Prevalence of Thalassemia in Patients With Microcytosis Referred for Hemoglobinopathy Investigation in Ontario A Prospective Cohort Study

Transcription

1 Hematopathology / PREVALENCE OF THALASSEMIA IN ONTARIO Prevalence of Thalassemia in Patients With Microcytosis Referred for Hemoglobinopathy Investigation in Ontario A Prospective Cohort Study John D. Lafferty, ART, 1 David S. Barth, FRCP(C), 2 Brian L. Sheridan, FRCP(C) 3 Andrew G. McFarlane, ART, 4 Linda M. Halchuk, MLT, 4 and Mark A. Crowther, FRCP(C) 1 Key Words: Prevalence; Thalassemia; Hemoglobinopathy; Microcytic anemia Abstract In Ontario, Canada, β-thalassemia is easily detected through measurement of hemoglobin A 2, but most laboratories do not do exhaustive DNA investigations for α-thalassemia. Therefore, the prevalence of thalassemia in microcytic samples for hemoglobinopathy investigation in Ontario is unknown. To address this, we performed a prospective cohort study in which samples referred for hemoglobinopathy investigation were also evaluated for α-thalassemia by DNA testing. Of 800 samples submitted, 664 were evaluable. Of the 664 patients represented, 163 (24.5%) were β-thalassemia major carriers, 68 (10.2%) were hemoglobin Bart s hydrops fetalis carriers and, in total, 361 (54.4%) had some form of thalassemia. We conclude that microcytosis due to thalassemia is common in Ontario and that major forms of thalassemia, including forms predisposing to hemoglobin Bart s hydrops fetalis and β-thalassemia major, are frequent. This illustrates the importance of adequate prenatal and laboratory investigation for these abnormalities in Ontario and other similar multiethnic jurisdictions worldwide. Thalassemia syndromes are among the most common inherited abnormalities worldwide. Clinical symptoms vary from mild microcytic anemia in heterozygous thalassemia trait to life-threatening anemia in homozygous or compound heterozygous thalassemia major. 1-3 Thalassemia is caused by genetic abnormalities of the α- or β-globin genes of hemoglobin A, resulting in decreased production of the corresponding α- or β-globin chains of the hemoglobin A molecule. 4 Abnormalities fall into 3 broad categories 1,2,5,6 : (1) gene deletions that eliminate the whole gene, (2) mutations within the gene that decrease globin production, and (3) variant hemoglobin that is produced in decreased amounts. α-thalassemia, which is common in individuals of Southeast Asian, African, Middle Eastern, and Mediterranean ethnic background, is usually caused by a deletion of 1 or more of the 4 inherited α-globin genes on chromosome 16 (αα/αα). Common single α-gene deletions are the rightward 3.7-kilobase (kb; -α 3.7 /) and leftward 4.2-kb (-α 4.2 /) deletions. Hemoglobin Constant Spring (α CS α/) is the most common hemoglobin variant causing α-thalassemia. These mutations manifest as heterozygous (eg, -α 3.7 /αα), homozygous (eg, -α 3.7 /-α 3.7 /), and compound heterozygous (eg, α CS α/-α 4.2 ) genotypes and vary phenotypically from normal to mild microcytic anemia. The most common 2 α-gene cis deletions are the Southeast Asian (-- SEA /), Filipino (-- FIL /) and Mediterranean (-- MED /) deletions. These can manifest as heterozygous α-thalassemia trait (eg, -- SEA /αα, -- FIL /αα, or -- MED /αα), which vary phenotypically from mild microcytosis to mild microcytic anemia, or as compound heterozygous hemoglobin H disease (eg, -α 3.7 /-- SEA ), which manifests as moderate microcytic anemia. Hemoglobin Bart s hydrops fetalis (α-thalassemia 192 Am J Clin Pathol 2007;127: Downloaded 192 from

