Molecular Pathology of Neoplastic Gastrointestinal Diseases

Visit the link below to download the full version of this book:

https://medidownload.com/product/molecular-pathology-of-neoplastic-gastrointesti
nal-diseases/

Click Download Now

Editors
Antonia R. Sepulveda, MD, PhD John P. Lynch, MD, PhD
Department of Pathology & Cell Biology Division of Gastroenterology
Columbia University University of Pennsylvania
New York, NY, USA Philadelphia, PA, USA

ISSN 1935-987X ISSN 1935-9888 (electronic)

ISBN 978-1-4614-6014-5 ISBN 978-1-4614-6015-2 (eBook)
DOI 10.1007/978-1-4614-6015-2
Springer New York Heidelberg Dordrecht London

Library of Congress Control Number: 2012954430

© Springer Science+Business Media New York 2013

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction
on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter developed. Exempted from this
legal reservation are brief excerpts in connection with reviews or scholarly analysis or material supplied specifically
for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work.
Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the
Publisher’s location, in its current version, and permission for use must always be obtained from Springer. Permissions
for use may be obtained through RightsLink at the Copyright Clearance Center. Violations are liable to prosecution
under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not
imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and
regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of publication, neither
the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may
be made. The publisher makes no warranty, express or implied, with respect to the material contained herein.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

Preface

Critical molecular mechanisms underlying gastrointestinal (GI) neoplasia have been substan-
tially unraveled in recent years. This has resulted from technological advances such as the
genome project data and large-scale “omic” methods, combined with the application of classic
molecular and chemical testing approaches and established procedures for pathologic evalua-
tion of tissue and cellular samples. This progress is leading to the development of new
approaches for treatments and, in parallel, novel diagnostic workups of gastrointestinal can-
cers, integrating specific molecular testing in routine pathology practice. Moreover,
identification of disease susceptibly genes has enabled the medical community to better man-
age and prevent diseases that have hereditary traits.
While significant advances have been harnessed, much remains to be learned in the spec-
trum of neoplastic diseases of the gastrointestinal tract. Critical elements of research that have
allowed progress in the various fields of GI neoplastic disease include the availability of ani-
mal models, cell culture models, and basic and translational research approaches utilizing
prospective or archived specimen material, and such advances are reviewed here.
In this book, we review the molecular aspects that characterize the spectrum of neoplasms that
affect the GI tract, providing the reader with up-to-date knowledge at the level of (1) the molecu-
lar basis of the individual neoplasms, spanning the carcinomas of esophagus, stomach, small
bowel, colon, and rectum; neuroendocrine tumors; and gastrointestinal stromal tumors; (2)
molecular testing approaches for diagnosis or for characterization of target genes for selective
targeted therapies, with a review of recommended guidelines for clinical application whenever
available; (3) molecular testing for hereditary predisposition or disease risk for GI cancers.
The last three chapters in the book are forward-looking, focused on the molecular mecha-
nisms of metastasis, detection of circulating tumor cells and nucleic acids, and the use of tumor
markers for gastrointestinal cancers. These are current areas of research interest and future clini-
cal practice and serve to complement the information reviewed for the individual neoplasms.
It is clear that the rapid pace of discovery is unmatched by the definitive validation of many
molecular alterations that are identified through ongoing basic and translational research of can-
cer. Given this scenario we felt it would be impractical to provide coverage of all areas of research
in each tumor type, and ultimately, authors for each of the chapters identified what in their opin-
ion are the most relevant topics to cover for each tumor type at the time of writing, realizing that
novel findings that may be clinically relevant may become a reality as the book is published.
Nevertheless, basic principles of molecular pathogenesis and diagnosis of GI cancers are exten-
sively covered and will remain a foundation for clinical practice as new knowledge emerges.
We expect that this book will be useful to a large spectrum of professionals, from patholo-
gists, laboratorians, clinical gastroenterologists and oncologists, and trainees at various levels
such as medical students, residents, fellows, and postdoctoral fellows, as well as investigators
interested in the area of gastrointestinal cancer.

