|Home | About | Journals | Submit | Contact Us | Français|
Iron overload disorders represent a heterogenous group of conditions resulting from inherited and acquired causes. If undiagnosed they can be progressive and fatal. Early detection and phlebotomy prior to the onset of cirrhosis can reduce morbidity and normalise life expectancy. We now have greater insight into the complex mechanisms of normal and disordered iron homeostasis following the discovery of new proteins and genetic defects. Here we review the normal mechanisms and regulation of gastrointestinal iron absorption and liver iron transport and their dysregulation in iron overload states. Advances in the understanding of the natural history of iron overload disorders and new methods for clinical detection and management of hereditary haemochromatosis are also reviewed. The current screening strategies target high-risk groups such as first-degree relatives of affected individuals and those with clinical features suggestive of iron loading. Potential ethical, legal and psychosocial issues arising through application of genetic screening programs need to be resolved prior to implementation of general population screening programs.
The commonest iron overload disorder in the general population is Type 1 hereditary haemochromatosis (HH). This autosomal recessive disorder is most commonly due to a homozygous C282Y mutation within the HFE gene on chromosome 6. HFE gene mutations are common and can lead to irreversible organ damage with complications of cirrhosis, diabetes and arthritis.1 The molecular mechanisms of iron absorption, the regulation of hepatic iron transport and homeostatic mechanisms have been further defined through the recent discovery of a number of new genes and proteins in various iron overload disorders in animal models and in humans. Early treatment of HH associated iron overload prevents end-organ damage, which has led to discussion regarding population based screening for the disease.
The body has no effective means of excreting iron and thus the regulation of absorption of dietary iron from the duodenum plays a critical role in iron homeostasis in the body. The body absorbs 1 to 2 mg of dietary iron a day and this is balanced with losses via sloughed intestinal mucosal cells, menstruation and other blood losses. Most of the iron in the body is distributed between red blood cell haemoglobin, the liver, muscle and macrophages of the reticuloendothelial system. Whilst iron is essential for cellular metabolism and aerobic respiration, cellular iron overload leads to toxicity and cell death via free radical formation and lipid peroxidation and thus iron homeostasis requires tight regulation.2,3
Dietary iron is found in haem and ionic (non-haem) forms and their absorption occurs at the apical surface of duodenal enterocytes via different mechanisms. Dietary non-haem iron primarily exists in an oxidised (Fe3+) form that is not bioavailable and must first be reduced to the Fe2+ form before it is transported across the intestinal epithelium. The responsible ferrireductase enzyme is a membrane bound haemoprotein called Dcytb expressed in the brush border of the duodenum.4 Fe2+ is then transported into the cell by a transporter called divalent metal transporter 1 (DMT1) which also traffics other metal ions such as zinc, copper and cobalt by a proton coupled mechanism.5,6
Haem iron is absorbed into the enterocyte by a different, as yet unidentified, haem receptor. A cell surface receptor called FLVCR is a major transport facilitator for feline leukemia virus subgroup C and has been shown to transport cytoplasmic haem in the human erythrocyte and may also impact on haem transport in the intestine and liver.7 Once internalised in the enterocyte, iron is released from haem by haem oxygenase and then either stored or transported out of the enterocyte across the basolateral membrane via mechanisms similar to that of ionic iron.
Ferroportin 1 (Fpn1, also known as Ireg1, MTP 1) is the only putative iron exporter identified to date.8–11 Ferrous iron once exported across the basal membrane by Fpn1, is then oxidised by a multi-copper oxidase protein called hephaestin before being bound by plasma transferrin.12 Fpn1 is also the putative iron exporter in macrophages and hepatocytes.13
The liver is the main storage organ for iron. In iron overload, free radical formation and generation of lipid peroxidation products may result in progressive tissue injury and eventually cirrhosis or hepatocellular carcinoma. Iron is sequestrated in hepatocytes predominantly in the form of ferritin or haemosiderin. The uptake of transferrin-bound iron (TBI) by the liver from plasma is mediated by two transferrin receptors - transferrin receptor 1 (TfR1) and TfR2. In iron overload, TfR1 is down-regulated in hepatocytes and in untreated HH subjects there is a complete absence of TfR1 expression on hepatocytes.14,15 The haemochromatosis protein (HFE) is also expressed by hepatocytes and is likely to regulate TfR1- mediated uptake of TBI.
