrioxamine and it has been decided to administer vitamin C, the vitamin supplement should not be given until after several weeks’ treatment.

Desferrioxamine use during pregnancy: This is discussed in detail in the relevant Chapter 5: Management of Fertility and Pregnancy in ‚-thalassaemia), but Desferrioxamine is not generally recommended unless the risk of cardiac disease in the mother without chelation treatment is high.

Practical issues with subcutaneous infusion
Because regular use of desferrioxamine is critical to a good outcome, every effort should be made with each individual to help him or her to find the most convenient way to infuse the drug.

Strength of infusion The manufacturers of desferrioxamine recommend that each 500 mg vial of the drug is dissolved in at least 5 ml of water, giving a 10% solution. Concentrations in excess of this may increase the risk of local reactions at the site of infusion.

Site of infusion Care must be taken to avoid inserting needles near important vessels, nerves or organs. The abdomen is generally the best place. However because of local reactions such as erythema, swelling and induration, it is often necessary to ‘rotate’ the sites used for injection (see Figure 5). Some patients find that the skin over the deltoid or the lateral aspect of the thigh provides useful additional or alternative sites.

Type of needle The best needle to use will depend on the individual. Many patients are happy with butterfly needles of 25 gauge or smaller, which are inserted at an angle of about 45 degrees to the skin surface. The needle tip should move freely when the needle is waggled. Other patients prefer needles that are inserted vertically through the skin and are fixed with an adhesive tape attached to the needle (see Figure 6). Patient preference is highly variable and clinicians should explore the best type of needle for each patient in order to maximise compliance.

Type of infuser There are many types of infusers now available. Newer devices, including balloon pumps, are smaller, lighter, and quieter than their predecessors. For patients who find dissolving, mixing and drawing up desferrioxamine a problem, pre-filled syringes or balloons may be useful. Some pumps are designed to monitor compliance.

Local reactions Persistent local reactions may be reduced by varying the injection sites, lowering the strength of infusion or, in severe cases, by adding 5-10 mg of hydrocortisone to the infusion mixture.

Supporting compliance ''It is clear that compliance with therapy determines prognosis.'' However, desferrioxamine treatment is troublesome and time-consuming, and can be painful. Practical approaches to maximising compliance by decreasing local reactions and providing the most convenient pump system have been discussed above. Especially important, however, is support from family and the health care team. Compliance requires a sustained and secure relationship between doctor, patient and parents, and regular discussion and support are keys to maximising compliance. The reasons for poor compliance are varied. In some cases, parents cannot sanction the daily “ordeal” of chelation therapy for their child. In others, compliance may only become a problem when a child reaches adolescence. Sometimes a previously good complier may become less compliant when other life events or stresses become a burden (see Chapter 15: Psychosocial Support). Helping a patient to take control or ‘self-manage’ is often a useful approach of long-term benefit (see TIF’s book on ‘Compliance to Iron Chelation Therapy with Desferrioxamine’).

Monitoring compliance There is no perfect way to measure compliance. One successful approach may be to give patients a calendar, in which each infusion of desferrioxamine is noted down during the treatment. Some pumps can log usage. Another approach has been to keep a record of empty vials returned to the provider of the desferrioxamine.

Rescue therapy with continuous infusions In high risk cases, continuous infusion of desferrioxamine is potentially more beneficial than periodic infusions because it reduces the exposure to toxic free iron (NTBI), which returns to pre-treatment levels within minutes of stopping a continuous intravenous infusion (Porter, 1996). The route of administration is not critical, provided that as close to 24-hour exposure to chelation is achieved. Intensification of treatment through continuous, 24-hour intravenous administration of desferrioxamine via an implanted intravenous delivery system (e.g. Port-a-cath) (Davis, 2000) or subcutaneously (Davis, 2004) has been shown to normalise heart function, reverse heart failure, improve myocardial T2* (Anderson, 2002) and lead to long-term survival, provided treatment is maintained. In non-high risk cases, options such as encouraging the patient to improve compliance or an increase in dose should be explored before moving to 24-hour treatment.

Consideration for intensive therapy Intensification should be considered in the following cases: - persistently very high ferritin values* - liver iron > 15 mg/g dry weight * - significant cardiac dysrhythmias - evidence of failing left ventricular function - evidence of very severe heart iron loading (T2*<6 ms) iron loading may be desirable subcutaneous infusions.
 * severe iron overload
 * significant cardiac disease^:
 * prior to pregnancy or bone marrow transplantation, when rapid reversal of
 * If the only abnormalities are high ferritin or LIC, it would be usual to try to increase the dosing (for example, to 50- 60 mg/kg) or the duration or the frequency of

When cardiac disease is present, 24-hour intensive therapy(or combination therapy with desferrioxamine and deferiprone-see below) is necessary and simple increments in conventional 8-12-hour dosing are not recommended.