2 Hematopathology / ORIGINAL ARTICLE major) is caused by homozygous (eg, -- SEA /-- SEA ) or compound heterozygous (eg, -- SEA /-- FIL ) inheritance of 2 α-gene cis deletions and causes serious obstetric complications and fetal death. 1,5,7,8 Only people with genotypes containing 2 α- gene cis deletions, ie, α 0 -thalassemia trait and hemoglobin H disease, are carriers of hemoglobin Bart s hydrops fetalis. 7,9 Carrier status is most common in patients of Southeast Asian ethnic background in whom the frequency of the SEA deletion is between 4% and 14%. 7,8 β-thalassemia is common in persons of Mediterranean, Arab, African, Indian, and Southeast Asian ethnic background. It is usually the result of mutations within the β-globin gene that decrease β-globin production. The 2 inherited β- globin genes are found on chromosome 11. To date, many mutations in the β gene have been described ( Other common β-thalassemia syndromes include deletional δβ-thalassemia and the variants hemoglobin E and hemoglobin Lepore. β-thalassemia major is caused by homozygous or compound heterozygous inheritance of 2 β-thalassemia mutations and manifests as severe hypochromic, hemolytic anemia requiring lifelong transfusion and chelation therapy. Carrier genotypes of β-thalassemia major include β-thalassemia trait, δβ-thalassemia trait, hemoglobin E trait, hemoglobin Lepore trait, and hemoglobin E disease and vary phenotypically from mild microcytosis to mild microcytic anemia. 1,2,6 Identification of thalassemia is important for 2 reasons: (1) to identify pregnancies or planned pregnancies at risk of thalassemia major so that appropriate genetic counseling can be offered and (2) to prevent unnecessary and potentially harmful medical intervention and iron therapy in patients with microcytic anemia due to thalassemia. 5,10,11 We present the results of a prospective multicenter study performed in Ontario, Canada, which was designed to demonstrate the prevalence of thalassemia in patients with microcytosis who have samples submitted for hemoglobinopathy investigation. A priori we anticipated finding a high prevalence of thalassemia and finding unusual combinations of thalassemia genotypes given the heterogeneous nature of Ontario s population. Materials and Methods Two Ontario laboratories participated in the study (Toronto Medical Labs, Toronto, Canada, and Hamilton Regional Laboratory Medicine Program [HRLMP], Hamilton, Canada). Research ethics approval for the project was obtained at both participating sites before initiation. Samples were considered eligible if processed in 1 of the 2 referring laboratories, a hemoglobinopathy investigation was ordered, the mean corpuscular volume was less than 80 µm 3 (<80 fl), and the patient was 18 years or older. Samples with a volume of less than 2 ml were excluded. All specimens were collected into EDTA evacuated blood collection tubes. Initial testing was performed at the referring laboratory and included CBC count, hemoglobin variant detection, hemoglobin A 2 and hemoglobin F quantitation by BioRad Variant or Variant II high-performance liquid chromatography (BioRad Laboratories, Hercules, CA), and a standardized manual hemoglobin H inclusion body screen. The standardized hemoglobin H inclusion body screen used commercially prepared brilliant cresyl blue (BDH Chemicals, Middlesex, England) mixed in a ratio of 4 parts whole blood and 1 part stain. The mixture was incubated for 1 hour at 37 C before the preparation of 2 blood films. Each blood film was examined for 2 minutes for hemoglobin H inclusion bodies by experienced laboratory operators. 12 Following the reporting of initial results, specimens were shipped at 4 C to the HRLMP Hemoglobinopathy Reference Laboratory for further investigation. At HRLMP, testing included ferritin, free erythrocyte protoporphyrin, and gap polymerase chain reaction (PCR) for the 5 most common α- thalassemia deletions, ie, 3.7 kb, 4.2 kb, SEA, MED, and FIL. In addition, a novel screening test for α-thalassemia that is based on the detection of ζ-globin using an enzyme-linked immunosorbent assay was performed on all samples (United Biotech, Mountain View, CA). 