Cover Story

Support for Noninvasive Options May Help Boost Colon Cancer Screening

Despite colonoscopy’s status as the primary colorectal cancer screening strategy, many Americans are unwilling or unable to undergo the procedure, for a variety of reasons including comorbidities, inconvenience, lack of access, cost concerns or fear of the procedure. It is in these populations that the variety of noninvasive options become important and necessary tools.

According to experts interviewed by Healio Gastroenterology and Liver Disease, the variety of available noninvasive CRC screening options could play a key role helping these populations get screened. With about one in three Americans still failing to undergo screening as recommended, experts agreed that good stewardship of noninvasive screening strategies is needed among gastroenterologists and primary care providers to help achieve the National Colorectal Cancer Roundtable’s goal of screening 80% of the eligible population.

By giving patients more options, those who may not otherwise undergo colonoscopy could still receive some form of screening, according to David A. Johnson, MD, professor of medicine and chief of gastroenterology at Eastern Virginia School of Medicine, and co-author of the 2017 MSTF guideline on CRC screening.

“Because patients seem to respond to opportunities of choice rather than absolute direction, we need a strategic menu that will allow us options beyond the primary preferred option of screening colonoscopy,” Johnson told Healio Gastroenterology and Liver Disease.

David A. Johnson

To address this need, Johnson and colleagues on the U.S. Multi-Society Task Force (MSTF) ranked the available CRC screening options for average-risk individuals at the eligible screening age using three tiers, “based on performance features, costs, and practical considerations.”

Considered the “cornerstones of screening,” first-tier options include colonoscopy every 10 years and annual fecal immunochemical testing (FIT). Second-tier options with certain disadvantages relative to first-tier tests include CT colonography every 5 years, the FIT-fecal DNA test (Cologuard, Exact Sciences) every 3 years, and flexible sigmoidoscopy every 5 to 10 years. The task force designated capsule colonoscopy every 5 years as a third-tier test due to its limited evidence and barriers to access, and suggested against the use of the Septin9 serum assay (Epi proColon, Epigenomics) due to lack of evidence.

Given that FIT is noninvasive, has a one-time sensitivity of 79% for detecting cancer, “fair sensitivity” for detecting advanced adenomas (about 30%), and low one-time cost (about $20), it is often the preferred test for programmatic screening, and the MSTF’s tier one ranking of the modality reflects its view of FIT as “an essential element of the CRC screening armamentarium for all practitioners.”

Positioning FIT

Unlike previous MSTF guidelines, the 2017 version no longer recommends the use of conventional guaiac fecal occult blood testing (gFOBT), as accumulating data show FIT detects CRC and advanced neoplasia with better sensitivity and comparable specificity vs. gFOBT. Results from a meta-analysis cited in an MSTF consensus statement suggested that FIT was superior to gFOBT for detecting both cancer (RR = 1.96; 95% CI, 1.2–3.2) and advanced neoplasia (RR = 2.28; 95% CI, 1.68–3.1).

“In the era of quality performance, FIT should replace FOBT; there’s no question,” Johnson said. “There may be lack of understanding or lack of uptake by some care providers, but it’s inappropriate at this point, as the evidence is quite strong.”

While the advantages of FIT include the convenience of performing it at home with no need for bowel prep, no medical restrictions, good sensitivity and low cost, it does carry notable limitations, including the need for repeated testing and follow-up colonoscopy in the event of a positive test. The MSTF noted this is a major drawback of stool-based tests, as achieving compliance with tests that need to be repeated at short intervals can be difficult in the U.S.

Another notable limitation of FIT is its lack of sensitivity for detecting serrated lesions, according to Johnson.

“We recognize that the serrated pathway has become an increasingly apparent pathway not only for primary colon cancer but also for interval colon cancers after people have had a colonoscopy,” he said.

A newer stool-based test approved in 2014, the FIT-fecal DNA test, can detect these high-risk lesions with better sensitivity, although with less specificity than FIT as well as a significantly higher cost, according to Seth A. Gross, MD, FACG, of NYU Langone Health.

Seth A. Gross

“The fecal DNA stool test is becoming more popular because it’s better at identifying advanced adenomas, whereas the primary aim of other noninvasive screening techniques is to diagnose colon cancer,” he said in an interview. “The stool DNA test is more sensitive to picking up advanced adenomas and sessile serrated polyps, which are important because in some situations they have a shorter pathway to moving onto colon cancer than a traditional precancerous tubular adenoma.”

FIT-fecal DNA

Cologuard, a tier 2 screening modality recommended every 3 years rather than every 1 year, is a good option, according to Thomas F. Imperiale, MD, of the Richard L. Roudebush VA Medical Center, the Regenstrief Institute, and the department of medicine at Indiana University School of Medicine.

“For people who prefer a less frequent noninvasive test, stool DNA FIT is the best option, though at a greater cost,” he said in an interview.

The test’s list price is $649, according to the Cologuard website. As of May 2017, it was covered by CMS and the major payors in the U.S., including Aetna, United Healthcare, Cigna, Humana and Tricare. Analysts said the test should be covered for about 95% of eligible patients this year.

Thomas F. Imperiale

“There are issues with the cost-effectiveness of this test, but the sensitivity of stool DNA is better than FIT alone,” Imperiale said.

In one study, the FIT–fecal DNA test showed a 1-time sensitivity for CRC of 92%, which is “the highest single-time testing sensitivity for cancer of any noninvasive, nonimaging CRC screening test,” according to the MSTF guideline. The study also showed it performed with 40% sensitivity for sessile serrated polyps greater than 1 cm in size.

“Where it has some drawback is with specificity, which is not as good as FIT, particularly in people older than age 65, but it is a very good option for people who don’t want to do a noninvasive test every year, and who are younger (< 65 years) in the screening age range,” Imperiale said.

The test manufacturer, Exact Sciences, is currently performing a longitudinal follow-up study “to make sure the 3-year interval is appropriate,” he added.

Comparative Studies Needed

The large number of different FIT tests currently available in the U.S. — 16, according to a recent report — poses a problem, as their efficacy relative to each other is currently unclear. To address this, a team of investigators across the U.S. recently announced they will compare the accuracy of different FIT tests to identify which is most effective. The study is funded through the Cancer Moonshot initiative.

