WO2001035095A2

WO2001035095A2 - Diagnostic test for hemochromatosis

Info

Publication number: WO2001035095A2
Application number: PCT/US2000/042108
Authority: WO
Inventors: Barry E. Rothenberg; James C. Barton; Luigi F. Bertoli
Original assignee: Individual
Current assignee: Individual
Priority date: 1999-11-12
Filing date: 2000-11-09
Publication date: 2001-05-17
Anticipated expiration: 2002-05-12
Also published as: WO2001035095A3; AU3269101A

Abstract

A method of diagnosing hemochromatosis or a predisposition thereto in a mammal is carried out by determining the mean corpuscular volume (MCV) value of a blood sample from the mammal. An increase in MCV compared to a normal control value indicates that the mammal has hemochromatosis or is predisposed to developing hemochromatosis.

Description

DIAGNOSTIC TEST FOR HEMOCHROMATOSIS

BACKGROUND OF THE INVENTION

The invention relates to diagnosis of iron disorders.

Hemochromatosis is a common hereditary disorder that affects approximately 0.5% of persons of western European descent. In various populations, 60-100% of cases are attributable to homozygosity for a missense mutation (cDNA nucleotide 845 G6A; C282Y) in HFE, a major histocompatibility class I gene on chromosome 6p. Some patients are compound heterozygotes for C282Y and another HFE allele (cDNA nucleotide 187 C6G; H63D), or are H63D homozygotes. However, C282Y and H63D are not known to occur on the same chromosome. Other persons with a hemochromatosis phenotype are homozygous for H63D, are heterozygous for C282Y or H63D, or are presumed to have an HFE or other mutation that is not presently detectable (wild-type; wt/wt). Regardless of HFE genotype, persons with a hemochromatosis phenotype usually have increased iron saturation of plasma transferrin, typically absorb increased quantities of iron, and often develop multisystem disease due to iron overload .

SUMMARY OF THE INVENTION

The invention is based on the discovery that peripheral blood erythrocyte parameters in hemochromatosis patients or those at risk of developing hemochromatosis is distinguished from the erythrocyte parameters of normal individuals. Hemoglobin (Hb), hematocrit (Hct), mean corpuscular volume (MCV), mean coφuscular hemoglobin (MCH), and mean corpuscular hemoglobin concentration (MCHC) values were found to be significantly higher in hemochromatosis patients compared to normal control individuals. Accordingly, the invention provides a methods of diagnosing hemochromatosis or a predisposition thereto by detecting altered erythrocyte parameters.

A method of diagnosing hemochromatosis or a predisposition thereto in a mammal is carried out by determining the mean corpuscular volume (MCV) value of a blood sample from the mammal. An increase of at least 5% compared to a normal control value indicates that the mammal has hemochromatosis or is predisposed to developing hemochromatosis. A normal control value or range is one obtained by testing an individual or pool of individuals who are known to be wt/wt. Individuals used to determine a normal control value or range are neither homozygous nor heterozygous for a genetic mutation associated with hemochromatosis such as C282Y or H63D of the HFE gene. Preferably, the increase is at least 7%, more preferably at least 7.5%, more preferably at least 8%, more preferably at least 9%, and most preferably at least 10% higher than a normal control value or range. Preferably, the erythrocyte profile of the tested individual does not indicate an anemic condition. For example, a red blood cell count (NBC), hematocrit (Hct), or hemoglobin concentration (Hb) should be within a normal range but the MCV value is elevated compared to a normal value or range to indicate a diagnosis of hemochromatosis. A method of diagnosing hemochromatosis or a predisposition thereto in a mammal is also carried out by determining that the MCV value of a blood sample from the mammal is at least 80 fL. Preferably, the value is at least 85 fL, and more preferably the value is in the range of 90-102 fL. An erythrocyte profile of hemochromatosis is distinguished from one previously thought to be associated with a hepatic disease in that the inventive profile diagnostic of hemochromatosis does not include an erythrocyte parameter, e.g., RBC, Hct, or Hb, which indicates anemia.

Hemochromatosis is also diagnosed by determining the mean coφuscular hemoglobin (MCH) value, mean coφuscular hemoglobin concentration (MCHC), Hb, or Hct of a blood sample from the mammal. An increase of at least 5% (preferably at least 7%, more preferably at least 7.5%, more preferably at least 8%, more preferably at least 9%, and most preferably at least 10%) compared to a normal control value indicates that the mammal has hemochromatosis or is predisposed to developing hemochromatosis. As above, the erythrocyte profile indicative of a diagnosis of hemochromatosis does not include a parameter such as RBC, Hct, or Hb which indicates an anemic condition. Hemochromatosis is diagnosed by determining that the MCH value of a blood sample from the mammal is at least 26 pg. More preferably, the value is at least 28 pg, such as a value in the range of 30-35 pg. Unlike the commonly-used transferrin saturation test (which must be done repeatedly in order to diagnose hemochromatosis), the invention provides a method of diagnosing hemochromatosis or a predisposition thereto in a mammal with a single determination of an MCV value or an MCH value from a blood sample from an individual. A single MCV or MCH value indicates whether the mammal, e.g., a white human of Western European descent, has hemochromatosis or is predisposed to developing hemochromatosis. Preferably, the values are optically determined, e.g., using a standard automated device for determining blood cell parameters. Another advantage of the methods described herein is that the blood sample is obtained from a fasting or non-fasting individual. Preferably, the individual is non- fasting. In contrast, standard transferrin saturation determinations must be carried out on a blood sample from a fasting mammal. MCV values which correspond to slight macrocytosis indicate a diagnosis of hemochromatosis or a predisposition thereto. A single MCV determination with a value that is greater than or equal to 101.0 fL indicates that the individual from which the blood sample was derived has hemochromatosis or is at risk of developing the disorder. Further testing, e.g., using a transferrin saturation test, or therapeutic intervention is indicated. Preferably, the MCV value is less than 120.0 fL; more preferably, the value is is less than 1 10.0 fL. For example, the MCV value is less than 106.0 fL.

MCH values diagnostic of hemochromatosis are generally greater than or equal to 33.0 pg. Such values indicate further testing and/or therapeutic intervention. Preferably, the mammal is male, and the MCH value is greater than or equal to 34.0 pg. Alternatively, the mammal is female, and the MCH value is greater than or equal to 32.0 pg.

Also within the invention is a method for identifying mammals with an increased risk for heavy metal poisoning, comprising determining the MCV value of a blood sample from a mammal. An increase in MCV value of the sample compared to a normal control value indicates that the mammal has an increased risk of developing poisoning upon exposure to a heavy metal, e.g., lead, cadmium, mercury, bismuth, and uranium. Similarly, a method for identifying mammals with an increased risk for heavy metal poisoning is carried out by determining the MCH value of a blood sample from a mammal, and an increase in the MCH value of the sample compared to a normal control value indicates that the mammal has an increased risk of developing poisoning upon exposure to a heavy metal. MCV and MCH values which are greater than or equal to 101 fL and greater than or equal to 32.0 pg, respectively, indicate an increased risk. The method is useful to determine risk of developing job-related disabilities. For example, workers which handle nuclear fuel or work in nuclear power plants may be exposed to and become at risk for developing uranium poisoning. Glassworkers may be exposed to bismuth. The test is also useful to determine increased risk of lead poisoning from lead paint or contaminated water.