13,14 All subsequent diagnoses of α-thalassemia by gap PCR were reported to the referring physician as a condition of research ethics board approval. Before initiating the study, we modeled the implications of a variety of sample sizes on our ability to detect important rates of individual hemoglobinopathies. We estimated, based on earlier studies, 15,16 that we would need to include samples from 1,200 patients to adequately test the diagnostic usefulness of our evaluative regimen. Subsequently, we found that the frequency of thalassemic states was significantly higher than in our previous studies, likely owing to sample preselection and changes in the demographic characteristics of our patient population (as a result of immigration from highprevalence regions). As a result, we were able to reduce our sample size to 800 without a reduction in our ability to demonstrate the frequency of different hemoglobinopathies. To reduce the likelihood that regional variation in ethnic distribution could affect our results, we tested samples from 2 large regional referral laboratories receiving samples from geographically distinct areas. The HRLMP serves a large urban population derived from the Mediterranean basin, Southeast Asia, and East Africa and patients referred from other urban areas in the province. The Toronto Medical Labs serves a diverse multicultural urban population derived from the Mediterranean basin, the Middle East, India, Africa, and Southeast Asia. In total, we sought 400 consecutive samples from each of these laboratories. Downloaded from Am J Clin Pathol 2007;127:

3 Lafferty et al / PREVALENCE OF THALASSEMIA IN ONTARIO Results Of 800 samples submitted for the study, 136 were excluded. The most common reasons for exclusion were inadequate DNA extracted for analysis (132 samples) and inadequate sample volume (4 samples) for complete hemoglobinopathy investigation. Table 1 β-thalassemia Syndromes Detected Syndrome β-thalassemia trait * 147 Hemoglobin E trait * 9 δβ-thalassemia trait * 3 Homozygous hemoglobin E * 2 Hemoglobin Lepore trait * 2 None detected 501 * Denotes β-thalassemia major carrier genotype. Table 2 α-thalassemia Syndromes Detected Syndrome Single α-gene deletion Heterozygous (-α 3.7 /αα) 110 Homozygous (-α 3.7 /-α 3.7 ) 37 Heterozygous (-α 4.2 /αα) 3 Homozygous (-α 3.7 /-α 4.2 ) 2 Homozygous (-α 4.2 /-α 4.2 ) 1 Hemoglobin Constant Spring/single α-gene deletion 2 (α CS α/-α 3.7 ) Heterozygous hemoglobin Constant Spring (α CS α/αα) 1 α-thalassemia trait -- SEA /αα * FIL /αα * MED /αα * 3 Hemoglobin H disease -α 3.7 /-- SEA* 7 -α IVS-1-5nt deletion /-- MED* 1 -α 3.7 /-- MCU* 1 -α 3.7 /-- FIL* 1 None detected 440 FIL, Filipino; MED, Mediterranean; SEA, Southeast Asian. * Denotes hemoglobin Bart s hydrops fetalis carrier genotype. Table 3 Nonthalassemic Hemoglobinopathies Detected Hemoglobinopathy Hemoglobin S 27 Hemoglobin C 7 Hemoglobin D 3 Rare variant 3 Hemoglobin G 1 Hemoglobin Hofu 1 None detected 622 Thalassemic states were identified frequently. The β-thalassemia syndromes detected are given in Table 1. Of 664 patients, 163 (24.5%) had carrier genotypes of β-thalassemia major. The α-thalassemia syndromes detected are given in Table 2. Of 664 patients, 224 (33.7%) had some form of α- thalassemia; 68 (10.2%) had 2 α-gene cis deletions and, thus, are carriers of hemoglobin Bart s hydrops fetalis. Of 664 patients, 42 (6.3%) had other hemoglobinopathies not usually associated with microcytosis Table 3. Table 4 lists the single and multiple mutation hemoglobinopathies detected. Overall, 314 patients (47.3%) were