Star Trek fans have long known that the future of medicine includes handheld gadgets that scan the human body, lighting up when they find an abnormality. Some technological prognosticators say that Stardate is not as far off as one might think.
Since coming on the scene in the early 1990s, spiral computed tomography (CT) scanning has become a common tool for diagnosing and staging a wide variety of cancers. The technology has become so good and so fast that a growing number of facilities are using the scanning technique to screen asymptomatic people for cancer, primarily lung and colorectal cancers.
To date, screening has been limited to high-risk individuals and patients willing to pay for the high-tech imagery. But advocates are confident that studies currently in the works will back up their claim that CT screening catches cancers often enough and early enough to improve patient outcomes. And it is only a matter of time before earthlings will head to the doctor for their routine body scan.
Spiral CT makes a difference
When symptoms indicate an underlying cancer, CT often is the first modality doctors turn to in making a diagnosis. By improving the detection rates for a number of cancers including pancreatic, hepatic and lung, spiral CT scanners have made CT an easy diagnostic choice.
Spiral CT generates 2D and 3D images that, when combined with advanced imaging software, create a “virtual reality” landscape of the body’s internal structures. The faster scanning speed also allows images to be obtained within a single breath-hold, reducing radiation exposure and eliminating motion artifacts.
These advantages have some radiologists turning to spiral CT scanning to screen for cancer as well. So far, lung and colorectal cancer screenings have taken the lead. Lung cancer, the most common and deadly cancer, kills more than 150,000 people a year — 85 percent of whom smoke. Because lung cancer usually spreads beyond the lungs before symptoms appear, five-year survival rates are only 14 percent. So, detecting lung cancer early through screening could save many lives.
Chest X-rays, studied in the past, did not catch cancers early enough to improve survival rates. Even those at high risk, such as smokers, are not screened. Screening for lung cancer using spiral CT, while still in its infancy, is gaining ground.
Colorectal cancer screening is recommended for those at high risk and individuals over 50 years of age. Despite colorectal cancer’s high survival rate when detected early and its frequency as the third most common cancer, the unpleasant and uncomfortable colonoscopy screening procedures currently available keep most people away. Spiral CT advocates say their procedure, virtual colonoscopy, is a more comfortable alternative that also catches more pre-cancerous abnormalities. Only a few facilities, however, currently offer the procedure and it is not recommended for patients at high risk for colon cancer.
“To get something into a screening modality, it has to be low cost and have rapid access,” says Michael Plunkett, M.D., chairman of the department of radiology at Abbott Northwestern Hospital (Minneapolis).
Spiral CT screening still is young, and insurance companies have not jumped in to help pay for it. Even if the procedure saves lives, Plunkett says he is not optimistic that insurance companies will come around. The money saved from catching cancer early in a few people would have to outweigh the cost of mass screenings.
Lung cancer has the best shot at meeting screening criteria, Plunkett says, adding that other cancers, such as pancreatic, may not stand up to such scrutiny. Plunkett also hesitates to endorse virtual colonoscopy, at least for now.
“At this time, [conventional] colonoscopy is the main screening modality and has much higher sensitivity and specificity,” he says, adding that “within five years, spiral CT and doing these virtual reality walks through the colon will be common.”
Virtual catches on
Some call it virtual colonoscopy; others call it CT colonography. Whatever the name, David Vining, M.D., associate professor at Wake Forest University Baptist Medical Center (Winston-Salem, N.C.), says virtual colonoscopy is catching on.
“The technology is there to do this,” says Vining. “The biggest challenge we face in terms of the radiologists is teaching old dogs new tricks.”
Vining, considered the founder of virtual colonoscopy, has been performing the procedure since the early 1990s. Because insurance does not reimburse the procedure as a screening method, most patients who undergo virtual colonoscopy come in with abdominal pain and are scanned as a diagnostic tool. Others are part of clinical trials or pay out of their own pockets.
Despite the procedure’s $750 price tag, the screening method is popular, Vining says, with people flying in from around the world.
To date, only Wake Forest University Baptist Medical Center, Boston (Mass.) University Medical Center and the Mayo Clinic (Rochester, Minn.) perform the procedure for money. Other facilities are participating in a multi-institutional study headed by the Medical University of South Carolina (Charleston, S.C.). The two-year study will compare the effectiveness of virtual colonoscopy and conventional colonoscopy as screening methods.
Vining says such studies are important, but adds that anecdotal evidence already suggests virtual colonoscopy catches more abnormalities than colonoscopy, which has been shown to have a 24 percent miss rate. The negative predictive value — or the ability to predict that a patient is free of cancer — is 98 percent for virtual colonoscopy. However, the positive predictive value is 60 percent. If virtual colonoscopy finds a significant polyp, colonoscopy finds it 60 percent of the time.
That creates a dilemma, Vining says.
“If we call a polyp and they don’t find it in colonoscopy, the question still remains: Was that a false positive on our part or a false negative on the part of the colonoscopy?” he adds. “That’s hard to resolve. No one has given up their colon to let us resolve that issue.”
C. Daniel Johnson, M.D., and Abraham H. Dachman, M.D., of the Mayo Clinic, agree that more studies must be done, but believe the procedure — which they refer to as CT colonography — has great potential for colorectal screening. In the August issue of Radiology, the doctors wrote that in addition to greater abnormality detection, increased patient compliance will be a key to the method’s ultimate success as a screening modality.
