Efficient detection of hereditary nonpolyposis colorectal cancer gene carriers by screening for tumor microsatellite instability before germline genetic testing

Background & Aims: The optimal strategy for the detection of hereditary non polyposis colorectal cancer (HNPCC) gene carriers remains uncertain. We eva luated whether microsatellite instability (MSI) analysis or MSH2 and MLH1 p rotein immunostaining of tumors will screen individuals efficiently for ger mline MSH2 and MLH1 testing. Methods: We performed a case-series study of 1 14 eligible families enrolled in our high-risk colorectal cancer (CRC) regi stry, Medical history data were collected on probands and relatives. MSI an alysis was performed on proband tumors, and MSH2 and MLH1 protein immunosta ining was assessed. Denaturing gradient gel electrophoresis was used to ide ntify germline MSH2 or MLH1 mutations in probands found to have tumors with high-frequency MSI, Results: Tumor tissue and adequate clinical data were available in 109 of the 114 families. Amsterdam criteria and Bethesda guide lines were met by 23% and 70% of the families, respectively. High-frequency MSI was identified in the proband tumors in 47 of the 109 families (43%). Germline MSH2 and MLH1 gene testing was carried out in the probands of 32 o f 47 families with MSI-H tumors. Mutations were detected in 16 families (9 in MSH2 and 7 in MLH1) and sequence variants of uncertain significance in 5 families (1 in MSH2 and 4 in MLH1). Germline mutations or sequence variant s of uncertain significance were detected in 15 of 19 (79%) of our Amsterda m families and in 6 of 13 (46%) of our non-Amsterdam families with MSI-H tu mors. MSH2 and MLH1 protein immunostaining was assessed in 38 of the 47 MSI -H tumors. Unequivocal loss of hMLH1 expression was found in 20 tumors and loss of MSH2 expression in 9 tumors. Corresponding loss of protein expressi on was seen in 17 of 18 (94%) of tumors from probands with germline mutatio ns or variants. Conclusions: The detection of high-frequency MSI or the los s of MSH2 or MLH1 immunostaining in CRCs are both useful criteria for selec ting high-risk patients who should be tested for germline mutations in MSH2 or MLH1.