Unformatted text preview: GASTROENTEROLOGY 2002;123:862– 876 SPECIAL REPORTS AND REVIEWS
Emerging Concepts in Colorectal Neoplasia
JEREMY R. JASS,* VICKI L. J. WHITEHALL,‡ JOANNE YOUNG,‡ and BARBARA A. LEGGETT‡
*Department of Molecular and Cellular Pathology, University of Queensland Medical School; and ‡Conjoint Gastroenterology Laboratory,
Queensland Institute of Medical Research, Queensland, Australia An understanding of the mechanisms that explain the
initiation and early evolution of colorectal cancer should
facilitate the development of new approaches to effective prevention and intervention. This review highlights
deﬁciencies in the current model for colorectal neoplasia in which APC mutation is placed at the point of
initiation. Other genes implicated in the regulation of
apoptosis and DNA repair may underlie the early development of colorectal cancer. Inactivation of these genes
may occur not by mutation or loss but through silencing
mediated by methylation of the gene’s promoter region.
hMLH1 and MGMT are examples of DNA repair genes
that are silenced by methylation. Loss of expression of
hMLH1 and MGMT protein has been demonstrated immunohistochemically in serrated polyps. Multiple lines
of evidence point to a “serrated” pathway of neoplasia
that is driven by inhibition of apoptosis and the subsequent inactivation of DNA repair genes by promoter
methylation. The earliest lesions in this pathway are
aberrant crypt foci (ACF). These may develop into hyperplastic polyps or transform while still of microscopic size
into admixed polyps, serrated adenomas, or traditional
adenomas. Cancers developing from these lesions may
show high- or low-level microsatellite instability (MSI-H
and MSI-L, respectively) or may be microsatellite stable
(MSS). The suggested clinical model for this alternative
pathway is the condition hyperplastic polyposis. If colorectal cancer is a heterogeneous disease comprising
discrete subsets that evolve through different pathways,
it is evident that these subsets will need to be studied
individually in the future. he central hypothesis explored in this review is that
colorectal cancer is a heterogeneous disease. By deﬁning subsets of colorectal cancer, it should be possible
to develop more targeted approaches to prevention and
treatment. Because genetic ﬁndings provided the initial
clues to the heterogeneous nature of colorectal cancer, the
review begins with a critical outline of the molecular
background. The hypothesis is then developed through
the stepwise correlation of molecular mechanisms with
morphological observations and culminates in a model of T “serrated neoplasia.” It is suggested that serrated polyps
are initiated by inhibition of apoptosis and by this mechanism are primed for further evolution through the disruption of DNA repair.
The autosomal dominant condition familial adenomatous polyposis (FAP) has long been viewed as the hereditary counterpart of sporadic colorectal cancer. The discovery of the genetic basis for FAP1– 4 and the
demonstration of APC alterations in sporadic colorectal
neoplasms5 established APC mutation as a critical ratelimiting step in the initiation of colorectal neoplasia.6
Kinzler and Vogelstein7 speculated that this single gene
serves as the “gatekeeper” of epithelial proliferation and
takes pride of place in the stepwise model of colorectal
tumorigenesis. By contrast, the “caretaking” DNA mismatch repair genes underlying the disorder hereditary
nonpolyposis colorectal cancer (HNPCC)8 –12 were envisaged to accelerate the progression of neoplasia following
the step of initiation.7 The tumorigenic pathway in
HNPCC would therefore proceed at a rapid pace but
would not differ qualitatively from the basic genetic
model. This ﬁts well with clinical and pathologic observations regarding the “aggressive” nature of HNPCC
adenomas.13 Recently, the concept of “gatekeeper” and
“caretaker” genes has become blurred. APC may act as a
“caretaker,”14 whereas DNA mismatch repair genes may
inﬂuence cell proliferation by signaling either cell cycle
arrest or apoptosis in the presence of DNA damage. Loss
of DNA mismatch repair function may therefore result in
continuation of cell proliferation despite a background of
After the discovery of the DNA mismatch repair genes
and the associated phenotype characterized by extensive
Abbreviations used in this paper: ACF, aberrant crypt foci; CIMP, CpG
island methylator phenotype; FAP, familial adenomatous polyposis;
HNPCC, hereditary nonpolyposis colorectal cancer; MSI, microsatellite
instability; MSI-H, microsatellite instability-high; MSI-L, microsatellite
instability-low; MSS, microsatellite stable.
© 2002 by the American Gastroenterological Association
doi:10.1053/gast.2002.35392 September 2002 COLORECTAL CANCER DNA microsatellite instability (MSI),17–19 it was appreciated that most cancers with the “mutator” phenotype
were not inherited but resulted from somatic inactivation
of a mismatch repair gene. The usual mismatch repair
gene implicated in this subgroup of sporadic colorectal
cancers was shown to be hMLH1,20 and the usual inactivating mechanism was methylation of the promoter
region of the gene.21–23 Before this mechanism was uncovered, studies were beginning to delineate multiple
differences between the 15% of cancers with MSI and the
remainder that were microsatellite stable (MSS). As compared with MSS cancers, MSI cancers are more likely to
be proximally located, to present at a more advanced age,
to occur in women, and to be associated with a favorable
prognosis.17,19,24,25 At the morphologic level, they are
more likely to be mucinous, to be poorly differentiated,
and to show lymphocytic inﬁltration.26 –28 MSI cancers
are also associated with normal immunostaining patterns
for -catenin29 (this applies speciﬁcally to sporadic MSI
cancers) and p53.30,31
Genetic features of the MSI subset of colorectal cancer
include diploid DNA content and a lack of loss of
heterozygosity at loci harboring tumor suppressor genes
such as APC, TP53, and candidate loci on chromosome
18q.29,32,33 There are also reports of lower frequencies of
APC34,35 and TP5336 mutation. Different tumor suppressor genes may serve as surrogates for APC and TP53
in MSI colorectal cancers, notably TGF RII,37 IGF2R,38
and BAX.39 These genes contain short mononucleotide
repeats within their encoding regions [e.g., poly(A)n].
Such repetitive nucleotide sequences are intrinsically unstable or prone to be copied incorrectly during the DNA
synthetic phase of the cell cycle. For example, there may
be deletion of an adenine in the replicated copy. The
parent and replicated complementary copy may still
anneal to form double-stranded DNA, but the longer
parent strand will form a loop that serves to alert the 863 DNA mismatch repair system. In the absence of DNA
repair proﬁciency, the replication error is not repaired.
Instead, the imperfectly replicated copy becomes a template in the next round of cell division, generating a
permanent mutation within daughter cells. The mutant
allele is a different size and can be separated from the
wild-type strand by gel electrophoresis. This is also the
basis for the demonstration of mutated band shifts in
nonencoding microsatellite regions. The most commonly
employed microsatellite markers comprise either mononucleotide [e.g., poly(A)n] or dinucleotide [e.g., poly(CA)n] repeats.
A more recently recognized molecular alteration found
frequently in MSI cancers, although not restricted to this
group, is the CpG island methylator phenotype
(CIMP).40 Dense aggregates of CpG sites (cytosine-guanine dinucleotide sequence) may occur in the promoter
regions of genes and are termed CpG islands. Extensive
methylation of cytosine bases is associated with promoter
silencing. Cancers demonstrating methylation and silencing of multiple genes are described as CIMP positive.40 Modiﬁcation of gene expression by such means as
DNA methylation (as opposed to structural DNA alteration or physical loss of the gene) is described as an
epigenetic change. Genes other than hMLH1 that may be
methylated in human tumors include ER,41 p16,42 p14,43
HPP1/TPEF,44 MGMT,45 THBS1,42 APC,46 COX-2,47
CDH1,48 RIZ1,49 and RASSF1A50 (Table 1). The extent
to which methylation of a particular gene results in
functional silencing and subsequent tumor suppression
in colorectal cancer (and other tumors) is currently under
intense investigation. This review will focus on 3 of these
genes: hMLH1, MGMT, and HPP1/TPEF.
For the preceding reasons, there are grounds for separating colorectal cancers into 2 largely nonoverlapping
groups: MSI and MSS. These have also been termed the
“mutator” (with reference to the frequent mutations Table 1. Genes Methylated in Colorectal Cancer
Gene Function Reference hMLH1
ER (estrogen receptor)
HPP1/TPEF (hyperplastic polyposis protein 1/transmembrane protein,
epidermal growth factor, follistatin)
MGMT (O-6-methylguanine-DNA methyltransferase)
THBS1 (thrombospondin 1)
APC (adenomatous polyposis coli)
RIZ1 (retinoblastoma protein-interacting zinc ﬁnger)
RASSF1A (Ras-association/Ras-effector Nore 1) DNA mismatch repair
Growth and differentiation
p53 regulation 21
Cell signalling 44
50 864 JASS ET AL. occurring in the absence of DNA mismatch repair) and
“suppressor” (with reference to the frequent loss of loci
harboring “classical” tumor suppressor genes) pathways.
It has been assumed, however, that the divergence into
MSI and MSS pathways occurs after the step of APC
mutation and initiation of a microadenoma.7,51 It is
important to establish whether APC mutation necessarily occurs at the point of initiation of colorectal cancer or
whether some pathways may be initiated independently
of this mechanism. An understanding of the earliest steps
in neoplastic evolution is relevant to development of
targeted chemopreventive compounds and polyp surveillance programs. Below, it will be argued that APC
mutation may not be the initiating step in MSI cancers
or in some MSS cancers. Similar-Appearing Adenomas May
Be Biologically Different
In FAP, only a small proportion of the many
thousands of adenomas will transform into cancers, and
the transition may take decades.52 Additionally, a unique
mode of adenomatous growth involving the fusion of
microadenomas into a polyclonal mass has been documented in this condition.53 In HNPCC, the ratio of
adenoma to carcinoma is close to unity, and evolution to
cancer appears to be rapid as well as frequent.54,55 An
intermediate position is observed in common forms of
colorectal cancer. These observations indicate that similar-appearing adenomas may be biologically distinct and
characterized by different rates of malignant conversion.
