Need for Virtual Colonoscopy colon screening

Need for VC colon screening trumps physician turf battles

SAN FRANCISCO – Virtual colonoscopy is safe and cost-effective and should be implemented for widespread colorectal cancer screening despite the inevitable turf struggles that arise between physician subspecialties, according to the radiologist who led the multicenter National CT Colonography Trial (ACRIN 6664).

Speaking on Saturday to a roomful of gastroenterologists, clinical oncologists, and surgeons at the 2009 Gastrointestinal Cancers Symposium, Dr. C. Daniel Johnson said that examining the growing ranks of unscreened patients 50 years and older is more important than waging turf battles between radiologists, the primary drivers of VC screening, and gastroenterologists, who perform the majority of optical colonoscopy exams, the principal screening method in the U.S.

Johnson is a professor of radiology at the Mayo Clinic in Scottsdale, AZ. The annual symposium is cosponsored by the American Society of Clinical Oncology (ASCO), the American Gastroenterological Association (AGA) Institute, the American Society for Radiation Oncology (ASTRO), and the Society of Surgical Oncology (SSO).

What the trial showed

The National CT Colonography Trial “was performed because there was a lot of controversy over how good this test really was” following the publication of widely divergent results in earlier multicenter studies, he said. Those studies, by Pickhardt et al, published in 2003, and by Cotton et al andRockey and colleagues in 2004, reported sensitivities for clinically significant polyps of 92%, 55%, and 59%, respectively.

For the National CT Colonography Trial, published last September (New England Journal of Medicine, September 18, 2008, Vol. 359:12, pp. 1208-1217), Johnson and colleagues screened 2,600 patients using state-of-the-art screening techniques, MDCT, and mandatory training of readers and reported 84% per-polyp (90% per-patient) sensitivity for the detection of adenomatous polyps 1 cm or larger, similar to that reported by optical colonoscopy for both large- and medium-sized adenomas.

“If we included the false positives that were detected at CT colonography, we would send about 12% of our patients to colonoscopy if you used a 6-mm threshold, and most patients would have been spared the cost, risk, and inconvenience of colonoscopy,” Johnson said.

As virtual colonoscopy, or CT colonography (CTC), moves from research into clinical use, high quality will need to be assured by using comprehensive reader training, stool and fluid tagging, and mechanical insufflation to ensure a well-distended colon, which he called critical for the maintenance of high-quality results.

In ACRIN 6664 “we had a highly compliant patient population that followed the instructions for the prep. I don’t want to underestimate the importance of that, as well as modern imaging methods,” Johnson said.

Controversy surrounding VC screening

As for the radiation dose, the average exam delivered about 5 mSv, he said, similar to that of a barium enema exam and about half the dose delivered by other abdominal CT exams. That 5 mSv compares to average annual background radiation of 3 mSv in the U.S. and radiation as high as 12 mSv in Denver and Santa Fe, NM.

“Airline personnel get even higher doses than those living in Denver, and there have been no reports of a higher incidence of cancers in those populations,” Johnson said. In addition, seven studies have examined the effects of long-term radiation exposure on the job in more than 100,000 workers, with no adverse effects reported, he said.

As a result of this evidence, the Health Physics Society has said that the so-called linear no threshold model is an oversimplification of the risks of ionizing radiation exposure, particularly at small, repetitive doses, and that the adverse effects of radiation “have not been consistently demonstrated below 100 mSv,” Johnson said. At that level, the health effects “are either too small to be observed or nonexistent,” he said.

Extracolonic findings

Johnson discussed another controversial topic in virtual colonoscopy — extracolonic findings — with some researchers questioning VC’s ability to determine significant findings accurately and deal with them safely and cost-effectively.

Multiple studies at the Mayo Clinic, and by Gleucker et al and Pickhardt et al, show that “overall you pick up about 6% to 10% of lesions that really merit additional investigation,” Johnson said. The studies show that extra imaging costs recommended as a result of the findings add between $31 and $67 per examination for evaluating findings that occur mainly in the kidneys, ovaries, chest, and abdominal aortic aneurysms.

