Bruce R. Bacon

HFE Genotype in Patients with Hemochromatosis and Other Liver Diseases

Hereditary hemochromatosis is a common inherited disorder of iron metabolism that affects between 1 in 200 and 1 in 400 persons of northern European descent (1). With early diagnosis and appropriate treatment, survival of patients is normal (1, 2). Recently, researchers identified a novel MHC class 1-like gene, HFE, which contains two missense mutations (3). Eighty-three percent of 178 typical patients with hemochromatosis were homozygous for one of these mutations (Cys282Tyr [C282Y]) (3). Subsequent studies from the United States, Australia, France, and Italy showed homozygosity for the C282Y mutation in 64% to 100% of patients with hemochromatosis (4-7). Heterozygosity for the second mutation (His63Asp [H63D]) is seen in 15% to 20% of the general population; this mutation is not believed to cause the same extent of progressive iron loading (3-7). The estimated allelic frequency of these mutations in white populations is 0.04 for the C282Y mutation and 0.14 for the H63D mutation (1). The presence of C282Y homozygosity and direct (elevated hepatic iron concentration) or indirect (elevated transferrin saturation or ferritin level) evidence of increased iron stores constitute the current gold standard for a definitive diagnosis of hemochromatosis (1, 8). Approximately 40% to 50% of patients with alcoholic liver disease (9), chronic viral hepatitis (10), and nonalcoholic steatohepatitis (11) have abnormal results on blood iron studies. About 5% to 10% of these patients have a modestly increased hepatic iron concentration, but not to the degree seen in typical patients with hemochromatosis. Clinicians have suspected that some of these patients are heterozygous for hemochromatosis; however, without pedigree studies (using HLA haplotyping) of a family with hemochromatosis, this interpretation has been only speculative (10, 11). A high prevalence of C282Y heterozygosity was found in patients with nonalcoholic steatohepatitis (12, 13). In patients with hepatitis C (14-16) and patients with alcoholic liver disease (17), researchers have found an incidence of C282Y heterozygosity equivalent to that in control populations. Recently, a genetic test for hemochromatosis that analyzes the C282Y and H63D mutations has become available. This test allows genotyping of patients who have typical hemochromatosis and those who have liver disease with or without abnormal results on iron studies. We evaluated the contribution of HFE genotyping to the diagnosis of hemochromatosis and determined the prevalence of HFE mutations in a group of patients with liver disease. Methods Patients with Hemochromatosis Between September 1990 and September 1997, clinical hemochromatosis was newly diagnosed in 66 patients by using one of two criteria: 1) a compatible result on liver biopsy (iron deposits of 2+, 3+, or 4+, predominantly in hepatocytes) and a hepatic iron index greater than 1.9 mmol/kg per year [18-22] or 2) HLA identity to a proband. A subset of these 66 patients (n=44) was included in a previous study (3) that identified the HFE gene. Patients with Liver Disease Between January 1996 and September 1997, we performed HFE genotyping on 132 patients with various liver diseases for whom we had also obtained a hepatic iron concentration. Nineteen of these patients were referred for suspected hemochromatosis on the basis of abnormal results on iron studies. The cause of liver disease was thoroughly evaluated in these 132 patients by examination of history of alcohol consumption, viral serologic studies for hepatitis B and C, autoimmune markers (antinuclear antibody, anti-smooth-muscle antibody, and antimitochondrial antibody), and markers of inherited metabolic diseases (transferrin saturation, ferritin level, ceruloplasmin level, and 1-antitrypsin level and protein typing). Laboratory Studies To aid in initial diagnosis, fasting transferrin saturation (reference range, 0.16 to 0.5), ferritin level (reference range, 15 to 200 g/L in women and 30 to 300 g/L in men), routine chemistry panel, and complete blood count were obtained for all patients. Specific serologic studies were also done, as appropriate. All studies were performed at routine clinical laboratories. After obtaining informed consent from all patients, we performed standard percutaneous liver biopsy. Sections of liver tissue were prepared in a routine manner. Staining was done with hematoxylin and eosin, the periodic acid-Schiff test after diastase digestion, Masson trichrome stain, Sweet reticulin stain, and Perls Prussian blue stain. Iron deposits in hepatocytes were graded from 0 to 4+ (23). Hepatic iron concentration was measured on a portion of the liver biopsy sample, as described by Torrance and Bothwell (24). The upper limit of normal used in our laboratory is 26.8 mmol/kg dry weight (1500 g/g dry weight). Hepatic iron index was calculated as hepatic iron concentration (mmol/kg) divided by age (years) (18). HFE genotyping for the C282Y and H63D mutations was performed by oligonucleotide ligation assays on polymerase chain reaction-amplified genomic samples of DNA taken from each patient (3). After obtaining informed consent, we drew blood for HFE genotyping from all patients with liver disease and from all patients with hemochromatosis (before the test became commercially available) by using protocols approved by the institutional review board of Saint Louis University. Statistical Analysis Data are presented as the median and range. Statistical comparisons among the cumulative distributions of the groups were performed by using the exact Komolgorov-Smirnov two-sample test. A P value less than 0.01 was considered statistically significant. Role of the Funding Sources The funding sources had no role in the collection, analysis, or interpretation of the data or in the decision to submit the paper for publication. Results Genotype by Patient Group Of the 66 patients who had clinically diagnosed hemochromatosis, 60 (91%) were C282Y homozygotes, 2 (3%) were compound heterozygotes, 1 (1.5%) was a C282Y heterozygote, 2 (3%) were H63D heterozygotes, and 1 (1.5%) was negative for both mutations. Of the 132 patients with liver disease, 80 had chronic hepatitis C, 19 had abnormal results on serum iron studies and had been referred for evaluation of iron overload or suspected hemochromatosis, 17 had nonalcoholic steatohepatitis, 4 had primary biliary cirrhosis or primary sclerosing cholangitis, and 12 had other liver disorders. Of these 132 patients, 6 (5%) were C282Y homozygotes, 8 (6%) were compound heterozygotes, 6 (5%) were C282Y heterozygotes, 5 (4%) were H63D homozygotes, 20 (15%) were H63D heterozygotes, and 87 (66%) were negative for both mutations. In the group of 19 patients with abnormal results on iron studies who were referred for evaluation of iron overload, 5 (26.5%) were C282Y homozygotes, 1 (5%) was a C282Y heterozygote, 1 (5%) was an H63D homozygote, 5 (26.5%) were H63D heterozygotes, and 7 (37%) were negative for both mutations. Iron Status by HFE Genotype To define and analyze the iron status of all patients for whom genotyping had been performed, we grouped patients with hemochromatosis and patients with liver disease together and divided them according to genotype (Table 1). Additional information about patients with certain genotypes is presented in Tables 2, 3, and 4. Table 1. HFE Genotype, Age, Serum Iron Studies, and Hepatic Iron Studies in 198 Patients with Hemochromatosis and Other Liver Diseases C282Y Homozygotes Of the 66 patients who were homozygous for the C282Y mutation (C282Y/C282Y), 40 were men and 26 were women. Median age at diagnosis was 45 years and 49 years, respectively. At diagnosis, men and women had similar transferrin saturations (P>0.2) and ferritin levels (P=0.07). The hepatic iron concentration (median, 159.1 mmol/kg dry weight [8910 g/g dry weight]; P<0.01) and the hepatic iron index (median, 3.25 mmol/kg per year; P<0.01) were higher in C282Y homozygotes than in the groups of patients with other genotypes (Figure 1). The hepatic iron concentration and the hepatic iron index were similar in male and female C282Y homozygotes (P>0.2 [data not shown]). On biopsy, all patients had hepatic iron deposition of grade 2+ to 4+. Three women and 9 men had substantial fibrosis or cirrhosis on liver biopsy; 6 of these patients had alanine aminotransferase levels or aspartate aminotransferase levels above the upper limit of normal. The median age of patients with fibrosis was 54 years (range, 40 to 72 years). The youngest patient with cirrhosis was a 40-year-old man with concomitant chronic hepatitis C. The youngest patients with cirrhosis or substantial fibrosis who were homozygous for the C282Y mutation and did not have contributing factors were a 47-year-old man and a 47-year-old woman. Figure 1. Hepatic iron concentration ( top ) and hepatic iron index ( bottom ) according to HFE genotype. Ten of 66 patients (15% [CI, 7.5% to 26%]) who were C282Y homozygotes had a hepatic iron index less than 1.9 mmol/kg per year. Specific findings of these patients are detailed in Table 2. Six of these 10 patients were men; 1 was a blood donor. Four of these patients (patients 1, 2, 5, and 8) were identified by HLA haplotyping in family studies; the other 6 were identified from the group of patients with liver disease who were not suspected of having hemochromatosis at the time of initial evaluation. These patients were thought to be heterozygotes or to have concomitant nonalcoholic steatohepatitis. Nine of the 10 patients had a transferrin saturation of at least 45%, and all 10 patients had a hepatic iron concentration above the upper limit of normal. All patients had hepatic iron deposition of grade 2+ to 4+, and only 1 had minimal fibrosis on biopsy. Table 2. Characteristics of C282Y Homozygotes with a Hepatic Iron Index Less Than 1.9 mmol/kg per Year When C282Y homozygosity is used as the gold standard for the diagnosis of hemochromatosis, the sensitivity

