Comparison of clinical schemas and morphologic features in predicting Lynch syndrome in mutation-positive patients with endometrial cancer encountered in the context of familial gastrointestinal cancer registries
Ryan P.,Mount Sinai Hospital |
Aronson M.,Familial Gastrointestinal Cancer Registry |
Ferguson S.E.,Princess Margaret Hospital |
Bapat B.,Familial Gastrointestinal Cancer Registry |
And 8 more authors.
Cancer | Year: 2012
BACKGROUND: Endometrial cancer (EC) is the most common extraintestinal malignancy in Lynch syndrome (LS) and often is the sentinel malignancy, yet there is no consensus regarding LS-EC detection algorithms. In this study, the authors determined the efficacy of family/personal history and tumor morphology in predicting LS in a cohort of patients with EC who had mutation-proven LS. METHODS: Amsterdam II (AmII) criteria, revised Bethesda guidelines (rBG), and Society of Gynecologic Oncologists (SGO) clinical screening criteria were applied to the pedigrees of 76 patients with mutation-proven LS who had pathology-proven EC. When tumors were tested for microsatellite instability (MSI) phenotype status or mismatch-repair protein-immunohistochemical (MMR-IHC) expression, those results also were reviewed, and LS-associated histopathologic features were documented in 38 available patients. RESULTS: Of 76 patients, 36%, 58%, 71%, and 93% would have been selected for further testing for LS by pedigree screening at the time of EC diagnosis with rBG, AmII, SGO 20%-to-25%, and SGO 5%-to-10% criteria, respectively. Ninety percent (18 of 20 tumors) of tested ECs had high MSI, and 96% (22 of 23 tumors) had abnormal MMR-IHC expression. At least 1 LS-EC morphologic feature was present in 16 of 38 tumors (42%). CONCLUSIONS: Clinical screening criteria had variable efficacy for the identification of LS-associated EC, and SGO 5%-to-10% criteria performed best. Characteristic pathologic features were present in a minority of patients. Although a high proportion of LS-ECs had the MSI phenotype and were MMR deficient, the specificity of these tests and of clinical screening for LS in unselected patients with EC has been poorly described. Prospective studies to determine the optimal combination of these screening modalities are required. © 2011 American Cancer Society. Source
Serrano P.E.,McMaster University |
Grant R.C.,Samuel Lunenfeld Research Institute |
Berk T.C.,Familial Gastrointestinal Cancer Registry |
Kim D.,University of Toronto |
And 9 more authors.
Annals of Surgery | Year: 2015
Objective: To describe the natural history and outcomes of surveillance of duodenal neoplasia in familial adenomatous polyposis (FAP) Background: Duodenal cancer is the most common cause of death in FAP. Methods: Cohort study of patients prospectively enrolled in an upper endoscopic surveillance protocol from 1982 to 2012. The duodenum was assessed by side-viewing endoscopy and classified as stage 1 to 5 disease. Endoscopic and/or operative interventions were performed according to stage. Results: There were 218 patients in the protocol (98 with advanced stage). They had a median of 9 endoscopies (range: 2-25) over a median of 11 years (range: 1-26). Median age at diagnosis of stage 3 disease (adenoma: 2.1-10 mm) was 41 years and stage 4 disease (adenoma >10 mm) was 45 years. Median time from first esophagogastroduodenoscopy to stage 4 disease was 22.4 years. The risk of stage 4 disease was 34.3% [95% confidence interval (CI) 23.8-43.4] at 15 years. In multivariate analysis, sex, type of colorectal surgery, years since colorectal surgery, and stage were significantly associated with risk of progression to stage 4 disease. Five of 218 (2.3%) patients developed duodenal cancer at median age of 58 years (range: 51-65). The risk of developing duodenal cancer was 2.1% (95% CI: 0-5.2) at 15 years. Conclusions: Patients with advanced duodenal polyposis progress in the severity of disease (size and degree of dysplasia); however, the rate of progression to carcinoma is slow. Aggressive endoscopic and surgical intervention, especially in the presence of large polyps and high-grade dysplasia, appears to be effective in preventing cancer deaths in FAP. Copyright © 2014 Wolters Kluwer Health, Inc. All rights reserved. Source
Bakry D.,University of Toronto |
Aronson M.,Familial Gastrointestinal Cancer Registry |
Durno C.,Familial Gastrointestinal Cancer Registry |
Rimawi H.,Jordan University of Science and Technology |
And 23 more authors.
