- Email: [email protected]
Screening for hereditary haemochromatosis within families and beyond
RESEARCH LETTERS
Screening for hereditary haemochromatosis within families and beyond C Anne McCune, David Ravine, Mark Worwood, Helen A Jackson, H Martyn Evans, David Hutton Screening programmes for haemochromatosis that include follow-up identification of relatives are claimed to be cost effective. We assessed uptake of screening by first-degree relatives of two groups of index cases: people homozygous for the C282Y mutation ascertained by genetic screening of blood donors; and patients presenting clinically with haemochromatosis. Only 40 (24%) of 165 relatives of blood donors had been tested. By contrast, testing uptake in 121 relatives of patients diagnosed clinically was more than double that (53%), despite unstructured provision of genetic information. A substantial number of untested relatives had undiagnosed iron overload. Overall efficacy of population screening for haemochromatosis is undermined by these observations.
Lancet 2003; 362: 1897–98
Hereditary haemochromatosis is a recessive disorder of iron accumulation. Most affected individuals are homozygous for an aminoacid substitution (C282Y) arising from a single mutation in the haemochromatosis gene (HFE). In the UK, more than 90% of patients with this disorder are C282Y homozygotes, as are about one in 150 people in the general population.1 Advocates of screening point to the high frequency of C282Y mutations in northern Europeans and availability of an effective treatment. However, results of research-based population screening programmes have suggested that the clinical penetrance of the gene is substantially lower than previously reported, with increasing uncertainty and controversy2,3 about the overall clinical importance of homozygosity for this mutation. By contrast, there is a powerful case for screening relatives of cases presenting clinically. Siblings have the highest risk (one in four) of iron overload and should be offered screening (http://www.bcshguidelines.com). Several groups have addressed the cost-effectiveness of population screening.4 When studies include follow-up identification of relatives, cost-effectiveness is substantially improved.4 However, none provides guidance on how to offer testing to relatives or address the probable uptake within families. We aimed to look into how information about genetic risk is transmitted through families, and to assess uptake of genetic screening within a large population of young healthy adults—the perceived ideal target population for haemochromatosis screening. We investigated first-degree relatives of two groups of index cases: C282Y homozygotes ascertained through a genetic-screening programme of healthy blood donors;1 and haemochromatosis patients who presented clinically. Identification of 72 individuals homozygous for the C282Y mutation from 10 556 blood donors in 1998 is described elsewhere.1 Written information about haemochromatosis
and inheritance was provided at enrolment. We contacted every homozygous individual and offered them an interview with a doctor to discuss results and arrange treatment as necessary. Inheritance of haemochromatosis was also explained, and we provided detailed information about family testing with an offer to counsel and test relatives. We supplied an additional factsheet reinforcing the genetic risk to close relatives. Patients identified within the three health authorities in south Wales were contacted after their hospital consultant had given permission.5 All patients (age 18–70 years) had presented clinically, rather than by family or population screening, and all were homozygous for C282Y. After obtaining ethics approval from BroTaf, Gwent, and Iechyd Morgannwg regional ethics committees, we sent letters in 2001 to all index cases asking them to consider discussing the present study with available first-degree relatives (parents and adult siblings), including those previously tested for haemochromatosis. A doctor interviewed consenting relatives, usually at home, because most people were reluctant to visit hospital since this journey often entailed travelling some distance; blood samples were taken at this visit. People living outside south Wales were interviewed by telephone, with blood samples taken locally and forwarded by post. Tests arranged on all samples included: transferrin saturation, serum ferritin, and HFE genotyping. We deemed transferrin saturation greater than 50% or serum ferritin greater than 300 g/L (greater than 200 g/L in premenopausal women) abnormal. Family details were available for 66 of the 72 individuals homozygous for C282Y (median age at screening 38 years [IQR 28–45], and 61 families responded to the letter inviting their participation. 