Original article Cost analysis of biomarker testing for mismatch repair deficiency in node-positive colorectal cancer E. Barrow 1 , R. McMahon 2 , D. G. Evans 3 , E. Levine 4 and J. Hill 1 Departments of 1 General Surgery and 2 Pathology, Manchester Royal Infirmary, 3 Department of Clinical Genetics, St Mary’s Hospital, and 4 Department of Clinical Oncology, Christie Hospital, Manchester, UK Correspondence to: Miss E. Barrow, Research Office, Department of Clinical Genetics, St Mary’s Hospital, Hathersage Road, Manchester M13 9WL, UK (e-mail: emma.barrow@doctors.org.uk) Background: Microsatellite instability (MSI) in colorectal cancer is caused by defective DNA mismatch repair (MMR). It is present in 15 per cent of sporadic colorectal cancers owing to epigenetic mutL homologue 1 (MLH1) inactivation. The evidence suggests that patients with tumours caused by defective DNA MMR do not benefit from 5-fluorouracil (5-FU)-based chemotherapy. Methods: The proportion of cancers with defective DNA MMR identified by MSI analysis or immunohistochemistry was calculated from published data. The cost of analysis was compared with the potential savings if 5-FU-based chemotherapy was not administered to these patients. Results: Some 16·3 per cent of sporadic colorectal cancers had defective DNA MMR. Immunostaining for MLH1 and mutS homologue 2 (MSH2) had a sensitivity of 92·4 per cent and a specificity of 99·6 per cent for identifying MSI-high tumours. The strongest predictive variable was right-sidedness, with positive and negative predictive values of 0·329 and 0·948 respectively. If 5-FU-based chemotherapy were not administered, potential savings of up to £1·2 million per 1000 patients tested could be made. Costs would be higher if alternative chemotherapeutic regimens were substituted as a result of testing. Conclusion: Knowledge of MMR status may enable participation in trials of non-5-FU-based chemotherapy. The cost of MMR testing may be offset by more efficient use of chemotherapy. Presented to a meeting of the European Society of Coloproctology, Portomaso, Malta, September 2007, and published in abstract form as Colorectal Dis 2007; 9(Suppl 3): 33 Paper accepted 14 January 2008 Published online 6 May 2008 in Wiley InterScience (www.bjs.co.uk). DOI: 10.1002/bjs.6172 Introduction The current lifetime risk of colorectal cancer is estimated to be one in 20 1 , but the incidence of the disease continues to rise. Although reductions in mortality may reflect improved treatment, for all colorectal cancers diagnosed the age- standardized 5-year survival rate is 40 per cent 2 . Improved success in the treatment of colorectal can- cer requires a better understanding of its development and biological behaviour. Dukes’ staging and histo- logical grade correlate well with prognosis, and these criteria are currently used to guide adjuvant ther- apy. However, molecular studies have shown colorec- tal cancer to be a biologically heterogeneous disease. There is currently great interest in the identification of biomarkers as prognostic indicators and to inform decisions regarding adjuvant therapy. Research work into such markers has yet to be translated into clinical practice 3 . One of these biomarkers, DNA microsatellite instability (MSI), results from failure of repair of DNA mismatches immediately after DNA replication. A panel of five microsatellite markers has been validated. Insertion or deletion mutations at these loci cause instability of these markers. Tumours may be characterized as MSI-high (MSI-H) if two or more of the five markers show instability, MSI-low (MSI-L) if only one of the five markers shows instability, and microsatellite stable (MSS) if there are no unstable loci 4 . Hereditary non-polyposis colorectal cancer or Lynch syndrome is caused by germline mutations in the mismatch repair (MMR) genes mutL homologue 1 Copyright  2008 British Journal of Surgery Society Ltd British Journal of Surgery 2008; 95: 868–875 Published by John Wiley & Sons Ltd