TfR2 is highly expressed in human liver and is likely to play an important role in liver iron loading in iron overload states.16 Unlike TfR1, TfR2 lacks an iron response element and thus is not reciprocally regulated in response to the level of plasma iron. Instead, TfR2 protein expression is regulated by transferrin saturation. TfR2 protein is up-regulated in iron overload and in the Hfe knockout mouse model of HH and may contribute to increased TBI uptake by the liver in iron overload.17,18 It has a 30-fold lower affinity than TfR1 for TBI but has a higher capacity to transport TBI into the hepatocyte. In normal and iron loaded conditions, expression of TfR2 exceeds that of TfR1 suggesting that TfR2 plays an important role in hepatic iron loading in HH.19
As transferrin becomes saturated in iron overload states, excess iron is also found as non transferrin-bound iron (NTBI) and is likely to play an important role in hepatocyte iron loading in HH and other iron overload conditions. NTBI is extremely toxic and is cleared rapidly from plasma by the liver.20 It has been shown that humans and mice lacking transferrin develop massive iron overload in non-haematopoietic tissues such as the liver and pancreas.21,22 In Hfe knockout mice plasma NTBI is increased and hepatocyte NTBI uptake is increased 2.5 fold.23 NTBI is reduced and transported across the hepatocyte membrane via a carrier-mediated process consistent with DMT1.24 Fpn1 is likely to mediate the transport of iron out of hepatocytes, which is then oxidised by caeruloplasmin and bound to transferrin.8,25
The absorption of iron is dependent on the body’s iron stores, hypoxia and rate of erythropoiesis. Two models have been proposed to explain how the absorption of iron is regulated. These models have been termed i) the crypt programming model and ii) the hepcidin model.25 The crypt programming model proposes that the crypt cells sense body iron levels, which in turn regulate the absorption of dietary iron via the mature villus enterocytes. The second model proposes that liver hepcidin, which is regulated by a number of factors such as liver iron levels, inflammation, hypoxia and anaemia, is secreted into the blood and interacts with villus enterocytes to regulate the rate of iron absorption. There is evidence to support both models and it is possible that both control mechanisms may contribute to the regulation of iron absorption.
The crypt programming model proposes that enterocytes in the crypts of the duodenum take up iron from the plasma. The intracellular iron level of the crypt cells corresponds to the body’s iron stores, which in turn determines the amount of iron absorbed from the gut lumen as these crypt cells migrate upwards to become absorptive cells at the brush border.26 The crypt cells express both TfR1 and TfR2 which mediate the cellular uptake of transferrin-bound iron from plasma.