Because individual colorectal adenomas occurring in the
condition FAP have limited potential for malignant conversion, a similar behavior would be expected for sporadic colorectal adenomas if these were also initiated by
biallelic inactivation of APC. On the basis of these
observations, extrapolation of the FAP model to sporadic
colorectal cancer should not be accepted uncritically. How Common Is APC Mutation in
Initiation and Progression of
Is APC mutation the invariable ﬁrst genetic alteration in colorectal neoplasms? The gene clearly has a
major role in directing epithelial growth and differentiation. As a component of the WNT or wingless cellsignalling pathway, a normal function of APC protein is
to bind the key effector molecule -catenin. When APC
is inactivated, -catenin translocates from the lateral cell
membrane to the nucleus, where it drives the transcription of multiple genes implicated in tumor growth and
invasion.56 In FAP adenomas, either somatic mutation of GASTROENTEROLOGY Vol. 123, No. 3 APC or loss of heterozygosity at chromosome 5q is
virtually universally present, even within early lesions.57,58 It has been assumed that the same would apply
to sporadic adenomas. In fact, relatively low frequencies
of APC mutation are reported in early sporadic neoplasms. In dysplastic aberrant crypt foci (ACF; see below), ﬂat tubular adenomas, and polypoid tubular adenomas, the frequency of APC mutation is reported as
0%, 7%, and 36%, respectively.59 – 61 A higher frequency
(77%) is reported in villous adenomas.62 A subsequent
study demonstrated a trend for APC alteration with
respect to grade of epithelial dysplasia and villous architecture and a signiﬁcant association with size of adenomas.63 In 3 of 7 sporadic adenomas with loss of heterozygosity (LOH) at 5q and focal high-grade dysplasia, the
LOH was restricted to the high-grade portion of the
lesion.64 These data implicate APC inactivation in progression and growth rather than initiation of colorectal
adenoma. The ﬁndings ﬁt with the suggested role of
APC mutation in the development of chromosomal instability.14
In sporadic colorectal cancer, APC mutation is again
not universal, occurring in approximately 60% of
cases.18,33–35,65 A higher frequency of APC mutation is
observed in rectal cancer (82%).66 A relatively low frequency of APC mutation is described in ﬂat colorectal
cancers (35%),60 sporadic high-level MSI (MSI-H) cancers (39%),33–35,65 and HNPCC cancers (44%).18,33,65
The shortfall of APC mutation in HNPCC is in part
made up by oncogenic -catenin mutation that is found
in around 30% of HNPCC cancers.67,68 However, -catenin mutation is rarely detected in sporadic MSI-H cancer.34 In fact, -catenin mutation is infrequent in all
colorectal cancers other than those complicating
HNPCC.68 AXIN2 mutation provides a further mechanism for -catenin activation in some MSI-H cancers.69
These data indicate that APC mutation is not obligatory
in all cases of colorectal cancer, even as a late event. DNA
methylation would provide an alternative mechanism of
silencing APC46 but is found in only 18% of colorectal
carcinomas and not necessarily in those lacking APC
mutation.70 Additionally, the normal pattern of -catenin immunolocalization (along lateral cell membrane) in
most examples of sporadic MSI-H colorectal cancer29,71
indicates that the WNT (wingless) signalling pathway
(in which APC and -catenin participate) remains largely
intact in this subgroup of colorectal cancers. Notwithstanding the possible involvement of WNT signalling
genes downstream of -catenin, the assumption that inactivation of the WNT pathway initiates the “vast majority” of colorectal neoplasms7 does not agree with the September 2002 facts. Nevertheless, WNT pathway disruption must occur during the evolution of most if not all non-MSI-H
colorectal cancers. MSI and Methylation: A Basis for
In the introduction, a case was made for classifying colorectal cancer into 2 groups based on the presence
or absence of DNA MSI. It was also noted that cancers
may be distinguished according to the presence or absence of DNA methylation. Because MSI usually arises in
sporadic colorectal neoplasia as a consequence of methylation and inactivation of the DNA mismatch repair
gene hMLH1, there will be an association between DNA
MSI and methylation, but the overlap of “mutator” and
“methylator” phenotypes is not exact. In particular, some
cancers with extensive DNA methylation do not show
the mutator phenotype.42 Additionally, cancers show
different degrees of MSI and different degrees of methylation. This adds a layer of complexity to the molecular
classiﬁcation of colorectal cancer that must be addressed.
A typical panel of microsatellite markers indicates that
MSI cancers are distributed bimodally with a breakpoint
at around 40%.72 Cancers with instability at 30%– 40%
of markers or more have been deﬁned as showing MSIH.73 The 5 microsatellite markers comprising the NCI
panel include 2 mononucleotide (BAT25 and BAT26)
and 3 dinucleotide (D5S346, D2S123, and D17S250)
markers. Cancers showing instability in 2 or more of
these markers are classiﬁed as MSI-H.73 Most (but not
all, see below) cancers classiﬁed as MSI-H according to
this criterion display the clinical, pathologic, and molecular features of MSI cancers as described above. Additionally, they are characterized by methylation of
hMLH1 and other genes that may be methylated as part
of the CIMP. Some MSI-H cancers may be CIMP negative, for example, HNPCC cancers and cancers arising
through somatic mutation of a DNA mismatch repair
Cancers with MSI-L and with instability limited to
dinucleotide markers do not show the full complement of
MSI characteristics but may share the feature of DNA
methylation.75 Importantly, these cancers are not characterized by methylation of hMLH1, but many show
methylation of the DNA repair gene MGMT (see below).75 Cancers with instability involving 2 NCI panel
dinucleotide markers but no mononucleotide markers are
technically MSI-H but do not show MSI-H features
(clinical, pathologic, or molecular) and should probably
be grouped as MSI-L.28 This misdiagnosis arises because COLORECTAL CANCER 865 the dinucleotide markers in the NCI panel are relatively
sensitive to MSI-L status.72
The proportion of non–MSI-H cancers that is MSI-L
depends on the number of markers that is used. If the
NCI panel is employed,73 up to 10% of cancers are
MSI-L. A study using 44 markers showed MSI-L in 68%
of non-MSI-H cancers and found that this was distributed as a nonrandom quantitative trait. There was an
excess of samples with 10%–25% unstable microsatellites and an excess with no instability.76 As noted above,
the higher range MSI-L cancers are more likely to be
CIMP positive and show methylation of MGMT.75
Around 40% of colorectal cancers show MGMT methylation, and most are not MSI-H.45,75 Methylation of
MGMT is described in 64% of MSI-L cancers and 26%
of MSS cancers.75
Below, it will be argued that most sporadic MSI-H
cancers develop through a pathway that is independent of
APC mutation. However, the same argument may extend to subsets of MSI-L and perhaps MSS cancers,
particularly those with evidence of DNA methylation. In
other words, the importance of the APC-driven model
may diminish even within the subset of cancers with
chromosomal instability. It is also evident that the classiﬁcation of colorectal cancer on the basis of MSI per se
is unsatisfactory and should probably be based on the
mechanism underlying the genetic instability.
Genes that may be methylated have been classiﬁed as
type A and type C. Methylation of type A genes, for
example, estrogen receptor (ER), is age related and occurs
in normal colorectal mucosa as well as in cancer.41 Methylation of type C genes is cancer related and is likely to
lead to pathogenic gene silencing in the case of hMLH1,
MGMT, p16, p14, RIZ1, and HPP1/TPEF.22,44,45,49,77,78
Like mutation, methylation is considered to be irreversible in vivo and exhibits somatic inheritance.79 In normal
colorectal mucosa, methylation patterns vary from crypt
to crypt.80 In the case of frequently methylated genes, a
high proportion of crypts will be affected, whereas, in the
case of infrequently methylated genes (type C genes),
methylation will be more sporadic. When methylation
results in silencing that gives rise to a growth advantage,
clonal expansion may occur. It is possible that some ACF
(see below) may represent the morphologic expression of
methylation-induced clonal expansion. ACF
The term ACF was initially applied to the microscopic epithelial lesions observed in experimental animals exposed to carcinogens.81,82 Similar lesions have
been identiﬁed in the mucosal surface of human colon 866 JASS ET AL. after methylene blue staining.83,84 ACF in humans have
been classiﬁed as dysplastic and nondysplastic.84,85 Dysplastic ACF are equivalent to microadenomas and probably account for about 5% of all ACF.85 Most nondysplastic ACF often show the histologic ﬁnding of crypt
serration in which the epithelium is folded and adopts a
saw-tooth outline. These ACF are indistinguishable from
minute hyperplastic polyps. Serration is a relatively easily recognized morphologic alteration that probably
arises as a consequence of inhibition of apoptosis (see
below under “top-down, bottom-up models” for the underlying mechanisms). The mean number of ACF identiﬁed in a series of 12 resection specimens from subjects with colorectal cancer has been calculated as 0.37
per square cm.83 If the surface area of the colorectum
is estimated as 1000 square cm, the mean number of
ACF per subject was 370. Plainly, the majority of ACF
cannot develop into macroscopic polyps, let alone
It has been suggested that dysplastic ACF are initiated
by APC mutation and may progress to adenoma, whereas
nondysplastic ACF are initiated by K-ras mutation, and
some may progress to hyperplastic polyps.85 An alternative view is that nondysplastic ACF may progress
through the advent of adenomatous transformation.86
This view is supported and extended in a large study of
dysplastic and nondysplastic ACF from subjects with and
without FAP.59 With respect to dysplastic ACF from
subjects without FAP, 0 of 15 (0%) showed APC mutation, 17 of 25 (68%) showed K-ras mutation, and 0 of
9 (0%) showed -catenin mutation. An identical mutational spectrum was found in nondysplastic ACF from
subjects without FAP.59 These data reinforce the concept
that APC mutation may not be the initiating event in
most examples of early sporadic colorectal neoplasia but
by no means discount the possibility of APC mutation
occurring as an early event (for example, driving the
progression of dysplastic ACF or the adenomatous transformation of nondysplastic ACF). Because around 65% of
colorectal cancers lack mutation of K-ras,33,34,87 this
change could not substitute for APC mutation as the
initiating event in most examples of sporadic colorectal
Despite the lack of evidence for mutation of APC or
-catenin in ACF, it may be noted that the latter may
nevertheless show altered -catenin expression by immunohistochemistry.88 The modulation is most apparent in
dysplastic ACF. Possible alternative mechanisms to mutation of APC or -catenin include dysregulated upstream
WNT proteins or induction of nitric oxide.89 GASTROENTEROLOGY Vol. 123, No. 3 Serrated Polyps of the Colorectum
The preceding data imply that neither K-ras nor
APC mutation is necessarily implicated in the initiation
of colorectal cancer. If this is the correct interpretation,
then the gap must be ﬁlled by an alternative mechanism.
