Published 31 May 2009, doi:10.1136/bmj.b2084
Cite this as: BMJ 2009;338:b2084

Editorials

Sigmoidoscopy screening for colorectal cancer

May reduce mortality, but longer term results are awaited

Evidence supporting colorectal screening, aside from randomised trials of faecal occult blood testing, comes mainly from observational studies. In the linked paper (doi:10.1136/bmj.b1846) interim results of the Norwegian Colorectal Cancer Prevention (NORCCAP) trial are presented.¹ NORCCAP is one of three ongoing trials of once only screening sigmoidoscopy.^{1 2 3} The findings suggest that the intervention may be effective in reducing mortality from colorectal cancer.

The NORCCAP trial randomised 55 736 people aged 55-64 years to usual care or to once only flexible sigmoidoscopy with or without a single round of immunochemical faecal occult blood testing. The primary outcome for this first report is 7 year cumulative incidence of colorectal cancer (anywhere in the colon); the secondary outcome is 6 year mortality from colorectal cancer. Colonoscopy, which was done in 21% of patients screened, was recommended for people with a positive screening result, defined as any polyp 10 mm or larger in diameter, any histologically verified adenoma or carcinoma, or a positive faecal occult blood test. Follow-up was obtained from the Cancer Registry of Norway and the Norwegian Cause of Death Registry, both of which are virtually 100% complete.

Of 13 653 people invited for screening sigmoidoscopy, 8846 (65%) underwent sigmoidoscopy; 41 092 individuals served as controls. The overall incidence of colorectal cancer did not differ significantly between screened and control groups (134.5 v 131.9 cases per 100 000 person years). The incidence of rectosigmoid cancer among attendees was reduced by 27% (58 v 79 per 100 000 person-years; P=0.10). The intention-to-screen analysis showed no significant difference in mortality from colorectal cancer (hazard ratio 0.73; 95% confidence interval 0.47 to 1.13), while the per protocol analysis showed a significant reduction in mortality for any colorectal cancer (0.41; 0.21 to 0.82) and for rectosigmoid cancer (0.24; 0.08 to 0.76). Compared with the control and non-attending groups, people with screen detected colorectal cancer tended to have earlier stage disease and had a lower case-fatality rate, though neither outcome was compared statistically.

In view of the challenges of doing clinical trials of cancer screening, how valid are these interim results? Several aspects of NORCCAP indicate that it is a high quality study, including its design (a population based, single invitation, randomised controlled trial) and use of blinded adjudication of colorectal cancer stage and cause of death. Follow up for both groups was achieved through two reliable national registries. Analysis by intention-to-screen provides an estimate of the effectiveness of sigmoidoscopy in the real world, where patients may or may not adhere. NORCCAP investigators thoughtfully considered control group contamination and length-time bias (the unintentional, preferential detection of slower growing, less aggressive lesions that have better prognoses) as explanations for some findings but present data that make both seem unlikely. If the study and findings are accepted as rigorous and valid, what explains the lack of a difference in cumulative incidence of colorectal cancer between screened and control groups, and how important is this finding?

Previous data on sigmoidoscopy screening come largely from high quality case-control studies suggesting a 60-80% reduction in mortality from distal but not from proximal colorectal cancer.^{4 5} A case-control study suggested that endoscopic procedures (sigmoidoscopy and colonoscopy) of the large bowel reduced incidence of colorectal cancer by 50% for up to six years.⁶ The Telemark clinical trial showed an 80% reduction in cancer incidence with sigmoidoscopy screening alone, but no effect on mortality from colorectal cancer, and an increase in all-cause mortality, due mostly to cardiovascular disease.⁷ The reduction in cancer incidence was observed after 13 years of follow-up. Cumulatively, these data suggest that sigmoidoscopy should be expected to reduce incidence and mortality related to colorectal cancer, at least for lesions within reach of the sigmoidoscope.

NORCCAP chose a high but appropriate bar by including the (prevalent) cancers initially detected by sigmoidoscopy in the measure of overall cumulative incidence. Excluding these prevalent cancers would have biased the results in favour of screening because of inability to exclude them in the unscreened control group. Using overall (as opposed to distal) incidence further raises the bar, since sigmoidoscopy would be expected to have a greater effect on incidence of distal cancers. With a smaller benefit on overall incidence expected, more time may be needed to detect it.

Since screening is expected to increase the detection of early stage curable cancer, we might expect a reduction in cancer mortality to occur before the reduction in incidence that results from removal of adenomas. In a trial of annual guaiac-based faecal occult blood test screening, a reduction in mortality from colorectal cancer was reported seven years earlier than a reduction in incidence.^{8 9} Although the difference between the groups was not significant, the reduction in overall mortality from colorectal cancer in NORCCAP is encouraging. The per protocol analyses more clearly indicate a reduction in mortality, but these findings may be prone to selection bias. Longer follow up will clarify these effects.

