A Genetic Contribution to Inflammatory Bowel Disease in Iceland: A Genealogic Approach

Transcription

1 CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2004;2: A Genetic Contribution to Inflammatory Bowel Disease in Iceland: A Genealogic Approach INGA REYNISDOTTIR,* DANIEL F. GUDBJARTSSON,* JOHANN HEIDAR JOHANNSSON, ILEANA MANOLESCU,* KRISTLEIFUR KRISTJANSSON,* KARI STEFANSSON,* JEFFREY GULCHER,* and SIGURDUR BJORNSSON *decode Genetics, Inc., Reykjavík; National University Hospital at Hringbraut, Reykjavík; and National University Hospital at Fossvogur, Reykjavík, Iceland Background & Aims: Both genetic and environmental factors play a role in the development of Crohn s disease (CD)and ulcerative colitis (UC), collectively known as inflammatory bowel disease (IBD). The aim of this study was to estimate the genetic component in IBD in Iceland. Methods: A population-based sample, representing everyone diagnosed with IBD in Iceland from 1950 to 1996, was studied using a computerized population-wide genealogic database. The relationships among the patients were analyzed by calculating the kinship coefficient and the relative risk. Results: The kinship coefficients for the patients were significantly greater than the mean kinship coefficient for the controls (P < 10 6 ). The risk ratio for siblings of IBD, UC, and CD patients was 5.0 (P < 0.001), 5.9 (P < 0.001), and 4.1 (P 0.033), respectively. The cross-risk ratio for siblings of UC patients developing CD (or vice versa)was 2.6 (P 0.015). Conclusions: The results indicate that the IBD patients are more closely related than the controls, which strongly supports the involvement of a genetic component in the development of IBD in Icelandic patients. We find that the increase in risk for relatives of UC probands to develop UC, or relatives of CD probands to develop CD, is greater than the increase in risk for relatives of UC probands to develop CD, or relatives of CD probands to develop UC. Crohn s disease (CD) and ulcerative colitis (UC) are chronic inflammatory disorders of unknown cause, collectively termed inflammatory bowel disease (IBD). Although no single environmental factor has proven to be the causative agent, it is likely that both environmental and genetic factors contribute. For a recent review on IBD see Yang et al. 1 The risk for spouses of affected probands is no higher than the population risk, regardless of onset before or after marriage, favoring factors other than late environmental exposure in disease development. Numerous studies indicate that genetic factors are involved in the pathogenesis of IBD. There are basically 4 lines of evidence for this. First, twin studies show that the concordance rate is higher in monozygotic twins than in dizygotic twins. 2 6 Second, there is a familial aggregation of patients with IBD, and relatives of IBD patients are at an increased risk for developing IBD Population-based studies have reported a 5% to 10% frequency of IBD in family members of IBD probands. That is much higher than the rate in the general population, which ranges from 0.2% to 0.5%. 8,11 16 Third, there are ethnic and racial differences in IBD frequency. 7,17,18 Finally, mutations have been identified in the NOD2 (or CARD15) gene, suggesting that the gene confers susceptibility to CD Three separate nationwide studies of the incidence of UC and CD in Iceland have been conducted by the same investigators for the 47-year period from 1950 to 1996; 2 of the studies were retrospective 22,23 and the third study was prospective. 24 The incidence of both diseases was found to have increased throughout the study period, the incidence rate of UC increasing from 2.8/100,000 from 1950 to 1959 to 16.5/100,000 from 1990 to 1994, and the incidence of CD increasing from 0.5/100,000 from 1950 to 1959 to 5.5/100,000 from 1990 to The incidence of UC in Iceland is one of the highest in Europe. 25 Increases in incidence rates have been observed in many Western countries, and in some it has levelled off. 2 Some of the increase may be owing to improved diagnostic techniques but most of the increase appears to be real. 26 It has been suggested that environmental factors play a role in the increasing incidence. 26,27 There is a unique opportunity to assess the genetic contribution to UC and CD in Iceland. Complete records of all patients diagnosed with IBD in Iceland from 1950 to 1996 are available from previous studies This reduces the probability of ascertainment bias. Further- Abbreviation used in this paper: KC, kinship coefficient by the American Gastroenterological Association /04/$30.00 PII: /S (04)