TfR1 is ubiquitously expressed and transferrin mediated iron uptake is thought to occur in most cell types.11 HFE however is highly expressed in crypt cells.27 HFE is a MHC class 1-like molecule which interacts with β2-microglobulin and forms a complex with TfR1. Its role in the regulation of TfR1 mediated transferrin-bound iron (TBI) uptake still remains unclear.28 It has been shown that HFE competitively inhibits the binding of TBI to TfR1, reduces the cycling time of the HFE/TfR1- TBI complex through the cell and reduces the rate of iron release from transferrin inside the cell. Conversely, when HFE and β2-microglobulin are over-expressed in Chinese Hamster Ovary cells, uptake of TBI was enhanced due to increased recycling of TfR1 through the cell and produces the opposite effect of higher intracellular iron concentrations.29 In human HH and in the Hfe knockout mouse model, the lack of HFE results in decreased TBI uptake from plasma into the enterocyte suggesting that HFE usually functions by enhancing TBI uptake from the plasma into the crypt cell via the TfR1 and may also inhibit the release of iron from the cell via ferroportin 1.30,31
TfR2 is restricted to hepatocytes, duodenal crypt cells and erythroid cells suggesting a more specialised role in iron metabolism.16 It is expressed at lower levels in the duodenum and does not interact with HFE in vitro.19,32 The role of TfR2 in the genesis of iron overload in the liver may be more important than its role in absorption of iron in the duodenum. The amount of iron absorbed by the enterocyte in the villus region is determined by regulation of the expression of the iron transporters DMT1 and Fpn1 in response to iron and other signalling peptides such as hepcidin. The intracellular iron concentration controls the interaction of cytosolic iron regulatory proteins (IRPs) 1 and 2 with iron regulatory elements (IREs) in the 3' and 5' regions of mRNA.33,34 IRP1 contains an iron-sulfur cluster and in the presence of iron acts as an aconitase (interconverting citrate and isocitrate). In the absence of iron, IRP1 binds to IREs. IRP2 undergoes iron-dependent degradation in iron-replete cells and is not available to bind to IREs. IRP2 is sensitive to degradation in the presence of nitrous oxide (NO), whereas IRP1 is activated by NO.34 IRP1 operates as an iron sensor only in high oxygen environments whereas IRP2 is the major sensor of iron in mammalian cells at physiological oxygen tensions.35 When IRP1 and IRP2 bind to the IRE in the 3'-untranslated region of TfR1 or DMT1 mRNA, the transcript is stabilised, translation proceeds, and the proteins are synthesised. Thus, a high IRP binding activity reflects low body iron stores and results in up-regulation of DMT1 and TfR1 levels in the duodenum and increased dietary iron absorption.36–38 When IRPs bind to the 5'-untranslated region of ferritin mRNA, translation of the transcript is blocked and synthesis is halted. Thus, ferritin levels are reciprocally regulated – being increased in iron replete states and decreased in iron deplete states. Fpn1 contains an IRE and is regulated by intracellular iron levels as well as post-translational by hepcidin. HFE and TfR2 do not contain IRE and their expression is not iron regulated.16,39
Hepcidin (also known as HAMP, LEAP 1) is a 25 amino acid cysteine rich peptide with antimicrobial properties that has recently been identified as an important regulator of iron homeostasis. It is almost exclusively synthesised by hepatocytes and is cleared by the kidney.40 The USF2 knockout mouse, which does not express hepcidin, was found to have iron overload resembling HH where iron deposition was significant in the liver, pancreas, heart and kidneys but not in the spleen.41 Conversely, transgenic mice constitutively expressing hepcidin have profound iron deficiency.42 Hepcidin expression is increased with iron loading of animals suggesting it is a compensatory response to limit iron absorption from the intestine and when injected into mice it inhibits iron absorption. It also inhibits macrophage iron release.43–46 Hepcidin expression decreases in response to anaemia and hypoxia independent of iron loading.47 Hepcidin has antimicrobial properties and expression is increased in mice and humans with inflammation, suggesting that it may also play an important part in the causation of anaemia of chronic disease.47 Hepcidin expression is increased during the acute phase response and is independent of the effects of HFE, TfR2 or β2-microglobulin and is mediated by the inflammatory cytokine IL-6.48,49 Recently, Fpn1 has been identified as a putative hepcidin receptor; binding of hepcidin to Fpn1 results in internalisation of Fpn1 and loss of Fpn1 function.50 Thus it is hypothesised that when hepcidin levels are increased in iron overload or inflammation, iron release from intestinal crypt cells and macrophages is reduced. In contrast, in iron deficiency or HH when hepcidin levels are reduced, it is likely that Fpn1 expression and iron release from intestinal cells and macrophages is increased.46,50 There is emerging evidence that hepcidin may act directly on mature villus enterocytes rather than crypt enterocytes. There are several situations where iron absorption can be modulated more rapidly (within hours) than can be accounted for via the mechanism that involves the programming and maturation of crypt enterocytes (lag time of days). This is reflected in the decrease in iron absorption that accompanies an acute phase response and the increase in iron absorption that is seen following transfusion with reticulocytes.51,52 A recent study of induced haemolysis and anaemia in rats revealed a four day delay before a significant increase in iron absorption was observed. Hepatic hepcidin expression did not decrease until day 3 which coincided with increased duodenal expression of DMT1, Dcytb and Fpn1, suggesting a direct effect of the hepcidin pathway on mature villus enterocytes rather than crypt cells. The hepcidin response lag may represent the time required for the body to assess iron needs following the stimulus to increase erythropoiesis or changes in iron status.53,54
Hepcidin production in the hepatocyte may be regulated by the uptake of transferrin bound iron via TfR1/HFE and TfR2 or by the level of expression of HFE and TfR2 protein in hepatocytes and this might explain why mutations in the HFE and TfR2 genes result in iron overload.1 Further studies are required to characterise the molecular mechanisms of the actions of hepcidin linking iron stores to intestinal iron absorption.