A method for identifying mammals with increased physical endurance is also carried out by determining MCV or MCH values of a blood sample from a mammal. An increase in MCV or MCH value of the sample compared to a normal control value indicates that the mammal has an increased capacity for physical endurance. MCV and MCH values which are greater than or equal to 101 fL and greater than or equal to 32.0 pg, respectively, indicate an increased capacity for physical endurance.

The present method diagnostic methods for diagnosing hemochromatosis overcomes the insensitivity of prior art diagnostic methods, because the standard value ranges of Hb, Hct, MCV, and MCH of earlier protocols are not true "normal ranges". Earlier protocols did not take into consideration that at least 25% of the population contains at least one mutation associated with hemochromatosis. The data reported herein indicates that the inclusion of such individuals in the determination of the previously-accepted standard ranges of values results in ranges that are not reflective of a normal control, but are approximately 5-10% higher. The diagnostic methods of the invention identify hemochromatosis patients or those at risk of developing the disease with standard hematological tests, thereby extracting the maximum information from the red blood cell indices.

Methods of measuring erythrocyte parameters are well known, but the inteφretation of the results to diagnose hemochromatosis is new . Hemochromatosis was not diagnosed using the previous assays because the values of hemochromatosis patients would fall into the "normal ranges". However, the normal ranges of the prior art were skewed because of the inadvertent inclusion of individuals which are homozygous or heterozygous for hemochromatosis-associated mutations (and which elevated Hb, Hct, MCV, or MCH values compared to individuals without a hemochromatosis-associated mutation

Other features and advantages of the invention will be apparent from the following description of the preferred embodiments thereof, and from the claims.

BRIEF DESCRIPTION OF THE DRAWING Fig. 1 is a line graph showing the frequency of mean coφuscular hemoglobin values in C282Y/C282Y probands and wt/wt controls.

DESCRIPTION OF THE PREFERRED EMBODIMENTS Peripheral blood erythrocyte parameters and HFE genotypes in 94 hemochromatosis probands and 132 Caucasian normal controls were evaluated. Mean red blood cell (RBC) counts in probands and controls were not significantly different. However, mean values of Hb, Hct, MCV, MCH, and MCHC were significantly higher (e.g., at least 5% higher) in C282Y/C282Y probands (ii = 60) than in wild-type (wt/wt) controls (n = 65). Probands with other HFE genotypes also had increased mean erythrocyte parameters (other than RBC). A proband is the first individual in a family identified to be affected by hemochromatosis. Forward and peripheral blood smears prepared before therapeutic phlebotomy to remove iron revealed that erythrocytes in many probands had increased diameters and were well-filled with hemoglobin. Erythrocyte parameters were similar in C282Y/C282Y probands with and without hepatomegaly, elevated serum concentrations of hepatic enzymes, hepatic cirrhosis, diabetes mellitus, arthropathy, or hypogonadism. Among C282Y/C282y probands, significantly greater values of MCV (but not other erythrocyte parameters) occurred among those who had transferrin saturation values ofat least 75% or iron overload at diagnosis. After iron depletion, the mean MCV, MCH, and MCHC values of C282Y/C282Y probands decreased, but remained significantly greater than values in wt/wt controls. Mean values of pre-phlebotomy MCH and MCHC were lower in HLA-A3-positive than in HLA- A3-negative C282Y/C282Y probands. These data indicate that increased values of mean Hb, Hct, MCV, MCH, and MCHC in hemochromatosis probands are due to increased iron uptake and hemoglobin synthesis by immature erythroid cells.

Red blood cell (RBC), Hb, Hct, MCV, MCH, MCHC parameters are measured using methods well known in the art of hematology, e.g. using an automated analyzer such as the Cell-Dyne 1700. Other analyzers such as a Model S particle analyzer (U.S. Pat. No. 3,549,994) are widely used to generate patient hematological data. Preferably, macrocytosis is not detected by visual examination (e.g., using microscopy) of peripheral blood smears.

MCV is directly detected or is derived as the quotient of hematocrit and red blood cell count, i.e., MCV=HCT/RBC. Similarly, MCH is the quotient of haemoglobin and red blood cell count, i.e., MCH=Hb/RBC.

PCV is the ratio of the volume of packed red cells to the total volume. For example, an electric counter makes a direct measurement of the red cell count (RBC/μL) and mean red cell volume:

MCV (fL) = PCV (L/L)

RBC (μL) x lO^"9

MCHC (g/dL) = Hb (g/dU

PCV (L/L) MCH (pg) = Hb (g/dL

(RBC /μL) ) x lO^"7

The hemochromatosis index described herein provides a highly informative discriminant between hepatic disease and hemochromatosis, such two conditions having very different prognoses, but yet frequently being confused during diagnosis using current diagnostic indices. The data described herein indicates that individuals with a RBC, Hct or Hb level that is not indicative of an anemic condition, but having a MCV, MCH, or Hb level that is at least 5% above the upper reference range of a standard normal value is diagnostic of hemochromatosis or a predisposition thereof. For example, an individual with a RBC range between 4.20-6.30 x 10⁶ μL (or other erythrocyte parameters consistent with a nonanemic condition) but with an elevated MCV, MCH, Hct or Hb level that is elevated is identified as having hemochromatosis or at risk of developing the condition.

Selection of Hemochromatosis Probands and Normal Control Volunteers

All persons who participated were Caucasians at least 18 years of age. Hemochromatosis probands were identified during routine medical care delivery who had completed iron depletion with therapeutic phlebotomy; probands with complications of iron overload (hepatic cirrhosis, diabetes mellitus, hemochromatosis-associated arthropathy, or hypogonadotrophic hypogonadism) were not excluded. At the time of diagnosis of hemochromatosis, each proband had a normal serum folate concentration, serum vitamin B 12 concentration, and thyroid profile. Volunteer normal control subjects unrelated to our hemochromatosis probands were recruited from the general population. Probands and control subjects who had alcoholism, increased erythrocyte mass, anemia other than that attributed to complications of iron overload, unexplained reticulocytosis, marrow infiltrative disease, leukemia or other malignancy, evidence of cold agglutinins or cryoglobulins, drug use associated with erythrocyte abnormalities, pregnancy, malabsoφtion, vegetarianism, dialysis requirements, transfusion dependency, or inherited disorders of DNA synthesis known to cause macrocytosis, pseudomacrocytosis, or megaloblastosis were excluded. Diagnosis of Hemochromatosis and Evaluation of Iron Overload

The working diagnostic criterion for hemochromatosis of the American College of Pathologists was used: elevated fasting transferrin saturation (at least 60% males, at least 50% females). In exceptional cases (n = 3), transferrin saturation values were not consistently elevated, but otherwise unexplained iron overload consistent with hemochromatosis was demonstrated by analysis of hepatic biopsy specimens and calculation of hepatic iron index. HFE genotyping was not used to establish the diagnosis of hemochromatosis. Iron overload was defined as evidence of systemic iron overload demonstrated by otherwise unexplained elevated serum ferritin concentration (at least 300 ng/mL in men, at least 200 ng/mL in women), increased hepatic iron content determined using hepatic biopsy specimens, or at least 4 g of iron mobilized by phlebotomy. For each proband, the pre-phlebotomy serum transferrin saturation values and serum ferritin concentrations were tabulated, and the units of blood removed by therapeutic phlebotomy (1 unit of blood = 450 - 500 mL, equivalent to -200 mg of iron) recorded. Complications of iron overload were assessed using standard methods.

