REVIEW

Genetic testing in colorectal cancer: who, when, how and why

Nicholas O. Davidson

Division  of  Gastroenterology,  Washington  University  School  of  Medicine,  

St.  Louis,  MO,  U.S.A.
(Received for publication on November 8, 2006)
(Revised for publication on January 23, 2007)
(Accepted for publication on February 15, 2007)

Abstract. Colorectal cancer (CRC) is among the most prevalent and preventable forms of
cancer worldwide, accounting for over 600,000 deaths in 2005. Both genetic and environ-
mental factors contribute to cancer etiology and estimates suggest that at least one third of
CRC has a familial component. There is increased awareness of a strong genetic component
to CRC risk, with the identification of several high penetrance alleles that predict increased
CRC susceptibility. These include familial adenomatous polyposis (FAP), linked to mutations
or deletions of the APC tumor suppressor gene, as well as Lynch syndrome (formerly known
as hereditary non-polyposis colorectal cancer or HNPCC), which is linked to mutations or
deletions of one or more mismatch repair genes including MLH1, MSH2 and MSH6. In addi-
tion, mutations in genes encoding key signaling molecules have been linked to autosomal
dominant hamartomatous syndromes that are associated with increased susceptibility to
CRC. These include Peutz-Jeghers syndrome, which is linked to mutations in STK11/LKB and
Juvenile polyposis, which is linked to mutations in the genes encoding SMAD4 and BMPR1A.
In addition to these high penetrance autosomal dominant alleles, recessive mutations in the
MYH mismatch repair gene are associated with a phenotype similar to FAP. With the wide-
spread availability of genetic testing for these alleles, physicians will be faced with a com-
plex array of choices in terms of advocating who should be tested, when should such testing
take place, how it should be conducted and interpreted and why it changes the management
and outcomes for the patient and his or her family. (Keio  J  Med  56  (1)  :  14－ 20,  March  2007)

Key words: tumor suppressor gene, hereditary mutations, high penetrance alleles, screening

Introduction and important new advances that support this suggestion

and will outline some of the challenges in implementa-
Colorectal cancer (CRC) is the second leading cause tion of this paradigm in practice.
of cancer related death in the United States and inflicts a
major toll worldwide.1 Advances in elucidating the ge- Who should be tested?
netic basis for many common polygenic disorders - which
have emerged coincident with completion of the human Genetic testing offers the opportunity to confirm a
genome sequencing project - have heightened awareness suspected diagnosis of several forms of hereditary CRC.
that many common cancers also have strong hereditary The goal of this review is to outline specific guiding
or familial components. Indeed, a recent review2 has sug- principles for each of the hereditary CRC syndromes
gested that up to one third of patients with CRC have an for which genetic mutations have been defined and for
identifiable familial component. Collectively then, these which genetic screening is available [for detailed re-
advances suggest that our view of CRC should be modi- views, see.3-5 As a prelude to such discussion it is worth
fied to reflect the consensus that it is a genetic disease. considering some general themes in considering who
This article will attempt to summarize the background should be offered genetic testing. Genetic testing can

Reprint requests to: Nicholas O. Davidson, M.D., Division of Gastroenterology, Box 8124, Washington University School of Medicine, 660 South Euclid
Avenue, St. Louis, MO 63110, U.S.A., E-mail: [email protected]

14
Keio J Med 2007; 56 (1): 14－ 20 15

Table 1 Proposed categories of hereditary colorectal cancer syndromes for which genetic testing is possible

Polyposis Syndromes Non-polyposis Syndromes

Adenomatous Polyposis Hyperplastic Polyposis Hamartomatous Polyposis Lynch Syndrome Familial Non-polyposis
CRC

Classical FAP - Autosomal dominant ●K-RAS, B-RAF Peutz-Jeghers MMR Gene Mutations Undefined genetic basis
Mutations
● APC Mutations ● STK11/LKB Mutations. ● MLH1 Mutations
Autosomal recessive ● MSH2 Mutations
● MYH Mutation ● MSH6 Mutations