2 September 2004 GENETIC CONTRIBUTION TO IBD IN ICELAND 807 more, a comprehensive genealogic database at decode Genetics, Inc. allows us to analyze the relationships between all of the patients. Here we report an epidemiologic study that evaluates the contribution of genetic components in the pathogenesis of IBD in Iceland. Two tests were applied to analyze the relatedness of the patients, both collectively (IBD) and separately (UC or CD) calculation of kinship coefficients and relative risk ratios. Materials and Methods Patients The study group was a population-based sample including all patients diagnosed with IBD in Iceland for the 47-year period from 1950 to The patients were born between 1885 and This group consisted of a total of 1162 patients including 934 with UC, 220 with CD, and 8 with indeterminate colitis from 3 different studies by the same principal investigator A diagnosis of indeterminate colitis was assigned to patients when it was not possible to distinguish between CD and UC. The diagnoses of all the patients were reviewed independently and fulfilled accepted diagnostic criteria. All the patients have had at least 1 year of follow-up evaluation and many patients have had decades of re-evaluation and confirmation of their final diagnosis The 8 indeterminate colitis patients were included only in our calculations for the collective group of patients, but were omitted when we analyzed UC and CD separately. Genealogy Database decode Genetics, Inc. has created a population-based genealogy database covering the past 10 centuries in Iceland It includes all living Icelanders, approximately 280,000, and a large proportion of their ancestors. To ensure the patients anonymity in our study their social security numbers were sent to the Data Protection Commission of Iceland for encryption before arriving at the laboratory. 31 The genealogy database is encrypted in the same manner by the same commission. The birth year of individuals as well as their year of death are grouped into 5-year periods to increase the security. Matched Controls We used matched controls to estimate significance in the statistical analysis. We randomly selected 1000 sets of controls. For each set of controls, every control was selected to match one subject. The matching was based on being born in the same 5-year interval, and having the same number of ancestors (up to fourth degree) in the genealogic database. Kinship Coefficients We incorporate a kinship coefficient (KC), a measure of identical-by-descent sharing, to assess whether the affected individuals are more related than a set of matched controls. 28 The KC for a pair of individuals is the probability that, for a particular autosomal locus, 2 randomly selected alleles, one from each individual, were identical by descent. 32 We determined the average pair-wise KC for the set of affected individuals and compared this with the distribution of the average pair-wise KC for sets of matched controls. Because of the large size of the pedigrees there was a computational problem in accurately determining the KC. Monte Carlo simulations were used to approximate the average pair-wise kinship coefficient for each group. A total of 100,000 simulations were performed to ensure that the Monte Carlo errors had a negligible effect on the results. 28,29 Empiric P values were calculated by counting how many sets of controls had average pair-wise KCs larger than the set of affecteds being tested. The contribution of close relatives dominates the KC values. To show that increased relatedness extends beyond first-degree relatives, we also report KCs with all first-degree relationships removed. Risk Ratio The risk ratio 33 gives the increase in risk for a relative of an affected individual (proband), as compared with the risk for the relative without any information about the proband. If A denotes the event that the proband is affected and B denotes the event that the relative is affected, then the risk ratio is defined as R P[B A] P[B]. For example, for a pair of siblings, the risk ratio measures the conditional probability that the second sibling is affected, provided that the first one is affected, divided by the population frequency of the disease. Much care must be taken to choose an appropriate estimator for the risk ratio that will suit our situation. 