Two serrated pathways of sporadic colorectal neoplasia
have been proposed, one culminating as MSI-H cancers90
and the second as MSI-L cancers.29,91 Below it will be
argued that a similar, two-step mechanism initiates both
serrated pathways: ﬁrst, the inhibition of apoptosis and,
second, the disruption of a DNA repair mechanism. In
these pathways, “serrated polyps” are conceived as a
morphologic continuum encompassing nondysplastic
ACF, hyperplastic polyps, admixed polyps (comprising
hyperplastic and dysplastic components), and serrated
adenomas.92 The dysplastic component of an admixed
polyp may include traditional adenoma or serrated adenoma. The link between hyperplastic polyps and serrated
adenomas has a surprisingly long history,93 although the
serrated adenomas as illustrated were labeled “villose”
adenomas.93 Both serrated pathways have been linked to
the epigenetic silencing of genes through the methylation of CpG islands within the promoter region.92,94
Both MSI95 and DNA methylation96,97 have been demonstrated at the early stage of ACF, and it is possible that
these alterations may serve as markers for ACF with
potential for progression. Such ACF are the likely precursors of hyperplastic polyps showing DNA methylation and MSI.90,91,97–99 MSI in hyperplastic polyps is
usually low level, but MSI-H has been described.98
Higher frequencies of both MSI-L and MSI-H have
been documented in admixed polyps and serrated adenomas.90,98 In one study, 58% and 25% of admixed
polyps were MSI-L and MSI-H, respectively.91 This relatively high frequency of MSI may be explained by the
fact that large and right-sided polyps were well represented in the series. Admixed polyps were initially conceived as collisions of typical hyperplastic and adenomatous polyps.100 At least some represent adenomas or
serrated adenomas developing within hyperplastic polyps. This has been inferred through the demonstration of
identical microsatellite mutations in the hyperplastic and
adenomatous components of the admixed polyps.91 Serrated adenomas show the architectural serration typical
of a hyperplastic polyp and the cytological characteristics
of an adenoma. Those occurring in the proximal colon
and appendix are often sessile or ﬂat and may be mistaken for a large hyperplastic polyp. The distinction
between a large hyperplastic polyp and serrated adenoma
may be extremely difﬁcult and warrants the development
of reproducible criteria and new biomarkers. Size ( 1 September 2002 cm), multiplicity ( 20), and proximal colonic location
are suggested features for achieving the distinction in the
meantime.101 Serrated adenomas occurring in the distal
colon are more likely to be polypoid with a tubulovillous
or villous architecture and may be misdiagnosed as tubulovillous or villous adenomas.102 In summary, the
spectrum of serrated neoplasia is broad and heterogeneous. Below, it is argued that serrated polyps serve as
the usual precursors of both MSI-H and MSI-L colorectal
cancers. Serrated Route to MSI-H Cancer
A key pathogenic mechanism in the pathogenesis
of sporadic MSI-H colorectal cancer is methylation and
loss of expression of the DNA mismatch repair gene
hMLH1. This loss of expression is observed in MSI-H
admixed polyps and serrated adenomas as well as in
MSI-H cancers.90,94,99 Loss of hMLH1 protein is occasionally observed in nondysplastic crypts within hyperplastic polyps,94 and methylation of hMLH1 has been
described within nondysplastic ACF.97 The silencing of
hMLH1 is an early event, and the association of this
change with dysplasia indicates that it is likely to serve
as a rate-limiting step driving the transition from hyperplasia to dysplasia. Analysis of extracted DNA from
microdissected dysplastic subclones in admixed polyps
and serrated adenomas with loss of hMLH1 protein has
demonstrated not only MSI-H but mutation of the same
target genes with repetitive coding sequences as are
mutated in MSI-H cancers, for example, TGF RII,
IGF2R, and BAX.90 K-ras mutation is uncommon in
proximally located serrated polyps of all types as well as
in MSI-H cancers.98,103 Methylation of a novel gene,
HPP1/TPEF, has been suggested as an early event.44
Common to both serrated polyps and sporadic MSI-H
cancer is a mixed mucinous phenotype combining both
gastric (MUC5AC) and intestinal (MUC2) mucin.104,105
There is little evidence implicating traditional adenomas
or the traditional mutational spectrum in the evolution
of sporadic MSI-H cancer.94,106 By contrast, traditional
adenomas serve as precursors of cancer in HNPCC,106,107
and adenoma removal results in prevention of cancer in
this syndrome.54 Cogent evidence for the MSI-H serrated
pathway to colorectal cancer34,35,90,91,94,98,99,104 –106,108 is
summarized in Table 2. Serrated Route to MSI-L Colorectal
The evidence supporting a serrated MSI-L pathway does not add up to the strong case for a serrated
MSI-H pathway, but a careful appraisal of the underlying COLORECTAL CANCER 867 Table 2. Evidence for Serrated Pathway to MSI-H Colorectal
Serrated adenomas adjacent to MSI-H cancers90,108
Molecular changes in serrated polyps (particularly in dysplastic
Loss of expression of hMLH190,94,99
Methylation of hMLH199
Mutation of TGF RII, IGF2R, and BAX 90
Hyperplastic polyps containing dysplastic subclones (admixed
polyps) in which the molecular changes associated with the
MSI-H pathway can be demonstrated90
Increased frequency of hyperplastic polyps in subjects with MSI-H
Similar mucinous phenotype in serrated polyps and MSI-H
Absence of traditional adenomas with MSI-H106
Absence of classical mutational spectrum in MSI-H cancers34,35 mechanisms fortiﬁes the argument in favor of the existence of a serrated MSI-L pathway. Features shared by
serrated polyps and MSI-L cancers include MSI-L,91 alterations at chromosome 1p,109 –111 a high frequency of
K-ras mutation,29,33,111 and a serrated architecture29 (but
in only a subset of MSI-L cancers). The MSI-L pathway is
more common in the left colon and rectum, the most
common site for hyperplastic polyps.29 The MSI-L pathway is associated with silencing of the DNA repair gene
O-6-methylguanine DNA methyltransferase (MGMT),
again by promoter methylation. Loss of expression of
MGMT has been observed in serrated polyps75 and methylation of MGMT within nondysplastic ACF.97 The
function of the “suicide” enzyme MGMT is to remove
promutagenic methyl adducts from guanine nucleotides.112 Several methylating compounds predispose to
the development of methylguanine adducts, including
4-(methylnitrosamine)-1-(3-pyridyl)-1-butatone (a component of tobacco smoke)113 and N-nitroso bile acid
conjugates.114 O-6-methylguanine adducts occur most
frequently in the normal mucosa of the distal colon and
K-ras mutation occurs in 50% MSI-L cancers vs. 30%
MSS cancers.29,33,111 TP53 and APC mutations occur at a
frequency similar to microsatellite stable cancer.33 In the
case of K-ras and TP53, the mutational spectrum associated with MGMT inactivation is narrow, mainly G:C
to A:T.116,117 This is explained by the failure to repair
methylguanine:thymine mismatches. The mismatches
arise because DNA polymerase misreads methylguanine as adenine. Therefore, during DNA replication,
thymine is mispaired with methylguanine. In a second
round of DNA replication, the mismatch may be converted to a stable mutation, for example, when adenine is
paired with thymine (hence giving rise to G to A tran- 868 JASS ET AL. sition).118 The sequence comprising methylation damage
mismatch following a round of DNA replication and
mutation following a second round of DNA replication
may be summarized as the following: G:C➝mG:
One of the mechanisms for repairing mG:T mismatches is through excision of thymine, perhaps by the
thymine DNA glycosylase MBD4.119 Involvement of
MBD4 in the MSI phenotype is suggested by the ﬁnding
of frameshift mutations in this DNA repair gene.120 The
single base gap that results from thymine excision is
detected by endonucleases that excise the remaining
sugar-phosphate residue as a prelude to “short-patch”
DNA repair.121 However, during the attempted repair of
DNA, the same mispairing may occur (polymerase
placing T opposite mG). The mismatch leads to a relative delay in further DNA extension, and this has been
hypothesized to predispose to sister chromatid exchange,
double strand breaks, chromosomal instability, and apoptosis. This mechanism of mismatch repair-dependent
toxicity to O-6-methylguanine lesions has been described as “futile cycles of repair.”122–124
Loss of expression of MGMT has been described in
traditional adenomas125 and within dysplastic subclones
in serrated polyps (Figure 1).75 The latter observation
implies that methylation of MGMT is not necessarily the
initiating event in the evolution of serrated polyps. Additionally, the correlation between MGMT methylation
with MSI-L status is not exact. As noted above, MGMT
methylation is associated with the high end of the range
of MSI-L status.75 Other DNA repair genes may be
implicated in the initiation of this pathway, including
the base excision repair genes methyl purine glycosylase
(MPG),126 MBD4 (see above), and polymerase .127,128
In the situation in which MGMT, and perhaps cooperating repair enzymes, are inactivated, the generation of
numerous mG:T mismatches would be the predicted
outcome. However, these may be repaired, at least to a
degree, by the hMSH2-hMSH6 heterodimer that recognizes mismatched methylguanine and instigates longpatch repair following the excision of a run of nucleotides
around the mismatch site.129 Neither hMSH2 nor
hMSH6 is a target for promoter methylation or is commonly mutated outside HNPCC.23 Because long-patch
excision occurs preferentially in the daughter strand containing the mismatched thymine, the mismatch may be
reintroduced, leading ultimately to destabilization of the
genome and apoptosis (futile cycles of repair). However,
apoptosis may also be induced more directly by damagesensing mismatch repair proteins.15,16 Increased longpatch repair could also generate deletion or insertion type GASTROENTEROLOGY Vol. 123, No. 3 mutations, particularly in the error-prone sites represented by the repetitive tracts of DNA found in microsatellite regions. Failure to repair such mismatches by a
repair system that is overloaded by the generation of
multiple mG:T mispairs may, in some instances, convert
unstable mG:T mismatches into stable frameshift mutations (detected as MSI-L).