Juvenile Polyposis ● PMS2 Mutations

● SMAD4/BMPR1A

Mutations
Attenuated FAP - Autosomal
dominant
● APC (I1307K)

● Autosomal Cowden Syndrome

Recessive
● MYH Mutation ● PTEN Mutations

be offered to “at risk” family members in order to clas- non-polyposis forms of hereditary CRC, specifically
sify carrier status for known alleles and to assign a likely Lynch syndrome (Table 1). The detailed application of
risk of disease. There are several categories of patients these genetic tests, as well as their interpretation and
who clearly merit such genetic testing for CRC. The first usefulness, differs, however, depending on the syndrome
group consists of patients suspected of syndromic risk under consideration and the mutant alleles involved. For
based on clinical findings.3, 5 This would include an early the purposes of the present discussion, this review will
age of onset (<40 years), the presence of multiple (>10) focus predominantly on testing for familial adenomatous
polyps, synchronous or metachronous cancers (either polyposis and hereditary non-polyposis CRC. Readers
CRC and/or associated cancers) and/or a family history are referred to excellent reviews for expanded discussion
positive for any of these features. A second group con- of Juvenile polyposis and other rare hamartomatous syn-
sists of patients within a family carrying a clinical diag- dromes (for example see5).
nosis of hereditary CRC, but for whom the pathogenic
mutation is unknown. In this instance, identification of the Familial Adenomotous Polyposis
mutant allele in the proband could be useful in screening
other members of the family in order to exclude carriers. In familial adenomatous polyposis (FAP), the clas-
A third group would be patients within a family known sical phenotype involves development of hundreds to
to have a genetic basis for CRC and in which the mutant thousands of adenomatous polyps carpeting the colonic
allele is known. In this instance, genetic testing of other mucosa. This phenotype becomes manifest typically by
family members would be of utility in excluding carri- age 35 years and demonstrates an autosomal dominant
ers and in counseling affected members. Clinical genetic inheritance pattern with high penetrance.2 FAP is found
testing of patients who manifest these characteristics will in ~1 per 7-10,000 births in the United States population
be discussed further below under the relevant disease and accounts for less than 1% of all CRC. In its classi-
category cal form, almost 100% of these adenomatous polyps are
considered to be cancer prone. FAP is typically a mono-
Genetic Testing for Defined Hereditary CRC syn- genic disease, the result of mutations or deletions of the
dromes: What can we test for and how do we decide adenomatous polyposis coli (APC) gene. In its classical
who should get tested? form, FAP is relatively easy to recognize and genetic
testing is important since its autosomal dominant inheri-
Screening for carrier status in hereditary CRC syn- tance pattern with high penetrance together imply a high
dromes can be offered for several genetic mutations, in- likelihood of vertical transmission. Accordingly the fol-
cluding familial adenomatous polyposis (FAP) and ham- lowing groups of individuals should be screened. First,
artomatous polyposis syndromes including Peutz-Jeghers individuals with some but not all the features of classical
syndrome and Juvenile polyposis syndrome (Table 1). In FAP should be tested. This would include individuals
addition, genetic testing can be offered to patients with with more than 10 but less than 100 polyps. Secondly,
16 Davidson NO: Genetic testing in colorectal cancer

