Four non-endoscopic diagnostics for colonic neoplasia may disrupt colonoscopy

SAN FRANCISCO — Non-endoscopic diagnostics for colonic neoplasia may prove to be disruptive to the practice of screening colonoscopy, according to a presenter at the 6th annual American Gastroenterological Association Tech Summit.

“Compared to other current forms of screening, [colonoscopy] probably has a greater potential for cancer prevention because of its ability to detect cancer precursor lesions and remove them, and we have evidence that supports infrequent use if a high quality exam is negative,” David A. Lieberman, MD, AGAF, from the Oregon Health and Science University, said in a presentation. However, colonoscopy “is invasive, and it would never be conceived of as the ideal test for population-based screening,” he said, due to its being expensive, resource-intensive and associated with risk for serious adverse events, especially in individuals older than 65 years.

Furthermore, although colonoscopy is the most commonly used form of screening in the United States, screening rates are lower compared with the European Union. “If we look at what has been accomplished with cervical cancer screening [and] mammography, we will see that there is a ways to go with colonoscopy.” This “screening gap represents an opportunity for innovative ways of doing this,” Lieberman said.

An ideal alternative would be a noninvasive, low-risk, low-cost and no prep test requiring readily available resources, he said. “I would suggest that there has some disruption in this universe and, potentially, there is going to be some reduction in the use of colonoscopy in the future.”

Fecal immunochemical test

According to Lieberman, the data on fecal immunochemical testing (FIT) “looks pretty good.” A recent meta-analysis “suggests cancers can be detected with 79% sensitivity,” and in a prospective study, 74% of cancers were detected with FIT. However, the rate of detection of advanced adenomas was “less than 50% in almost all of these studies, so in terms of early cancer detection, [FIT is] pretty good, but in terms of cancer prevention, maybe not so good.”

Behavioral factors play a major role in the effectiveness of FIT, Lieberman said. “The issues surround the testing program because this is a program that requires programmatic adherence to both positive tests, [where] patients need a colonoscopy because they have an increased risk of cancer, and for negative tests, since it doesn’t eliminate the possibility of colon cancer, so they need to come back for repeat testing.”

In another trial that compared FIT with colonoscopy, colonoscopy was superior in detecting advanced neoplasia in patients who were adherent, which is “no surprise,” Lieberman said, “but this study confirms this in a way that we had not clearly known before.”

With FIT, “there is some potential for colon cancer prevention, but it’s going to be less than some other [methods],” he said.

Stool DNA

The principle behind stool DNA, Lieberman said, is in detecting “the needle in the haystack” of mutated cells sloughed into stool. In a recent “landmark study” comparing stool DNA with FIT (after which all the patients underwent colonoscopy), colon cancer detection rate was 92% vs. 74%, respectively, “and when you look at earlier stages of cancer it is even slightly better. But an important caveat here is the detection rate for advanced adenomas, [which] is still under 50% for both of these tests.”

Stool DNA appears to be an effective screening test, Lieberman said, and although it has been approved by the FDA and CMS, “the appropriate interval for testing is still uncertain. The recommendation is a 3-year interval,” based in part on its superior sensitivity compared with FIT, “meaning we shouldn’t need to do it as often.” Cost-effectiveness is an additional issue that will be resolved by decision modeling in the future, he added.

PAGE BREAK

Imaging

A number of studies published in the early 2000s showed that CT colonography is “not bad at detecting polyps that are greater than 10 mm,” Lieberman said, but “the cost is very sensitive to the threshold for referral to colonoscopy.” CT colonography also requires “a very good bowel prep,” and if patients are referred for a subsequent colonoscopy that they cannot have it the same day, the second bowel prep required may be undesirable to the patient.

Furthermore, the detection rate for flat serrated lesions is unclear, he said. Although serrated lesions also are difficult to detect with colonoscopy, “it would be very challenging to detect such a lesion with CT colonography or any imaging method unless you had some method for labeling, and that may be something that could happen in the future, but my guess is that this will be difficult.” It is thought that “this pathway accounts for more than 20% of colon cancers, and it certainly accounts for a number of cancers we see after patients have had colonoscopy,” he said.

Another issue associated with imaging is that of ionizing radiation, as patients undergoing repeat imaging tests are exposed to more radiation, he said. However, as it is not clear whether this is a significant risk to patients, this is “an existential risk.”

Magnetic resonance (MR) colonography also has been studied, but most extensively in Europe, Lieberman said. “There has been little data that has been generated over the last decade about MR here in the U.S. It could work, but our experience with it is much more limited.”

Serum testing

According to Lieberman, serum testing could be described as “the holy grail because it would be ideal to have a noninvasive test. You just come in, get a blood draw and then you get risk stratified.”

There are a number of possible pathways for serum testing, he said, including specific genetic markers, proteomics or genetic fingerprinting (identifying a pattern of genes associated with colon cancer), and metabolomics (identifying metabolites altered by neoplasia).

The most studied areas in this field include molecular tests, which “are all somewhat promising in terms of cancer sensitivity, but not so great for adenoma sensitivity, so it’s fair to say that this is still a work in progress,” he said.

One cohort study demonstrated relatively low sensitivity for detecting stage 1 to stage 3 cancer with a test based on a specific methylation marker that may be altered in colon cancer, Lieberman said. “It gets a lot better [at] stage 4 cancer, but we don’t want to diagnose at stage 4.” Another study compared stool DNA and the same serum methylation marker, which showed stool DNA was more accurate than serum tests in detecting early stage cancers and large adenomas, he said. The lead author of this study, David A. Ahlquist, MD, from the Mayo Clinic, proposed that “a plasma marker is probably going to require some degree of vascular invasion, whereas a stool marker is really just requiring exfoliation of cells, and that is why he thinks the stool DNA may be a more accurate vehicle to detect these lesions,” according to Lieberman.

Colonoscopy is not likely to “go away,” Lieberman concluded, “but as we think about where we want to go in the future … ultimately, we would like to have a noninvasive test that can accurately detect early stage cancer but also identify high-risk individuals before they get invasive cancer.” – by Adam Leitenberger

Reference:

Lieberman DA. Presented at: American Gastroenterological Society Tech Summit; March 19-20, 2015; San Francisco.

Disclosure: Lieberman reports he is on the scientific advisory board for Exact Sciences.