A statistical test for detection of ancestral genetic contributions to disease occurrence in finite populations

Intergenerational and Genealogical Approaches for the Study of Longevity in the Saguenay-Lac-St-Jean Population

The mechanisms of longevity have been the subject of investigations for a number of years. Although the role of genetic factors is generally acknowledged, important questions persist regarding the relative impact of environmental exposures, lifestyle characteristics, and genes. The BALSAC population register offers a unique opportunity to study longevity from an intergenerational and genealogical point of view. Individuals from the Saguenay-Lac-St-Jean population who died at age 90 or older between 1950 and 1974 were selected from this database (n = 576), along with a control group of individuals born in the same period who died between 50 and 75 years of age. For these subjects and controls, spouses' ages at death and parental ages at death and at their birth were investigated using regression analysis. Genealogical reconstructions were carried out for each individual, and various analyses were performed on both groups. Both fathers' and mothers' mean ages at death were significantly higher among the longer-lived cases than among controls whereas spouses' ages at death and parental ages at birth had no effect. Regression analysis confirmed the positive effect of both fathers' and mothers' age at death. Mean kinship coefficients for the parents' generations displayed significant differences, indicating that kinship was higher among subjects than controls (this effect was stronger among the oldest 10% of the subjects). Frequencies and genetic contributions of ancestors were very similar for the two groups, and none of these ancestors appeared more likely to have introduced genetic variants involved in longevity patterns in this French Canadian population.

Molecular and genealogical characterization of the R1443X BRCA1 mutation in high-risk French-Canadian breast/ovarian cancer families

The Quebec population contains about six-million French Canadians, descended from the French settlers who colonized "Nouvelle-France" between 1608 and 1765. Although the relative genetic contribution of each of these founders is highly variable, altogether they account for the major part of the contemporary French-Canadian gene pool. This study was designed to analyze the role of this founder effect in the introduction and diffusion of the BRCA1 recurrent R1443X mutant allele. A highly conserved haplotype, observed in 18 French-Canadian families and generated using 17 microsatellite markers surrounding the BRCA1 locus, supports the fact that the R1443X mutation is a founder mutation in the Quebec population. We also performed haplotyping analysis of R1443X carriers on 19 other families from seven different nationalities; although the same alleles are shared for three markers surrounding the BRCA1 gene, distinct haplotypes were obtained in four families, suggesting multiple origins for the R1443X mutation. Ascending genealogies of the 18 French Canadian families and of controls were reconstructed on an average depth of 10 generations. We identified the founder couple with the highest probability of having introduced the mutation in the population. Based on the descending genealogy of this couple, we detected the presence of geographical concentration in the diffusion pattern of the mutation. This study demonstrates how molecular genetics and demogenetic analyses can complement each other to provide findings that could have an impact on public health. Moreover, this approach is certainly not unique to breast cancer genetics and could be used to understand other complex traits.

A genealogical study of Alzheimer disease in the Saguenay region of Quebec

We performed an analysis of inbreeding and kinship among the ascending genealogies of 205 autopsy-confirmed Alzheimer disease (AD) subjects recruited in the Saguenay area of Québec. We hypothesized that if some traits pertaining to the disease were determined by inherited factors, and if the corresponding genes were not too frequent in the population, it might be possible to detect some clusters of patients related to common ancestors and presenting a level of kinship and/or inbreeding higher than is observed in the unaffected population of the same age. In view of the heterogeneity of the disease, we also verified if some of the factors investigated could be associated more specifically with subsets of cases based on age of onset and on apolipoprotein E (APOE) genotype. Results were compared with those obtained on 205 controls matched for gender, place and year of birth. We found that late-onset AD cases with an APOE-epsilon 4 were significantly more inbred than controls and that this increase was explained by the high level of inbreeding of a few cases whose parents were related at the first-cousin level. This could possibly indicate the implication of a recessive element in a small subset of AD cases in the Saguenay population. We also found that late-onset epsilon 4+ cases were significantly more closely related among themselves than with controls. This increase in kinship may be attributable to the presence of the epsilon 4 allele or to some other unidentified genetic factor possibly acting in conjunction with APOE-epsilon 4.

Geographic clustering of testicular cancer incidence in the northern part of The Netherlands

Geographic variations in testicular cancer incidence may be caused by differences in environmental factors, genetic factors, or both. In the present study, geographic patterns of age-adjusted testicular cancer incidence rates (IRs) in 12 provinces in The Netherlands in the period 1989-1995 were analysed. In addition, the age-adjusted IR of testicular cancer by degree of urbanization was evaluated. Cancer incidence data were obtained from the Netherlands Cancer Registry. The overall annual age-adjusted IR of testicular cancer in The Netherlands in the period 1989-1995 was 4.4 per 100000 men. The province Groningen in the north of the country showed the highest annual IR with 5.8 per 100000 men, which was higher (P < 0.05) than the overall IR in The Netherlands (incidence rate ratio (IRR) 1.3, 95% confidence interval (CI) 1.1-1.6). The highest IR in Groningen was seen for both seminomas and non-seminomas. In addition, Groningen showed the highest age-specific IRs in all relevant younger age groups (15-29, 30-44 and 45-59 years), illustrating the consistency of data. The province Friesland, also situated in the northern part of the country, showed the second highest IR of testicular cancer with 5.3 cases per 100000 men per year (IRR 1.2, 95% CI 1.0-1.5, not significant). This mainly resulted from the high IR of seminoma in Friesland. Analysis of age-adjusted IRs of testicular cancer by degree of urbanization in The Netherlands showed no urban-rural differences at analysis of all histological types combined, or at separate analyses of seminomas and non-seminomas. Geographic clustering of testicular cancer seems to be present in the rural north of The Netherlands with some stable founder populations, which are likely to share a relatively high frequency of genes from common ancestors including genes possibly related to testicular cancer. Although this finding does not exclude the involvement of shared environmental factors in the aetiology of testicular cancer, it may also lend support to a genetic susceptibility to testicular cancer development. Testicular cancer cases in stable founder populations seem particularly suitable for searching for testicular cancer susceptibility genes because such genes are likely to be more frequent among affected men in such populations.

Consanguinity and multiple sclerosis in Orkney

For all patients with multiple sclerosis in Orkney who were alive in 1974 or who had died during 1958 to 1974, pedigree data were analysed. The relationship between their parents, expressed by kinship coefficients, was compared with that between parents of matched controls. The closer relationship between parents of patients suggests that it is the genes that are shared by a patient's parents, and that he inherits from both, that influence his susceptibility to multiple sclerosis.

Testing for familial aggregation of a dichotomous trait

Familial aggregation of a disease is usually tested using a chi square statistic on a sample of pairs of related individuals. Standard sampling designs often produce dependent observations (multiple pairs from the same family), possibly inflating the numerical value of the statistic over the value independent observations would produce. This article presents a derivation of the chi square statistic for familial aggregation, allowing for dependency because of the inclusion of the same and related persons in multiple pairs. For intraclass aggregation, the standard chi square statistic is appropriate. For interclass aggregation, an adjustment to the standard chi square statistic is needed; the adjustment depends on the disease frequency and the intraclass aggregation within each class.