European Journal of Cancer | Year: 2014
Background Constitutional mismatch repair deficiency (CMMRD) is a devastating cancer predisposition syndrome for which data regarding clinical manifestations, molecular screening tools and management are limited. Methods We established an international CMMRD consortium and collected comprehensive clinical and genetic data. Molecular diagnosis of tumour and germline biospecimens was performed. A surveillance protocol was developed and implemented. Results Overall, 22/23 (96%) of children with CMMRD developed 40 different tumours. While childhood CMMRD related tumours were observed in all families, Lynch related tumours in adults were observed in only 2/14 families (p = 0.0007). All children with CMMRD had café-au-lait spots and 11/14 came from consanguineous families. Brain tumours were the most common cancers reported (48%) followed by gastrointestinal (32%) and haematological malignancies (15%). Importantly, 12 (30%) of these were low grade and resectable cancers. Tumour immunohistochemistry was 100% sensitive and specific in diagnosing mismatch repair (MMR) deficiency of the corresponding gene while microsatellite instability was neither sensitive nor specific as a diagnostic tool (p < 0.0001). Furthermore, screening of normal tissue by immunohistochemistry correlated with genetic confirmation of CMMRD. The surveillance protocol detected 39 lesions which included asymptomatic malignant gliomas and gastrointestinal carcinomas. All tumours were amenable to complete resection and all patients undergoing surveillance are alive. Discussion CMMRD is a highly penetrant syndrome where family history of cancer may not be contributory. Screening tumours and normal tissues using immunohistochemistry for abnormal expression of MMR gene products may help in diagnosis and early implementation of surveillance for these children. © 2014 Published by Elsevier Ltd. Source
Roberts A.,Queensland Institute of Medical Research |
Nancarrow D.,Queensland Institute of Medical Research |
Clendenning M.,Queensland Institute of Medical Research |
Buchanan D.D.,Queensland Institute of Medical Research |
And 33 more authors.
Familial Cancer | Year: 2011
Causative genetic variants have to date been identified for only a small proportion of familial colorectal cancer (CRC). While conditions such as Familial Adenomatous Polyposis and Lynch syndrome have well defined genetic causes, the search for variants underlying the remainder of familial CRC is plagued by genetic heterogeneity. The recent identification of families with a heritable predisposition to malignancies arising through the serrated pathway (familial serrated neoplasia or Jass syndrome) provides an opportunity to study a subset of familial CRC in which heterogeneity may be greatly reduced. A genome-wide linkage screen was performed on a large family displaying a dominantly-inherited predisposition to serrated neoplasia genotyped using the Affymetrix GeneChip Human Mapping 10 K SNP Array. Parametric and nonparametric analyses were performed and resulting regions of interest, as well as previously reported CRC susceptibility loci at 3q22, 7q31 and 9q22, were followed up by finemapping in 10 serrated neoplasia families. Genome-wide linkage analysis revealed regions of interest at 2p25.2-p25.1, 2q24.3-q37.1 and 8p21.2-q12.1. Finemapping linkage and haplotype analyses identified 2q32.2-q33.3 as the region most likely to harbour linkage, with heterogeneity logarithm of the odds (HLOD) 2.09 and nonparametric linkage (NPL) score 2.36 (P = 0.004). Five primary candidate genes (CFLAR, CASP10, CASP8, FZD7 and BMPR2) were sequenced and no segregating variants identified. There was no evidence of linkage to previously reported loci on chromosomes 3, 7 and 9. © 2010 The Author(s). Source