identified as having a single α- or β-thalassemia genotype, 24 (3.6%) had α- and β-thalassemia genotypes, 23 (3.5%) had an α- or a β-thalassemia Table 4 Various Single and Multiple Mutation Hemoglobinopathies Detected Hemoglobinopathy β-thalassemia trait 124 and (-α 3.7 /αα) 15 and (-α 3.7 /-α 3.7 ) 4 and (-- MED /αα) 2 α-thalassemia trait -- SEA /αα FIL /αα MED /αα 1 δβ-thalassemia trait 3 Single α-gene deletion Heterozygous (-α 3.7 /αα) 78 Homozygous (-α 3.7 /-α 3.7 ) 29 Heterozygous (-α 4.2 /αα) 3 Homozygous (-α 3.7 /-α 4.2 ) 2 Homozygous (-α 4.2 /-α 4.2 ) 1 Hemoglobin Constant Spring/single α-gene 1 deletion (α CS α/-α 3.7 ) Hemoglobin S trait 10 and (-α 3.7 /αα) 10 and (-α 3.7 /-α 3.7 ) 4 Hemoglobin E trait 8 and (-α 3.7 /αα) 1 Hemoglobin C trait 4 and (-α 3.7 /αα) 3 Hemoglobin D trait 3 Hemoglobin Lepore trait 2 Hemoglobin G and (-α 3.7 /αα) 1 Hemoglobin Hofu and hemoglobin H disease 1 (-α 3.7 /-- SEA ) Hemoglobin S disease and (-α 3.7 /αα) 1 Hemoglobin S/β-thalassemia and (-α 3.7 /αα) 1 Hemoglobin S/β+-thalassemia 1 Unidentified hemoglobin variant 1 and (-α 3.7 /αα) 1 and (-- SEA /αα) 1 Hemoglobin E disease and (α CS α/-α 3.7 ) 1 and hemoglobin H disease (-α 3.7 /-- SEA ) 1 Hemoglobin H disease -α 3.7 /-- SEA 5 -α 3.7 /-- FIL 1 -α 3.7 /-- MCU 1 -α IVS-1-5nt deletion /-- MED 1 None detected 284 FIL, Filipino; MED, Mediterranean; SEA, Southeast Asian. 194 Am J Clin Pathol 2007;127: Downloaded 194 from

4 Hematopathology / ORIGINAL ARTICLE genotype in combination with a nonthalassemic hemoglobin variant, 18 (2.7%) had a single nonthalassemic hemoglobin variant, and 1 (0.2%) had α- and β-thalassemia genotypes in combination with a nonthalassemic hemoglobin variant. Table 5 lists the causes of microcytosis identified. Of 664 patients, 362 (54.5%) had thalassemia as a cause of microcytosis. In 287 (43.2%), microcytosis due solely to thalassemia; 75 cases (11.3%) were due to a combination of thalassemia and iron deficiency defined by a ferritin level of less than 18 ng/ml (<18 µg/l); 166 cases (25.0%) were due solely to iron deficiency and 61 (9.2%) to iron deficiency and/or anemia of chronic disease defined as a ferritin level of more than 17 ng/ml (>17 µg/l) and a free erythrocyte protoporphyrin level of more than 85 µg/dl (>1.5 µmol/l) of RBCs. No obvious cause of microcytosis was detected in 75 patients (11.3%). Discussion Microcytic anemia is the most common hematologic abnormality encountered by physicians. The study results confirm that the current ethnic diversity of the Ontario population makes thalassemia a common cause of microcytosis; a thalassemic state was detected in 54.5% of the patients in this cohort. Of 664 patients, 231 (34.8%) were carriers of a thalassemic state that put them at risk of having children with thalassemia major: 163 (24.5%) β-thalassemia major and 68 (10.2%) hemoglobin Bart s hydrops fetalis. This high prevalence of thalassemia carrier genotypes demonstrates the importance of investigating for microcytosis and, if present, thalassemia trait in patients who are pregnant or planning a pregnancy and suggests that newborn screening for α-thalassemia by using hemoglobin Bart s detection by high-performance liquid chromatography may provide useful information for subsequent pregnancies. In our cohort, of 362 patients with thalassemia, 75 (20.7%) had concomitant iron deficiency and another 47 (13.0%) had multiple thalassemia syndromes or thalassemia Table 5 Causes of Microcytosis Cause Iron deficiency 166 α-thalassemia syndrome 149 β-thalassemia syndrome 119 Iron deficiency or anemia of chronic disease 61 α-thalassemia syndrome and iron deficiency 50 β-thalassemia syndrome and iron deficiency 19 α-thalassemia and β-thalassemia syndromes 19 α-thalassemia, β-thalassemia syndrome, and 6 iron