93. Hahnloser D, Pemberton JH, Wolff BG, et al. Results at up to 20 years after ileal pouch–anal

anastomosis for chronic ulcerative colitis. Br J Surg 2007;94:333–340.

94. Holubar SD, Cima RR, Sandborn WJ, et al. Treatment and prevention of pouchitis after ileal pouch–

anal anastomosis for chronic ulcerative colitis. Cochrane Database Syst Rev 2010:CD001176.

95. Lake JP, Firoozmand E, Kang JC, et al. Effect of high-dose steroids on anastomotic complications

after proctocolectomy with ileal pouch–anal anastomosis. J Gastrointest Surg 2004;8:547–551.

96. Branco BC, Sachar DB, Heimann TM, et al. Adenocarcinoma following ileal pouch–anal anastomosis

for ulcerative colitis: review of 26 cases. Inflamm Bowel Dis 2009;15:295–299.

1724

Chapter 67

Colonic Polyps and Polyposis Syndromes

Robert S. Bresalier and C. Richard Boland

Key Points

1 Colorectal neoplasia develops through multistep carcinogenesis that involves the gradual

accumulation of genetic and epigenetic alterations in the genome. This process usually requires the

presence of some forms of genomic or epigenetic instability, and neoplasms may require several

decades to evolve from their earliest stages to fully advanced disease.

2 Genetic and familial factors play an important role in the genesis of both the sporadic (common) and

syndromic forms of colorectal neoplasia, such as familial adenomatous polyposis and Lynch

syndrome.

3 Colorectal carcinoma usually evolves gradually from colorectal adenomas in clinically identifiable

stages; removal of adenomas reduces the incidence of colorectal carcinoma.

4 The genetic bases of familial adenomatous polyposis and Lynch syndrome (previously called

hereditary nonpolyposis colorectal cancer) have been identified, which facilitate the early

identification and diagnosis of affected individuals.

5 A wide range of clinical heterogeneity is present in the familial colorectal cancer syndromes, and

attenuated forms of familial adenomatous polyposis and Lynch syndrome can be subtle and a

challenge to the clinician.

COLORECTAL POLYPS

The gastrointestinal tract accounts for more neoplastic disease than any other organ system in the body.

In North America, neoplasms of the colon and rectum have attracted the greatest interest because of

their relatively high incidence, and because appropriate intervention can dramatically modify the

morbidity and mortality associated with them. The adenoma is the most common precursor of colorectal

cancer, and early removal of adenomatous polyps can interrupt the natural history of the disease and

prevent death.

Colorectal cancers can develop through one of at least three molecular pathways, and each variety

appears to have some unique clinical and pathologic features.1–3 The adenoma–carcinoma sequence is

virtually canonical at this time and describes the common pathway taken by neoplasms that have

“chromosomal instability.” It has subsequently been proposed that “serrated” polyps, especially those in

the right colon, may be precursors of colon cancers with the microsatellite instability (MSI) phenotype.

It is becoming increasingly important to understand the clinical behavior of colorectal neoplasms in the

context of the genetic and molecular bases of these lesions.

Classification of Colorectal Polyps

The term polyp (from the Greek polypous, “morbid excrescence”) refers to a macroscopic protrusion of

the colonic mucosa into the bowel lumen. This can result from abnormal growth of the mucosa or from

a submucosal process that causes the mucosa to protrude into the lumen. Mucosal polyps can be sessile,

protruding directly from the colonic wall, or pedunculated, extending from the mucosa through a

fibrovascular stalk.

Mucosal polyps in the colon can be categorized as neoplastic, with malignant potential, and

nonneoplastic, with no malignant potential (Table 67-1). Neoplastic polyps include benign adenomatous

polyps that may evolve to carcinoma, adenomatous polyps that contain foci of intramucosal carcinoma

(carcinoma in situ), and adenomatous polyps in which carcinoma has penetrated the muscularis mucosae

(invasive carcinoma). A serrated polyp (sessile serrated adenoma [SSA] and traditional serrated

adenoma [TSA]) appears to originate in a manner unique from typical adenomas, but is also a

premalignant lesion. Instead of evolving through the traditional multistep pathway, SSAs in the

1725

proximal colon appear to evolve through a pathway that involves both BRAF mutation and progressive

methylation of DNA that result in the silencing of multiple genes. The fine details of this pathway

remain to be elucidated. Sometimes a polyp is found in which carcinoma has completely obliterated the

adenomatous tissue from which it arose (polypoid carcinoma).

Nonneoplastic mucosal polyps include hyperplastic polyps (HPs) (a form of serrated polyp which does

not directly progress to cancer), juvenile polyps, Peutz–Jeghers hamartomas, and a variety of

inflammatory polyps, including those associated with inflammatory bowel disease. Any submucosal

lesion can expand to push the mucosa into the bowel lumen and thus appear as a polypoid lesion.

Examples include lipomas, leiomyomas, colitis cystica profunda, pneumatosis cystoides intestinalis,

lymphoid aggregates, primary or secondary lymphomas, carcinoid tumors, and other metastatic

neoplasms.

Neoplastic Mucosal Polyps

Most colorectal cancers arise in pre-existing adenomatous polyps. Neoplastic mucosal epithelium

evolves through a series of progressive, cumulative molecular and cellular steps that lead to altered

proliferation, cellular accumulation, and glandular disarray. In some instances, the polyp also achieves

the ability to invade and metastasize through the adenoma-to-carcinoma sequence.

