Part IV: CASE STUDIES: Using Human Genome Epidemiology Information to Improve Health Chapter 28

Hereditary Hemochromatosis

Giuseppina Imperatore, Rodolfo Valdez, Wylie Burke

References

Background

Disease

Hereditary hemochromatosis (HHC, OMIM #235200) is an inherited disorder of iron metabolism characterized by an increased absorption of iron from the diet. Over time the excess iron accumulates in body tissues, a condition known as iron overload, and can lead to organ damage. Iron accumulation occurs primarily in the liver, pancreas, heart, joints, and pituitary gland. This may result in organ failure including liver cirrhosis, primary liver cancer, impotence, arthritis, diabetes, or cardiomyopathy. The disease onset is insidious and often characterized by common non-specific symptoms such as fatigue, arthralgia, and abdominal pain. For this reason, it can be undetected for a long time and it is usually diagnosed when advanced organ damage has already occurred. Two methods of screening for the detection of early stage of HHC are available: serum iron measures and molecular testing to detect mutations in the HHC gene, called HFE. Iron overload due to HHC can be detected before the appearance of organ damage. The most commonly used test to identify persons at risk of developing iron overload disease is the percent transferrin saturation (TS) (1). An elevated TS usually occurs well before HHC clinical symptoms. The first step for ascertaining HHC is measuring a non-fasting TS. If this results elevated (usually > 45%), the test should be repeated after an overnight fast (2). If fasting TS is also elevated then more tests need to be performed to check for the presence of increased iron stores. First, serum ferritin levels should be measured. Serum ferritin levels above 300 μg/L in men and post-menopausal women and ≥ 200 μg/L in pre-menopausal women indicate primary iron overload. Liver biopsy or quantitative phlebotomy confirms the diagnosis of HHC and quantifies the degree of iron overload (2). Treatment of iron overload consists in removing the excess iron through repeated phlebotomy, which improves survival in symptomatic persons (3-6). If phlebotomy is initiated before the development of cirrhosis, survival rate of individuals with HHC is similar to that of the general population (3,7,8).

Epidemiology

The lack of a standardized disease definition makes the estimate of the prevalence of HHC complicated. HHC can be defined by the presence of genetic mutations in the HFE gene, or biochemical markers of iron metabolism, or the presence of clinical symptoms. The genetic analysis identifies persons carrying one or two copies of the two known mutations in the HFE gene, C282Y and H63D. In the U.S., a recent population-based study estimated that among whites the frequency of HFE genotypes containing two mutations (C2982Y/C282Y, C282Y/H63D, and H63D/H63D) is about 5% (9). Among these genotypes, however, the highest risk for iron overload disease occurs with the C282Y homozygous genotype (10). The preponderance of clinically diagnosed HHC cases with C282Y/C282Y genotype, despite the fact that it is much rarer than other HFE genotypes, is evidence of the higher risk associated with this genotype. In the U.S., for example, among whites the prevalence of homozygosity for C282Y mutation is 0.30% (95% CI 0.12-0.82), about 1 in 333 individuals (9). Similar estimates were reported among members of a health maintenance organization where the prevalence of C282/C282Y genotype in whites was 0.4% (11). If we assume that about 81% of affected individuals in the U.S. are homozygous for C282Y (12), based on mutation analysis the estimate of HHC in the white population of the U.S. ranges between 37 in 10,000 (30/0.81) to 62 in 10,000 (50/0.81).

In population-based intervention trials, the estimated prevalence of homozygosity based on phenotype, defined as biochemical evidence of iron overload, is 50 per 10,000 (95% confidence interval 17-84) for men and 62 per 10,000 (95% confidence interval 27-97) for women (13). In primary care settings among whites, the estimated prevalence of clinically proven or liver biopsy proven HHC is 54 per 10,000 (14). A higher prevalence (80/10,000) was obtained in one study when elevated TS alone was used for defining HHC (15). This may simply reflect the fact that a significant proportion of unaffected or heterozygous individuals have TS levels above the cutoff, especially when TS thresholds of 50% are used (13). Lower estimates (3 to 19 per 10,000) are derived from autopsy studies and review of death records. In 1992, the HHC-associated mortality rate in the American population was reported at 1.8 deaths per million (16), far lower than the estimated prevalence of HHC. Similarly, a study using data from the National Hospital Discharge Survey from 1979 through 1997 estimated that the rate of hemochromatosis-associated hospitalizations was 2.3 per 100,000 persons in the United States (17). There are, therefore, fewer people requiring treatment or dying from HHC than is predicted by the frequency of the HHC mutations. This may reflect the fact that the disease is under-diagnosed, or that the penetrance (the likelihood that a person carrying a given genotype will develop clinical disease) of the genotype is low, or both.

Genetics

More than twenty years ago, Simon et al. (18) described HHC as an autosomal recessive disorder linked to the HLA-A3 complex on the short arm of chromosome 6. In 1996, Feder et al. mapped the HFE gene on the short arm of chromosome 6 (6p21.3) and described two missense mutations of this gene (C282Y and H63D) that accounted for the majority of HHC patients in their study (19).

HHC due to mutations of the HFE gene occurs commonly among whites, especially those of northern European descent (20). HFE mutations so far identified, however, do not account for all cases of hemochromatosis (12). For example, in Southern Europe, non-HFE related iron overload disorders have been described due to mutations of the transferrin receptor 2 (TFR2) and the ferroportion gene (SLC11A3) (21-22). Therefore, the genetics of HHC is complex.

