The Good, the Bad, and the Indeterminate in CRC Screening

Pam Harrison

March 04, 2019

Screening adults from the age of 50 onward to detect early colorectal cancer (CRC) has long been recommended in the United States. A study among older adults in the Veterans Affairs healthcare system found that colonoscopy reduced CRC mortality by up to 70% despite suboptimal uptake.

Any argument against the use of screening colonoscopy would therefore seem churlish, would it not?

Well, yes and no. Although it is likely that colonoscopy will emerge as the best screening strategy in the battle against CRC, "if you want randomized trial evidence supporting colonoscopy, there won't be anything out there for at least 3 to 4 more years," Michael Bretthauer, MD, University of Oslo, Norway, told Medscape Medical News.

Bretthauer is the principal investigator of a population-based CRC screening study being conducted in Poland, Norway, the Netherlands, and Sweden. Interim results from this study, published in 2016, indicated that 0.5% of more than 31,000 participants who underwent colonoscopy had CRC at the time of screening. In addition, some 30% of those screened had adenomas on screening, and almost half had some form of polyp.

So, it's no wonder that colonoscopy reduces CRC mortality rates. Most CRCs develop from benign adenomas, and their removal during colonoscopy has been shown to prevent about 80% of CRCs.

However, many of the participants in the study by Bretthauer and colleagues refused to undergo colonoscopy. Acceptance rates ranged from a low of about 23% in the Netherlands to a high of about 61% in Norway. The old truism that the best pill in the world won't work if people refuse to take it becomes critical in any appraisal of the benefits of colonoscopy, as several trial investigators pointed out.

"To carefully evaluate the balance of benefits and harms of colonoscopy screening, randomized trials are imperative," Bretthauer and colleagues write.

Because that evidence is not yet in, European guidelines do not yet recommend colonoscopy as a screening strategy for the average-risk person.

With evidence from randomized trials in support of colonoscopy still some years away, it's not unreasonable to discuss the potential limitations of colonoscopy, even where expertly applied. Intuitively, most would feel that a test that allows physicians to visualize the entire colon must be superior to methods that only visualize half the colon (flexible sigmoidoscopy) or that assess blood in the stool (the guaiac-based fecal occult blood test [gFOBT] or the fecal immunochemical test [FIT]).

However, it depends on which part of the colon practitioners explore — the proximal colon, or the distal colon, which includes the sigmoid colon and the rectum.

"Honestly speaking, when you think about the proximal side of the colon, the lesions or precursor lesions that might be there are much more difficult to detect and much more difficult to resect than those in the distal colon," Michael Kaminski, MD, PhD, Institute of Oncology, Warsaw, Poland, told Medscape Medical News.

Older observational studies have suggested that deaths from CRC that are prevented by colonoscopy are almost entirely limited to deaths from cancers that develop on the left side of the colon, not the right side.

Even if newer studies refute these findings — for example, Bretthauer and colleagues reported high detection rates for low- and high-risk adenomas in both the proximal and distal colon — there is no question that there is significant variation among operators. How competent individual endoscopists are at screening colonoscopy can affect the likelihood that they will detect any lesions at all, regardless of their location.

"When you look at this from a population perspective, the thing that counts the most is how much a test is accepted by society," Kaminski reiterated.

In Europe, acceptance of screening colonoscopy could be a long way off, given relatively recent evidence that many Europeans are risk-adverse to a test as invasive as colonoscopy.

Screening Sigmoidoscopy

In contrast to colonoscopy, there is good evidence that CRC screening with flexible sigmoidoscopy reduces both incidence and mortality. Flexible sigmoidoscopy is still in use in some countries and is still recommended in the United States, although it is rarely used.

In one of the latest of these trials — the Prostate, Lung, Colorectal, and Ovarian Cancer trial — US investigators reported that at a median follow-up of approximately 16 years for CRC incidence and 17 years for CRC mortality, the incidence was 18% lower and mortality was 25% lower among adults who underwent flexible sigmoidoscopy at baseline and again at 3 or 5 years.

Unfortunately, the reduction in mortality was limited to lesions detected in the distal colon; there was no significant effect on mortality from lesions in the proximal colon. Importantly as well, the reduction in mortality was also significantly greater in men than in women (P = .04), as well as in patients aged 65 to 74 years at baseline compared to younger patients (P = .01).

"Flexible sigmoidoscopy is still one of the most effective proven screening [methods] for CRC that we have," co–principal investigators Eric Miller, PhD, and Paul Pinsky, PhD, both of the National Cancer Institute, Rockville, Maryland, told Medscape Medical News.

That, however, has not stopped the precipitous decline in the use of flexible sigmoidoscopy as a screening tool in the United States. Currently, only about 5% of all CRC screening is done by this method. "In the US culture, we tend to like to do overkill," acknowledged Pinsky. At least part of the reason for its dead-in-the-water status is the assumption that being able to visualize the whole colon (with colonoscopy) is better than being able to visualize only half (as with sigmoidoscopy).