“Our number one goal is to identify which FIT test is the most accurate,” Navkiran Shokar, MD, MPH, MA, of Texas Tech University Health Sciences Center El Paso, said in a press release. “But we are also hoping to build awareness about these newer, more convenient FIT tests and get more people screened for colorectal cancer.”

Similarly, a new study published in the American Journal of Gastroenterology comparing three fecal occult blood tests in more than 1,000 patients showed that the InSure FIT (Clinical Genomics) performed with superior sensitivity vs. the OC FIT-CHEK and Hemoccult II SENSA (26.3% vs. 15.1% and 7.4%, respectively, for detecting advanced colorectal neoplasia). Another recent study published in Gastroenterology compared nine quantitative FITs and found wide variability when using sensitivity and specificity thresholds recommended by the manufacturers, leading investigators to conclude that this heterogeneity in diagnostic performance “can be overcome to a large extent by adjusting thresholds to yield defined levels of specificity or positivity rates.”

Beyond FIT, published randomized trials directly comparing the outcomes of different CRC screening tests are also needed, and Johnson and colleagues on the MSTF noted that because of this lack of data, they took “practical considerations” into account when designating modalities into their given tiers.

Test availability appears to have factored into the tiering criteria in this guideline, according to Imperiale.

“Even though it seems like we’ve got a lot of good choices, what really is most readily available and best known in the U.S. at this time is colonoscopy and FIT,” he said. “Flexible sigmoidoscopy has, by far, the strongest and most consistent evidence supporting efficacy than any screening currently available, but no one’s getting screened with it in the U.S. It’s shown fairly good incidence and mortality reductions, but it’s essentially unavailable here, so the tiering criteria goes beyond published evidence about effectiveness. Unlike flexible sigmoidoscopy, there are no completed randomized controlled trials involving colonoscopy, which is why four trials involving colonoscopy are ongoing worldwide.”

Among these four large, long-term randomized comparative trials currently underway, Imperiale and colleagues recently completed enrollment in the CONFIRM study, which will compare the efficacy of screening colonoscopy vs. FIT for preventing CRC mortality among 50,000 average-risk VA patients over 10 years.

“We really don’t know which strategy is better, colonoscopy every 10 years or a FIT test every year,” Imperiale said in an interview. “We don’t know which is better in terms of uptake, adherence, and CRC morbidity and mortality reduction.”

Imperiale noted that CONFIRM is one of four ongoing global randomized trials comparing screening colonoscopy with FIT or other less invasive strategies, some of which have already reported interim results. One is the COLONPREV study in Spain comparing colonoscopy vs. FIT, with 10-year outcomes expected in 2021, and another is the Nord-ICC trial involving centers in Poland, the Netherlands, Norway, Sweden and Iceland, comparing colonoscopy with standard of care, with 15-year results expected in 2026.

Uptake

Patient compliance with the different tests is an important consideration of these large studies, for obvious reasons, according to Imperiale.

“People are very interested in comparing uptake, because a screening test can’t be effective if people don’t do it,” Imperiale said. “The most effective it can be is the extent to which it’s taken up, so if 30% of the colonoscopy group gets colonoscopy, the most it can reduce colon cancer incidence or mortality is by 30%. If 80% of the FIT group gets FIT done regularly, and it detects 75% of all CRC, FIT will outperform colonoscopy in the long-run. But we just don’t know the answers to these basic questions about uptake, adherence and effectiveness.”

Interim results from the COLONPREV study have provided some data on this topic, showing a 39% higher participation rate with FIT and the same CRC detection rate as colonoscopy (0.1% in each), but other data are conflicting. In contrast, a randomized clinical trial published last year in JAMA showed higher screening completion rates with colonoscopy vs. FIT outreach over 3 years, although both interventions were superior to usual care (38.4% for colonoscopy, 28% for FIT, 10.7% for usual care).

A major challenge to improving uptake of CRC screening is understanding and adequately addressing all the reasons why people aren’t being screened, according to Douglas G. Adler, MD, professor of medicine in the division of gastroenterology and hepatology at the University of Utah School of Medicine.

“There are lots of reasons why people don’t get screening colonoscopy,” he said. “Their primary care provider may not think to offer it. They may fear it will be uncomfortable, painful or embarrassing. And bowel prep is inconvenient. ... These are all classic barriers.”

Douglas G. Adler

People who live in more remote areas also face the added challenge of access, Adler said.

“Many small towns have no gastroenterologist,” he said. “If people live in rural Wyoming or Montana or Nevada, if the nearest GI doctor is hours away, they’re just not going to prioritize getting a colonoscopy if they have to designate 2 days to drive 4 hours, get a colonoscopy and stay overnight at a hotel.”

Cost-effectiveness

Cost concerns are yet another barrier preventing people from undergoing screening colonoscopy, and issues of cost-effectiveness underlie the usefulness of all screening strategies, according to Imperiale.

“Screening rates in the U.S. have been increasing more in community health centers and safety net hospitals in cities, largely because of FIT, but adherence with colonoscopy has leveled off and has even gone down in some cases because of co-pays and deductibles,” he said.

Although the Affordable Care Act has improved access to screening colonoscopy by fully covering it, patients are often liable for a co-pay if polyps are found and removed, as the procedure is then coded as a diagnostic colonoscopy. Similarly, health plans may deem colonoscopy diagnostic after a positive noninvasive test or if the patient is symptomatic.

While concerns about out-of-pocket costs may prevent patients from being screened, cost-effectiveness studies have consistently shown that CRC screening by any modality is cost-effective compared with no screening, given the high costs of CRC treatment. Both colonoscopy and FIT perform well in these cost-effectiveness analyses compared with other screening tests, but the newer options including CT colonography, FIT–fecal DNA and capsule colonoscopy “could reach cost-effectiveness by substantially increasing compliance,” according to the MSTF guideline.

The cost-effectiveness of screening approaches is important to consider, particularly when thinking about CRC screening on a global scale, according to Jonathan A. Leighton, MD, of the Mayo Clinic in Arizona.

“In many other countries their screening programs include the FIT test annually with or without flexible sigmoidoscopy every 5 to 10 years. It is probably the most cost-effective strategy overall, and the one that most countries can afford to do, because colonoscopy by itself does get expensive and is more resource-intensive. Cost is certainly one of the reasons annual FIT was recommended as a first-tier test, and it has a fairly high one-time sensitivity and a low first-time cost, all of which makes it attractive,” he said.

