Frequency of familial colon cancer and hereditary nonpolyposis colorectal cancer (Lynch syndrome) in a large population database

Background and aims: Estimates have been made concerning the fraction of colorectal cancer (CRC) cases that meet Amsterdam I criteria but not Amsterdam II criteria. The aim of this study was to determine in a population setting what fraction of CRC cases can be considered familial high-risk, what fraction of these meet Amsterdam I or II criteria, and what fraction of CRC cases overall meet Amsterdam I and II criteria. Methods: The Utah Population Data Base (UPDB), which links Utah genealogies to the Utah Cancer Registry, was used to examine the aims of the study. Familial high-risk was operationally defined as CRC occurring at an age < 50 years or as a part of a first-degree relative pair. A subset of Amsterdam positive cancers was tested for microsatellite instability (MSI) to determine what fraction of Amsterdam families was likely to have hereditary nonpolyposis colorectal cancer (HNPCC). Results: Of the 6,628 CRC cases in the UPDB, 24.5% met the criteria for familial high-risk. Of these, 2.6% met Amsterdam I criteria and 5.5% Amsterdam II. Of total data base CRC cases, 0.8% met Amsterdam I criteria and 2.3% Amsterdam II. In a subset of colon tumors from Amsterdam families, 70% were MSI stable. Conclusions: Although nearly 25% of CRC cases in our population data base met a simple definition of familial high-risk, only a small fraction of these and a smaller fraction of total CRC cases met Amsterdam I or II criteria. Less than half of a limited set of tumors from Amsterdam families were MSI positive.