34 For our data structure, the following estimator for the risk ratio is appropriate: N j 1 N j 1 a j n j n j x j r j x j n j x j where j runs over groups of people where both the numerator (P[B A]) and the denominator (P[B]) in the ratio are constant, n j is the number of people in group j, x j is the number of affected people in group j, r j is the number of relatives in group j (note that a single person may be the relative of 1 proband, and thus may be counted more than once), and a j is the number of relatives in group j that are affected (again counting more than once relatives of more than 1 proband). In our analyses, we have taken j to run over groups of people born in the same 5-year period, for all relationships except spouses. For the spouses, extra care must be taken because if the risk for

3 808 REYNISDOTTIR ET AL. CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 2, No. 9 developing a disease is not completely independent of sex we may end up with a biased estimate of the risk ratio. In the case of IBD, the risk is close to being independent of sex so that this bias would not be expected to affect our estimates, but we still prefer using the unbiased approach as described earlier. For the spouses, we let j run over all groups of people sharing the same 5-year period and the same sex. Testing against the null hypothesis of there being no familial component can be performed by counting how many matched control groups have higher risk ratios than the patient group. The frequency of control groups surpassing the patient group gives an empiric P value. By using a variance stabilizing square root transform, an approximate confidence interval may be constructed. 30 If 2.5 and 97.5 are the 2.5% and 97.5% quantiles of the risk ratio in the matched control groups, and ˆ is the estimate of the risk ratio, then ˆ , ˆ will be an approximate 95% confidence interval for. It is trivial to extend the risk ratio to examine the relationship between one disease phenotype and another (which we call the cross-risk ratio). Instead of conditioning on the proband having the same disease as the relative, we may as well condition on the proband having an entirely different disease. Let the cross-risk ratio be defined as: Figure 1. An ancestral pedigree of patients with IBD. UC patients (, ) and Crohn s disease patients (l, L) are within the same pedigree. Men are represented by squares and women are represented by circles. For clarity, the unaffected siblings are not drawn. Table 1. Kinship Coefficients X3Y P[relative develops disease Y proband develops disease X]. P[relative develops disease Y] Because of the symmetry in the definition of X3Y, we have X3Y Y3X. Results Patients Controls SD P value IBD 2.25e e e IBD 1 st 2.03e e e UC 2.38e e e UC 1 st 2.12e e e CD 2.51e e e CD 1 st 2.41e e e NOTE. The KC measures the extent of allelic sharing by descent within a group of individuals. The KC for the IBD, UC, and CD patients is 9, 11.7, and 16.8 SD, respectively, from the mean average KC for the controls. 1 st, first-degree relationships removed. The majority of IBD patients in Iceland appear to fall into families. One of the families is shown in Figure 1. The ancestral pedigree shows 36 patients whose common ancestors were born in the early 1700s. It is one of many extended pedigrees that includes both UC and CD patients. Not all families are mixed; there are families that include only UC patients and others that include only CD patients. In this study a nationwide populationbased sample of IBD in Iceland was investigated using the genealogic database 30 to analyze the relatedness of the patients and, therefore, the genetic contribution to the development of the disease. Kinship Test The KC measures the likelihood of allelic sharing by descent from common ancestors. The mean KC of the IBD group is , but the control groups have a mean average KC of (SD, 0.047e 4, P 0.001; Table 1). The KC for the IBD patients was 9 SDs greater than the mean KC for the controls. The KCs for the UC and CD patients were 11.7 and 16.8 SDs greater than the mean KC for the controls, respectively (Table 1). These results were significant for all 3 groups (P 0.001). They suggest that the patients were more likely to have alleles derived from common ancestors than the control groups. The contribution of close familial relationships to the KC may outweigh that of distant relationships. To eliminate this and to focus only on contributions from more distant relationships, all first-degree relationships were