Suggested dosing A dose of at least 50 mg/kg/day and not exceeding 60 mg/kg/day is recommended as a 24-hour infusion (Davis, 2000 and 2004). Higher doses have been used by some clinicians however DFO is not licensed at these doses and the risk of retinopathy increases. Addition of vitamin C is recommended only when acute heart dysfunction has settled, which usually occurs by three months of continuous treatment (Anderson, 2004). As ferritin falls, the dose but preferably not the duration of treatment can be reduced, in line with the therapeutic index (see above).

Management of in-dwelling intravenous lines Infection and thrombosis of the catheter may occur. Careful aseptic procedures must be followed in order to prevent possible infection by Staphylococcus epidermidis and aureus, which when established are difficult to eradicate and often removal of the infusion system becomes necessary. The risk of thrombosis and infection is likely to be greater in centres that do not have regular experience in the use of long-term indwelling lines. Use of prophylactic anticoagulation is advised as line-thrombosis is relatively common in thalassaemia major (Davis, 2000). As development of a thrombosis can occur at the tip of the catheter, it is advisable, if possible to avoid placing the tip in the right atrium.

Intravenous desferrioxamine with blood transfusion This has been used as a supplement to conventional therapy (e.g. 1 g over 4 hours piggy-backed into the infusion line), but its contribution to iron balance is very limited. Special attention must be given to avoiding accidental boluses due to desferrioxamine collecting in the dead space of the infusion line. Co-administration of desferrioxamine and blood can lead to errors in interpreting side effects such as acute fever, rashes, anaphylaxis and hypotension during blood transfusion. Desferrioxamine should never be added directly into the blood unit.

Deferiprone (Ferriprox®, Kelfer®,L1)
Deferiprone is an orally absorbed iron chelator that began clinical trials in the UK in the 1980. It was first licensed for use in thalassaemia in India, followed by the European Union and other countries outside the US and Canada, in the late 1990s.

Pharmacology Three molecules of deferiprone are required to bind one iron atom, and the efficiency of iron binding decreases with falling concentrations of iron or of chelator. The drug is rapidly metabolised and inactivated in the liver by glucuronidation of one of its iron binding sites (Kontoghiorghes, 1998). At currently used doses, about 6% of the drug binds iron before it is excreted or metabolised (6% efficiency) (Aydinok, 2005). Unlike desferrioxamine, iron excretion is almost exclusively in the urine.

Evidence of effectiveness of deferiprone There are considerable accumulated publications about the effects of deferiprone. Most of these have not been randomised controlled trials, making comparison with desferrioxamine difficult.

Effects on serum ferritin Four prospective randomised trials compare the effects of deferiprone on serum ferritin at baseline and at follow-up (Maggio, 2002; Gomber, 2004; Pennell, 2006; Ha, 2006). Pooled analysis shows a statistically significant decrease in serum ferritin at six months in favour of desferrioxamine (Gomber, 2004; Ha, 2006), with no difference between the two drugs at 12 months (Maggio, 2002; Pennell, 2006). There are numerous nonrandomised cohort studies demonstrating a lowering of serum ferritin at doses of 75 mg/kg/day administered in three doses. The effect on serum ferritin at this dose appears greater at higher baseline ferritin values. In these studies significant decreases in serum ferritin are seen in patients with baseline values above 2,500 Ìg/L (Al-Refaie et al., 1992; Agarwal, 1992; Oliveiri, 1995) but not with values below 2,500 Ìg/L (Olivieri,1995; Hoffbrand,1998; Cohen, 2000).

Effects on liver iron Four trials that measure change in liver iron concentration (LIC) from baseline after a period of treatment with deferiprone compared with desferrioxamine are available (Olivieri, 1997a; Maggio, 2002; Pennell, 2006; Ha, 2006). One study showed increases in LIC at 33 months of 5 mg/g dry wt with deferiprone (n=18) and 1 mg/g dry wt with desferrioxamine (n=18) (Olivieri, 1997). A second study showed an average decrease in LIC at 30 months with both deferiprone (n=21) and desferrioxamine (n=15) (Maggio, 2002). A third study found decreases in LIC at one year of 0.93 mg/g dry wt with deferiprone (n=27) and 1.54 mg/g dry wt with desferrioxamine (n=30) (Pennell, 2006).

Another study reported decreases in LIC at six months with both deferiprone (6.6 mg/g dry wt, n=6) and desferrioxamine (2.9mg/g dry wt, n= 7) (Ha, 2006). In a non-randomised prospective study, using Deferiprone, LIC increased from baseline by 28% at two years and by 68% at three years of treatment (Fischer, 2003). In other studies where only single biopsies were performed after several years of Deferiprone treatment, LIC has been found to be above 15 mg/g dry wt in variable proportions of patients: 11% (Del Vecchio, 2002), 18% (Tondhury, 1998) and 58% (Hoffbrand et al, 1998).