New York, NY, USA Antonia R. Sepulveda, MD, PhD

Philadelphia, PA, USA John P. Lynch, MD, PhD

v
Contents

Part I Introduction

1 Mechanisms of Gastrointestinal Carcinogenesis ................................................... 3

Frank I. Scott and John P. Lynch

Part II Epithelial Neoplastic Disease

2 Targeted Therapies and Molecular Diagnostics

of Gastrointestinal Cancers ..................................................................................... 33
Davendra Sohal, Antonia R. Sepulveda, and Weijing Sun
3 Molecular Pathology of Barrett’s Metaplasia
and Esophageal Adenocarcinoma .......................................................................... 43
Mamoun Younes
4 Molecular Pathology of Squamous Carcinomas of the Esophagus ..................... 53
Rohinton S. Tarapore and Jonathan P. Katz
5 Molecular Mechanisms and Pathology of Gastric Carcinogenesis:
Sporadic Cancers ..................................................................................................... 67
Shuko Harada and Antonia R. Sepulveda
6 Hereditary Diffuse Gastric Cancer and Other Gastric Cancers
Associated with Hereditary Predisposition Syndromes........................................ 83
Fátima Carneiro, Carla Oliveira, and Raquel Seruca
7 Cancer Predisposition Syndromes of the Gastrointestinal Tract ........................ 109
Ian S. Hagemann and Antonia R. Sepulveda
8 Molecular Pathology of Colon and Small Bowel Cancers: Sporadic Type......... 131
Asif Rashid
9 Cancers of the Rectum and Anal Canal................................................................. 141
Jenia Jenab-Wolcott and Bruce Giantonio
10 Molecular Pathology of Inflammatory Bowel Disease-Associated Neoplasia ....... 173
Takeshi Uehara, Deqin Ma, and Antonia R. Sepulveda

Part III Non-epithelial Neoplastic Tumors

11 Molecular Biology and Pathology of Gastrointestinal Stromal Tumors ............. 181

Paul J. Zhang
12 Neuroendocrine Tumors .......................................................................................... 193
Joseph R. Pisegna

vii
viii Contents

Part IV Metastatic Disease and Tumor Markers

13 Molecular Mechanisms of Tumor Metastasis ........................................................ 213

Andrew D. Rhim, Davendra Sohal, and Hiroshi Nakagawa
14 Circulating Tumor Cells and Nucleic Acids for Tumor Diagnosis ...................... 229
Loren Joseph
15 Serological Markers of Digestive Tract Cancers ................................................... 249
Jorge L. Sepulveda

Index .................................................................................................................................. 271

Contributors

Fatima Carneiro, M.D., Ph.D. Department of Pathology and Cancer Genetics Group,
Institute of Molecular Pathology and Immunology of the University of Porto, Porto, Portugal
Bruce Giantonio, M.D. Department of Hematology/Oncology, University of Pennsylvania,
Philadelphia, PA, USA
Ian S. Hagemann, M.D., Ph.D. Department of Pathology and Immunology, Washington
University, St. Louis, MO, USA
Shuko Harada, M.D. Department of Pathology, University of Alabama at Birmingham,
Birmingham, AL, USA
Jenia Jenab-Wolcott, M.D., Ph.D. Delaware County Memorial Hospital, Drexel Hill, PA,
USA
Loren Joseph, M.D. Department of Pathology, The University of Chicago, Chicago, IL,
USA
Jonathan P. Katz, M.D., Ph.D. Division of Gastroenterology, University of Pennsylvania,
Philadelphia, PA, USA
John P. Lynch, M.D., Ph.D. Department of Medicine, Division of Gastroenterology, Hospital
of the University of Pennsylvania, Philadelphia, PA, USA
Deqin Ma, M.D., Ph.D. Department of Pathology, University of Iowa Hospitals and Clinics,
Iowa City, IA, USA
Hiroshi Nakagawa, M.D., Ph.D. Department of Medicine, Gastroenterology Division,
University of Pennsylvania School of Medicine, Philadelphia, PA, USA
Carla Oliveira, Ph.D. Cancer Genetics Group, Institute of Molecular Pathology and
Immunology of the University of Porto and Medical Faculty of the University of Porto,
Porto, Portugal
Joseph R. Pisegna, M.D. Division of Gastroenterology and Hepatology, VA Greater Los
Angeles Healthcare System, Los Angeles, CA, USA
Asif Rashid, M.D., Ph.D. Department of Pathology, The University of Texas M. D. Anderson
Cancer Center, Houston, TX, USA
Andrew D. Rhim, M.D. Department of Medicine, Gastroenterology Division, University of
Pennsylvania School of Medicine, Philadelphia, PA, USA
Frank I. Scott, M.D. Division of Gastroenterology, University of Pennsylvania, Philadelphia,
PA, USA
Antonia R. Sepulveda, M.D., Ph.D. Department of Pathology & Cell Biology, Columbia
University, New York, NY, USA