4 September 2004 GENETIC CONTRIBUTION TO IBD IN ICELAND 809 Table 2. Risk Ratios IBD to IBD UC to UC CD to CD UC to CD/CD to UC Relationship Kinship P value 95% CI P value 95% CI P value 95% CI P value 95% CI Siblings ( ) ( ) ( ) ( ) Parent/child ( ) ( ) ( ) ( ) Grandparent/ grandchild ( ) ( ) ( ) ( ) Avuncular ( ) ( ) ( ) ( ) Cousins ( ) ( ) ( ) ( ) Spouses ( ) ( ) ( ) ( ) NOTE. The risk ratios for IBD, UC, and CD were calculated for relatives of probands with IBD, UC, and CD, respectively. The risk ratios for UC/CD were calculated for probands with CD/UC. The P values reported are empiric P values based on how many matched control sets produced risk ratios higher than the subject set. The 95% confidence intervals presented are based on using a square root variance stabilizing transform. CI, confidence interval. removed from consideration. The results were similar as before; the KC for the patients was significantly higher than the average KC for the control groups (Table 1). Risk Ratio The risk ratios were calculated for IBD to IBD, UC to UC, and CD to CD, for first-, second-, and third-degree relatives as well as for spouses. We also calculated the cross-risk ratio for UC to CD, which is the same as the cross-risk ratio for CD to UC, as was stated earlier (see Materials and Methods section). Table 2 contains the results of these calculations. The risk ratio for siblings of IBD, UC, and CD patients was 5.0 (P 0.001), 5.9 (P 0.001), and 4.1 (P 0.033), respectively. For all the within-phenotype tests, the risk ratios decreased as the relationship became more distant. The pattern was a little bit distorted in the CD to CD case, but this likely was owing to the relatively small size of the CD subject group, as reflected in the large confidence intervals. The within-phenotype risk ratios were almost all significantly different from 1 for the first-, second-, and third-degree relatives. The exceptions were cousins for IBD to IBD, and UC to UC, and the grandparent/ grandchild relationship for CD to CD. That is not to say that the latter risk ratios are not greater than 1; the lack of significance may be owing to small sample sizes. The risk ratio for spouses was only significantly different from 1 in the CD to CD case, in which it was marginally significant (P 0.045). The cross-risk ratios for UC to CD or CD to UC were notably different. In those relationships, only the sibling risk ratio was significantly different from 1, and the estimates of the other risk ratios were not significant. Discussion Many previous epidemiologic studies have implicated a genetic component in the development of IBD. In addition to familial aggregation and increased risk for relatives of probands, 7 10 the disease prevalence is higher in some ethnic groups 7,17,18 and concordance is higher among monozygotic than dizygotic twins. 2 6 Our results lead to the same conclusion, namely that there must be a genetic component involved in the cause of IBD. Previous work has examined ethnically homogeneous populations in well-defined geographic locations. Assessment of familial occurrence of IBD in Copenhagen County, which represents about 10% of the population of Denmark, clearly showed familial aggregation and an increase in relative risk. 15 Similarly, other studies have collected patients within cities or counties. 8,13,14,35 The occurrence of IBD within the families varies somewhat among these studies; nevertheless all of them show familial aggregation of IBD patients. Our approach has some distinct advantages compared with previous studies. Although previous work was performed with ethnically homogeneous groups in welldefined geographic areas, they did not include a nationwide population-based sample. Ascertainment bias is a major problem that is reduced substantially by having a population-based sample. Biases in probabilities of affecteds being diagnosed still can cause problems but the selection of patients from the pool of diagnosed affecteds will not cause a bias because all those that are available are selected. The patients in our study included everyone diagnosed with IBD in Iceland from 1950 to Another advantage is the genealogic information that exists in Iceland. The genealogy project at decode Genetics, Inc. involves the compilation of existing genealogic information into an electronic database Therefore, this population-based database allowed us to investigate all the relationships among the IBD patients and allowed us to distinguish between sporadic and familial cases.

