Published online before print February 12, 2004, doi: 10.1161/01.STR.0000117966.47696.3A
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2004;35:e59.)
© 2004 American Heart Association, Inc.
Letters to the Editor |
Familial Intracranial Aneurysms
Department of Neurology, Rudolf Magnus Institute of Neuroscience
Department of Biomedical Genetics, University Medical Center, Utrecht, the Netherlands
To the Editor:
With interest we have read the study by Wills et al on 346 Finnish families with familial intracranial aneurysms.1 The authors were able to collect an impressively large number of intracranial aneurysms families, defined as at least 2 members with the diagnosis of intracranial aneurysms.
In their study, the authors describe different characteristics of the collected families, including the determination of the patterns of inheritance. The authors describe that 198 (57.2%) families were consistent with an autosomal recessive pattern of inheritance, 126 (36.4%) with an autosomal dominant pattern of inheritance, and 19 (5.5%) with an autosomal dominant pattern of inheritance with incomplete penetrance. In 3 (0.9%) families the pattern of inheritance was found to be complex and not consistent with a clear pattern of inheritance.1
However, it is not clear how these patterns of inheritance were determined, as no prespecified criteria for the definition of the modes of inheritance were outlined. Six representative pedigrees of the 346 families were shown in the article. As a representative example of a family with an autosomal recessive pattern of inheritance, the authors showed family no. 10. In this family no. 10, individuals from both the first and second generation were not affected while 10 out of the 19 individuals of the third generation (of 3 different pairs of parents) were affected, compatible with a segregation ratio of 53% (95% CI 29 to 76). This is in strong contrast to the 25% that is expected in families with a true autosomal recessive disease gene. Furthermore, autosomal recessive inheritance in this family is unlikely, as this requires that all parents of the affected children should have carried a disease-causing recessive gene. This would mean that the 3 individuals of the second generation who have affected children and are presumed gene carriers all should have reproduced with a spouse who also carried the same mutant gene. Given the approximately 2% rate of intracranial aneurysms in the general population,2 including predominantly nonfamilial and for a smaller part familial cases, this seems highly unlikely.
We used this family as an example for the 6 representative pedigrees shown in the article. The assignment of the patterns of inheritance to the other 5 representative pedigrees can also be debated. Therefore, we would like to point out that the determination of the patterns of inheritance to the 346 collected families in this study may not always be correct. Because not all pedigrees are given and no prespecified criteria for the definition of the modes of inheritance are outlined, we are uncertain about the other assigned patterns of inheritance in this study.
As research on the genetic factors that predispose to intracranial aneurysms is already hampered by the late onset,3 low penetrance,2 and high case fatality4 of the disease, genetic studies should be performed with precision and should not build on uncertain modes of inheritance. Therefore, since the different modes of inheritance in intracranial aneurysms cannot be unambiguously defined, genome-wide scans should rely on nonparametric, model-free methods. Recently, a genome-wide scan for intracranial aneurysm susceptibility genes in the Japanese population reported positive evidence for linkage to 7q11,5 while another Finnish study found linkage to 19q.6 These studies used nonparametric, model-free methods to overcome the problem of possible complex etiology.
References
1. Wills S, Ronkainen A, van der Voet M, Kuivaniemi H, Helin K, Leinonen E, Frösen J, Niemelä M, Jääskeläinen J, Hernesniemi J, Tromp G. Familial intracranial aneurysms: an analysis of 346 multiplex Finnish families. Stroke. 2003; 34: 1370–1375.
2. Rinkel GJE, Djibuti M, Algra A, van Gijn J. Prevalence and risk of rupture of intracranial aneurysms. Stroke. 1998; 29: 251–256.
3. Longstreth WT Jr, Nelson LM, Koepsell TD, van Belle G. Clinical course of spontaneous subarachnoid hemorrhage: a population-based study in King County, Washington. Neurology. 1993; 43: 712–718.
4. Hop JW, Rinkel GJE, Algra A, van Gijn J. Case-fatality rates and functional outcome after subarachnoid hemorrhage: a systematic review. Stroke. 1997; 28: 660–664.
5. Onda H, Kasuya H, Yoneyama T, et al. Genomewide-linkage and haplotype-association studies map intracranial aneurysm to chromosome 7q11. Am J Hum Genet. 2001; 69: 804–819.[CrossRef][Medline] [Order article via Infotrieve]
6. Olson JM, Vongpunsawad S, Kuivaniemi H, Ronkainen A, Hernesniemi J, Ryynanen M, et al. Search for intracranial aneurysm susceptibility gene(s) using Finnish families. BMC Med Genet. 2002; 3: 7.[CrossRef][Medline] [Order article via Infotrieve]
Response
Center for Molecular Medicine and Genetics, Wayne State University School of Medicine, Detroit, Mich
We are delighted that Drs Ruigrok, Rinkel, and Wijmenga read our report1 with such attention to detail. We agree that assigning a mode of inheritance unambiguously to an individual family, based solely on inspection or segregation ratio, is difficult at best and impossible at worst. Indeed, the literature is replete with examples where identification of the molecular mechanism for a disease showed that it was due to de novo sporadic mutations where it had been assumed to be recessive2–5 (see also Online Mendelian Inheritance in Man,6 MIM number: 259400: 11/24/1998), and some diseases can demonstrate autosomal dominant, pseudo-autosomal dominant, and recessive inheritance, all from mutations in the same gene.7 For this reason, among others, we chose with great deliberation the phrasing "family X is consistent with autosomal dominant, recessive, ..." since we understand that to be a more conservative and less emphatic statement than "family X is autosomal dominant, recessive, ...."
