Elsevier

The Lancet

Volume 353, Issue 9150, 30 January 1999, Pages 391-399
The Lancet

Seminar
Colorectal cancer

https://doi.org/10.1016/S0140-6736(98)07127-XGet rights and content

Summary

Colorectal cancer is a leading cause of morbidity and mortality with about 300 000 new cases and 200 000 deaths in Europe and the USA each year.1, 2 Published trials have established a role for chemotherapy in colorectal cancer, in the adjuvant setting for Dukes C colon cancer, with an absolute survival benefit of about 5% and in advanced colorectal cancer, for which it improves quality of life and increases survival by 6–12 months. For rectal cancer, radiotherapy decreases rates of local recurrence and, in locally advancd disease, successfully palliates pain, tenesmus, and bleeding. The evolving understanding of colorectal carcinogenesis, in particular recognition of vital genes that may be mutated or lost during tumour development, has been translated into innovative gene therapy techniques. Finally it is increasingly apparent that surgical site specialisation and a multidisciplinary approach (including surgeons, pathologists, and oncologists) may lead to optimum outcomes for patients with colorectal cancer.

Section snippets

Pathogenesis

Most large bowel cancers arise within pre-existing adenomatous polyps or adenomas. These lesions are common. True incidence is difficult to calculate in symptom-free and unselected populations, but necropsy studies have shown a prevalence of about 35% in Europe and the USA, with lower rates (10–15%) in Asia and Africa. Adenomas are classified by histological architecture as tubular, tubulovillous, or villous. Villous change is associated with a higher malignant potential, as are large (up to

Molecular biology and genetics

Relative colorectal cancer risk is defined by genetic predisposition and environmental factors, including a diet low in fibre, vegetables, and folate, and high in fat, red meat, and alcohol, sedentary occupation, and cigarette smoking.

Recognition of the genetic component of colorectal cancer is growing. Mutations are present as inherited germline defects or arise in somatic cells secondary to environmental insult. There are two main inherited predisposition syndromes: familial adenomatous

Sporadic colorectal cancer

Vogelstein and colleagues3 have proposed a multistep model for the genetic events in the progression of sporadic colorectal cancer. Colorectal cancer occurs mainly in the elderly, which is consistent with the theory that a cell must accumulate a combination of four or five defects, including mutational activation of oncogenes and inactivation of tumour-suppressor genes, to undergo full malignant transformation. If one or more of these defects are present at birth as germline abnormalities,

FAP

FAP is an autosomal dominant disorder in which multiple adenomatous polyps develop in the colon during the second and third decades of life. These polyps are histologically identical to those preceding sporadic colorectal cancer, and, although individually each has little risk of malignant transformation, the numbers (reaching thousands in some patients) mean that the risk of colorectal cancer is almost 100% by age 40 years.4

The gene that causes FAP (the adenomatous polyposis coli [APC] gene)

HNPCC

HNPCC describes an autosomal dominant predisposition to colorectal cancer that lacks the excess polyposis seen in FAP. The disorder is thought to account for up to 6% of colorectal cancer cases. HNPCC is classified into two Lynch syndromes, and an international group has defined a set of diagnostic criteria—the Amsterdam criteria10, 11 (panel). Clinical features of HNPCC include an association with other tumours (endometrium, ovary, hepatobiliary tract, skin), a predilection for the right side

Screening and prevention

The premises underlying any cancer screening programme include a sensitive and specific screening assay and a likelihood that early detection of disease will result in more patients being seen who can be cured by surgical intervention. Patients at increased risk of colorectal cancer cannot be defined by family history. Molecular genetic studies looking for relevant germline mutations could elucidate relative risk for patients, in particular families (for example in families with FAP the

Faecal occult blood tests

The faecal occult blood test is a cheap investigation and causes the patient little discomfort. There are several types of test, including immunological and haemporphyrin assays, but the most commonly used is the Haemoccult test. Filter paper impregnated with guiac undergoes phenolic oxidation in the presence of haemoglobin in the stool after the addition of hydrogen peroxide. The test detects 10 mL of blood loss per day in 67% of patients and more than 20 mL blood loss per day in 80–90% of

Flexible sigmoidoscopy

Flexible sigmoidoscopy is quick and simple and gives high sensitivity and specificity but has the disadvantage that only 65% of adenomas and early cancers can be visualised because of limited access. Results from two case-control studies are, however, promising. Newcomb25 reviewed medical records of 66 patients with colorectal cancer and 196 controls. Individuals who had undergone a single assessment by sigmoidoscopy had lower mortality risks from cancers of the rectum and distal colon than

Colonoscopy

The disadvantage of flexible sigmoidoscopy is that only the distal 65 cm of the colon can be examined. This limitation may not, however, be critical, since evidence suggests that distal adenomas have a higher potential than proximal adenomas for malignant progression. Adenomas found on sigmoidoscopy could select patients for more extensive investigation (barium enema or colonoscopy). The alternative primary intervention is full colonoscopy. Although this procedure can decrease mortality by up