5 810 REYNISDOTTIR ET AL. CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 2, No. 9 The within-phenotype risk ratios for spouses were only significant for CD, and then only marginally significant (P 0.045). This supports our claim that ascertainment is not a major problem in our dataset, and it suggests that whatever genetic factors may be contributing to IBD are not likely to be shared by spouses. It should be noted that our estimate of the risk ratio for spouses of CD patients is based on observing only 2 spouses with CD out of a total of 173 spouses, as is reflected in the wide confidence interval. Two previous studies have suggested an environmental role in the cause of IBD because couples with IBD have been observed more often than expected. 36,37 Unfortunately, our data does not enable us to replicate their findings because of relatively few observed spouses and our lack of information about disease status at the time of marriage. There is a high degree of familial aggregation of IBD patients in Iceland. The kinship calculation results show the familial segregation as well as extension of risk beyond the nuclear family for IBD, UC, and CD. Firstand second-degree relatives of IBD, UC, and CD probands have significantly and substantially increased risk for developing IBD, UC, and CD, respectively. An increased risk for first-degree relatives may reflect shared genetic and environmental components. However, the increased risk for second-degree relatives and the fact that the KC remains significant after subtraction of first-degree relatives suggests that these observations are not simply caused by environment only or by ascertainment bias. The risk ratios are not significant for thirddegree relatives of the probands other than CD; however, the majority of the second-degree relationships (avuncular and grandparent-grandchild) have risk ratios significantly different from 1. The cross-risk ratio between UC and CD siblings is significant (P 0.015), but none of the other types of relationships are. The cross-risk ratios also clearly are lower than the within-disease risk ratios. The risk for relatives of CD probands of developing CD is higher than that of relatives of UC probands developing UC, which in turn is higher than that of relatives of IBD probands developing IBD. This is consistent with our finding of the sibling cross-risk ratios being substantially lower than the within-disease sibling risk ratios. Our data suggest that there are environmental or genetic components common to both phenotypes, but that such components may be substantially smaller than the withinphenotype risk factors. Our calculations of the relative risk within phenotype risk and cross-phenotype risk are similar to observations by others on prevalence and risk in relatives of probands with UC and CD. The trend seems to be that the highest risk is observed for siblings and it is lower for parents and offspring (see Yang and Rotter 7 and McConnell and Vadheim 38 ) and it is reduced further when the studies are extended to include second-degree relatives. 12,15 An increased cross-risk between CD and UC also is observed in other populations, 9,12,15,39 41 but it does not extend beyond the first-degree relatives. 12,15 However, most studies report that there is a strong disease concordance. Relatives of CD probands are more likely to develop CD rather than UC and vice versa. The cross-risk between CD and UC suggests that there could be a shared genetic component between the 2 disease phenotypes. Linkage studies support that there are common genetic factors between CD and UC, as well as separate genes for the 2 phenotypes (see Taylor et al. 42 ). The IBD1 locus on chromosome 16 is seen in CD families only, 43,44 whereas the IBD2 locus on chromosome 12 appears to have a greater role in UC families The loci IBD3 on 6p, 47,48 IBD6 on 19p13, 49 IBD7 on 1p36, 50,51 and IBD8 on 16p 52 all are seen with both