56 families were finally available for testing. Of 180 eligible relatives, 165 (92%) agreed to take part. 105 (64·4%) of these knew of their relative’s diagnosis (table). However, only a third were aware of their own risk; overall, a quarter had been tested for haemochromatosis. After interview, 163 (99%) chose to proceed with testing. 25 C282Y homozygotes were identified (15 women), of whom three women and seven men (40%) also had a raised transferrin saturation and serum ferritin. Family details were available for 60 clinical probands (median age at diagnosis 51 years [IQR 45–58]). 121 (85%) of 143 eligible first-degree relatives were interviewed (table). By contrast with relatives of blood donors, almost all knew of the index case’s diagnosis (97%). Of these, 66 (55%) also had some awareness of their own risk and most had been tested. Like relatives of donors, a substantial proportion had not appreciated their own high genetic risk despite awareness of the disorder in their relative. All previously untested relatives elected to be investigated. 34 were C282Y homozygotes (19 women), of whom eight women and 12 men (59%) also had a raised transferrin saturation and serum ferritin. Overall, 59 people homozygous for C282Y were recorded, and half had excessive iron stores. As a result, 16 new referrals were made for venesection. Relatives of cases discovered by genetic Relatives of patients testing (n=165) (n=121)
Outcome Aware of own genetic risk and tested for haemochromatosis 40 (24%)* Aware of own genetic risk but not tested for haemochromatosis 16 (10%)† Aware of diagnosis in first-degree relative but no knowledge of own genetic risk 49 (30%) Unaware of diagnosis in first-degree relative 60 (36%)
64 (53%) 2 (2%)‡ 51 (42%) 4 (3%)
Data are number of people (%). *Difference between proportions of relatives tested in the two groups, p<0·0001. †Advised against testing by health-care professional (n=7), unsure how to arrange testing (n=1), not yet arranged testing (n=5), active decision not to test (n=3). ‡Advised against testing by health-care professional.
Uptake of testing in first-degree relatives of C282Y homozygotes discovered by genetic testing and after clinical presentation
THE LANCET • Vol 362 • December 6, 2003 • www.thelancet.com
1897
For personal use. Only reproduce with permission from The Lancet publishing Group.
RESEARCH LETTERS
Why was the difference in screening uptake between groups so notable? Perhaps concern for a sick family member was absent in relatives of cases detected by screening, and our attempt to provide good quality information seemed no match for the absence of stimulus provided by a clinical disorder. Although a simple observation, the possibility of a muted response from high-risk family members has not been considered in the debate about population screening for haemochromatosis. Nevertheless, overall outcome in relatives of clinical cases cannot be regarded as satisfactory, and a substantial number had undiagnosed iron overload. With our approach, which ensured skilled counselling and testing, uptake of screening in both groups of relatives rose to more than 98%. Such a proactive approach offers a possible solution to the difficulty of poor screening uptake in relatives. Whether such a high uptake could be achieved in a routine service setting remains to be seen. In assessing screening of a hypothetical cohort of 10 000 blood donors, Adams and colleagues4 assumed that all identified probands would have two siblings, each at 25% risk. Screening was regarded as cost-effective, and including siblings within the model substantially reduced costs. Similarly, Baer and colleagues4 reported a favourable cost of US$17 000 per case of haemochromatosis identified, as long as affected relatives were included in cost calculations. We have shown that high uptake of testing by relatives cannot be assumed. Blood donors are thought to be well motivated, and one might expect higher levels of testing in their relatives. However, even in this group we reported unacceptably low rates of uptake. This rate is improved by a proactive approach, which has additional cost implications that, until now, have not been considered. Future assessments of general population screening must take these costs into account. Contributors C A McCune, D Ravine, and M Worwood designed the study, analysed data, and prepared the report. C A McCune obtained data and wrote the first draft of the report. D Hutton and H A Jackson contributed to design of the study and provided data on index cases (blood donors and patients). H M Evans contributed to design and ethical aspects of the study. All authors reviewed the report and approved the final version.
1898
Conflict of interest statement None declared.