Allelic Variants

The HFE gene codes for a 343 residue type I transmembrane protein that associates with class I light chain beta2-microglobulin (19). This protein binds to the transferrin receptor and reduces its affinity for iron-loaded transferrin by 5- to 10-fold (23). The localization of the HFE protein in the crypt cells of the duodenum (the site of dietary iron absorption) and its association with the transferrin receptor in those cells are consistent with a role in regulating iron absorption (24-25). The observation that HFE-deficient mice (HFE gene knockout model) develop iron overload similar to that seen in human HHC provides further evidence that the HFE protein is involved in r iron homeostasis (26). The C282Y mutation results from a G-to-A transition at nucleotide 845 of the HFE gene (845G->A) that produces a substitution of cysteine for a tyrosine at the amino acid position 282 in the protein product. This substitution alters the HFE protein structure and beta2-microglobulin association, disrupting its transport to and presentation on the cell surface (27). In the H63D mutation, a G replaces C at nucleotide 187 of the gene (187C->G), causing aspartate to substitute for histidine at the amino acid position 63 in the HFE protein. The H63D mutation does not seem to prevent beta2-microglobulin association or cell surface expression (24), indicating that the C282Y mutation results in a greater loss of protein function than does H63D (28).
In addition to C282Y and H63D, nine other missense mutations causing amino acid substitutions have been documented. In one, a substitution of a cysteine for serine at the amino acid position 65 (S65C) has been implicated in a mild form of HHC (29). A number of intronic polymorphisms have also been found (30). One polymorphism occurs within the intron 4 (5569G-A) of the HFE gene in the binding region of the primer originally described by Feder et al. (19). One laboratory reported that when a polymerase chain reaction (PCR)-based restriction endonuclease digestion assay is used, the presence of this polymorphism might cause C282Y heterozygosity to be misdiagnosed as C282Y homozygosity (31-32). However, three groups could not replicate this finding (33-34). Beutler et al. reported that a mutation in intron 3 (IVS3-48c) can also lead to misdiagnose heterozygotes for the C282Y mutation as homozygotes (35).

Genotype Prevalence

A number of studies have reported on both the general population and the probands frequencies of the HFE genotype. Recently, a review has summarized the results of these studies according to the geographic origin of the populations studied (12). In the general population, a total of 6,203 samples from European countries revealed on average a C282Y homozygous and heterozygous prevalence of 0.4 percent and 9.2 percent, respectively. However, C282Y homozygosity has not been reported in the general population of Southern or Eastern Europe. The frequency of the C282Y heterozygosity varies from 1 to 3 percent in Southern and Eastern Europe to as high as 24.8 percent in Ireland. In North America, among 3,752 samples the HFE genotype distribution had a similar pattern: C282Y/C282Y genotype was the rarest with a frequency of 0.5 percent and C282Y heterozygosity was present in 9.0 percent of the samples. In the Asian, Indian subcontinent, African/Middle Eastern, and Australasian populations, C282Y homozygotes were not found and the frequency of C282Y heterozygosity was very low (range: 0 to 0.5 percent). C282Y/H63D compound and H63D homozygosity each accounted for 2 percent of the European general population and 2.5 percent and 2.1 percent in the American populations, respectively. The heterozygous frequency of the H63D mutation was 22 percent in Europe and 23 percent in North America.

Hanson et al. (12) recently reviewed 17 studies reporting the frequency of the HFE genotypes among patients with clinically diagnosed HHC. Most of the studies used case definitions that included diagnostic evidence of iron overload from either liver biopsy or quantitative phlebotomy. The exceptions were a French (29) and a U.S. study (36), which used a case definition of persistently elevated TS or elevated serum ferritin. In all case series, the majority of patients had the homozygous C282Y genotype. However, there was some variability across studies. For example, among 2,229 European HHC patients, the estimated prevalence of homozygosity for the C282Y genotype ranged from 52 percent (37) to 96 percent (38). In North America, among 588 patients the C282Y homozygosity ranged between 67 to 95 percent. Heterozygosity for the H63D mutation and compound heterozygosity (C282Y/H63D) each accounted for 6 percent of European cases and 4 percent of cases in North America. Overall, 3.6 percent (95 percent CI: 2.9, 4.3) of the patients had the C282Y/wild genotype, and 1.5 percent (95 percent CI: 1.1, 2.1) had the H63D/H63D genotype. Worldwide, among 2,929 patients 6.9 percent (95 percent CI: 6.0, 7.9) were homozygous for the wild allele. These findings suggest that non?genetic influences, additional HFE mutations; or variation at additional genes affecting iron metabolism, as recently reported, may also cause or modulate iron overload (21,22,39).

Gene-gene and Gene-environment Interactions

The clinical expression of HHC is influenced by a variety of factors, both genetic and environmental. In HFE knockout mice, mutations of other genes involved in iron metabolism, such as beta2-microglobulin, transferrin receptor, and transmembrane iron import molecule (DTM1), strongly modify the amount of liver iron (40). It is, therefore, conceivable, that similar gene-gene interactions may influence the course of HHC in humans. The finding of the C282Y heterozygote genotype among some persons classified as being affected with HHC is also suggestive of the influence of other yet-to-be-identified HFE mutations; or of the combination of the C282Y heterozygous state with environmental modifying factors (e.g., high iron intake, viral hepatitis or alcohol abuse); or of a second genetic disorder (e.g., beta-thalassemia trait, iron loading anemia) that could account for clinical disease (41-44).

