Politics matters in the real world of virtual colonoscopy

Nov 4, 2003

The colonoscopy may be virtual, but the politics shaping its approval for screening can be excruciatingly real. At the 2003 International Symposium on Virtual Colonoscopy in Boston, radiologist Dr. Joseph Ferrucci from Boston University School of Medicine offered evidence, and some politics of his own, in a discussion of VC's future.

"I'm going to talk about politics, which somebody defined as the science of how who gets what and why," he said. "If you ask how we're going to proceed to accepting virtual colonoscopy, I think we arguably are going to need data, plus some attention to the real world of politics." The ultimate incorporation of VC into the screening arsenal is inevitable, he said; what's unknown is how, why, and when it will happen.

"For radiologists there's a two-part approach to getting virtual colonoscopy approved," Ferrucci said. "One (is) to offer a critique of the gold standard of colonoscopy: honestly, fairly, wide-eyed. And (second is) to see whether or not we shouldn't simply seek parity with the other approved tests for colorectal cancer screening."

There is no question that VC is being introduced into a competitive environment, he said, but the dismissals of virtual colonoscopy by its detractors, particularly in the gastroenterology community, have been particularly strident. An American College of Gastroenterology brochure produced a couple of years ago is a case in point, he said. "Facing Reality: The Facts about 'Virtual' Colonoscopy" (click here for PDF file) is a "fairly trenchant" example of patient information used to discredit the virtual exam, Ferrucci said.

It's not as though conventional colonoscopy could be used to screen everyone who needs it, said Ferrucci, citing the comments of Dr. Seth Glick from the University of Pennsylvania Medical Center in Philadelphia. In the preceding talk, Glick had noted that while conventional colonoscopy is the gold standard for judging VC's accuracy in every clinical trial, it remains the most expensive and potentially hazardous colorectal cancer screening option available. And there aren't nearly enough endoscopists to screen everyone who will need it (73 million people in the U.S. by 2010), according to estimates.

Questions about conventional colonoscopy

More to the point, conventional colonoscopy has yet to demonstrate a clear benefit in a low-prevalence screening population, Ferrucci said, citing a New England Journal of Medicine study on the use of colonoscopy to screen asymptomatic adults for colorectal cancer.

The study, by Dr. David Lieberman and colleagues at the Portland VA Medical Center, examined 3,121 asymptomatic patients (96.8% men, mean age 69.2) using conventional colonoscopy. The results showed neoplastic lesions in 37.5% of the patients, an adenoma 10 mm or larger or a villous adenoma in 7.9%, an adenoma with high-grade dysplasia in 1.6%, and invasive cancer in 1% (NEJM, July 20, 2000, Vol. 343:3, pp. 162-168).

But colonoscopy found few large cancers. There was no evidence of neoplasia at all in 62.5% of patients, and in the remaining 38.5% of patients who did have lesions, 84.5% of the findings were smaller than 10 mm.

"Less than 10% of the lesions detected or removed were either premalignant or an actual malignant lesion," Ferrucci said. "I could say we detected and removed these ultra-small adenomas to turn the clock back in these patients, decreasing the future risk of colorectal cancer, but in 85% of patients there was no definite benefit," he said. "Reassurance, a clean bill of health, but no specific added value. I make this point because there's so much written about cost effectiveness, quality of life, number of years, so if you look at colonoscopy and what's in the literature, it's kind of interesting."

As for cost, Lieberman and colleagues removed small, innocuous polyps in 50% of the patients, which would have entailed a separate charge for polypectomy and increased the risk of complications, Ferrucci said. A pathology report for each lesion removed raises the cost even higher. And polypectomy and pathology aren't included in mathematical cost models, which has the effect of understating the true cost of colonoscopy, Ferrucci said.

"There's probably a net surcharge...of several hundred dollars per case," he said. "So we have to bear in mind what actually happens in an average-risk screening population." In such populations, the lower cost and risk profile of virtual colonoscopy make it cheaper and potentially more efficacious.

Nor do the VA results appear to be a fluke. A new colonoscopy screening study, also led by Lieberman, found an even lower prevalence of clinically significant lesions in a much larger population. Among 9,109 average-risk men who underwent screening colonoscopy, polyps larger than 9 mm were seen at colonoscopy in just 5.1% of patients overall, the authors wrote (American Journal of Gastroenterology, Oct. 2003, Vol. 98:10, pp. 2312-2316).

Comparison with other screening tests

Ferrucci showed a simple chart comparing the relative pluses and minuses of the various colorectal cancer screening methods in use today, including the fecal occult blood test (FOBT), flexible sigmoidoscopy, double-contrast barium enema (DCBE), conventional or optical colonoscopy (OC), and virtual colonoscopy (VC).

	FOBT	Flex Sig	DCBE	OC	VC
Accuracy	-	+/-	+/-	+	+
Cost	+	+	+	-	-
Risk	+	+	+	-	+
Acceptance	+/-	+/-	+/-	+/-	+/-
Availability	+	+	+	+	+/-
Other	-	+	-	+	+/-

Among the more salient statistics, FOBT is by far the lowest-cost screening method, but detects only about a quarter of advanced cancers. Flexible sigmoidoscopy finds about 70% of cancers (76% combined with FOBT), but misses lesions beyond the reach of the scope. In the National Polyp Study, DCBE missed 50% of all polyps 1 cm and larger. Major studies have shown virtual colonoscopy to be about 75%-90% sensitive for the detection of colorectal polyps 1 cm and larger. (Some of these statistics are more fully elucidated in a PowerPoint presentation by Lieberman to an FDA advisory committee in March 2002.)

Thus, VC stands up quite well in comparison with the other tests, "and, importantly, according to acceptance and compliance," Ferrucci said. "Many people have said that the best test for colon screening is one that you're willing to have. And when the new (exam's) features are as good as the old ones, why set the bar needlessly high? 'A new test need not outperform in all or even any category,'" Ferrucci read from a study discussion. "A relative advantage in one, plus acceptable performance in others, should suffice."

Evidence does matter, Ferrucci said, and radiologists are becoming more fluent in the language of multiple randomized clinical trials and meta-analyses. But along with excellent results, they'll need the political support of others in order to make the virtual exam a reimbursable reality.

The real decision-makers

"Who is going to be deciding about this screening test? It's not going to be radiologists, it's going to be our colleagues, especially in gastroenterology, other medical specialists, public health experts, insurance carriers, and colon cancer patient advocacy groups," Ferrucci said. "We in radiology know that the reason PET scanning was approved was not because of oncologists, not because of radiologists, but because of women with breast cancer. They're the ones who got to Congress and got PET scanning on the list."

In Ferrucci's view, the key groups to decide VC's screening future in the U.S. will be the American Cancer Society (ACS), the Agency for Healthcare Research and Policy (AHCPR), the American Gastroenterological Association (AGA), and, of course, the Centers for Medicare and Medicaid Services (CMS).

"As radiologists we're going to have to encourage virtual colonoscopy and invite others to participate in its evaluation," he concluded. "I think the requirements for acceptance are ... suitable data, review of the data, and national guidelines for colorectal cancer screening. We in the radiology community must get on with the business of creating practice standards and accreditation programs. And I'm not too worried about money. The reimbursement will follow."

By Eric Barnes
AuntMinnie.com staff writer
November 5, 2003
Related Reading

Group credits 3-D reading for best-ever VC results, October 15, 2003

Toward a CRC screening strategy: gastroenterologists take the stage, June 3, 2002