Abstract

Approximately 80 per cent of patients with colorectal cancer have sporadic disease whereas the remaining 20 per cent seem to have a genetic component. Hereditary nonpolyposis colorectal cancer (HNPCC) is the most common autosomal dominant hereditary syndrome predisposing to colorectal cancer. Various methods have been described to screen for HNPCC and to directly test for mismatch repair gene mutations. This study evaluates the initial results of 1) microsatellite instability (MSI) and immunohistochemistry (IHC) staining of tumors and 2) genetic sequencing for mismatch repair gene mutations in patients suspected to have HNPCC. Appropriate patients for HNPCC testing were identified through a high-risk colorectal cancer clinic. Of those patients screened only those who met Amsterdam criteria (AC) for HNPCC or were young age onset (YAO) (<40 years of age) were eligible for testing. The tumors underwent testing for MSI and had IHC performed in those patients with available tumor specimens. MSI was performed on the five markers approved by the NIH consensus conference. MSI-High (MSI-H) was defined as two or more markers being unstable. IHC was done with commercially available stains for MLH1 and MSH2. All patients had sequencing of the MLH1 and MSH2 genes performed to search for mutations by a commercial laboratory. Genetic counseling was provided and written informed consent was obtained. Fourteen patients were part of kindreds that met the AC. An additional 10 patients were <40 years of age at diagnosis of colorectal cancer but lacked any family history. Testing for MSI and IHC was performed on those available tissue blocks. Of the AC patients five had MSH2 mutations and two had MLH1 variants. Of the five with MSH2 mutations three of four had MSI-H tumors and all four had loss of expression of MSH2 on IHC. Of the MLH1 variants only one had MSI-H tumor and lacked expression of MLH-1 on IHC. Of those patients with no mutation identified three of six had MSI-H tumors. For those patients YAO no genetic mutations were identified. Two of the seven had MSI-H tumors. Genetic testing for HNPCC even in those patients fulfilling the rigid AC yielded mutations in only five of 14 patients with variants of unknown significance being found in an additional two patients. Only one MSH2 variant of unknown significance was identified in the 10 YAO patients, which would suggest that screening in this group of patients with MSI and/or IHC would be appropriate.