ix
x Contributors

Jorge L. Sepulveda, M.D., Ph.D. Department of Pathology & Cell Biology, Columbia
University, New York, NY, USA
Raquel Seruca, M.D., Ph.D. Cancer Genetics Group, Institute of Molecular Pathology and
Immunology of the University of Porto and Medical Faculty of the University of Porto,
Porto, Portugal
Davendra Sohal, M.D., M.P.H. Solid Tumor Oncology, Taussig Cancer Institute, Cleveland
Clinic, Cleveland, OH, USA
Weijing Sun, M.D. Department of Medicine, Hematology-Oncology, University of Pittsburgh
School of Medicine, UPMC Cancer Pavilion, Pittsburgh, PA, USA
Rohinton S. Tarapore, M.S., Ph.D. Department of Gastroenterology, University of
Pennsylvania, Philadelphia, PA, USA
Takeshi Uehara, M.D. Department of Laboratory Medicine, Shinshu University School of
Medicine, Matsumoto, Japan
Mamoun Younes, M.D. Department of Pathology, University of Texas Medical School at
Houston, Houston, TX, USA
Paul J. Zhang, M.D. Department of Pathology and Lab Medicine, University of Pennsylvania
Medical Center, Philadelphia, PA, USA
 Part I
Introduction
 Mechanisms of Gastrointestinal
Carcinogenesis 1
Frank I. Scott and John P. Lynch

Introduction malignancies, focusing on the key principles of oncogenesis

and molecular steps involved in initiation, evolution, and
The burden of neoplastic diseases of the gastrointestinal tract progression of gastrointestinal malignancy.
is ever growing. Gastrointestinal malignancies are a leading
cause of morbidity worldwide. GLOBOCAN Data from 2008
estimate the overall annual incidence of GI malignancies to be Basic Concepts of Cancer Pathogenesis
3.8 million cases per year, with a mortality of 3.2 million per-
sons per year.1,2 Given these extraordinary numbers of affected Cancer Is a Disease of Gene Mutations
patients, a tremendous amount of resources have been
expended in a quest to understand, prevent, and treat these Foremost among the results of the modern molecular revolu-
diseases over the past 40 years. We now know that malignan- tion is the recognition that many human disease conditions
cies of the gastrointestinal tract are heterogeneous in composi- are, fundamentally, caused by mutations in genes.
tion, involving a complex interaction between environmental Gastrointestinal cancers are no different. Cancers arise
and host factors. This interaction can result in conversion of because cells have inherited or acquired mutations in critical
normal mucosa to precursor lesions, premalignant lesions, and genes that regulate cell proliferation, differentiation, and
eventual frank malignancy. Despite differences between apoptosis. These cells experience disorganized cell division
malignancies in differing cell types and tissues, there are some and unregulated growth, which also predisposes the cell to
common themes shared amongst them. further mutation events. This disordered, unregulated growth
Cells that have undergone neoplastic transformation lose leads to the formation of clinically observable tumors.
the ability to respond appropriately to signals regulating cell Genetic mutations involved in neoplastic transformation fall
differentiation, replication, migration, and apoptosis. into two categories: (1) gain-of-function events, which typi-
Neoplasms can develop in any organ of the GI tract and from cally involve oncogene or proto-oncogene activation, and (2)
any tissue in the body and can result in extrinsic compression loss-of-function events, typically involving disruption of
or invasion, consumption of metabolic resources, and meta- tumor suppressors.3 A number of different molecular events
static spread to distant sites in the body. Tumors can there- can produce gain-of-function (proviral insertion, gene
fore significantly impact patient health, regardless of whether amplification, chromosomal translocations, point mutations,
they are benign or malignant. and small deletions) or loss-of-function changes (gene dele-
This chapter will aim to demonstrate the underlying tions, point mutations, chromosomal rearrangements, and
molecular mechanisms and genetic basis for gastrointestinal epigenetic gene silencing). One important focus of current
research efforts is to utilize novel sequencing and microarray
technologies to minutely map and catalogue the many genetic
F.I. Scott, M.D. and epigenetic changes that occur in human cancers.
Division of Gastroenterology, University of Pennsylvania,
Philadelphia, PA, USA
e-mail: [email protected]
Carcinogenesis Is a Multistep Process
J.P. Lynch, M.D., Ph.D. (*)
Department of Medicine, Division of Gastroenterology,
The development of gastrointestinal malignancy usually
Hospital of the University of Pennsylvania,
4156 Curie Blvd/600 CRB, Philadelphia, PA, USA occurs in a multistep fashion, with malignant neoplastic
e-mail: [email protected] cells arising from dysplastic tissue (Fig. 1.1). In the colon,