It is likely that the silencing of DNA repair genes
results in an acquired resistance to DNA damage.122–124
In a DNA-damaging environment, such cells will neither stop cycling to repair the damage (through loss of
signaling to a cell cycle checkpoint) nor will they undergo apoptosis (through loss of apoptotic signaling).15
The cells will therefore acquire a selective growth advantage over their normal counterparts.130 DNA repair
genes may therefore serve as “gatekeepers” in place of
APC.16,131 The silencing of the “caretaking” function of
DNA repair genes may lead to the accumulation of
mutations that add to a cell’s growth advantage and
accelerate progression to cancer. Therefore, DNA repair
genes may serve as both “gatekeepers” and “caretakers,”
a combined role that is shared by APC.14 It is also
possible, however, that initiating events may be mediated through mutation of K-ras or the epigenetic silencing of genes implicated in the control of differentiation,
the cell cycle, or apoptosis, such as HPP1/TPEF (see
below), p16, or p14.44,77 In the following section, it is
suggested that inhibition of apoptosis is the most important initiating event in the serrated pathway. Top-down and Bottom-up Models
Based on microreconstruction studies in the colorectal mucosa of subjects with FAP, the earliest morphologic evidence of adenomatous neoplasia has been identiﬁed as a bud arising from the side of a parent crypt.132
The bud migrates (in concert with the normal epithelium of the parent crypt) while elongating to form a
short tubule composed of an immature and proliferating
epithelium. The unicryptal adenoma so formed assumes
a superﬁcial position in the mucosa where it “drops” from
the surface epithelium. Further budding from the base of
the neoplastic tubule results in lateral and downward
growth and the formation of microadenoma. This has
been described as the “top-down” model of adenomatous
morphogenesis.133 Repeated budding and branching
eventually give rise to an exophytic or polypoid growth.
In serrated polyps (hyperplastic polyps and serrated
adenomas), the proliferative zone remains in the lower
mucosal compartment. Cells mature and cease dividing
as they migrate into the upper crypt and surface epithelium.134 The term “bottom-up” may be used to describe Figure 1. Admixed serrated polyp comprising dysplastic subclone
with branching crypts that arises in a hyperplastic polyp. Nuclei in the
dysplastic subclone show loss of expression of MGMT (stromal cells
remain positive). Immunostaining for MGMT (clone MT3.1,
NeoMarkers). Figure 2. “Bottom-up” model (A ) in which the proliferative zone (P)
has extended beyond the limit of normal but grades into maturing (M)
surface epithelium. “Top-down” model (B) in which the surface as well
as crypt epithelium is proliferative (P ). A represents the topographical
organization of a serrated adenoma (note saw-tooth outline of crypt
epithelium). Here, the fundamental defect is inhibition of apoptosis
and failure of superﬁcial exfoliation of cells. B represents a microadenoma with adjacent normal crypts in which the fundamental defect is
dysregulated proliferation. Inhibition of apoptosis and dysregulated
proliferation are 2 sides of the same coin, both leading to an increase
in cell numbers. However, inhibition of apoptosis primes a lesion for
the subsequent loss of DNA repair proﬁciency and may, therefore, be
the more efﬁcient mechanism for initiating tumorigenesis. Figure 3. This Figure synthesizes the concept of alternative pathways to colorectal cancer based on different underlying mechanisms. Both
panels depict diminutive polyps (less than 2 mm) that represent contrasting pathways. The upper lesion was an incidental ﬁnding in a right
hemicolectomy for a sporadic MSI-H colorectal cancer. Despite its minute size, advanced molecular changes were demonstrated, including
methylation of both HPP1 and hMLH1, MSI-H, and loss of expression of hMLH1 and hPMS2 by immunohistochemistry (not shown). The lesion
shows high-grade epithelial dysplasia, yet a serrated crypt outline can still be discerned. The epithelium appears eosinophilic as a result of a
relatively abundant cytoplasm and the presence of clear or vesicular nuclei. The lower lesion is a microadenoma from a subject with familial
adenomatous polyposis and, therefore, implicates the gene APC. This is a “top-down” lesion in which the dysplastic glands are continuous with
surface epithelium. The epithelium appears blue overall because of the presence of crowded, pseudostratiﬁed, and hyperchromatic nuclei. The
left hand vertical arrow indicates that the lesions are representative of a biological continuum in which DNA methylation becomes increasingly
evident. The horizontal arrows indicate key rate-limiting genetic alterations that determine the principal pathways. 870 JASS ET AL. the retention of the normal topographical organization of
the basal proliferative and superﬁcial maturing compartments (Figure 2). Nevertheless, serrated adenomas and
some hyperplastic polyps may show aberrant extension of
the proliferative zone.134
Dysregulation of the WNT/wingless pathway through
mutation of APC or -catenin is the basis of the asymmetrical budding and proliferative abnormality that
characterizes the “top-down” model of colorectal neoplasia.133,135 By contrast, “bottom-up” neoplasia will implicate the dysregulation of apoptosis. Down-regulation of
the apoptosis receptor Fas (CD95) has been demonstrated
in serrated polyps,136 and K-ras mutation, found in a
subset of hyperplastic polyps, is known to inhibit Fas
expression.137 It is likely that the speciﬁc underlying
defect in “bottom-up” or serrated neoplasia centers upon
a type of apoptosis that is peculiar to epithelial surfaces
and is triggered by epithelial exfoliation. This type of
apoptosis has been termed anoikis.138,139 Different mechanisms may disrupt anoikis, including mutation of Kras139 and inactivation (by methylation) of the putative
antiadhesion gene HPP1/TPEF.44 It has been known for
years that senescent cells accumulate within the epithelial surface of hyperplastic polyps.140 More recently, it
has been shown that these cells contain denatured cytokeratin 18, a degenerative change associated with programmed cell death.94 The preceding observations underlie the inhibition of surface epithelial shedding, and
the resultant cellular accumulation leads in turn to the
characteristic buckling of crypt epithelium known as
It may be speculated that disruption of apoptosis
serves a permissive role with respect to the subsequent
inactivation of a DNA repair mechanism. This would
explain why the immunohistochemical demonstration of
MGMT or hMLH1 extinction occurs in serrated polyps
but not in normal mucosa (Figure 1). In the normal
colorectal mucosa of subjects with HNPCC, one might
expect to observe scattered single crypts with loss of
expression of the appropriate mismatch repair protein
(through inactivation of the wild-type copy of the gene),
yet this has not been described. Conceivably, loss of
DNA repair might drive rapid neoplastic progression. If
this were correct, one would expect to observe numerous
neoplasms in subjects with HNPCC, as is the case in
FAP. It may be inferred, therefore, that the disruption of
DNA repair is generally followed by extensive DNA
damage that triggers apoptosis. Loss of DNA repair
proﬁciency may be tolerated, however, if it occurs in
lesions previously primed with impaired apoptosis. The
fact that HPP1/TPEF is methylated and shows extinc- GASTROENTEROLOGY Vol. 123, No. 3 tion of expression in entire hyperplastic polyps and not in
subclones (as may be observed in the case of MGMT and
hMLH1) means that the alteration occurs early.44 However, the suggestion that HPP1/TPEF is implicated in
the regulation of apoptosis is based on analogy with
genes of similar structure and not on functional data.44 Hyperplastic Polyposis: A Model for
Sporadic Colorectal Cancer?
If FAP is not an appropriate model for explaining
all pathways to sporadic colorectal cancer, is there an
alternative familial condition that could serve such a
role? Hyperplastic polyposis presents a plausible model
for the following reasons: (1) Polyps in this condition
may show MSI and silencing of relevant DNA repair
genes including hMLH1.90,98,99 (2) CpG island methylation is demonstrated in DNA extracted from hyperplastic polyps in a subset of subjects with hyperplastic polyposis.97 The ﬁnding of methylation is concordant
within multiple polyps in such cases,97 whereas discordant ﬁndings occur in subjects with multiple adenomas.125 (3) The requisite plasticity in methylator pathways is evident in hyperplastic polyposis in which all
types of epithelial polyp may occur (hyperplastic, admixed, serrated adenoma, and traditional adenoma), and
cancers may be MSI-H, MSI-L, or MSS (even within the
same subject).90,142 (4) The condition hyperplastic polyposis may be familial.98,143
The genetic basis for hyperplastic polyposis is unknown. One mechanism could be a germline mutation or
polymorphism conferring an increased tendency to DNA
damage and/or methylation within colorectal epithelial
cells. Association studies employing candidate genes
known to play a role in the control of DNA methylation
and demethylation would be a reasonable strategy for
identifying genes responsible for hyperplastic polyposis
and attenuated counterparts of this condition. Candidates would be expected to synergize with a polymorphism in the methylene tetrahydrofolate reductase gene
that is associated with MSI-H colorectal cancer.144 Conclusion and Clinical Implications
A signiﬁcant proportion of colorectal cancer may
not be initiated by mutation of APC, as is generally
supposed, but through the epigenetic silencing of alternative genes implicated in apoptosis and DNA repair
mechanisms. Epithelial hyperplasia and serration are
early morphological changes within this alternative pathway. Although not initiating this pathway, APC mutation may occur early and generate subclones with an
adenomatous morphology. However, APC mutation is September 2002 not observed in most sporadic MSI-H cancers and is
found in only a subset of other types of colorectal cancer.
The evolutionary paradigm placing APC mutation at the
point of initiation of colorectal cancer applies in the
speciﬁc case of FAP and probably to other adenomaprone individuals. Uncritical acceptance of the inevitable
role of APC mutation in the initiation of colorectal
neoplasia will impede new understanding and the development of effective chemopreventive strategies that target the earliest molecular events. If it is accepted that
APC mutation is not an inevitable event in colorectal
cancer, either at inception or during subsequent evolution, then APC mutation may serve as a prognostic
marker. The lowest frequency of APC mutation is in the
good prognosis MSI-H subset of colorectal cancer. More
interestingly, -catenin activation is pivotal in driving
signalling pathways that culminate in inﬁltrative and
budding growth at the advancing tumor margin.145 Tumor budding is now recognized as an independent prognostic factor in colorectal cancer, second only to lymph
node spread.146 APC mutation is the main mechanism
for activating -catenin and may serve as the underlying
basis for both tumor budding and adverse prognosis in
Neoplastic pathways may be classiﬁed on the basis of
genetic instability, DNA methylation, morphology, and
whether the earliest changes implicate the control of
proliferation or apoptosis. These approaches are synthesized in Figure 3. The lower panel shows a “top-down”
microadenoma from a subject with FAP and, therefore,
implicating APC inactivation. The upper panel shows a
small but severely dysplastic MSI-H serrated adenoma in
which there is methylation and loss of expression of both
HPP1/TPEF and hMLH1 (not illustrated). The proportion of cancers in the “methylator” group may increase if
additional DNA repair genes are shown to be rendered
pathogenic by promoter region methylation.
Based on experience with HNPCC, it is possible that
pathways driven by the inactivation of DNA repair genes
will acquire the requisite genomic alterations in relatively quick succession, and progression during the precancerous stage of growth will be correspondingly rapid.