Type 1 haemochromatosis is the classical and commonest of the primary HH syndromes (Table). It is an autosomal recessive disorder resulting in iron overload and variable multi-organ dysfunction. The HFE gene was identified on chromosome 6 by Feder et al in 1996.39 90–95% of affected patients are homozygous for the missense mutation that results in the substitution of tyrosine for cysteine at amino acid 282 (C282Y).1
A more common mutation is the substitution of aspartate for histidine at amino acid 63 (H63D); this mutation may contribute to minor increases in iron levels but rarely iron overload in the absence of C282Y. 1–5% of subjects with HH may be compound heterozygous for both the C282Y and H63D mutations.55,56 A novel mutation of HFE S65C results in the substitution of serine for cysteine and may also result in mild iron overload in the compound heterozygous state with C282Y.57
The HFE C282Y mutation results in the disruption of a critical disulde bridge which decreases the affinity of HFE for β2 microglobulin and TfR1 and the mutant protein is retained in the Golgi complex.58 This results in decreased uptake of plasma TBI by the duodenal crypt cells in the Hfe knockout mouse model.30 This impaired uptake of iron would lead to a “false” iron deficiency of the duodenum despite the increasing total body iron stores which then leads to the up-regulation of IRP, DMT1 and Fpn1 expression and increased iron absorption and trafficking from the duodenal lumen into the circulation. This is supported by the finding that DMT1 and Fpn1 are over-expressed in HH.59,60
The interactions between hepcidin and HFE still remain unclear. Hepcidin expression was decreased despite significant increases in iron loading in the Hfe knockout mouse and in humans with HH.61,62 The lack of up-regulation of hepcidin by the liver in HH implies that HFE may play an important role in the regulation of hepcidin. This is further supported by the observation that constitutive over-expression of hepcidin by Hfe knockout mice prevents iron overload.62 When HFE knockout mice were crossed with USF2 mouse which lacked hepcidin, liver iron accumulation was enhanced in the double knockout mice compared with the mice lacking Hfe alone.63,64 These observations support the concept that HFE and hepcidin are part of the same regulatory axis and HFE might function as a sensor of iron ‘upstream’ from hepcidin.11 Exogenous hepcidin has been shown to act directly on the duodenum in both Hfe knockout and wild-type mice to reduce iron absorption.45 Therefore the lack of up-regulation of hepcidin in HH is likely to lead to increased iron absorption.