Effects on heart function One prospective one-year study found that in patients with normal left ventricular ejection fraction, deferiprone given at high doses (92 mg/kg) improved heart function (Pennell, 2006). In another randomised study over one year, no difference in LVEF or other measures of LV function was seen with either deferiprone at 75 mg/kg/day or desferrioxamine (Maggio, 2002). A prospective study of the effects of deferiprone monotherapy on patients with abnormal LVEF or symptomatic heart disease has not been reported.

Effects on heart iron The effect of deferiprone monotherapy on heart iron has been reported in two prospective studies. One study found significant improvement in T2* after one year at 92 mg/kg of deferiprone daily. Patients with starting T2* values of between 8 and 20 ms showed an average increase from 13 ms to 16.5 ms in the deferiprone group, and 13.3 to 14.4 ms in the desferrioxamine group (Pennell, 2006). In another randomised study, of deferiprone and desferrioxamine administered at standard doses over one year, no change in heart iron estimated by T2 was reported for either drug (Maggio, 2002). A recent study using a new technique multislice, multiecho T2* demonstrated improved T2* values in the Deferiprone group compared to the Desferrioxamine group (Pepe, 2006).

Effects on survival and complications of cardiac disease In six randomised prospective comparisons with desferrioxamine, mortality was not reported as an outcome measure while in a seventh, one death reported in the Deferiprone arm, but not in the Desferrioxamine arm was considered as due to cardiac complications (Ha, 2006). In a retrospective cohort analysis of patients treated with deferiprone or desferrioxamine, no deaths were reported (n=157) in the Deferiprone arm in contrast to the desferrioxamine-treated patients (Borgna- Pignatti, 2006a), although some caution was expressed by the authors with regard to the interpretation of these results. In this analysis the author noted that there were no cardiac events in 750 patient years of exposure to Deferiprone in more than 150 patients.

Compliance with deferiprone One study comparing compliance with deferiprone and desferrioxamine found rates of 95% and 72% respectively (Olivieri, 1990), while another found 94% and 93% respectively (Pennell, 2006).

Two important points to be taken into consideration are (i) compliance with any treatment tends to be higher in studies than in routine use, and (ii) although compliance with oral treatment is expected to be better, it cannot be taken for granted requiring, as in the use of Desferrioxamine, constant

Unwanted effects with deferiprone
Neutropenia, agranulocytosis and thrombocytopenia The most serious and potentially fatal adverse effect of deferiprone is agranulocytosis (absolute neutrophil count, or ANC*, <500/mm3). The condition may occur with thrombocytopenia, but also isolated thrombocytopenia has occasionally been reported. Onset of agranulocytosis is variable, from a few months to nine years. In a prospective trial where weekly neutrophil counts were undertaken and where deferiprone was discontinued when the ANC was <1,500/mm3, agranulocytosis developed in 0.2/100 patient years and milder forms of neutropenia (ANC 500-1500/mm3) occurred in about 2.8/100 patient years (Cohen, 2000 and 2003). Recently, 46 cases of agranulocytosis, were reported, in Europe with nine related deaths (Swedish Orphan, safety alert, 2006). Five of these cases were in patients who had been prescribed the drug for an unspecified ‘off label’ indication, and several were not receiving weekly blood count monitoring. Swedish Orphan has subsequently issued the following recommendations on the use of deferiprone:

''ANC* should be monitored every week or more frequently if there are signs of infection; concomitant treatment that could affect the white cell count should be avoided; if severe neutropenia or agranulocytosis develop, the drug should be stopped and not reintroduced, and the use of GM CSF should be considered in the case of agranulocytosis; offlabel use of the drug should be avoided.''

Gastrointestinal symptoms Nausea and change in appetite (loss or gain) occur in 3-24% of patients (Ceci 2002; Cohen et al, 2000).

Effects on liver Variable fluctuation in liver enzymes has been reported. About a quarter of patients show ALT fluctuation of twice the normal upper limit (Cohen, 2000). One prospective randomised study showed no significant endof- study changes in liver enzymes with deferiprone or desferrioxamine (Pennell, 2006). An observational report of fibrosis after treatment for three or more years (Olivieri, 1998) has not been supported by other reports (Tondury, 1998; Hoffbrand, 1998; Wanless, 2002). A relevant prospective randomised study investigating the progression to fibrosis, using Deferiprone for one year, showed no difference as compared with Desferrioxamine, over the same period and no difference in baseline and end-oftreatment liver function tests (Maggio, 2002).