A rapid transition from normal to cancer will apply not
only to the serrated pathway but also to some de novo
cancers and cancers developing in ﬂat adenomas. As is
the case in HNPCC, cancers that develop within a short
time frame may not be any more aggressive (in terms of
such measures as spread and prognosis) than cancers
having a long precancerous phase. It is likely that the
clinical and pathologic diversity that is now apparent in
the spectrum of colorectal neoplasia will be explained by COLORECTAL CANCER 871 the order in which genes are inactivated through the
epigenetic mechanism of DNA methylation. These alterations may occur within morphologically normal
crypts. Further understanding of the early processes, and
speciﬁcally the mechanisms that would accelerate tumor
progression, will occur through meticulous clinical, morphologic, and molecular correlations conducted on
minute lesions, microscopic lesions, and normal-appearing crypts. By understanding these mechanisms, it is
hoped that subjects prone to rapidly evolving forms of
colorectal neoplasia will be identiﬁed and offered effective preventive measures.
The synthesis of molecular and morphologic observations described in this review has stripped the hyperplastic polyp of its long-presumed innocence. However, the
fact that the vast majority of hyperplastic polyps will
never progress to cancer has not altered. It is impractical
to advocate the removal of every minute hyperplastic
lesion. On the other hand, more attention might be
given to subjects with “high-risk” hyperplastic polyps.
High-risk features would include multiplicity (more
than 20), size (greater than 10 mm), proximal location,
associated polyps with dysplasia, and a family history of
colorectal cancer. New diagnostic criteria and markers
are required to distinguish innocent hyperplastic polyps
from their serrated counterparts with a malignant potential that belies their deceptively bland morphology.
In summary, this review has shown that a single
stepwise model for the evolution of colorectal cancer that
is initiated by APC mutation may not only explain
relatively little, but may impede new understanding that
is relevant to the prevention, screening, prognosis, and
treatment of colorectal cancer. References
1. Bodmer WF, Bailey CJ, Bodmer J, Bussey HJR, Lucibello FC,
Rider SH, Scambler P, Sheer D, Solomon E, Spurr NK. Localization of the gene for familial adenomatous polyposis on chromosome 5. Nature 1987;328:614 – 616.
2. Leppert M, Dobbs M, Scambler P, O’Connell P, Nakamura Y,
Stauffer D, Woodward S, Burt R, Hughes J, Gardner E, Lathrop
M, Wasmuth J, Lalouel J-M, White R. The gene for familial
polyposis maps to the long arm of chromosome 5. Science
3. Groden J, Thliveris A, Samowitz W, Carlson M, Gelbert L, Albertsen H, Joslyn G, Stevens J, Spirio L, Robertson M, Sargeant L,
Krapcho K, Wolff E, Burt R, Hughes JP, Warrington J, McPherson
J, Wasmuth J, Le Paslier D, Abderrahim H, Cohen D, Leppert M,
White R. Identiﬁcation and characterization of the familial adenomatous polyposis gene. Cell 1991;66:589 600.
4. Kinzler KW, Nilbert MC, Su L-K, Vogelstein B, Bryan TM, Levy DB,
Smith KJ, Preisinger AC, Hedge P, McKechnie D, Finniear R,
Markham A, Groffen J, Boguski MS, Altschul SF, Horii A, Ando H,
Miyoshi Y, Miki Y, Nishisho I, Nakamura Y. Identiﬁcation of FAP
locus genes from chromosome 5q21. Science 1991;253:661–
5. Solomon E, Voss R, Hall V, Bodmer WF, Jass JR, Jeffreys AJ, 872 6. 7.
8. 9. 10. 11. 12. 13. 14. 15.
16. 17. 18. 19.
20. 21. JASS ET AL. Lucibello FC, Patel I, Rider SH. Chromosome 5 allele loss in
human colorectal carcinomas. Nature 1987;828:616 – 619.
Vogelstein B, Fearon ER, Hamilton SR, Kern SE, Preisinger AC,
Leppert M, Nakamura Y, White R, Smits AMM, Bos JL. Genetic
alterations during colorectal-tumor development. N Engl J Med
Kinzler KW, Vogelstein B. Lessons from hereditary colorectal
cancer. Cell 1996;87:159 –170.
Fishel R, Lescoe MK, Rao MRS, Copeland NG, Jenkins NA,
Garber J, Kane M, Kolodner R. The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis
colon cancer. Cell 1993;75:1027–1038.
Leach FS, Nicolaides NC, Papadopoulos N, Liu B, Jen J, Parsons
R, Peltomaki P, Sistonen P, Aaltonen LA, Nystrom-Lahti M, Guan
X-Y, Zhang J, Meltzer PS, Yu J-W, Kao F-T, Chen DJ, Cerosaletti
KM, Fournier REK, Todd S, Lewis T, Leach RJ, Naylor SL, Weissenbach J, Mecklin J-P, Jarvinen H, Petersen GM, Hamilton SR,
Green J, Jass J, Watson P, Lynch HT, Trent JM, de la Chapelle A,
Kinzler KW, Vogelstein B. Mutations of a mutS homolog in
hereditary nonpolyposis colorectal cancer. Cell 1993;75:1215–
Bronner CE, Baker SM, Morrison PT, Warren G, Smith LG, Lescoe MK, Kane M, Earabino C, Lipford J, Lindblom A, Tannergard
P, Bollag RJ, Godwin AR, Ward DC, Nordenskjold M, Fishel R,
Kolodner R, Liskay RM. Mutation in the DNA mismatch repair
gene homologue hMLH1 is associated with hereditary nonpolyposis colon cancer. Nature 1994;368:258 –261.
Papadopoulos N, Nicolaides NC, Wei Y-F, Ruben SM, Carter KC,
Rosen CA, Haseltine WA, Fleischmann RD, Fraser CM, Adams
MD, Venter JC, Hamilton SR, Petersen GM, Watson P, Lynch HT,
Peltomaki P, Mecklin J-P, de la Chapelle A, Kinzler KW, Vo¨
gelstein B. Mutation of a mutL homolog in hereditary colon
cancer. Science 1994;263:1625–1629.
Papadopoulos N, Nicolaides NC, Liu B, Parsons R, Lengauer C,
Palombo F, D’Arrigo A, Markowitz S, Willson JKV, Kinzler KW,
Jiricny J, Vogelstein B. Mutations of GTBP in genetically unstable
cells. Science 1995;368:1915–1917.
Lynch HT, Smyrk T, Jass JR. Hereditary nonpolyposis colorectal
cancer and colonic adenomas: aggressive adenomas? Semin
Surg Oncol 1995;11:406 – 410.
Fodde R, Kuipers J, Rosenberg C, Smits R, Kielman M, Gaspar
C, van Es JH, Breukel C, Wiegant J, Giles RH, Clevers H. Mutations in the APC tumor suppressor gene cause chromosomal
instability. Nat Cell Biol 2001;3:433– 438.
Fishel R. Mismatch repair, molecular switches, and signal transduction. Genes Dev 1998;12:2096 –2101.
Fishel R. The selection for mismatch repair defects in hereditary
nonpolyposis colorectal cancer: revising the mutator hypothesis. Cancer Res 2001;61:7369 –7374.
Ionov Y, Peinado MA, Malkhosyan S, Shibata D, Perucho M.
Ubiquitous somatic mutations in simple repeated sequences
reveal a new mechanism for colonic carcinogenesis. Nature
Aaltonen LA, Peltomaki PS, Leach FS, Sistonen L, Pylkkanen J-P,
Mecklin H, Jarvinen SM, Powell J, Jen SR, Hamilton GM, Pe¨
tersen KW, Kinzler B, Vogelstein B, de la Chapelle A. Clues to
the pathogenesis of familial colorectal cancer. Science 1993;
Thibodeau SN, Bren G, Schaid D. Microsatellite instability in
cancer of the proximal colon. Science 1993;260:816 – 819.
Thibodeau SN, French AJ, Cunningham JM, Tester D, Burgart LJ,
Roche PC, McDonnell SK, Schaid DJ, Vockley CW, Michels VV,
Farr GHJ, O’Connell MJ. Microsatellite instability in colorectal
cancer: different mutator phenotypes and the principle involvement of hMLH1. Cancer Res 1998;58:1713–1718.
Kane MF, Loda M, Gaida GM, Lipman J, Mishra R, Goldman H,
Jessup JM, Kolodner R. Methylation of the hMLH1 promoter GASTROENTEROLOGY Vol. 123, No. 3 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. correlates with lack of expression of hMLH1 in sporadic colon
tumors and mismatch repair-defective human tumor cell lines.
Cancer Res 1997;57:808 – 811.
Herman JG, Umar A, Polyak K, Graff JR, Ahuja N, Issa JPJ,
Markowitz S, Von Willson JK, Hamilton SR, Kinzler KW, Kane
MF, Kolodner RD, Vogelstein B, Kunkel TA, Baylin SB. Incidence
and functional consequences of hMLH1 promoter hypermethylation in colorectal carcinoma. Proc Natl Acad Sci U S A 1998;
95:6870 – 6875.
Cunningham JM, Christensen ER, Tester DJ, Kim C-Y, Roche PC,
Burgart LJ, Thibodeau SN. Hypermethylation of the hMLH1 promoter in colon cancer with microsatellite instability. Cancer Res
Malkhosyan SR, Yamamoto H, Piao Z, Perucho M. Late onset
and high incidence of colon cancer of the mutator phenotype
with hypermethylated hMLH1 gene in women. Gastroenterology
Wright CM, Dent OF, Barker M, Newland RC, Chapuis PH, Bokey
EL, Young JP, Leggett BA, Jass JR, Macdonald GA. The prognostic signiﬁcance of extensive microsatellite instability in sporadic
clinicopathological stage C colorectal cancer. Br J Surg 2000;
Kim H, Jen J, Vogelstein B, Hamilton SR. Clinical and pathological characteristics of sporadic colorectal carcinomas with DNA
replication errors in microsatellite sequences. Am J Pathol
Ruschoff J, Dietmaier W, Luttges J, Seitz G, Bocker T, Zirngibl H,
Schlegel J, Schackert HK, Jauch KW, Hofstaedter F. Poorly
differentiated colonic adenocarcinoma, medullary type: clinical,
phenotypic, and molecular characteristics. Am J Pathol 1997;
Jass JR, Do K-A, Simms LA, Iino H, Wynter C, Pillay SP, Searle
J, Radford-Smith G, Young J, Leggett B. Morphology of sporadic
colorectal cancer with DNA replication errors. Gut 1998;42:
Jass JR, Biden KG, Cummings M, Simms LA, Walsh M, Schoch
E, Meltzer S, Wright C, Searle J, Young J, Leggett BA. Characterisation of a subtype of colorectal cancer combining features
of the suppressor and mild mutator pathways. J Clin Pathol
Fujiwara T, Stoker JM, Watanabe T, Rashid A, Longo P, Eshleman JR, Booker S, Lynch HT, Jass JR, Green JS, Kim H, Jen J,
Vogelstein B, Hamilton SR. Accumulated clonal genetic alterations in familial and sporadic colorectal carcinomas with widespread instability in microsatellite sequences. Am J Pathol
Samowitz WS, Holden JA, Curtin K, Edwards SL, Walker AR, Lin
HA, Robertson MA, Nichols MF, Gruenthal KM, Lynch BJ, Leppert MF, Slattery ML. Inverse relationship between microsatellite instability and K-ras and p53 gene alterations in colon
cancer. Am J Pathol 2001;158:1517–1524.