deficiency No cause determined 75 in conjunction with a nonthalassemic hemoglobin variant. This confirms that thalassemia, hemoglobin variants, and iron deficiency are independent events, and the diagnosis of iron deficiency or an easily detected thalassemia (eg, β-thalassemia trait diagnosed based on an elevated hemoglobin A 2 level) does not exclude the presence of a more difficult to diagnose coincident abnormality (eg, α-thalassemia). This is important in patients with microcytosis who are pregnant or planning a pregnancy, in whom a thorough hemoglobinopathy investigation, including DNA analysis for common forms of α-thalassemia, is required to identify patients at risk of thalassemia major. Our study had significant limitations. Most important, our patient samples do not truly represent the prevalence of thalassemia in microcytic patients in our population because we analyzed only samples submitted for hemoglobinopathy investigations to 2 tertiary care laboratories. Many patients with simple microcytic anemia due to iron deficiency would have been excluded from the referral population because a thalassemia investigation is often a secondary investigation in patients with microcytosis. Finally, we did not collect data on patient ethnicity or, in fact, extensive patient demographics to satisfy the requirements of our research ethics board. Obtaining such data would have required specific consent from individual patients, making the study unmanageably complex. Strengths of this study included a comprehensive laboratory investigation with routine PCR analysis for α-thalassemia on all samples, irrespective of the presence or absence of other hemoglobinopathies, thalassemias, or iron deficiency. To our knowledge, this is the first study to perform such a systematic analysis on a large cohort of mixed ethnicity patients evaluated outside hyperendemic regions for thalassemia trait. A similar study, performed in Montreal, excluded patients with iron deficiency, β-thalassemia trait, and other thalassemic hemoglobinopathies. 17 This protocol in our study would have excluded 75 cases of α-thalassemia, resulting in an erroneous 20.7% reduction in the number of thalassemia cases detected. Of these, 13 would have been carriers of hemoglobin Bart s hydrops fetalis, resulting in an erroneous 19.1% reduction in the number of these carriers detected. Furthermore, by including samples from 2 referral laboratories, we maximized the ethnic heterogeneity of the underlying population, thus more accurately reflecting the typical ethnic diversity found in large urban centers in Ontario. We conclude that there is a high prevalence of clinically important thalassemic genotypes, including genotypes placing patients at risk for hemoglobin Bart s hydrops fetalis and β-thalassemia major, in patients with microcytosis referred for hemoglobinopathy investigations in large urban centers in Ontario. Such genotypes were commonly found in patients with other disorders that might provide a sufficient explanation for microcytosis, eg, iron deficiency Downloaded from Am J Clin Pathol 2007;127:

5 Lafferty et al / PREVALENCE OF THALASSEMIA IN ONTARIO or β-thalassemia trait. This demonstrates the concerning potential for underdiagnosis of clinically important thalassemic genotypes when investigations are discontinued on the basis of finding another cause for microcytosis. We further conclude that these data support the assertion that a formal adequate prenatal screening strategy needs to be implemented in the province. 