| |||||||||||||||
|
Home
|
For other articles and previous issues click here. June 27, 2005 Full Sail or Into the Wind?By Dan Harvey Radiology Today Vol. 6 No. 13 P. 10 The ACRIN trial hopes to better define the effectiveness and role of CT colonography as a screening tool. At first glance, computed tomography colonography (CTC), also called virtual colonoscopy, offers a desirable alternative to conventional colonoscopy. From a patient’s perspective, CTC promises a less-invasive and more comfortable option. To some clinicians, using a CT scanner to produce 3-D images of the colon seems to be equally as effective as conventional colonoscopy. Still, conventional colonoscopy remains the gold standard for colorectal screening, and for good reason. If a problematic polyp appears, the physician is well-positioned to deal with it. However, in light of dismal colon cancer screening statistics, clinicians view CTC as a possible way to make screening more acceptable to members of the at-risk population. “Mainstreaming” the technology, it is hoped, could bring more people in for screening and, in turn, help decrease the incidence of colorectal cancer. Many still consider CTC an emerging technology that poses significant challenges and raises substantive questions that must be addressed before widespread adoption of the technology. Recent studies have sought to demonstrate that CTC can be an accurate method for detecting colorectal polyps, but variable results raised questions about the procedure’s viability. The answers involve more than just determining whether CTC is effective. Enter ACRIN The multicenter trial, funded by the National Cancer Institute and led by principal investigator C. Daniel Johnson, MD, of the Mayo Clinic in Rochester, Minn., seeks to compare the effectiveness of state-of-the-art CTC with conventional colonoscopy for detecting growths in the colon. Controls built into this study will differentiate it from previous comparison studies, which contained significant variation in both methods and results. Findings were either heartening or dismaying as far as CTC’s viability is concerned, but still left unanswered questions due to the design of the studies. So it became clear that a large, multicenter study was necessary—one that would involve the latest techniques and technology. Many observers believe CTC needs to be put through such a large-scale trial before it can be considered for widespread clinical application. Quality control standards will be high. The ACRIN study involves the most current equipment and procedures. All participating clinicians will be trained and tested to ensure that they possess highly developed skills in interpreting the CT results. New Technology Researchers across all sites hope to address the unresolved issues associated with CTC. Some of the issues, says Yee, the principal investigator at the San Francisco site, include the use of tagging of stools, software usage for data evaluation, and how to interpret virtual colonoscopy—“whether to use the primary 2-D vs. primary 3-D interpretation method,” she says. Comparing how studies are read is a major issue in this trial, according to Peter Zimmerman, MD, associate clinical professor of radiological sciences and lead investigator for the ACRIN study at the David Geffen School of Medicine at the University of California, Los Angeles. “A large component of the test involves exactly how the study is interpreted on the computer workstation with different software,” he says. “This involves a comparison between what is called the primary 2-D reading method and primary 3-D method, where you use 3-D reformatting to make the images appear as if you’re traveling through the center of the colon itself, which is what is called the endoluminal view. Researchers will read studies in both ways, comparing the two methods.” Main Goals More specific goals include determining whether CTC provides clinicians with sufficient information to identify problems in the colon, validating the widespread clinical application of CTC, and evaluating the capability of CTC to detect lesions of 1 centimeter or larger in the colon. Polyp size is of primary importance, as polyps measuring less than 1 centimeter have less than a 1% chance of becoming cancerous. Polyps larger than 1 centimeter have a 10% chance of becoming cancerous over a 10-year period and 20% over 20 years. Many polyps can be found within the colons of the over-50 screening population. If problematic polyps can be detected early, the risk of cancer—and death from cancer—is greatly reduced. So, the effectiveness and attractiveness of CTC loom large. According to the study protocol, investigators are seeking to address “aspects of central importance to the clinical application of CTC in several interrelated but independent parts that will be conducted in parallel.” As such, the trial will be conducted in four parts: • Part I — The clinical performance of the CTC examination will be prospectively compared in a blinded fashion with colonoscopy. • Part II — Optimization of the CT technique will be performed in view of new technological advances in CT technology. • Part III — Lesion detection will be optimized by studying the morphologic features of critical lesion types and in the development of a database for computer-assisted diagnosis. • Part IV — Patient preferences and cost-effectiveness implications of observed performance outcomes will be evaluated using a predictive model. Currently, investigators are recruiting participants. The goal is to recruit more than 2,300 volunteers at 15 sites nationwide during a one-year accrual period. Participants will include adults aged 50 or older who are scheduled for a screening colonoscopy and have not received one in the past five years. All will undergo CTC virtual and conventional colonoscopy. The perceived reticence among the screening population is a major concern of ACRIN investigators. Colorectal cancer is the third-most-common cancer and second-leading cause of cancer death in the United States. Most colon cancers develop from polyps. Screenings to detect and remove polyps early are considered the best way to prevent colon cancer. The American Cancer Society recommends a colonoscopy every 10 years for adults aged 50 and older. Though the benefits have been clearly articulated, most Americans do not undergo colonoscopy. Many see colonoscopy as one of life’s more unpleasant experiences. “Less than 30% of the adult population that should be screened actually gets screened,” Yee says. “Even given the current screening recommendations, the majority of the population doesn’t come in. Clearly, we need a more acceptable screening tool.” Many physicians are sizing up CTC as such a tool. It is more comfortable and less invasive. Risk of perforation or tearing of the colon is eliminated. In addition, CTC requires no sedation or recovery time, unlike conventional colonoscopy. Previous Studies However, in another major study, conducted at the Duke University Medical Center in Durham, N.C.