Menko and colleagues review the genetic counseling that occurs
after the presymptomatic diagnosis of hereditary nonpolyposis
colorectal cancer (HNPCC) is made. The discovery of the mechanism
by which tumors develop in this setting and the cloning of the
genes responsible for the disease have made possible the DNA-based
diagnosis of a disease, which, less than 5 years ago, was a controversial
entity. In a remarkably short period, the nature of HNPCC has
been greatly elucidated, its relationship to defective DNA mismatch
repair has become manifest, and the means to diagnose it in a
presymptomatic state has been developed .
The first generation of diagnostic tests are now available, (Luce,
et al, 1995) and no doubt, more sensitive and specific tests are
on the way. Before long, diagnostic tests will become commercially
available, and we will have to grapple with a number of possibly
unforeseen issues. Physicians will be in a position to
inform healthy adults with no premorbid phenotype that they are
at extraordinary risk of developing cancer during their lifetime,
and that this will tend to occur two to three decades earlier
than the rest of the population. Although physicians have been
counseling patients about their risk of neoplasia for many decades,
they may be armed with a degree of certainty that was never previously
possible. Will we be ready for this?
An Important Step Forward, But...
The presymptomatic diagnosis of certain diseases, eg, Huntington's
disease, poses a formidable challenge, in that no effective therapy
is available to prevent their morbid and lethal consequences.
In the case of HNPCC, on the other hand, effective intervention
is available that can dramatically reduce its adverse consequences.
Accurate diagnosis within high-risk families can relieve some
individuals of the need for lifelong diagnostic tests, including
frequent colonoscopy, and can help convince affected individuals
to accept proper management even when they are feeling well.
DNA-based diagnosis should, therefore, be considered an important
step forward in the management of familial colon cancer. Certainly,
there will be great enthusiasm to include these new approaches
in our diagnostic armamentarium. However, we should consider the
possibility that risks may accompany the ability to make this
diagnosis. Can such knowledge have side effects?
The paper by Menko et al makes a number of important points. First,
the diagnosis of HNPCC can now be made in smaller families in
which the Amsterdam criteria would not likely be met. New mutations
can be diagnosed and interpreted properly. In such instances,
HNPCC has important implications for an individual and his or
her offspring, but the risk does not extend to siblings or parents.
Once a genetic diagnosis is confidently made, colonoscopic screening
and appropriate gynecologic screening may be initiated, as outlined
in the authors' Table 5. Moreover, psychological intervention
and support may be started in the presymptomatic period, permitting
the affected individual to deal with risk and familial issues
independently of problems associated with personal health risks.
This information will surely empower the properly informed patient
with the means to reduce the risks of developing metastatic cancer.
Complexities of DNA-Based Diagnosis May Not Be Readily Apparent
However, the ability to make DNA-based diagnoses is fraught with
complexities that may not be immediately apparent to the primary-care
physician, surgeon, gastroenterologist, or oncologist. Numerous
ambiguities may cloud the interpretation of a genetic diagnosis.
Each of the different genetic approaches to the diagnosis of HNPCC
has its own inherent strengths and liabilities, as the authors
indicate. Every approach can miss the offending mutation, which
results in a false-negative diagnosis.
Also, it is likely that we have not yet uncovered all of the genes
that can produce HNPCC. The DNA mismatch repair system requires
many proteins, of which four have been linked to HNPCC and several
others have not. Furthermore, at least one of these genes (hPMS-1)
has been linked to only one HNPCC family, and due to the nature
of this mutation, it is possible that it does not actually cause
the HNPCC. On the other side of the equation, it would appear
that other members of the DNA mismatch repair system should be
capable of causing HNPCC, but no families have yet been discovered
with germ-line mutations in these genes. It will take time to
establish with certainty all of the genes that may result in HNPCC.
Some diagnostic approaches will always miss certain classes of
mutations. For example, direct sequencing of genes (the most rigorous
but labor-intensive approach to the diagnosis), will miss mutations
that affect gene expression, may miss alterations at intron-exon
borders that can adversely affect splicing, and will miss large
genetic deletions. Screening techniques, such as DGGE and SSCP,
are intrinsically less sensitive than direct sequencing and are
more likely to result in underdiagnosis. The protein truncation
test (also called in vitro transcription/translation, or IVTT)
detects only mutations that result in the translation of a pathologically
foreshortened protein that is missing its carboxy-terminus .
False-Positive Diagnoses Also a Risk
False-positive diagnoses are an equal risk. As mentioned above,
direct sequencing is likely to be the most sensitive approach
for the detection of germ-line mutations in the DNA repair genes.
However, interpretation of the sequences may not be simple. For
example, certain variations in DNA sequence will be silent and
not result in any change in the protein structure. Oth er variations
may change the amino acid sequence of the protein (ie, a missense
mutation), but this alteration may be "conservative"
and not seriously affect its function. Therefore, the more sensitive
direct-sequencing approach may detect a "polymorphism"
that represents a deviation from the most common human sequences
but still may not cause disease.
Investigators familiar with DNA sequencing and molecular biology
may recognize these sequence variations as polymorphisms, but
this may not be obvious to the uninitiated. Worse yet, mutations
may come in a variety of "strengths," some of which
are more powerful than others and may cause a more severe phenotype.