“Extracolonic findings are potentially a problem for us — I want to be frank about that,” Johnson said. “Most of the findings are inconsequential to clinical care; they have the potential to increase exam costs, patient inconvenience, and concern,” he said. “I think the radiology community needs to take the next step and develop a framework for categorizing, reporting, and better managing these findings.”

VC’s safety

In daily clinical performance, VC has proved an extremely safe exam, Johnson said. A survey from the Working Group on Virtual Colonoscopy found that in nearly 22,000 VC cases reviewed for safety, there were no perforations in the screening population; however, there were two cases among patients who had diagnostic exams. There were also two cases of renal failure and no myocardial infarctions or deaths, for an overall complication rate of about 1 in 5,000 exams, much lower than that of conventional colonoscopy, he said.

As for costs, Mayo Clinic statisticians found the cost of CTC to be about 50% higher than that of flexible sigmoidoscopy and slightly more than half the cost of conventional colonoscopy, Johnson said.

Results have not yet been published for what may be the most comprehensive evaluation to date, a cost survey by the U.S. National Institutes of Health’s (NIH) Cancer Intervention and Surveillance Modeling Network (CISNET) that is examining the costs and benefits of all colorectal screening exams. Preliminary results were presented last November at a meeting sponsored by the NIH.

The CISNET cost models, which incorporate the adenoma-carcinoma sequence, as well as the potential multiplicity of polyps, all found that that CTC screening every five years is a cost-effective strategy compared to no screening, and that the costs of incremental screening are modest, Johnson said. All of the models showed life-year gains at a modest cost using CTC as a primary screening strategy.

“But because there are a number of colorectal cancer tests that are available to patients, all of the CTC strategies were dominated by other approaches, although the cost increment between those strategies was small,” he said.

Taken as a whole, the data make it clear that there is an effective new test that can be made available to the public in an effort to get more individuals screened, he said. “The real question is: Will this new test translate into increased screening by the public? And frankly, the data are mixed.”

What patients think

“With colonoscopy, there’s obviously good and bad,” he said. “With colonoscopy, patients need to have drivers, you have to have an IV, you’re sedated, you’re out of work for the day. The good news is that you don’t really remember much about it, so you’re willing to come back in 10 years. On the other hand, with CTC there’s no driver required, there’s no sedation, and you return to work right away … but the bad news is that you can remember that you had some transient discomfort when your colon was distended for those 10 minutes.”

Also, as many as 12% of screening patients will need to have colonoscopy anyway, and unless same-day colonoscopy is performed, patients will need to have another bowel prep, he said.

Interestingly, among several studies that have evaluated both exams for patient preferences, those led by radiologists have all found that patients preferred CTC, while studies led by gastroenterologists all concluded that colonoscopy was the preferred method.

Maybe there was some bias in the way the questions were asked, he said. Also, in studies where CTC was preferred, the subjects received the antispasmolytic drug glucagon prior to the exam, which relaxes the colon. Studies have also shown that bowel prep, usually required in both CTC and optical colonoscopy, is the most uncomfortable part of the exam, Johnson said.

But VC offers the potential of patient-friendly cleansing methods, which forego laxatives in favor of tagging, and sometimes utilize electronic cleansing to remove the tagged stool with comparable sensitivity to CTC exams performed with cathartic cleansing, he said.

“We’re hoping that this is kind of the future of CT colonography, where we can access those patients who are unwilling to have a colon screening test because of the disincentive of the cathartic prep.”

CTC meets all of the criteria for a screening test, including performance, cost-effectiveness, acceptance, and safety, he said.

“In my opinion, CTC is ready for prime time,” he said. “I hope this doesn’t turn into a tug-of-war between specialties, in which there’s a self-interest to promote their own screening technique. Because I don’t think we need a tug-of-war, I think that energy is poorly used. What we should do is consider this an opportunity to work as a team to screen more of the eligible 35 million Americans who are not screened with current guidelines.”

By Eric Barnes staff writer
January 20, 2009