There is also evidence that sex plays a primary role in the clinical manifestation of HHC. Family studies based on HLA-typing indicate that the frequency of affected brothers and sisters is similar, as expected for an autosomal recessive disorder, but the proportion of females among probands diagnosed on the basis of clinical symptoms is 11 to 35 percent, rather than the expected 50 percent (3,4,45). In a large study, the prevalence of iron overload, as determined by liver biopsy or phlebotomy, was twice as frequent in males as females (46). This sex difference has been attributed to the lower degree of iron overload in women because of menstruation, pregnancy and lactation.

Other possible modifiers include chronic blood loss (gastrointestinal bleeding, regular hematuria, helminthic or other parasitic infections) and regular blood donation, alcohol abuse, excessive iron intake, or vitamin C intake. Tannates, phytates, oxalates, calcium and phosphates also modify HHC because they are known to bind iron and inhibit iron absorption (47). Chronic viral hepatitis B and C and metals such as zinc and cobalt may also influence expression of HHC (47–48). Iron modulates the course of hepatitis B (57), and iron reduction has been shown to decrease the severity of chronic hepatitis C while increasing the likelihood of response to antiviral therapy. Hepatitis C virus infection and HFE mutations have also been identified as risk factors for porphyria cutanea tarda (49).

Laboratory Tests

Serum tests for iron status

The value of serum iron measures or HFE mutation analysis in screening for individuals at high risk for developing serious clinical manifestations of HHC is difficult to assess because of uncertainties about the natural history of the disease. Thus, the phenotype of interest must be specified before assessing the validity of each test for screening. For example, the phenotype might be defined by biochemical evidence of iron overload (e.g., hepatic iron index >1.9 or removal of more than 4 grams of iron by quantitative phlebotomy), or by clinical symptoms compatible with iron overload in combination with biochemical evidence of iron overload, or by evidence of serious end-stage organ disease in combination with biochemical evidence of iron overload.

The marker for serum iron status most used is percent transferrin saturation (TS). This test can be used as a phenotypic screening test to identify persons with biochemical evidence of iron overload. The cutoff TS values recommended for screening have varied from 45 to 70 percent (1,14,15,50). Using data collected in family studies and screening trials, the performance of TS as a screening test (e.g., detection rate, false positive rate, and positive and negative predictive values) has been estimated for different TS cutoff levels. For example, based on published parameters, screening at a TS cutoff level of 50% would identify about 94% and 82% of men and women with HHC, respectively, along with a number of false positives (about 6% of males and 3% of females screened). Assuming an HHC genotype prevalence of about 50 in 10,000, the odds of being affected given a positive result (OAPR) would be about 1 to 12 for males and 1 to 8 for females, corresponding to positive predictive values of 8 percent and 11 percent, respectively. Diagnostic testing (e.g., liver biopsy or quantitative phlebotomy) is recommended for persons with a positive screening result (either a single test result or persistently elevated TS) and no other identifiable explanation for increased body iron stores (e.g., chronic anemias, liver disease related to alcohol abuse or hepatitis). In persons diagnosed to have iron overload related to HHC by such a screening and diagnostic process, the probability of developing at least one clinical symptom can be estimated from family studies and screening trials to be about 50-70% for males and 40-50% for females. It is worth noting that most complications recorded in such studies were common and nonspecific clinical manifestations of the disease such as joint pain and diabetes. In the absence of control groups, the proportion of complications attributable to HHC is difficult to determine; as a result, the probability of developing clinical complications may be considerably lower.

Penetrance appears to be consistently lower in women than in men at all ages. However, as many as 40% of genetically susceptible younger individuals of both sexes do develop biochemical evidence of iron overload; many also have non-specific symptoms compatible with early iron overload. A smaller proportion, not well defined, may develop serious complications such as diabetes, cirrhosis or cardiomyopathy.

Ferritin is an intracellular iron storage protein and serum ferritin (SF) concentration significantly correlates with body iron stores (1ng/mL = 10 mg of stored iron). SF values, but not TS values, are associated with HHC clinical signs, and SF concentrations are higher for those with clinical manifestations (13). SF has been used as a second screening test in many trials, and it can be very effective in reducing the number of false positives (46), if cutoffs appropriate for age and sex are used. Elevation of the SF concentration in HHC must be differentiated, however, from other liver disorders such as alcoholic liver disease, chronic viral hepatitis, and nonalcoholic steatohepatitis. Serum ferritin is also an acute phase reactant and can be elevated as a result of inflammatory conditions.

HFE Gene Mutation Analysis

HFE mutation analysis identifies persons carrying one or two copies of either of the two known mutations, C282Y and H63D. Since the majority of clinically diagnosed probands are homozygous for C282Y, individuals with this genotype are considered to be at the highest risk for iron overload disease. However, approximately 20% of HHC cases occur in persons with other HFE genotypes, and as many as 7% have no identifiable mutation (12). The penetrance of the different HFE genotypes – that is, the likelihood that persons carrying a given HFE genotype will develop manifestations of iron overload – can only be roughly estimated from published data. The data suggest that a large proportion of individuals with the C282Y homozygous genotype will develop biochemical evidence of iron overload during their lifetime; we can only speculate how many will develop clinical symptoms related to iron overload (perhaps about half), or who will die from complications of iron overload (likely to be a small proportion).