The game is still young, and we are still in the process of trying
to understand the implications of different types of mutations.
If clinicians unfamiliar with the principles of molecular biology
can obtain test results without assistance, errors can be made,
and improper counseling may be offered.
The authors also raise the issue of indirect gene tracking, such
as linkage analysis or other techniques that study the segregation
of DNA polymorphisms linked to a disease-causing gene in a family.
This approach may be potentially hazardous as well, depending
on the propinquity of the marker locus to the disease-causing
gene. This approach is usually used when a gene has been mapped
but has not yet been cloned. Linkage analysis should be used with
HNPCC More Complicated Than Initially Thought
The more we learn about HNPCC, the clearer it becomes that the
disease is more complex than initially thought. It was traditionally
assumed that every gene carrier would develop cancer at an early
age. It now appears that variable penetrance occurs in this disease.
Skipped generations may occur [3,4], and some individuals may
develop cancer later in life, without any explanation .
Examples have now been reported in which HNPCC was diagnosed
in a parent after it had been diagnosed in his or her child .
Moreover, Liu et al  found HNPCC in several individuals who
were selected for study after the exclusion of a family
history of HNPCC.
It is not known whether the variable penetrance is related to
different "strengths" of the germ-line mutation, the
influence of environmental factors, or even the modulating influence
of another gene, as appears to be the case in familial adenomatous
polyposis . We must be prepared for the accumulation of perplexing
details about this disease.
Management Still in Evolution
The management of patients with a DNA-based diagnosis of HNPCC
is still in evolution. Interestingly, Menko et al suggest that
colonoscopy should be carried out every 2 to 3 years in "at-risk"
individuals, as opposed to every 1 to 2 years in known gene carriers.
The authors are attempting to cope with the utilization of resources
for patients who are at 50% risk of the disease (ie, asymptomatic
offspring of HNPCC patients for whom a genetic diagnosis has not
been possible) vs a more intense regimen than those who are known
to have HNPCC. Is this entirely rational? If 1- to 2-year colonoscopic
intervals are appropriate for patients known to carry a gene for
HNPCC, perhaps this approach should be used in patients at 50%
risk until the diagnosis can be definitively excluded.
It will be very difficult to exclude the diagnosis of HNPCC until
the affected gene is identified in a family. Thus, knowing more
about the disease, but being unable to definitively rule it out
in some patients, may increase the physician's anxieties about
the disease and actually increase the use of preventive measures
in such families. Finally, the authors stop short of recommending
the most ambitious form of management, prophylactic colectomy
(or perhaps subtotal colectomy). Although surgery would surely
be rejected by many patients, others may prefer a surgical approach
to a lifetime of colonoscopy. Time will tell.
Psychological and Social Consequences: A Pandora's Box?
Finally, the authors explore the important psychological and social
consequences of a genetic diagnosis. They appropriately point
out that the information provided by a definitive diagnosis of
HNPCC frequently improves a person's outlook on life, whether
the result proves that the individual carries such a mutated gene
or rules out this possibility. A definitive answer is often gratifying,
since individuals can make plans for coping with their circumstance,
and 50% of those tested will experience relief from their anxieties.
It comes as a surprise to many individuals who are not involved
in genetic counseling that not all patients who are informed that
they carry a diseased gene react adversely. Conversely, some patients
who are informed that they do not carry such a gene feel
Moreover, there are important social consequences of genetic counseling
that go beyond the personal psychological counseling. Will this
information remain confidential? Will a gene carrier find that
he or she is no longer considered employable or find it more difficult
to obtain health or life insurance? Will there be aberrations
in the social perception of individuals who carry mutated DNA
mismatch repair genes, even if effective management can offer
a reasonably good likelihood of leading a long and healthy life?
At this point in time, a Pandora's box has been opened, and it
remains to be seen how full it is.
1. Marra G and Boland CR: Hereditary nonpolyposis colorectal cancer
(HNPCC): The syndrome, the genes, and an historical perspective.
J Natl Cancer Inst 87:1114-1125, 1995.
2. Luce MC, Marra G, Chauban DP, et al: In vitro transcription/translation
assay for the screening of hMLH1 and hMSH2 mutations in familial
colon cancer. Gastroenterology 109:1368-1374, 1995.
3. Mecklin JP, Jarvinen HJ, and Peltokallio P: Cancer family syndrome:
Genetic analysis of 22 Finnish kindreds. Gastroenterology 90:328:333,
4. Vasen HFA, Taal BG, Griffioen G, et al: Clinical heterogeneity
of familial colorectal cancer and its influence on screening protocols.
Gut 35:1262-1266, 1994.
5. Menko FH, Te Meerman GJ, Sampson JR: Variable age of onset
in hereditary nonpolyposis colorectal cancer. Clinical implications.
Gastroenterology 104:946-947, 1993.
6. Liu B, Farrington SM, Petersen GM, et al: Genetic instability
occurs in the majority of young patients with colorectal cancer.
Nature Medicine 1:348-353, 1995.
7. MacPhee M, Chepenik KP, Liddell RA, et al: The secretory phospholipase
A2 gene is a candidate for Mom1 locus, a major modifier of ApcMin-induced
intestinal neoplasia. Cell 81:957-966, 1995.