Comment on the Phosphodiesterase 4D Replication Study by Bevan et al
To the Editor:
We observe with interest the results of the study by Bevan et al,1 where they attempted to replicate our findings associating the PDE4D gene with stroke. We appreciate the effort by the authors and their presentation of their data in some detail. The latter allowed us to compare their results with ours, including some unpublished results. In our original publication,2 the main findings are summarized in Figure 4, where haplotypes are grouped into 3 categories: the at-risk haplotypes, the “wild-type,” and the protective haplotype. Although the definition of the at-risk haplotypes involved multiple markers, the protective haplotype, in our data, is characterized by one single-nucleotide polymorphism, specifically allele A of SNP45. This makes it particularly suitable for a replication in a different population. In Table I of Bevan et al, allele A has frequency of 15.4% in 621 stroke patients and 18.7% in 848 controls. We calculated that this corresponds to 191 copies of allele A and 1051 Gs in stroke patients, and 317 As and 1379 Gs in controls. Applying Fisher exact test gives a 2-sided probability value of 0.02 and a 1-sided probability value of 0.01. For the same allele counts, estimated risk of A relative to G is 0.79, or equivalently risk of G relative to A is 1.26. This agrees well with our estimated relative risk of 1.33 for allele G for all stroke patients (in Table 1 of Gretarsdottir et al2). Hence, even though Bevan et al did not consider this result as significant, which could be attributable to the use of somewhat different statistical tests and adjustment for multiple comparisons, we are nonetheless encouraged by these results, especially because we have obtained similar results for SNP45 in a replication study of ours (Gretarsdottir et al, unpublished data, 2005). In summary, we appreciate the difficulties of getting definitive results for a replication study when the effect of the variant may only be modest and proper care has to be taken to adjust for multiple testing of variants and phenotypes. And we agree with the authors that further investigation is needed to develop a better understanding of the role of PDE4D in stroke.
Bevan S, Porteous L, Sitzer M, Markus HS. Phosphodiesterase 4D gene, ischemic stroke, and asymptomatic carotid atherosclerosis. Stroke. 2005; 36: 949–953.
Gretarsdottir S, Thorleifsson G, Reynisdottir ST, Manolescu A, Jonsdottir S, Jonsdottir T, Gudmundsdottir T, Bjarnadottir S, Einarsson OB, Gudjonsdottir HM, Hawkins M, Gudmundsson G, Gudmundsdottir H, Andrason H, Gudmundsdottir AS, Sigurdardottir M, Chou TT, Nahmias J, Goss S, Sveinbjörnsdottir S, Valdimarsson E, Jakobsson F, Agnarsson U, Gudnason V, Kong A, Thorgeirsson G, Fingerle J, Gurney M, Frigge ML, Kong A, Stefansson K, Gulcher JR. The gene encoding phosphodiesterase 4D confers risk of ischemic stroke. Nat Genet. 2003; 35: 131–138.