Evaluation of Erythrocyte and other Blood Parameters

Peripheral blood specimens obtained by standard antecubital venipuncture from probands and control subjects were analyzed using a Cell-Dyne7 1200 automated blood counter (Abbott Laboratories, Chicago, IL). In probands, serial complete blood counts were obtained before and during iron depletion therapy with phlebotomy. Erythrocyte parameters were also measured in each proband at least three months after the completion of iron depletion therapy at a time when the serum ferritin concentration was 20 - 50 ng/mL. Serum iron parameters, serum concentration ofhepatic enzymes, and other blood chemistry determinations were performed using automated clinical laboratory methods. Hepatic biopsy specimens were stained with hematoxylin and eosin, acid ferrocyanide, and Masson's trichrome techniques; additional portions of specimens were analyzed for iron content using atomic absoφtion spectrometry. Peripheral blood smears were prepared with Wright-Giemsa staining, and reticulocytes were enumerated using New Methylene Blue technique. Bone marrow aspirates and biopsy specimens were stained with Wright-Giemsa, hematoxylin and eosin, and acid ferrocyanide technique, as indicated. HFE genotypes and human leukocyte antigen (HLA) immunophenotypes were determined according to standard methods.

Statistical Considerations The data set consisted of observations on 94 hemochromatosis probands and 132 normal control subjects. There were sufficient numbers of C282Y/C282Y probands (n = 60) and normal control subjects (including 65 who had a wt/wt HFE genotype) to permit meaningful analysis of subgroup data. Normal reference ranges for erythrocyte parameters measured using the Cell-Dyne7 1700 supplied by the manufacturer are: erythrocyte (RBC) count 4.20 - 6.30 x 10⁶/FL; hemoglobin (Hb) 12.0 - 18.0 g/dL; hematocrit (Hct) 37.0 -51.0%; mean coφuscular volume (MCV), 80.0 - 97.0 fL; mean coφuscular hemoglobin (MCH) 26.0 - 32.0 pg; mean coφuscular hemoglobin concentration (MCHC) 31.0-36.0 g/dL; mid erythrocyte width distribution (RDW) 1 1.9 - 14.5%. The reference range for reticulocyte counts in our laboratory is 0.5 - 1 .5%. Descriptive data are presented as percentages or as means ± 1 S.D. (range). Comparisons were performed using unpaired two-tailed t-test, chi- square analysis, or the correlation coefficient (r), as appropriate. A value of p < 0.05 was defined as statistically significant.

General Characteristics of Study Subjects: Hemochromatosis Probands The mean age of our 94 probands was 51 ± 13 years (range 20 - 80 years); 58 (61

.7%) were men and 36 (38.3%) were women. There was no significant difference in the mean ages of men and women. Fifteen probands (16.0%) had hepatic cirrhosis, 10(10.6%) had diabetes mellitus, 21 (22.3%) had hemochromatosis-associated arthropathy, and 15 (16.0%) had hypogonadotrophic hypogonadism. Iron overload was more severe and complications of iron overload were more frequent in C282Y/C282Y probands, on the average, than in probands with other HFE genotypes. Frequencies of HFE genotypes among the present probands were similar to those previously reported from unselected hemochromatosis probands from the same geographic area (Table 1).

*C282Y, cDNA nucleotide 845 G→A; nucleotide 187 C→G; wt, wild-type (normal)

HFE allele.

Normal Control Subjects

The mean age of our control subjects was 52 ± 15 years (range 18 - 86 years); 53 (40.2%) were men and 79 (59.8%) were women. There was no significant difference in the mean ages of men and women. Frequencies of HFE genotypes among the control subjects were similar to those previously reported from normal persons from the same geographic area (Table 1).

Erythrocyte Parameters of Study Subjects: Hemochromatosis Probands

Among all probands, males had significantly greater mean values of RBC, Hb, and Hct than women (4.83 ± 0.47 x 10⁶/FL vs. 4.36 ± 0.42 10⁽7FL, 15.8 ± 1.4 g/dL vs. 14.2 ± 1.0 g/dL, and 46.3 ± 4.5% vs. 41.9 ± 3.3%; p < 0.0001, p < 0.0001, and p < 0.01 , respectively). Mean values of MCV, MCH, and MCHC were not significantly different in men and women. Mean values of RBC, Hb, and Hct were similar among probands of different HFE genotypes, and values in most probands were within the corresponding reference ranges (Table 2). However, 41 (43.6%), 49 (61.7%), and 8 (8.5%) of all probands had values of MCV, MCH, and MCHC, respectively, that were greater than the corresponding reference ranges (Table 2). Peripheral blood smears made before the initiation of therapeutic phlebotomy revealed that erythrocytes in 71.7% of C282Y/C282Y probands and in 64.9% of all probands had mildly or moderately increased diameters and were well-filled with hemoglobin ("thick" macrocytes). Significant numbers of "thin" macrocytes (with or without central "targets"), poikilocytosis, anisocytosis, acanthocytosis, or nuclear hypersegmentation of granulocytes were not observed. Mean RDW values were within the reference range in probands grouped by HFE genotype. Pre-treatment reticulocyte counts in each of 43 probands and bone marrow aspirate and biopsy specimens in each of 8 probands were normal or non-diagnostic.

Table 2: Erytlirocyte Parameters and Transferrin Saturation Values in 94 Untreated

Hemochromatosis Probands.*

respectively above and below the corresponding reference ranges are shown in parentheses.

IC282Y, cDNA Nucleotide 845 G→A; H63D, cDNA nucleotide 187 C→G; wt, wild- type (normal) HFE allele.

Normal Control Subjects

Men had significantly greater mean values of RBC, Rb, and Hct than women (5.00 ±0.47 x 10°/FL vs. 4.39 ± 0.40 x lO⁶/FL; 15.3 x 1.4 g/dL vs. 13.3 ± 1.0 g/dL; and 45.4 ± 4.4% vs. 39.8 ± 3.4%; p < 0.0001 for each comparison). However, mean values of MCV, MCH, and MCHC were not significantly different in men and women. Mean values of REC, Hb, and Hct and were similar among normal control subjects of different HFE genotypes, and most values were within the corresponding reference ranges (Table 3). Although the mean values of MCV, MCH, and MCHC were not increased above the corresponding reference ranges in any HFE genotype group, individual values of MCH or MCHC were elevated in some control subjects who inherited the HFE genotypes C282Y/wt or C282Y/H63D (Table 3). Peripheral blood smears did not reveal significant moφhologic abnormalities of erythrocytes other than "thick" macrocytes that occurred in some control subjects with HFE genotypes C282Y/wt or C282Y/H63D. Mean RDW values were within the reference range in control subjects grouped by HFE genotype.