A.R. Sepulveda and J.P. Lynch (eds.), Molecular Pathology of Neoplastic Gastrointestinal Diseases, Molecular Pathology Library 7, 3
DOI 10.1007/978-1-4614-6015-2_1, © Springer Science+Business Media New York 2013
4 F.I. Scott and J.P. Lynch

Fig. 1.1 Colon cancer develops in a multistep fashion, progressing ment of adenocarcinoma (c, f). An example of the typical progression
from normal mucosa (a, d) (seen here endoscopically and histologi- at the genetic level is also demonstrated (g) with examples of mutations
cally), with microscopic foci of dysplasia as mutations accumulate. that can occur at various points in the progression from normal mucosa
Subsequently, adenomas form (b, e), with further accumulation of to adenoma to carcinoma. MSI microsatellite instability. Pathology
mutations. Neoplastic transformation can then occur with the develop- images courtesy of Antonia R. Sepulveda, MD, PhD
1 Mechanisms of Gastrointestinal Carcinogenesis 5

Fig. 1.2 The progression from normal squamous cell mucosa (a, e) to environmental insults resulting in chronic inflammation. This
esophagitis (f) to Barrett’s esophagus (b, g), followed by low-grade inflammation results in Barrett’s metaplasia, with subsequent altered
dysplasia (LGD) (h), high-grade dysplasia (HGD) (c, i), and finally to gene expression and subsequent development of various mutations and
esophageal adenocarcinoma (d, j), as seen here endoscopically and his- eventual neoplastic transformation. Endoscopic images courtesy of Dr.
tologically, is not as uniform at a genetic level as in colon cancer. Gary W. Falk, M.D., M.Sc. Pathology images courtesy of Antonia R.
Despite this, the process is similar (k), with genetic predispositions and Sepulveda MD, PhD

this progression from normal mucosa to dysplastic foci to Similar progressions from normal mucosa to dysplasia to
eventual invasive carcinoma is represented by the adenoma– malignancy have been described for esophageal cancer, gas-
carcinoma sequence, and has been demonstrated in patho- tric cancer, and pancreatic cancer. In esophageal adenocarci-
logic, epidemiologic, and animal studies.4–7 The earliest noma, malignancy is preceded by a premalignant metaplasia
histologically definable lesions in this sequence are aberrant known as Barrett’s esophagus. Barrett’s is an intestinal-type
crypt foci (ACF). ACF are crypts that appear larger and metaplasia that arises in the setting of chronic gastroesopha-
thicker than normal, with increased luminal diameter and an geal reflux disease.21 Barrett’s metaplasia is not dysplastic,
opening that can be slit-like or serrated. They were first but can progress to dysplasia and adenocarcinoma in a
identified by methylene blue stain in azoxymethane-treated significant number of patients (Fig. 1.2). Similar patterns of
mice.8–10 It is estimated that 65–95% of human ACF are progression from established precursor lesions to dysplasia
hyperplastic, but a significant proportion are dysplastic and to neoplasia have been documented in the stomach (gastri-
are considered to be similar to adenomatous polyps.11–13 tis > atrophy > intestinal metaplasia > dysplasia > cancer) and
Genetic analysis of these lesions has demonstrated that they in pancreatic cancer (PanIN1a > PanIN1b > PanIN2 > PanIN3)
share many of the same mutations present in adenomas.8,13–20 as well.22,23
6 F.I. Scott and J.P. Lynch

Role of Evolutionary Processes in Neoplastic Table 1.1 The hallmarks of carcinogenesis