A screening study at a health maintenance organization in southern California represents the only controlled study to evaluate penetrance of the C282Y/C282Y genotype (11). The study included 41,038 adults attending a health appraisal clinic (a clinic providing assessment of health status and prevention options, attended voluntarily) with a mean age of 57 years, of whom 152 subjects (0.4%) had the C282Y/C282Y genotype. Of these, 45 had previously been diagnosed with HHC (30%); for most, the diagnosis had been made on the basis of screening. The study evaluated 124 subjects with the C282Y/C282Y genotype, including all those not previously diagnosed with HHC and 17 for whom data were available prior to diagnosis. TS was elevated in 75% of men and 40% of women and serum ferritin was elevated in 76% of men and 54% of women with the C282Y/C282Y genotype. Compared with control subjects (22 394 white and Hispanic participants on whom questionnaire data were available and who did not have any HFE mutations), persons homozygous for C282Y were more likely to have a history of a“liver problem or hepatitis” (8% vs. 4%), elevated serum aspartate aminotransferase (8% vs. 4%), and elevated plasma collagen IV, a measure of mild liver fibrosis (26 % vs. 11%), but were no more likely to have a history of fatigue, joint pain, impotence, skin pigmentation, or diabetes. Among the full cohort of 152 subjects with the C282Y/C282Y genotype, only one, an alcoholic, had a clinical history of end-stage HHC. Two others, out of 119 with complete data, had markedly abnormal laboratory values suggestive of severe liver fibrosis. On the basis of these data, the authors concluded that the likelihood of significant clinical disease in persons with the C282Y/C282Y genotype was 1%.

This study has a potential selection bias that could have resulted in an under-estimate of penetrance: subjects were drawn from a preventive care setting, potentially selecting against patients with clinical disease. The limited clinical findings among the large group of subjects with the C282Y/C282Y genotype argues for low penetrance of the genotype, even if it ultimately proves to be above 1%. In keeping with this conclusion, the study found the prevalence of the C282Y/C282Y genotype was the same among older and younger subjects (11); high penetrance would be expected to result in premature mortality for some people with the genotype, resulting in a lower prevalence of the genotype at older ages. The penetrance of all other HFE genotypes is estimated to be much lower than that of C282Y/C282Y (10).

Implications of Genetic Testing

Screening for HHC using HFE mutation analysis could involve testing for both HFE mutations, or only for C282Y. If both mutations are tested, about 5-6 percent of persons of northern European descent will have a test result indicating the presence of two HFE mutations. However, about 0.5 percent of the general screened population with the C282Y homozygous genotype will be at high risk of iron overload. Another 2 percent of the general population will have the compound heterozygous (C282Y/H63D) genotype, but only about 1 in a 100 of these persons would be expected to develop significant iron loading. If testing is limited to C282Y, about 10 percent of the northern European population would be identified as heterozygote, but only 0.5 percent of the population homozygous for C282Y would be at high risk for iron overload. Either approach would identify the majority of persons at risk for hemochromatosis, though the risk for some persons would be low and difficult to quantify. In other populations – eg, southern European – this screening approach may identify a smaller proportion of persons at risk. For either approach, costs and sequelae of screening are influenced by decisions concerning the provision of counseling and/or clinical follow-up. For example, the follow-up offered to all persons with the C282Y homozygous genotype should include counseling about the uncertainty of their prognosis, and the possibility that risk of clinical complications is low. For persons with other genotypes, risk of iron overload disease is known to be low, and appropriate counseling and follow-up have not been established. Decisions concerning the counseling needs of C282Y heterozygotes would have an important effect on the cost and outcome of a screening program, since these persons constitute a substantial proportion of the population (about 9 percent in populations of northern European descent). Clinical follow-up or counseling to address their potential risk for iron overload related to alcohol abuse or other risk factors, as well as the potential risk to family members, would be costly. In addition, there is currently no data to assess the value of such intervention.
Family-based detection represents an important alternative approach to identifying people with iron overload. When a diagnosis of HHC is made, it also identifies family members who represent a group with a markedly higher a priori risk of iron overload disease than the general population. Therefore, it is reasonable to consider assessment of iron status in relatives and to monitor them for symptoms suggestive of iron overload.

HFE genotyping provides a one-time test to determine which relatives of an identified proband have an increased risk of iron overload. These relatives can be offered ongoing surveillance, while others can be reassured. However, genotyping may also cause confusion about clinical status and adverse labeling, so the value of genotyping as a method for family-based detection of HHC is not entirely clear. Siblings of an affected person with the homozygous C282Y genotype have a 25% chance of sharing the same high risk genotype; for siblings who do not share the genotype, this single test can greatly reduce the risk.

However, HHC has occurred in some people with other HFE genotypes (e.g., C282Y/H63D, C282Y/+) (12), suggesting the need for caution in the interpretation of a “negative” test result. But even the implications of a “positive” result are not straightforward; current penetrance data make risk of disease hard to calculate even for relatives with a C282Y/C282Y genotype, and argue against making a diagnosis of HHC on the basis of genotype alone. In the uncommon instance of a proband with a different HFE genotype, genotypic studies of relatives are even more difficult to assess, given the very low penetrance of genotypes other than C282Y/C282Y.

