1. szöveg

Genetic epidemiology

Jaakko Kaprio

 

Genetic epidemiology is the study of the aetiology, distribution, and control of disease in groups of relatives and of inherited causes of disease in populations. From its parent disciplines of genetics and epidemiology, it has inherited the key elements of studying defined populations while investigating the roles of genes and the environment in relation to each other.

The primary goal of genetic epidemiology is resolving the genetic architecture of a disease –that is, establishing whether it has a genetic component and the relative size of that genetic effect in relation to environmental effects. In this context the environment is understood to encompass everything non-genetic, from the intrauterine environment to physical and chemical effects and to behavioural and social aspects. Effects from different environmental categories are insufficiently taken into account in most genetic epidemiological studies.

The estimation of the genetic component comes from family studies, in which the disease risk in relatives of a patient is compared with the general risk of disease in the population. Naturally, it is important that the patients studied are representative of the population. However, an increased risk in family members does not necessarily indicate that the disease has an inherited component accounted for by genetic variation. Familial aggregation can be due to non-genetic factors in the family environment, such as the physical environment of the home and the family's socioeconomic status. Interindividual differences in religiosity, a protective factor against alcohol misuse in many societies, are largely accounted for by non-genetic familial factors. Stratification of risk by degree of relatedness (first degree versus second degree relatives) and comparisons with unrelated individuals living in the same household (typically spouses) can help distinguish between genetic and non-genetic familial effects. A thorough family history provides excellent information about possible genetic risk in families.

Families with several diseased family members, in particular those with large pedigrees, are particularly informative, both for establishing that genes matter and for identifying the specific genes. Such families are rare for the common diseases now at the centre of genetic epidemiological research. Other traditional designs for distinguishing non-genetic family effects from genetic effects have been studies of twins and adoptees, but study of half-sibs, who are increasingly common with higher divorce rates, is also valuable. Combinations of designs, such as the inclusion of parents and sibs in twin studies, can permit more incisive estimation of the role of genetic factors and account for assortative mating and transmission of non-genetic effects from parents to offspring.

A model of complex disease

After the size of a genetic component has been established, we seek to establish how many genes are contributing to the disease. In complex diseases many genes act through several intermediate phenotypes to increase disease risk, but the same genes can also influence other diseases. Environmental factors can independently affect the risk of disease, but also act through the intermediate phenotypes. For diseases such as coronary heart disease, we know something of the genetics of such intermediate phenotypes such as blood lipids, haemostatic factors, and blood pressure.

In any single gene conferring disease susceptibility there are generally multiple alleles that affect disease risk to different degrees. For example, the cystic fibrosis gene has over 800 mutations associated with the disease. A decade of research has indicated that the genotype poorly predicts phenotype, bringing new complexity to the diagnosis of cystic fibrosis while permitting identification of carriers of the cystic fibrosis gene. Such multiplicity of mutations and disease associated alleles is more the rule than the exception. Also, mutations in the cystic fibrosis gene are associated with other phenotypes such as male infertility and allergic bronchopulmonary aspergillosis. For other diseases, multiple genes are known to be involved. Migraine has been shown to have a genetic component in family and twin studies, but identification of migraine genes has so far been restricted to a rare subtype of migraine, familial hemiplegic migraine. Calcium channel genes on chromosomes 1 and 19 account for many but not all cases of familial hemiplegic migraine. To complicate matters further, not even all family members with a mutation have manifest migraine, and these disease mutations have not been convincingly associated with the common forms of migraine.

Complex interactions are probably important in explaining differences in disease prevalence among populations. For example, population based twin studies in the Nordic countries in the 1990s suggest that the heritability of asthma is about 70%. Strictly comparative earlier studies are not available, but twin studies from the 1970s suggested that the heritability was under 50%, and at the same time asthma has increased in prevalence. While susceptibility genes for asthma cannot have changed in the population during one generation, their expression and interaction with environmental factors may have changed, and may be reversed if the appropriate environmental factors can be identified and eliminated.

BMJ 2000;320:1257