Other well defined but rare genetic disorders of iron metabolism result from mutations outside the HFE gene. HH type 2 or juvenile haemochromatosis is a rare autosomal-recessive disorder resulting in iron overload in the second and third decades of life in the same pattern as adult HH.1 Some individuals with HH type 2 have been shown to possess mutations in the hepcidin or hemojuvelin genes.65,66 HH type 3 is a disorder resulting from mutations in the transferrin receptor-2 gene located on chromosome 7q22. It was first described in southern Italy and is inherited in an autosomal-recessive fashion.67 HH type 4 was identified originally in three families in Italy, is inherited in autosomal-dominant fashion, and is caused by mutations in the ferroportin gene (IREG1, MTP1 or SLC11A3).68,69 These disorders are uncommon and should not routinely be screened for, but should be considered if there is HH which can not be explained by the more common HFE gene mutations. Other inherited iron overload disorders such as acaeruloplasminaemia, atransferrinaemia, neonatal iron overload, and autosomal dominant haemochromatosis are extremely rare conditions and account for a negligible proportion of the overall burden of inherited iron overload syndromes.70,71
In conditions of ineffective erythropoiesis such as thalassaemia, hereditary sideroblastic anaemias and certain myelodysplastic syndromes, the hyperplastic erythroid marrow may stimulate increased iron absorption to an extent that causes clinical iron overload. There is a direct correlation between the level of erythropoiesis and iron absorption.72
Each unit of blood contains 200–250 mg of iron. Chronic blood transfusion therapy is required in the treatment of children with beta thalassaemia, bone marrow failure and treatment of complications of sickle cell anaemia. With hypertransfusion, the iron load initially accumulates in the reticuloendothelial macrophages but is deposited subsequently in the parenchymal cells of the liver, heart, pancreas and endocrine tissue as seen in Type 1 haemochromatosis. Phlebotomy is not usually a treatment option because of the underlying anaemia. Chelation therapy with deferoxamine administered as a continuous infusion is often the only option. Side effects of deferoxamine include susceptibility to severe fungal infections, visual and auditory neurotoxicity.73 An oral iron chelator, deferiprone, has been studied but has limitations in its efficacy.74,75
Iron overload in sub-Saharan Africa was believed originally to result from the ingestion of large amounts of iron derived from traditional home brewed beer fermented in non-galvanised steel drums. However, only a small number of these beer drinkers get iron overload, suggesting that a genetic predisposition may be involved. Several lines of evidence indicate that this gene locus is not related to the HFE gene but the exact putative locus has not been identified.76 Heterozygosity for a common polymorphism (Q248H) in the Fpn1 gene was identified in a proportion of African and African-American subjects with iron overload.77 However a recent study of 19 southern African families with iron overload failed to show evidence that the ferroportin (Q248H) mutation is responsible for the condition.78
Increased hepatic iron deposition has been recognised in association with cirrhosis not related to haemochromatosis. It is seen more commonly in non-biliary causes of cirrhosis such as alcoholic liver disease, chronic viral hepatitis and non-alcoholic steatohepatitis.79 Whilst the level of hepatic iron deposition is not as high as those levels found in haemochromatosis, up to 7% of subjects with non-HFE related chronic liver disease and iron overload have hepatic iron levels in the range reported for haemochromatosis. The mechanism by which iron accumulates in the liver in these conditions is not well understood. Heterozygosity for the C282Y mutation in the HFE gene may play a role.80 In cirrhosis there is a reduced synthesis of transferrin and a relative increase in the plasma NTBI levels which may contribute to hepatic iron overload and toxicity.81 Carbon tetrachloride induced hepatocyte injury and regeneration results in increased iron uptake, partly due to increased TfR1expression on regenerating hepatocytes.82 Decreased incorporation of iron into red cells as well as decreased red cell survival in cirrhosis related hypersplenism, combined with portocaval shunting may also contribute to liver iron overload.83,84
Heterozygosity and homozygosity for the C282Y mutation of the HFE gene occurs in up to 15% and 0.5% respectively, of subjects of northern European descent but is less frequent in southern and eastern European populations.25,86 The estimated prevalence of the mutation is lowest in Pacific Islanders (0.012%) and Asians (0.000039%), and is also rare in indigenous populations of Africa, South America and Australia.2 As many as 95% of subjects of northern European descent with HH are homozygous for the C282Y mutation.88 The H63D mutation is more common with a carrier frequency of 15% to 40% in European populations and over 5% in Asian and the Middle Eastern populations.89 Among northern European populations homozygosity for H63D ranges from 2% to 4%, heterozygosity from 20% to 25% with 2% to 4% having compound heterozygous mutations, carrying one allele with the C282Y mutation and one allele with the H63D mutation.56,90
Controversy exists whether the definition of HH for screening purposes should be based on genetic or biochemical criteria. The genotypic definition requires the presence of the homozygous C282Y HFE mutation. The compound heterozygous C282Y/H63D mutation is not universally included as a case definition as it accounts for less than 10% of all HH cases and rarely leads to liver disease.1 Further complicating use of the genetic definition is that not all C282Y homozygotes develop iron overload or organ damage. Therefore, C282Y homozygous subjects are best regarded as those who are at increased risk of development of phenotypic expression and thus require monitoring up to the age of 50 or 60 years for evidence of iron overload.
