Bowel test 'slashes cancer deaths'

A new bowel cancer test “cuts deaths by 40 per cent”, according to The Daily Telegraph. The newspaper estimates that offering the brief, one-off test to people aged 55 could save 3,000 lives a year and prevent a further 5,000 people from developing bowel cancer.

Several other newspapers also reported the results of a well-conducted, landmark trial into the screening test, which involves inserting a flexible camera into the bowel to check for abnormal growths (polyps) that may become cancerous. Those who attended screening were 33% less likely to develop bowel cancer and 43% less likely to die from it than those not invited for screening.

Several aspects of this test need consideration, such as the relatively low uptake rate (only 53% of those who were approached attended screening) and the invasive method used. However, bowel cancer is the third most common type of cancer in England, affecting around 31,000 people and causing around 13,000 deaths each year. Given this test’s potential to save thousands of lives through early treatment, its overall benefits seem clear.

Where did the story come from?

This study was carried out by Professor Wendy Atkin and colleagues of Imperial College London and other institutions in the UK. The study was funded by the Medical Research Council, NHS, Cancer Research UK and KeyMed. It was published in the peer-reviewed medical journal The Lancet.

The media accurately reflected the findings of this research.

What kind of research was this?

This study was a multicentre randomised controlled trial, which was carried out in 14 areas across the UK. It assessed whether the incidence of colorectal cancer and related deaths could be reduced by screening, using a technique called flexible sigmoidoscopy. This involves inserting a flexible camera tube into the back passage to view the rectum and lower bowel.

A randomised controlled trial (RCT) is the best way to test the effect of an intervention, in this case screening, on an outcome, such as colorectal cancer diagnosis and death. This RCT had additional strengths due to the large number of people it followed and the fact that they were recruited from all around the UK.

What did the research involve?

Between 1994 and 1999, researchers recruited adults aged 55-64 from 14 locations across the UK. Participants had no history of bowel disease or cancer, no symptoms suggestive of bowel cancer, no family history of bowel cancer and had not had a sigmoidoscopy or colonoscopy in the past three years.

Of the 368,142 eligible people who were mailed a questionnaire asking if they were interested in bowel screening, 194,726 (52.9%) said they were interested. After a further assessment for eligibility, 170,432 went on to be randomised into the study groups: either the screening appointment group (57,237 people) or the control group (113,195 people), who received no invitation to screening.

During screening appointments, sigmoidoscopy was used to identify any polyps (abnormal growths from the bowel wall that may develop into cancer). Polyps were either biopsied or referred for further investigation and management.

Data on cancer occurrences and date of death from 1999 and beyond were available through the NHS Central Register (NHSCR) and directly through the cancer registries, Hospital Episodes Statistics and the NHS Bowel Cancer Screening Programme databases. Further information on cause of death was obtained from death certificates and the Office for National Statistics (ONS). Different sources were cross-referenced to check their validity and steps were taken to independently confirm the accuracy of recorded causes of death. Incidence of colorectal cancer 10 years after screening was another outcome of interest examined.

What were the basic results?

Of those randomised to the screening group, 71% (40,674 people) attended their appointment. At screening, 95% (38,525 people) were discharged and not considered to be at risk, while high-risk polyps were detected in 5% of those screened (2,131 people). People with polyps went on to have either further investigations or management.

Over the course of screening and follow-up (average 11.2 years), 2,524 participants (1.5% of those randomised) were diagnosed with colorectal cancer (1,818 in the control group and 706 in the intervention group). In total, 727 people were certified to have died from colorectal cancer (538 in the control group and 189 in the intervention group).

Invitation to screening had reduced the risk of diagnosis of colorectal cancer by 23% (hazard ratio [HR] 0.77, 95% confidence interval [CI] 0.70 to 0.84) and the risk of dying from colorectal cancer by 31% (HR 0.69, 95% CI 0.59 to 0.82).

Those who attended their invited screening session (i.e. disregarding those who did not attend) had a 33% lower risk of a colorectal cancer diagnosis than those in the control group (HR 0.67, 95% CI 0.60 to 0.76). Screening attendees also had a 43% lower risk of death from colorectal cancer (HR 0.57, 95% CI 0.45 to 0.72). The researchers report that the incidence of colorectal cancer in non-attendees was very similar to that in those not invited (the controls).

The researchers calculated that 191 people needed to be screened to prevent one colorectal cancer diagnosis by the end of the study period. To prevent one death from colorectal cancer, 489 people would need to be screened.

How did the researchers interpret the results?

The researchers concluded that “flexible sigmoidoscopy is a safe and practical test”. They say that offering a single test between the ages of 55 and 64 “confers a substantial and long lasting benefit”.

Conclusion

This well-conducted, landmark trial investigated the effect of inviting healthy adults with an average age of 60 to attend a one-off lower-bowel examination, using flexible sigmoidoscopy, to screen for cancer.

Colorectal (large bowel) cancer is reportedly the third most common cancer worldwide and accounts for 600,000 deaths every year. As cancerous and pre-cancerous bowel growths are often symptomless in the early stages, visual detection during this early stage may be expected to reduce diagnoses of cancer and deaths due to the cancer.

The strengths of this study include the large number of people included in the trial and the average 11-year follow-up. While only 1.5% of the participants included developed colorectal cancer and only 0.43% died from it, the study’s sample size was sufficiently large to offer a reliable statistical comparison of the screened and unscreened groups. This trial has demonstrated that inviting people to screening reduces their risk of being diagnosed with colorectal cancer by 23% and their risk of dying from colorectal cancer by 31%. If they attend screening when invited, their risk reduction is even greater (respectively 33% and 43%, the higher figures quoted by the media).

An important and unavoidable obstacle to a screening programme is uptake of the test among the general population. The NHS already has a bowel-screening programme in place, which offers screening to all men and women aged 60 to 69 every two years. This involves sending out a “faecal occult blood test kit” to use at home and send back for analysis. The screening test detects small amounts of blood in faeces (invisible to the naked eye) which, if present, may be due to the presence of polyps or tumours that need further examination (often using colonoscopy). The pilot study that led to the introduction of the current screening programme found uptake in the pilot population to be about 57%.

The invasive nature of this new screening examination may be unacceptable to some people. This trial only included people who responded “yes” when asked if they would attend bowel screening if invited. Only 52.9% of them said yes, and they were subsequently invited. Of these people, only 71% attended when sent an appointment. From this experience, uptake in the eligible population as a whole could potentially be lower than 50%. Uptake and other important factors relating to screening, including possible anxiety associated with the screening process and waiting for results, will need to be carefully considered whenever a new programme is introduced or changes are made to an existing one.

The importance of these findings should not be underestimated, and the potential benefits of this screening programme are worthy of serious consideration. As the chief executive of Cancer Research UK said: “This is one of those rare occasions to use the word breakthrough. It is extremely rare to see clinical trial results as compelling as these.”