Acknowledgments We thank the blood donors, patients, and their families for taking part in this study; Jeanne Kingston (University of Wales College of Medicine) for HFE genotyping of blood samples; Kimberley Carter (University of Wales College of Medicine) for coordinating sample storage; Richard Ellis (University Hospital of Wales) for assays for iron and ferritin; and Carole Rabaiotti (University of Wales College of Medicine) for invaluable secretarial support. This work was funded by a grant awarded by the Wales Office of Research and Development in July, 2001. The sponsor of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. 1
2
3
4
5
Jackson HA, Carter K, Darke C, et al. HFE mutations, iron deficiency and overload in 10 500 blood donors. Br J Haematol 2001; 114: 474–84. Beutler E. The HFE Cys282Tyr mutation as a necessary but not sufficient cause of clinical hereditary hemochromatosis. Blood 2003; 101: 3347–50. Ajioka RS, Kushner JP. Clinical consequences of iron overload in hemochromatosis homozygotes. Blood 2003; 101: 3351–53. Cappuccio J, Phatak PD. Cost-effectiveness of screening for hemochromatosis. In: Barton JC, Edwards CQ, eds. Hemochromatosis: genetics, pathophysiology, diagnosis and treatment. Cambridge: Cambridge University Press, 2000: 525–34. McCune CA, Al-Jader LN, May A, Hayes SA, Jackson HA, Worwood M. Hereditary haemochromatosis: only 1% of adult HFE C282Y homozygotes in South Wales have a clinical diagnosis of iron overload. Hum Genet 2002; 111: 538–43.
Department of Haematology, University of Wales College of Medicine, Heath Park, Cardiff CF14 4XN, UK (C A McCune MBBS, Prof M Worwood PhD); Western Australian Institute for Medical Research, School of Medicine and Pharmacology, University of Western Australia, Perth, Australia (Prof D Ravine MD); Department of Haematology, Royal Gwent Hospital, Newport, UK (H A Jackson MBChB); John Snow College, University of Durham Queen’s Campus, Thornaby, Stockton-on-Tees, UK (Prof H M Evans PhD); The Welsh Blood Service, Ely Valley Road, Talbot Green, Pontyclun, UK (D Hutton MBChB) Correspondence to: Dr Anne McCune (e-mail: [email protected])
THE LANCET • Vol 362 • December 6, 2003 • www.thelancet.com
For personal use. Only reproduce with permission from The Lancet publishing Group.
Screening for hereditary haemochromatosis within families and beyond C Anne McCune, David Ravine, Mark Worwood, Helen A Jackson, H Martyn Evans, David Hutton Screening programmes for haemochromatosis that include follow-up identification of relatives are claimed to be cost effective. We assessed uptake of screening by first-degree relatives of two groups of index cases: people homozygous for the C282Y mutation ascertained by genetic screening of blood donors; and patients presenting clinically with haemochromatosis. Only 40 (24%) of 165 relatives of blood donors had been tested. By contrast, testing uptake in 121 relatives of patients diagnosed clinically was more than double that (53%), despite unstructured provision of genetic information. A substantial number of untested relatives had undiagnosed iron overload. Overall efficacy of population screening for haemochromatosis is undermined by these observations.