individuals should be tested who manifest clinically MYH mutations, particularly those in whom APC testing
defined FAP (ie adenomatous colonic polyposis), but is negative. This is important since there appear to be
where the mutation has not been defined within the fam- both low penetrance as well as high penetrance MYH al-
ily. This would also include individuals with no family leles, suggesting that yet to be understood genetic and/or
history of FAP. However, in this regard it is important environmental modifiers may play a role in modulating
to bear in mind that up to 25% of cases of FAP arise the attenuated phenotype.2 Finally, clinicians should be
as spontaneous APC mutations and these patients will aware that the presence of multiple (>20) colonic polyps
therefore have no family history of the disease.5 Thirdly, is consistent with a diagnosis of hyperplastic polyposis
relatives within an FAP cohort family should be tested syndrome, rather than AFAP. In this situation, biopsy of
once the founder mutation is known, since genetic test- the polyps will reveal hyperplastic rather than adenoma-
ing will be informative in predicting or excluding carrier tous features.1 6 Hyperplastic polyposis is an important
status. This is a crucial issue since a negative genetic test entity to recognize since its genetic basis and molecular
result for a family member who has a proband with an pathogenesis are distinct from that of FAP but fam-
identified mutation in the APC gene means that the fam- ily members are at increased risk of developing CRC.17
ily member (who tests negative) is at no greater risk for There is an emerging literature pointing to the impor-
colon cancer than the general population, and the impact tance of the hyperplastic polyposis pathway in a subset
of genetic screening clearly has a major effect on the pa- of familial CRC. In this alternative pathway, the serrated
tients’ perceptions and needs for future screening.6 polyp represents the precursor lesion rather than the
There are several important issues to keep in mind adenomatous polyp.1 8 Individuals and families with hy-
when considering genetic testing for FAP. It is crucial perplastic polyposis and CRC tend to present later and to
that patients and family members fully understand the demonstrate BRAF mutations along with microsatellite
implications of genetic testing, including the value of instability as a result of hMLH1 promoter methylation
both positive and negative results. Even with the best (see below).18 This alternate pathway has been described
technology, disease-causing mutations in the APC gene in both European and North American populations
are detectable in ~85% patients with phenotypic classical and up to 16% of adenocarcinomas of the cecum and
FAP. In addition, as noted above, while the typical in- ascending colon may reveal a serrated phenotype.1 9, 2 0
heritance pattern is autosomal dominant, approximately Although much remains to be understood concerning the
25% of cases arise as spontaneous new mutations and molecular pathways that influence the progression of the
these patients will not have an immediate relative with serrated polyp, it is important for clinicians to recognize
colonic polyposis.3, 5 In these cases, where the proband this distinctive manifestation of CRC risk.
presents with a phenotype consistent with classical FAP Important considerations need to be kept in mind when
but with either a negative family history or with fea- considering genetic testing for FAP. It is essential that
tures consistent with an autosomal recessive pattern of patients and their immediate family members be of-
inheritance (for example, skipped generations often in fered genetic counseling, in order that they understand
the presence of multiple affected members of a single the implications and potential limitations as well as the
generation) it is worth considering testing for MYH mu- possible benefits.2 1 This includes acknowledgement of
tations.7, 8 The MYH gene encodes a homolog of a bacte- the positive and negative predictive values to such tests.
rial DNA excision repair gene and mutations within this There are also important psychological issues that must
gene lead in turn to the accumulation of mutations within be confronted for these families, including access to
the APC gene. Two common mutations in the MYH counselors experienced in explaining in detail the con-
gene (Y165C, G382D) account for the majority of cases cepts involved. Patients undergoing genetic testing need
where this proves to be the defective allele in phenotypic to provide informed consent and genetic testing should
FAP.9, 10 be conducted through an approved facility, where ac-
There is also a distinct category of subjects with an at- curate interpretation is assured.21 Information and useful
tenuated form of FAP (AFAP), characterized by a later material can be accessed through the following links:
age of onset (over age 40), fewer adenomatous polyps www.genetests.org
(generally <50) and with a lower cumulative CRC risk. www.geneclinics.org
Some of these patients will be discovered to harbor mu- www3.ncbi.nlm.nih.gov
tations in the extreme 3’ or 5’ end of the APC gene.11, 12 www.hereditarycc.org
Others, particularly those of Ashkenazi Jewish extrac-
tion, will be found to harbor the I1307K mutation.1 3, 1 4 Peutz Jeghers Syndrome
This particular mutation is highly prevalent among
Ashkenazim and carriers have a lifetime CRC risk in the Peutz Jeghers syndrome (PJS) is an autosomal domi-
range of 10-20%.15 Other considerations in patients pre- nant cancer syndrome with an incomplete penetrance
senting with an attenuated phenotype include testing for pattern and a variable phenotype.2 2 The pathological
Keio J Med 2007; 56 (1): 14－ 20 17