What are the implications of the current findings? Since site specific cancer mortality is generally considered the most appropriate end point for evaluating screening interventions,¹⁰ we should be encouraged by NORCCAP’s interim findings. Evidence to date strongly suggests that one time screening sigmoidoscopy can reduce incidence and mortality from distal colorectal cancer and may be a legitimate strategy. The magnitude and duration of benefit have yet to be ascertained, along with acceptability in different populations and the effort and cost needed for implementation. In the meantime, data on risk reduction may be useful for clinicians as they discuss screening options with their patients, for people estimating cost effectiveness, and for policy makers. We await the further results of this landmark trial and of the other ongoing trials of sigmoidoscopy screening.

Cite this as: BMJ 2009;338:b2084

¹ Indiana University Medical Center, Indianapolis, IN, USA, ² Regenstrief Institute, Indianapolis, IN 46202-5121, USA

Research: doi:10.1136/bmj.b1846

---

Competing interests: None declared.

Provenance and peer review: Commissioned, not externally peer reviewed.

References

1. Hoff G, Grotmol T, Skovlund E, Bretthauer M, for the Norwegian Colorectal Cancer Prevention Study Group. Risk of colorectal cancer seven years after flexible sigmoidoscopy screening: randomised controlled trial. BMJ 2009;338:b1846[Abstract/Free Full Text]
2. Segnan N, Senore C, Andreoni B, Aste H, Bonelli L, Crosta C, et al. Baseline findings of the Italian multicenter randomized controlled trial of "once-only sigmoidoscopy"—SCORE. J Natl Cancer Inst 2002;94:1763-72.[Abstract/Free Full Text]
3. UK Flexible Sigmoidoscopy Screening Trial Investigators. Single flexible sigmoidoscopy screening to prevent colorectal cancer: baseline findings of a UK multicenter randomized trial. Lancet 2002;359:1291-300.[CrossRef][Web of Science][Medline]
4. Selby JV, Firedman GD, Quesenberry CP Jr, Weiss NS. A case-control study of screening sigmoidoscopy and mortality from colorectal cancer. N Engl J Med 1992;326:653-7.[Abstract]
5. Newcomb PA, Norfleet RG, Storer BE, Surawicz TS, Marcus PM. Screening sigmoidoscopy and colorectal cancer mortality. J Natl Cancer Inst 1992;84:1572-5.[Abstract/Free Full Text]
6. Muller AD, Sonnenberg A. Prevention of colorectal cancer by flexible endoscopy and polypectomy. A case-control study of 32,702 veterans. Ann Intern Med 1995;123:904-10.[Abstract/Free Full Text]
7. Thiis-Evensen E, Hoff GS, Sauar J, Langmark F, Majak BM, Vatn MH. Population-based surveillance by colonoscopy: effect on the incidence of colorectal cancer. Telemark Polyp Study I. Scand J Gastroenterol 1999;34:414-20.[CrossRef][Web of Science][Medline]
8. Mandel JS, Bond JH, Church TR, Snover DC, Bradley GM, Schuman LM, et al. Reducing mortality from colorectal cancer by screening for fecal occult blood. Minnesota Colon Cancer Control Study. N Engl J Med 1993;328:1365-71.[Abstract/Free Full Text]
9. Mandel JS, Church TR, Bond JH, Ederer F, Geisser MS, Mongin SJ, et al. The effect of fecal occult-blood screening on the incidence of colorectal cancer. N Engl J Med 2000;34:1603-7.
10. Cuzick J, Cafferty FH, Edwards R, Moller H, Duffy SW. Surrogate endpoints for cancer screening trials: general principles and an illustration using the UK Flexible Sigmoidoscopy Screening Trial. J Med Screen 2007;14:178-85.[Abstract/Free Full Text]

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    StumbleUpon    Technorati    What's this?

Relevant Article

Risk of colorectal cancer seven years after flexible sigmoidoscopy screening: randomised controlled trial

Geir Hoff, Tom Grotmol, Eva Skovlund, Michael Bretthauer, and for the Norwegian Colorectal Cancer Prevention Study Group
BMJ 2009 338: b1846. [Abstract] [Full Text] [PDF]

This article has been cited by other articles:

• (2009). More Evidence Supports Flexible Sigmoidoscopy Screening. JWatch Gastroenterology 2009: 3-3 [Full text]  
• (2009). Colorectal Cancer After Screening with Flexible Sigmoidoscopy. JWatch General 2009: 5-5 [Full text]  

Online poll

Find out more >>