—which compared CTC, conventional colonoscopy, and air-contrast barium enema—CTC only demonstrated a 59% sensitivity for polyps 10 millimeters or larger. The Duke study used 2-D interpretation with 3-D problem solving, whereas the Pickhardt study used a converse approach. Subjects in the Duke study had positive fecal occult blood, while Pickhardt’s subjects were asymptomatic. In addition, the Duke study was developed earlier than the Pickhardt study and it didn’t have the benefit of later, improved technology. “The previous studies were done using different parameters,” says Yee. “The researchers used older models of CT scanners, and some studies didn’t use tagging. Also, there was some controversy as to whether some of the radiologists used primary 2-D vs. primary 3-D interpretation. Training of the radiologist was an issue as well.” In its “AGA Future Trends Report: CT Colonography,” published in the September 2004 issue of Gastroenterology, the American Gastroenterological Association (AGA) summarized the following variations in earlier major studies: • techniques used to prepare/cleanse patients and perform the studies; • differences in CTC technology (both hardware and software); and • variability in the manner in which CTCs have been read. Moreover, the AGA attributed the variable results of the studies to these variations. “Many different components go into such tests and are very important,” says Zimmerman. “That makes it very complicated. CT imaging is one of those components, although it is pretty much being done the same way in most places. The method of tagging the intestinal content is another way that it can be very variable. Software interpretations are another factor. There are several different commercial systems out there, and some people believe this has a major impact on accuracy. That is a major factor very difficult to control because different places use different software packages. The one that Dr. Pickhardt used, called Viatronix, is the one that some people believe is the one that everyone should be using.” One issue that led to uncertainty about software is that software packages have undergone rapid evolution. “Only a few years ago, none of them were really very adequate,” Zimmerman says. “Now, they’ve become much more adequate, but they still change so quickly that you’re never quite sure which one is the best to use. So, comparing different computer software programs and viewing methods can be very complicated things in themselves. The ACRIN study will be addressing that to a certain degree.” Despite variable results of past studies, Yee believes those efforts aren’t without merit. “I think all studies add to our body of knowledge,” she says. “With some of the studies, you pick up the positive. With others, you learn from the mistakes.” As results from the previous tests were published, enthusiasm for CTC technology rose and fell. Yee places herself at the higher end of that spectrum. “I have a great deal of enthusiasm for virtual colonoscopy because we have been using it since 1997,” she says. “I’ve seen the technique evolve and I’ve trained a lot of radiologists on how to interpret it. When properly performed and read by a trained radiologist, it does an excellent job of detecting clinically significant polyps.” At this point in time, Zimmerman is not quite as enthusiastic. “CTC is predicated on many different things, but it is probably sufficient to say that it is still not clear that it is a widely generalizable technique,” he says. “But Dr. Pickhardt produced good results in a large study using the 3-D technique, and he believes the technique is ready to roll on its own.” Widespread Prospects Yee, who is optimistic about the technology’s widespread adoption, is one of the principle drafters of the guidelines. “The guidelines will review how to perform, interpret, and report virtual colonoscopy,” she says. Zimmerman feels that such an approach makes sense. “At the very least, we would need to determine what would be the minimally acceptable technology, interpretation method or time, and software that you would need to accurately determine virtual colonoscopy.” — Dan Harvey is a freelance writer based in Wilmington, Del. He is a frequent contributor to Radiology Today.
“The ACRIN trial is going to accrue hundreds and hundreds of CTC [computed tomography colonoscopy] studies, and what we’re going to do is take a subset of cases—those where polyps have been proven by confirmatory optical colonoscopy—that will be useful to research on computer-aided polyp detection,” says Ronald M. Summers, MD, PhD, chief of the imaging process group at the National Institutes of Health. Summers leads the CAD research component of the ACRIN trial. By the end of the trial, Summers hopes to have roughly 200 proven positive cases that can be used by CAD researchers. “As cases are accrued at the different imaging centers throughout the United States, the cases that have proven polyps that are 7 millimeters or larger will be identified and put into this database,” he says. The ultimate goal, Summers says, is to make it easier and less expensive for researchers, both in academia and industry, to develop computer-aided polyp detection systems. Beyond the ACRIN study, Summers has submitted a proposal to ACRIN to conduct a blinded evaluation of different CAD systems using the database. So far, the ambitious proposal has passed the first round of review. “If the proposal makes it past future rounds of review, I will be asking academic and industry partners to join me in a blinded comparison of CAD systems,” says Summers. Summers has been at the vanguard of research involving CAD. He believes CAD could potentially improve accuracy to a significant degree. “It can reduce interpretation error,” he says. “I see it from a patient-oriented perspective. One of the main benefits I see is that CAD will help patients by improving the accuracy of CTC interpretation. So, more polyps will be recognized by physicians and, hopefully, fewer people will go on to unnecessary optical colonoscopy. I should add that these benefits are unproven at this time. We have very little good data in the literature about how well CAD works, but these are our hopes for the technology.” — DH |
 |
|
3801 Schuylkill Rd • Spring City, PA 19475 Publishers of Radiology Today All rights reserved. |