Arthropathy The frequency of arthropathy varies greatly between studies, from as low as 4.5% at one year (Cohen, 2000) to 15% after four years (Cohen, 2003) in a predominantly European patient group, and as high as 33-40% in a study of patients in India (Agarwal et al, 1992; Choudhry et al, 2004). It is not yet clear whether these differences reflect environmental or genetic differences, or differences in iron overload between populations at the start of treatment. supervision and patient support.

Symptoms range from mild non-progressive arthropathy, typically in the knees, controllable with non-steroidal antiinflammatory drugs to (more rarely) severe erosive arthropathy that may progress even after treatment is stopped. Cases involving other joints, such as wrists, ankles and elbows, and avascular necrosis of the hips, have also been described.

''Treatment should be stopped where joint symptoms continue despite a reduction in deferiprone dose and are not controlled by non-steroidal antiinflammatory drugs.''

Neurological effects Neurological complications are very rare and have been typically associated with unintentional overdosing. Rare neurological effects have included cognitive effects, nystagmus, walking disorders, ataxia, dystonia and impaired psychomotor skills. These effects appear to improve on cessation of treatment.

Effects on eye and ear There have been isolated reports of loss of vision (central scotoma). One study reported continued audiometric deterioration after switching from desferrioxamine to deferiprone (Chiodo, 1997). It is therefore advisable to monitor for CNS, audiometric and visual function in patients on regimes containing deferiprone.

Other effects Zinc deficiency during deferiprone therapy has also been observed in some patients, especially those with diabetes.

'''As a result of the various unwanted effects, 20-30% of patients are unable to sustain long-term treatment with deferiprone (Hoffbrand, 1998).'''

'''Frequency of adverse events compared with desferrioxamine''' Adverse effects have been reported in four randomised studies comparing deferiprone with desferrioxamine. One trial has reported data that allows comparison of the probability of an adverse event with deferiprone and desferrioxamine (Maggio, 2002), establishing a statistically significant two-fold difference between deferiprone (34%) and desferrioxamine (15%), but no difference between temporary or permanent treatment withdrawal.

Pregnancy Deferiprone is teratogenic in animals and must never be given to patients attempting to conceive. Until more is known, potentially fertile sexually active women and men taking deferiprone must use contraception. Deferiprone should not be used in pregnant women.

Recommended treatment regimens with deferiprone According to the official European licensing Agency (EMEA*), Deferiprone could be used as a second line drug, for removing iron in patients who are unable to use Desferrioxamine or in whom DFO therapy has proven ineffective.

Standard dosing and frequency The daily dose of deferiprone that has been evaluated most thoroughly is 75 mg/kg/day, given in three doses. In the EU, the drug is 53 (*) EMEA: European Agency for the Evaluation of Medicinal Products

licensed for doses up to 100mg/kg/day but formal safety studies of this dose are limited. The standard dose of 75mg/kg/day administered in three separate doses is therefore recommended.

Dose escalation with deferiprone. Doses of 100mg/kg/day have been given in at least one prospective study (Pennell, 2006), with no increase in side-effects reported. High dose monotherapy with deferiprone has not yet been prospectively evaluated for safety and effectiveness for patients with abnormal heart function, and combination therapy with deferiprone and desferrioxamine (see below) or intensive therapy with desferrioxamine as a 24-hour infusion should be recommended for this group of patients.

Age of commencement Although there have been some retrospective reports of its use in children, the safety and efficacy of this drug has not been formally evaluated in children under 10 years of age.

Use of vitamin C The effect of vitamin C on iron excretion with deferiprone is not clear and is thus not recommended.

Safety monitoring Weekly blood counts are necessary throughout treatment so that a falling white cell count can be detected early and treatment stopped before overwhelming sepsis develops. If severe neutropenia or agranulocytosis develops, re-challenge is contra-indicated. Recent reports of eight deaths from agranulocytosis in patients treated in Europe, cited above, only emphasise the importance of scrupulous monitoring of the blood count throughout treatment.

Combined Desferrioxamine and Deferiprone
A variety of regimens involving combinations of deferiprone and desferrioxamine have been used by clinicians, either in the context of a formal trial or on an ad hoc basis, usually when monotherapy with desferrioxamine or deferiprone has failed to control iron overload or its effects.

