Sample Size Requirements for Association Studies of Gene-Gene Interaction

W. J. Gauderman
2002 American Journal of Epidemiology  
In the study of complex diseases, it may be important to test hypotheses related to gene-gene (G ϫ G) interaction. The success of such studies depends critically on obtaining adequate sample sizes. In this paper, the author investigates sample size requirements for studies of G ϫ G interaction, focusing on four study designs: the matched-case-control design, the case-sibling design, the case-parent design, and the case-only design. All four designs provide an estimate of interaction on a
more » ... raction on a multiplicative scale, which is used as a unifying theme in the comparison of sample size requirements. Across a variety of genetic models, the case-only and case-parent designs require fewer sampling units (cases and case-parent trios, respectively) than the casecontrol (pairs) or case-sibling (pairs) design. For example, the author describes an asthma study of two common recessive genes for which 270 matched case-control pairs would be required to detect a G ϫ G interaction of moderate magnitude with 80% power. By comparison, the same study would require 319 case-sibling pairs but only 146 trios in the case-parent design or 116 cases in the case-only design. A software program that computes sample size for studies of G ϫ G interaction and for studies of gene-environment (G ϫ E) interaction is freely available (http://hydra.usc.edu/gxe). Am J Epidemiol 2002;155:478-84. 478 by guest on November 29, 2015 http://aje.oxfordjournals.org/ Downloaded from * A sampling unit is a pair for the case-control and case-sibling designs, a trio for the case-parent design, and a case for the case-only design. † Assumptions: a recessive model at both loci, with Pr(G ϭ 1) ϭ 0.40 at GSTM1 and Pr(H ϭ 1) ϭ 0.25 at GSTT1 and main-effect relative risks R g ϭ 1 and R h ϭ 1. ‡ Compared with the case-control design; ratios above (below) 1.0 indicate greater (lesser) efficiency. Downloaded from * A sampling unit is a pair for the case-control and case-sibling designs, a trio for the case-parent design, and a case for the case-only design. † Compared with the case-control design; ratios above (below) 1.0 indicate greater (lesser) efficiency. Downloaded from * A sampling unit is a pair for the case-control and case-sibling designs, a trio for the case-parent design, and a case for the case-only design. † Assumed dominance models are dominant for locus G and recessive for locus H. ‡ Compared with the case-control design; ratios above (below) 1.0 indicate greater (lesser) efficiency.
doi:10.1093/aje/155.5.478 pmid:11867360 fatcat:6wptak2srvgs3pvpva2kdh3sqq