Lothe RA, Peltomaki P, Meling GI, Aaltonen LA, Nystrom-Lahti M,
Pylkkanen L, Heimdal K, Andersen TI, Moller P, Regnum TO.
Genomic instability in colorectal cancer: relationship to clinicopathological variables and family history. Cancer Res 1993;53:
Konishi M, Kikuchi-Yanoshita R, Tanaka K, Muraoka M, Onda A,
Okumura Y, Kishi N, Iwama T, Mori T, Koike M, Ushio K, Chiba
M, Nomizu S, Konishi F, Utsunomiya J, Miyaki M. Molecular
nature of colon tumors in hereditary nonpolyposis colon cancer,
familial polyposis, and sporadic colon cancer. Gastroenterology
Salahshor S, Kressner U, Påhlman L, Glimelius B, Lindmark G,
Lindblom A. Colorectal cancer with and without microsatellite
instability involves different genes. Genes Chromosom Cancer
Olschwang S, Hamelin R, Laurent-Puig P, Thuille B, De Rycke Y, September 2002 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. Li YJ, Muzeau F, Girodet J, Salmon RJ, Thomas G. Alternative
genetic pathways in colorectal carcinogenesis. Proc Natl Acad
Sci U S A 1997;94:12122–12127.
Simms LA, Radford-Smith G, Biden KG, Buttenshaw R, Cummings M, Jass JR, Young J, Meltzer SJ, Leggett BA. Reciprocal
relationship between the tumor suppressors p53 and BAX in
primary colorectal cancers. Oncogene 1998;17:2003–2008.
Markowitz S, Wang J, Myeroff L, Parsons R, Sun L, Lutterbaugh
J, Fan RS, Zborowska E, Kinzler KW, Vogelstein B, Brattain M,
Willson JKV. Inactivation of the type II TGF- receptor in colon
cancer cells with microsatellite instability. Science 1995;268:
Souza RF, Appel R, Yin J, Wang S, Smolinski KN, Abraham JM,
Zou TT, Shi YQ, Lei J, Cottrell J, Cymes K, Biden K, Simms L,
Leggett B, Lynch PM, Fraizer M, Powell SM, Harpaz N, Sugimura
H, Young J, Meltzer SJ. The insulin-like growth factor II receptor
gene is a target of microsatellite instability in human gastrointestinal tumours. Nat Genet 1996;14:255–257.
Yamamoto H, Sawai H, Weber TK, Rodriguez-Bigas MA, Perucho
M. Somatic frameshift mutations in DNA mismatch repair and
proapoptosis genes in hereditary nonpolyposis colorectal cancer. Cancer Res 1998;58:997–1003.
Ahuja N, Mohan AL, Li Q, Stokler JM, Herman JG, Hamilton SR,
Baylin SB, Issa J-P. Association between cPG island methylation
and microsatellite instability in colorectal cancer. Cancer Res
Issa J-PJ, Ottaviano YL, Celano P, Hamilton SR, Davidson NE,
Baylin SB. Methylation of the oestrogen receptor CpG island
links ageing and neoplasia in human colon. Nat Genet 1994;7:
Toyota M, Ahuja N, Ohe-Toyota M, Herman JG, Baylin SB, Issa
J-PJ. CpG island methylator phenotype in colorectal cancer. Proc
Natl Acad Sci U S A 1999;96:8681– 8686.
Robertson KD, Jones PA. The human ARF cell cycle regulatory
gene promoter is a CpG island which can be silenced by DNA
methylation and down-regulated by wild-type p53. Mol Cell Biol
Young JP, Biden KG, Simms LA, Huggard P, Karamatic R, Eyre
HJ, Sutherland GR, Herath N, Barker MA, Anderson GJ, Fitzpatrick D, Ramm GA, Jass JR, Leggett B. HPP1: a transmembrane protein commonly methylated in colorectal polyps and
cancers. Proc Natl Acad Sci U S A 2001;98:265–270.
Esteller M, Hamilton SR, Burger PC, Baylin SB, Herman JG.
Inactivation of the DNA repair gene O 6-methylguanine-DNA
methyltransferase by promoter hypermethylation is a common
event in primary human neoplasia. Cancer Res 1999;59:793–
Xiong Z, Wu AH, Bender CM, Tsao JL, Blake C, Shibata D, Jones
PA, Yu MC, Ross RK, Laird PW. Mismatch repair deﬁciency and
CpG island hypermethylation in sporadic colon adenocarcinomas. Cancer Epidemiol Biomarkers Prev 2001;10:799 – 803.
Toyota M, Shen L, Ohe-Toyota M, Hamilton SR, Sinicrope FA,
Issa J-PJ. Aberrant methylation of the Cyclooxygenase 2 CpG
island in colorectal tumors. Cancer Res 2000;60:4044 – 4048.
Grady WM, Willis J, Guilford PJ, Dunbier AK, Toro TT, Lynch H,
Wiesner G, Ferguson K, Eng C, Park JG, Kim SJ, Markowitz S.
Methylation of the CDH1 promoter as the second genetic hit in
hereditary diffuse gastric cancer. Nat Genet 2000;26:16 –17.
Du Y, Carling T, Fang W, Piao Z, Sheu JC, Huang S. Hypermethylation in human cancers of the RIZ1 tumor suppressor gene, a
member of a histone/protein methyltransferase superfamily.
Cancer Res 2001;61:8094 – 8099.
Burbee DG, Forgacs E, Zochbauer-Muller S, Shivakumar L, Fong
K, Gao B, Randle D, Kondo M, Virmani A, Bader S, Sekido Y,
Latif F, Milchgrub S, Toyooka S, Gazdar AF, Lerman MI,
Zabarovsky E, White M, Minna JD. Epigenetic inactivation of COLORECTAL CANCER 51. 52.
53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 873 RASSF1A in lung and breast cancers and malignant phenotype
suppression. J Natl Cancer Inst 2001;93:691– 699.
Tomlinson I, Ilyas M, Johnson V, Davies A, Clark G, Talbot I,
Bodmer W. A comparison of the genetic pathways involved in
the pathogenesis of three types of colorectal cancer. J Pathol
Muto T, Bussey HJR, Morson BC. The evolution of cancer of the
rectum. Cancer 1975;36:2251–2276.
Novelli MR, Williamson JA, Tomlinson IPM, Elia G, Hodgson SV,
Talbot IC, Bodmer WF, Wright NA. Polyclonal origin of colonic
adenomas in an XO/XY patient with FAP. Science 1996;272:
Jarvinen HJ, Aarnio M, Mustonen H, Aktan-Collan K, Aaltonen
LA, Peltomaki P, De La Chapelle A, Mecklin J-P. Controlled
15-year trial on screening for colorectal cancer in families with
hereditary nonpolyposis colorectal cancer. Gastroenterology
2000;118:829 – 834.
Vasen HFA, Nagengast FM, Meera Khan P. Interval cancers in
hereditary non-polyposis colorectal cancer (Lynch syndrome).
Korinek V, Barker N, Morin PJ, van Wichen D, de Weger R,
Kinzler KW, Vogelstein B, Clevers H. Constitutive transcriptional
activation by a -catenin-Tcf complex in APC / colon carcinoma. Science 1997;275:1784 –1787.
Rees M, Leigh SEA, Delhanty JDA, Jass JR. Chromosome 5
allele loss in familial and sporadic colorectal adenomas. Br J
Ichii S, Horii A, Nakatsuru S, Furyama J, Utsunomiya J, Nakamura Y. Inactivitation of both APC alleles in an early stage of
colon adenomas in a patient with familial adenomatous polyposis (FAP). Hum Mol Genet 1992;1:387–390.
Takayama T, Ohi M, Hayashi T, Miyanishi K, Nobuoka A, Nakajima T, Satoh T, Takimoto R, Kato J, Sakamaki S, Niitsu Y.
Analysis of K-ras, APC, and -catenin in aberrant crypt foci in
sporadic adenoma, cancer, and familial adenomatous polyposis. Gastroenterology 2001;121:599 – 611.
Umetani N, Sasaki S, Masaki T, Watanabe T, Matsuda K, Muto
T. Involvement of APC and K-ras mutation in non-polypoid colorectal tumorigenesis. Br J Cancer 2000;82:9 –15.
Kim JC, Koo KH, Lee DH, Roh SA, Kim HC, Yu CS, Kang GH.
Mutations at the APC exon 15 in the colorectal neoplastic
tissues of serial array. Int J Colorectal Dis 2001;16:102–107.
De Benedetti L, Sciallero S, Gismondi V, James R, Baﬁco A,
Biticchi R, Masetti E, Bonelli L, Heouaine A, Picasso M, Groden
J, Robertson M, Risio M, Caprilli R, Bruzzi P, White RL, Aste H,
Santi L, Varesco L, Ferrara GB. Association of APC gene mutations and histological characteristics of colorectal adenomas.
Cancer Res 1994;54:3553–3556.
Mulkens J, Poncin J, Arends JW, De Goeij AF. APC mutations in
human colorectal adenomas: analysis of the mutation cluster
region with temperature gradient gel electrophoresis and clinicopathological features. J Pathol 1998;185:360 –365.
Zauber NP, Sabbath-Solitare M, Marotta SP, Bishop DT. K-ras
mutation and loss of heterozygosity of the adenomatous polyposis coli gene in patients with colorectal adenomas with in situ
carcinoma. Cancer 1999;86:31–36.
Huang J, Papadopoulos N, McKinley AJ, Farrington SM, Curtis
LJ, Wyllie AH, Zheng S, Willson JK, Markowitz SD, Morin P,
Kinzler KW, Vogelstein B, Dunlop MG. APC mutations in colorectal tumors with mismatch repair deﬁciency. Proc Natl Acad
Sci U S A 1996;93:9049 –9054.