Jonathan A. Leighton

The cost-effectiveness of FIT makes it an especially viable way to expand the screening population, given recent calls to increase the percentage of individuals screened, Leighton said.

Lowering the Screening Age

Screening colonoscopy is also out of reach for many younger patients in whom CRC rates are notably on the rise, prompting some to call for lowering the screening age for average-risk individuals from 50 to 45 years. Similarly, racial disparities in CRC risk and screening has led the U.S. MSTF on CRC Screening to suggest lowering the screening age among black Americans to 45 in their recent guideline, but the best course of action to address it remains a matter of debate.

“I am definitely concerned about the evidence that CRC is occurring more frequently in patients under the age of 50, so I don’t think we can ignore that, and the risk appears to be higher in black people in particular,” Leighton said. “So, based on that we certainly need to consider lowering the screening age in certain sub-populations, but it requires further study.”

Others have made the same call. A recent study presented at UEG Week showed marked increases in the detection of neoplasia, polyps and adenomas in colonoscopy patients aged 45 to 49 years vs. those aged 40 to 44 years, which led investigators to recommend that CRC screening should begin at age 45 years.

According to Leighton, FIT shows the most promise for expanding the screening population in a cost-effective way.

“Lowering the screening age may turn out not to be as cost-effective, but from the standpoint of the individual patient I think it’s something that we have to strongly consider,” he said. “It at least needs to be further studied so that we can make an educated decision, but in the meantime, I think doing FIT plus or minus flexible sigmoidoscopy, which is what’s done in many parts of the world, would be a very reasonable approach to do it initially rather than necessarily going straight to colonoscopy.”

Imperiale does not believe that lowering the screening age is the appropriate response to this trend. In a recent editorial, he commented on a study of patients in the Korean CRC screening program, which found that average-risk patients aged 35 to 49 years with a positive FIT showed a higher risk for advanced colorectal neoplasia confirmed by colonoscopy vs. patients in their 50s. The investigators concluded that the results may justify beginning FIT screening at age 35.

In his editorial, Imperiale wrote that “lowering the screening age now would be an emotional response to a problem that needs reason and a greater understanding of etiologic factors, risk factors, and the balance of benefits and harms under varying conditions,” and that immediate efforts should instead focus on patient education and prevention.

“When you look at the numbers in that article, the ratio of false-to-true positive was 9 to 1, so you’d have to perform colonoscopy on 10 people to find one with an advanced neoplasm, nearly all of which were not cancer,” he said. “Most of what was found were non-advanced adenomas that would put people into surveillance programs beginning at young ages; the fact that we don’t know the natural history of advanced or non-advanced adenomas is a huge drawback. We know the natural history is not that bad, but studies from Germany suggest that the younger you are, the lower the transition rates from non-advanced adenoma to advanced adenoma and from advanced adenoma to cancer. Lowering the screening age could very well tip the benefit-to-risk ratio towards risk. The downsides are potentially substantial and for these reasons, I don’t think this would work in the U.S.”

While modeling studies show that it really is best to start screening around age 50 on average, the number itself is somewhat arbitrary, and there are inherent inefficiencies in choosing a single age to start screening for everyone, Imperiale said.

“If we could tailor the age at which to begin screening we would do better as far as efficiency, except it becomes more complicated to start younger in one demographic group and start older in another,” he said. “For example, men and women probably don’t require colon cancer screening starting at the same age. Men’s risks for advanced adenoma and colon cancer are 1.5 to 2 times that of women irrespective of age until around 70, where women start to catch up, but we don’t screen men earlier or women later given the challenges to implementing it.

“Even though the risk of colon cancer is rising in young people, it’s rising slowly and the risk is still an order of magnitude less than it is in people who are 50 and older. So, to say we should start screening earlier could cause a lot of mischief ... and it’s quite possible that the frequency of major complications would outweigh curable-stage cancer detection.”

Other Imaging Options

While colonoscopy and stool-based tests are the current mainstays of CRC screening in the U.S., other tier 2 and 3 imaging modalities play important roles in screening certain niches of individuals.

CT colonography, for example, is a tier 2 test that is recommended every 5 years, and has essentially replaced the use of double-contrast barium enema. It carries a lower risk for perforation than colonoscopy, and detects adenomas greater than 1 cm with sensitivity ranging from 82% to 92%.

This option is primarily used to screen patients who cannot undergo colonoscopy due to comorbidities, and to perform a same-day follow-up exam after an incomplete colonoscopy, according to Leighton.

“The nice thing about CT colonography is that it can visualize the entire colon but it doesn’t require sedation and you can safely perform it in anticoagulated patients,” he said. “Some people are concerned about the radiation, but it does use a low dose. A major advantage is you can often perform CT colonography to rule out any significant lesions on the same day as a colonoscopy after an incomplete exam.”

However, according to the MSTF, CT colonography has a limited impact, and is disadvantaged by the need for frequent follow-up colonoscopies and management of incidental extracolonic findings. “However, CT colonography appeals to a niche of patients who are willing to undergo bowel preparation and are concerned about the risks of colonoscopy,” the MSTF wrote.

Another emerging option that may be appropriate for a subgroup of patients who decline conventional colonoscopy is capsule colonoscopy, like Medtronic’s Pillcam, which the FDA approved in 2014 for patients with prior incomplete colonoscopies or those who cannot undergo colonoscopy or sedation. However, capsule colonoscopy is not approved for screening average-risk individuals, and lack of reimbursement currently impedes its widespread use, according to Gross.

“Until we have a CPT code for it, I don’t think that it will be widely available in the community, because the cost ends up going to the patient,” he said.

While the tier 3 option may be reasonable for patients who have incomplete colonoscopy, those who simply decline colonoscopy may not prefer capsule colonoscopy as an alternative, as it requires a more extensive bowel prep than conventional colonoscopy, Gross noted.

“One of the things that turns people off from colonoscopy is the bowel prep, and when you do the colon capsule it’s a more rigorous bowel prep,” he said. “So, I think there’s value in it, but again it’s going to be appropriate for a small subset of patients that can’t undergo traditional optical colonoscopy.”