phenotypes whereas IBD4 on 14q11-q12 53,54 and IBD5 on 5q ,55 are observed in CD and CD early onset, respectively. Other suggestive loci have been observed with the combined IBD phenotype on chromosomes 7q and 3p, 45 chromosomes 3q and 4q, 50 and chromosomes 1, 10, and A linkage to chromosome X was seen for UC. 56 Some of these loci have been replicated in other populations but others have not. In summary, our results indicate that patients with IBD in Iceland are more closely related than controls, which suggests that there is a genetic contribution to the susceptibility to IBD in Iceland. Our data also suggest that CD and UC may share common genetic and/or environmental risk factors. References 1. Yang H, Taylor KD, Rotter JI. Inflammatory bowel disease. I. Genetic epidemiology. Mol Genet Metab 2001;74: Binder V. Genetic epidemiology in inflammatory bowel disease. Dig Dis 1998;16: Subhani JM, Hamiliton MI. Review article: the management of inflammatory bowel disease during pregnancy. Aliment Pharmacol Ther 1998;12: Thompson NP, Driscoll R, Pounder RE, Wakefield AJ. Genetics versus environment in inflammatory bowel disease: results of a British twin study. BMJ 1996;312: Tysk C, Lindberg E, Jarnerot G, Floderus-Myrhed B. Ulcerative colitis and Crohn s disease in an unselected population of monozygotic and dizygotic twins. A study of heritability and the influence of smoking. Gut 1988;29: Halfvarson J, Bodin L, Tysk C, Lindberg E, Jarnerot G, Laharie D, Debeugny S, Peeters M, Van Gossum A, Gower-Rousseau C, Belaiche J, Fiasse R, Dupas JL, Lerebours E, Piotte S, Cortot A, Vermeire S, Grandbastien B, Colombel JF. Inflammatory bowel disease in a Swedish twin cohort: a long-term follow-up of con-

6 September 2004 GENETIC CONTRIBUTION TO IBD IN ICELAND 811 cordance and clinical characteristics. Inflammatory bowel disease in spouses and their offspring. Gastroenterology 2003;124: Yang H, Rotter J. Genetic aspects of idiopathic inflammatory bowel disease. In: Kirsner J, Shorter R, eds. Inflammatory bowel disease. Baltimore: Williams and Wilkins, 1995: Russel MG, Pastoor CJ, Janssen KM, van Deursen CT, Muris JW, van Wijlick EH, Stockbrugger RW. Familial aggregation of inflammatory bowel disease: a population-based study in South Limburg, The Netherlands. The South Limburg IBD Study Group. Scand J Gastroenterol Suppl 1997;223: Peeters M, Nevens H, Baert F, Hiele M, de Meyer AM, Vlietinck R, Rutgeerts P. Familial aggregation in Crohn s disease: increased age-adjusted risk and. Gastroenterology 1996;111: Lee JC, Lennard-Jones JE. Inflammatory bowel disease in 67 families each with three or more affected first-degree relatives [see comments]. Gastroenterology 1996;111: Mayberry JF, Rhodes J, Newcombe RG. Familial prevalence of inflammatory bowel disease in relatives of patients with Crohn s disease. BMJ 1980;280: Meucci G, Vecchi M, Torgano G, Arrigoni M, Prada A, Rocca F, Curzio M, Pera A, de Franchis R. Familial aggregation of inflammatory bowel disease in northern Italy: a multicenter study. The Gruppo di Studio per le Malattie Infiammatorie Intestinali (IBD Study Group). Gastroenterology 1992;103: Monsen U, Bernell O, Johansson C, Hellers G. Prevalence of inflammatory bowel disease among relatives of patients with Crohn s disease. Scand J Gastroenterol 1991;26: Monsen U, Brostrom O, Nordenvall B, Sorstad J, Hellers G. Prevalence of inflammatory bowel disease among relatives of patients with ulcerative colitis. Scand J Gastroenterol 1987;22: Orholm M, Munkholm P, Langholz E, Nielsen OH, Sorensen IA, Binder V. Familial occurrence of inflammatory bowel disease. N Engl J Med 1991;324: Bernstein CN, Blanchard JF, Rawsthorne P, Wajda A. Epidemiology of Crohn s disease and ulcerative colitis in a central Canadian province: a population-based study. Am J Epidemiol 1999;149: Yang H, McElree C, Roth MP, Shanahan F, Targan SR, Rotter JI. Familial empirical risks for inflammatory bowel disease: differences between Jews and non-jews. Gut 1993;34: Probert CS, Jayanthi V, Rampton DS, Mayberry JF. Epidemiology of inflammatory bowel disease in different ethnic and religious groups: limitations and aetiological clues. Int J Colorectal Dis 1996;11: Hugot JP, Chamaillard M, Zouali H, Lesage S, Cezard JP, Belaiche J, Almer S, Tysk C, O Morain CA, Gassull M, Binder V, Finkel Y, Cortot A, Modigliani R, Laurent-Puig P, Gower-Rousseau C, Macry J, Colombel JF, Sahbatou M, Thomas G. Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn s disease. Nature 2001;411: Ogura Y, Bonen DK, Inohara N, Nicolae DL, Chen FF, Ramos R, Britton H, Moran T, Karaliuskas R, Duerr RH, Achkar JP, Brant SR, Bayless TM, Kirschner BS, Hanauer SB, Nunez G, Cho JH. A frameshift mutation in NOD2 associated with susceptibility to Crohn s disease. Nature 2001;411: Cuthbert AP, Fisher SA, Mirza MM, King K, Hampe J, Croucher PJ, Mascheretti S, Sanderson J, Forbes A, Mansfield J, Schreiber S, Lewis CM, Mathew CG. The contribution of NOD2 gene mutations to the risk and site of disease in inflammatory bowel disease. Gastroenterology 2002;122: Bjornsson S. Inflammatory bowel disease in Iceland during a 30-year period, Scand J Gastroenterol Suppl 1989;170: Bjornsson S, Johannsson JH, Oddsson E. Inflammatory bowel disease in Iceland, A retrospective nationwide epidemiologic study. Scand J Gastroenterol 1998;33: Bjornsson S, Johannsson JH. Inflammatory bowel disease in Iceland, : a prospective, nationwide, epidemiological study. Eur J Gastroenterol Hepatol 2000;12: Shivananda S, Lennard-Jones J, Logan R, Fear N, Price A, Carpenter L, van Blankenstein M. Incidence of inflammatory bowel disease across Europe: is there a difference between north and south? Results of the European Collaborative Study on Inflammatory Bowel Disease (EC-IBD). Gut 1996;39: Russel MG, Stockbrugger RW. Epidemiology of inflammatory bowel disease: an update. Scand J Gastroenterol 1996;31: Hugot JP, Zouali H, Lesage S, Thomas G. Etiology of the inflammatory bowel diseases. Int J Colorectal Dis 1999;14: Gudmundsson H, Gudbjartsson DF, Frigge M, Gulcher JR, Stefansson K. Inheritance of human longevity in Iceland. Eur J Hum Genet 2000;8: Ingvarsson T, Stefansson SE, Hallgrimsdottir IB, Frigge ML, Jonsson H Jr, Gulcher J, Jonsson H, Ragnarsson JI, Lohmander LS, Stefansson K. The inheritance of hip osteoarthritis in Iceland. Arthritis Rheum 2000;43: Sveinbjornsdottir S, Hicks AA, Jonsson T, Petursson H, Gugmundsson G, Frigge ML, Kong A, Gulcher JR, Stefansson K. Familial aggregation of Parkinson s disease in Iceland. N Engl J Med 2000;343: Gulcher JR, Kristjansson K, Gudbjartsson H, Stefansson K. Protection of privacy by third-party encryption in genetic research in Iceland. Eur J Hum Genet 2000;8: Lange K. Genetic identity coefficients. Mathematical and statistical methods for genetic analysis (statistics for biology and health). New York: Springer Verlag, 1977: Risch N. Linkage strategies for genetically complex traits. I. Multilocus models. Am J Hum Genet 1990;46: Guo SW. Inflation of sibling recurrence-risk ratio, due to ascertainment bias and/or overreporting. Am J Hum Genet 1998;63: Sedlack RE, Whisnant J, Elveback LR, Kurland LT. Incidence of Crohn s disease in Olmsted County, Minnesota, Am J Epidemiol 1980;112: Bennett RA, Rubin PH, Present DH. Frequency of inflammatory bowel disease in offspring of couples both. Gastroenterology 1991;100: Laharie D, Debeugny S, Peeters M, Van Gossum A, Gower-Rousseau C, Belaiche J, Fiasse R, Dupas JL, Lerebours E, Piotte S, Cortot A, Vermeire S, Grandbastien B, Colombel JF. Inflammatory bowel disease in spouses and their offspring. Gastroenterology 2001;120: McConnell RB, Vadheim CM. Inflammatory bowel disease. In: King RA, Rotter JI, Motulsky AG, eds. The genetic basis of common diseases. New York: Oxford University Press, 1992: Weterman IT, Pena AS. Familial incidence of Crohn s disease in The Netherlands and a review of the literature. Gastroenterology 1984;86: Roth MP, Petersen GM, McElree C, Vadheim CM, Panish JF, Rotter JI. Familial empiric risk estimates of inflammatory bowel disease in Ashkenazi Jews. Gastroenterology 1989;96: Orholm M, Fonager K, Sorensen HT. Risk of ulcerative colitis and Crohn s disease among offspring of patients with chronic inflammatory bowel disease. Am J Gastroenterol 1999;94: Taylor KD, Yang H, Rotter JI. Inflammatory bowel disease. II. Gene mapping. Mol Genet Metab 2001;74: Hugot JP, Laurent-Puig P, Gower-Rousseau C, Olson JM, Lee JC, Beaugerie L, Naom I, Dupas JL, Van Gossum A, Orholm M,