Transformation Self-sufficiency in proliferative signals
KRAS, cyclin D1, WNT, MYC
The recognition that cancer is a disease of gene mutations Ability to ignore anti-proliferative signals
RB, TP53, PTEN, APC
and that biologically there is a recognizable, stepwise pattern
Evasion of apoptotic cues
leading to cancer led to the current standard model wherein TP53, BAX, bcl-2, FAS
carcinogenesis is a complex process requiring multiple Telomere maintenance
sequential genetic mutations.24–27 Prior to the development of TERT, TERC
this theory, it was thought that tumor initiation and progres- Self-directed angiogenesis
sion were distinct biologic and genetic processes.28 The mul- VEGF, PDGF, THBS1
tistep hypothesis has been supported by growing data that Local invasion and metastatic spread
E-cadherin
suggest that tumors are clonal, derived from a single pro-
genitor that has acquired the necessary mutations to undergo
neoplastic transformation.24,29 Mutational events are cur-
rently believed to be random events, with only rare muta- properties. These properties, while not necessarily unique to
tions leading to inappropriate activation of growth promoters malignant cells, are all required for a successfully trans-
or silencing of tumor suppressor pathways. These rare muta- formed cell. These six cardinal features were described in a
tions thus impart upon these cells a unique survival advan- seminal review by Hanahan and Weinberg in 2000.30 They
tage over normal adjacent cells.3,30 These mutations are are (1) self-sufficiency in proliferative signals, (2) ability to
acquired gradually over a period of time. The survival and ignore anti-proliferative signals, (3) inability to undergo
competitive advantages of certain mutations permit a clonal apoptotic cell death or able to evade normal apoptotic cues,
expansion. Subsequent acquired mutations may impart a fur- (4) telomere maintenance, permitting limitless replication
ther advantage and another round of clonal expansion, ulti- potential, (5) self-directed angiogenesis, improving nutrition
mately leading to the formation of a neoplasm.24,29,31 and oxygenation to the developing tumor, and, lastly, (6) the
The number of genetic mutations required for clonal capacity to invade tissues locally and metastasize distally
expansion and malignant transformation remains unknown, (Table 1.1).
and is thought to vary based on the tissue studied, gene com- This is a very flexible model for carcinogenesis, and is
binations employed, and an individual’s genetic background better able to describe an individual person’s cancer, as any
including gene polymorphisms. It is currently estimated that number of gene mutations can yield the same required can-
for the majority of malignancies the average number of cer cell property. Moreover, these properties or hallmark fea-
required mutations is between four and seven.30 This raises tures can be obtained in different orders but still yield a
an important problem, particularly for epithelial tissues that cancer. Several recent studies also indicate that the ability to
are constantly turning over. Few cells except stem cells are modulate the immune response should be added to this list as
retained long enough to acquire 3–7 mutations in critical a seventh cancer hallmark feature.36,37 In addition, this model
genes, when mutations are believed to be largely random argues that a beneficial strategy toward cancer therapeutics
events. This observation then has been used to support the would be to attack these required features. We will discuss
theory that stem cells are involved in carcinogenesis.32,33 An these cardinal features in greater detail later in this chapter.
alternate, unproven hypothesis is that certain mutations can
alter population dynamics, with certain non-stem cells
acquiring mutations that promote survival or that alter pat- Biological Factors Contributing
terns of migration or differentiation.34,35 This remains an to Gastrointestinal Carcinogenesis
active area of research.
Mutations in genes that are crucial for cell proliferation,
differentiation, and apoptosis result in disorganized cell
Successfully Transformed Cancer Cells division and unregulated growth, which also predispose the
Share a Finite Set of Qualities cell to further mutation. These mutations can be inherited
or sporadic. Exposure to various environmental factors,
While there can be great variety in the processes by which including chronic inflammatory conditions and carcino-
cells acquire cancer-causing mutations, and the genes gens, can promote these mutations. We will briefly discuss
involved can vary considerably between tissues, it has also the role of heredity, acquired sporadic mutations, and envi-
been observed that cells that have been successfully trans- ronmental factors in the pathogenesis of gastrointestinal
formed typically have obtained a subset of well-defined cancers.
1 Mechanisms of Gastrointestinal Carcinogenesis 7

