Penetrance of mutations in the familial Wilms tumor gene FWT1.

Wilms tumor is an embryonal kidney cancer that affects one in 10 000 children. Epidemiologic studies have shown that 1%–3% of cases of Wilms tumor are familial and that a predisposition to Wilms tumor is probably caused by rare germline mutations acting in a dominant fashion (1). The risks of Wilms tumor conferred by mutations in these genes are poorly characterized, with estimates of their penetrance ranging from 18% to 63% (2–4). These estimates represent an average of the risks of all genes that predispose an individual to Wilms tumor and, therefore, are influenced by population-dependent variation in the prevalence of mutations in different genes. Constitutional mutations in the WT1 gene on chromosome 11p13 predispose an individual to Wilms tumor and are associated with genitourinary abnormalities. Germline WT1 mutations have been reported in only four families (three with two cases of Wilms tumor and one with three cases of Wilms tumor) and do not appear to be a common cause of familial predisposition to the disease [reviewed in (5)]. Several genetic syndromes (e.g., the Beckwith– Wiedemann syndrome) have been associated with a predisposition to Wilms tumor, but such syndromes only rarely result in familial cases of the disease [reviewed in (5)]. We previously localized a familial Wilms tumor predisposition gene, designated FWT1, to chromosome 17q12– q21 by genetic linkage analysis of a French–Canadian pedigree, MON 480 (6). Subsequently, the existence of the locus was confirmed by extension of this family (logarithm of odds [LOD] score 4 5.7) and by strong evidence of linkage (LOD score 4 2.7) in an unrelated family, K1104, from Utah (5). Unlike mutations in WT1, mutations in FWT1 are not commonly associated with developmental abnormalities and do not appear to predispose an individual to early-onset Wilms tumor (5). Moreover, on the basis of currently available information, FWT1 does not appear to function as a tumor suppressor gene (7). In this study, we have estimated the penetrance of FWT1 mutations and have thus provided the first gene-specific estimate of the risks conferred by a Wilms tumor susceptibility gene. The penetrance of FWT1 was estimated by use of the two families (MON 480 and K1104) in which Wilms tumor is clearly linked to the gene. None of the other 17 Wilms tumor families in our series generate an LOD score greater than 0.4 when markers in the vicinity of FWT1 are used [(5); Rahman N: unpublished data]. Both FWT1-linked pedigrees were systematically extended so that samples from all individuals willing to participate, irrespective of their disease status, were collected. Genotyping of microsatellite markers D17S250, D17S806, and D17S1820 was used to determine the carrier status of 113 potential carriers of the FWT1 mutations segregating in the two families (67 individuals from MON 480 and 46 individuals from K1104). These markers are approximately ordered cen–D17S250–6.5 cM–D17S806–6 cM–D17S1820–tel (where cen 4 centromere, cM 4 centimorgan, and tel 4 telomere) and are located within the 14-cM interval containing FWT1 defined by recombinants