Lancet 2003; 362: 1897–98
Hereditary haemochromatosis is a recessive disorder of iron accumulation. Most affected individuals are homozygous for an aminoacid substitution (C282Y) arising from a single mutation in the haemochromatosis gene (HFE). In the UK, more than 90% of patients with this disorder are C282Y homozygotes, as are about one in 150 people in the general population.1 Advocates of screening point to the high frequency of C282Y mutations in northern Europeans and availability of an effective treatment. However, results of research-based population screening programmes have suggested that the clinical penetrance of the gene is substantially lower than previously reported, with increasing uncertainty and controversy2,3 about the overall clinical importance of homozygosity for this mutation. By contrast, there is a powerful case for screening relatives of cases presenting clinically. Siblings have the highest risk (one in four) of iron overload and should be offered screening (http://www.bcshguidelines.com). Several groups have addressed the cost-effectiveness of population screening.4 When studies include follow-up identification of relatives, cost-effectiveness is substantially improved.4 However, none provides guidance on how to offer testing to relatives or address the probable uptake within families. We aimed to look into how information about genetic risk is transmitted through families, and to assess uptake of genetic screening within a large population of young healthy adults—the perceived ideal target population for haemochromatosis screening. We investigated first-degree relatives of two groups of index cases: C282Y homozygotes ascertained through a genetic-screening programme of healthy blood donors;1 and haemochromatosis patients who presented clinically. Identification of 72 individuals homozygous for the C282Y mutation from 10 556 blood donors in 1998 is described elsewhere.1 Written information about haemochromatosis
and inheritance was provided at enrolment. We contacted every homozygous individual and offered them an interview with a doctor to discuss results and arrange treatment as necessary. Inheritance of haemochromatosis was also explained, and we provided detailed information about family testing with an offer to counsel and test relatives. We supplied an additional factsheet reinforcing the genetic risk to close relatives. Patients identified within the three health authorities in south Wales were contacted after their hospital consultant had given permission.5 All patients (age 18–70 years) had presented clinically, rather than by family or population screening, and all were homozygous for C282Y. After obtaining ethics approval from BroTaf, Gwent, and Iechyd Morgannwg regional ethics committees, we sent letters in 2001 to all index cases asking them to consider discussing the present study with available first-degree relatives (parents and adult siblings), including those previously tested for haemochromatosis. A doctor interviewed consenting relatives, usually at home, because most people were reluctant to visit hospital since this journey often entailed travelling some distance; blood samples were taken at this visit. People living outside south Wales were interviewed by telephone, with blood samples taken locally and forwarded by post. Tests arranged on all samples included: transferrin saturation, serum ferritin, and HFE genotyping. We deemed transferrin saturation greater than 50% or serum ferritin greater than 300 g/L (greater than 200 g/L in premenopausal women) abnormal. Family details were available for 66 of the 72 individuals homozygous for C282Y (median age at screening 38 years [IQR 28–45], and 61 families responded to the letter inviting their participation. 56 families were finally available for testing. Of 180 eligible relatives, 165 (92%) agreed to take part. 105 (64·4%) of these knew of their relative’s diagnosis (table). However, only a third were aware of their own risk; overall, a quarter had been tested for haemochromatosis. After interview, 163 (99%) chose to proceed with testing. 25 C282Y homozygotes were identified (15 women), of whom three women and seven men (40%) also had a raised transferrin saturation and serum ferritin. Family details were available for 60 clinical probands (median age at diagnosis 51 years [IQR 45–58]). 121 (85%) of 143 eligible first-degree relatives were interviewed (table). By contrast with relatives of blood donors, almost all knew of the index case’s diagnosis (97%). Of these, 66 (55%) also had some awareness of their own risk and most had been tested. Like relatives of donors, a substantial proportion had not appreciated their own high genetic risk despite awareness of the disorder in their relative. All previously untested relatives elected to be investigated. 34 were C282Y homozygotes (19 women), of whom eight women and 12 men (59%) also had a raised transferrin saturation and serum ferritin. Overall, 59 people homozygous for C282Y were recorded, and half had excessive iron stores. As a result, 16 new referrals were made for venesection. Relatives of cases discovered by genetic Relatives of patients testing (n=165) (n=121)
Outcome Aware of own genetic risk and tested for haemochromatosis 40 (24%)* Aware of own genetic risk but not tested for haemochromatosis 16 (10%)† Aware of diagnosis in first-degree relative but no knowledge of own genetic risk 49 (30%) Unaware of diagnosis in first-degree relative 60 (36%)
64 (53%) 2 (2%)‡ 51 (42%) 4 (3%)
Data are number of people (%). *Difference between proportions of relatives tested in the two groups, p<0·0001. †Advised against testing by health-care professional (n=7), unsure how to arrange testing (n=1), not yet arranged testing (n=5), active decision not to test (n=3). ‡Advised against testing by health-care professional.
Uptake of testing in first-degree relatives of C282Y homozygotes discovered by genetic testing and after clinical presentation
THE LANCET • Vol 362 • December 6, 2003 • www.thelancet.com
1897
For personal use. Only reproduce with permission from The Lancet publishing Group.
