Natural History and Management of HFE-hemochromatosis

Eng K. Gan, M.B.B.S.; Lawrie W. Powell, M.B.B.S., M.D., Ph.D., F.R.A.C.P., F.R.C.P.; John K. Olynyk, B.Med.Sc., M.B.B.S., M.D., F.R.A.C.P.


Semin Liver Dis. 2011;31(3):293-301. 

In This Article

What is the Natural History of HH?

The relationship between iron overload and liver cirrhosis was first postulated by Trousseau in 1865.[6] The term "hemochromatosis" was introduced into the medical literature by a German pathologist, Frederich Daniel von Recklinghausen in 1889, who was also the first to document the link between hemochromatosis and iron overload.[7] HH began to emerge as a unique clinical entity when Joseph Sheldon detailed the phenotypic clinicopathologic aspects of 311 cases. He concluded that HH was rare, affecting 1 in 2,000 to 5,000 patients and speculated this interesting disorder was due to an inborn error of metabolism inherited in a recessive mode. Patients did not live past the average of 2 years from diagnosis and died from complications of infections or diabetes prior to the invention of insulin.[8]

Accompanying the clinical observations and escalating in complexity, iron metabolism research has contributed to our understanding since identification of the first iron metabolism proteins in 1937.[9] When Dame Sheila Sherlock pioneered liver biopsy in 1945, it was quickly applied and used to diagnose hemochromatosis.[10,11] Thereafter, Davies and Arrowsmith introduced treatment which remains current to the present day when they published their experience of repeated phlebotomies in three cases of hemochromatosis.[12]

In 1955, Finch and Finch provided a comprehensive review of hemochromatosis. The concept of "iron-storage disease," where abnormal absorption of iron from the diet or from blood transfusion causing multiorgan tissue damage, was born. Perturbations of plasma iron indices were universal in symptomatic patients and recognized to be the earliest phenotypic abnormalities seen. Two-thirds of the clinical cases were diagnosed at midlife. Despite hepatomegaly and biopsy-proven fibrosis, more than half the decompensated patients exhibited minimal liver biochemical test abnormalities. Liver cancer risk was increased in older patients. Skin pigmentation, diabetes, and cardiac abnormalities were common extrahepatic abnormalities.[13]

Hemochromatosis was definitively shown to be an inherited autosomal recessive HLA-linked disease in 1975.[14] Thereafter, population studies, HLA assessment, and careful family screening contributed to increased numbers of cases being detected, providing a revised incidence in the order 0.3% to 1.2% in Caucasian populations.[15] Almost all cirrhotic patients exhibited weakness, lethargy, and loss of libido. Extrahepatic cardiac and endocrine complications were also more common in cirrhotics.[16] Niederau et al reported excess mortality from liver cancer (mortality ratio 119), diabetes (mortality ratio 14), cardiomyopathy (mortality ratio 14), and cirrhosis (mortality ratio 10). However, nondiabetics and noncirrhotics had a similar mortality profile to the general population.[17]

A major milestone was achieved in 1996 when the candidate gene for HH, termed HFE, was identified on chromosome 6.[1] Two common mutations were identified in the gene product: first, tyrosine was substituted for cysteine at amino acid 282 (C282Y); and second, aspartate was substituted for histidine at amino acid 63 (H63D). In typical HH, homozygosity for C282Y was observed in 85 to 90% of patients of northern European descent,[18,19] whereas in southern Europe as many as 30% of cases were heterozygous or wild-type for this mutation.[20]

With the recent cloning of the HFE gene, the majority of cases are diagnosed at an earlier stage in the disease. This is often well before clinical or biochemical evidence of the disease has developed. The development of disease can be considered in four stages, which have variable durations and do not always progress:[21]

  1. Genetic propensity for iron loading. No overt abnormality apart from raised serum transferrin saturation as the main finding.

  2. Asymptomatic iron overload (0–5 g parenchymal iron overload).

  3. Early symptoms of iron overload—lethargy, arthralgia.

  4. Iron overload with attendant organ damage—cirrhosis, diabetes, destructive arthritis.

Hence, it is paramount for clinicians to adequately diagnose and screen for patients in stages 1 to 3, prior to the development of end-organ damage.

