Sensitivity and specificity of clinical criteria for hereditary non-polyposis colorectal cancer associated mutations in MSH2 and MLH1

Background and aims-There are multiple criteria for the clinical diagnosis of hereditary non-polyposis colorectal cancer (HNPCC). The value of several of the newer proposed diagnostic criteria in identifying subjects with mut ations in HNPCC associated mismatch repair genes has not been evaluated, an d the performance of the different criteria have not been formally compared with one another. Methods-We classified 70 families with suspected hereditary colorectal canc er (excluding familial adenomatous polygosis) by several existing clinical criteria for HNPCC, including the Amsterdam criteria, the Modified Amsterda m criteria, the Amsterdam II criteria, and the Bethesda criteria. The resul ts of analysis of the mismatch repair genes MSH2 and MLH1 by full gene sequ encing were available for a proband with colorectal neoplasia in each famil y. The sensitivity and specificity of each of the clinical criteria for the presence of MSH2 and MLH1 mutations were calculated. Results-Of the 70 families, 28 families fulfilled the Amsterdam criteria, 3 9 fulfilled the Modified Amsterdam Criteria, 34 fulfilled the Amsterdam II criteria, and 56 fulfilled at least one of the seven Bethesda Guidelines fo r the identification of HNPCC patients. The sensitivity and specificity of the Amsterdam criteria were 61% (95% CI 43-79) and 67% (95% CI 50-85). The sensitivity of the Modified Amsterdam and Amsterdam II criteria were 72% (9 5% CI 58-86) and 78% (95% CI 64-92), respectively. Overall, the most sensit ive criteria for identifying families with pathogenic mutations were the Be thesda criteria, with a sensitivity of 94% (95% CI 88-100); the specificity of these criteria was 25% (95% CI 14-36). Use of the first three criteria of the Bethesda guidelines only was associated with a sensitivity of 94% an d a specificity of 49% (95% CI 34-64). Conclusions-The Amsterdam criteria for HNPCC are neither sufficiently sensi tive nor specific for use as a sole criterion for determining which familie s should undergo testing for MSN2 and MLH1 mutations. The Modified Amsterda m and the Amsterdam II criteria increase sensitivity, but still miss many f amilies with mutations. The most sensitive clinical criteria for identifyin g subjects with pathogenic MSN2 and MLH1 mutations were the Bethesda Guidel ines; a streamlined version of the Bethesda Guidelines may be more specific and easier to use in clinical practice.