Pharmacology In principle, chelators can be given at the same time as each other (simultaneously) or following one another (sequentially). There is considerable variation in the way in which sequential treatment can and has been administered. Some investigators have used the term ‘alternating therapy’ to describe the use of two drugs administered on alternate days, reserving the term ‘sequential therapy’ for when desferrioxamine is given at night and deferiprone during the day. In practice regimes may involve a component of ‘sequential’ and ‘alternating’ therapy, such as when desferrioxamine is given three times a week (alternate nights) and deferiprone every day. Most regimes have tended to give deferiprone daily, at standard doses, combined with varying frequency and dosing of desferrioxamine. The pharmacology of combinations of chelators may be fundamentally different depending on whether the drugs are present in cells or plasma at the same time. By giving desferrioxamine at night and deferiprone by day (sequentially), 24-hour exposure to iron chelation can be achieved (similar to that achieved with 24-hour desferrioxamine infusion or once daily deferasirox. (For more on deferasirox (Exjade), see below). This has the theoretical advantage of 24-hour protection from labile (redox active) iron (Cabantchik, 2005). If the drugs are given at the same time (simultaneously), they may interact in a process that involves the ‘shuttling’ of iron, which may lead to additional chelation of iron from cells or plasma and so improved iron removal. However, there is also a possibility of chelation from metalloenzymes, leading to increased drug-related toxicity.

In short, the simultaneous use of these drugs has not been tested formally in large enough patient groups to allow firm, evidence-based recommendations about efficacy and safety.

However, data from several prospective studies indicate that sequential (or alternating) use of these chelators can be used to achieve control of iron overload and improvement in heart iron measurements.

Evidence of efficacy of combined treatments
Effects of sequential use on serum ferritin Four randomised studies have compared levels of serum ferritin in patients using combined treatments with those under other treatment regimes. One study (Mourad et al, 2003) found the decrease in serum ferritin achieved with five days of desferrioxamine monotherapy (n=11) to be similar to that achieved with two nights of desferrioxamine plus seven days of deferiprone at 75 mg/kg (n=14). Another randomised study, involving 30 patients and three different treatments (Gomber et al, 2004), found that the decrease in serum ferritin was greatest with five nights of desferrioxamine, albeit not significantly different from that achieved with a combined treatment of desferrioxamine two nights a week plus deferiprone seven days a week. A third randomised study, involving 60 patients (Galanello, 2006), found no difference in the level of decrease in serum ferritin in patients randomised to combined treatment (two days desferrioxamine at 33 mg/kg + seven days deferiprone at 75 mg/kg) or to desferrioxamine five nights a week at 33 mg/kg.

Taken together, these studies suggest that serum ferritin can be controlled with a relatively small dose of desferrioxamine given twice a week, when combined with deferiprone at standard doses (75 mg/kg/day). In a more recent randomised study of 65 patients (Tanner, 2007), serum ferritin was decreased more by combined treatment (desferrioxamine five days a week plus deferiprone seven days a week) than with standard desferrioxamine monotherapy (40 mg/kg five times a week).

Effects of sequential use on liver iron One randomised study, assessing the effects on liver iron of combined treatment compared to desferrioxamine monotherapy (n=60), found LIC of <7 mg/g dry wt at baseline – a figure maintained, on average, in both arms of the study (Galanello, 2006). Another prospective randomised study, comparing the effect of desferrioxamine monotherapy administered subcutaneously five times a week with that of deferiprone administered daily at 75 mg/kg daily or deferiprone at 75 mg/kg daily, plus twiceweekly desferrioxamine, found that the decrease in liver iron was highest in the desferrioxamine monotherapy group and lowest in the deferiprone monotherapy group, with sequential combination treatment showing an intermediate effect (Aydinok, 2005). A further randomised study, comparing deferiprone plus desferrioxamine five times a week with desferrioxamine monotherapy five times a week (n=65), found that an improvement in liver T2* (as a surrogate measure of LIC) was greater in the combination arm (Tanner, 2007).

Effects of sequential use on heart function In the above-mentioned randomised controlled study of 65 patients (Tanner, 2007), with baseline LVEF >56% changes in LVEF improved by approximately 2.5% in the combination arm and 0.5% in the desferrioxamine monotherapy arm. Two observational studies have also reported changes in heart function under combined treatment. In 79 patients treated with a variable desferrioxamine regimen plus deferiprone at 75 mg/kg seven days a week for a variable time, there was an improvement in LVEF measured by echocardiography (Origa, 2005). In an observational study of 42 patients with sequential use of treatment over three to four years (deferiprone 75 mg/kg/day plus desferrioxamine two to six days a week), the LV shortening fraction improved (Kattamis, 2006).

Effects of sequential use on cardiac iron In a randomised controlled study of 65 patients with moderate heart iron loading (T2* 8-20 ms), myocardial T2* changes with combined deferiprone 75 mg/kg seven days a week plus desferrioxamine five days a week were compared with patients on standard desferrioxamine five times a week (Tanner, 2007). T2* improved in both groups but was significantly greater (6 ms) with combined treatment than with desferrioxamine monotherapy (3 ms). In an observational study, the T2 of the heart improved with combined therapy (Kattamis, 2006).

