Bowel cancer is mainly a sporadic disease but people with a family history are at an increased risk of developing bowel cancer.
Hereditary bowel cancer due mainly to HNPCC and APC, with a small proportion occurring as a result of other inherited conditions (see below) (Murday and Slack, 1989). A family history of colorectal adenoma or carcinoma without any defined hereditary syndrome also increases the lifetime risk of developing colorectal cancer in an individual (see Table 1.1) (Houlston et al., 1990).

Table 1.1 Family history and lifetime risk of developing colorectal cancer (Houlston et al., 1990)

Family History Lifetime Risk
No family history 1 in 50 (2%)
One first degree relative 1 in 17 (6%)
One first degree relative + one second degree relative 1 in 12 (8%)
One first degree relative under the age of 45 years 1 in 10 (10%)
Two first degree ralatives 1 in 6 (17%)


HNPCC (Hereditary Non-Polyposis Colorectal Cancer)

This is the most common cause of hereditary colorectal cancer and occurs as a result of defects in mismatch repair genes (vide infra).
In 1913, an American pathologist, Aldred Scott Warthin, described a number of families with an increased frequency of cancers in successive generations compared to the general population (Warthin, 1913). Cancer of the uterus in the female members of the families and of the gastrointestinal tract (mainly stomach) in the male members appeared to be the most common. Other sites included the breast, mouth and lips. It was also apparent from Warthin’s numerous family-tree charts that cancers tended to occur at an earlier age in successive generations.
In 1966, an American physician, Henry T. Lynch described two families with preponderance for colorectal cancers and this was labelled Cancer Family Syndrome (Lynch et al., 1966).
Subsequently, one of Warthin’s “cancerous fraternities” termed ‘family G’ was further updated and was shown to have an increasing number of colorectal and uterine tumours in successive generations but there appeared to be fewer stomach cancers compared to Warthin’s original description, probably as a result of the general decline in this tumour type (Lynch and Krush, 1971).

There appears to be two distinct syndromes of familial inheritance of colorectal cancer in patients without polyposis. These were initially described by Lynch and bear the eponym Lynch syndrome I and II. Both of these are now referred to as ‘hereditary non-polyposis colorectal cancer’ (Lynch et al., 1985).
Lynch syndrome I is characterised by an autosomal dominant inheritance, predominance of proximal colon tumours, early age of presentation, more frequent metachronous and synchronous tumours and relatively improved survival.
Lynch syndrome II has the above features and in addition includes extracolonic tumours, mainly endometrial but also includes ovary, stomach, small bowel, hepatobiliary tract and transitional cell carcinoma of the urinary tract (Lynch, 1999)

There are clinical and genetic criteria for diagnosing HNPCC and these are detailed below:

1) Clinical Criteria 

The initial clinical criteria suggested by the International Collaborative Group on HNPCC (ICG-HNPCC) are outlined in Table 1.2 and are referred to as the classic criteria (or Amsterdam Criteria I) (Vasen et al., 1991). Subsequently this has been revised (Amsterdam Criteria II) because of families with the mismatch repair mutation genotype but who do not have colorectal cancer (Table 1.3) (Vasen et al., 1999).
Table 1.2 Amsterdam Criteria I (1990) (Vasen et al., 1991)

Classic ICG-HNPCC Criteria

Amsterdam Criteria I

At least three relatives with CRC with the following criteria

  • One should be a first degree relative of the other two
  • ≥ 2 successive generations should be affected
  • ≥ 1 patient with CRC should have been diagnosed before age 50
  • FAP must be excluded
  • All tumours should be histologically confirmed
Table 1.3 Amsterdam Criteria II (1999) (Vasen et al., 1999)

Revised ICG-HNPCC Criteria

Amsterdam Criteria II

At least three relatives with an HNPCC-associated cancer (CRC, endometrium, small bowel, ureter or renal pelvis) with the following criteria

  • One should be a first degree relative of the other two
  • ≥ 2 Successive generations should be affected
  • ≥ 1 Patient with cancer diagnosed before age 50
  • FAP must be excluded in cases of CRC
  • All tumours must be histologically confirmed
2) Genetic Criteria 