Comparisons of Hemochromatosis Probands and wt/wt Normal Control Subjects

Mean values of erythrocyte parameters in probands grouped by HFE genotype (Table 2) were compared with those of wt/wt normal control subjects (Table 3). In C282Y/C282Y probands, these mean values were significantly increased: Hb (p = 0.0001), Hct (p = 0.008), MCV (p = 0.0001), MCH (p = 0.0001), and MCHC (p = 0.016). Values of MCH in C282Y/C282Y probands and in wt/wt normal control subjects are displayed in Fig. 1. The mean value of MCH was significantly increased in C282Y/H63D probands (p = 0.04), and the mean value of MCHC was increased in C282Y/wt probands (p = 0.02). Similar trends in mean erythrocyte parameters were also observed in probands with other HFE genotypes

(Table 2). However, these values were not significantly different from the respective values in wt/wt normal control subjects, possibly due to the small numbers of probands in these subgroups. Further, the percentages of individual MCHC values above the reference range for our blood cell counter in C282Y/wt, C282Y/H63D, and H63D/wt probands were greater than that of C282Y/C282Y probands. However, these apparent differences may also be attributable to the small numbers of probands in these non-C282Y/C282Y subgroups.

Table 3: Erythrocyte Parameters in 132 Unrelated Normal Control Subjects.*

HFE genotype! RBC,107μL Hb,g/dL Hct, % MCV, fL MCH, pg MCHC,

(n) g/dL

C282Y/wt (24) 4.51±0.43 14.0±1.5 41.0±4.2 31.2±3.0 34.0±1.3 34.1±1.3 (0; 20.8) (0; 4.2) (0; 16.7) (0; 0) (20.8; 0) (20.8; 0)

C282Y1H63D 4.78±0.65 14.9±2.2 43.7±6.7 91.9±5 .3 31 .2±1.7 34.0±1.2 (14) (0;21.4) (0;0) (7.1; 14.3) (14.3;0) (0;0) (28.6;35.6)

H63D/wt(23) 4.70±0.48 14.1±1.5 42.9±5.1 91.2±5.0 30.0±2.1 32.9±1.5 (0; 13.0) (4.3; 0) (4.3; 4.3) (8.7; 0) (13.0;8.7) (0; 4.3)

H63D/H63D (6) 4.32±0.25 13.5±0.7 40.2±2.8 93.1±3.2 31.2±1.2 33.5±1.0 (33.3;0) (0;0) (0;0) (0;0) (16.7;0) (0;0) wt/wt (6) 4.67±0.54 14.0-1=1.4 42.1±4.3 90.4±3.8 30.2±1.9 33.3---1.4 (0; 18.5) (0; 3.1) (0; 6.2) (1.5; 0) (10.8;3.1) (3.1 ;9.2)

All Controls 4.64±0.52 14.1±1.5 42.1 ±4.7 91.0±4.1 30.4±1.8 33.5±1.4 (132) (1.5; 17.4) (0.8; 2.3) (1.5; 6.8) (3.8; 0) (12.1;3.0) (8.3;9.1)

*Data are displayed as mean ± 1 S.D. Percentages of probands with values respectively above and below the corresponding reference ranges are shown in parentheses. tC282Y, cDNA nucleotide 845 G→A; H63D, cDNA nucleotide 187 C→G; wt, wild- type (normal) HFE allele.

Relationships of Erythrocyte Parameters to Various Factors in C282Y/C282Y Probands

Men had significantly greater mean values of RBC (4.69 ± 0.48 x 10⁶/μL vs. 4.41 ± 0.41 x 10⁶/μL; p = 0.02), Hb (15.4 ± 1.4 g/dL vs. 14.4 ±0.8 g/dL; p = 0.002), and Hct (45.6 + 4.6% vs. 42.2 ±3.2%; p = 0.002) than women. Mean values of MCV, MCH, MCHC, and transferrin saturation did not differ significantly in men and women. The arithmetic mean of serum ferritin concentrations was marginally greater in men (p = 0.05), and the number of units of blood removed by phlebotomy was significantly greater in men than women (37 ± 26 units vs. 21 ± 20 units; p = 0.01 ).

Iron Parameters (transferrin saturation, serum ferritin concentration, and phlebotomy units)

C282Y/C282Y probands were grouped by values of transferrin saturation at diagnosis (<75% (n = 18) and > 75% (n = 42)); the mean transferrin saturation values in these groups were 61 ± 12% and 92 ± 8%, respectively. In the group with higher mean transferrin saturation, the mean MCV was greater than in those with lower mean transferrin saturation (98.0 ± 4.3% vs. 94.8 ± 5.3%; p = 0.02); other mean erythrocyte parameters did not differ significantly between these groups. The mean volume of blood removed by phlebotomy was marginally greater in the high transferrin saturation group (33 ± 25 units vs. 19 ± 20 units, respectively; p = 0.05). C282Y/C282Y probands were grouped according to the iron removed by therapeutic phlebotomy (< 4 g of iron and >4 g of iron; n = 26 and n = 34, respectively); mean values of iron parameters were significantly greater in the latter group. Their mean MCV was also greater (98.6 ± 4.1 fL vs. 95.3 ± 4.9 fL, respectively; p = 0.006); other mean erythrocyte parameters did not differ significantly between these groups. However, there was no significant correlation (r) between erythrocyte and iron parameters when data from all C282Y/C282Y probands were analyzed. Taken together, these data indicated that the relationship between erythrocyte and iron parameter data in C282Y/C282Y probands is not a coiltinuous one, but that markedly elevated iron parameters may be associated with unusually high values of MCV.

Complications of Iron Overload

Erythrocyte parameters did not differ significantly in C282Y/C282Y probands with and without hepatomegaly, elevated serum concentrations of hepatic enzymes, or cirrhosis. However, the arithmetic means of the serum ferritin concentration and numbers of units of phlebotomy required to achieve iron depletion were significantly greater in those with hepatic cirrhosis than in those without cirrhosis (2,877 ± 1,769 ng/dL vs. 824 ± 657 ng/dL and 60 ± 32 units vs. 22 ± 17 units; p = 0.0001 for each comparison). The occurrence of diabetes mellitus or hemochromatosis-associated arthropathy in C282Y/C282Y probands was not associated with significantly different mean values of erythrocyte parameters than was observed in probands unaffected with these complications. All of our C282Y/C282Y probands with hypogonadotrophic hypogonadism were men; their mean erythrocyte parameters did not differ significantly from those of male C282Y/C282Y probands without hypogonadism.