Safety of combined treatment Formal safety data on combined treatment are limited. A meta-analysis of the incidence of agranulocytosis with combined regimes compared with deferiprone monotherapy suggests that the risk may be increased several-fold, although the number of patients that qualify for evaluation is small (Macklin, IND submission to FDA, 2004). The increased incidence appeared to occur mostly in those regimes where the drugs were administered simultaneously. In a recently reported prospective study, one case of agranulocytosis and two of neutropenia were seen at one year in the combination arm, including 32 patients (Tanner, 2007), while no increase in arthropathy was observed in the same group of patients.

Conclusions and possible treatment regimens ''The above-mentioned studies suggest that some combined regimens can control iron overload in the liver and heart where monotherapy is not having the desired effects. In general, if a patient is not doing well with monotherapy, combined treatment offers an additional approach (as does intensive therapy with at least 50 mg/kg/day of desferrioxamine for as many hours a day as is practicable-see above). For patients with very high levels of heart iron or cardiac dysfunction, 24-hour treatment with desferrioxamine and daily therapy with deferiprone should be strongly considered.''

Deferasirox (Exjade)
Deferasirox was developed by Novartis as a once-daily, oral monotherapy for the treatment of transfusional iron overload. The drug has been licensed as first-line monotherapy for thalassaemia major in over 70 countries worldwide, including the US (2005) and the EU (2006). The average follow up in large-scale prospective trials at the time of writing is three years.

Pharmacology This is an orally absorbed iron chelator, with two molecules binding each iron atom. The tablet is dissolved in water (or apple juice) using a non-metallic stirrer, and consumed as a drink once daily, preferably before a meal. Metabolic iron balance studies show iron to be excreted almost entirely in the faeces, with less than 0.1% of the drug eliminated in urine (Nisbet-Brown, 2003). Metabolism occurs predominantly by glucuronidation in the liver. Due to the long plasma half-life (nine to 11 hours), once-daily administration provides 24-hour chelation from labile plasma iron (Nisbet-Brown, 2003; Galanello, 2003). The efficiency of chelation is 28%, over a wide range of doses and levels of iron loading.

Evidence of effectiveness of deferasirox Deferasirox has undergone preclinical and clinical evaluation that has included largescale prospective randomised studies involving over 1,000 patients, to assess safety, efficacy and the dose response effects of treatment. At this time, evidence of effectiveness is confined to serum ferritin and liver iron.

Dose effect on serum ferritin A dose-dependent effect on serum ferritin has been observed in several studies (Cappellini, 2006; Piga, 2006). A prospective randomised study comparing the effects of deferasirox in 296 thalassaemia major patients with that of desferrioxamine in 290 patients found that 20 mg/kg daily of deferasirox stabilised serum ferritin close to 2,000 Ìg/L. At 30 mg/kg, serum ferritin was reduced, with an average fall of 1,249 Ìg/L over one year (Cappellini, 2006). Longer-term analysis of ferritin trends shows that the proportion of patients with ferritin values <1,000 Ìg/L and less than 2,500Ìg/L is decreasing progressively with time (Porter, 2006).

Dose effect on liver iron and iron balance In the same prospective study, iron balance was achieved at 20 mg/kg/day, with mean LIC constant over one year (Cappellini, 2006). Negative iron balance was achieved at 30 mg/kg/day, while mean LIC fell by 8.9 mg/g dry wt (equivalent to a decrease in body iron of 94 mg/kg body weight) over one year. These are average trends and a closer analysis shows that the blood transfusion rate influences the response to treatment (Cohen, 2005). Thus for patients in the high or low transfusion category (see Table 7), the average dose required to achieve iron balance is accordingly adjusted up or down from 20mg/kg/day (Cohen, 2005). Some patients will still fail to achieve negative iron balance at a daily dose of 30 mg/kg/day of deferasirox, and studies are currently underway to assess the effectiveness and safety of higher doses.

A more moderate reduction in LIC occurred in children under six years old, despite the administration of an average dose of 21.9 mg/kg in this subgroup. However, these patients had the highest mean transfusional iron intake.

Effects on heart iron and heart function The effects of deferasirox on heart function and estimated heart iron were not evaluated formally as part of the drug registration process, and formal prospective studies on both heart function and heart iron are now in progress. Retrospective analysis of effects on myocardial T2* after one year and two years of treatment suggests that this measure can be improved in a significant proportion of patients with pre-existing abnormal T2* values (Porter, 2005). Patients with normal LVEF showed no change in this measure over one year (Porter, 2005).

Unwanted effects with deferasirox
Gastrointestinal effects Gastrointestinal disturbances – typically mild and transient – occurred in 15% of patients and included abdominal pain, nausea and vomiting, diarrhoea and constipation, lasting a median of less than eight days. These symptoms rarely required dose adjustment or discontinuation.

