Colon cancer will affect approximately 140,000 persons and kill an estimated 55,000 people in 1997, making colon cancer the second deadliest cancer in the United States. Hereditary nonpoly-posis colon cancer (HNPCC) accounts for approximately 3% to 10% of colon cancers.
This autosomal dominant condition is characterized by an earlier age of onset of colon cancer (in the 40s versus about age 65 for sporadic colon cancer), a tendency for multiple tumors to form either simultaneously or over time, and a tendency for tumors to occur proximal to the splenic flexure.
The pathology of the tumors is more likely to be mucinous and poorly differentiated, but in contrast to this histologic grade, which would connote a more aggressive tumor, HNPCC colon cancers have a better clinical outcome than sporadic tumors matched for stage. [This will be discussed by Dr. Risto Sankila of the Finnish Cancer Registry in the next article in this series.]
In addition, there is an increased incidence of certain noncolonic tumors in some HNPCC families, including those of the female reproductive tract, stomach, and urinary tract.
Criteria for Studies of HNPCC
In 1991 in Amsterdam, the International Collaborative Group on HNPCC developed uniform criteria for collaborative studies on HNPCC. (These criteria were not designed for the diagnosis of HNPCC.) The requirements are:
- The family includes three or more relatives with histologically verified colorectal cancer, one of whom is a first-degree relative of the other two, and familial adenomatous polyposis has been excluded.
- The colorectal cancers involve at least two successive generations.
- At least one of the colorectal cancers has been diagnosed before the age of 50 years.
However, these criteria are restrictive in that they exclude patients with extra-colonic tumors and late-onset variants, thus overlooking some HNPCC families by this definition.
The heritable nature of HNPCC (and its variant, the Muir-Torre syndrome) is caused by germline mutations in one of the several genes involved in the DNA mismatch repair (MMR) system, an editing mechanism for polymerase errors that occur during DNA replication.
The MMR genes are so named for their gene products' ability to recognize and direct repair of nucleotide mispairs and misalignment at short repetitive se-quences of DNA (called microsatellites) whose length was not accurately copied during DNA replication. Repair by the MMR system occurs on the newly synthesized DNA strand, as documented by in vitro repair of mispairs on the strand containing nicks.[5,6]
Base mispairing can lead to nucleo-tide transitions or transversions, altering the authentic genetic sequence. If the mismatch occurs in the coding region for a particular gene, the newly introduced point mutation may affect the expression and/or function of that particular gene.
Perhaps more importantly, some of the approximately 100,000 DNA micro-satellite sequences scattered throughout the human genome may be altered in length when this system is defective. A microsatellite may become lengthened in a daughter cell if there is nucleotide-pairing slippage (looping) along the newly synthesized strand during DNA synthesis, or it may become shortened if the template strand microsatellite has slippage during DNA replication.
This alteration in microsatellite length is termed microsatellite instability (MIN)[7-9] and can be identified by electrophoretic resolution of amplified microsatellite DNA sequences. MIN is relatively easy to detect in the laboratory, and when found in the DNA of a tumor, it indicates the presence of a hypermutable phenotype.
Occasionally, microsatellites are present in the coding region (exons) of critical growth regulatory genes. This has been best demonstrated with the transforming growth factor-beta type II receptor (TGF-beta RII); when this receptor is bound by TGF-beta in the colonic epithelium, cellular proliferation is inhibited.
Mutation of TGF-beta RII occurs with defective MMR, commonly resulting in length changes in the polyadenine mononucleotide repeat microsatellite (A10) within the gene, producing a frameshift mutation and rendering the receptor inactive.[10,11] This mutation removes the growth brake provided by TGF-beta, allowing the colon cells to undergo clonal expansion.
The evolutionarily conserved genes that comprise the MMR system and that have been found mutated in HNPCC families include hMSH2[12,13], hMLH1[14,15], hPMS1, and hPMS2. Although other components of the MMR system have been identified, germline mutations of these components have yet to be found in HNPCC families.
When a physical deformation remains in the newly replicated DNA double helix (caused by the mispairing of nucleo-tides or by slippage and looping at microsatellite loci), a complex termed hMutS-alpha (a heterodimer of hMSH2 and hMSH6 proteins) identifies the error and binds the DNA at this site.[17-20]
Subsequently, hMutS-alpha recruits the hMutL-alpha complex, a heterodimer of hMLH1 and hPMS2 proteins, which targets the newly synthesized daughter DNA strand for "long patch" excision repair.
It remains unclear how hPMS1 interacts within the complex. Loss of any of the four components of the MMR system will inactivate or attenuate repair. Germline mutations in the hMLH1 and hMSH2 genes account for the majority (approximately 90%) of HNPCC families identified to date.