HLA-A Types

Forty-five C282Y/C282Y probands were positive for HLA- A3 (14 presumed HLA- A3 homozygotes and 31 HLA- A3 heterozygotes); 15 probands did not express HLA-A3. Mean values of pre-phlebotomy MCH and MCHC were lower in HLA-A3-positive than in HLA- A3 -negative probands (33.8 ± 1.1 pg vs. 34.5 ± 1.4 pg, p = 0.045; and 31.9 + 2.1 g/dL vs. 33.2 ±2.1 g/dL, p = 0.038, respectively). Other erythrocyte and iron parameters did not differ significantly between HLA-A3-positive and -negative probands. Effect of Therapeutic Phlebotomy on Erythrocyte Parameters in Hemochromatosis Probands

Post-treatment erythrocyte parameters were measured in the same 94 hemochromatosis probands studied before therapeutic phlebotomy. Measurements were made at least three months after completion of iron depletion therapy when the serum ferritin concentration was 20- 50 ng/mL (Table 4). In treated C282Y/C282Y probands, the mean RBC value was lower than in wt/wt controls (4.38 ± 0.34 x 10⁶/μL probands vs. 4.67 ± 0.54 x 10⁶/μL controls, p = 0.0008). However, mean values of Hb and Hct in C282Y/C282Y probands (Table 4) were similar to those of wt/wt normal control subjects (Table 3). In C282Y/C282Y probands, mean post-treatment values of MCV, MCH, and MCHC remained significantly greater than corresponding measurements in wt/wt normal control subjects (p = 0.0001, 0.0001, and 0.008, respectively) (Tables 3 and 4). In probands with other HFE genotypes, similar but less pronounced abnormalities persisted after phlebotomy therapy. Among all 94 probands, post-treatment values of MCV, MCH, and MCHC greater than the upper limit of the corresponding reference ranges were detected in 26 (27.6%), 42 (44.7%), and 4(4.3%) probands, respectively.

Table 4: Erythrocyte Parameters in 94 Treated Hemochromatosis Probands.*

phlebotomy. In the first few weeks after the initiation of weekly phlebotomy therapy, transient increases in values of MCV attributed at a time when the serum ferritin concentration was 20-50 ng/mL. Pre-phlebotomy values of these probands are displayed in Table 3. fC282Y, cDNA nucleotide 845 G→A; H63D, cDNA nucleotide 187 C→G; wt, wild- type (normal) HFE allele. Diagnosis of Hemochromatosis

In C282Y/C282Y hemochromatosis probands, mean values of Hb, Hct, MCV, and MCH were 7.1 - 8.9% greater, on the average, than those of wt/wt normal control subjects. These changes occurred in association with a 1 .8% increase in mean

MCHC, and without a significant increment in RBC count. This represents an average of approximately five grams of additional circulating hemoglobin per proband (equivalent to -170 mg of iron) than is found in wt/wt normal controls, based on a blood volume of five liters and the assumption that plasma volume is normal in persons with hemochromatosis. As in persons with iron deficiency, MCV and MCH were more sensitive than MCHC in detecting erythrocyte changes in our C282Y/C282Y hemochromatosis probands before or after therapeutic phlebotomy. In hemochromatosis homozygotes, the appearance of marrow erythroid cells, rates of erythrocyte production and destruction, and biochemical analyses of hemoglobin are usually normal. Taken together, these data indicate that increased entry of iron into hemochromatosis erythrocytes is subsequently incoφorated into hemoglobin in the cells.

Previously, elevated values of MCV and macrocytosis were thought to be diagnostic of hepatic disease. The results reported herein indicate that this is not correct for hemochromatosis. The profile of erythrocyte parameters described herein as being diagnostic for hemochromatosis differs from the profile indicative of hepatic disease (i.e., anemic conditions). In aggregate, the probands did not have anemia, "thin" macrocytes, "target" macrocytes, or acanthocytosis, unlike many persons hepatic disease such as chronic liver disease. Erythrocyte parameters in C282Y/C282Y probands did not differ significantly in the presence or absence of hepatomegaly, elevated serum concentrations ofhepatic enzymes, or hepatic cirrhosis. Further, other persons with hemochromatosis previously reported, similar to our C282Y/C282Y probands, also had a normal mean Hct (41.6%; range 32 - 53% (n = 30)) and an elevated mean MCV (99.8 fL; range 91 - 112 fL (n = 20)) . Before ; iron depletion therapy, 44% and 52% of our probands had values of MCV and MCH above the normal reference ranges, respectively. After treatment, 28% and 45% had elevated values of MCV and MCH, respectively. Because hemochromatosis is common, this disorder is a frequent cause of elevated "normal range" values of MCV, MCH, and "thick" macrocytosis among western Caucasians that is not generally recognized. As a result, many previously accepted "normal" ranges are really elevated due to the presence of hemochromatosis patients in the tested normal pool. The data reported herein is useful to correct the standard range values for such parameters such as MCV or MCH that are affected hemochromatosis.

Approximately 10% of diferric transferrin is usually present in the plasma of normal persons and this satiates the iron requirements of erythropoiesis. However, transferrin saturation is typically elevated in untreated persons with hemochromatosis, including probands. Some hemochromatosis heterozygotes also have elevated values of transferrin saturation. After phlebotomy therapy, mean transferrin saturation values are lower but usually remain elevated. In the probands tested, mean MCV, MCH, and MCHC also decreased after phlebotomy therapy, but remained significantly elevated. Biochemical and electrophoretic characteristics of transferrin in hemochromatosis are normal, and there are no reports of possible mutations in the transferrin gene or its regulatory elements that could account for unusual cases of hemochromatosis. Thus, the increased transferrin saturation characteristic of hemochromatosis could cause increased iron uptake into developing erythroid cells by a transferrin-dependent mechanism. Non-transferrin-bound iron (NTBI) occurs in the plasma of most untreated hemochromatosis patients, and is often detectable after iron depletion therapy. NTBI is also present in the plasma of some hemochromatosis heterozygotes who have normal transferrin saturation. Further, NTBI can enter erythroid cells via a transferrin-independent pathway , although the extent to which this mechanism functions in hemochromatosis is unknown. Pre- incubation of a wide variety of cell types with ferric ammonium citrate results in marked stimulation of ⁵⁹Fe incoφoration from ⁵⁹Fe-transferrin at concentrations greater than those required for saturation of the transferrin receptor. However, this phenomenon has never been documented using erythroid cells. This nonetheless suggests that NTBI in the plasma of iron- overloaded patients may also promote increased loading of erythroid cells with iron derived from diferric transferrin in vivo.

In several extra-intestinal, non-erythroid cell types in hemochromatosis, transferrin receptor function and regulation appear to be normal. However, mRNA for transferrin receptor is inappropriately increased in the duodenum in hemochromatosis. A corresponding increase in erythroid cell surface transferrin receptor expression could explain increased transferrin-mediated iron uptake by erythroid cells in hemochromatosis. Interactions oft ransferrin receptor, transferrin, and mutant and wt HFE proteins observed in cultured cells could also explain some of the differences of iron absoφtion observed in persons with hemochromatosis and in normal subjects. Because HFE protein is not expressed in normal marrow erythroid cells or in K562 erythroleukemia cells, interaction of intracellular HFE with transferrin and transferrin receptor may not facilitate iron transport in erythroid cells. However, a role for soluble HFE protein in modifying erythroid cell iron uptake cannot be excluded. A divalent metal cation transporter (Nramp2, DCT1, DMTI) probably accounts for the increased entry of iron into enterocytes in hemocliromatosis. The same transporter removes transferrin-bound iron from endocytic vesicles in erythroid cells, and functions abnormally in the presence of mutant HFE protein or other factors that occur in hemochromatosis.

C282Y/C282Y probands who were HLA-A3-negative had higher mean MCH and MCHC values than HLA-A3-positive probands, and some probands without detectable HFE mutations also had abnormal erythrocyte parameters. This indicates that genetic factors other than HFE also influence peripheral blood erythrocyte parameters in hemochromatosis.

