Statistics from Altmetric.com
See article on page 540
Conditions which arise from single gene defects demonstrate a simple mendelian mode of inheritance. There are however a large number of common conditions in which genetic factors are thought to be involved but are not clearly passed from one individual to another and merely cluster in families. Such observations raise a number of questions including: what is the size of the genetic contribution to the disorder and how might the susceptibility gene(s) involved in the development of the condition be identified ?
The National Academy of Sciences-National Research Council twin registry of almost 16 000 twin pairs reported concordance rates for cirrhosis of 16.9% in monozygotic twins and 5.3% in dizygotic twins, implying a genetic predisposition to this complication of alcohol abuse.1 Although it is possible that twins share not only genes but a similar environment, with greater (in utero) environmental sharing in monozygotic twins, there is increasing epidemiological2 and laboratory3 evidence to support a genetic basis to familial clustering. However, the concordance rate for alcoholic liver disease in monozygotic twins falls well below 100%, highlighting the role of environmental as well as genetic factors.
The magnitude of the genetic contribution to disease is not easy to assess in common disorders. The overall recurrence risk ratio, or lambda (λ)s of Risch,4 which is the ratio of the risk of a second sibling developing a disease where the first sibling already has the disease divided by the population prevalence of the disease, can be used to estimate the degree of familial clustering and, therefore, quantify the genetic risk. A λs of 1 implies no genetic contribution whereas in single gene defects such as haemophilia, the ratio exceeds 1000. Common complex disorders including for example diabetes, epilepsy, psoriasis, and multiple sclerosis, where both genetic and environmental effects are contributing to disease development, demonstrate a λs of <20. The ratio for alcoholic cirrhosis is not known although it is likely to be of a similar order of magnitude. In general terms the smaller the λs the harder it is to identify not only the gene(s) involved in the disease but also separate the primary aetiological mutation within the gene from the many polymorphisms which exist within the gene region.
In the search for genetic susceptibility loci for common complex disorders, three main approaches have been used: population based case control studies, intrafamilial association studies, and classical linkage analysis. Case control studies are a sensitive method of gene detection and the collection of subjects is relatively straightforward. However, they require a candidate gene to be studied and are prone to the generation of inconsistent results due to false positives arising from population mismatch or as a result of a random chance event because of small numbers. Intrafamilial association studies also require a candidate gene but eliminate false positives which arise from population mismatch by examining transmission of susceptibility alleles from parents to disease affected offspring.5 As parental DNA is required in addition to that of the index case, these studies are not as simple to conduct as case control studies. Linkage analysis is a powerful but complex tool for detecting major genes and is generally used as a means of systematically screening the genome in family based data sets. Its use is restricted because the collection of sib-pairs and multiple family members may be extremely difficult (as is the case with alcoholic liver disease) and its ability to detect genes of “modest” effect has proved limited.6
Candidate gene studies in alcoholic liver disease have recently moved from genes encoding ethanol metabolism to immune response genes because of their potential role in disease pathogenesis. In this issue, Groveet al (see page 540) have performed a population based, case control study to determine if the polymorphism associated with low interleukin 10 (IL-10) production is associated with and, therefore, a risk factor for advanced alcoholic liver disease (AALD). The A allele of the single nucleotide polymorphism (SNP) at position −627 (C-A) in the promoter of theIL-10 gene was found to be more common in patients with AALD compared with either normal controls or heavy drinkers with no evidence of liver disease, resulting in a relative risk for the development of AALD of 2.04 and 1.76, respectively. This paper is the first report of an association between polymorphism of the cytokine gene IL-10 and AALD. Although it is not possible with this type of study to determine if the associated allele is the primary disease causing mutation or is acting as a marker in linkage disequilibrium with a nearby as yet unknown aetioloigcal mutation, even in a neighbouring gene, there is good circumstantial functional evidence to implicate the IL-10gene in AALD. The data presented are convincing, in that the total number of subjects used in the study was significant (621) and two sets of appropriate controls were used as comparators. The potential pitfalls of the population based case control approach in general and those specific to AALD which may have resulted in false positive results are carefully discussed by the authors. As they point out, family based AALD data sets would be useful to exclude population mismatching but their collection would be problematic. Replication of the data of Grove et al in a further independent data set is however vital, even if this is in the form of a second population based case control study. While confirmatory findings in an ethnically distinct population as opposed to similar ethnicity would be desirable, this is not essential. Replication of these data would strengthen the argument for subsequent functional molecular biological studies to help “nail” the primary aetiological disease causing mutation.
As with the study of all complex diseases, establishment of large multiple independent data sets is vital in attempts at identifying susceptibility loci which are likely to be exerting small but clinically significant effects. It is unlikely that any one group will have sufficient numbers of patients to establish more than one data set and, therefore, working collaborations between groups should be encouraged. With the ongoing emergence of more detailed genetic maps, including the availability of in silicocandidate SNP markers and the development of technology capable of performing large scale genotyping, those groups who have established large data sets will be the first to identify susceptibility genes.
See article on page 540
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.