HNPCC is due to mutations of DNA mismatch repair (MMR) genes and are characterised by the phenomenon called microsatellite instability. Microsatellite instability (MSI) is defined as ‘a change of any length due to either insertion or deletion of repeating units in a microsatellite within tumour DNA when compared to DNA from normal tissue’ (Boland et al., 1998). MSI is produced by the inability of mutated MMR genes to repair errors of mismatched nucleotides which occur during normal DNA replication.
MMR genes are also referred to as caretaker genes because they repair errors that take place throughout the genome during DNA replication. This results in an accumulation of genetic defects that may ultimately result in the development of cancer.

In comparison, the tumour suppressor gene APC is believed to act as a gatekeeper gene (Kinzler and Vogelstein, 1996). Mutations in this gene lead to Familial Adenomatous Polyposis (FAP), in which individuals develop thousands of benign adenomas. The accumulation of further mutations within these adenomas eventually leads to malignant transformation. This is discussed later.

MSI has been shown in most HNPCC tumours and in about 15% of sporadic colorectal tumours. Tumours with MSI can be subdivided into 3 groups depending on the number of markers showing microsatellite instability.
Tumours are classified into MSI-high (MSI-H), if two or more markers show instability; MSI-low (MSI-L), if only one marker shows instability and MSI-stable (MSS), if none of the markers show instability (Boland et al., 1998).
The majority of patients with HNPCC have microsatellite instability (MSI) and tumours tend to be MSI-H (see below).

The revised Bethesda guidelines (Table 1.4) provide criteria that determine which patients in the general population should be screened for microsatellite instability in order to identify individuals at risk for HNPCC (Umar et al., 2004). These revised guidelines were published following a HNPCC workshop held in 2002, at the National Cancer Institute in Bethesda, Maryland. The new guidelines supersede previously published criteria in the original Bethesda guidelines (Rodriguez-Bigas et al., 1997).

It was also suggested at the 2002 workshop that a pentaplex panel of 5 quasi-monomorphic mononucleotide markers should be included in the evaluation of MSI because the previously recommended panel of 5 microsatellite markers (2 mononucleotide and 3 dinucleotide repeats) would tend to underestimate the frequency of microsatellite instability.

Table 1.4 Revised Bethesda guidelines

Revised Bethesda guidelines
  • Colorectal cancer diagnosed in a patient younger than age 50
  • The patient has synchronous or metachronous colorectal or HNPCC-associated tumours regardless of age of presentation
  • A patient younger than age 60 with colorectal cancer that has histological characteristics associated with MSI-high tumours
  • A patient who has one or more first-degree relatives that had an HNPCC-related tumor at age 50 or younger
  • A patient who has two or more first- or second-degree relatives that had HNPCC-related tumors at any age


FAP (Familial Adenomatous Polyposis)

This is an AD (autosomal dominant) disease characterised by the development of multiple adenomatous colorectal polyps commonly by the second decade of life. The number of polyps varies from 100s to 1000s and the incidence is about 1 in 10,000 individuals.

The lifetime risk of malignant transformation in at least one of the polyps is nearly 100% and usually occurs at a mean age of 40 years. This is referred to as classic FAP and is due to mutations within the APC (Adenomatous Polyposis Coli) gene; located on chromosome 5q 21-22 (Bodmer et al., 1987; Leppert et al., 1987).
Different mutations of the APC gene lead to varying clinical entities and other syndromes such as attenuated FAP (Spirio et al., 1993) and Gardner’s syndrome (vide infra).

Other clinical features of FAP include:

  • Duodenal adenomatous polyps

50 to 90% of patients develop duodenal polyps but less than 5% develop duodenal cancer. They tend to occur in the second and third part of the duodenum and periampullary polyps appear to have a higher malignant potential (Kadmon et al., 2001). After panproctocolectomy, patients with FAP still have an increased relative risk of dying and this is due mainly to death from upper gastrointestinal malignancy (Wallace and Phillips, 1998).

