Follow-up study on a susceptibility locus for schizophrenia on chromosome 6q

Evidence for suggestive linkage to schizophrenia with chromosome 6q markers was previously reported from a two-stage approach. Using nonparametric affected sib pairs (ASP) methods, nominal p-values of 0.00018 and 0.00095 were obtained in the screening (81 ASPs; 63 independent) and the replication (109 ASPs; 87 independent) data sets, respectively. Here, we report a follow-up study of this 50cM 6q region using 12 microsatellite markers to test for linkage to schizophrenia. We increased the replication sample size by adding an independent sample of 43 multiplex pedigrees (66 ASPs; 54 independent). Pairwise and multipoint nonparametric linkage analyses conducted in this third data set showed evidence consistent with excess sharing in this 6q region, though the statistical level is weaker (p=0.013). When combining both replication data sets (total of 141 independent ASPs), an overall nominal p-value=0.000014 (LOD=3. 82) was obtained. The sibling recurrence risk (lambdas) attributed to this putative 6q susceptibility locus is estimated to be 1.92. The linkage region could not be narrowed down since LOD score values greater than three were observed within a 13cM region. The length of this region was only slightly reduced (12cM) when using the total sample of independent ASPs (204) obtained from all three data sets. This suggests that very large sample sizes may be needed to narrow down this region by ASP linkage methods. Study of the etiological candidate genes in this region is ongoing.