Example 1 : Evaluation of MCV and MCH to screen for hemochromatosis The most widely used screening test for hemochromatosis is the transferrin saturation test. This test was used in a standard strategy and with a standard case definition to initially identify hemochromatosis probands in routine medical care. Other tests for hemochromatosis include measuring unbound iron-binding capacity, serum iron concentration, or serum ferritin level, but these parameters are less sensitive, are less specific, or have not been evaluated as extensively as the transferrin saturation test. The data described below indicate that elevated values of MCV and MCHC (compared to non-hemochromatosis patients) can be used as a reliable test for hemochromatosis screening.

Using cut-off points of > 99.0 fL and > 97.0 fL, respectively, the sensitivity of MCV was 33% - 48% for men and 39% - 44% for women. Sensitivity of MCH was 57% - 70% for men and 39% - 50% for women at the corresponding cut-off points. In contrast, sensitivity of the transferrin saturation test is somewhat greater: 86% - 94% in men and 67% - 82% in women (cut-offs of > 60% and > 50%, respectively) when HLA haplotyping of family members is used as the standard case definition. However, specificity of MCV was 92% and 95% - 99% in men and women, respectively, and of MCH 79% - 98% and 86% - 98%, respectively. These values approach those of transferrin saturation testing: 93% - 99% specificity in men and 95% - 99% in women (transferrin saturation cut-offs of > 60% and > 50%, respectively). Using MCV, PV+ of 88% - 100% was observed in men and women; using MCH, PV+ of 75% - 77% was observed in men and 68% - 72% in women. In contrast, PV+ of an initial transferrin saturation screening test may be as low as 4% when HLA typing is used as the standard case definition; this increases to 68% when repeat testing is performed on a fasting specimen. An important advantage of the tests described herein is that unlike transferrin saturation, values of MCV and MCH are stable and not affected by fasting, because they reflect events which have influenced erythropoiesis for three - four months. Serial determinations of MCV and MCH demonstrated the relative stability of these indices in hemochromatosis. Unlike the standard transferrin saturation testing diagnostic tests, repeat measurement of MCV or MCH is unnecessary to improve PV+ for screening for individuals to identify individuals with hemochromatosis or at risk of developing the disorder.

Of American outpatients undergoing diagnostic testing, 1.7% - 3.6% had values of MCV > 100 fL. These observations suggest that the frequency of elevated MCV in persons receiving medical care is greater than the estimated prevalence of hemochromatosis in the general population. This could increase the ratio of false-positive to true-positive cases and thus reduce the PV+ of MCV or MCH testing. On the other hand, PV+ may be increased somewhat because control subjects and probands who had disorders (other than hemochromatosis) which cause macrocytosis, pseudomacrocytosis, or megaloblastosis were eliminated as part of the study design. The PV+ of screening tests for hemochromatosis also increases with increasing prevalence of this disorder. Hemochromatosis prevalence is greater in persons receiving testing or treatment for conditions associated with complications of iron overload than in the general population. For example, hemochromatosis occurred in 3.4% - 4.6% of white persons undergoing percutaneous liver biopsy, 1.2% - 3.5% of persons in a rheumatology clinic, and 0.5% - 1.0% of persons attending diabetes clinics. Persons diagnosed with hemochromatosis in routine medical care also have more severe iron overload and more frequent complications, on the average, than those detected by transferrin saturation screening of large populations. The PV+ of a single transferrin saturation test was 41% in persons with liver disease. In patients with diabetes mellitus, PV+ of initial and repeat transferrin saturation testing were 63% and 80%, respectively, when elevated hepatic iron stores were used as the standard case definition. As demonstrated herein, PV+ of > 80% and > 68% was obtained in all probands when MCV and MCH cut-off points of > 97.0 fL and >32.0 pg were used, respectively. Taken together, these observations indicate that the PV+ of MCV or MCH testing is greater in persons receiving medical care than in blood donors or in the general population.

The laboratory method used to determine MCV and MCH could affect the use of these indices for hemochromatosis screening. The coefficients of variation for MCV and MCH determined by automated and manual methods are approximately 1 % and 10%, respectively. Macrocytosis can be detected by microscopy of peripheral blood smears, but this method fails to identify approximately one-third of cases with increased mean MCV demonstrable with automated blood cell counters. Estimates of erythrocyte hemoglobin content by microscopy are unreliable. Automated determinations of MCV and MCH, e.g., using automated optical devices (Coulter Electronics, Hialeah, FL) are preferably used. Establishing and adhering to statistically valid reference ranges pertinent to the age of reference subjects, the specific automated blood cell counter, and the study population is necessary. As with other laboratory tests, maintaining appropriate standardization of automated blood cell counters in clinical practice is important. The data indicate that MCV or MCH cut-off points > 97.0 fL or >32.0 pg, respectively, are associated with greater sensitivity and PV+ for detection of C282Y homozygotes than persons who have "non-classical" HFE genotypes. C282Y homozygotes have iron overload of greater severity than persons with "non-classical" hemochromatosis, on the average. However, mean values of MCV and MCH in hemochromatosis are not significantly affected by the severity of iron overload, elevated serum concentrations of hepatic enzymes, hepatic cirrhosis, diabetes mellitus, arthropathy, or hypogonadism, although very severe iron overload may be associated with unusually high values of MCV in some cases. Taken together, these observations indicate that use of a MCV or MCH criterion allows the diagnosis of persons with hemochromatosis receiving medical care before they develop complications of iron overload. Thus, another advantage of the test is earlier detection of hemochromatosis or a predisposition to develop the disorder. Further, there is no specific additional cost for the measurement of MCV or MCH in persons who have undergone complete blood count testing for other reasons.

The data tabulated below consists entirely of observations from white persons of western European descent. In patients in Brittany with a syndrome of liver iron overload and normal transferrin saturation, mean values of MCV were normal, although some patients had elevated values. This suggests that using an elevated MCV or MCH value for screening could lead incidentally to the ascertainment of some persons with this disorder. However, forms of hereditary anemia which cause microcytosis and hypochromia of erythrocytes are common in persons of southern European and sub-Saharan African descent who have non- HFE hemochromatosis and other relatively common forms of iron overload. Thus using an elevated MCV or MCH criterion may be a less effective screening maneuver to identify persons at risk to have these iron overload disorders than those who may have hemochromatosis.

Hemochromatosis is a common cause of mildly or moderately increased MCV and MCH in white persons which (prior to the invention) was often unrecognized. No etiology was detected in 23% of persons with macrocytosis, suggesting that hemochromatosis could account for some unexplained occurrences of macrocytosis. Patients who have been identified as having elevated MCV or MCH according to the invention should undergo testing of fasting transferrin saturation in addition to routine diagnostic evaluation for non- hemochromatosis etiologies. Evaluation for possible iron overload should be performed thereafter, if indicated. Extreme macrocytosis is unusual in hemochromatosis, but is often associated with deficiency of vitamin B12 or folate. Anemia is rare in untreated persons with hemochromatosis, but is common among persons with increased MCV and MCH due to other causes. Hemochromatosis, like many etiologies of increased MCV or MCH, can be treated successfully in many patients. An advantage of the invention is that patients which require therapy are identified at an early stage in the development of the disorder, thereby allowing early intervention with improved success.