The phenotypic case definition of HH relies on biochemical evidence of iron overload as determined by fasting serum transferrin saturation (TRS) and ferritin levels. TRS is more sensitive than ferritin for phenotypic identification and is used as the initial screening test. Blood should be collected in the fasting state and the test should be repeated if the result is elevated.91 A fasting TRS level of greater than 45% confers a sensitivity of at least 98% for identification of C282Y homozygotes and 22% for identification of heterozygotes. Choosing a higher TRS threshold level of 60% for screening decreases the sensitivity to 79–86%.92,93 Conversely if a lower cut-off for TRS were used, there is minimal increase in sensitivity but specificity and positive predictive values decrease.94 In view of the progressive increase in TRS over time in adults homozygous for the C282Y mutation, the addition of an age threshold of 40 years may also increase specificity and sensitivity of biochemical screening from a population screening perspective.95 Unsaturated iron-binding capacity has performance characteristics comparable to TRS with sensitivity ranging from 78–90% and specificity of 90– 97% and may be a suitable alternative to TRS for screening purposes.92
Examination of the penetrance of the C282Y homozygous mutation is complicated by varying definitions of disease expression (biochemical versus clinical), variable lengths of lead-time and confounding genetic and environmental factors that influence disease expression. In addition, ascertainment bias adds to the difficulty of interpreting studies examining phenotypic expression. Case series are likely to overestimate the severity of individual cases; blood donor derived statistics underestimate case severity due to exclusion of individuals with clinical disorders or laboratory abnormalities; kinship studies may be misleading because of the tendency to overestimate penetrance and are unable to independently assess the effects of environmental exposure and concomitant modifier genes.96,97 Despite this, recent studies of community genetic screening and penetrance of C282Y homozygosity have helped to clarify the preventable disease burden. These studies indicate that phenotypic development of significant clinical sequelae is not universal and may well be comparatively rare.98
Large cross-sectional cohort studies from Europe, United Kingdom, and North America have suggested biochemical penetrance in up to 75% of subjects but relatively low clinical penetrance.87,99–101 Beutler et al. screened over 41,000 individuals in a health clinic setting for HH. Of the 152 C282Y homozygotes 75% of the men and 40% of the women had a TRS greater than 50%. Common (but non-specific) symptoms of HH such as fatigue, arthralgia and impotence were no more prevalent among C282Y homozygotes than among subjects lacking the mutation. 8% of C282Y homozygotes had raised liver aminotransaminases and less than 1% of C282Y homozygotes developed clinically overt haemochromatosis (defined as the combination of elevated liver enzymes, raised serum ferritin, diabetes, heart failure and possible skin pigmentation).99 Similarly, McCune and colleagues estimated only 1.2% of all C282Y homozygotes in South Wales had been clinically diagnosed with HH, of which 51% had evidence of significant iron overload and 11% had cirrhosis.100 Biochemical screening of over 65,000 people in Norway found 0.4% of subjects with HH and iron overload, of which only 1.5% had cirrhosis.101
In the recently published cross-sectional HEIRS study of nearly 100,000 subjects the C282Y homozygous mutation was associated with a three fold increased risk of self-reported liver disease but not diabetes, arthropathy or heart disease. Other studies have documented higher rates of biochemical and clinical expression. Longitudinal population based studies have shown that 50% of individuals homozygous for the C282Y mutation had clinical features of HH such as arthritis, skin pigmentation and liver disease.86 In addition, TRS was 45% or greater in over 90% of individuals. Long term natural history studies have demonstrated that a single screening event with TRS at a median age of 31 years would miss 60% of untreated adult HH subjects, all of whom subsequently developed biochemical iron overload and 30% developed biopsy proven significant liver fibrosis or cirrhosis.95 Andersen and colleagues reported similar tendencies for the TRS values to increase over a 25 year follow up of C282Y homozygous individuals.102 However, none of the homozygotes developed overt clinical haemochromatosis and there was no rise in the prevalence of arthritis compared with controls. A recent study demonstrated that 8% of male and 1.7% of female asymptomatic C282Y homozygotes detected at health checks or by family screening had hepatic cirrhosis when evaluated clinically and with liver biopsy. Thus, penetrance of the disorder may be underestimated without appropriate clinical evaluation.103 In addition, the apparent variation in clinical phenotype may be associated with modulating factors such as higher meat and alcohol intake as well as other environmental factors.104 Overall, undiagnosed and untreated HH may lead to cirrhosis and reduced life expectancy in up to 10% of individuals. The observations of variable phenotypic expression help explain the discordance between the gene frequency and the prevalence of diagnosed cases of iron overload attributed to HH.