Much debate has ensued regarding the true prevalence and penetrance of HH. Prevalence is dependent on how the cases are defined and which diagnostic tests are used. Cross-sectional studies performed prior to 2006 demonstrated highly variable biochemical and clinical expression.[3,22–24] After 2005, longitudinal studies became the state-of-the-art approach for determination of clinical expression of HH, as they addressed the limitations of point-in-time cross-sectional analyses.[4,25,26]

The most notable of the longitudinal studies emerged from Australia; it was comprised of 31,192 persons of northern European descent who were recruited from the general community in 1990–1994. Health outcomes in 1,438 persons of variable HFE genotype (known as the HealthIron cohort), including all C282Y homozygotes and randomly selected groups of other genotypes, were evaluated over a period of 12 years. In this study, there were 203 C282Y homozygotes, 242 compound heterozygotes, 337 C282Y heterozygotes, and 157 H63D heterozygote/homozygotes. At inception, consistent with previous cross-sectional studies, biochemical penetrance (as defined by serum ferritin level) of C282Y homozygotes was high; 84% of male and 65% of female C282Y homozygotes had serum ferritin levels elevated above the upper limits of normal. Thirty-seven percent of male and 3% of female C282Y homozygotes had ferritin levels exceeding 1000 μg/L.[3,27,28]

In untreated participants after 12 years of follow-up, less than 50% of male and 20% of female C282Y homozygotes exhibited ferritin levels exceeding 1000 μg/L if their baseline ferritin values at initial assessment were less than 1000 μg/L. Moreover, if the baseline value at the age of 55 years was normal, less than 15% of C282Y homozygotes progressed to levels exceeding 1000 μg/L after 12 years. From these data we infer that the majority of C282Y homozygotes who are likely to develop clinically significant iron overload-related disease, would have done so by the age of 55 years.[27,28] The likelihood of developing iron overload-related disease (defined as documented iron overload combined with one of the following: cirrhosis, liver fibrosis, hepatocellular carcinoma, elevated aminotransferrase, physician-diagnosed symptomatic hemochromatosis, or arthropathy of the second and third metacarpophalangeal joints) was determined using regression modeling.[27] Iron overload-related disease was present in 28% of men and 1% of woman at the age 65 years.[27,28] The risk of iron overload-related disease in C282Y/H63D was insignificant unless they had other liver disease risk factors.[29]

Genetic and environmental factors modify the disease penetrance. For example, C282Y homozygous individualts with rs884409 single nucleotide polymorphism (SNP) in CYBRD1 (the gene encoding the duodenal reductase DCYTB) demonstrate substantially lower levels of serum ferritin compared with those who did not possess the mutation.[30,31] Several transferrin SNPs (rs1830084, rs3811647, rs1799852, and rs2280673) are also found to significantly affect iron metabolism in persons with C282Y genotypes.[32] Polymorphisms in the profibrogenic genes (e.g., tumor-growth factor-β [TGF-β]) may determine the course of the hepatic phenotype.[33] Interactions with other genes of iron metabolism including HAMP (hepcidin), HJV (hemojuvelin), BMP (bone morphogenic protein), and Hp (haptoglobin) may also modify iron overload.[34–37]

Important environmental factors that may accelerate iron loading include excessive alcohol consumption and meat intake in postmenopausal women.[38,39] Conversely, eating non-citrus fruit may confer protection against iron loading.[40] Alcohol is a strong cofactor that accelerates the development of liver injury in HH.[41,42] Concurrent viral hepatitis and hepatic steatosis may also enhance the risk of liver injury in patients with iron loading.[43,44] Interestingly, some HLA haplotypes (such as HLA-A*03, B*14) may confer longer survival in C282Y homozygotes due to earlier age of diagnosis and less-severe iron loading phenotype.[45]

HFE gene mutations have also been implicated as risk factors for malignancy. H63D homozygosity results in a threefold increase in incidence and earlier age of onset of colorectal cancer in MMR gene mutation-carriers.[46] C282Y homozygosity doubles the risk of breast cancer in women and colorectal cancer in both genders.[47] In addition, a recent long-term study of phlebotomy therapy demonstrated a reduction in the risk of cancer in patients receiving phlebotomy compared with those who did not.[48] Thus, there is renewed interest in the role of iron in neoplasia and a potential beneficial role of iron-reduction therapies to reduce the risk of malignancy, especially colorectal cancer.[49]