Facing the challenge of colorectal Ca screening
Nationwide screening for colorectal cancer is expected to be rolled out next year
GP Dr Lorna Gold and consultant surgeon Mr Paul Stonelake discuss the subject
emind us of the epidemiology of colorectal cancer in the UK.
Colorectal cancer is the second commonest cause of cancer death in the UK, lung cancer being the commonest. Around 33,000 people are diagnosed with bowel cancer every year, and there are around 17,000 deaths per year. Mortality is largely stage dependent, so early diagnosis is worthwhile. Operative mortality is significant in these patients, especially with emergency presentation. Men and women are equally affected, but there appears to be an increasing frequency in men of around 1 per cent every year.
Only 5 per cent of cases occur under the age of 50, and the incidence increases with age. Over 75 per cent occur sporadically, but the risk is increased in patients with a family history (15-20 per cent of cases), genetic conditions such as familial adenomatous polyposis (1 per cent) and hereditary non-polyposis colorectal cancer (5 per cent), and longstanding colitis (1 per cent).
What symptoms should make us consider bowel cancer and use the fast-track
The high-risk symptoms and signs are summarised in the guidelines for referral. There are some local differences in age criteria. We use the following in Birmingham:
* rectal bleeding with change in bowel habit to looser stools and/or increased frequency of defaecation persistent for six weeks
* rectal bleeding without anal symptoms over the age of 50
* change in bowel habit to looser stools and/or increased frequency, without rectal bleeding, over age 50
* iron deficiency anaemia without an obvious cause (Hb<11g l="" in="" men="" or="">11g><10g l="" in="" postmenopausal="">10g>
* a definite palpable right-sided abdominal mass
* a definite palpable rectal (not pelvic) mass.
It should be noted that change in bowel habit to constipation (decreased frequency and harder stools) is not an indication for urgent referral. Abdominal pain without clear evidence of intestinal obstruction also does not require a fast-track appointment
Should the GP refer every patient who presents with recent onset rectal bleeding?
From the above criteria, patients under 50 with rectal bleeding need not be referred via the two-week system as the incidence of cancer in this group is extremely rare, although routine referral may be appropriate to manage the cause of the bleeding, most commonly haemorrhoids.
Patients of any age with bleeding associated with anal symptoms (pain, swelling, prolapse, itching) in the absence of other features suggesting bowel cancer need not be referred urgently. In general, bright red blood noticed on wiping is very likely to have an anal cause. Darker blood and/or blood mixed with the stool is more likely to be colonic or rectal and investigation is more urgent.
What should the GP tell the patient to
expect when they attend the rapid access colorectal clinic?
Typically, the initial consultation will include a history and abdominal examination, digital rectal examination and rigid proctosigmoidoscopy. Along with many colorectal units, we have moved to directly book flexible sigmoidoscopy for high-risk patients with rectal bleeding.
This involves a phosphate enema followed by inspection of the rectum, sigmoid and descending colon in the unsedated patient. This is an efficient means of investigating rectal bleeding, the cause for which is usually evident from the examination.
The majority of patients with benign problems can be dealt with on a 'one-stop' basis. For example, banding of haemorrhoids can be carried out at the same visit.
The one-in-15 patients who are found to have neoplasia are counselled by colorectal clinical nurse specialists at the point of diagnosis and then guided through the subsequent investigations, surgery and adjuvant therapy.
Patients whose symptoms include a change in bowel habit can expect to have their entire colon examined by either barium enema or colonoscopy, following bowel preparation.
How is colorectal cancer classified, and what is the prognosis for the different stages?
With some modifications, we still use the staging method of Dukes, which classifies the tumour according to extent of spread through the bowel wall (A, or B if through the wall), lymph node involvement (C1 or C2 depending on extent of lymph node involvement) and D when distant metastases are present. (see stage-dependent survival rates below).
It can be seen that there is a high likelihood of cure if the disease is treated early, but unfortunately the majority of patients present with more advanced disease.
For any given stage, operative mortality is around 20 per cent higher if patients require emergency rather than elective surgery. These facts have driven the recent initiatives for earlier diagnosis and screening.
What is known about the genetic aspects of colorectal cancer?
Familial adenomatous polyposis and hereditary non-polyposis colorectal cancer, the genetic conditions already mentioned, usually have a very strong family history.
Many patients seek advice about their cancer risk if there is colorectal cancer in the family. Generally their perception of risk is exaggerated, but there should be a low threshold for investigating patients with bowel symptoms and a family history of bowel cancer.
In asymptomatic individuals, screening with five-yearly colonoscopy should be considered where there has been more than one first-degree relative with bowel cancer, or just one who developed the disease before age 40.
Who is currently offered screening?
Patients are currently offered screening colonoscopy if there is a strong family history, a personal history of previous bowel cancer or polyps, and colitis after around 10 years from onset of symptoms. National guidelines have just been published.
A national screening programme has been under discussion for many years. Why has its introduction been delayed for so long?
The National Screening Committee reviewed the evidence and suggested pilot studies be set up. Although the evidence that screening reduces
mortality from colorectal cancer is not in doubt, there are still questions about resources and a national programme must be fully costed.
What procedure was used in the recent pilot study of colorectal cancer screening in Warwickshire, and what were the results?
The pilot studies used faecal occult blood testing. The technique involves inviting patients aged 50-69 for two-yearly testing using six squares of test 'card'.
If more than four test positive, the patient is referred for colonoscopy. If four or fewer test positive, the test is repeated following a meat-free diet.
Only 2 per cent of patients test positive, and only 6 per cent of these will be found to have bowel cancer on further investigation. If the results of the randomised studies are reproduced, a
10-20 per cent reduction in mortality from bowel cancer can be expected in the screened population. Patients are informed that a negative test does not exclude bowel cancer but that only one in every 1,000 patients tested negative will develop an 'interval' bowel cancer (ie, before the next screening round).
It has been suggested that providing faecal samples and having lower gastrointestinal endoscopy, possibly unnecessarily, would fall outside many people's definition of acceptable tests. Has this been a problem in pilot studies?
This has, perhaps surprisingly, not been a major problem. With the increased awareness that will inevitably follow the introduction of a national programme, it is likely that uptake will, if anything, improve further.
Is there a definite timescale for the introduction of colorectal cancer screening, and do any potential problems still need to be addressed before it can be rolled out nationally?
An evaluation team is due to report on the two pilot studies in Coventry and Dundee early in 2003. Results are not yet available and a final decision on a national screening programme can be expected following this report.
The evaluation team will report on the cost implications for introduction of screening and the effects this will have
on resources for facilities such as colonoscopy.
Stage-dependent survival rates
Dukes stage Proportion Five-year survival
A 11 per cent Approx 90 per cent
B 35 per cent 65 per cent
C 26 per cent 40 per cent
D 29 per cent 3 per cent
·Guidelines for colorectal cancer screening in high-risk groups.
This sets out national guidelines for screening colonoscopy.
·Kronborg O et al. Randomised study of screening for colorectal cancer with faecal-occult-blood test.
·Hardcastle JD et al. Randomised controlled trial faecal-occult-blood test screening for colorectal cancer.
Evidence that colorectal cancer screening reduces mortality.