  • Gastric polyps

There are two main types of gastric polyps that occur in FAP. Fundic-gland hamartomatous polyps occur in about 50% of patients and are located in the fundus and body of the stomach; they have very little, if any, malignant potential. The other type is the adenomatous polyps which occur in only about 10% of patients and are located mainly in the antrum; these have some malignant potential (Offerhaus et al., 1999).

  • Osteomas

These are benign tumours of bone which commonly occur on the skull and may present in childhood even before gastrointestinal polyposis is evident. They do not have malignant potential but may need to be removed for cosmetic reasons.

  • Dental abnormalities

A variety of disorders of dentition occur including absent or extra teeth, odontomas and dentigerous cysts.

  • Congenital Hypertrophy of the Retinal Pigment Epithelium (CHRPE)

These are asymptomatic discrete, round, pigmented lesions of the retina. They are usually multiple and bilateral in association with FAP but may occur as single lesions sporadically. They are present in up to 80% of FAP patients and are present from birth, so may be used as a clinical indicator of inheritance of APC mutation, even before colonic polyps occur. These lesions are usually associated with mutations between codons 463 and 1387 of the APC gene (Olschwang et al., 1993).

  • Soft tissue tumours

Epidermoid cysts, fibromas and lipomas are common in FAP. They may require excision for cosmetic reasons.

  • Desmoid tumours

These are benign fibrous tumours which are locally invasive but without metastatic potential. They occur commonly within the abdomen and the abdominal wall, usually at the site of abdominal scars and after intra-abdominal operations. They are more frequent in females and regression can sometimes occur with anti-oestrogen therapy, cyclo-oxygenase inhibitors or chemotherapy.

Attenuated FAP

This is characterised by fewer colonic polyps than in classic FAP; usually less than a hundred. The mean age of diagnosis of colorectal cancer is 50 to 55 years and the polyps tend to be located more proximally in the bowel. Other features of classic FAP may also occur but CHRPE lesions and desmoid tumours are rare.

Gardner syndrome

This syndrome is phenotypically related to FAP and is characterised by gastrointestinal polyps associated with mesenchymal tumours such as osteomas and epidermoid cysts (Gardner and Richards, 1953). It was believed to be a completely separate clinical entity but is now known to be due to mutations in the APC gene which also causes classic FAP.

Turcot’s syndrome

Turcot’s syndrome is characterised by the association of various central nervous system tumours, especially medulloblastoma, with colorectal adenomatous polyps. The first documented cases of Turcot’s syndrome were described in 1959 by Jacques Turcot in a teenage brother and sister (Turcot et al., 1959). It appears that this clinical syndrome can arise from two distinct types of germ line aberrations which include mutation of the APC gene or mutation of a mismatch repair gene (Hamilton et al., 1995).

Other syndromes predisposing to colorectal polyps:-

Cowden Disease (CD)

This is an AD condition predisposing to a variety of proliferative lesions including gastrointestinal polyps, breast and thyroid cancer. It was first described in 1963 and named after the family name of the proband who was a 20 year old lady presenting with various developmental abnormalities, multiple hamartomas as well as breast and thyroid cancer (Lloyd and Dennis, 1963). Cowden disease has been found to be due to germline mutations in PTEN (a tumour suppressor gene located on chromosome 10q, see below).
Bannayan-Riley-Ruvalcaba (BRR) syndrome: this disease shares phenotypic similarities with Cowden’s disease and is due to deletions in the same tumour suppressor gene; PTEN (Marsh et al., 1997).

Juvenile Polyposis Syndrome (JPS)

This is also an AD condition with variable penetrance and causing hamartomatous gastrointestinal polyps with a predisposition to malignant transformation.
JPS has been found to be due to germline mutations in DPC4/SMAD4 (Deleted in Pancreatic Carcinoma locus 4/Small Mothers Against Decapentaplegic locus 4, which is located on chromosome 18q21.1. This gene has a sequence similar to the Drosophila gene, MAD, and is involved in transforming factor-beta signalling pathways) (Howe et al., 1998). In addition, BMPR1A (Bone Morphogenetic Protein Receptor 1A), which is located on chromosome 10q22.3 may be involved in a subset of cases (Howe et al., 2001).