Table 1.2 Signaling pathways involved in familial cancer syndromes tumor suppressor genes is necessary for the genesis of
of the GI tract tumors, but the processes causing this are unknown.44 The
Gene Associated gastrointestinal malignancy reason for this is that the cells retain sufficient function from
APC Familial adenomatous polyposis (FAP) the remaining normal allele to prevent neoplastic transfor-
DNA mismatch repair Hereditary nonpolyposis colon cancer mation under most circumstances. Molecular studies of the
syndrome (HNPCC)
second allele found that they are frequently inactivated by
Base excision repair MYH adenomatous polyposis (MAP)
several processes including direct mutation, epigenetic
ATM Gastric adenocarcinoma
silencing, or gene deletion. A second unanswered question is
BRCA2 Pancreatic adenocarcinoma
why neoplastic transformations only occur in a limited set of
FA (Fanconi’s anemia) Hepatocellular carcinoma
CDH1/E-cadherin Gastric adenocarcinoma
tissues even though the inherited mutations reside in every
cell of the body. One possibility is that these tissues may be
uniquely sensitive to or dependent on these genes. The rea-
sons for this selective sensitivity remain unknown. In addi-
tion, the penetrance of these mutations can be highly variable
Heredity in part due to the presence of genetic modifiers. Therefore,
there remains much to be learned about the process of car-
A familial predisposition for cancer has been recognized for cinogenesis, even when an underlying inherited mutation has
some time. Familial cancer syndromes are recognized for been identified.
esophageal, gastric, colonic, and pancreatic cancer. Despite
the considerable differences between the tissues, familial
cancer syndromes share several common features. First, Sporadic Carcinogenesis
these genetic tendencies are primarily due to inherited (ger-
mline) mutations. The mutations confer a significantly While familial cancer syndromes have been a primary
increased risk for a particular type of cancer. Moreover, those research focus, the majority of all cancers are sporadic.
with a familial cancer syndrome typically develop the cancer Compared to familial cancers, sporadic malignancies tend to
at a much younger age than those afflicted with the sporadic present later in life. This has led to the hypothesis that these
type of malignancy. Some germline mutations confer a risk tumors are the result of accumulated mutations in the course
for multiple neoplasms, often in different tissues. In addi- of a normal life, due to either carcinogen exposures or even
tion, while penetrance for any inherited predisposition can errors that occur during DNA replication. It has been esti-
be variable, it is usually the case that in familial cancer syn- mated that, over decades, these processes can result in the
dromes greater than 50% of those carrying the gene mutation accumulation of several mutations that can collectively result
will develop a cancer. Lastly, familial cancer syndromes are in neoplastic transformation.24,25 Importantly, although famil-
typically due to inherited mutations in critical tumor sup- ial and sporadic cancers are mechanistically quite distinct,
pressor genes.38 many of the mutations are common to both processes. Thus,
The inherited colon cancers Hereditary Nonpolyposis though they may arrive by different means, the path to can-
Colorectal Cancer (HNPCC), also known as Lynch syn- cer shares many common elements that will prove useful
drome, and Familial Adenomatous Polyposis (FAP) illustrate diagnostically and therapeutically.
another important principle. HNPCC and FAP kindreds are One critical difference between familial and sporadic
responsible for approximately 5% of all colon cancers. forms of cancer is that the cancer risk is genetically defined
Research into the causative etiologies of both of these syn- for the former but not the latter. Moreover, the sporadic can-
dromes has illuminated gene pathways involved in the major- cer risk can be modified by dietary habits, environmental
ity of sporadic colon cancers.4,39,40 This is true for other exposures, and even exercise and fitness levels. Dietary fac-
familial cancer syndromes as well; the study of these disor- tors, like high salt diets or the Western high-fat diet, have
ders has revealed significant insights into the pathogenesis of been associated with an increased incidence of gastric cancer
all malignancies, since many of the genes involved in heredi- and colorectal cancer, respectively.45–47 Folate supplementa-
tary cancers are also mutated in sporadic cancers.41–43 Several tion has been associated with a decreased risk of colorectal
common signaling and regulatory pathways are now known cancer in a small number of patients.48,49 Anti-inflammatory
to be responsible for the majority of familial cancer syn- medications such as aspirin and other NSAIDs have been
dromes (Table 1.2). associated with a decreased risk of esophageal, gastric, pan-
A number of important questions remain unresolved. creatic, and colorectal cancer in both human50–54 and animal
What are the molecular events that precede the development studies of neoplasia.55–61 Body habitus, in particular degree
of malignancy? It is known that biallelic inactivation of of obesity, can increase the risk of cancer of the esophagus,
8 F.I. Scott and J.P. Lynch