Several lines of evidence support the assumption that colorectal adenocarcinomas arise from

adenomatous polyps. The descriptive epidemiology of colonic adenomas parallels that of carcinomas,

but the benign lesions occur earlier. Adenomas are rare in geographic regions with a low prevalence of

colon cancer, and the distribution of adenomas in the colon parallels that of carcinomas. Adenomas

often occur in anatomic proximity to colon cancers, and cancer risk is proportional to the number of

adenomas present synchronously or metachronously in a patient. Cancer is often present in polyps

removed endoscopically or surgically, and the risk for cancer is proportional to the degree of dysplasia

or atypia in the polyp. Conversely, histologically evident residual adenomatous tissue may be found

surrounding carcinomas. Most important, results from several studies indicate that the systematic

removal of adenomatous polyps during screening sigmoidoscopy or surveillance colonoscopy decreases

the risk for the development of colorectal cancers and death from this disease.

CLASSIFICATION

Table 67-1 Classification of Colorectal Polyps

1726

Pathogenesis

Molecular Biology

1 Genetic changes that lead to the development of adenomas (and carcinomas) can be loosely organized

into three broad categories: alterations in proto-oncogenes, loss of tumor-suppressor gene activity, and

abnormalities of genes involved in DNA repair (Fig. 67-1).1–3 Several different mechanisms are involved

in altering these genes. Typically, oncogenes become overactive by mutation, rearrangement, or

amplification. Tumor-suppressor genes become inactivated by mutation or deletion, or silenced through

promoter methylation. Tumor-suppressor genes require inactivation of both copies of alleles, making

this process more complex. It is now clear, however, that the development of adenoma and carcinoma is

always associated with the progressive accumulation of genetic changes, and this process is termed

multistep carcinogenesis.1–4

In familial adenomatous polyposis (FAP), Lynch syndrome (previously called hereditary nonpolyposis

colorectal cancer [HNPCC]), and other familial syndromes, the first genetic alteration is inherited in the

germline, and therefore present in every cell. Environmental factors or “accidents” occurring in the DNA

from faulty replication or DNA damage are then required for the additional genetic mutations, termed

somatic mutations, as the process moves toward cancer. The clinical phenotype for the inherited

predispositions to gastrointestinal cancer is initially normal. The germline mutation provides the first

alteration or “hit,” which greatly increases the likelihood of neoplasia and that the disease will occur at

a younger age. FAP and Lynch syndrome account for about 3% to 4% of all colorectal cancers, but it is

estimated that as many as 30% of colorectal cancers occur in the context of a positive family history,

which is probably mediated by numerous subtle differences in DNA sequence (called single nucleotide

polymorphisms, or SNPs) that exert small changes in risk.

“Sporadic” polyps and cancers are associated with multiple somatic mutations, all of which are caused

by spontaneous decay of DNA, exogenous insults, or are accidents that occur during the replication of

DNA. The occurrence of most cases of colorectal cancer has been attributed to the high frequency of

cancer in the colon and rectum compared to other organs is a consequence of the number of stem cell

divisions that occurs in human colon. To some degree, the occurrence of colorectal cancer is often a

matter of “bad luck.”5

Destabilization of the genome is a prerequisite to carcinogenesis. This most commonly involves

chromosomal instability, which is manifested as widespread chromosomal deletions, duplications, and

rearrangements that produce aneuploidy. Alternatively, increased rates of mutations, often in tandemly

repeated DNA sequences known as microsatellites (i.e., microsatellite instability or MSI) or a form of

epigenetic instability called the CpG island methylator phenotype (CIMP), in which genes are

inappropriately silenced by promoter methylation1–3 are mechanisms that can lead to progressive

multistep carcinogenesis. Genomic (or epigenetic) instability generates a large number of random

alterations, which are often silent or lead to the extinction of the cell. However, occasional genetic

alterations enhance growth and survival, and the successive accumulation of those changes will

eventually facilitate the evolution of a neoplastic cell.

Cellular proto-oncogenes are a group of evolutionarily conserved genes that play a role in signal

transduction and the normal regulation of cell growth. They function by being turned on when growth

is stimulated and then turned off when sufficient growth has been achieved. Inappropriate activation of

these genes, usually through mutation that activates the protein or rearrangement of the promoter

sequences, leads to unregulated growth-regulatory signaling and excessive cellular proliferation. The

prototypical oncogene in CRC is in KRAS, and mutations in this gene are progressively more frequent in

larger, more advanced adenomatous polyps. The tumor-suppressor gene APC on chromosome 5q is

critical to the evolution of the colorectal adenoma. Allelic losses of chromosome 5q occur early during

carcinogenesis in the colon, and this is frequently the initial step in the evolution of colorectal

neoplasia. APC acts as the “gatekeeper” of colonic epithelial proliferation, and inactivation of this gene

will lead to net cellular proliferation and initiation of neoplasia in the colon. The protein produced by

the APC gene plays a central role in regulating the intracellular concentrations of the transcription

factor, β-catenin. β-catenin is highly expressed in cells at the base of the colonic crypt, where it

stimulates proliferation. As the cell commits to terminal differentiation, the APC gene is expressed,

which downregulates β-catenin by phosphorylation of the protein, targeting it for degradation.

Inactivating alterations in the p53 tumor-suppressor gene are a critical step in the transition from

adenoma to carcinoma, as the p53 protein normally prevents cells with damaged DNA from progressing

from the G1 to the S phase in the cell cycle. Alterations in APC, p53, and KRAS are the most common

alterations in most CRCs. However, a subset of about 15% of CRCs are “hypermutated” and have a

1727