Skin rashes These occurred in (11%) of patients and were typically pruritic, maculopapular and generalised, but occasionally confined to palms and soles of the feet. A rash typically developed within two weeks of starting treatment. A minority of patients required permanent discontinuation of therapy, and mild rashes often resolved without dose modification.

Increase in serum creatinine An increase in serum creatinine ≥30% on at least two consecutive readings was observed in 38% of patients receiving deferasirox, most frequently at doses of 20 mg/kg and 30 mg/kg (Cappellini, 2006). These increases were sometimes transient and generally within the normal range, never exceeding two times the upper limit of normal (ULN), and were more frequent in the population of patients having the most dramatic decrease in LIC and serum ferritin. In the randomised study, a dose reduction of 33-50% was planned if at least two consecutive increases in serum creatinine were >33% above baseline. As the creatinine spontaneously normalised in a number of cases, dose reductions were instituted in only 13%. In about 25% of those cases, the creatinine then returned to baseline, while in the rest it remained stable or fluctuated between baseline and the maximum increase observed prior to dose reduction. With follow up of a median of three years at the time of writing, no evidence of progressive renal dysfunction has been reported where the above doses and modifications are used. Further work on the mechanism of creatinine increases is being undertaken.

Effects on the liver Overall a decrease in ALT* was seen, which paralleled improvements in LIC (Deugnier, 2005). Two patients out of 296 developed elevated ALT values greater than twice the ULN while receiving deferasirox for one year, which the investigator reported as related to the administration of the drug.

Other effects No agranulocytosis, arthropathy or growth failure was associated with deferasirox administration. Comparing 296 patients who received deferasirox in a one-year prospective randomised study with 290 patients receiving desferrioxamine, deafness, neurosensory deafness or hypoacusis were reported as adverse events in eight patients on deferasirox and seven in desferrioxamine. Cataracts or lenticular opacities were reported as adverse events in two patients on deferasirox and five on desferrioxamine (Cappellini, 2006).

Convenience and impact on quality of life Studies comparing satisfaction and convenience of DFS with DFO in thalassaemia major show a significant and sustained preference for DFS (Cappellini, 2006). Total withdrawals in deferasirox-treated patients were 6% at one year compared with 4% with desferrioxamine (Cappellini, 2005). This compares with a dropout rate of 15% at one year with deferiprone (Cohen, 2000). Based on thalassaemia patient reported preferences for DFS and DFO, published compliance data with DFO and the probability of complications from iron overload in relation to compliance with DFO, the cost effectiveness per qualityadjusted life year (QALY) gained is 4.1 per patient for DFO and 8.1 per patient for deferasirox.

(*)ALT: Alanine L-Aminotransferase

Recommended treatment regimens with deferasirox
Recommended dosing The drug is taken orally as a suspension in water, once daily, preferably before a meal. A starting dose of 20 mg/kg is recommended for thalassaemia major patients who have received 10-20 transfusion episodes and currently receive standard transfusion at rates of 0.3-0.5 mg of iron/kg/day. In those patients in whom there is a higher rate of iron intake from transfusion (>0.5 mg/kg/day) or in patients with pre-existing high levels of iron loading, where a decrease in iron loading is clinically desirable, 30 mg/kg/day is recommended. For patients with a low rate of iron loading (<0.3 mg/kg/day), a dose of 10-15 mg/kg may be sufficient to control iron loading.

Age of commencement Prospectively randomised studies of deferasirox in children as young as two years of age have been carried out (Cappellini, 2006; Galanello, 2006). A fall in LIC was seen across all age groups analysed, with no age-related adverse effects: in particular, no adverse effects on growth, sexual development or bones were seen (Piga, 2006). The drug also appears to be palatable to children at this young age. On the basis of present knowledge, the criteria for starting treatment (ferritin level, age, number of transfusions) do not differ from those of desferrioxamine.

Other indications and contraindications Deferasirox is contraindicated in patients with renal failure or significant renal dysfunction. For patients with evidence of significant heart dysfunction (e.g. LVEF below reference range) there is very limited clinical experience and treatment cannot be recommended at this time for patients with heart failure or poor LV function. The combined use of deferasirox with other iron chelators has also not been formally assessed and therefore cannot be recommended at this time. The drug should not be used in pregnant women.

''Experience of use in patients with pre-existing renal disease (baseline creatinine outside reference range) is insufficient at this time to recommend its use. As the median follow up in large-scale studies is three years at this time, vigilance in monitoring for possible longterm effects is still advisable.''