Duration of hepatic iron exposure increases the risk of significant fibrosis in hereditary hemochromatosis: a new role for magnetic resonance imaging.

OBJECTIVES:Hepatic fibrosis is a complication of hereditary hemochromatosis. The aim of this study was to determine whether the product of the magnitude and duration of hepatic iron exposure is related to the risk of significant fibrosis.METHODS:Receiver-operating characteristic curve analysis to determine the utility of hepatic iron concentration (HIC) and age in the diagnosis of low- or high-grade fibrosis was undertaken retrospectively in 60 subjects who had undergone liver biopsy for assessment of hereditary hemochromatosis. A prospective pilot study was then conducted in 10 additional subjects to evaluate utility of magnetic resonance imaging (MRI) measurements of HIC to predict fibrosis.RESULTS:Eighteen subjects had high-grade fibrosis while 42 subjects had low-grade fibrosis. Hepatic iron concentration alone was highly sensitive (100%) but of limited specificity (67%) in diagnosis of high-grade fibrosis. The product of [HIC × age] had a sensitivity and specificity of 100% and 86%, respectively, for diagnosis of high-grade fibrosis. Magnetic resonance imaging measurements also provided accurate assignment of subjects into fibrosis severity groups.CONCLUSIONS:Duration of exposure to iron is important in the development of hepatic fibrosis in hereditary hemochromatosis. The product of HIC and age is highly sensitive and specific for diagnosis of high-grade fibrosis and can be obtained using MRI.