Hemochromatosis or a predisposition thereto was diagnosed as follows.

Selection of Hemochromatosis Probands and Normal Control Volunteers

Hemochromatosis probands were diagnosed during routine medical care delivery who had completed iron depletion with therapeutic phlebotomy. Probands with complications of iron overload (hepatic cirrhosis, diabetes mellitus, hemochromatosis-associated arthropathy, or hypogonadotrophic hypogonadism) were not excluded. At the time of diagnosis of hemochromatosis, each proband had a normal serum folate concentration, serum vitamin B12 concentration, and thyroid profile. Volunteer normal control subjects unrelated to our hemochromatosis probands were recruited from the general public and outpatient populations. Probands and control subjects who had (non-hemochromatosis) disorders known to cause increased values of MCV or MCH, macrocytosis, pseudomacrocytosis, or megaloblastosis were excluded. These disorders include alcoholism, increased erythrocyte mass, anemia other than that attributed to complications of iron overload, unexplained reticulocytosis, marrow infiltrative disease, leukemia or other malignancy, evidence of cold agglutinins or cryoglobulins, drug use associated with erythrocyte abnormalities, pregnancy, malabsoφtion, vegetarianism, dialysis requirements, transfusion dependency, and inherited disorders of DNA synthesis.

Evaluation of Iron Overload

The working diagnostic criterion for hemochromatosis of the American College of Pathologists: elevated fasting transferrin saturation was used (> 60% males, >50% females) on at least two occasions in the absence of other known causes. In exceptional cases (n = 3), transferrin saturation values were not consistently elevated, but otherwise unexplained iron overload consistent with hemochromatosis was demonstrated by analysis of hepatic biopsy specimens and calculation of hepatic iron index. HFE genotyping was not used to establish the diagnosis of hemochromatosis. Iron overload was defined as evidence of systemic iron overload demonstrated by otherwise unexplained elevated serum ferritin concentration (>300 ng/mL in men, >200 ng/mL in women), increased hepatic iron content determined using hepatic biopsy specimens, or iron > 4 g mobilized by phlebotomy. For each proband, the pre- phlebotomy serum transferrin saturation values and serum ferritin concentrations were tabulated, and the units of blood removed by therapeutic phlebotomy (1 unit of blood = 450 - 500 mL, equivalent to -200 mg of iron). Complications of iron overload were assessed as previously described. For each proband, the following parameters were tabulated the pre- phlebotomy serum transferrin saturation values and serum ferritin concentrations, and the units of blood removed by therapeutic phlebotomy.

Evaluation of Erythrocyte Parameters

Peripheral blood specimens obtained by antecubital venipuncture from probands and control subjects were analyzed using a Cell-Dyne® 1700 automated blood counter (Abbott Laboratories, Chicago, IL). Serum iron parameters (serum transferrin saturation and ferritin concentration), serum concentration of hepatic enzymes, and other blood chemistry determinations were performed using standard automated clinical laboratory methods.

Hepatic biopsy specimens were stained with hematoxylin and eosin, acid ferrocyanide, and Masson's trichrome techniques; additional portions of specimens were analyzed for iron content using atomic absoφtion spectrometry. HFE genotypes were determined using known methods.

Statistical Analysis The data set consisted of observations on 94 hemochromatosis probands and 132 normal control subjects. Sixty probands had a C282Y/C282Y HFE genotype; 65 normal control subjects had a wt/wt HFE genotype. The reference ranges for MCV and MCH measured using the Cell-Dyne® 1700 supplied by the manufacturer were 80.0 - 97.0 fL and 28.0 - 32.0 pg, respectively. Accordingly, values of MCV and MCH > 97.0 fL and > 32.0 pg fL, respectively, were the minimum cut-off-off points of diagnostic pertinence to all persons receiving routine medical care in the population examined. Sensitivity, specificity, and positive and negative predictive values (PV+ and PV-, respectively) were calculated for hemochromatosis using various cut-off points for MCV (> 97.0) and MCH (> 32.0 pg).

Characteristics of Study Subjects

The mean age of our 94 probands was 51 ± 13 years (range 20 - 80 years); 58 (61.7%) were men and 36 (38.3%) were women. There was no significant difference in the mean ages of men and women. Fifteen probands (16.0%) had hepatic cirrhosis, 10 (10.6%) had diabetes mellitus, 21 (22.3%) had hemochromatosis-associated arthropathy, and 15 (16.0%) had hypogonadotrophic hypogonadism. Iron overload was more severe and complications of iron overload were more frequent in C282Y/C282Y probands, on the average, than in probands with other HFE genotypes. Frequencies of HFE genotypes among the present probands were similar to those from unselected hemochromatosis probands from in the same geographic area. The mean age of our control subjects was 52 ± 15 years (range 18 - 86 years); 53 (40.2%) were men and 79 (59.8%) were women. There was no significant difference in the mean ages of men and women. Frequencies of HFE genotypes among the control subjects were similar to those previously reported from normal persons in the same geographic area. In probands and in normal control subjects, mean values of MCV were not significantly different in men and women. Forty-one probands (43.6%) and 5 normal control subjects (3.8%) had values of MCV that were greater than the reference range. No probands or normal controls had values of MCV below the reference range. Sensitivity, Specificity, and Predictive Values of MCV: Men vs. Women Sensitivity was 32.8% - 48.3% for men and 38.9% - 44.4% for women (cut-off points of > 99.0 fL and > 97.0 fL, respectively). Sensitivity was slightly greater for men at MCV cut-off points > 99.0 fL or greater (Table 5).

Table 5: Accuracy of MCV in Male and Female Hemochromatosis Patients

PV+ and PV- denote positive predictive value and negative predictive value, respectively. Specificity, PV+, and PV- were calculated using corresponding date from 53 control men and 79 control women.

Most values of specificity were slightly greater for women than men. PV+ at cut-off points of > 99.0 fL were slightly greater for women. PV- were greater for female probands at all cut-off points; at a cut-off point of > 97.0 fL, we observed values of 79.0% for women and 62.0% for men (Table 5).

Probands with "Classical" vs. "Non-classical" HFE Genotypes Sensitivity of MCV was greater for C282Y/C282Y probands than for probands who had "non-classical" HFE genotypes at all cut-off points (Table 6). At the > 97.0 fL cut-off point, sensitivity values of MCV for C282Y/C282Y probands and "non-classical" probands were 53.3% and 35.3%, respectively. Similarly, PV+ for C282Y/C282Y probands was greater at all MCV cut-off points. At cut off points > 101.0 fL, sensitivity values were greater for nonclassical probands (Table 2) Specificity was > 93.9% for all cut-off points for both groups of probands. PV+ was 11.8% and 6.0% for classical and nonclassical probands, respectively,for a cut- ff point > 101.0 fL and overall hemochromatois frequency 0.015 (80% classical and 20 % nonclassical cases). Values of PV+ were 100.0% for MCV > 103.0 fL for both proband groups PV- was > 99.0% for classical and nonclassical probands for all cut-off points and hemochromatosis frequencies (Table 6)

Table 6

Classical hemochromatosis probands are those with and HFE genotype C282Y/C282Y. Nonclassical probands are those with HFE genotypes other than HFE genotype C282Y/C282Y. Sensitivity and specificity values were calculated by comparing classical and nonclassical proband data with that of 132 control subjects, respectively.