Kapiteijn E, Liefers GJ, Los LC, Kranenbarg EK, Hermans J,
Tollenaar RA, Moriya Y, van de Velde CJ, van Krieken JH. Mechanisms of oncogenesis in colon versus rectal cancer. J Pathol
Miyaki M, Iijima T, Kimura J, Yasuno M, Mori T, Hayashi Y, Koike
M, Shitara N, Iwama T, Kuroki T. Frequent mutation of -Catenin 874 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. JASS ET AL. and APC genes in primary colorectal tumors from patients with
hereditary nonpolyposis colorectal cancer. Cancer Res 1999;
59:4506 – 4509.
Mirabelli-Primdahl L, Gryfe R, Kim H, Millar A, Luceri C, Dale D,
Holowaty E, Bapat B, Gallinger S, Redston M. -Catenin mutations are speciﬁc for colorectal carcinomas with microsatellite
instability but occur in endometrial carcinomas irrespective of
mutator pathway. Cancer Res 1999;59:3346 –3351.
Liu W, Dong X, Mai M, Seelan RS, Taniguchi K, Krishnadath KK,
Halling KC, Cunningham JM, Qian C, Christensen E, Roche PC,
Smith DI, Thibodeau SN. Mutations in AXIN2 cause colorectal
cancer with defective mismatch repair by activating -catenin/
TCF signalling. Nat Genet 2000;26:146 –147.
Esteller M, Sparks A, Toyota M, Sanchez-Cespedes M, Capella
G, Peinado MA, Gonzalez S, Tarafa G, Sidransky D, Meltzer SJ,
Baylin SB, Herman JG. Analysis of adenomatous polyposis coli
promoter hypermethylation in human cancer. Cancer Res 2000;
60:4366 – 4371.
Wong NACS, Morris RG, McCondochie A, Bader S, Jodrell DI,
Harrison DJ. Cyclin D1 overexpression in colorectal carcinoma in
vivo is dependent on -catenin protein dysregulation, but not
k-ras mutation. J Pathol 2002;197:128 –135.
Dietmaier W, Wallinger S, Bocker T, Kullmann F, Fishel R,
Ruschoff J. Diagnostic microsatellite instability: Deﬁnition and
correlation with mismatch repair protein expression. Cancer Res
1997;57:4749 – 4756.
Boland CR, Thibodeau SN, Hamilton SR, Sidransky D, Eshleman
JR, Burt RW, Meltzer SJ, Rodriguez-Bigas MA, Fodde R, Ranzani
GN, Srivastava S. A National Cancer Institute Workshop on
microsatellite instability for cancer detection and familial predisposition: Development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer
Res 1998;58:5248 –5257.
Cunningham JM, Kim CY, Christensen ER, Tester DJ, Parc Y,
Burgart LJ, Halling KC, McDonnell SK, Schaid DJ, Walsh Vockley
C, Kubly V, Nelson H, Michels VV, Thibodeau SN. The frequency
of hereditary defective mismatch repair in a prospective series
of unselected colorectal carcinomas. Am J Hum Genet 2001;
Whitehall V, Walsh MD, Young J, Leggett BA, Jass JR. Methylation of 0-6-methylguanine DNA methyltransferase characterises a subset of colorectal cancer with low level DNA microsatellite instability. Cancer Res 2001;61:827– 830.
Halford S, Sasieni P, Rowan A, Wasan H, Bodmer W, Talbot I,
Hawkins N, Ward R, Tomlinson I. Low-level microsatellite instability occurs in most colorectal cancers and is a nonrandomly
distributed quantitative trait. Cancer Res 2002;62:53–57.
Esteller M, Tortola S, Toyota M, Capella G, Peinado MA, Baylin
SB, Herman JG. Hypermethylation-associated inactivation of
p14ARF is independent of p16INK4A methylation and p53 mutational status. Cancer Res 2000;60:129 –133.
Liang G, Robertson KD, Talmadge C, Sumegi J, Jones PA. The
gene for a novel transmembrane protein containing epidermal
growth factor and follistatin domains is frequently hypermethylated in human tumor cells. Cancer Res 2000;60:4907– 4912.
Pfeifer GP, Steigerwald SD, Hansen RS, Gartler SM, Riggs AD.
Polymerase chain reaction-aided genomic sequencing of an X
chromosome-linked CpG island: methylation patterns suggest
clonal inheritance, CpG site autonomy, and an explanation of
activity state stability. Proc Natl Acad Sci U S A 1990;87:8252–
Yatabe Y, Tavare S, Shibata D. Investigating stem cells in
human colon by using methylation patterns. Proc Natl Acad Sci
U S A 2001;98:10839 –10844.
Bird RP. Observation and quantiﬁcation of aberrant crypts in the
murine colon treated with a colon carcinogen: preliminary ﬁndings. Cancer Lett 1987;37:147–151. GASTROENTEROLOGY Vol. 123, No. 3 82. Bird RP, McLellan EA, Bruce WR. Aberrant crypts, putative precancerous lesions, in the study of the role of diet in the aetiology of colon cancer. Cancer Surv 1989;8:189 –200.
83. Roncucci L, Stamp D, Medline A, Cullen JB, Bruce WR. Identiﬁcation and quantiﬁcation of aberrant crypt foci and microadenomas in the human colon. Hum Pathol 1991;22:287–294.
84. Roncucci L, Medline A, Bruce WR. Classiﬁcation of aberrant
crypt foci and microadenomas in human colon. Cancer Epidemiol Biomarkers Prev 1991;1:57– 60.
85. Jen J, Powell SM, Papadopoulos N, Smith KJ, Hamilton SR,
Vogelstein B, Kinzler KW. Molecular determinants of dysplasia
in colorectal lesions. Cancer Res 1994;54:5523–5526.
86. Nascimbeni R, Villanacci V, Mariani PP, Di Betta E, Ghirardi M,
Donato F, Salerni B. Aberrant crypt foci in the human colon. Am J
Surg Pathol 1999;23:1256 –1263.
87. Slattery ML, Curtin K, Anderson K, Ma KN, Ballard L, Edwards S,
Schaffer D, Potter J, Leppert M, Samowitz WS. Associations
between cigarette smoking, lifestyle factors, and microsatellite
instability in colon tumors. J Natl Cancer Inst 2000;92:1831–
88. Hao XP, Pretlow TG, Rao JS, Pretlow TP. -Catenin expression is
altered in human colonic aberrant crypt foci. Cancer Res 2001;
89. Mei JM, Hord NG, Winterstein DF, Donald SP, Phang JM. Differential formation of beta-catenin/lymphoid enhancer factor-1
DNA binding complex induced by nitric oxide in mouse colonic
epithelial cells differing in adenomatous polyposis coli (Apc)
genotype. Cancer Res 2000;60:3379 –3383.
90. Jass JR, Iino H, Ruszkiewicz A, Painter D, Solomon MJ, Koorey
DJ, Cohn D, Furlong KL, Walsh MD, Palazzo J, Bocker Edmonston T, Fishel R, Young J, Leggett BA. Neoplastic progression
occurs through mutator pathways in hyperplastic polyposis of
the colorectum. Gut 2000;47:43– 49.
91. Iino H, Jass JR, Simms LA, Young J, Leggett B, Ajioka Y, Watanabe H. DNA microsatellite instability in hyperplastic polyps,
serrated adenomas, and mixed polyps: a mild mutator pathway
for colorectal cancer? J Clin Pathol 1999;52:5–9.
92. Jass JR. Serrated route to colorectal cancer: back street or
super highway? J Pathol 2001;193:283–285.
93. Goldman H, Ming S, Hickock DF. Nature and signiﬁcance of
hyperplastic polyps of the human colon. Arch Pathol 1970;89:
94. Jass JR, Young J, Leggett BA. Hyperplastic polyps and DNA
microsatellite unstable cancers of the colorectum. Histopathology 2000;37:295–301.
95. Heinen CD, Shivapurkar N, Tang Z, Groden J, Alabaster O.
Microsatellite instability in aberrant crypt foci from human colons. Cancer Res 1996;56:5339 –5341.
96. Sakurazawa N, Tanaka N, Onda M, Esumi H. Instability of X
chromosome methylation in aberrant crypt foci of the human
colon. Cancer Res 2000;60:3165–3169.
97. Chan AO-O, Broaddus RR, Houlihan PS, Issa J-PJ, Hamilton SR,
Rashid A. CpG island methylation in aberrant crypt foci of the
colorectum. Am J Pathol 2002;160:1823–1830.
98. Rashid A, Houlihan S, Booker S, Petersen GM, Giardiello FM,
Hamilton SR. Phenotypic and molecular characteristics of hyperplastic polyposis. Gastroenterology 2000;119:323–332.
99. Hawkins NJ, Ward RL. Sporadic colorectal cancers with microsatellite instability and their possible origin in hyperplastic polyps and serrated adenomas. J Natl Cancer Inst 2001;93:1307–
100. Longacre TA, Fenoglio-Preiser CM. Mixed hyperplastic adenomatous polyps/serrated adenomas. A distinct form of colorectal
neoplasia. Am J Surg Pathol 1990;14:524 –537.
101. Jass JR. Hyperplastic polyps of the colorectum—innocent or
guilty? Dis Colon Rectum 2001;44:163–166. September 2002 102. Jass JR. Serrated adenoma of the colorectum. Curr Diag Pathol
103. Ajioka Y, Watanabe H, Jass JR, Yokota Y, Kobayashi M, Nishikura K. Infrequent K-ras codon 12 mutation in serrated adenomas of human colorectum. Gut 1998;42:680 – 684.
104. Biemer-Huttmann A-E, Walsh MD, McGuckin MA, Ajioka Y, Wa¨
tanabe H, Leggett BA, Jass JR. Immunohistochemical staining
patterns of MUC1, MUC2, MUC4, and MUC5AC mucins in hyperplastic polyps, serrated adenomas, and traditional adenomas of the colorectum. J Histochem Cytochem 1999;47:1039 –
105. Biemer-Huttmann A-E, Walsh MD, McGuckin MA, Young J, Leg¨
gett BA, Jass JR. Mucin core protein expression in colorectal
cancers with high levels of microsatellite instability indicates a
novel pathway of morphogenesis. Clin Cancer Res 2000;6:
106. Loukola A, Salovaara R, Kristo P, Moisio A-L, Kaariainen H,
Ahtola H, Eskelinen M, Harkonen N, Julkunen R, Kangas E,
Ojala S, Tulikoura J, Valkamo E, Jarvinen H, Mecklin J-P, de la
Chapelle A, Aaltonen LA. Microsatellite instability in adenomas
as a marker for hereditary nonpolyposis colorectal cancer. Am J
Pathol 1999;155:1849 –1853.