We applied fairly standard criteria to determine which mode of inheritance should be designated to each particular family.8 From a genetic point of view, the autosomal genetic loci that contribute to disease (a categorical trait), whether simple Mendelian or complex, must be inherited in 1 of a few modes: recessive where 2 copies are necessary to cause, or contribute to, disease (the haploid copy is sufficient for normal status), dominant where, for instance, haploinsufficiency causes or contributes to disease, and codominant where the heterozygote has a phenotype or risk that is distinguishable from both the homozygotes. The ability to define the mode of inheritance is clouded, but not invalidated, by late age at onset, partial penetrance, nongenetic disease, and possibly a requirement for alleles at >1 locus.
The difficulty in analyzing mode of inheritance for human pedigrees based on their structure has been recognized for many years (see References9–11). A fundamental and recurring theme in segregation analysis has been the requirement for a sampling scheme that allows for an unbiased collection of both multiplex (familial) and simplex (singleton or sporadic) cases. Segregation analyses are based on a number of assumptions that, if violated, invalidate the results. One is that since human families are sampled conditional on the identification of a proband, it is necessary to remove the proband from consideration or condition the likelihood on the proband, to arrive at a valid conclusion that can be extrapolated. Our data set did not meet some of the assumptions and therefore was not suitable for segregation analysis. Also, segregation analyses are unable to assign a mode of inheritance to specific families; they can, however, determine the most likely mode of inheritance in the sample of families. Therefore, the mode of inheritance for a specific family may well be incorrect, particularly when dealing with complex diseases that most likely will have underlying locus heterogeneity. If a covariate exists that can identify subgroups, the mode of inheritance for the subgroup samples may be determined.10
The authors provide an interesting and illustrative analysis of family 10. They suggest that assigning a recessive mode of inheritance is untenable because the segregation ratio is implausibly high and that the disease allele frequency has to be implausibly high to have a sufficient number of carriers of mutations in the gene (note that they need not carry the same alleles). As mentioned above, when determining the segregation ratio, the effect of sampling on the proband needs to be accounted for. This can be done by removing the proband from consideration; therefore, the ratio reduces to 9 of 18, exactly 0.5 or the segregation ratio expected for a dominant disease (96% CI 0.25 to 0.75). Concluding that the family is consistent with a dominant mode of inheritance is not necessarily justified. Since none of the 3 related parents are affected, it is necessary to incorporate incomplete penetrance. If the assumed penetrance is too high (even as low as 0.7), an incongruous situation arises, namely that the expectation is that at least 1 of the 3 parents and their 2 living siblings should be affected. On the other hand, if the assumed penetrance is too low (even as high as 0.7), a similarly incongruous situation arises in that the observed proportion of affected offspring is too high. Multilocus models that model the population data, with or without incomplete penetrance and whether allowing for phenocopies (with penetrances for sporadic cases) or not, yield even less consistent scenarios. In such a situation, it is preferable to accept the least complicated hypothesis, namely that the family is consistent with an autosomal recessive mode of inheritance albeit with an unusual sampling. Similar arguments can be made for the other debatable assignments.
The authors suggest that since research on intracranial aneurysms is already hampered by late age at onset, low penetrance, and high case fatality, we do a disservice to ourselves and the community by performing research without precision and built on uncertain modes of inheritance. They suggest we follow the lead of ourselves12 and others13 and perform the studies with nonparametic, model-free methods. It is worth clarifying that although these methods are referred to as nonparametric, they in fact are parameterized, although not in terms of mode of inheritance (ie, model-free). It is also worth noting that what the models gain in robustness they lose in precision and power; that is, although these models allow for complexity, they do not "overcome" it. We will continue to use such methods,14 and modifications thereof,15–17 in addition to investigating all aspects of our data18–20 in our endeavor to elucidate in part the etiology of intracranial aneurysms.
References
1. Wills S, Ronkainen A, van der Voet M, Kuivaniemi H, Helin K, Leinonen E, Frösen J, Niemelä M, Jääskeläinen J, Hernesiemi J, Tromp G. Familial intracranial aneurysms: an analysis of 346 multiplex Finnish families. Stroke. 2003; 34: 1370–1374.
