Article Text

Mismatch repair deficit: a gain for diagnostic histopathology
  1. Academic Department of Pathology
  2. St Mark’s Hospital
  3. Northwick Park
  4. Watford Road
  5. Harrow HA1 3UJ, UK

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

See article on page 409

Cawkwell and colleagues, in this issue (see page 409), describe an immunohistochemical method which enables the status of a colorectal carcinoma with respect to the two mismatch repair (MMR) genes, hMLH1 and hMSH2, to be ascertained from paraffin wax sections produced by any diagnostic histopathology laboratory.

Despite many advances in the pathological management of colorectal cancer since Dukes and Bussey1—for example, taking account of the number of lymph nodes containing metastatic tumour or the nature of the infiltrating margin of the tumour,2staging systems have not hitherto provided precise information about the biology and likely behaviour of the tumour in individual cases. By adopting the relatively simple methodology proposed by Cawkwell and colleagues, it should now be possible to assign patients reliably to one of the two main specific categories of MMR status, thereby identifying individuals who are likely to have the hereditary non-polyposis colon cancer (HNPCC), Lynch syndrome, genotype. Although there is a theoretical possibility that HNPCC could be due to a defect at one of the other MMR genes (e.g. PMS1, PMS2, hMSH6/GTBP, and hMSH3), all patients with HNPCC so far documented have a deficit at either MLH1 or MSH2.

This is a major advance. At a stroke, a small but significant group of patients with a relatively good prognosis but at risk of developing metachronous colorectal and other carcinomas can now be identified. It should be noted that this does not just mean the 2 or 3% of patients with colorectal cancer who are genetically HNPCC family members. Up to 13% of sporadic colorectal cancers also display a MMR deficient phenotype,3 which, as in HNPCC, is both associated with a better prognosis4 ,5 and is a marker for increased risk of development of metachronous colorectal cancers. It is also reported that patients with such “sporadic” replication error (RER) positive colorectal cancers may have, after all, a family history of colorectal cancer.4

Importantly, it has also been shown that RER positive colorectal cancers, both familial and “sporadic”, are relatively resistant to chemotherapy6 and, by implication, radiotherapy. It will now be possible to identify such patients in advance of therapy and plan treatment accordingly.

Perhaps the most valuable application of this immunohistochemical test will be in the large group of patients who, because of their family history, are worried about their mutator phenotype status. Hitherto, it has been necessary to undertake specific tests on DNA from peripheral blood or specimens of tumour, testing for microsatellite instability using relatively inaccessible and therefore more expensive PCR technology.7 ,8 This has presented practical obstacles and, in order to keep numbers manageable, it has been necessary to restrict the availability of genetic testing to those with a family history suggestive of HNPCC. For the reasons outlined earlier, it has become clear that a good case can be made for all colorectal cancers to be screened for MMR deficiency. Use of this technology will not only add power to the geneticist’s elbow, but will permit the identification of an important subgroup of patients in the everyday practice of colorectal surgery. In the case of apparently sporadic colorectal cancers, in which the average age is 63 years and there is no strong family history, MMR deficient status will tell us only about that particular patient and no further genetic testing is indicated. However, when there are other factors such as young age or family history in a patient with a tumour lacking hMLH1 or hMSH2 protein, specific genetic studies are indicated, as a means of initiating an investigation of the whole family for cancer prevention. The use of a preliminary test such as this for screening would cast the net widely for MMR defective tumours, while keeping the workload of genetic laboratories manageable.

The general adoption of this technology would also provide a rationale for a more economical use of resources for follow up of patients after surgery, by focusing on those patients at risk of developing a metachronous cancer. The authors suggest that those with RER negative tumours could be discharged from follow up after three years. This would provide a rational basis for a policy which some surgeons have been following rather pragmatically for some time.


Linked Articles