Despite its limited use, Gross noted that colon capsule technology has significantly improved over time. In a meta-analysis of 14 studies involving 2,420 patients, he and colleagues showed that the sensitivity for detecting polyps greater than 6 mm (58% vs. 86%) and 10 mm (54% vs. 87%) increased substantially between first- and second-generation colon capsules.

“It’s an exciting advancement, and they’ve done so much more with the second-generation technology, in the sense of capturing more frames per second, so you’re able to get more images of the colon lining,” Gross said.

However, due to several disadvantages, the future role of capsule colonoscopy for screening remains uncertain, according to Leighton.

“In addition to the more intense bowel prep, the technical failure rate, at least in the U.S. study, was around 9%, so there’s a relatively high number of technical failures,” he said. “There is also some concern that it may have trouble detecting sessile serrated lesions. It has been shown to be useful after an incomplete colonoscopy, but usually not on the same day. Unlike CT colonography, where you can perform the procedure on the same day as a colonoscopy, it’s tough to do that with the capsule colonoscopy. So, I think the future is still in question. I would say that it is a reasonable option for patients who have incomplete optical colonoscopy.”

Similarly, Adler said the logistics of colonoscopy scheduling are a notable barrier preventing less invasive imaging strategies from being fit into the CRC screening algorithm.

“The biggest problem is the study is only diagnostic,” he said. “If a patient has a CT colonography or a capsule colonoscopy and a polyp is found, the provider can’t just call the GI lab and send the patient over for a colonoscopy. They’d have to be scheduled, likely for the following week unless they’re incredibly lucky, so the patient is getting two preps and two studies where maybe only one would have been enough. So, the idea of using a capsule to screen the colon for CRC or polyps is a phenomenal idea that just has really not taken off, because the logistics of the way colonoscopy works, while not insurmountable, are very difficult.”

Liquid Biopsy

Finally, while the MSTF recommended against the use of the Septin9 serum assay due to its inferior performance relative to FIT, experts agreed that an accurate blood test for CRC would be very valuable.

“Blood testing is kind of the holy grail,” Adler said. “A very high sensitivity, high specificity blood test with a good positive predictive value would be very widely adopted. However, it has been extremely difficult to develop such a test, and I suspect one would be used to detect cancer rather than polyps.”

While no such test is ready for primetime, experts agreed the effort is worthwhile. Recently at the ASCO Gastrointestinal Cancers Symposium, new data was presented showing a circulating tumor blood test (CellMax Life) detected precancerous lesions with 84% accuracy and cancerous lesions with 88% accuracy.

Investigators evaluated 620 individuals who were scheduled for routine colonoscopy or had a CRC diagnosis, 438 of whom had either adenomatous polyps or early- to late-stage colorectal cancers. The CellMax biomimetic platform (CMx) identified these lesions with 97.3% specificity. Sensitivity was 76.6% (95% CI, 67.9-83.5) for detection of circulating tumor cells in precancerous lesions and 86.9% (95% CI, 82.8-90.1) for stage I to stage IV cancers.

“The accuracy of the CellMax CTC blood test in terms of sensitivity and specificity is comparable or superior to stool based tests such as FOBT and FIT for detecting Stage I-IV cancers,” study investigators Wen-Sy Tsai, MD, PhD, assistant professor at Linkou Chang Gung Memorial Hospital in Taipei, Taiwan, and Ashish Nimgaonkar, MD, assistant professor of medicine at Johns Hopkins University, told Healio Gastroenterology and Liver Disease by email. “The key difference between this CTC blood test and stool based tests is its higher sensitivity in the detection of precancerous lesions. Stool based tests have roughly 40% or lower sensitivity in the detection of precancerous lesions.”

The test could cost less than $100, making it an affordable option, according to a company press release.

“Surveys conducted to find the cause of low screening rates demonstrated that most people would rather get a routine blood test than a routine colonoscopy or stool-based tests due to its convenience and non-invasive nature,” they said. “The advent of the CTC blood test is that it offers individuals a convenient and accurate test option so that they routinely get tested which is essential.”

The blood test is commercially available in Asia and the company is evaluating commercialization options in the U.S., Tsai and Nimgaonkar noted.

Education, Stewardship

While providers await the arrival of a viable blood test, supporting awareness and use of the currently available noninvasive screening options is important to maximize screening rates, according to Imperiale.

“Gastroenterologists need to be good stewards of colon cancer screening, so that their primary care counterparts don’t play the methods down,” he said. “Some providers may not give their patients a choice, perhaps because they don’t believe that these tests are effective, so we have to be good stewards of the different options for colon cancer screening and support primary care providers and patients in their choices.”

Adler agreed, but emphasized that patient education is just as important as patient choice.

“The biggest thing we can do is educate people,” he said. “Colon cancer, like a lot of malignancies, is asymptomatic until it’s very advanced, and I think that if people don’t really have a sense of the risk and danger of colon cancer they are less likely to be screened.”

Adler cited a billboard campaign in Utah cited by his patients as a reason for pursuing a colonoscopy. His patients, he said, also reached out to their primary care provider for the referral when they had not previously received it, “a good example of the patient driving the doctor,” he added.

“Campaigns to raise awareness among the public tend to be extremely successful,” Adler. “The biggest thing we can do is awareness and education. The goal was 80% by 2018, and we’re certainly far below that. So, a very large swath of this country who is eligible has still not had any colorectal cancer screening test. We’re in a good time in the sense that we have so many tests now at our disposal, it’s very hard for a patient to say they don’t want to be screened at all.”

Finally, Adler addressed the apprehension among some gastroenterologists that noninvasive testing may represent competition to their colonoscopy practice and a loss of screening colonoscopies.

“Noninvasive screening is actually good for GI business,” he said. “If you start screening more of the population, you’ll start finding more people who need colonoscopy, and you’ll probably be doing less diagnostic colonoscopy and more therapeutic colonoscopy, so I think it’ll actually be good for the population and good for GIs.” – by Adam Leitenberger

Disclosures: Adler, Gross and Johnson report no relevant financial disclosures. Imperiale reports that he receives grant support through Indiana University from Exact Sciences. Leighton reports consulting and research relationships with Given Imaging, Covidien and Medtronic related to colon capsule technology.