Furthermore, other studies have shown that flexible sigmoidoscopy, offered as a one-time screening test, effectively reduces CRC incidence and mortality in men, but it has little or no effect in women, at least in Norway. This may reflect the possibility that the risk of developing lesions in the proximal colon is higher in women than in men — although not dramatically so — and thus the risk of a proximal lesion being overlooked on colonoscopy and not being visualized on sigmoidoscopy is also greater, Miller suggested.

"In general, the belief is that colonoscopy should be more effective than sigmoidoscopy, but we really need proof of this, and this proof will not be available for many years to come," Kaminski reaffirmed.

Noninvasive Stool Tests for CRC Screening

In contrast to the invasiveness of both colonoscopy and sigmoidoscopy, noninvasive stool tests can be used for CRC screening. Both gFOBT and FIT are still fairly widely used in Europe, and both are recommended as a screening tool in the United States.

The newer FIT test is much more accurate in detecting cancer and adenomas than gFOBT, although it has not been compared to colonoscopy in a randomized control trial.

This will be rectified in a few years, as one of the principal investigators of the COLONPREV study, Enrique Quintero, MD, PhD, Hospital Universitario de Canarias, Tenerife, Spain, pointed out to Medscape Medical News. The co–principal investigator of the study is Antoni Castells, MD, PhD, Hospital Clinic de Barcelona, Spain.

COLONPREV is comparing FIT performed every 2 years to one-time colonoscopy in reducing CRC-related mortality in an average-risk population.

An interim report from this study, published in 2012, noted that roughly 26,000 men and women have been assigned to undergo colonoscopy, and roughly the same number have been assigned to undergo FIT. As has been true for other European trials, only about one quarter of participants assigned to undergo colonoscopy in the study agreed to go through with it. By contrast, more than one third of patients assigned to undergo FIT did so. Interestingly, CRC detection rates were identical in both groups, at 0.1%, although slightly more advanced adenomas were found in the colonoscopy group than in the FIT group, at 1.9% vs 0.9% (P < .001).

Importantly, though, far fewer nonadvanced — and possibly clinically insignificant — adenomas were found in FIT participants, at only 0.4%, compared to almost 10 times that number, at 4.2%, in the colonoscopy group (P < .001). Of those who were screened by FIT, roughly 7% tested positive. The majority of these patients subsequently underwent colonoscopy. The fact that colonoscopy detected more advanced and nonadvanced adenomas than FIT is understandable, given that it allows for direct visualization of the colon, Quintero acknowledged.

"However, the important thing is that the detection rate for early CRC was the same [between the two techniques]," he added. Also significant is that the first round of FIT in the study detected about half the number of advanced adenomas as were detected by colonoscopy. With additional FIT screens as participants move through the trial, more high-risk adenomas should be detected, Quintero also observed. Perhaps as importantly, FIT was far less likely to detect clinically insignificant, low-risk adenomas, sparing patients — and Spain's overwhelmed endoscopy clinics — the burden of having to pursue follow-up colonoscopy in a significant proportion of patients.

This, in a nutshell, is what Quintero sees as one of the major advantages of FIT.

In the United States, it is recommended that patients with low-risk adenomas undergo surveillance colonoscopy, effectively increasing the load for endoscopic units across the country and possibly interfering with the ability of symptomatic patients to access timely screening. Using FIT up front means that patients with clinically insignificant lesions will not be clogging up the queues, at least not right away, keeping the lines open for those who really need colonoscopy, he suggested.

The other important advantage of FIT is that people (at least in Europe) are apparently much more willing to participate in CRC screening with FIT.

"The lower participation rate in the colonoscopy group [seen in COLONPREV] and the recurrent nature of FIT screening may reduce the apparent advantage of colonoscopy," Quintero and colleagues opine.

Future Projections

In America, it is highly likely that colonoscopy will continue to be the dominant CRC screening method, regardless of what the randomized studies show.

Findings from sigmoidoscopy studies that show that women do not benefit much, if at all, from this approach have policy makers in the United Kingdom, Norway, and Italy scrambling to modify their recommendations for CRC screening. In those countries, recommendations will probably no longer include sigmoidoscopy because of its poor performance in women.

Kaminski noted that in Poland, a new screening project does not include sigmoidoscopy. Individuals are first offered colonoscopy, and if they refuse, they are offered FIT.

Although evidence from randomized trials is eagerly awaited by those with a vested interest in keeping CRC rates down, the medical community in the United States is already convinced colonoscopy is the only way to go.

"Even though there is no randomized trial evidence, most of us feel that there is as much circumstantial evidence for colonoscopy as there is for sigmoidoscopy in terms of its benefit," Pinsky said.

"So while colonoscopy is resource intensive, we don't want to lose sight of the fact that it's been a real success story in cancer overall, second probably only to cervical cancer in terms of the effect it has had on reducing the incidence of and mortality from CRC," he added.

Bretthauer is a member of the scientific advisory board of Exact Sciences and has received funding for equipment for studies from Olympus, Fujinon, Falk Pharma, and CCS Healthcare. Kaminski has received personal fees from Olympus, Alfa Sigma, Norgine, and Fujifilm. Quintero has acted as an advisor to Sysmex España. Miller and Pinsky have disclosed no relevant financial relationships.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.