Early detection is important as treatment with phlebotomy prevents end organ damage and restores normal life expectancy. Life expectancy is reduced in those with cirrhosis, patients who develop diabetes or those who could not be depleted of iron during the first 18 months compared to those without these complications. Life expectancy is not reduced in patients with haemochromatosis without cirrhosis compared to the age and sex matched normal population. Compared to the normal population, liver cancer was 219 times more frequent, cardiomyopathy 306 times more frequent and cirrhosis 13 times more frequent in patients with HH.105 However, given the variability of phenotypic expression and non-specific nature of symptoms such as lethargy, arthralgia or abdominal pains, the diagnosis can be difficult to make.3 High risk populations include those with a family history of HH, individuals with hepatomegaly, abnormal iron studies, abnormal liver function tests, porphyria cutanea tarda, late onset diabetes mellitus, peripheral arthritis, early onset impotence, infertility and dilated cardiomyopathy. The most useful biochemical tests of phenotype are the serum TRS and the ferritin level. A raised TRS is the earliest phenotypic marker for type 1 haemochromatosis and a fasting TRS of >45% will detect at least 98% of patients with iron overload.4,5 Screening is most beneficial when undertaken in individuals aged 40 years or older as individuals who are homozygous for C282Y may have relatively normal iron stores in their third and fourth decades yet progress and develop significant iron overload and organ injury in their fifth decade or later.86 Sequential measurements of TRS may be required over a period of many years to determine non-expression in younger C282Y homozygous adults.
Serum ferritin levels correlate with total body iron stores but are not as sensitive as TRS in detecting disease. The ferritin levels can vary significantly over time and may fluctuate in the presence of inflammation, hepatocellular necrosis and malignancy due to increase in its transcription via inflammatory cytokines. While a ferritin level less than 1000 μg/L is a useful negative predictor of severe fibrosis, well documented cases of cirrhosis and progressive fibrosis have been reported in HH subjects with ferritin levels well below 1000 μg/L.106
Persistently elevated TRS mandates genetic testing for type 1 HH. Only 1 in 700 individuals in a population of Anglo- Celtic descent will have clinical iron overload with no HFE mutation.1 However an Italian study demonstrated that the classical HFE gene mutations were only seen in a third of their population with iron overload.107
Liver biopsy has been the gold standard for the diagnosis of hepatic fibrosis and cirrhosis and for quantifying hepatic iron content. Liver biopsy should be considered in addition to genotyping if patients have at least one of the following features: ferritin level greater than 1000 μg/L, age greater than 40 years, hepatomegaly, or abnormal liver function tests.108 These parameters will identify the vast majority of patients with advanced fibrosis or cirrhosis. However, its clinical application in the diagnosis of haemochromatosis may be changing. There is an emerging role for the use of magnetic resonance imaging (MRI) as an accurate, non-invasive measurement of hepatic iron content and the possible detection of fibrosis. There is a high correlation between the mean liver proton transverse relaxation rates measured using MRI and the biochemical liver iron content. MRI provides a high degree of sensitivity and specificity for measurement of liver iron content with an area under the ROC curve greater than 0.98.109,110 Furthermore, MRI provided information regarding the stage of fibrosis. In a recent study of 18 HH subjects with high-grade and 42 HH subjects with low-grade fibrosis, MRI had a high sensitivity (100%) but low specificity (67%) in the diagnosis of high-grade fibrosis. However, the product of [HIC x age] had a sensitivity and specificity of 100% and 86%, respectively, for diagnosis of high-grade fibrosis and patients were accurately assigned to fibrosis groups via the use of MRI.111
Genetic testing for HH is accompanied by various ethical concerns including genetic testing in minors, newborn screening, non-paternity, discrimination by insurers and employers, and confidentiality.112 The focus of genetic counselling is to assist individuals and families understand and cope with a genetic disease. Relevant education and counselling has been shown to improve acceptance rates of genetic testing.113
