CMS Decision on CTC Prompts Restrategizing
The CMS decision to retain nonapproval status for CT colonography (CTC) as a screening method for Medicare patients was a bitter pill for radiology to swallow—all the more bitter because, prior to the February 11 decision, the mood had been optimistic. After years of battling on behalf of CTC, researchers and CTC advocates seemed finally to have turned a corner that would result in CMS approval.
Bibb Allen, Jr, MD
In a 2007 study, Kim et al¹ used a mathematical model to evaluate the cost effectiveness of CTC compared with optical colonoscopy (OC) and flexible sigmoidoscopy. In this study, with no threshold for polyp size, CTC was found least expensive of the screening methods, with a cost of $7,138 per life-year gained, compared with $7,407 for sigmoidoscopy and $9,180 for OC. With a polyp threshold of 6 mm, CTC was even more cost effective, at $4,361 per life-year gained. Then, in September 2008, the ACR released results of its long anticipated ACR Imaging Network (ACRIN) trial² comparing CTC with OC. The results showed CTC to have sensitivity rates of 90% and specificity rates of 86% for precancerous polyps 10 mm or larger. These rates were comparable with OC, the report says. Even before the ACRIN results were officially released, however, the American Cancer Society (ACS) gave CTC a huge boost by approving it as a screening option for colorectal cancer. To some, the ACS validation, in March 2008, seemed to signal that CMS would approve CTC for screening Medicare patients. Bibb Allen, Jr, MD, chair of the ACR's Commission on Economics, told last summer, "We don't know exactly how [CMS] will respond. We are optimistic for a favorable decision." A favorable CMS ruling, however, was not to be. In the announcement last month, CMS officials wrote, "The evidence is inadequate to conclude that CT colonography is an appropriate colorectal cancer screening test. CT colonography for colorectal cancer screening remains uncovered." The agency then called for additional comment over the next 30 days, leaving the door open for a reversal of its decision—but a reversal seemed unlikely, because the agency added that if a reversal occurred, CMS would then have to determine the cost effectiveness of CTC. In the week before the comment period ended on March 13, colorectal screening experts traveled to Capitol Hill to recommend the agency change its decision, stressing that seniors should have access to the minimally invasive screening test. To buttress its decision, CMS cited a series of votes by its Medical Evidence Development & Coverage Advisory Committee taken at a meeting on November 19, 2008. The committee members voted on seven key questions on CTC, including effectiveness, cost, and the likelihood of wider screening following CTC approval. Overall, the panel gave lukewarm scores to CTC. Some responses, the CMS report noted, "amounted to a vote of no confidence" in the procedure. In a series of review findings, CMS argued that CTC may be less effective than OC for polyps smaller than 10 mm. It also noted the issue of radiation exposure and questioned the expertise of radiologists in community settings who might be overseeing and interpreting the scans. The impact of extracolonic findings spotted by radiologists while performing CTC was also raised as an issue. Such findings occur often during CTC screening. One study, performed at Wake Forest University in 2006³, found that extracolonic findings such as lung nodules and kidney lesions (which are subject to independent workups) added an average of $231 to the cost of a CTC scan. If CMS approved CTC for screening, it would presumably pay for testing the extracolonic abnormalities for disease as well. Judy Yee, MD, is chief of radiology at the San Francisco VA Medical Center and is also professor and vice chair of the department of radiology and biomedical imaging at the University of California–San Francisco (UCSF). She has been a pioneer of CTC through studies at the VA and elsewhere. She has also developed national guidelines for performing CTC screening, and she teaches CTC at the UCSF medical school. "Currently, less than 40% of those who should be screened are actually getting screened," Yee says. "This is with the options of colonoscopy, barium enema, or flexible sigmoidoscopy." Most people continue to rely on a routine physician’s-office test, the fecal occult blood test (FOBT), but the FOBT will only detect bleeding cancers; it won't detect the all-important precancerous polyps that CTC finds, Yee says, because they don't usually bleed. In addition, she says, well-developed cancers often bleed only intermittently, and so are missed by the FOBT. Underscreening of the population over age 50 for colon cancer is something that CTC researchers like Yee and Pickhardt have stressed. OC is uncomfortable and time consuming, and most asymptomatic people avoid it. Because CTC requires no physical probe, the idea is that more people at risk of colon cancer will opt for it. More people will be screened, and more colon-cancer deaths will be prevented. Yee was disappointed by the CMS decision, but says that she wasn't surprised. "Obviously, I think this is not completely unexpected, given the economic climate," she says. "In the economic times we are in, everyone is looking at decreasing costs. I can understand why certain government bodies would not be anxious for another avenue of expense." She says that CMS held CTC "to a more rigorous approval level than some of the other screening tools," such as the double-contrast barium enema. "The double-contrast barium enema, which has an equal radiation dose and is not as sensitive, is covered," Yee notes, "yet here we are, with a better test and better supporting data than the double-contrast barium enema, but we're still refused." Time to Restrategize Yee says that CMS noncoverage of CTC already has radiologists contemplating new strategies for gaining approval in the future. Since CMS officials suggested that CTC studies, to date, had not addressed the Medicare-age population, Yee says that the ACRIN data might be reanalyzed to single out a cohort of Medicare-age patients in order to demonstrate the efficacy of CTC in that age group. Another strategy will probably involve starting a registry of CTC patients to document the technique's effectiveness in quality-controlled settings, Yee says. She also notes that some private payors, such as Blue Cross Blue Shield and Kaiser, have approved CTC as one screening option. Their patients will add to the CTC data pool. A recent Canadian study4 suggested that OC is good at preventing cancer deaths from left-side colon lesions, but that there was no such documentation for right-side or ascending colon lesions. The implication was that OC wasn't good at finding those right-side lesions. Yee suggests, however, that there are many factors in the Canadian study, such as lesion size and type, that are yet to be analyzed. The data are not in yet, she cautions. While most cancerous lesions (55% to 60%) occur in the left side of the colon, Yee says, about 25% occur on the right side. CTC is good at finding these right-side lesions, and that has been brought to the attention of CMS, Yee says. "It will be interesting to see what they do with this under review," she adds. Advanced visualization and computer-aided detection also have roles to play in the movement for broader acceptance of CTC. A recent study by a team of Dutch researchers at the University of Amsterdam5 found that novice CTC readers using 3D electronically cleansed endoluminal data sets detected more polyps than they did from 2D images. Expert readers, on the other hand, performed equally well in 3D and 2D. Yee says that the ACRIN trial showed that primary 3D and 2D renderings worked equally well for locating lesions. Both were comparable to OC, she says. She is also excited about the benefits of computer-aided detection. "Some workstations already have colon computer-aided detection," she says, "while others will be bringing it out this year. It increases sensitivity and specificity for lesion detection, particularly in novice readers." Computer-aided detection will definitely play a role in the CTC studies of the future. As Yee says, "This is another issue for CMS. Computer-aided detection is a big factor."