Despite colonoscopy’s status as the primary colorectal cancer screening strategy, many Americans are unwilling or unable to undergo the procedure, for a variety of reasons including comorbidities, inconvenience, lack of access, cost concerns or fear of the procedure. It is in these populations that the variety of noninvasive options become important and necessary tools.

According to experts interviewed by Healio Gastroenterology and Liver Disease, the variety of available noninvasive CRC screening options could play a key role helping these populations get screened. With about one in three Americans still failing to undergo screening as recommended, experts agreed that good stewardship of noninvasive screening strategies is needed among gastroenterologists and primary care providers to help achieve the National Colorectal Cancer Roundtable’s goal of screening 80% of the eligible population.

By giving patients more options, those who may not otherwise undergo colonoscopy could still receive some form of screening, according to David A. Johnson, MD, professor of medicine and chief of gastroenterology at Eastern Virginia School of Medicine, and co-author of the 2017 MSTF guideline on CRC screening.

“Because patients seem to respond to opportunities of choice rather than absolute direction, we need a strategic menu that will allow us options beyond the primary preferred option of screening colonoscopy,” Johnson told Healio Gastroenterology and Liver Disease.

David A. Johnson

To address this need, Johnson and colleagues on the U.S. Multi-Society Task Force (MSTF) ranked the available CRC screening options for average-risk individuals at the eligible screening age using three tiers, “based on performance features, costs, and practical considerations.”

Considered the “cornerstones of screening,” first-tier options include colonoscopy every 10 years and annual fecal immunochemical testing (FIT). Second-tier options with certain disadvantages relative to first-tier tests include CT colonography every 5 years, the FIT-fecal DNA test (Cologuard, Exact Sciences) every 3 years, and flexible sigmoidoscopy every 5 to 10 years. The task force designated capsule colonoscopy every 5 years as a third-tier test due to its limited evidence and barriers to access, and suggested against the use of the Septin9 serum assay (Epi proColon, Epigenomics) due to lack of evidence.

Given that FIT is noninvasive, has a one-time sensitivity of 79% for detecting cancer, “fair sensitivity” for detecting advanced adenomas (about 30%), and low one-time cost (about $20), it is often the preferred test for programmatic screening, and the MSTF’s tier one ranking of the modality reflects its view of FIT as “an essential element of the CRC screening armamentarium for all practitioners.”

PAGE BREAK

Positioning FIT

Unlike previous MSTF guidelines, the 2017 version no longer recommends the use of conventional guaiac fecal occult blood testing (gFOBT), as accumulating data show FIT detects CRC and advanced neoplasia with better sensitivity and comparable specificity vs. gFOBT. Results from a meta-analysis cited in an MSTF consensus statement suggested that FIT was superior to gFOBT for detecting both cancer (RR = 1.96; 95% CI, 1.2–3.2) and advanced neoplasia (RR = 2.28; 95% CI, 1.68–3.1).

“In the era of quality performance, FIT should replace FOBT; there’s no question,” Johnson said. “There may be lack of understanding or lack of uptake by some care providers, but it’s inappropriate at this point, as the evidence is quite strong.”

While the advantages of FIT include the convenience of performing it at home with no need for bowel prep, no medical restrictions, good sensitivity and low cost, it does carry notable limitations, including the need for repeated testing and follow-up colonoscopy in the event of a positive test. The MSTF noted this is a major drawback of stool-based tests, as achieving compliance with tests that need to be repeated at short intervals can be difficult in the U.S.

Another notable limitation of FIT is its lack of sensitivity for detecting serrated lesions, according to Johnson.

“We recognize that the serrated pathway has become an increasingly apparent pathway not only for primary colon cancer but also for interval colon cancers after people have had a colonoscopy,” he said.

A newer stool-based test approved in 2014, the FIT-fecal DNA test, can detect these high-risk lesions with better sensitivity, although with less specificity than FIT as well as a significantly higher cost, according to Seth A. Gross, MD, FACG, of NYU Langone Health.

Seth A. Gross

“The fecal DNA stool test is becoming more popular because it’s better at identifying advanced adenomas, whereas the primary aim of other noninvasive screening techniques is to diagnose colon cancer,” he said in an interview. “The stool DNA test is more sensitive to picking up advanced adenomas and sessile serrated polyps, which are important because in some situations they have a shorter pathway to moving onto colon cancer than a traditional precancerous tubular adenoma.”

FIT-fecal DNA

Cologuard, a tier 2 screening modality recommended every 3 years rather than every 1 year, is a good option, according to Thomas F. Imperiale, MD, of the Richard L. Roudebush VA Medical Center, the Regenstrief Institute, and the department of medicine at Indiana University School of Medicine.

PAGE BREAK

“For people who prefer a less frequent noninvasive test, stool DNA FIT is the best option, though at a greater cost,” he said in an interview.

The test’s list price is $649, according to the Cologuard website. As of May 2017, it was covered by CMS and the major payors in the U.S., including Aetna, United Healthcare, Cigna, Humana and Tricare. Analysts said the test should be covered for about 95% of eligible patients this year.

Thomas F. Imperiale

“There are issues with the cost-effectiveness of this test, but the sensitivity of stool DNA is better than FIT alone,” Imperiale said.

In one study, the FIT–fecal DNA test showed a 1-time sensitivity for CRC of 92%, which is “the highest single-time testing sensitivity for cancer of any noninvasive, nonimaging CRC screening test,” according to the MSTF guideline. The study also showed it performed with 40% sensitivity for sessile serrated polyps greater than 1 cm in size.

“Where it has some drawback is with specificity, which is not as good as FIT, particularly in people older than age 65, but it is a very good option for people who don’t want to do a noninvasive test every year, and who are younger (< 65 years) in the screening age range,” Imperiale said.

The test manufacturer, Exact Sciences, is currently performing a longitudinal follow-up study “to make sure the 3-year interval is appropriate,” he added.

Comparative Studies Needed

The large number of different FIT tests currently available in the U.S. — 16, according to a recent report — poses a problem, as their efficacy relative to each other is currently unclear. To address this, a team of investigators across the U.S. recently announced they will compare the accuracy of different FIT tests to identify which is most effective. The study is funded through the Cancer Moonshot initiative.