One wild-type allele of an MMR gene is generally sufficient to maintain normal MMR function. For colon cancer to develop in HNPCC patients, a second somatic event (in addition to the vertically transmitted mutant allele) must occur in the wild-type allele of a colonocyte. This completely inactivates both of the MMR genes and causes the hypermutable phenotype seen with HNPCC tumors.
Methods of Diagnosing HNPCC
Identification of genes involved in HNPCC has prompted efforts to diagnose this condition in presymptomatic patients (see table). There is no premor-bid clinical phenotype that identifies an HNPCC patient antecedent to the development of cancer.
Nonneoplastic cells in HNPCC patients have a normal phenotype, since the inactivation of only one allele of an MMR gene is not permissive of hypermutability. The presence of inactivating mutations in both alleles of MMR genes in tumors confirms Knudson's "two-hit" hypothesis for tumor-suppressor genes.
Although MIN should be a necessary finding for this condition, in fact, only 92% of HNPCC tumors show MIN. Furthermore, it has been demonstrated that only a minority of tumors with MIN actually come from HNPCC families.[26,27]
Adenomas in HNPCC patients often manifest MIN,[28,29] indicating that inactivation of the second allele of an MMR gene occurs as an early event in colorectal tumorigenesis. Unfortunately, the finding of MIN in a tumor is not perfectly sensitive and is quite nonspecific, so it is not a practical screening strategy for HNPCC.
Obtaining peripheral blood is the least invasive method to diagnose HNPCC. In HNPCC, the lymphocytes will have one mutated allele and one wild-type allele in one of the above-mentioned MMR genes.
Direct sequencing of the four MMR genes involved in HNPCC is a possible diagnostic strategy, but the number of genes involved and the number of exons in each gene make this clinically impractical at this time.
These problems extend to common mutation detection methods like single-stranded conformational polymorphism (SSCP) and denaturing gradient gel electrophoresis (DGGE).
Nevertheless, direct DNA sequencing of hMLH1 and hMSH2, the two most common genes involved, is the most reliable approach to diagnosis at this time. Interpretation of the sequence data, however, may prove difficult, since the full spectrum of disease-producing mutations and innocuous polymorphisms has not been catalogued.
The reported mutations in hMLH1 and hMSH2 include insertions, deletions, nonsense mutations, and some missense mutations.[21,25] Many variations in the DNA sequences of these genes are not associated with an increased risk for tumor formation. A large fraction of these mutations yield a truncated protein after translation, more so with hMSH2 than hMLH1.
This fact has led to the application of an in vitro transcription/translation assay (IVTT) for these genes. With the IVTT approach, RNA is extracted from the peripheral blood, reverse transcribed into complementary DNA, and expressed as protein.
The altered migration of a truncated protein on a polyacrylamide gel can provide a means of identifying which MMR gene contains an inactivating mutation. A study assaying hMLH1 and hMSH2 by the IVTT method was about 50% sensitive in patients who fulfilled the Amster-dam criteria for HNPCC.
Linkage analysis can identify current or future family members as carriers of a mutated MMR gene (and thus distinguishes the carrier as an HNPCC patient).
The mutated allele may be identified from affected family members (and excluded from unaffected family members) using microsatellites or restriction fragment length polymorphisms (RFLPs) that are linked to the MMR gene. This approach, however, requires obtaining blood from more than one generation of first-degree relatives, including the affected and unaffected members.
The MAMA Technique
Another novel but time-consuming approach for certain mutations missed by conventional techniques is the isolation of each allele from peripheral blood lymphocytes into a somatic cell hybrid, termed MAMA (monoallelic mutational analysis).
The MAMA method fuses individual chromosomes from human cells from a patient with an appropriate hamster cell line to individually analyze each allele for mutations that might be masked by the wild-type allele. However, the time required to perform this assay makes its use impractical outside of the research setting.
At this time, the use of the IVTT assay combined with direct DNA sequencing may be the most appropriate approach for finding germline mutations in HNPCC.
These tests can be applied to families who fit the Amsterdam criteria, as well as to the young patient (ie, less that 45 years of age) who develops colon cancer that has characteristics typical of HNPCC tumors (ie, right-sided, mucinous, poorly differentiated adenocarcinomas).
When available, neoplastic tissue, including paraffin(Drug information on paraffin)-embedded archival blocks, can be valuable for MIN analysis, since patients with HNPCC will usually have microsatellite alterations at multiple loci. Supporting information at the time of microsatellite analysis includes loss of heterozygosity of an MMR gene, which can be identified using appropriate microsatellite targets.
The aim of identifying the germline lesion in the patient is for screening remaining family members. There is no substitute for assessing the family history, including the ages of affected family members and the occurrence of extra-colonic tumors.
In clinically defined HNPCC families in which a mutation cannot be identified by current genetic techniques, or in family members who do not wish to be genetically tested, a regular screening program consisting of colonoscopy or barium enema and sigmoidoscopy significantly reduces the rate of tumor development and death.[32,33]