Sensitivity, Specificity, and Predictive Values of MCH: Men vs. Women Sensitivity was 56.9% - 70.1% for men and 39.1% - 50.0% for women (cut-off points of > 32.5 pg and > 32.0 pg, respectively); sensitivity values were also greater for men at all other cut-off points tabulated (Table 3). Specificity was 79.3% - 98.1% for men and 86.1% - 97.5% for women over the range of cut-off points tabulated. PV+ was 74.6% - 95.0% for men and 67.7% - 89.5% for women over the range of cut-off points tabulated. PV- was slightly greater for women than men at corresponding all cut-off points (Table 7). Table 7

Probands with "Classical" vs. "Non-classical" HFE Genotypes. Sensitivity was 56.7% - 70.0% for men and 50.0% - 56.7% for C282Y homozygotes (cut-off points of > 32.5 pg and > 32.0 pg, respectively), and was greater for C282Y homozygotes than probands with "non-classical" HFE genotypes at all other cut-off points tabulated (Table 8). Specificity was 82.6% - 98.5% for C282Y homozygotes and 82.6% - 98.5% for probands with "non- classical" HFE genotypes over the range of cut-off points tabulated. PV+ was 64.6% - 89.5% for C282Y homozygotes and 47.7% - 80.0% for probands with "non-classical" HFE genotypes over the range of cut-off points tabulated. PV- was slightly greater for probands with "non-classical" HFE genotypes at all corresponding cut-off points (Table 8).

Table 8

Other embodiments are within the following claims.

Claims

What is claimed is:CLAIMS

1. A method of diagnosing hemochromatosis or a predisposition thereto in a mammal, comprising determining the mean coφuscular volume (MCV) value of a blood sample from said mammal, wherein an increase of at least 5% compared to a normal control value indicates that said mammal has hemochromatosis or is predisposed to developing hemochromatosis.

2. The method of claim 1, wherein said increase is at least 10% greater compared to a normal control value.

3. The method of claim 1 , wherein an erythrocyte parameter of said mammal is not indicative of anemia.

4. The method of claim 3, wherein said erythrocyte parameter is red blood cell count (RBC), hematocrit (Hct), or hemoglobin concentration (Hb).

5. A method of diagnosing hemochromatosis or a predisposition thereto in a mammal, comprising determining the mean coφuscular volume (MCV) value of a blood sample from said mammal, wherein a single MCV value is diagnostic of hemochromatosis or a predisposition to developing hemochromatosis.

6. The method of claim 5, wherein said MCV value is optically determined.

7. The method of claim 5, wherein said blood sample is obtained from a non-fasting mammal.

8. The method of claim 5, wherein said mammal is a white human of western European descent.

9. The method of claim 5, wherein said MCV value is greater than or equal to 101.0 fL.

10. The method of claim 9, wherein said MCV value is less than 120.0 fL.

1 1. The method of claim 9, wherein said MCV value is less than 110.0 fL.

12. The method of claim 9, wherein said MCV value is less than 106.0 fL.

13. A method of diagnosing hemochromatosis or a predisposition thereto in a mammal, comprising determining the MCV value of a blood sample from said mammal, wherein a value of at least 80 fL indicates that said mammal has hemochromatosis or is predisposed to developing hemochromatosis.

14. The method of claim 13, wherein said value is at least 85 fL.

15. The method of claim 13, wherein said value is in the range of 90-102 fL .

16. The method of claim 13, wherein an erythrocyte parameter of said mammal is not indicative of anemia.

17. The method of claim 16, wherein said erythrocyte parameter is RBC, Hct, or Hb.

18. The method of claim 16, wherein said erythrocyte parameter of said mammal is RBC.

19. The method of claim 16, wherein said erythrocyte parameter is Hct.

20. The method of claim 16, wherein said erythrocyte parameter is Hb.

21. A method of diagnosing hemochromatosis or a predisposition thereto in a mammal, comprising determining the mean coφuscular hemoglobin (MCH) value of a blood sample from said mammal, wherein a single MCH value is diagnostic of hemochromatosis or a predisposition to developing hemochromatosis.

22. The method of claim 21, wherein said MCH value is optically determined.

23. The method of claim 21, wherein said blood sample is obtained from a non-fasting mammal.

24. The method of claim 21, wherein said mammal is a white human of western European descent.

25. The method of claim 21, wherein said MCH value is greater than or equal to 33.0 pg.

26. The method of claim 21, wherein said mammal is male and said MCH value is greater than or equal to 34.0 pg.

27. The method of claim 21 , wherein said mammal is female and said MCH value is greater than or equal to 32.0 pg.

28. A method of diagnosing hemochromatosis or a predisposition thereto in a mammal, comprising determining the mean coφuscular hemoglobin (MCH) value of a blood sample from said mammal, wherein an increase of at least 5% compared to a normal control value indicates that said mammal has hemochromatosis or is predisposed to developing hemochromatosis.

29. The method of claim 28, wherein said increase is at least 10% greater compared to a normal control value.

30. The method of claim 28, wherein an erythrocyte parameter of said mammal is not indicative of anemia.

31. The method of claim 30, wherein said erythrocyte parameter is red blood cell count (RBC), hematocrit (Hct), or hemoglobin concentration (Hb).

32. A method of diagnosing hemochromatosis or a predisposition thereto in a mammal, comprising determining the MCH value of a blood sample from said mammal, wherein a value of at least 26 pg indicates that said mammal has hemochromatosis or is predisposed to developing hemochromatosis.

33. The method of claim 32, wherein said value is at least 28 pg.

34. The method of claim 32, wherein said value is in the range of 30-35 pg.

35. The method of claim 32, wherein an erythrocyte parameter of said mammal is not indicative of anemia.

36. The method of claim 32, wherein said erythrocyte parameter is RBC, Hct, or Hb.

37. A method for identifying mammals with an increased risk for heavy metal poisoning, comprising determining the MCV value of a blood sample from a mammal, wherein an increase in MCV value of said sample compared to a normal control value indicates that said mammal has an increased risk of developing poisoning upon exposure to a heavy metal.

38. The method of claim 37, wherein said heavy metal is selected from the group consisting of lead, cadmium, mercury, bismuth, and uranium.

39. A method for identifying mammals with an increased risk for heavy metal poisoning, comprising determining the MCH value of a blood sample from a mammal, wherein an increase in the MCH value of said sample compared to a normal control value indicates that said mammal has an increased risk of developing poisoning upon exposure to a heavy metal.

40. The method of claim 39, wherein said heavy metal is selected from the group consisting of lead, cadmium, mercury, bismuth, and uranium.

41. A method for identifying mammals with increased physical endurance, comprising determining the MCV value of a blood sample from a mammal, wherein an increase in MCV value of said sample compared to a normal control value indicates that said mammal has an increased capacity for physical endurance.

42. A method for identifying mammals with increased physical endurance, comprising determining the MCH value of a blood sample from a mammal, wherein an increase in the MCH value of said sample compared to a normal control value indicates that said mammal has an increased capacity for physical endurance.