hallmark of PJS is the presence of distinctive hamarto- has a first degree relative with CRC diagnosed at age
matous polyps throughout the gastrointestinal tract in as- <50 years and/or colorectal adenomas diagnosed at age
sociation with pathognomonic hyperpigmented mucocu- <40 years.1 In addition, a patient with CRC diagnosed at
taneous spots in the perioral and buccal mucosa. Polyps age <60 years would be classified as Lynch syndrome if
typically appear during the first two decades of life and the tumor contained the mutational signature of defective
patients frequently present because of intussusception as mismatch repair (microsatellite instability, MSI), which is
a result of small intestinal hamartomas. In addition, pa- described below. Individuals diagnosed with Lynch syn-
tients with PJS are at greatly increased risk for a number drome, as well as their immediate family members, have
of cancers, including throughout the GI tract (esophagus, a greatly increased lifetime risk of developing colorectal
stomach, pancreas, small intestine and colon) as well as cancer and female carriers carry an equally substantial
at extraintestinal sites, principally breast.22 lifetime risk of developing endometrial cancer. These
PJS is caused by mutations or deletions in the STK11 features of Lynch syndrome make it particularly impor-
gene and informative mutations are found in 30-70% of tant to recognize in order to offer preventive screening
sporadic cases and ~70% individuals with a positive fam- for at-risk family members. These recommendations
ily history. Genetic testing for mutations in the STK11 include frequent colonoscopy (at 1-2 year intervals)
gene is recommended for at-risk individuals, principally starting at age 20-25 or at least 10 years earlier than the
the first degree relatives of probands with confirmed earliest age of onset of cancer in a family member.24 In
PJS. This includes consideration for testing children of addition, endometrial cancer surveillance is recommend-
affected family members, since there is a very high rate ed with frequent transvaginal ultrasound examination.2 4
of intussusception in carriers.2 2 Once the STK11 gene These screening recommendations make it imperative to
mutation is identified in a PJS patient, family members establish an unequivocal diagnosis of Lynch syndrome
and at-risk individuals can be tested specifically for that where possible.
mutation, since the results are highly informative. At risk The mutational signature of Lynch syndrome, MSI, is
family members in whom a mutation is not found in the present in most if not all cancers arising in patients with
STK11 gene are recommended to undergo small intesti- this disease complex and results from mutations, dele-
nal imaging and colonoscopy as well as periodic upper tions or defects in one of several DNA mismatch repair
endoscopy in order to identify and remove large ham- genes.25 The genes most frequently involved are hMLH1
artomatous polyps as well to undergo regular periodic and hMSH2 (which together account for ~90% of the
surveillance for other cancers (breast, testis, pancreas).22 germline mutations that account for Lynch syndrome) as
well as hMSH6 (which accounts for the majority (~7%)
Hereditary, non-polyposis colorectal cancers of the remaining mutations) and a small minority caused
(Lynch Syndrome) by hPMS2.4 Genetic testing through commercially avail-
able conventional DNA sequence analysis is currently
The non-polyposis forms of hereditary colon cancer offered only to detect mutations in hMLH1 and hMSH2
are much more common than the syndromes outlined genes.24
above, with estimates that it may account for 3-5% of all
cases of CRC.2 Lynch syndrome is an autosomal domi- Problems in diagnosing Lynch syndrome
nant disease with pleomorphic features characterized by
early onset of colon cancer (< age 40 years) often in as- Having outlined the basic elements of this disease
sociation with a family history positive for either colon complex, it is important to understand that reaching an
cancer or for other associated cancers, including endo- unambiguous diagnosis of Lynch syndrome is often
metrial, ovarian, brain, small intestinal, pancreatic, uri- quite challenging. There are at least three major reasons
nary tract cancers.23, 24 Criteria were established over 15 for this difficulty. First, clinical suspicion of Lynch
years ago to establish clinical criteria for the diagnosis syndrome largely rests on a clinical diagnosis based on
of this common form of hereditary CRC (the so-called information concerning the extended family history of
Amsterdam criteria). These included the presence of at a suspected proband. In this regard, a detailed family
least 3 members of a kindred with CRC (excluding FAP) history is often not documented in the patients’ medical
with at least one member being a first degree relative of record-- particularly with respect to the presence of as-
the other 2, that at least two generations be involved and sociated cancers (endometrial, ovarian, ureteric etc)
that one was age <50 years.1 These criteria have been in maternal or paternal relatives.2 6 The absence of this
modified sequentially and the current, so-called Modi- information might lead to the clinician overlooking pos-
fied Bethesda criteria now classify a patient as having sible Lynch syndrome in a patient presenting with new
Lynch syndrome if there is CRC diagnosed <50 years, or onset CRC. The second issue that has led to confusion
if the patient has a synchronous or metachronous CRC or and ambiguity with respect to establishing a molecular
if the patient has an extracolonic cancer (see above) di- diagnosis of Lynch syndrome is that less than half of
agnosed at any age, or if the patient with CRC (any age) patients will have a pathogenic mutation in one of the
18 Davidson NO: Genetic testing in colorectal cancer