pancreas, and colon.62,63 One note of caution here is that ingestion, as studies have demonstrated that significant
genetics may not be entirely removed from the risk for spo- amounts of aflatoxin ingestion work in synergy with the virus
radic carcinogenesis, as subtle genetic polymorphisms may to increase rates of hepatocellular carcinoma.73–75 Tobacco
interact with these dietary and environmental stresses to smoke is another common environmental exposure that has
increase the risk for cancer. Genome-wide association stud- been linked to gastrointestinal malignancies including head
ies are under way to identify these more modest genetic con- and neck and esophageal squamous cell cancers.76
tributions to cancer risk, but much more work needs to be Mechanistically, this is due to the 40 different mutagens and
done in this area. In summary, sporadic carcinogenesis is carcinogens in tobacco smoke that cause DNA damage.
mechanistically distinct from familial forms and the risk for Among these carcinogens are N-nitrosamines, which are
developing a sporadic cancer can be modified by behavior thought to directly stimulate colon cancer cell growth,
and dietary changes. benzo[a]pyrene, which has been associated with esophageal
cancer, and polycyclic aromatic hydrocarbons, which are
thought to promote KRAS mutations in pancreatic cancer.77
Carcinogen Exposure In summary, we are exposed to a number of environmental
carcinogens that can play a significant role in introducing
DNA is typically stable, but can be altered via chemical or DNA mutations and thereby promoting the development of
physical reactions. Some compounds, such as alkylating gastrointestinal neoplasia.
agents, can interact and modify DNA directly. These com-
pounds are known as carcinogens. Most agents, however, are
procarcinogens, and must first be metabolized to generate Inflammatory Microenvironment
active, electrophilic derivatives. These activated compounds
can then interact with DNA, modifying individual nucle- After several decades of observation and study, it was estab-
otides. If not repaired or removed by the cell’s DNA repair lished that chronic inflammatory conditions are associated
machinery, these modified nucleotides can then disrupt base with cancer in a number of tissues and organ systems. The
pairings during DNA replication. If these chemical gastrointestinal organs may be particularly vulnerable for
modifications are not recognized prior to DNA replication, reasons that are not clear at the present time. Every major
the mispaired DNA will introduce mutations into the genetic organ of the GI tract is subject to one or more disease pro-
code that will be passed on to subsequent daughter cells. cesses that can result in a chronic inflammatory condition.
Fortunately, evolution has endowed cells a plethora of mech- While the etiologies for these processes can vary consider-
anisms to protect DNA against potential mutagens64,65 via ably (chemical, viral, bacterial, genetic) all are known to
repairing mutations that occur66,67 or initiating apoptosis if increase the risk for cancer. These findings have prompted
necessary in order to prevent transmission of mutant DNA ongoing mechanistic studies into how inflammatory condi-
codes to daughter cells. tions promote carcinogenesis, and have suggested interven-
The gastrointestinal tract is continuously exposed to a tions that have begun to reduce the frequency of some GI
number of mutagens via ingestion, and these tissues may cancers.
therefore be particularly dependent upon these mechanisms A compelling demonstration of the association between
for protection. Certain cooking practices can contribute to chronic inflammatory conditions and cancer is observed in
development of small amounts of carcinogens within food patients with hereditary pancreatitis. Hereditary pancreatitis
that are then ingested.68 Food preservation and storage prac- is an autosomal dominant condition caused in most kindreds
tices may be particularly to blame for inadvertent exposures. by inherited mutations in the gene encoding cationic trypsino-
Nitrates are preservatives commonly found in processed gen (PRSS1).78,79 These mutations augment trypsinogen
meats, and historically were included in beer as well. These autoactivation or enhance trypsin stability, leading to epi-
compounds can function as procarcinogens if converted to sodes of acute pancreatitis.79 Soon after the recognition of
N-nitrosamines by gut bacteria. Chronic ingestion of these this inherited disorder, it was noted that several family kin-
compounds has been associated with increased rates of gas- dreds also experienced high rates of pancreatic cancer.80
tric and esophageal cancer.69–72 In contrast, corn, peanuts, Pancreatic cancers arise only after 20 or more years of recur-
and rice storage practices in the developing world frequently rent pancreatitis,81 suggesting that it is the chronic
lead to contamination by Aspergillus molds. These molds inflammation that causes the cancer, not the PRSS1 muta-
can produce aflatoxin, a procarcinogen that is metabolized tion. This pattern thus supports the hypothesis that chronic
by the cytochrome P450 system in the liver. After metabolic inflammation can promote carcinogenesis in the pancreas.
conversion, this chemical can form adducts with guanine Ulcerative colitis (UC) is a chronic, relapsing inflammatory
nucleotides, leading to mutagenesis. Individuals with chronic condition of the colon with a strong association with carcino-
active HBV infection are especially sensitive to aflatoxin genesis.82,83 Initial reports suggested a cumulative cancer risk
1 Mechanisms of Gastrointestinal Carcinogenesis 9