“Our number one goal is to identify which FIT test is the most accurate,” Navkiran Shokar, MD, MPH, MA, of Texas Tech University Health Sciences Center El Paso, said in a press release. “But we are also hoping to build awareness about these newer, more convenient FIT tests and get more people screened for colorectal cancer.”

Similarly, a new study published in the American Journal of Gastroenterology comparing three fecal occult blood tests in more than 1,000 patients showed that the InSure FIT (Clinical Genomics) performed with superior sensitivity vs. the OC FIT-CHEK and Hemoccult II SENSA (26.3% vs. 15.1% and 7.4%, respectively, for detecting advanced colorectal neoplasia). Another recent study published in Gastroenterology compared nine quantitative FITs and found wide variability when using sensitivity and specificity thresholds recommended by the manufacturers, leading investigators to conclude that this heterogeneity in diagnostic performance “can be overcome to a large extent by adjusting thresholds to yield defined levels of specificity or positivity rates.”

PAGE BREAK

Beyond FIT, published randomized trials directly comparing the outcomes of different CRC screening tests are also needed, and Johnson and colleagues on the MSTF noted that because of this lack of data, they took “practical considerations” into account when designating modalities into their given tiers.

Test availability appears to have factored into the tiering criteria in this guideline, according to Imperiale.

“Even though it seems like we’ve got a lot of good choices, what really is most readily available and best known in the U.S. at this time is colonoscopy and FIT,” he said. “Flexible sigmoidoscopy has, by far, the strongest and most consistent evidence supporting efficacy than any screening currently available, but no one’s getting screened with it in the U.S. It’s shown fairly good incidence and mortality reductions, but it’s essentially unavailable here, so the tiering criteria goes beyond published evidence about effectiveness. Unlike flexible sigmoidoscopy, there are no completed randomized controlled trials involving colonoscopy, which is why four trials involving colonoscopy are ongoing worldwide.”

Among these four large, long-term randomized comparative trials currently underway, Imperiale and colleagues recently completed enrollment in the CONFIRM study, which will compare the efficacy of screening colonoscopy vs. FIT for preventing CRC mortality among 50,000 average-risk VA patients over 10 years.

“We really don’t know which strategy is better, colonoscopy every 10 years or a FIT test every year,” Imperiale said in an interview. “We don’t know which is better in terms of uptake, adherence, and CRC morbidity and mortality reduction.”

Imperiale noted that CONFIRM is one of four ongoing global randomized trials comparing screening colonoscopy with FIT or other less invasive strategies, some of which have already reported interim results. One is the COLONPREV study in Spain comparing colonoscopy vs. FIT, with 10-year outcomes expected in 2021, and another is the Nord-ICC trial involving centers in Poland, the Netherlands, Norway, Sweden and Iceland, comparing colonoscopy with standard of care, with 15-year results expected in 2026.

Uptake

Patient compliance with the different tests is an important consideration of these large studies, for obvious reasons, according to Imperiale.

“People are very interested in comparing uptake, because a screening test can’t be effective if people don’t do it,” Imperiale said. “The most effective it can be is the extent to which it’s taken up, so if 30% of the colonoscopy group gets colonoscopy, the most it can reduce colon cancer incidence or mortality is by 30%. If 80% of the FIT group gets FIT done regularly, and it detects 75% of all CRC, FIT will outperform colonoscopy in the long-run. But we just don’t know the answers to these basic questions about uptake, adherence and effectiveness.”

PAGE BREAK

Interim results from the COLONPREV study have provided some data on this topic, showing a 39% higher participation rate with FIT and the same CRC detection rate as colonoscopy (0.1% in each), but other data are conflicting. In contrast, a randomized clinical trial published last year in JAMA showed higher screening completion rates with colonoscopy vs. FIT outreach over 3 years, although both interventions were superior to usual care (38.4% for colonoscopy, 28% for FIT, 10.7% for usual care).

A major challenge to improving uptake of CRC screening is understanding and adequately addressing all the reasons why people aren’t being screened, according to Douglas G. Adler, MD, professor of medicine in the division of gastroenterology and hepatology at the University of Utah School of Medicine.

“There are lots of reasons why people don’t get screening colonoscopy,” he said. “Their primary care provider may not think to offer it. They may fear it will be uncomfortable, painful or embarrassing. And bowel prep is inconvenient. ... These are all classic barriers.”

Douglas G. Adler

People who live in more remote areas also face the added challenge of access, Adler said.

“Many small towns have no gastroenterologist,” he said. “If people live in rural Wyoming or Montana or Nevada, if the nearest GI doctor is hours away, they’re just not going to prioritize getting a colonoscopy if they have to designate 2 days to drive 4 hours, get a colonoscopy and stay overnight at a hotel.”

Cost-effectiveness

Cost concerns are yet another barrier preventing people from undergoing screening colonoscopy, and issues of cost-effectiveness underlie the usefulness of all screening strategies, according to Imperiale.

“Screening rates in the U.S. have been increasing more in community health centers and safety net hospitals in cities, largely because of FIT, but adherence with colonoscopy has leveled off and has even gone down in some cases because of co-pays and deductibles,” he said.

Although the Affordable Care Act has improved access to screening colonoscopy by fully covering it, patients are often liable for a co-pay if polyps are found and removed, as the procedure is then coded as a diagnostic colonoscopy. Similarly, health plans may deem colonoscopy diagnostic after a positive noninvasive test or if the patient is symptomatic.

While concerns about out-of-pocket costs may prevent patients from being screened, cost-effectiveness studies have consistently shown that CRC screening by any modality is cost-effective compared with no screening, given the high costs of CRC treatment. Both colonoscopy and FIT perform well in these cost-effectiveness analyses compared with other screening tests, but the newer options including CT colonography, FIT–fecal DNA and capsule colonoscopy “could reach cost-effectiveness by substantially increasing compliance,” according to the MSTF guideline.

PAGE BREAK

The cost-effectiveness of screening approaches is important to consider, particularly when thinking about CRC screening on a global scale, according to Jonathan A. Leighton, MD, of the Mayo Clinic in Arizona.

“In many other countries their screening programs include the FIT test annually with or without flexible sigmoidoscopy every 5 to 10 years. It is probably the most cost-effective strategy overall, and the one that most countries can afford to do, because colonoscopy by itself does get expensive and is more resource-intensive. Cost is certainly one of the reasons annual FIT was recommended as a first-tier test, and it has a fairly high one-time sensitivity and a low first-time cost, all of which makes it attractive,” he said.