107. Iino H, Simms LA, Young J, Arnold J, Winship IM, Webb SI,
Furlong KL, Leggett B, Jass JR. DNA microsatellite instability
and mismatch repair protein loss in adenomas presenting in
hereditary non-polyposis colorectal cancer. Gut 2000;47:37–
108. Makinen MJ, George SMC, Jernvall P, Makela J, Vihko P, Kart¨
tunen TJ. Colorectal carcinoma associated with serrated adenoma—prevalence, histological features, and prognosis.
J Pathol 2001;193:286 –294.
109. Bomme L, Bardi G, Pandis N, Fenger C, Kronborg O, Heim S.
Clonal karyotypic abnormalities in colorectal adenomas: clues
to the early genetic events in the adenoma-carcinoma sequence. Genes Chromosom Cancer 1994;10:190 –196.
110. Bardi G, Pandis N, Fenger C, Kronborg O, Bomme L, Heim S.
Deletion of 1p36 as a primary chromosomal aberration in intestinal tumorigenesis. Cancer Res 1993;53:1895–1897.
111. Kambara T, Matsubara N, Nakagawa H, Notohara K, Nagasaka
T, Yoshino T, Isozaki H, Sharp GB, Shimizu K, Jass J, Tanaka N.
High frequency of low-level microsatellite instability in early
colorectal cancer. Cancer Res 2001;61:7743–7746.
112. Pegg AE. Mammalian O 6-alkylguanine-DNA alkyltransferase: regulation and importance in response to alkylating carcinogenic
and therapeutic agents. Cancer Res 1990;50:6119 – 6129.
113. Devereux TR, Belinsky SA, Maronpot RR, White CM, Hegi ME,
Patel AC, Foley JF, Greenwell A, Anderson MW. Comparison of
pulmonary O 6-methylguanine DNA adduct levels and Ki-ras activation in lung tumors from resistant and susceptible mouse
strains. Mol Carcinog 1993;8:177–185.
114. Puju S, Shuker DE, Bishop WW, Falchuk KR, Tannenbaum SR,
Thilly WG. Mutagenicity of N-nitroso bile acid conjugates in
Salmonella typhimurium and diploid human lymphoblasts. Cancer Res 1982;42:2601–2604.
115. Povey AC, Hall CN, Badawi AF, Cooper DP, O’Connor PJ. Elevated levels of the pro-carcinogenic adduct, O 6-methylguanine,
in normal DNA from the cancer prone regions of the large bowel.
116. Esteller M, Toyota M, Sanchez-Cespedes M, Capella G, Peinado
MA, Watkins DN, Issa J-PJ, Sidransky D, Baylin SB, Herman JG.
Inactivation of the DNA repair gene O6-Methylguanine-DNA
methyltransferase by promoter hypermethylation is associated
with G to A mutations in K-ras in colorectal tumorigenesis.
Cancer Res 2000;60:2368 –2371.
117. Esteller M, Risques RA, Toyota M, Capella G, Moreno V, Peinado MA, Baylin SB, Herman JG. Promoter hypermethylation of
the DNA repair gene O6-methylguanine-DNA methyltransferase COLORECTAL CANCER 118. 119. 120. 121. 122.
124. 125. 126. 127. 128. 129. 130. 131.
133. 134. 135. 136. 137. 875 is associated with the presence of G:C to A:T transition mutations in p53 in human colorectal tumorigenesis. Cancer Res
2001;61:4689 – 4692.
Horsfall MJ, Gordon AJ, Burns PA, Zielenska M, van der Vliet
GM, Glickman BW. Mutational speciﬁcity of alkylating agents
and the inﬂuence of DNA repair. Environ Mol Mutagen 1990;
Hendrich B, Hardeland U, Ng HH, Bird A. The thymine glycosylase MBD4 can bind to the product of deamination at methylated CpG sites. Nature 1999;401:301–304.
Bader S, Walker M, Hendrich B, Bird A, Bird C, Hooper M, Wyllie
A. Somatic frameshift mutations in the MBD4 gene of sporadic
colon cancers with mismatch repair deﬁciency. Oncogene
1999;18:8044 – 8047.
Wiebauer K, Jiricny J. Mismatch-speciﬁc thymine DNA glycosylase and DNA polymerase
mediate the correction of G:T
mispairs in nuclear extracts from human cells. Proc Natl Acad
Sci U S A 1990;87:5842–5845.
Fink D, Aebi S, Howell SB. The role of DNA mismatch repair in
drug resistance. Clin Cancer Res 1998;4:1– 6.
Karran P, Bignami M. DNA damage tolerance, mismatch repair
and genome instability. Bioessays 1994;16:833– 839.
Branch P, Aquilina G, Bignami M, Karran P. Defective mismatch
binding and a mutator phenotype in cells tolerant to DNA damage. Nature 1993;362:652– 654.
Rashid A, Shen L, Morris JS, Issa J-PJ, Hamilton SR. CpG island
methylation in colorectal adenomas. Am J Pathol 2001;159:
Vickers MA, Vyas P, Harris PC, Simmons DL, Higgs DR. Structure of the human 3-methyladenine DNA glycosylase gene and
localization close to the 16p telomere. Proc Natl Acad Sci U S A
Wang L, Patel U, Ghosh L, Banerjee S. DNA polymerase
mutations in human colorectal cancer. Cancer Res 1992;52:
4824 – 4827.
Dobashi Y, Shuin T, Tsuruga H, Uemura H, Torigoe S, Kubota Y.
DNA polymerase gene mutation in human prostate cancer.
Cancer Res 1994;54:2827–2829.
Berardini M, Mazurek A, Fishel R. The effect of O6-methylguanine DNA adducts on the adenosine nucleotide switch functions
of hMSH2-hMSH6 and hMSH2-hMSH3. J Biol Chem 2000;275:
Breivik J, Gaudernack G. Genomic instability, DNA methylation,
and natural selection in colorectal carcinogenesis. Semin Cancer Biol 1999;9:245–254.
Fishel R. Signaling mismatch repair in cancer. Nat Med 1999;
Nakamura S, Kino I. Morphogenesis of minute adenomas in
familial polyposis coli. J Natl Cancer Inst 1984;73:41– 49.
Shih I-M, Wang TL, Traverso G, Romans K, Hamilton SR, BenSasson S, Kinzler KW, Vogelstein B. Top-down morphogenesis
of colorectal tumors. Proc Natl Acad Sci U S A 2001;98:2640 –
Kang M, Mitomi H, Sada M, Tokumitsu Y, Takahashi Y, Igarashi
M, Katsumata T, Okayasu I. Ki-67, p53, and Bcl-2 expression of
serrated adenomas of the colon. Am J Surg Pathol 1997;21:
Wasan HS, Park H-S, Liu KC, Mandir NK, Winnett A, Sasieni P,
Bodmer WF, Goodlad RA, Wright NA. APC in the regulation of
intestinal crypt ﬁssion. J Pathol 1998;185:246 –255.
Tateyama H, Li W, Takahashi E, Miura Y, Sugiura H, Eimoto T.
Apoptosis index and apoptosis-related antigen expression in
serrated adenoma of the colorectum: the saw-toothed structure
may be related to inhibition of apoptosis. Am J Surg Pathol
Fenton RG, Hixon JA, Wright PW, Brooks AD, Sayers TJ. Inhibi- 876 138. 139.
142. 143. 144. JASS ET AL. tion of Fas (CD95) expression and Fas-mediated apoptosis by
oncogenic Ras. Cancer Res 1998;58:3391–3400.
Shanmugathasan M, Jothy S. Apoptosis, anoikis and their relevance to the pathobiology of colon cancer. Pathol Int 2000;
Frisch SM, Francis H. Disruption of epithelial cell-matrix interactions induces apoptosis. J Cell Biol 1994;124:619 – 626.
Kaye GI, Fenoglio CM, Pascal RR, Lane N. Comparative electron
microscopic features of normal, hyperplastic, and adenomatous human colonic epithelium. Variations in cellular structure
relative to the process of epithelial differentiation. Gastroenterology 1973;64:926 –945.
Jass JR. Relation between metaplastic polyp and carcinoma of
the colorectum. Lancet 1983;i:28 –30.
Leggett BA, Devereaux B, Biden K, Searle J, Young J, Jass
J. Hyperplastic polyposis: association with colorectal cancer.
Am J Surg Pathol 2001;25:177–184.
Jeevaratnam P, Cottier DS, Browett PJ, Van de Water NS, Pokos
V, Jass JR. Familial giant hyperplastic polyposis predisposing to
colorectal cancer: a new hereditary bowel cancer syndrome.
J Pathol 1996;179:20 –25.
Shannon B, Gnanasampanthan S, Beilby J, Iacopetta B. A poly- GASTROENTEROLOGY Vol. 123, No. 3 morphism in the methylenetetrahydrofolate reductase gene predisposes to colorectal cancers with microsatellite instability.
Gut 2002;50:520 –524.
145. Jung A, Schrauder M, Oswald U, Knoll C, Sellberg P, Palmqvist
R, Niedobitek G, Brabletz T, Kirchner T. The invasion front of
human colorectal adenocarcinomas shows co-localization of
nuclear beta-catenin, cyclin D1, and p16INK4A and is a region
of low proliferation. Am J Pathol 2001;159:1613–1617.
146. Ueno H, Murphy J, Jass JR, Mochizuki H, Talbot IC. Tumour
“budding” as an index to estimate the potential of aggressiveness in rectal cancer. Histopathology 2002;40:127–132. Received February 4, 2002. Accepted June 6, 2002.
Address requests for reprints to: Prof. J. R. Jass, Department of
Pathology, McGill University, Lyman Duff Medical Sciences Building,
3775 University Street, Montreal, Quebec, Canada H3A 2B4. e-mail:
[email protected]; fax: (514) 398 7446.
The authors thank Dr. Richard Fishel, Kimmel Cancer Center, Philadelphia, Pennsylvania, and Dr. Asif Rashid and Dr. Jean-Pierre Issa,
M. D. Anderson Cancer Center, Houston, Texas, for reading this review
and providing helpful criticism and information. ...
View Full Document