Purpose: Somatic mutations that result in the activation of the growth factor receptor KIT are commonly found in gastrointestinal stromal tumors (GISTs). Six families have been reported in which a germ-line mutation in KIT is associated with an autosomal dominant predisposition to the development of GISTs. Hyperpigmentation, urticaria pigmentosa, and dysphagia have been described in some, but not all, families. Preliminary correlations between the site of mutation and the clinical phenotype have been proposed, but the strength of these associations is not defined.Design: A large kindred with multiple GISTs, hyperpigmentation, and dysphagia was identified after the index case presented with multiple GISTs. A germ-line mutation in KIT (W557R) was identified in an affected cousin, after which a large family meeting was held and testing offered. Clinical data were obtained by interview and, whenever possible, medical record documentation.Results: To date, 19 individuals have been tested, and the mutation has been shown to cosegregate with the syndrome. The phenotypic expression, however, is variable. GISTs, often presenting as upper gastrointestinal bleeding, and hyperpigmentation are common, but not diagnosed in all documented or obligate carriers. Dysphagia is a less prevalent complaint. The diagnosis of GISTs appears to be made at a younger age in more recent generations. Metastatic disease is uncommon. Conclusions:A germ-line mutation in KIT resulting in an amino acid substitution in the juxtamembrane region is associated with a syndrome of GIST, hyperpigmentation, and dysphagia, although the prominence of each component varies.

Guidelines recommend patients be informed of their incidental results (IR) when undergoing genomic sequencing (GS), yet there are limited tools to support patients' decisions about learning IR. The aim of this study is to develop and test the usability of a decision aid (DA) to guide patients' selection of IR, and to describe patients' preferences for learning IR following use of the DA. We developed and evaluated a DA using an iterative, mixed-methods process consisting of (1) prototype development, (2) feasibility testing, (3) cognitive interviews, (4) design and programming, and (5) usability testing. We created an interactive online DA called the Genomics ADvISER, a genomics decision AiD about Incidental SEquencing Results. The Genomics ADvISER begins with an educational whiteboard video, and then engages users in a values clarification exercise, knowledge quiz and final choice step, based on a 'binning' framework. Participants found the DA acceptable and intuitive to use. They were enthusiastic towards GS and IR; all selected multiple categories of IR. The Genomics ADvISER is a new patient-centered tool to support the clinical delivery of incidental GS results. The Genomics ADvISER fills critical care gaps, given the health care system's limited genomics expertise and capacity to convey the large volume of IR and their myriad of implications.

No abstract

The genetic factors that cause clustering of colorectal cancers (CRCs) other than mutations in the mismatch repair (MMR) genes are not well understood. Clustering in families who lack MMR gene mutations may be attributable to low-penetrance mutations. Hypothetically, mono-allelic MYH mutations could contribute to the risk of CRC in these families. Using Fisher's exact test and logistic regression, we compared the frequency of the known disease-causing MYH mutations Y165C, G382D and 466delE in 137 probands (117 cases with CRC and 20 cases diagnosed on the basis of adenomatous polyps only) from families with three or more CRCs but negative for mutations in the MMR genes and in 967 healthy controls with comparable ethnic backgrounds. Of 137 cases, 6 (4.4%) carried mono-allelic MYH mutations compared with 16 of 967 (1.6%) controls. In addition, three bi-allelic MYH mutation carriers, who eventually developed MYH-associated polyposis, were also identified in families with pedigree structures consistent with dominant inheritance of CRC susceptibility. By Fisher's exact tests, there was a statistically different frequency of cases with any MYH mutation (mono- or bi-allelic carriers; P-value = 0.002) and of cases with mono-allelic MYH mutation (P = 0.04) compared with the controls. Using a logistic regression model, the unadjusted odds ratio associated with any MYH mutation was 4.14 (P-value < 0.001); for mono-allelic carriers, it was 2.79 (P-value = 0.04). Adjusting for ethnic backgrounds, gender and age, the odds ratio associated with any disease-causing MYH mutation was 3.23 (P-value = 0.01); for mono-allelic carriers, it was 1.99 (P-value = 0.20). Overall, the results support previous studies suggesting that mono-allelic mutations of MYH constitute low-penetrance CRC-causing alleles. These data further support a model in which low-penetrance alleles are enriched in MMR gene mutation-negative CRC families.

Germline mutations in MSH6 can cause HNPCC, which is associated with a tumor phenotype featuring MSI. However, tumors arising in persons with disease-causing mutations of MSH6 may or may not exhibit MSI. We used D-HPLC to screen for germline mutations in the promoter region, the coding region and the 3-UTR of MSH6. Eighty-four families, enrolled on the basis of Amsterdam I and II criteria (HNPCC families) and less stringent criteria (HNPCC-like families), were tested for MMR gene mutations; 27 families had a disease-causing mutation in MLH1 or MSH2, and the remaining 57 families were tested for mutations in MSH6. Two protein-truncating mutations were identified in each of 2 families fulfilling the Amsterdam I criteria, being present in persons affected with early-onset colorectal cancers exhibiting MSI. Immunohistochemical analysis showed that expression of both MSH2 and MSH6 proteins was lost in the cancer cells of the 2 mutation carriers but only MSH6 protein expression was lost in 2 adenomatous polyps. A third possibly disease-causing mutation was found in a person affected with a tumor that did not exhibit MSI. In addition, we found 4 new polymorphisms and determined that neither of the 2 studied by association analysis conferred susceptibility to colorectal or endometrial cancer. HNPCC is an autosomal dominant syndrome characterized by susceptibility to CRC with a mean age at diagnosis of approximately 45 years. 1 In HNPCC patients, the risk of developing cancers of the endometrium, ovary, stomach, small intestine, hepatobiliary tract, renal pelvis and brain is increased. 2,3 Germline mutation in one of the genes of the DNA MMR system is most often responsible for the syndrome. DNA-based genetic testing usually is offered to persons who satisfy certain family history criteria. The Amsterdam I criteria select families that have (i) 3 or more relatives affected with CRC in at least 2 consecutive generations, one person being a first-degree relative of the other 2; (ii) at least one CRC diagnosed before age 50; and (iii) absence of gastrointestinal polyposis. 4 Because these criteria did not include the noncolonic cancers that are associated with MMR gene mutations, the Amsterdam I criteria were modified to include cancers of the endometrium, small bowel and renal pelvis (Amsterdam II). 5 HNPCC is caused primarily by inactivating mutations in one of the genes that constitute the MMR system, most commonly MLH1 or MSH2. 6,7 Mutations in these 2 genes are associated with a tumor phenotype featuring accumulation of mutations at microsatellite repeat sequences, which is referred to as MSI. 8 -10 MSI results from failure of the MMR system to repair deletions or insertions of repeat units of microsatellite loci that spontaneously occur during DNA replication. A tumor is classified as MSI-high when a fraction greater than one-third of the microsatellite loci tested display novel locus-specific fragments; a tumor is classified as non-MSIhigh when a fraction Ͻ0.3 of the microsatellite loci tested exhibit novel fragments...