The prevention successes achieved by colorectal cancer screening have been well documented, but the debate persists regarding the optimal screening method, and new Canadian task force guidelines reignited the issue earlier this year.
A recurring theme underlies the ongoing and sometimes acrimonious discussion: To what extent should doctors insist on colonoscopy, rather than allowing patients to select a method that they are more likely to consistently follow?
Colorectal diagnoses and deaths continue to decline due to both improved treatment and to screening with one of several methods: a stool test, sigmoidoscopy, or colonoscopy. The tests, typically recommended for adults ages 50 and older, vary in terms of how much patient time and bowel preparation are required, as well as in cost and how frequently they need to be repeated. Moreover, randomized trials comparing strategies are limited and no randomized data are yet available about colonoscopy versus other methods, although several studies are now ongoing.
In March, physicians in Canada cited that lack of randomized evidence when they issued updated guidelines that recommended against colonoscopy in average-risk adults ages 50 to 74. Instead, the Canadian Task Force on Preventive Health Care suggested a stool test every 2 years or a sigmoidoscopy every decade. “Colonoscopy will be revisited once randomized findings become available, but that won't be for another 5 to 10 years,” said Maria Bacchus, MD, who chaired the working group that developed the guidelines, which were published March 15 in CMAJ. (A chart, “Comparison of guidelines for colorectal cancer screening methods and intervals,” reprinted from that paper, is Table.)
“There are lots of trials with interventions that people have thought would be effective but in the end turned out not to be effective. I suspect that it [colonoscopy] is going to be effective. But before making a recommendation, we want to confirm this and know how much more so. The bottom line is for people to get screened,” said Dr. Bacchus, echoing a sentiment also conveyed in the draft recommendations of the U.S. Preventive Services Task Force. Those recommendations, published in October 2015, did not rank the screening tests in any preferred order and instead stressed the importance of getting some kind of test. Nearly 28% of U.S. adults ages 50 and older have never been screened, according to the Centers for Disease Control and Prevention.
Often due to time constraints, primary care doctors may not raise the option, said Thomas Imperiale, MD, FACP, a gastroenterologist and a professor of medicine at Indiana University School of Medicine in Indianapolis. “If they discuss screening, some providers don't present options,” he said. “They'll say, ‘It's time for your colonoscopy.’”
Stool test considerations
After lung cancer, colorectal malignancies are the most common type of cancer that impacts both men and women, diagnosed in nearly 5% of adults and responsible for nearly 50,000 deaths annually, according to the National Cancer Institute. The median age at diagnosis is 68; two-thirds of patients live at least 5 years. But if the cancer is caught before it has spread beyond the colon or rectum, the 5-year survival rate is 90%.
Patients appear more likely to agree to screening via stool test, according to the results of a Spanish study published in the New England Journal of Medicine on Feb. 23, 2012, which found that participation rates were 34.2% for fecal immunochemical testing (FIT) versus 24.6% for colonoscopy.
In recent years, the FIT method has been increasingly preferred over the traditional fecal occult blood test (FOBT), said Douglas Corley, MD, PhD, a gastroenterologist and a research scientist at Kaiser Permanente Northern California in Oakland. FIT is more sensitive and the patient doesn't have to comply with dietary restrictions, such as avoiding red meat, that might trigger a false positive with FOBT, he said.
The Canadian task force, which only looked at randomized controlled trials to answer the question of effectiveness of screening tests, relied on the GRADE (Grading of Recommendations Assessment, Development, and Evaluation) system to assess the quality of the evidence and to determine the strength of the recommendations. Where randomized data were not available, the U.S. Preventive Services Task Force incorporated other findings, such as those from observational studies.
Neither task force stated a screening preference for FIT versus FOBT. The Canadians recommended some method of stool testing every 2 years and the U.S. task force's draft recommendations preferred annually.
In reaching its stool test recommendation, the Canadian task force cited a meta-analysis of studies involving FOBT finding that stool testing prevented 2.7 colorectal cancer deaths per 1,000 adults screened.
One advantage of stool testing is that it can be completed without the patient coming to the doctor's office, Dr. Corley said. He served as the principal investigator on a study that evaluated such a widespread outreach effort, in which FIT kits were mailed to slightly more than 670,000 Kaiser Permanente members in northern and southern California.
The results, published online Jan. 26 by Annals of Internal Medicine, found that 48.2% of participants returned the kits within the first year. Of those who started screening, they tended to continue—more than 75% completed the tests in the next 3 years that the kits were mailed.
Kaiser Permanente Northern California, which began mailing FIT kits to its members when they reach age 50, has achieved a notably high rate of colorectal screening, according to Dr. Corley. As of 2011, more than 80% of members ages 50 to 75 were up to date on colorectal screening, using FIT or some other method.
Some patients prefer to get a colonoscopy every decade so they don't have to worry about screening more frequently, Dr. Corley said. Other patients are fine with an annual stool test, as they'd prefer to avoid the time and bowel preparation involved with colonoscopy. “It becomes a matter of patient choice,” he said.
Plus, giving people options besides colonoscopy appears to improve screening rates, according to findings from a 2012 study involving 997 patients and published in Archives of Internal Medicine. The screening group that was offered colonoscopy only had the lowest rate (38%) of getting the test within 12 months compared with the group offered FOBT (67%) or the choice of either FOBT or a colonoscopy (69%). An analysis along racial and ethnic lines found that white patients were more likely to undergo a colonoscopy than nonwhite patients.
