USPSTF: Routine colonoscopy should cease at age 75

Updated colorectal cancer screening guidelines also include recommendations on method, frequency.

The U.S. Preventive Services Task Force has recommended that, in most cases, routine colorectal cancer screening should begin at age 50 and end at age 75 years, according to guidelines published recently online ahead of print in the Annals of Internal Medicine.

The task force updated previous guidelines published in 2002 based on a targeted systematic evidence review and a decision analytic modeling analysis. In the decision analysis, researchers found that decreasing the stopping age for routine colonoscopy from 85 years to 75 years decreased life-years gained by 1% to 4% but colonoscopy use, which carries several risks, decreased by 4% to 15%.

Although physicians may decide to screen patients older than 75 years of age, the task force recommended against screening in patients aged older than 85 years because of “moderate certainty that the benefits of screening do not outweigh the harms.”

Recommended interventions included the more sensitive fecal occult blood tests, flexible sigmoidoscopy with fecal occult blood tests or colonoscopy, according to the researchers. CT colonography and stool DNA testing did not receive recommendation. The task force wrote that insufficient evidence is available to assess the benefits and harms of these two screening options.

The guidelines also included recommendations for the intervals between screening options. The task force recommended that colonoscopy be conducted every 10 years; annual fecal occult blood screening between ages 50 and 75 years; and during the same age range, flexible sigmoidoscopy every five years, with a fecal occult blood test every two or three years.

The task force wrote that there is mounting evidence of increased sensitivity of fecal occult blood tests. They recommended what they said is a higher sensitive fecal occult blood test, Hemoccult SENSA (Beckman Coulter), or fecal immunochemical tests with comparable sensitivity. The Hemoccult II (Beckman Coulter) and flexible sigmoidoscopy alone were found to be less effective than other interventions.

Michael Pignone, MD, of the University of North Carolina and Harold Sox, MD, of the American College of Physicians in Philadelphia provided an accompanying editorial. They wrote that these recommendations diverge from the American Cancer Society guidelines, which recommend CT colonography and stool DNA testing. – by Rob Volansky

All from Ann Intern Med. 2008 Oct 6. [epub ahead of print.]

My own view, as an oncologist, is not really reflected in these guidelines. I believe that anything short of colonoscopy is unacceptable, unless virtual colonoscopy were to be elevated to that level. However, virtual colonoscopy is not yet generally available.

The issue here is this: are we making incremental improvements, or can we really eliminate a disease? Sigmoidoscopy and fecal occult blood tests will identify a subset of patients with early colon cancers but will miss many others. Fecal occult blood testing, while improving in its sensitivity and specificity, is hit or miss, and sigmoidoscopy is also imperfect. In screening algorithms, the recommendation for sigmoidoscopy is premised on the belief that if you do not see polyps in the distal colon, you can rest assured that there are no polyps in the proximal colon. That assumption has been disproven. If you rely on fecal occult blood tests, you may find a few cancers earlier, but you are also going to miss cases.

Having stated these criticisms, there is a need to aim for what is feasible. There is already poor adherence to guidelines, even without a rigorous colonoscopy recommendation. The difficulties with colonoscopy — including patient aversion, a shortage of capable endoscopists, risks and cost — make it unlikely that a recommendation of periodic colonoscopy for all individuals will be followed.

The ideal approach would be to enrich the population of people who need colonoscopy. An understanding of the molecular features of risk syndromes, for example, would potentially create a two-pronged screening strategy: periodic colonoscopy for high-risk individuals and less intrusive methods for the rest. Such an investment would pay dramatic dividends both in preventing colon cancer cases and saving the financial and personal costs of dealing with the disease.

– Alan Venook, MD

HemOnc Today Editorial Board member