KRAS mutation testing of metastatic colorectal cancer in Australia: where are we at?

AIM: To carry out a nationwide study of KRAS testing in metastatic colorectal cancer as reported by nine major molecular pathology service providers in Australia, including mutation frequencies and turnaround times that might impact on patient care. METHODS: Participating laboratories contributed information on KRAS mutation frequencies, including the G13D mutation type, as well as turnaround times for tumor block retrieval and testing. RESULTS: The KRAS mutation frequency observed by nine different test sites for a total of 3688 metastatic colorectal cancers ranged from 34.4% to 40.7%, with an average across all sites of 38.8%. The average frequency of the G13D mutation type among all cases was 8.0%. The median turnaround time was 17 days (range 0-191), with 20% of cases requiring more than 4 weeks for a KRAS test result. The major contributor to long turnaround times was the time taken to retrieve archived blocks of primary tumor, particularly from sources external to the test site. CONCLUSION: The frequency of KRAS mutations in metastatic colorectal cancer reported by the major Australian test sites is very similar to that reported by other large overseas studies. More widespread introduction of routine testing at the time of initial diagnosis should eliminate the long turnaround times currently being experienced in a significant proportion of cases. Future expansion of testing to include other KRAS and NRAS mutation hotspots may spur the introduction of next-generation sequencing platforms.

Correlation of mismatch repair genes immunohistochemistry and microsatellite instability status in HNPCC-associated tumours.

AIM: The aim of this study was to assess the performance of immunohistochemistry using antibodies for MLH1, MSH2, MSH6 and PMS2 mismatch repair gene proteins against microsatellite instability (MSI) testing. METHODS: Tumour samples included in this study were derived from referred patients for screening for hereditary non-polyposis colorectal cancer (HNPCC) and patients who had resections for colorectal cancer that were examined at our institution. MSI was assessed at nine loci (BAT25, BAT26, BAT40, D2S123, D10S197, D17S579, D18S34, D5S346 and D17S250) in all cases. Immunohistochemistry for MLH1 and MSH2 was performed in all cases. Staining for MSH6 and PMS2 was performed in selected cases only. RESULTS: There were 742 tumours including 661 colorectal lesions and 81 extracolonic tumours of the HNPCC spectrum. Among the 555 MSI-negative tumours, 554 showed an intact protein expression. Amongst the 187 MSI-positive tumours, 126 showed abnormal expression of MLH1 gene protein, 41 showed abnormal expression of MSH2 gene, three showed abnormal expression of MSH6 only, one showed abnormal expression of PMS2 gene protein only and one case showed abnormal expression of all four proteins. CONCLUSION: Immunohistochemistry offers an alternative method for assessment of MSI status which is fast and relatively inexpensive compared with MSI testing. We achieved a sensitivity rate of 92% and specificity of 99.8% for immunohistochemistry testing assessed against the MSI testing. It has to be accepted that a small fraction of MSI-positive cases will be missed by testing with immunohistochemistry alone.