2. Mallipeddi R, Bleck O, Mellerio JE, Ashton GH, Eady RA, McGrath JA. Dilemmas in distinguishing between dominant and recessive forms of dystrophic epidermolysis bullosa. Br J Dermatol. 2003; 149: 810–818.[CrossRef][Medline] [Order article via Infotrieve]
3. Randon J, Miraglia del Giudice E, Bozon M, Perrotta S, De Vivo M, Iolascon A, Delaunay J, Morle L. Frequent de novo mutations of the ANK1 gene mimic a recessive mode of transmission in hereditary spherocytosis: three new ANK1 variants: ankyrins Bari, Napoli II and Anzio. Br J Haematol. 1997; 96: 500–506.[CrossRef][Medline] [Order article via Infotrieve]
4. Sillence DO, Barlow KK, Garber AP, Hall JG, Rimoin DL. Osteogenesis imperfecta type II delineation of the phenotype with reference to genetic heterogeneity. Am J Med Genet. 1984; 17: 407–423.[CrossRef][Medline] [Order article via Infotrieve]
5. McKusick VA. Mendelian Inheritance in Man: A Catalog of Human Genes and Genetic Disorders. Baltimore, Md: Johns Hopkins University Press; 1994.
6. OMIM. Online Mendelian inheritance in man. Baltimore, Md: Johns Hopkins University. MIM Number. Available at http://www.ncbi.nlm.nih.gov/omim./ Accessed February 2, 2004.
7. Booth DR, Gillmore JD, Lachmann HJ, Booth SE, Bybee A, Soyturk M, Akar S, Pepys MB, Tunca M, Hawkins PN. The genetic basis of autosomal dominant familial Mediterranean fever. QJM. 2000; 93: 217–221.
8. Vogel F, Motulsky AG. Human Genetics: Problems and approaches. Heidelberg, Germany: Springer-Verlag; 1997.
9. Morton NE. Outline of Genetic Epidemiology. Basel, Switzerland: Karger; 1982.
10. Blangero J. Segregation analysis, complex. In: Elston RC, Olson JM, Palmer LJ, eds. Biostatistical Genetics and Genetic Epidemiology. West Sussex, UK: John Wiley & Sons Ltd; 2002: 696–708.
11. Majumder PP. Segregation analysis, classical. In: Elston RC, Olson JM, Palmer LJ, eds. Biostatistical Genetics and Genetic Epidemiology. West Sussex, UK: John Wiley & Sons Ltd; 2002: 693–696.
12. Olson JM, Vongpunsawad S, Kuivaniemi H, Ronkainen A, Hernesniemi J, Ryynänen M, Kim L-L, Tromp G. Search for intracranial aneurysm susceptibility gene(s) using Finnish families. BMC Med Genet. 2002; 3: 7.[CrossRef][Medline] [Order article via Infotrieve]
13. Onda H, Kasuya H, Yoneyama T, Takakura K, Hori T, Takeda J, Nakajima T, Inoue I. Genomewide-linkage and haplotype-association studies map intracranial aneurysm to chromosome 7q11. Am J Hum Genet. 2001; 69: 804–819.[CrossRef][Medline] [Order article via Infotrieve]
14. van der Voet M, Olson JM, Kuivaniemi H, Dudek DM, Skunca M, Ronkainen A, Jääskeläinen J, Hernesiemi J, Helin K, Leinonen E, Biswas M, Tromp G. Intracranial aneurysms in Finnish families: confirmation of linkage and refinement of the interval to chromosome 19p13.3. Am J Hum Genet. 2004;74:in press.
15. Olson JM. A general conditional-logistic model for affected-relative-pair linkage studies. Am J Hum Genet. 1999; 65: 1760–1769.[CrossRef][Medline] [Order article via Infotrieve]
16. Goddard KAB, Witte JS, Suarez BK, Catalona WJ, Olson JM. Model-free linkage analysis with covariates confirms linkage of prostate cancer to chromosomes 1 and 4. Am J Hum Genet. 2001; 68: 1197–1206.[CrossRef][Medline] [Order article via Infotrieve]
17. Olson JM, Goddard KA, Dudek DM. A second locus for very-late-onset Alzheimer disease: a genome scan reveals linkage to 20p and epistasis between 20p and the amyloid precursor protein region. Am J Hum Genet. 2002; 71: 154–161.[CrossRef][Medline] [Order article via Infotrieve]
18. Ronkainen A, Hernesniemi J, Ryynänen M, Puranen M, Kuivaniemi H. A ten percent prevalence of asymptomatic familial intracranial aneurysms: preliminary report on 110 magnetic resonance angiography studies in members of 21 Finnish familial intracranial aneurysm families. Neurosurgery. 1994; 35: 208–212.[Medline] [Order article via Infotrieve]
19. Ronkainen A, Hernesniemi J, Puranen M, Niemitukia L, Vanninen R, Ryynänen M, Kuivaniemi H, Tromp G. Familial intracranial aneurysms. Lancet. 1997; 349: 380–384.[CrossRef][Medline] [Order article via Infotrieve]
20. Ronkainen A, Miettinen H, Karkola K, Papinaho S, Vanninen R, Puranen M, Hernesniemi J. Risk of harboring an unruptured intracranial aneurysm. Stroke. 1998; 29: 359–362.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||