The colonoscopy tug of war
While both the U.S. task force guidelines and those from Canada reference the third option of sigmoidoscopy, there are typically just 2 options for patients and clinicians to consider: FIT and colonoscopy, Dr. Imperiale said. Sigmoidoscopy is considered less frequently, such as when the physician wants to examine at least the lower portion of the colon and the patient prefers to avoid the more extensive bowel preparation and sedation typically involved when examining the entire length through colonoscopy, he said. “Sigmoidoscopy examines a portion of the colon that for many people reflects what's going on in the rest of the colon, but that's not true for everyone,” he said.
Dr. Imperiale is one of the researchers with the CONFIRM (Colonoscopy Versus Fecal Immunochemical Test in Reducing Mortality From Colorectal Cancer) study, 1 of 4 ongoing colonoscopy studies cited by the Canadian task force members. The U.S.-based study, which is still recruiting with a goal of 50,000 veterans, will compare 1-time screening with colonoscopy and annual screening with FIT. (Dr. Imperiale is a local site investigator and a member of the trial's executive committee.)
CONFIRM's estimated completion date is 2027. Another, similarly designed study, involving more than 50,000 patients in Spain, is anticipated to wrap up a bit sooner, by 2021. Meanwhile, the colonoscopy debate has proven to be particularly fierce in the United States, with conflicting guidance. While the U.S. Preventive Services Task Force doesn't currently prioritize any method, the American College of Gastroenterology guidelines describe a colonoscopy every 10 years beginning at age 50 as the “preferred” screening strategy.
No other test stays a step ahead of cancer, removing worrisome polyps before they can develop into a malignancy, said David Bernstein, MD, FACP, a gastroenterologist and vice-chair of medicine at Northwell Health on Long Island, N.Y. If patients refuse a colonoscopy, they should be offered another screening option, he said.
But Dr. Bernstein said he was troubled by the ethics of randomizing patients to colonoscopy versus stool tests in ongoing research trials. “As a gastroenterologist, I have issues with the end points of these trials,” he said. “Why are we having a decline in colorectal cancers? Because we are taking out polyps before they become cancers. Mortality isn't necessarily the best end point.”
Until the randomized trial data are available, though, “We do not know whether colonoscopy is superior to annual FIT,” Dr. Imperiale said. In the meantime, risks and benefits should be considered with each patient, he said. Some people are averse to collecting the samples needed for stool testing. With others, the time needed for colonoscopy bowel prep and the procedure is a deterrent.
Plus, there's a small but still real risk of complications, Dr. Imperiale said. He cited a couple of studies, including 1 that looked at 97,091 outpatient colonoscopies in Canada and was published in 2008 in Gastroenterology. Researchers found that the rates of bleeding and perforation were 1.64 patients per 1,000 screened and 0.85 patient per 1,000 screened, respectively.
Another consideration is that advancing technology means that colonoscopy can detect smaller and smaller adenomas that will likely never develop into cancer, Dr. Imperiale said. But the detection of those adenomas leads to a surveillance callback cycle of colonoscopies every 3 to 5 years, he said, with considerable expense and uncertain patient benefit.
While the primary focus of the Canadian guidelines was to analyze randomized data for screening effectiveness, Dr. Bacchus acknowledged that practical resources also were considered, albeit secondarily. The wait times for colonoscopies in Canada have gotten longer in recent years, she said. While a gastroenterologist is required to perform a colonoscopy in Canada, a sigmoidoscopy can be done by a primary care doctor or a nurse practitioner.
Thus, patients with symptoms or at higher colorectal cancer risk because of family history are prioritized, Dr. Bacchus said. “They are definitely ahead in line to have a colonoscopy,” she said, noting that the number of specialists and facilities would have to be ramped up if screening colonoscopies became more routine.
Assessing individual risk
Primary care doctors also should keep in mind that not all “average-risk” patients have the same colorectal cancer risk, Dr. Imperiale said. “We are in the era of personalized medicine, are we not?”
To better sort out relative vulnerability, Dr. Imperiale and his research colleagues at the Indiana University School of Medicine and the Regenstrief Institute have developed a risk-scoring system that is based on colorectal cancer risk factors and is quick and simple enough to be used by a busy primary care doctor, he said. To develop the scoring system, patients were given risk scores based on 5 variables: age, gender, family colorectal cancer history, smoking history, and waist circumference.
The participants, who were ages 50 to 80 and getting their first screening colonoscopy, were then divvied into 1 of 4 risk categories. Among those classified as at low or very low risk based on their scores, advanced neoplasms were identified in just 7%, according to the findings, published Sept. 1, 2015, by Annals of Internal Medicine. Conversely, advanced neoplasms were identified in one-quarter of those in the highest-risk category.
Doctors who were shown the risk-scoring approach say that it would help them initiate a discussion about the pros and cons of stool testing versus colonoscopy, Dr. Imperiale said.
“I think it [the risk-scoring approach] also sends the message to the patient that we are looking at features that are unique to you,” Dr. Imperiale said. “We're not just saying, ‘It's time for your colon cancer screening. Do you want to have a colonoscopy?’”