Testing of offspring raises even more questions, because of the high carrier rate for HFE mutations (e.g., 9% for C282Y, 23% for H63D in populations of European descent) (12). If the parent with HHC is a C282Y homozygote, offspring have a 4.5% likelihood of inheriting the same genotype (calculated as: 100% chance of inheriting the C282Y allele from the affected parent x 9% chance that the other parent is a C282Y carrier x 50% chance of inheriting C282Y from the unaffected parent) and an 11.5% chance of inheriting a C282Y/H63D genotype. All other offspring will be C282Y carriers. Because disease occurs in middle age, there is no rationale for testing during childhood.

Genotype testing does not substitute for the serum iron studies needed to identify iron overload and it could expose the family member to a premature diagnosis, unnecessary treatment, and the potential for stigma and discrimination. These considerations underscore the need for more information about the clinical penetrance of HFE genotypes in HHC, and also about effective ways to counsel patients after genetic testing to ensure an accurate understanding of the results.

Potential Benefits and Harms Associated with Genetic Testing for HFE Mutations.

There are important potential benefits from early detection of affected HHC persons, including prevention of significant morbidity and mortality and long-term reduction in health care costs for those who would otherwise suffer from serious medical complications of hemochromatosis. TS screening has been used successfully in pilot studies, suggesting that this is a feasible screening approach. At the same time, universal screening for HHC would expose a large number of persons to the possibility of adverse psychological, social or economic consequences related to a diagnosis of HHC. As reviewed in previous sections, a majority of those identified through screening might remain healthy without treatment. The potential for loss of insurance or employment after a genetic diagnosis is a concern for consumers and policymakers (51-54). Legislative efforts to minimize such loss are being implemented (53), but the degree of protection they will provide is unknown. Although adverse outcomes after a diagnosis of HHC have been reported (55), no systematic study has been undertaken to further assess these outcomes. A genetic diagnosis may be stigmatizing, and has the additional effect of identifying a potential risk for family members. (56). The psychological burdens of a diagnosis of HHC may be reduced by counseling that the diagnosis does not imply a certainty of future disease, and that effective treatment is available to reduce future risk. Whether communication of this kind can change the stigmatizing potential of a diagnosis of HHC, or reduce the likelihood of discrimination, remains to be determined. These issues are not substantially different from those identified for other types of genetic testing (e.g., cystic fibrosis, cancer markers), and may influence decisions about the need for counseling procedures as part of a screening program. As with other aspects of HHC screening, judgments about the relevance of these issues to decisions about HHC screening must be made in the absence of definitive data.

Conclusions

Iron overload can be treated or prevented by phlebotomy, but treatment is often delayed, resulting in irreversible organ damage. Greater physician awareness of HHC may help reduce the morbidity and mortality of primary iron overload. HHC is usually diagnosed after a delay of several years, during this period care has been sought for the early non-specific symptoms of the disorder (57). Some persons with HHC are not diagnosed until life-threatening complications are present, e.g., diagnoses have occurred after a liver transplant for end stage cirrhosis (58). The delay in diagnosis of hemochromatosis suggests that physicians lack awareness of this disorder, or have a low index of suspicion when symptoms compatible with the early stages of the disease are present and even sometimes when late complications are present. With early diagnosis, preventive therapy can be instituted in the form of regular phlebotomy. If treatment is begun before cirrhosis or diabetes has occurred, the prognosis is good. However, late and missed diagnoses lead to under utilization of this readily accessible preventive treatment.

A number of questions remain about the benefits and risks of identifying and treating asymptomatic persons at high risk for HHC. “Universal” or “population-based” screening refers to screening performed across an entire population of mainly asymptomatic individuals not referred for testing due to symptoms of the disease. This could be accomplished through public health screening programs, or as part of routine testing within primary health care settings. This should be clearly distinguished from the alternative approach of “case finding” or “enhanced case detection”, which could include iron status testing and/or HFE mutation analysis in targeted populations, such as persons at increased risk due to an affected family member or persons who present with clinical complaints consistent with a diagnosis of HHC. Generally accepted criteria for an effective population-based screening test include the following:

• The disorder screened for must be well-defined and represent an important health problem. The natural history of the disorder should be understood.
  This is, in fact, a key question. Is the disorder for which we plan to screen hereditary hemochromatosis (HHC) or iron overload? HHC, as defined by the HFE genotype, better fits the criterion of a “well-defined” disorder. HHC is a serious, treatable disorder with life-threatening complications. However, important questions about natural history, particularly age-related penetrance, remain and recent data raise the possibility that very few people with the genotype will develop clinically significant disease. Iron overload, as defined by persistently elevated TS and diagnostic evidence of increased body iron stores, however, encompasses a broad range of disorders, and requires a complex diagnostic protocol to differentiate HHC from other acquired (e.g., chronic anemias, alcoholic liver disease) and inherited causes (e.g., juvenile hemochromatosis, non-HFE hemochromatosis). Because the natural history of hemochromatosis has not been systematically studied, questions remain concerning the persons most likely to benefit from early treatment.
• The prevalence must be known and the disorder common enough that population-based screening will be cost-effective.
  Based on evidence from family studies, screening trials and mutation analysis, the prevalence of HHC, as defined by genotype, can be estimated to be about 50 to 60 per 10,000; lower estimates (3 to 19 per 10,000) are derived from autopsy studies and review of death records. Estimates of phenotype prevalence in screening trials, as defined by diagnostic testing (e.g., liver biopsy or quantitative phlebotomy) for iron overload, range from about 26 to more than 50 per 10,000. The prevalence of life-threatening complications due to HHC is not well established and may be much lower.
• Suitable screening test(s) with known performance (detection rate, false positive rate, OAPR) must be available.
  As discussed before, two well-characterized screening tests are available, TS and HFE mutation analysis. Both tests are readily available (though there are licensing issues with regard to HFE testing in the US). While consensus may have to be reached on the distribution of TS measurements in homozygous, heterozygous and unaffected individuals, enough information is available to allow reasonably accurate prediction of detection rates and false positive rates using different cutoff levels of TS (alone or in conjunction with serum ferritin). A consistent set of data is available with regard to expected distributions of HFE genotype frequencies among patients and in the general population. However, screening performance cannot be fully assessed due to uncertainties about the age-related penetrance of HHC.
• For individuals identified as screen positive, there must be an adequate and acceptable protocol for diagnosis and an effective treatment.
  Protocols for diagnosis and treatment of HHC are available. However, limited data are available on the outcome of treatment in persons with HHC who are asymptomatic at the time of diagnosis.
• Adequate facilities must be available to support screening, diagnosis and treatment.
  It is likely that laboratories capable of supporting screening and diagnostic testing methodologies either exist now, or could be developed within a reasonable time frame. However, a number of logistical issues need to be considered further, including: the potential burden of time and cost to health care providers to educate patients about HHC, offer testing, and follow up on positive screens; the anticipated level of patient compliance with screening, diagnosis, and treatment; proposed mechanisms of long-term follow-up and maintenance of iron status or genotype information in patients’ medical records.
• Costs have been examined and are reasonable.
  TS-based screening is likely to be cost-effective even given unfavorable assumptions (e.g., low prevalence or rate of progression to serious disease, low compliance). Screening for the C282Y mutation may be cost-effective due to a lower positive rate than TS screening, with corresponding reduced costs and burden of intervention. However, published studies to date have not considered all costs of screening, both medical and societal; in particular, the physician and health system effort required for long-term follow-up and the personal psychological and economic consequences of screening have not been evaluated (47,50,59,60).
• The approach must be considered ethical and must be acceptable to health care providers and consumers.
  Potential benefits from early detection of affected persons are significant, including prevention of significant morbidity and mortality and long-term reduction in health care costs. TS screening has been used successfully in pilot studies. Concern has been raised about potential harms of screening, including psychological morbidity, stigmatization and discrimination (e.g., insurance or employment) resulting from a diagnosis of HHC, particularly if the risk of clinical complications is low among those diagnosed by screening.

Health policies regarding use of genetic information

At this time genetic testing for HFE mutations is not recommended for population-based screening for HHC (61,62) due to the uncertainty about disease prevalence and penetrance, the optimal care for asymptomatic persons found to carry HFE mutations, and the psychosocial impact of genetic testing. Mutational analysis of the HFE gene may be useful in confirming the diagnosis of HHC in persons with elevated iron measures or for identifying relatives at risk to develop HHC of patients with HHC due to the known mutations

Agenda for future HuGE research

The discovery of the HFE gene represents an important step in understanding the nature of HHC. However, much remains to be learned about this disorder. It is crucial to know the age- and sex-specific penetrance of the HFE mutations. Future efforts should also be directed towards the identification of environmental modifiers and assessment of their interactions with HFE genotypes. Moreover, as more and more genes involved in iron metabolism are being discovered, attempts should be made to understand the complex interplay of the HFE genotype and these genes. Future studies should be performed to assess the effectiveness of interventions to reduce disease burden in asymptotic individuals carrying susceptibility genotypes. In addition, more information is needed regarding the social, ethical, and psychological outcomes of genetic screening for HHC.

