Vol. 18 • Issue 7 • Page 16
Colorectal carcinoma represents the second most common cause of cancer deaths in the U.S.1 The majority of cases of colon cancer (>85 percent) occur through the typical tumor suppressor gene pathways, often involving alterations in the adenomatous polyposis coli (APC) gene, followed by subsequent mutations in the KRAS oncogene and other important tumor suppressor genes. Although this is the most common mechanism of tumorigenesis, a second pathway through faulty DNA mismatch repair systems is also recognized and accounts for the remaining 15 percent.2-4
In the tumors with DNA mismatch repair defects, up to 30 percent are thought to be hereditary. The remaining tumors are sporadic, but still have DNA mismatch repair defects. The hereditary tumors occur as part of Lynch Syndrome, hereditary nonpolyposis colorectal carcinoma (HNPCC) or other variant syndromes, such as Turcot and Muir-Torre syndromes.5,6 In contrast to other polyposis syndromes, such as Familial adenomatous polyposis in which patients present with hundreds of colon polyps, patients with hereditary HNPCC have a finite number of adenomas. Both genetic syndromes are associated with a high risk of developing colon cancer. Patients with HNPCC have an 85 percent lifetime risk of developing colonic carcinoma.7 It is clinically important to identify patients with hereditary colon cancer, as their family members are also at risk. The patients themselves are at increased personal risk for both second colon cancers and other tumors (Table).
Guidelines have been proposed to identify patients at risk for HNPCC, including the Amsterdam criteria and more recently the Bethesda Guidelines, and the revised Bethesda Guidelines.8,9 All of these schemes indicate that younger age at presentation with colon cancer (under 50) and family history are key clinical features.
The most recent guidelines have also taken note of pathology literature that shows key histologic features associated with HNPCC (right-sided tumors, mucinous or poorly differentiated tumors, Crohn's-like inflammation and tumor-infiltrating lymphocytes).10
Testing for DNA Mismatch Repair
The enzymes involved in the DNA mismatch repair system are MLH1, MSH2, MSH6, PMS1 and PMS2. Alterations in the genes that encode for these enzymes are the usual culprits that cause DNA mismatch repair defects. These alterations can be genomic or epigenetic in nature. For example, in the hereditary conditions, germline mutations in the MSH2 or MSH6 gene are the most common findings. In sporadic tumors with DNA mismatch repair defects, methylation of MLH1 is a common finding. The end result of all of these defects is inadequate or absent enzyme production and failure of the DNA mismatch repair system. The germline mutations in the genes that encode the enzymes of the DNA mismatch repair defects can be detected through sequencing assays. These are labor-intensive and typically require genetic counseling and informed consent.
One of the main results of a faulty DNA mismatch repair system is an accumulation of replication errors in the DNA. When a cell is dividing and the DNA is replicating, errors in the base pair matching process are relatively common. Among the most susceptible sites for replication errors are mono- and dinucleotide repeat units, otherwise known as short tandem repeats (STR) or microsatellites. The replication errors in these STRs result in an abnormal number of repeat units in the daughter cells as the cell divides. In normal cells, with an intact DNA mismatch repair system, these errors are corrected. However, in patients with a defective DNA mismatch system, the enzymes do not function properly and the errors at microsatellites go uncorrected; hence, they are said to display microsatellite instability (MSI).
Assays for MSI utilize a PCR-based approach to determine the number of repeat units in the STR in the tumor, as compared to normal tissue from the same patient. The vast majority of tumors occurring in the setting of DNA mismatch repair defects will display MSI, regardless of whether the tumor is hereditary or sporadic.
Microsatellite instability testing involves PCR with primers flanking the areas of the STR being tested.11 Relatively pure samples of tumor and normal tissue can be best obtained through microdissection of tissue sections. DNA is extracted from tumor and normal tissue and the genotype for these areas are compared for a series of microsatellites. In the tumor tissue with MSI, there will be novel PCR products not present in the patient's normal tissue. The most straightforward analysis method involves capillary electrophoresis;11 SI will appear as different peaks or new amplicons in comparison to the normal DNA profile.
The first MSI assays were based on a panel of microsatellites (2 mononucleotides and 3 dinucleotides) established by an NCI-panel (BAT-25, BAT-26, D2S123, D5S346, and D17S250). Microsatellite instability is graded as either "high" or "low" based on the percentage of markers that show the characteristic patterns (>30 percent is high, and <30 percent is low). It has been shown, however, that the mononucleotide repeats are more sensitive and specific than the dinucleotide repeats for MSI-related tumors.12 Newer kits are available composed of multiplexed mononucleotides for ease of use.
Microsatellite unstable colonic carcinomas often have loss of protein expression of the affected enzyme in the DNA mismatch repair system. Loss of expression of MLH1, MSH2, MSH6 and PMS2 can be identified by using immunohistochemical stains.13 The loss of protein expression correlates fairly well with the molecular mutational findings, though some authors have suggested that the immunostain interpretation can be difficult and subjective. Key features of the immunostain profile include the fact that MSH2 loss is highly correlated with germline mutations in the gene and hereditary tumors. MLH1 loss is not so specific, since many of the cases with loss of this enzyme are sporadic and due to promoter methylation, not germline mutations.14
Future Testing Options
New information indicates that some sporadic tumors with MSI can be identified through a specific genetic profile.8,12 Around 40 percent of these tumors have a BRAF gene mutation. They also often have detectable promoter methylation of MLH1. Assays are being developed as adjunctive testing for MSI high tumors to narrow the spectrum of people who will require full sequencing of the genes to identify germline mutations. Because whole gene sequencing is labor intensive and requires consenting and genetic counseling of patients, these additional assays may become part of our standard testing algorithm.15,16
Dr. Hunt is associate professor of Pathology, Cleveland Clinic Lerner College of Medicine; section head, Surgical Pathology; and director, Head and Neck Pathology and Anatomic Molecular Pathology, Cleveland Clinic.
For a list of references, go to www.advanceweb.com/labmanager