Peutz-Jeghers syndrome (PJS)

This disorder was first described in 1896 by Sir Jonathan Hutchinson, an English pathologist and surgeon. However, a Dutch physician, Johannes L. A. Peutz, in 1921 and an American physician, Harold J. Jeghers, in 1944 were responsible for characterising the various features of the syndrome ( The term ‘Peutz-Jeghers syndrome’ was first introduced into medical literature in 1954.

Peutz-Jeghers syndrome is an autosomal dominant condition, characterised by the association of hamartomatous gastrointestinal polyps and mucocutaneous pigmentation. Polyps are most frequent in the small intestine, particularly the jejunum, but can occur elsewhere in the gastrointestinal tract including the colon.

Mucocutaneous hyperpigmented dark blue/brown macules present usually in young children and may fade by adulthood. Macules occur around the mouth, buccal mucosa, eyes, nose, perianal area and fingers.

Patients are at a relatively high risk of developing tumours at various sites including the colon (Giardiello et al., 2000).

The gene responsible for PJS has been found to reside at 19p13.3 by linkage analysis in most families with PJS (Olschwang et al., 1998), this gene has been denoted as LKB1 (Hemminki et al., 1998) and STK11 (serine/threonine protein kinase 11) (Jenne et al., 1998).

Bowel Cancer Surgery

Surgery for Bowel Cancer.


The geographical distribution and variation in cancer incidence in various immigrant populations (discussed earlier) imply a strong association with environmental aetiological factors.


There appears to be a significant role for dietary factors in the pathogenesis of colorectal cancer. Dietary factors are very difficult to study in humans due to the complex nature of human diet in addition to several interactions with other social factors such as cigarette smoking, alcohol intake and physical activity.
Certain metabolic enzymes are known to affect the bioactivation or detoxification of mutagens found in the diet.
Dietary mutagens may be ingested as carcinogens which may not be effectively cleared from the body as a result of mutations or polymorphisms of detoxifying enzymes. Others are ingested as pro-carcinogens which may be activated by polymorphisms or aberrations within metabolising enzymes. Thus genes that control metabolising enzymes may also be important in the effect of dietary factors in the development of sporadic colorectal cancer.

The effect of dietary factors has been well reviewed by Burnstein MJ in 1993 (Burnstein, 1993) and Sandler RS in 1996 (Sandler, 1996). The various factors are summarised below.


This is probably the most studied component of the diet and overall there appears to be an increased colorectal cancer risk with low fibre intake. Conflicting studies exist and the particular source of fibre studied does seem to influence relative risk of developing colorectal cancer. Fibre derived from fruit and vegetables appears to exert a more protective effect than that from cereal grains.
Fibre increases faecal bulk and may dilute potential carcinogens in the gastrointestinal tract as well as speeding up colonic transit and thus reducing period of contact between potential dietary carcinogens and the colonic mucosa. The theory that high fibre diet may be protective was popularised by Denis Burkitt FRS, an Irish surgeon who served in Uganda for many years, where he noted that the incidence of colorectal cancer was very low. He suggested that this was due to the high fibre diet that the people in this region enjoyed (Burkitt, 1971).


There are also several conflicting studies regarding the effect of fat intake and development of colorectal cancer. Overall, there appears to be an increased frequency of colorectal cancer in communities with a high fat intake, particularly saturated fat. This effect may be as a result of increased secondary bile acids which have been shown experimentally to promote colonic tumorigenesis. A high fat diet may also promote the growth of organisms capable of degrading bile acids to carcinogens. Studies have also shown a greater excretion of bile acids in the faeces of patients with colorectal cancer compared to controls.
Calcium intake has been found to lead to a reduced risk of developing colorectal cancer, possibly by its ability to inactivate bile salts.

Red meat:

Large consumption of red meat has been shown to be a risk factor for developing colorectal cancer. Heterocyclic amines released during cooking of meat may be responsible for the mutagenic effect of meat.