Heterogeneity between surgeons

Surgeons have played an important historical part in establishing clinical audit for postoperative morbidity and mortality. Despite this role, there is substantial variability between surgeons in the outcomes they achieve. A review of patients in Scotland managed by 13 surgeons, none of whom had a special interest in surgery for colorectal cancer, showed adjusted hazard ratios for all patients of 0·59-1·61.29 Similar variability was seen for outcome data from a series of more than 3000 patients

Surgery for colonic tumours

The colon and rectum are anatomically distinct and present different difficulties to the surgeon. The principles of radical surgical resection for colonic tumours have generally remained unchanged in the past few decades. The procedure involves ligation of the major vascular pedicle, obtaining tumour-free margins, and resection of any contiguous involved organ. Ligation of the vascular pedicle enables wide excision of lymph nodes draining the tumour, which is helpful in histopathological

Surgery for rectal cancer

Resection margins and surgical technique are especially important in determining outcome for rectal tumours compared with colon tumours, since anatomical constraints make surgery to remove tumours more complicated.

Williams and colleagues31 reported that only three of 50 patients undergoing abdominoperineal resection had involved lymph nodes distal to the gross tumour within the mesorectum. Despite this finding, MacFarlane and colleagues32 advocate complete excision of the mesorectum for rectal

The role of the pathologist

Pathological investigation of resected samples provides essential prognostic information that facilitates treatment decisions, such as the offer of adjuvant chemotherapy. The Dukes and Astler-Collins staging systems are widely used, but there is increasing international uptake of the tumour node metastisis (TNM) system. Additional factors that one would expect to be reported routinely in the pathology report include tumour grade, evidence of vascular invasion or lymphocyte infiltration, and the

Follow-up for colorectal cancer

There is continuing debate on the best method of follow-up for patients with colorectal cancer and current practices vary widely from single clinical surveillance to repeat laparotomy. This variety is of importance, since between 3% and 10% of patients have been estimated to be eligible for potentially curative surgery if relapse is detected at an early stage. However, randomised trials show no survival advantage for patients who are followed up intensively, such as by regular ultrasound,

Chemotherapy and radiotherapy

The anatomical differences between colon and rectal cancer and the ease of surgical intervention determine the pattern of recurrence, which defines the most appropriate adjuvant and relapse treatments. About 50% of recurrences of rectal cancer occur in the pelvis and are, therefore, amenable to local radiotherapy. By contrast, relapse of colon cancer is generally at distant sites-liver, lungs, and bone-and systemic chemotherapy seems more appropriate.

Chemotherapy

Fluorouracil has been the mainstay of chemotherapy for colorectal cancer for four decades. It is a prodrug which is converted intracellularly to various metabolites that inhibit synthesis of thymidine, DNA, and RNA (figure 2). Insights into its molecular pharmacology have led to several strategies to modulate its cytotoxic effects. The most important is coadministration with folinic acid, which increases the degree of inhibition of thymidylate synthase, depletes cellular thymidine, and induces

Systemic chemotherapy for advanced disease

Five randomised controlled trials of fluorouracil-based chemotherapy compared with the best supportive care in advanced colorectal cancer showed consistently better survival (by 3–6 months) and equal or better quality of life with early active treatment before symptoms occurred (figure 3).36 The overall benefits of chemotherapy are probably underestimated because of the high crossover to chemotherapy from the control groups when symptoms occurred (60% in the Nordic Study Group).

Many

Hepatic arterial chemotherapy for metastatic disease

Regional chemotherapy depends on the premises that most cytotoxic agents have steep dose-response curves and that high drug concentrations can be generated within a target organ (eg, liver) or body cavity (eg, peritoneum) because of differential drug clearance. Up to 20% of patients who relapse after an apparently curative resection of colorectal cancer present with disease macroscopically confined to the liver, which suggests that hepatic arterial chemotherapy could be a useful treatment.

Adjuvant chemotherapy

Despite 80% of patients with colon cancer having complete macroscopic clearance of their disease by resection, 50% of these patients develop recurrence, presumably because of disseminated micrometastases present at the time of surgery. Adjuvant chemotherapy is used to eradicate these circulating cancer cells before they become established and refractory to intervention. The Intergroup trial38 tested adjuvant chemotherapy after surgery with stage B colorectal cancer compared with surgery

Rectal cancer and radiotherapy

In advanced rectal cancer, radiotherapy can improve pain, staunch haemorrhage, and lessen tenesmus. In patients with inoperable local advanced disease, radiotherapy can convert 35–75% to resectability.44 Radiotherapy has been assessed as an adjuvant therapy. In the largest trial of radiotherapy before surgery (the Swedish Rectal Cancer Trial45), administration of 25 Gy over five fractions produced a relative decrease in local recurrence rate of 61% and an improvement in overall survival (58 vs

