Fine mapping and confirmation of linkage to chromosome 9q22 in colorectal neoplasia kindreds. C. Gray-McGuire1,4, R.C. Elston1,4, S.D. Markowitz4,5,6, J. Potter7, N. Lindor8, G.L. Wiesner2,3,4,5. 1) Dept Epi and Biostat, Case Western Reserve Univ, Cleveland, OH; 2) Dept Genetics, Case Western Reserve Univ, Cleveland, OH; 3) Center for Human Genetics, Case Western Reserve University and Univ Hospitals of Cleveland, Cleveland, OH; 4) Ireland Comprehensive Cancer Center, Western Reserve University and Univ Hospitals of Cleveland, Cleveland, OH; 5) Dept of Medicine, Case Western Reserve University and University Hospitals of Cleveland, Cleveland, OH; 6) Howard Hughes Medical Institute, Cleveland, OH; 7) Fred Hutchinson Cancer Institute and Colon Cancer Family Registry, Seattle, WA; 8) Mayo Clinic and Colon Cancer Family Registry, Rochester MN.
Colorectal Cancer is the second leading cause of cancer in American adults and presents a substantial public health concern. While ~5% of cases are accounted for by familial adenomatus polyposis and hereditary nonpolyposis colon cancer, 20% of cases have a family history of colon cancer. This study is a follow-up of a genome scan for colorectal neoplasia predisposing genes in a European American sample comprising 185 sibling pairs ascertained through two or more sibs with a history of colorectal cancer (Wiesner et al. PNAS, 2003). The primary area of interest from the genome scan was chromosome 9q22.2-31.2. For this study, genotypes for 10 additional markers were generated for 1) a revised version of the sample mentioned above (182 sibling pairs), referred to as ROC and 2) a confirmation collection (CC) of 90 sibling pairs. These fine mapping regions were analyzed using the weighted Haseman-Elston regression (dependent variable is a weighted average of the squared sib-pair trait difference and the squared mean-corrected sum) and the sibling pair mean test. The marker with the strongest signal in the ROC and ROC+CC was D9s1786 (-log p-values =3.25, and 3.38, respectively) the same marker indicated in the genome scan. The mean test yielded -log p-values of 3.30, and 3.79 for the ROC and ROC+CC collections, respectively. These results both confirm the presence of a susceptibility gene on chromosome 9 as well as narrow the region most likely to contain this susceptibility gene to 9q22.31-9q22.33.