mismatch repair genes.4 DNA sequencing of the two tant category of patients should be offered genetic test-
leading candidate genes (hMLH1 and hMSH2) led to the ing for Lynch syndrome.30 This is important since a high
realization that many of the sequence variations noted proportion of young patients with CRC do not meet strict
were silent polymorphisms or missense mutations of un- clinical criteria for Lynch syndrome. In addition, young
known significance.4 This is an important problem since patients with CRC tend to have less right sided predomi-
the emerging information from several centers supports a nance of their colon cancers than the typical Lynch syn-
divergence in the classification of Lynch syndrome into drome patients.30 However, almost three quarters of the
classical genotypic families with MSI and/or an identi- tumors from young patients with CRC will show MSI
fied mismatch repair gene mutation and families with and approximately half of these will have mismatch re-
phenotypic features only.2 7 A rational approach to this pair gene mutations. Finally, over 40% of young patients
issue is discussed in detail below. The third issue that with CRC will develop a second cancer within 12 years,
complicates establishing an unequivocal diagnosis of the majority of these being GI tract cancers.30
Lynch syndrome is the observation that the mutational
signature of DNA mismatch repair (MSI) is present in New information concerning suspected
~15% of sporadic CRC.2 In this setting, MSI is associ- Lynch syndrome patients who do not carry
ated with acquired silencing (through promoter meth- the molecular fingerprint of MSI
ylation) of the hMLH1 gene and not a heritable mutation
in the germline. As noted above, promoter hypermeth-      Three new studies have added considerable insight into
ylation is an important epigenetic mechanism of gene the management and genetic classification of patients
silencing-as exemplified by MSI appearing in the set- with phenotypic Lynch syndrome.2 9, 3 1, 3 2 These studies
ting of hyperplastic polyposis and the serrated adenoma addressed the question of whether the disease process
to adenocarcinoma transition. was fundamentally distinct in patients based on the pres-
ence or absence of MSI and the findings provide impor-
Suggested algorithm for genetic testing of patients tant new insights for the management of these patients
and family members with suspected Lynch syndrome and their at-risk family members. As discussed above,
patients with classic Lynch syndrome have a constella-
In view of the aforementioned problems and complica- tion of clinical features in conjunction with the molecu-
tions in establishing a diagnosis of Lynch syndrome, it is lar fingerprint of MSI, frequently accompanied by altera-
extremely important to elicit a detailed family history of tions in the expression of mismatch repair genes.27 There
other cancers, the age of diagnosis and where possible to is, however, another group of patients who manifest the
obtain the tumor specimen for MSI testing. Tumors that clinical features of Lynch syndrome (early onset CRC,
manifest MSI should be examined by immunohistochem- positive family history, etc) but where the tumor does
ical staining for the protein products of the mismatch not demonstrate MSI and the genetic basis for the can-
repair genes (ie MLH1, MSH2, MSH6 and PMS2).2 8 cer susceptibility is unknown. There are important fea-
Using this approach, the clinician will be able to detect tures concerning this subset of patients, who have been
loss of staining of one of the candidate mismatch repair referred to as familial non-polyposis colorectal cancer
genes and would then conclude that there is likely to be type X.31 These include the finding that the standardized
an inactivating mutation in the corresponding gene. Loss incidence ratios for CRC are elevated in first degree rela-
of staining of a mismatch repair gene, coupled with MSI tives (as in classical Lynch syndrome) but that the inci-
would then be a reasonable basis to pursue mutational dence of other Lynch syndrome associated cancers (en-
analysis of the corresponding gene in order to identify dometrial, ovarian, stomach etc) were not higher than in
informative mutations that could then be used to screen an age-adjusted general population.3 1 This information
at-risk family members. A recent report has outlined a is of particular relevance in terms of counseling female
useful web-based questionnaire that uses a balanced al- family members who may be at risk, since their need for
gorithm of clinical information (age of diagnosis, gender, intensive screening for endometrial cancer surveillance
location of tumor, presence of synchronous or metachro- is no longer recommended. In addition, the average age
nous tumors, family history of CRC or endometrial can- at which CRC was detected in the family members from
cer) as well as immunohistochemical staining to predict the familial non-polyposis colorectal cancer syndrome
the likelihood of Lynch syndrome.2 9 This questionnaire cohort (type X) was approximately 60 years, compared
can be accessed at: to less than age 50 in classical Lynch syndrome.3 1 This
http://www1.hgu.mrc.ac.uk/Softdata/MMRpredict.php information is particularly useful in terms of providing
guidelines for the initiation of screening colonoscopy
Other considerations for who should be offered (5-10 years younger than the earliest CRC diagnosis in
genetic testing for Lynch syndrome the family) and for the intervals between screening (ev-
ery 5 years). These patients and their family members
Any patient with CRC diagnosed at <40 years impor- should be counseled that they do not have a sinister can-
Keio J Med 2007; 56 (1): 14－ 20 19