Cytotoxic agents

Raltitrexed is a quinazoline antifolate that directly inhibits thymidylate synthase. It gains entry to cells via the reduced folate carrier system and is polyglutamated to a potent, long-life inhibitor of thymidylate synthase. Raltitrexed administered by intravenous bolus (3 mg/m2) every 3 weeks has been compared with bolus fluorouracil and folinic acid (daily for 5 days every 4 weeks) in three randomised trials, including about 1500 patients with advanced colorectal cancer. Tumour response

Immunotherapy

Colorectal cancer has never been a very obvious target for immunotherapeutic intervention. Many cytokines have been tested and found to be inactive. Riethmuller and colleagues59 did a randomised study of a murine monoclonal antibody, which binds a tumour-specific cellsurface glycoprotein (17-1a), compared with a control as an adjuvant therapy in 189 patients with stage C colorectal cancer. Survival was significantly lengthened in the antibody-treated group (5-year survival rates 51 vs 36%,

Gene therapy

Cancer gene therapy encompasses many genetic manipulation strategies, including insertion of genes encoding cytokines, expression of tumour-suppressor genes to inhibit proliferation, and insertion of genes such as p53 that might lead to sensitisation to conventional cytotoxic therapy. Phase I trials have been completed of adenoviruses used to deliver the tumour-supressor p53 gene by hepatic arterial infusion.62 The virus was well tolerated and will enter Phase II trials in combination with

Biological agents

There are several innovative therapies in preclinical and early clinical development for colorectal cancer that are based on understanding of molecular and cellular carcinogenic events. These therapies include small-molecular-weight inhibitors of the K-ras oncogene, which have an antiproliferative effect by prevention of posttranslational modification of the mutant protien. Agents used include potent inhibitors of the matrix metalloproteinase enzymes essential for tumour invasion and metastasis

References (64)

  • RS Midgley et al.

    Systemic adjuvant chemotherapy for colorectal cancer

  • B Vogelstein et al.

    Genetic alterations during colorectal-tumour development

    N Engl J Med

    (1988)
  • L Cawkwell et al.

    The molecular biology and genetics of colorectal cancer

  • RJ Scott et al.

    Novel germline APC mutation in a large familial adenomatous polyposis kindred displaying variable phenotypes

    Gut

    (1995)
  • H Nagase et al.

    Correlation between the location of germline mutations in the APC gene and the number of colorectal polyps in familial adenomatous polyposis patients

    Cancer Res

    (1992)
  • PJ Morin et al.

    Activation of β-catenin-Tcf signalling in colon cancer by mutations in β-catenin or APC

    Science

    (1997)
  • R Kemler

    From cadherins to catenins: cytoplasmic protein interactions and regulation of cell adhesion

    Trends Genet

    (1993)
  • HFA Vasen et al.

    The international collaborative group on hereditary non-polyposis colorectal cancer (ICG-HNPCC)

    Dis Colon Rectum

    (1991)
  • R Fishel et al.

    The human mutator gene homolog MSH2 and its association with hereditary non-polyposis colon cancer

    Cell

    (1993)
  • CE Bronner et al.

    Mutation in the DNA mismatch repair homolog hMLH1 is associated with hereditary non-polyposis colon cancer

    Nature

    (1994)
  • NC Nicolaides et al.

    Mutations of two PMS homologues in hereditary non-polyposis colorectal cancer patients

    Nature

    (1994)
  • LA Aaltonen et al.

    Clues to the pathogenesis of familial colorectal cancer

    Science

    (1993)
  • J Doran et al.

    Bleeding patterns in colorectal cancer: the effect of aspirin and implications for faecal occult blood testing

    Br J Surg

    (1982)
  • JS Mandel et al.

    Reducing mortality from colorectal cancer by screening for fecal occult blood

    N Engl J Med

    (1993)
  • SJ Winawer et al.

    Progress report on controlled trial of fecal occult blood testing for the detection of colorectal neoplasia

    Cancer

    (1980)
  • J Kewenter et al.

    A randomised trial of fecal occult blood testing for early detection of colorectal cancer: results of screening and re-screening of 51,325 subjects

  • M Tokiue et al.

    Detection of K-Ras gene mutations in the stool sample of patients with colorectal tumours

    Gut

    (1994)
  • PA Newcomb et al.

    Screening sigmoidoscopy and colorectal cancer mortality

    J Natl Cancer Inst

    (1992)
  • JV Selby et al.

    A case-control study of screening sigmoidoscopy and mortality from colorectal cancer

    N Engl J Med

    (1992)
  • DF Ransohoff et al.

    Sigmoidoscopic screening in the 1990s

    JAMA

    (1993)
  • DH Bennett et al.

    Early diagnosis and screening

  • C McArdle et al.

    Impact of variability among surgeons on postoperative morbidity and mortality and ultimate survival

    BMJ

    (1991)
  • Cited by (0)

    View full text