Genotype-Specific Recurrence Risks as Indicators of the Genetic Architecture of Complex Diseases

A statistic is introduced that relates discoveries made in genome-wide association (GWA) studies to patterns of disease risks among relatives. The genotype-specific recurrence risk (GSR) is the genotype-specific risk to relatives of known relationship to affected probands. The GSRs can be used for three purposes. (1) They can provide an independent test of whether an allele identified in a GWA study is associated with the disease. (2) They can provide a test of whether interactions among loci affecting the disease are multiplicative. (3) They can be used by genetic counselors to incorporate information from GWA studies for predicting the risk to relatives of known genotype. Under a multiplicative model of disease causation, the GSRs for a locus are the genotypic risks in probands for that locus multiplied by λR/λjR, where λR is Risch's recurrence risk ratio and λjR is the contribution to λR from the locus of interest. If there is saturation of risk with increasing numbers of causative alleles, then observed GSRs for individuals with high-risk genotypes will be lower than predicted by the multiplicative model.