RESEARCH LETTERS
Why was the difference in screening uptake between groups so notable? Perhaps concern for a sick family member was absent in relatives of cases detected by screening, and our attempt to provide good quality information seemed no match for the absence of stimulus provided by a clinical disorder. Although a simple observation, the possibility of a muted response from high-risk family members has not been considered in the debate about population screening for haemochromatosis. Nevertheless, overall outcome in relatives of clinical cases cannot be regarded as satisfactory, and a substantial number had undiagnosed iron overload. With our approach, which ensured skilled counselling and testing, uptake of screening in both groups of relatives rose to more than 98%. Such a proactive approach offers a possible solution to the difficulty of poor screening uptake in relatives. Whether such a high uptake could be achieved in a routine service setting remains to be seen. In assessing screening of a hypothetical cohort of 10 000 blood donors, Adams and colleagues4 assumed that all identified probands would have two siblings, each at 25% risk. Screening was regarded as cost-effective, and including siblings within the model substantially reduced costs. Similarly, Baer and colleagues4 reported a favourable cost of US$17 000 per case of haemochromatosis identified, as long as affected relatives were included in cost calculations. We have shown that high uptake of testing by relatives cannot be assumed. Blood donors are thought to be well motivated, and one might expect higher levels of testing in their relatives. However, even in this group we reported unacceptably low rates of uptake. This rate is improved by a proactive approach, which has additional cost implications that, until now, have not been considered. Future assessments of general population screening must take these costs into account. Contributors C A McCune, D Ravine, and M Worwood designed the study, analysed data, and prepared the report. C A McCune obtained data and wrote the first draft of the report. D Hutton and H A Jackson contributed to design of the study and provided data on index cases (blood donors and patients). H M Evans contributed to design and ethical aspects of the study. All authors reviewed the report and approved the final version.
1898
Conflict of interest statement None declared.
Acknowledgments We thank the blood donors, patients, and their families for taking part in this study; Jeanne Kingston (University of Wales College of Medicine) for HFE genotyping of blood samples; Kimberley Carter (University of Wales College of Medicine) for coordinating sample storage; Richard Ellis (University Hospital of Wales) for assays for iron and ferritin; and Carole Rabaiotti (University of Wales College of Medicine) for invaluable secretarial support. This work was funded by a grant awarded by the Wales Office of Research and Development in July, 2001. The sponsor of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. 1
2
3
4
5
Jackson HA, Carter K, Darke C, et al. HFE mutations, iron deficiency and overload in 10 500 blood donors. Br J Haematol 2001; 114: 474–84. Beutler E. The HFE Cys282Tyr mutation as a necessary but not sufficient cause of clinical hereditary hemochromatosis. Blood 2003; 101: 3347–50. Ajioka RS, Kushner JP. Clinical consequences of iron overload in hemochromatosis homozygotes. Blood 2003; 101: 3351–53. Cappuccio J, Phatak PD. Cost-effectiveness of screening for hemochromatosis. In: Barton JC, Edwards CQ, eds. Hemochromatosis: genetics, pathophysiology, diagnosis and treatment. Cambridge: Cambridge University Press, 2000: 525–34. McCune CA, Al-Jader LN, May A, Hayes SA, Jackson HA, Worwood M. Hereditary haemochromatosis: only 1% of adult HFE C282Y homozygotes in South Wales have a clinical diagnosis of iron overload. Hum Genet 2002; 111: 538–43.
Department of Haematology, University of Wales College of Medicine, Heath Park, Cardiff CF14 4XN, UK (C A McCune MBBS, Prof M Worwood PhD); Western Australian Institute for Medical Research, School of Medicine and Pharmacology, University of Western Australia, Perth, Australia (Prof D Ravine MD); Department of Haematology, Royal Gwent Hospital, Newport, UK (H A Jackson MBChB); John Snow College, University of Durham Queen’s Campus, Thornaby, Stockton-on-Tees, UK (Prof H M Evans PhD); The Welsh Blood Service, Ely Valley Road, Talbot Green, Pontyclun, UK (D Hutton MBChB) Correspondence to: Dr Anne McCune (e-mail: [email protected])
THE LANCET • Vol 362 • December 6, 2003 • www.thelancet.com
For personal use. Only reproduce with permission from The Lancet publishing Group.