Jonathan A. Leighton

The cost-effectiveness of FIT makes it an especially viable way to expand the screening population, given recent calls to increase the percentage of individuals screened, Leighton said.

Lowering the Screening Age

Screening colonoscopy is also out of reach for many younger patients in whom CRC rates are notably on the rise, prompting some to call for lowering the screening age for average-risk individuals from 50 to 45 years. Similarly, racial disparities in CRC risk and screening has led the U.S. MSTF on CRC Screening to suggest lowering the screening age among black Americans to 45 in their recent guideline, but the best course of action to address it remains a matter of debate.

“I am definitely concerned about the evidence that CRC is occurring more frequently in patients under the age of 50, so I don’t think we can ignore that, and the risk appears to be higher in black people in particular,” Leighton said. “So, based on that we certainly need to consider lowering the screening age in certain sub-populations, but it requires further study.”

Others have made the same call. A recent study presented at UEG Week showed marked increases in the detection of neoplasia, polyps and adenomas in colonoscopy patients aged 45 to 49 years vs. those aged 40 to 44 years, which led investigators to recommend that CRC screening should begin at age 45 years.

According to Leighton, FIT shows the most promise for expanding the screening population in a cost-effective way.

“Lowering the screening age may turn out not to be as cost-effective, but from the standpoint of the individual patient I think it’s something that we have to strongly consider,” he said. “It at least needs to be further studied so that we can make an educated decision, but in the meantime, I think doing FIT plus or minus flexible sigmoidoscopy, which is what’s done in many parts of the world, would be a very reasonable approach to do it initially rather than necessarily going straight to colonoscopy.”

PAGE BREAK

Imperiale does not believe that lowering the screening age is the appropriate response to this trend. In a recent editorial, he commented on a study of patients in the Korean CRC screening program, which found that average-risk patients aged 35 to 49 years with a positive FIT showed a higher risk for advanced colorectal neoplasia confirmed by colonoscopy vs. patients in their 50s. The investigators concluded that the results may justify beginning FIT screening at age 35.

In his editorial, Imperiale wrote that “lowering the screening age now would be an emotional response to a problem that needs reason and a greater understanding of etiologic factors, risk factors, and the balance of benefits and harms under varying conditions,” and that immediate efforts should instead focus on patient education and prevention.

“When you look at the numbers in that article, the ratio of false-to-true positive was 9 to 1, so you’d have to perform colonoscopy on 10 people to find one with an advanced neoplasm, nearly all of which were not cancer,” he said. “Most of what was found were non-advanced adenomas that would put people into surveillance programs beginning at young ages; the fact that we don’t know the natural history of advanced or non-advanced adenomas is a huge drawback. We know the natural history is not that bad, but studies from Germany suggest that the younger you are, the lower the transition rates from non-advanced adenoma to advanced adenoma and from advanced adenoma to cancer. Lowering the screening age could very well tip the benefit-to-risk ratio towards risk. The downsides are potentially substantial and for these reasons, I don’t think this would work in the U.S.”

While modeling studies show that it really is best to start screening around age 50 on average, the number itself is somewhat arbitrary, and there are inherent inefficiencies in choosing a single age to start screening for everyone, Imperiale said.

“If we could tailor the age at which to begin screening we would do better as far as efficiency, except it becomes more complicated to start younger in one demographic group and start older in another,” he said. “For example, men and women probably don’t require colon cancer screening starting at the same age. Men’s risks for advanced adenoma and colon cancer are 1.5 to 2 times that of women irrespective of age until around 70, where women start to catch up, but we don’t screen men earlier or women later given the challenges to implementing it.

PAGE BREAK

“Even though the risk of colon cancer is rising in young people, it’s rising slowly and the risk is still an order of magnitude less than it is in people who are 50 and older. So, to say we should start screening earlier could cause a lot of mischief ... and it’s quite possible that the frequency of major complications would outweigh curable-stage cancer detection.”

Other Imaging Options

While colonoscopy and stool-based tests are the current mainstays of CRC screening in the U.S., other tier 2 and 3 imaging modalities play important roles in screening certain niches of individuals.

CT colonography, for example, is a tier 2 test that is recommended every 5 years, and has essentially replaced the use of double-contrast barium enema. It carries a lower risk for perforation than colonoscopy, and detects adenomas greater than 1 cm with sensitivity ranging from 82% to 92%.

This option is primarily used to screen patients who cannot undergo colonoscopy due to comorbidities, and to perform a same-day follow-up exam after an incomplete colonoscopy, according to Leighton.

“The nice thing about CT colonography is that it can visualize the entire colon but it doesn’t require sedation and you can safely perform it in anticoagulated patients,” he said. “Some people are concerned about the radiation, but it does use a low dose. A major advantage is you can often perform CT colonography to rule out any significant lesions on the same day as a colonoscopy after an incomplete exam.”

However, according to the MSTF, CT colonography has a limited impact, and is disadvantaged by the need for frequent follow-up colonoscopies and management of incidental extracolonic findings. “However, CT colonography appeals to a niche of patients who are willing to undergo bowel preparation and are concerned about the risks of colonoscopy,” the MSTF wrote.

Another emerging option that may be appropriate for a subgroup of patients who decline conventional colonoscopy is capsule colonoscopy, like Medtronic’s Pillcam, which the FDA approved in 2014 for patients with prior incomplete colonoscopies or those who cannot undergo colonoscopy or sedation. However, capsule colonoscopy is not approved for screening average-risk individuals, and lack of reimbursement currently impedes its widespread use, according to Gross.

“Until we have a CPT code for it, I don’t think that it will be widely available in the community, because the cost ends up going to the patient,” he said.

While the tier 3 option may be reasonable for patients who have incomplete colonoscopy, those who simply decline colonoscopy may not prefer capsule colonoscopy as an alternative, as it requires a more extensive bowel prep than conventional colonoscopy, Gross noted.

PAGE BREAK

“One of the things that turns people off from colonoscopy is the bowel prep, and when you do the colon capsule it’s a more rigorous bowel prep,” he said. “So, I think there’s value in it, but again it’s going to be appropriate for a small subset of patients that can’t undergo traditional optical colonoscopy.”