Fruits and vegetables:

Most studies show that fruits and vegetables have a protective effect on colonic mucosa. This effect may partly be due to fibre content but probably more importantly as a result of a high content of anti-carcinogens. These include folate, which is a methyl donor, and various antioxidants such as vitamin A, C, E and selenium.
Selenium may also inhibit cell proliferation, impair tumour metabolism and affect immune function.


High alcohol consumption has been shown to increase the relative risk of developing bowel cancer, particularly those involving the rectum.
Several mechanisms have been postulated; alcohol antagonises methyl donors such as folate and methionine and can thus affect DNA synthesis. Alcohol may also catalyse bioactivation of pro-carcinogens as well as inhibit the detoxification of mutagens by the liver and immune system.

Other risk factors for sporadic colorectal cancer include:


Individuals with history of previous polypectomy for adenomatous polyps or colectomy for colorectal tumours are at an increased risk of developing metachronous large bowel tumours. The frequency of metachronous tumours has been found to be about 4% with a variable time interval from 2 to 30 years (Morson, 1974).


Ulcerative colitis increases the risk of developing colorectal adenocarcinoma. Risk is particularly increased in patients who have had the disease for over ten years, in whom the disease involves large segments of the colon and in those who have an early age of onset of colitis (Bernstein et al., 1994).
There is some evidence that large bowel Crohn’s disease may also be associated with development of colorectal cancer (Ribeiro et al., 1996).


Cigarette smoking has been shown to be associated with an increased relative risk of developing colorectal adenomas and to a lesser extent, colorectal cancer (Giovannucci, 2001), suggesting that cigarette smoking exerts its effect at an early stage of the adenoma-carcinoma sequence. However, colorectal cancer is known to develop from adenomas after a significant lag-time, thus current epidemiological studies will show the association with adenomas much earlier than with cancers; since cigarette smoking has only become widespread in Western societies in the last few decades.
Recently, smoking of cigars has been associated with development of rectal cancer, whereas cigarette smoking was found to be associated with more proximal colon cancers (Sharpe et al., 2002).


Pelvic irradiation may be associated with a significant risk of developing colorectal cancer, usually after a latent period of 10 years. This is commonly seen in women who have had irradiation for gynaecological malignancy and subsequently develop rectosigmoid cancer. Although interpretation of this association is compounded by the fact that it is known that patients who have had a gynaecological malignancy are at higher risk for developing colorectal cancer irrespective of whether they had radiotherapy. The possible association of radiation and colorectal cancer was well reviewed by Sandler and Sandler (Sandler and Sandler, 1983).


There appears to be a decreased risk of colorectal cancer in postmenopausal women receiving hormone replacement therapy (Grodstein et al., 1999).
Oestrogens decrease the concentration of bile acids in the gastrointestinal tract, which are believed to have mutagenic potential.
Exogenous oestrogen use may also stimulate oestrogen (Issa et al., 1994) and vitamin D receptors found in the gastrointestinal tract that are believed to have tumour suppressor functions (Smirnoff et al., 1999).


Some operations have been implicated including cholecystectomy (Sandler, 1993) and ureterosigmoidostomy (Stewart et al., 1982).
Patients who have undergone a cholecystectomy have a higher level of faecal bile and this may explain their increased risk. Growth factors such as EGF (epidermal growth factor) in the urine of patients who have had an ureterosigmoidostomy may stimulate mucosal proliferation with subsequent neoplastic transformation


Clinical presentation

Wide range of clinical presentations which may be subclassified according to anatomical site of the tumor:

Right colon
Anaemia - lethargy, dizziness, blurring of vision, fainting spells, collapse
Right sided or central abdominal pain
Right sided abdominal mass
Weight loss
Pyrexia of Unknown Origi

Left colon
Left sided or lower abdominal pain
Change in bowel habits
Passing blood or mucus
Weight loss

Rectal bleeding
Mucus discharge
Tenesmus: a feeling of incomplete evacuation
Sacral/ Perineal pain


Blood tests
Barium enema
CT colonography

Page 1 of 2

Book A Consultation

Spire Elland Hospital

Longlands Rooms