cer predisposition syndrome. In particular, since women Cohen Z, Bapat B: Genotype-phenotype correlations in attenuated
family members with Lynch syndrome carry a 40-60% adenomatous polyposis coli. Am J Hum Genet 1998; 62(6): 1290
lifetime risk of endometrial cancer, this diagnosis carries －1301
12. Heppner Goss K, Trzepacz C, Tuohy TM, Groden J: Attenuated
major implications for their continued surveillance.33 APC alleles produce functional protein from internal translation
initiation. Proc Natl Acad Sci U S A 2002; 99(12): 8161－8166
Summary comments 13. Drucker L, Shpilberg O, Neumann A, Shapira J, Stackievicz R,
Beyth Y, Yarkoni S: Adenomatous polyposis coli I1307K muta-
tion in Jewish patients with different ethnicity: prevalence and
Newer advances in the molecular genetics of colorectal phenotype. Cancer 2000; 88(4): 755－760
cancer pathogenesis will undoubtedly continue to evolve. 14. Zauber NP, Sabbath-Solitare M, Marotta SP, Bishop DT: The
As public awareness of the human genome project ex- characterization of somatic APC mutations in colonic adenomas
pands and commercialization of the tools for diagnostic and carcinomas in Ashkenazi Jews with the APC I1307K variant
genomic applications continue to expand, it will be im- using linkage disequilibrium. J Pathol 2003; 199(2): 146－151
15. Prior TW, Chadwick RB, Papp AC, Arcot AN, Isa AM, Pearl
portant for physicians to stay well informed about the DK, Stemmermann G, Percesepe A, Loukola A, Aaltonen LA, De
limitations, interpretation and utility of these approaches La Chapelle A: The I1307K polymorphism of the APC gene in
and to be able to apply them in a scientifically sound colorectal cancer. Gastroenterology 1999; 116(1): 58－63
manner to the management of their patients. 16. Chow E, Lipton L, Lynch E, D’Souza R, Aragona C, Hodgkin L,
Brown G, Winship I, Barker M, Buchanan D, Cowie S, Nasioulas
S, du Sart D, Young J, Leggett B, Jass J, Macrae F: Hyperplastic
Acknowledgements polyposis syndrome: Phenotypic presentations and the role of
MBD4 and MYH. Gastroenterology 2006; 131: 30－39
This work was supported through grants from the Na- 17. Huang CS, O’Brien MJ, Yang S, Farraye FA: Hyperplastic pol-
tional Institutes of Health (HL-38180, DK-56260 and yps, serrated adenomas, and the serrated polyp neoplasia pathway.
Am J Gastroenterol 2004; 99: 2242－2255
DK-52574).
18. Young J, Jass JR: The case for a genetic predisposition to ser-
rated neoplasia in the colorectum: hypothesis and review of the
literature. Cancer Epidemiol Biomarkers Prev 2006; 15(10): 1778
References
－1784
1. Potter JD: Colorectal cancer: molecules and populations. J Natl 19. Laiho P, Kokko A, Vanharanta S, Salovaara R, Sammalkorpi H,
Cancer Inst 1999; 91(11): 916－932 Jarvinen H, Mecklin JP, Karttunen TJ, Tuppurainen K, Davalos
2. de la Chapelle A: Genetic predisposition to colorectal cancer. Nat V, Schwartz S, Jr., Arango D, Makinen MJ, Aaltonen LA: Ser-
Rev 2004; 4: 769－780 rated carcinomas form a subclass of colorectal cancer with distinct
3. Giardiello FM, Brensinger JD, Petersen GM: AGA technical molecular basis. Oncogene 2007; 26(2): 312－320
review on hereditary colorectal cancer and genetic testing. Gastro- 20. Makinen MJ: Colorectal serrated adenocarcinoma. Histopathology
enterology 2001; 121(1): 198－213 2007; 50(1): 131－150