Despite its limited use, Gross noted that colon capsule technology has significantly improved over time. In a meta-analysis of 14 studies involving 2,420 patients, he and colleagues showed that the sensitivity for detecting polyps greater than 6 mm (58% vs. 86%) and 10 mm (54% vs. 87%) increased substantially between first- and second-generation colon capsules.

“It’s an exciting advancement, and they’ve done so much more with the second-generation technology, in the sense of capturing more frames per second, so you’re able to get more images of the colon lining,” Gross said.

However, due to several disadvantages, the future role of capsule colonoscopy for screening remains uncertain, according to Leighton.

“In addition to the more intense bowel prep, the technical failure rate, at least in the U.S. study, was around 9%, so there’s a relatively high number of technical failures,” he said. “There is also some concern that it may have trouble detecting sessile serrated lesions. It has been shown to be useful after an incomplete colonoscopy, but usually not on the same day. Unlike CT colonography, where you can perform the procedure on the same day as a colonoscopy, it’s tough to do that with the capsule colonoscopy. So, I think the future is still in question. I would say that it is a reasonable option for patients who have incomplete optical colonoscopy.”

Similarly, Adler said the logistics of colonoscopy scheduling are a notable barrier preventing less invasive imaging strategies from being fit into the CRC screening algorithm.

“The biggest problem is the study is only diagnostic,” he said. “If a patient has a CT colonography or a capsule colonoscopy and a polyp is found, the provider can’t just call the GI lab and send the patient over for a colonoscopy. They’d have to be scheduled, likely for the following week unless they’re incredibly lucky, so the patient is getting two preps and two studies where maybe only one would have been enough. So, the idea of using a capsule to screen the colon for CRC or polyps is a phenomenal idea that just has really not taken off, because the logistics of the way colonoscopy works, while not insurmountable, are very difficult.”

PAGE BREAK

Liquid Biopsy

Finally, while the MSTF recommended against the use of the Septin9 serum assay due to its inferior performance relative to FIT, experts agreed that an accurate blood test for CRC would be very valuable.

“Blood testing is kind of the holy grail,” Adler said. “A very high sensitivity, high specificity blood test with a good positive predictive value would be very widely adopted. However, it has been extremely difficult to develop such a test, and I suspect one would be used to detect cancer rather than polyps.”

While no such test is ready for primetime, experts agreed the effort is worthwhile. Recently at the ASCO Gastrointestinal Cancers Symposium, new data was presented showing a circulating tumor blood test (CellMax Life) detected precancerous lesions with 84% accuracy and cancerous lesions with 88% accuracy.

Investigators evaluated 620 individuals who were scheduled for routine colonoscopy or had a CRC diagnosis, 438 of whom had either adenomatous polyps or early- to late-stage colorectal cancers. The CellMax biomimetic platform (CMx) identified these lesions with 97.3% specificity. Sensitivity was 76.6% (95% CI, 67.9-83.5) for detection of circulating tumor cells in precancerous lesions and 86.9% (95% CI, 82.8-90.1) for stage I to stage IV cancers.

“The accuracy of the CellMax CTC blood test in terms of sensitivity and specificity is comparable or superior to stool based tests such as FOBT and FIT for detecting Stage I-IV cancers,” study investigators Wen-Sy Tsai, MD, PhD, assistant professor at Linkou Chang Gung Memorial Hospital in Taipei, Taiwan, and Ashish Nimgaonkar, MD, assistant professor of medicine at Johns Hopkins University, told Healio Gastroenterology and Liver Disease by email. “The key difference between this CTC blood test and stool based tests is its higher sensitivity in the detection of precancerous lesions. Stool based tests have roughly 40% or lower sensitivity in the detection of precancerous lesions.”

The test could cost less than $100, making it an affordable option, according to a company press release.

“Surveys conducted to find the cause of low screening rates demonstrated that most people would rather get a routine blood test than a routine colonoscopy or stool-based tests due to its convenience and non-invasive nature,” they said. “The advent of the CTC blood test is that it offers individuals a convenient and accurate test option so that they routinely get tested which is essential.”

The blood test is commercially available in Asia and the company is evaluating commercialization options in the U.S., Tsai and Nimgaonkar noted.

PAGE BREAK

Education, Stewardship

While providers await the arrival of a viable blood test, supporting awareness and use of the currently available noninvasive screening options is important to maximize screening rates, according to Imperiale.

“Gastroenterologists need to be good stewards of colon cancer screening, so that their primary care counterparts don’t play the methods down,” he said. “Some providers may not give their patients a choice, perhaps because they don’t believe that these tests are effective, so we have to be good stewards of the different options for colon cancer screening and support primary care providers and patients in their choices.”

Adler agreed, but emphasized that patient education is just as important as patient choice.

“The biggest thing we can do is educate people,” he said. “Colon cancer, like a lot of malignancies, is asymptomatic until it’s very advanced, and I think that if people don’t really have a sense of the risk and danger of colon cancer they are less likely to be screened.”

Adler cited a billboard campaign in Utah cited by his patients as a reason for pursuing a colonoscopy. His patients, he said, also reached out to their primary care provider for the referral when they had not previously received it, “a good example of the patient driving the doctor,” he added.

“Campaigns to raise awareness among the public tend to be extremely successful,” Adler. “The biggest thing we can do is awareness and education. The goal was 80% by 2018, and we’re certainly far below that. So, a very large swath of this country who is eligible has still not had any colorectal cancer screening test. We’re in a good time in the sense that we have so many tests now at our disposal, it’s very hard for a patient to say they don’t want to be screened at all.”

Finally, Adler addressed the apprehension among some gastroenterologists that noninvasive testing may represent competition to their colonoscopy practice and a loss of screening colonoscopies.

“Noninvasive screening is actually good for GI business,” he said. “If you start screening more of the population, you’ll start finding more people who need colonoscopy, and you’ll probably be doing less diagnostic colonoscopy and more therapeutic colonoscopy, so I think it’ll actually be good for the population and good for GIs.” – by Adam Leitenberger

Disclosures: Adler, Gross and Johnson report no relevant financial disclosures. Imperiale reports that he receives grant support through Indiana University from Exact Sciences. Leighton reports consulting and research relationships with Given Imaging, Covidien and Medtronic related to colon capsule technology.