18 This strategy should include adequate laboratory investigation for α- and β-thalassemia syndromes irrespective of other causes of microcytosis. This is especially important for patients of high-risk ethnic backgrounds but should not be restricted to these groups given the increasingly complex ethnic diversity of Ontario s population. The variety of single and multiple hemoglobinopathies seen in this study attests to this complex ethnic diversity. These results are of immediate relevance to hematologists, hematopathologists, and pathologists who supervise laboratories performing hemoglobinopathy testing, particularly in areas of Canada, the United States, and the industrialized world where there is significant ethnic heterogeneity similar to that found in Ontario. From 1 McMaster University, Hamilton Regional Laboratory Medicine Program, St Joseph s Healthcare, Hamilton; 2 University of Toronto, Toronto Medical Labs, Princess Margaret Hospital, Toronto; 3 MDS Laboratories, Hamilton; and 4 Hamilton Regional Laboratory Medicine Program, St Joseph s Healthcare, Hamilton, Canada. Funded by the Thalassemia Foundation of Canada, Toronto. Address reprint requests to Dr Crowther: Room L208, St Joseph s Hospital, 50 Charlton Ave East, Hamilton, Canada L8N 4A6. Acknowledgments: We acknowledge the staff of the following laboratories for their assistance in performing the sample selection, analysis, and data collection for this study: Hamilton Regional Laboratory Medicine Program, Hemoglobinopathy Reference Laboratory, St Joseph s Healthcare; McMaster University Medical Centre, Hamilton; and the Toronto Medical Laboratory, Princess Margaret Hospital, Toronto, Canada. References 1. Dumars KW, Boehm C, Eckman JR, et al. Practical guide to the diagnosis of thalassemia. Am J Med Gen. 1996;62: Rund D, Rachmilewitz E. β-thalassemia. N Engl J Med. 2005;353: Lafferty J, Waye JS, Chui DH, et al. Good practice guidelines for laboratory investigation of hemoglobinopathies. Lab Hematol. 2003;9: Weatherall DJ. The thalassemias: the role of molecular genetics in an evolving global health problem. Am J Hum Genet. 2004;74: Waye JS, Chui DHK. The alpha-globin gene cluster: genetics and disorders. Clin Invest Med. 2001;24: Olivieri N. The β thalassemias. N Engl J Med. 1999;341: Chui DHK, Waye JS. Hydrops fetalis caused by α thalassemia: an emerging health care problem. Blood. 1998;91: Chui DHK, Fucharoen S, Chan V. Hemoglobin H disease: not necessarily a benign disorder. Blood. 2003;101: Modell M, Modell B. Genetic screening for ethnic minorities. Br Med J. 1990;300: Lafferty J. The laboratory diagnosis of α thalassemia. Can J Med Lab Sci. 1998;60: Bain BJ, Amos RJ, Bareford D, et al. Guideline: the laboratory diagnosis of haemoglobinopathies. Br J Haematol. 1998;101: Dacie JV, Lewis SM. Investigations of abnormal haemoglobins and thalassemia. In: Practical Haematology. 6th ed. Edinburgh, Scotland: Churchill Livingstone; 1995: United Biotech. α-thalassemia (-SEA) ζ Globin Enzyme Immunoassay [package insert]. Mountain View, CA: United Biotech; Lafferty JD, Crowther MA, Waye JS, et al. A reliable screening test to identify adult carriers of the (-- SEA ) alpha 0 - thalassemia deletion: detection of embryonic zeta-globin chains by enzyme-linked immunosorbent assay. Am J Clin Pathol. 2000;114: Lafferty JD, Crowther MA, Ali MAM, et al. The evaluation of various mathematical indices and their efficacy in discriminating between thalassemic and nonthalassemic microcytosis. Am J Clin Pathol. 1996:106: Ali MAM, Lafferty J. The clinical significance of hemoglobinopathies in the Hamilton region: a twenty year review. Clin Invest Med. 1992;15: Bergeron J, Weng X, Robin L, et al. Prevalence of α-globin gene deletions among patients with unexplained microcytosis in a North American population. Hemoglobin. 2005;29: Basran RK, Patterson M, Walker L, et al. Prenatal diagnosis of hemoglobinopathies in Ontario, Canada. Ann N Y Acad Sci. 2005;1054: Am J Clin Pathol 2007;127: Downloaded 196 from