7 812 REYNISDOTTIR ET AL. CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 2, No. 9 Bonaiti-Pellie C, Weissenbach J, Mathew CG, Lennard-Jones JE, Cortot A, Colombel JF, Thomas G. Mapping of a susceptibility locus for Crohn s disease on chromosome 16. Nature 1996;379: Cavanaugh J. International collaboration provides convincing linkage replication in complex disease through analysis of a large pooled data set: Crohn disease and chromosome 16. Am J Hum Genet 2001;68: Satsangi J, Parkes M, Louis E, Hashimoto L, Kato N, Welsh K, Terwilliger JD, Lathrop GM, Bell JI, Jewell DP. Two stage genomewide search in inflammatory bowel disease provides evidence for susceptibility loci on chromosomes 3, 7 and 12. Nat Genet 1996;14: Parkes M, Barmada MM, Satsangi J, Weeks DE, Jewell DP, Duerr RH. The IBD2 locus shows linkage heterogeneity between ulcerative colitis and Crohn disease. Am J Hum Genet 2000;67: Hampe J, Shaw SH, Saiz R, Leysens N, Lantermann A, Mascheretti S, Lynch NJ, MacPherson AJ, Bridger S, van Deventer S, Stokkers P, Morin P, Mirza MM, Forbes A, Lennard-Jones JE, Mathew CG, Curran ME, Schreiber S. Linkage of inflammatory bowel disease to human chromosome 6p. Am J Hum Genet 1999;65: Dechairo B, Dimon C, van Heel D, Mackay I, Edwards M, Scambler P, Jewell D, Cardon L, Lench N, Carey A. Replication and extension studies of inflammatory bowel disease susceptibility regions confirm linkage to chromosome 6p (IBD3). Eur J Hum Genet 2001;9: Rioux JD, Silverberg MS, Daly MJ, Steinhart AH, McLeod RS, Griffiths AM, Green T, Brettin TS, Stone V, Bull SB, Bitton A, Williams CN, Greenberg GR, Cohen Z, Lander ES, Hudson TJ, Siminovitch KA. Genomewide search in Canadian families with inflammatory bowel disease reveals two novel susceptibility loci. Am J Hum Genet 2000;66: Cho JH, Nicolae DL, Gold LH, Fields CT, LaBuda MC, Rohal PM, Pickles MR, Qin L, Fu Y, Mann JS, Kirschner BS, Jabs EW, Weber J, Hanauer SB, Bayless TM, Brant SR. Identification of novel susceptibility loci for inflammatory bowel disease on chromosomes 1p, 3q, and 4q: evidence for epistasis between 1p and IBD1. Proc Natl Acad Sci U S A 1998;95: Cho JH, Nicolae DL, Ramos R, Fields CT, Rabenau K, Corradino S, Brant SR, Espinosa R, LeBeau M, Hanauer SB, Bodzin J, Bonen DK. Linkage and linkage disequilibrium in chromosome band 1p36 in American Chaldeans with inflammatory bowel disease. Hum Mol Genet 2000;9: Hampe J, Frenzel H, Mirza MM, Croucher PJ, Cuthbert A, Mascheretti S, Huse K, Platzer M, Bridger S, Meyer B, Nurnberg P, Stokkers P, Krawczak M, Mathew CG, Curran M, Schreiber S. Evidence for a NOD2-independent susceptibility locus for inflammatory bowel disease on chromosome 16p. Proc Natl Acad Sci U S A 2002;99: Duerr RH, Barmada MM, Zhang L, Pfutzer R, Weeks DE. Highdensity genome scan in Crohn disease shows confirmed linkage to chromosome 14q Am J Hum Genet 2000;66: Ma Y, Ohmen JD, Li Z, Bentley LG, McElree C, Pressman S, Targan SR, Fischel-Ghodsian N, Rotter JI, Yang H. A genome-wide search identifies potential new susceptibility loci for Crohn s disease. Inflamm Bowel Dis 1999;5: Rioux JD, Daly MJ, Silverberg MS, Lindblad K, Steinhart H, Cohen Z, Delmonte T, Kocher K, Miller K, Guschwan S, Kulbokas EJ, O Leary S, Winchester E, Dewar K, Green T, Stone V, Chow C, Cohen A, Langelier D, Lapointe G, Gaudet D, Faith J, Branco N, Bull SB, McLeod RS, Griffiths AM, Bitton A, Greenberg GR, Lander ES, Siminovitch KA, Hudson TJ. Genetic variation in the 5q31 cytokine gene cluster confers susceptibility to Crohn disease. Nat Genet 2001;29: Hampe J, Schreiber S, Shaw SH, Lau KF, Bridger S, Macpherson AJ, Cardon LR, Sakul H, Harris TJ, Buckler A, Hall J, Stokkers P, van Deventer SJ, Nurnberg P, Mirza MM, Lee JC, Lennard-Jones JE, Mathew CG, Curran ME. A genomewide analysis provides evidence for novel linkages in inflammatory bowel disease in a large European cohort. Am J Hum Genet 1999;64: Address requests for reprints to: Inga Reynisdottir, decode Genetics, Inc., Sturlugata 8, 101 Reykjavík, Iceland. ingar@decode.is; fax: (354) The following authors (IR, DFG, IM, KK, KS, and JG)are current or former employees of decode Genetics, Inc., and, as such, receive monthly salaries and other benefits from the company. The following investigators (JHJ and SB)at the National University Hospital of Iceland (NUI)have entered into agreements with decode Genetics, Inc., which stipulate that scientific research funds with which the investigators and NUI are affiliated will receive certain financial contributions from decode should the company be able to commercialize products that may result from this research. The following author (JHJ)owns stock in the company worth less than $1,000.