4. Lynch HT, de la Chapelle A: Hereditary colorectal cancer. N Engl 21. Petersen GM, Brensinger JD, Johnson KA, Giardiello FM: Genet-
J Med 2003; 348(10): 919－932 ic testing and counseling for hereditary forms of colorectal cancer.
5. Doxey BW, Kuwada SK, Burt RW: Inherited polyposis syn- Cancer 1999; 86(11 Suppl): 2540－2550
dromes: molecular mechanisms, clinicopathology, and genetic 22. Giardiello FM, Trimbath JD: Peutz-Jeghers syndrome and man-
testing. Clin Gastroenterol Hepatol 2005; 3: 633－641 agement recommendations. Clin Gastroenterol Hepatol 2006; 4:
6. Johnson KA, Trimbath JD, Petersen GM, Griffin CA, Giardiello 408－415
FM: Impact of genetic counseling and testing on colorectal cancer 23. Vasen HFA, Boland CR: Progress in genetic testing, classifi-
screening behavior. Genet Test 2002; 6(4): 303－306 cation, and identification of Lynch Syndrome. JAMA 2005;
7. Al-Tassan N, Chmiel NH, Maynard J, Fleming N, Livingston 293(16): 2028
AL, Williams GT, Hodges AK, Davies DR, David SS, Sampson 24. Lynch HT, Boland CR, Gong G, Shaw TG, Lynch PM, Fodde R,
JR, Cheadle JP: Inherited variants of MYH associated with so- Lynch JF, de la Chapelle A: Phenotypic and genotypic heteroge-
matic G:C-->T:A mutations in colorectal tumors. Nat Genet 2002; neity in the Lynch Syndrome: diagnostic, surveillance and man-
30(2): 227－232 agement implications. Eur J Hum Genet 2006; 14: 390－402
8. Sampson JR, Dolwani S, Jones S, Eccles D, Ellis A, Evans DG, 25. Aaltonen LA, Peltomaki P, Mecklin JP, Jarvinen H, Jass JR,
Frayling I, Jordan S, Maher ER, Mak T, Maynard J, Pigatto F, Green JS, Lynch HT, Watson P, Tallqvist G, Juhola M, et  al: Rep-
Shaw J, Cheadle JP: Autosomal recessive colorectal adenoma- lication errors in benign and malignant tumors from hereditary
tous polyposis due to inherited mutations of MYH. Lancet 2003; nonpolyposis colorectal cancer patients. Cancer Res 1994; 54(7):
362(9377): 39－41 1645－1648
9. Sieber OM, Lipton L, Crabtree M, Heinimann K, Fidalgo P, Phil- 26. Grover S, Stoffel EM, Bussone L, Tschoegl E, Syngal S: Physi-
lips RK, Bisgaard ML, Orntoft TF, Aaltonen LA, Hodgson SV, cian assessment of family cancer history and referral for genetic
Thomas HJ, Tomlinson IP: Multiple colorectal adenomas, classic evaluation in colorectal cancer patients. Clin Gastroenterol Hepa-
adenomatous polyposis, and germ-line mutations in MYH. N Engl tol 2004; 2(9): 813－819
J Med 2003; 348(9): 791－799 27. Boland CR: Decoding hereditary colorectal cancer. New Eng J
10. Gismondi V, Meta M, Bonelli L, Radice P, Sala P, Bertario L, Med 2006; 354(26): 2815－2817
Viel A, Fornasarig M, Arrigoni A, Gentile M, Ponz de Leon M, 28. Hampel H, Frankel WL, Martin E, Arnold M, Khanduja K, Kue-
Anselmi L, Mareni C, Bruzzi P, Varesco L: Prevalence of the bler P, Nakagawa H, Sotamaa K, Prior TW, Westman J, Panescu
Y165C, G382D and 1395delGGA germline mutations of the J, Fix D, Lockman J, Comeras I, de la Chapelle A: Screening for
MYH gene in Italian patients with adenomatous polyposis coli the Lynch syndrome (hereditary nonpolyposis colorectal cancer).
and colorectal adenomas. Int J Cancer 2004; 109(5): 680－684 N Engl J Med 2005; 352(18): 1851－1860
11. Soravia C, Berk T, Madlensky L, Mitri A, Cheng H, Gallinger S, 29. Barnetson RA, Tenesa A, Farrington SM, Nicholl ID, Cetnarskyj
20 Davidson NO: Genetic testing in colorectal cancer