References

1. Witte DL, Crosby WH, Edwards CQ, et al. Practice guideline development task force of the College of American Pathologists. Hereditary hemochromatosis. Clin Chim Acta 1996;245:139-200.
2. Powell WL, George KD, McDonnell SM, et al. Diagnosis of hemochromatosis. Ann Intern Med 1998;129:925-93.
3. Niederau C, Fischer R, Purschel A, et al. Long-term survival in patients with hereditary hemochromatosis. Gastroenterology 1996;110:1107-19.
4. Niederau C, Fischer R, Sonnenberg A, et al. Survival and causes of death in cirrhotic and in noncirrhotic patients with primary hemochromatosis. N Engl J Med 1985;313:1256-62.
5. Bomford A, Williams R. Long term results of venesection therapy in idiopathic haemochromatosis. Q J Med 1976;45:611-23.
6. Powell LW, Kerr JF. Reversal of “cirrhosis” in idiopathic haemochromatosis following long- term intensive venesection therapy. Australas Ann Med 1970;19:54-7.
7. Adams PC, Speechley M, Kertesz AE. Long-term survival analysis in hereditary hemochromatosis. Gastroenterology 1991;101:368-72.
8. Fargion S, Mandelli C, Piperno A, et al. Survival and prognostic factors in 212 Italian patients with genetic hemochromatosis. Hepatology 1992;15:655-9.
9. Steinberg KK, Cogswell ME, Chang JC, et al. Prevalence of C282Y and H63D mutations in the hemochromatosis (HFE) gene in the United States. JAMA 2001:285;2216-2222.
10. Burke W, Imperatore G, McDonnell SM, et al. Contribution of different genotypes to iron overload disease: a pooled analysis. Genet Med 2000;2:271-7.
11. Beutler E, Felitti V, Koziol J, et al. Penetrance of the 845GàA (C282Y) HFE hereditary haemochromatosis mutation in the USA. Lancet 2002; 359:211-18.
12. Hanson EH, Imperatore G, Burke W. HFE gene and hereditary hemochromatosis: a HuGE Review. Am J Epidemiol 2001;154-193-206.
13. Bradley LA, Haddow JE, Palomaki GE. Population screening for haemochromatosis: expectations based on a study of relatives of symptomatic probands. J Med Screen 1996;3:171-7.
14. Phatak PD, Sham RL, Raubertas RF, et al. Prevalence of hereditary hemochromatosis in 16,031 primary care patients. Ann Intern Med 1998;129:954-61.
15. McDonnell SM, Hover A, Gloe D, et al. Population-based screening for hemochromatosis using phenotypic and DNA testing among employees of health maintenance organizations in Springfield, Missouri. Amer J Med 1999;107:30-7.
16. Yang Q, McDonnell SM, Khoury MJ, et al. Hemochromatosis-associated mortality in the United States from 1979 to 1992: an analysis of multiple-cause mortality data. Ann Intern Med 1998;129:946-53.
17. Brown AS, Gwinn M, Cogswell ME, et al. Hemochromatosis-associated morbidity in the United States: a analysis of the National Hospital Discharge Survey, 1979-1997. Genet Med 2001;3:109-111.
18. Simon M, Bourel M, Fauchet R, et al. HLA and “non-immunological” disease: idiopathic haemochromatosis [letter]. Lancet 1976;2:973-4.
19. Feder JN, Gnirke A, Thomas W, et al. A novel MHC class I-like gene is mutated in patients with hereditary haemochromatosis. Nat Genet 1996;13:399-408.
20. Merryweather-Clarke AT, Pointon JJ, Jouanolle AM, et al. Geography of HFE C282Y and H63D Mutations. Genetic Testing 2000;4:183-198.
21. Camaschella C, Roetto A, Cali A, et al. The gene TFR2 is mutated in a new type of haemochromatosis mapping to 7q22 Nat Genet 2000; 25: 14-15.
22. Montosi G, Donovan A, Totaro A, et al. Autosomal-dominant hemochromatosis is associated with a mutation in the ferroportion (SLC11A3) gene. J Clin Invest 2001;108:619-623.
23. Feder JN, Penny DM, Irrinki A, et al. The hemochromatosis gene product complexes with the transferrin receptor and lowers its affinity for ligand binding. Proc Natl Acad Sci U S A 1998;95:1472-7.
24. Waheed A, Parkkila S, Saarnio J, et al. Association of HFE protein with transferrin receptor in crypt enterocytes of human duodenum. Proc Natl Acad Sci U S A 1999;96:1579-84.
25. Parkilla S, Waheed A, Britton RS, et al. Immunohistochemistry of HLA-H, the protein defective in patients with hemochromatosis, reveals unique pattern of expression in hereditary hemochromatosis. Proc Natl Acad Sci USA 1997;94:2534-39.
26. Zhou XY, Tomatsu S, Fleming RE, et al. HFE gene knockout produces mouse model of hereditary hemochromatosis. Proc Natl Acad Sci U S A 1998;95:2492-7.
27. Lebron JA, Bennet MJ, Vaughn DE, et al. Crystal structure of the hemochromatosis protein HFE and characterization of its interaction with transferrin receptor. Cell 1998;93:111-23.
28. Feder JN, Tsuchihashi Z, Irrinki A, et al. The hemochromatosis founder mutation in HLA-H disrupts beta2- microglobulin interaction and cell surface expression. J Biol Chem 1997;272:14025-8.
29. Mura C, Raguenes O. HFE mutations analysis in 711 hemochromatosis probands: Evidence for S65C implication in mild form of hemochromatosis. Blood 1999;93:2502-5.
30. Pointon JJ, Wallace D, Merryweather-Clarke AT, et al. Uncommon mutations and polymorphisms in the hemochromatosis gene. Genetic Testing 2000;4:151-61.
31. Jeffrey GP, Chakrabarti S, Hegele RA, et al. Polymorphism in intron 4 of HFE may cause overestimation of C282Y homozygote prevalence in haemochromatosis. Nat Genet 1999;22:325-6.
32. Sommerville MJ, Sprysak KA, Hicks M, et al. An HFE intronic variant promotes misdiagnosis of hereditary hemochromatosis [letter]. Am J Hum Genet 1999;65:924-6.
33. Noll WW, Belloni DR, Stenzel TT, et al. Polymorphism in intron 4 of HFE does not compromise haemochromatosis mutation results [letter; comment]. Nat Genet 1999;23:271-2.
34. Elsea SH, Leykam V. HFE polymorphism and accurate diagnosis of C282Y hereditary hemochromatosis carriers [letter]. Blood 2000;95:2453-5.
35. Beutler E, Gelbart T. A common intron 3 mutation (IVS3-48c–>g) leads to misdiagnosis of the c.845G–>A (C282Y) HFE gene mutation. Blood Cells Mol Dis 2000;26:229-233.
36. Barton JC, Shih WW, Sawada-Hirai R, et al. Genetic and clinical description of hemochromatosis probands and heterozygotes: evidence that multiple genes linked to the major histocompatibility complex are responsible for hemochromatosis. Blood Cells Mol Dis 1997;23:135-45.
37. Moirand R, Jouanolle AM, Brissot P, et al. Phenotypic expression of HFE mutations: a French study of 1110 unrelated iron-overloaded patients and relatives. Gastroenterology 1999;116:372-7.
38. Brissot P, Moirand R, Jouanolle AM, et al. A genotypic study of 217 unrelated probands diagnosed as “genetic hemochromatosis” on “classical” phenotypic criteria. J Hepatol 1999;30:588-93.
39. Whitfield JB, Cullen LM, Jazwinska EC, et al. Effects of HFE C282Y and H63D polymorphisms and polygenic background on iron stores in a large community sample of twins. Am J Hum Genet 2000;66:1246-58.
40. Levy JE, Montross LK, Andrews NC. Genes that modify the hemochromatosis phenotype in mice. J Clin Invest 2000;105:1209-16.
41. Bulaj ZJ, Griffen LM, Jorde LB, et al. Clinical and biochemical abnormalities in people heterozygous for hemochromatosis. NEJM 1996;335:1799-1805.
42. Edwards CQ, Skolnick MH, Kushner JP. Coincidental non-transfusional iron overload and thalassemia minor: association with HLA-linked hemochromatosis. Blood 1981,58:844-848.
43. Edwards CQ, Griffen LM, Goldgar DE, et al. HLA-linked hemochromatosis alleles in sporadic porphyria cutanea tarda. Gastroenterology 1989;97: 972-981.
44. Piperno A, Sampietro M, Pietrangelo A, et al. Heterogeneity of hemochromatosis in Italy. Gastroenterology 1998; 114:996-1002.
45. Adams PC, Deugnier Y, Moirand R, et al. The relationship between iron overload, clinical symptoms, and age in 410 patients with genetic hemochromatosis. Hepatology 1997;25:162-6.
46. Niederau C, Niederau CM, Lange S, et al. Screening for hemochromatosis and iron deficiency in employees and primary care patients in Western Germany. Ann Intern Med 1998;128:337-45.
47. Barton JC, McDonnell SM, Adams PC, et al. Management of hemochromatosis. Hemochromatosis Management Working Group. Ann Intern Med 1998;129:932-9.
48. Burke W, Press N, McDonnell SM. Hemochromatosis: genetics helps to define a multifactorial disease. Clin Genet 1998;54:1-9.
49. Bonkovsky HL, Poh-Fitzpatrick M, Pimstone N, et al. Porphyria cutanea tarda, hepatitis C, and HFE gene mutations in North America. Hepatology 1998;27:1661-9.
50. Adams PC, Gregor JC, Kertesz AE, et al. Screening blood donors for hereditary hemochromatosis: decision analysis model based on a 30-year database. Gastroenterology 1995;109:177-88.
51. Hudson KL, Rothenberg KH, Andrews LB, et al. Genetic discrimination and health insurance: an urgent need for reform. Science 1995; 270:391-393.
52. Lapham EV, Kozma C, Weiss JO. Genetic discrimination: perspectives of consumers. Science 1996; 274: 621-624.
53. Rothenberg K, Fuller B, Rothstien M et al. Genetic information and the workplace: legislative approaches and policy challenges. Science 1997: 275: 1755-1757.
54. Holtzman NA, Watson MS eds.Task Force on Genetic Testing. Promoting Safe and Effective Genetic Testing in the United States: Final Report. Johns Hopkins University Press: Baltimore MD, 1998.
55. Alper JS, Geller LN, Barash CI et al. Genetic discrimination and screening for hemochromatosis. J Public Health Policy 1994;15: 345-358.
56. Markel H. The stigma of disease: implications of genetic screening. Am J Med 1992; 93: 209-215.
57. McDonnell SM, Preston BL, Jewell SA, et al. A survey of 2,851 patients with hemochromatosis: symptoms and response to treatment. Am J Med 1999;106:619-24.
58. Kowdley KV, Trainer TD, Saltzman JR, et al. Utility of hepatic iron index in American patients with hereditary hemochromatosis: a multicenter study. Gastroenterology 1997;113:1270-1277.
59. Balan V, Baldus W, Fairbanks V, et al. Screening for hemochromatosis: a cost-effectiveness study based on 12,258 patients. Gastroenterology 1994;107:453-9.
60. Phatak PD, Guzman G, Woll JE, et al. Cost-effectiveness of screening for hereditary hemochromatosis. Arch Intern Med 1994;154:769-76.
61. Burke W, Thomson E, Khoury M, et al. Hereditary hemochromatosis: gene discovery and its implications for population-based screening. JAMA 1998;280:172-178.
62. EALS International Consensus Conference on Haemochromatosis. Jury Document. J Hepatol 2000;33:496-504

Address correspondence to Dr. Khoury at
Office of Genomics and Disease Prevention
Centers for Disease Control and Prevention
6 Executive Park, Mail Stop E-82
Atlanta, Georgia 30329