Screening for Colorectal Cancer


Synopsis: Hemoccult testing is not ideal as a screening tool, but flexible sigmoidoscopy and colonoscopy leave much to be desired in terms of cost.

Sources: Kronborg O, et al. Lancet 1996;348:1467-1471; Hardcastle JD, et al. Lancet 1996;348:1472-1478; Lieberman D, Sleisenger MH. Lancet 1996;348:1463-1464.

Two associated recent articles in the Lancet address the very important and topical issue of colon cancer screening.

The first study, by Kronborg et al from Denmark, was performed in 140,000 people, aged 45-75 years, who were followed over 10 years. The subjects were randomized to either screening with hemoccult-II tests or no screening. Colorectal cancer mortality, including deaths attributable to complications from colorectal cancer treatment, was significantly lower in the screening group than in the controls (mortality ratio, 0.82).

The second study, from Hardcastle et al in the United Kingdom, involved 152,850 individuals from the Nottingham area who were randomly allocated to either fecal-occult-blood screening (hemoccult) or to no screening and no intervention between 1981 and 1991. Screening group participants who had a positive test were offered full colonoscopy. At follow-up, they reported a 15% reduction in cumulative colorectal cancer mortality in the screening group.


The implications of these studies and several other recent studies1,2 on various strategies for colorectal cancer screening, which include a flexible sigmoidoscopy and colonoscopy, are carefully reviewed in the accompanying editorial.3 First and foremost, the editorial points out the relatively modest gain in mortality associated with hemoccult screening. For the two studies, this ranged between a 15% and an 18% reduction in colorectal cancer mortality. The major factors responsible for this relatively disappointing result were the poor sensitivity of the fecal-occult blood test and poor compliance among the screened groups. While recent modifications of hemoccult tests may improve sensitivity and specificity, it is clear that hemoccult testing falls far short of what would be regarded as ideal criteria for a screening tool. This has led others to recommend flexible sigmoidoscopy and perhaps even colonoscopy as a superior screening technique.4-6 There is indeed now a considerable body of evidence to suggest that colonoscopy may be the optimal screening technique in terms of sensitivity and specificity, although it leaves a lot to be desired in terms of cost. For example, other studies have suggested that sigmoidoscopy can reduce mortality from colon cancer located within the reach of the sigmoidoscope by 60%—colonoscopy would be expected to detect 90-95% of cancers.7,8 Given the enormous cost that would be involved in routine screening by colonoscopy, another approach that is beginning to be addressed is to attempt to identify those who are at greatest risk and to reserve colonoscopic screening for these groups. Very exciting advances in our understanding of the genetics and molecular biology of colon cancer may lead to the development of less invasive markers for risk.

These two large and important studies could be interpreted in one of two ways. On the one hand, they conclusively demonstrate a benefit in terms of a reduction in colon cancer mortality from fecal-occult blood testing. On the other hand, and perhaps more impressively, they demonstrate the limitations of this approach. If we are truly to have an effect on the incidence of and mortality from this common cancer, other approaches such as flexible sigmoidoscopy and colonoscopy will be necessary. Unfortunately, my suspicion is that the debate on this issue will move more and more from the research lab and physicians’ offices to the corporate board room.


1. Mandel JS, et al. N Engl J Med 1993;378;1365-1371.

2. Allison JE, et al. N Engl J Med 1996;334:155-159.

3. Lieberman D, Sleisenger MH. Lancet 1996;348: 1463-1464.

4. Winawer SJ, et al. N Engl J Med 1993;329:1977-1981.

5. Selby JV, et al. N Engl J Med 1992;326:653-657.

6. Winawer SH, et al. N Engl J Med 1996;334:82-87.

7. Lieberman DA. Gastroenterol 1995;109:1781-1790.

8. Levin B, Bond JM. Gastroenterol 1996;111:1381-1384.