Summary of Iron Overload and its treatment: donor Hct = volume of RBC). On average 200mg iron/donor unit; iron/kg/year or 0.32-0.64 mg/kg/day; considered: (i) iron overload; (ii) inflammation; (iii) hepatitis; and/or (iv) liver damage. When serum ferritin is low, the following should be considered: (i) low body iron; (ii) vitamin C deficiency. In thalassaemia intermedia, ferritin underestimates the degree of iron overload. Ferritin levels related to low risk are below 2,500 mg/l, preferably below 1,000 mg/l; Very low risk = <1.8 mg/g dry weight Low to moderate risk = 1.8 - 7 mg/g dry weight; Moderately high to high risk = 7 - 15 mg/g dry weight; Very high risk = > 15 mg/g dry weight; Total body iron stores = 10.6 x LIC (mg/g dry weight); LIC is measured by: a) Liver biopsy – indicated if ferritin levels deviate from expected trends, if coexistent hepatitis and if uncertain response to chelation; b) SQUID – not universally available; c) MRI – R2.
 * 1.08 mg of iron in 1ml of pure red cells (HCT = 100%);
 * Rate of iron loading: volume of RBC x 1.08 (annual transfusion requirements x
 * Recommended transfusion 100-200 ml/kg/year is equivalent to 116-232 mg
 * Serum ferritin broadly related to body iron. When high, the following should be
 * Ranges of LIC reflecting levels of RISK:-
 * Cardiac iron reflected by heart function tests and measured by MRI T2*;
 * Urinary iron – used to monitor desferrioxamine or deferiprone dose effects. Variability in daily excretion, and
 * NTBI and LPI – not yet routinely used.

Desferrioxamine: mg/kg; the first trimester and can be used in the second and third trimesters, in selected cases; Indications: a) Persistently high serum ferritin; b) LIC > 15 mg/g dry weight; c) Significant heart disease, and; d) Prior to pregnancy or bone marrow transplantation Dose: 50 mg/kg/day (up to 60 mg/kg/day)
 * Initiate treatment after first 10-20 transfusions or ferritin level above 1,000 μg/l;
 * If before 3 years of age monitoring of growth and bone development is recommended;
 * Therapeutic index = mean daily dose (mg/kg) (Mean daily dose = actual dose of each infusion x doses/7 days) /ferritin (mg/l). Keep index < 0.025 at all times;
 * Standard treatment: a) Slow subcutaneous infusion over 8-12 hours, b) 10% desferrioxamine solution (5 ml water for each 500 mg vial), and c) infusion pump (several types available);
 * Standard dose: a) children 20-40 mg/kg (not exceeding 40 mg/kg, until growth has ceased), and b) adults 50-60 mg/kg. Infuse 8-12 hours 6 nights minimum per week;
 * Alternative route: subcutaneous bolus – two S.C. boluses/day to a total daily dose of 45
 * Vitamin C-dose limited to 2-3 mg/kg/day given orally at the time of infusion;
 * Pregnancy – desferrioxamine can be used in pregnancy. It should be interrupted during
 * Intensive chelation with desferrioxamine – continuous 24-hourly infusions IV or SC.
 * In-dwelling catheters: danger of infection and thrombosis.

Deferiprone:


 * Standard dose: 75 mg/kg/day in 3 divided dose (up to 100 mg/kg/day, but as yet not enough information);
 * Children above 10 years of age;
 * Vitamin C concomitant treatment not recommended;
 * Weekly blood counts (more frequently if signs of infection);
 * Pregnancy – stop treatment. It is recommended that sexually active patients should use contraception;

COMBINATION THERAPY. In patients for whom monotherapy with desferrioxamine or deferiprone is not controlling body levels of iron or myocardial iron or in the presence of significant heart disease, combined regimes offer an alternative that can reduce iron levels in both the liver and heart. No recommendations as to which is the more effective combination can be made at present.

CAUTION: agranulocytosis may be more frequent in combination therapy, especially in simultaneous use.

Deferasirox:

Starting dose 20 mg/kg/day. After 10-20 transfusions (iron intake (0.3-0.5 mg/kg/day); If pre-existing iron overload (or iron intake > 0.5 mg/kg/day), the dose of 30 mg/kg/day is recommended. For patients with low rate of iron loading (<0.3 mg/kg/day), lower doses may be sufficient to control iron loading; some patients will still fail to achieve negative iron balance at a daily dose of 30mg/kg/day of deferasirox, and studies are currently underway to assess the effectiveness and safety of higher doses;
 * Recommended dose:
 * Administration: Tablet dissolved in water (or apple juice), using a non-metallic stirrer. Taken once a day before a meal.
 * Continuous Monitoring
 * Use in children > 2 (FDA) and >6 (EMEA) years of age
 * Contraindicated in renal failure or significant renal dysfunction;
 * Cannot be given during pregnancy