as high as 60% in UC patients.84 More recently, this risk has animals. In these initial studies, viral genes were identified
been better defined, and in particular the elements of disease that could induce neoplastic transformation. These genes
extent, duration, and severity are now recognized to impact were called oncogenes.90 Subsequent research demonstrated
the risk for the colon cancer in UC patients.83 Disease dura- similar sequences in the human genome. These human genes
tion is the best-established risk factor, as the annual and are now known as ‘proto-oncogenes’, and further research
cumulative incidences for colon cancer in UC increase from has demonstrated that proto-oncogenes and their regulators
0.2% and 1.6%, respectively, after 10 years of disease to are frequently mutated in human malignancies, resulting in
1.2% and 18.4%, respectively, after 30 years of UC.85 Disease their constitutive activation.91 Since the initial discovery of
extent is also a clinically relevant factor. Pancolitis imparts a the human oncogene HRAS,92 more than 80 human proto-
20-fold greater risk for colon cancer than disease limited to oncogenes have been identified, and new candidates are con-
the rectum.85,86 Together, these correlations support the con- stantly being evaluated.
clusion that the chronic inflammatory process can promote Proto-oncogenes fall into four broad categories: peptide
carcinogenesis. growth factors, receptor and non-receptor tyrosine kinases,
Lastly, epidemiological studies examining aspirin and signal transduction proteins associated with cell membranes,
nonsteroidal anti-inflammatory use and cancer incidence and nuclear transcription factors.93 All have been found to
have generally confirmed that frequent users of aspirin or play significant roles in cell cycle regulation, proliferation,
other NSAIDS have reduced rates of esophageal, gastric, and and differentiation. Several proto-oncogenes can be activated
colon cancers.50,53,54,87 Meta-analyses have further concluded simultaneously in gastrointestinal cancers, and continued
that regular aspirin or NSAID use may be chemoprotective functionality is mandatory for tumor growth. Disruption of
for gastric and esophageal cancers50 or in patients with famil- these stimulatory signals not only results in inhibition of
ial adenomatous polyposis (FAP), an inherited predisposi- tumor growth but can also trigger induction of apoptosis and
tion to colon cancer.88,89 In summary, there is substantial cell death.94 This has led to the suggestion that cancers are
epidemiologic, biologic, and pharmacologic evidence impli- “oncogene dependent”, and therapeutic strategies to take
cating the inflammatory environment as a causative factor advantage of this dependency are being explored.
for many human gastrointestinal cancers. Several types of mutations involving proto-oncogenes
can lead to the development of neoplasia.3,93 Gene rearrange-
ments, point mutations, and insertion/deletion events can
Molecular Mechanisms Promoting Neoplastic permanently alter the conformation of the gene product,
Transformation resulting in chronically active proteins. These same mecha-
nisms can interfere with or inactivate proto-oncogene regula-
As was discussed previously, there are a great variety of tory domains. Mutations can also increase oncogene mRNA
mutations cancer cells can accumulate. This has challenged production. All of these mutations, therefore, result in either
our ability to understand the process mechanistically and loss of proto-oncogene regulator function or inappropriate
develop effective therapeutic modalities. However, Hanahan proto-oncogene expression. The result is uncontrolled cell
and Weinberg in 2000 suggested an alternative model, recog- division, proliferation, and delayed differentiation.
nizing six hallmark features that all cancers need to acquire While several proto-oncogenes are typically activated
to become a “successful” cancer.30 These are (1) self- simultaneously in transformed cells, the combination of
sufficiency in proliferative signals, (2) ability to ignore anti- genes is highly variable between tissues and even within
proliferative signals, (3) inability to undergo apoptotic cell the same cancer. Critical growth-signaling pathways often
death or able to evade normal apoptotic cues, (4) telomere will contain several proto-oncogenes in epistatic succes-
maintenance, permitting limitless replication potential, (5) sion. Phenotypically, these mutations would all produce the
self-directed angiogenesis, improving nutrition and oxygen- same effect. One example is the epithelial growth factor
ation to the developing tumor, and, lastly, (6) the capacity to signaling cascade. EGF, the receptor EGFR, KRAS, JUN
invade tissues locally and metastasis distally. We will explore (c-jun), and MYC (c-myc) are all linked in this signaling
several of these topics in greater depth next. pathway. KRAS mutations are present in up to 90% of pan-
creatic cancers and 50% of colorectal cancers, but are much
less common in cancers of the esophagus. Therefore, while
Self-Sufficiency in Growth Signals 90% of esophageal squamous cell carcinomas do not have
KRAS mutations, these tumors still exhibit increased EGFR
The concept that genetic mutations could result in gained activity due to EGFR gene amplification. In summary, one
functionality and unrestrained cell proliferation was first essential feature of neoplastic cells is gain-of-function
appreciated via the study of DNA viruses and retroviruses mutations of proto-oncogenes, with subsequent self-
that were known to be associated with tumors in humans and sufficiency in proliferation.