The benefits of obtaining parental consent on behalf of their children for testing may be questionable in HH as therapeutic intervention is not required at a young age.114 Competent adolescents should have the privilege to assert a degree of autonomy in providing informed consent to have a genetic test. Two committees of the American Academy of Paediatrics have advised against genetic testing of potential carriers until over 18 years of age.115,116
Minimisation of the effects of genetic discrimination must be considered for any proposed screening program for HH.117 In particular, there are concerns regarding privacy infringements that could result in discrimination such as reluctance of health care carriers to pay for phlebotomy therapy for asymptomatic individuals, loss of health, life and disability insurance and unemployment. It is disturbing that Shaheen et al. was able to demonstrate that up to 20% of individuals with HH who had no end organ damage associated with iron overload reported difficulties in acquiring insurance.118 Legislation restricting misuse of genetic information has not been universally enacted. The Genetic Information Nondiscrimination Act of 2005 has not yet been approved by the US House of Representatives despite being passed by the US Senate in 2003.119 In Australia and several European countries health insurance is population-rated and discrimination is illegal. For life insurance, an agreement has been reached with the insurance industry that considerably reduces the risk of discrimination.120 There is a scale of legislation thus far endorsed ranging from moratoria regulating the use of genetic information and testing in insurance and employment to outright prohibition of any use of genetic information by insurers (Austria, Belgium, Denmark, Estonia, France, Luxembourg, and Norway).121 Policies toward providing appropriate directives to protect personal data derived from genetic testing are being developed and should be given priority before implementing population-based screening programs.122
Non-paternity concerns give rise to confidentiality implications and misinformation regarding genetic risk for family members of individuals having a genetic test for the HFE mutation. Unless determination of paternity is the purpose of the test the American Society of Human Genetics recommends against informing family members about misattribution of paternity.123
In terms of equitable access, it is unclear whether excluding ethnic groups of non-northern European background from genetic testing is discriminatory, despite apparent cost saving issues at a population screening level.124 The implications of this may be more significant in a multicultural population with ethnic mixing. At the individual level, those without adequate health insurance may be denied a rebate for genetic testing adding a potential barrier to widespread screening implementation.
Dramatic advances have occurred in the field of iron metabolism since the discovery of the HFE gene in 1996. New understanding of iron transport, storage and regulatory mechanisms through the discovery of new proteins and genetic defects in both animal models and humans reveals the heterogeneity of the conditions that cause iron overload. Although our understanding of these mechanisms is far from complete, correlation of functional changes in iron transport with genetic, transcriptional and translational events provides us with insights into the disorders of iron overload and the body’s complex iron homeostatic mechanisms. Long term follow up studies and large population studies have revealed the epidemiology and natural history of hereditary haemochromatosis whilst advances in diagnostic tests including gene testing and MRI provide new approaches to a more accurate and less invasive way of diagnosing this condition.
HH is a common disorder which can result in serious organ damage in a minority of subjects if left untreated. Genotypic screening followed by serum iron studies in the presence of C282Y homozygosity is mandatory in first-degree relatives following the diagnosis in a proband. Widespread population based screening is difficult to justify due to the relatively low phenotypic penetrance and low prevalence of mutations in certain ethnic groups. Targeted screening in ethnic groups at higher risk of mutations and phenotypic expression such as males of northern European origin may be reasonable.
Competing interests: None declared