R, Porteous ME, Campbell H, Dunlop MG: Identification and sur- Lower cancer incidence in Amsterdam-I criteria families without
vival of carriers of mutations in DNA mismatch-repair genes in mismatch repair deficinecy. Familial colorectal cancer Type X.
colon cancer. N Engl J Med 2006; 354(26): 2751－2763 JAMA 2005; 293(16): 1979－1985
30. Durno C, Aronson M, Bapat B, Cohen Z, Gallinger S: Family his- 32. Dove-Edwin I, De Jong AE, Adams J, Mesher D, Lipton L, Sasi-
tory and molecular features of children, adolescents, and young eni P, Vasen HFA, Thomas HJW: Prospective results of surveil-
adults with colorectal carcinoma. Gut 2005; 54(8): 1146－1150 lance colonscopy in dominant familial colorectal cancer with and
31. Lindor NM, Rabe K, Petersen GM, Haile R, Casey G, Baron without Lynch Syndrome. Gastroenterology 2006; 130: 1995 －
J, Gallinger S, Bapat B, Aronson M, Hopper J, Jass J, LeMer- 2000
chand L, Grove J, Potter J, Newcomb P, Terdiman JP, Conrad P, 33. Offit K, Kauff ND: Reducing the risk of gynecologic cancer in the
Moslein G,Goldberg R, Ziogas A, Anton-Culver H, de Andrade Lynch Syndrome. New Eng J Med 2006; 354(3): 293－295
M, Siegmund K, Thibodeau SN, Boardman LA, Seminara D: