We completed fine mapping of nine positional candidate regions for attention-deficit/hyperactivity disorder (ADHD) in an extended population sample of 308 affected sibling pairs (ASPs), constituting the largest linkage sample of families with ADHD published to date. The candidate chromosomal regions were selected from all three published genomewide scans for ADHD, and fine mapping was done to comprehensively validate these positional candidate regions in our sample. Multipoint maximum LOD score (MLS) analysis yielded significant evidence of linkage on 6q12 (MLS 3.30; empiric P=.024) and 17p11 (MLS 3.63; empiric P=.015), as well as suggestive evidence on 5p13 (MLS 2.55; empiric P=.091). In conjunction with the previously reported significant linkage on the basis of fine mapping 16p13 in the same sample as this report, the analyses presented here indicate that four chromosomal regions--5p13, 6q12, 16p13, and 17p11--are likely to harbor susceptibility genes for ADHD. The refinement of linkage within each of these regions lays the foundation for subsequent investigations using association methods to detect risk genes of moderate effect size.

Attention-deficit hyperactivity disorder (ADHD) is one of the most common childhood behavioral disorders. Genetic factors contribute to the underlying liability to develop ADHD. Reports implicate variants of genes important for the synthesis, uptake, transport and receptor binding of dopamine in the etiology of ADHD, including DRD4, DAT1, DRD2, and DRD5. In the present study, we genotyped a large multiplex sample of ADHD affected children and their parents for polymorphisms in genes previously reported to be associated with ADHD. Associations were tested by the transmission disequilibrium test (TDT). The sample is sufficient to detect genotype relative risks (GRRs) for putative risk alleles. The DRD4 gene 120-bp insertion/deletion promoter polymorphism displayed a significant bias in transmission of the insertion (v 2 ¼ 7.58, P ¼ 0.006) as suggested by an analysis of a subset of these families. The seven repeat allele of the DRD4 48-bp repeat polymorphism (DRD4.7) was not significantly associated with ADHD in the large sample in contrast to our earlier findings in a smaller subset. We replicate an association of a dinucleotide repeat polymorphism near the DRD5 gene with ADHD by showing biased nontransmission of the 146-bp allele (P ¼ 0.02) and a trend toward excess transmission of the 148-bp allele (P ¼ 0.053). No evidence for an association was found for polymorphisms in DRD2 or DAT1 in this sample. The DRD5 146-bp (DRD5.146) allele and the DRD4 240-bp (DRD4.240) allele of the promoter polymorphism emerge as the two DNA variants showing a significant association in this large sample of predominantly multiplex families with ADHD, with estimated GRRs of 1.7 and 1.37, respectively.

In this unique sample of parents from multiply affected ADHD families, novelty seeking and the 48-bp DRD4 variant were associated with a lifetime history of ADHD. However, the association between novelty seeking and ADHD does not appear to be due to variation in the 48-bp DRD4 variant.

Sibling correlations, familial clustering and candidate gene associations provide strong support for verbal working memory as a candidate endophenotype for ADHD. More complex models of, and larger sample sizes for, genetic association with cognitive functions are encouraged for future study.