Vining says that bowel preparation, while still necessary, is milder compared to other screening methods. The procedure also uses a smaller tube and the scan is done in approximately 20 seconds. After the procedure, image analysis and reporting take approximately 10 minutes.
Another benefit is the procedure’s ability to find diseases outside the colon, such as gallstones, kidney stones, abdominal aortic aneurysms and cancers in other organs — problems colonoscopy cannot catch. This benefit can save the patient months of tests and uncertainty.
Colonoscopy’s advantage, however, is its coupled diagnostic and therapeutic value, in that the gastroenterologist can take out any abnormalities or polyps when they are found. As a screening method, Vining says that benefit is countered by the small number of people who need a polyp removed. In the asymptomatic population, only 20 percent have polyps and just 10 percent of those are significant.
Despite virtual colonoscopy’s advantages, Vining says the study will have to become less expensive than colonoscopy to be economically feasible. Colonoscopy costs about $300 to perform, but despite virtual colonoscopy’s current inflated price tag, it costs just $150 to perform. The procedure has room to move down, and Vining is confident it will to remain competitive.
Although Vining touts the benefits of virtual colonoscopy, he predicts a synergistic relationship for the two procedures in the future.
Lung cancer screening
When it comes to diagnosing lung cancer, what is more effective — a CT scan or a chest X-ray? A CT scan, so say results of the Early Lung Cancer Action Program (ELCAP), which were published in the July 10, 1999 issue of The Lancet. ELCAP is a product of researchers at Weill Medical College of Cornell University (New York), New York University Medical Center (New York), and McGill University (Montreal, Quebec City, Canada).
 |
 |
| With the use of CT scans, researchers at Weill Medical College of Cornell University can show the size of a tumor in a lung (left) and its growth three months later (right). | |
In the study, 1,000 heavy smokers who had no cancer symptoms were given chest X-rays and low-dose CT scans. According to the researchers, CT scans can catch substantially smaller tumors than chest X-rays. CT scans were four times more likely to find malignant tumors than chest X-rays and six times more likely to find stage I cancers. If caught before it has a chance to spread, the five-year survival rate for lung cancer is 60 percent, according to the American Cancer Society. Only 15 percent of lung cancers are caught that early, however, and the survival rate drops to 2 percent after it has spread.
Although the results are promising, no public health agency has issued recommendations regarding lung cancer screening. David Yankelevitz, M.D., professor of radiology at Weill Medical College and an ELCAP author, says the results are so encouraging that hospitals are beginning to advertise lung screening to the public.
“Clearly, there is a need for more guidelines to be made,” says Yankelevitz. “As to whether we will find early lung cancers, there’s no doubt. It’s strictly a matter of quantifying how beneficial it is.”
Not everyone agrees. National Cancer Institute (NCI of the National Institutes of Health, Bethesda, Md.) officials are urging caution, saying that no outcome studies have been done to prove that finding cancers early saves lives or that too many false positives will not do more harm than good.
“People at NCI are worried that there is going to be an overdiagnosis problem and a lot of people are going to be getting biopsies and thoracotomies just based on the fact that radiologists have a really hard time interpreting those scans,” says NCI spokesman Brian Vastag.
Advertising an unproven procedure also raises the concern that CT screening will become the de facto standard, making it hard for people to accept being randomized for clinical trials.
In September, NCI began enrolling 3,000 people in a randomized study to compare chest X-rays and spiral CT scans. The one-year study will assess patients’ follow-up care to see if CT scans catch early cancers or if too many unnecessary biopsies or surgeries are performed.
Yankelevitz welcomes the NCI study and others. Many questions remain: how many patients will have cancer? Which population groups should be screened? How frequently should they be screened? And what is the best way to manage patients?
But Yankelevitz disagrees that the procedure is dangerous. “People keep talking about it like you keep finding all of these abnormalities,” he says. “Well, that’s not true.”
In the ELCAP study, 26 people out of 1,000 were recommended for biopsies, of which 25 were malignant and one benign. Two patients requested thoracotomies despite recommendations against them. Yankelevitz points out that the ELCAP study published the baseline, or first scan, results, but that it is the follow-up scans that are important. Comparing the procedure to mammography, Yankelevitz says that baseline “abnormalities” are more difficult to read, but they also are regarded with less suspicion. In the ELCAP study, about 27 percent had abnormalities in the baseline scan, but only 2 percent to 3 percent had abnormalities in the repeat scans.
“Once you’ve gotten past the baseline, it’s very easy to do this test,” Yankelevitz says. Results of the one-year follow-up to the ELCAP study have not yet been published.
Yankelevitz agrees that learning how to manage patients, particularly following the baseline scan, is an important step needing further study. At Weill Medical, where procedures are done under a strict research protocol, most abnormalities bigger than one centimeter are biopsied. Abnormalities less than one centimeter are monitored with follow-up scans to determine if they are growing. The screening (not covered by insurance) costs $300 at Weill Medical.
Other imaging modalities also play a role in managing patients, but those roles are changing as technology improves, Yankelevitz says.
For example, PET scanning, which works well for tumors larger than one centimeter, will not be useful if CT scanners are picking up six millimeter tumors. Spiral CTs, able to view two millimeter sections today, will view smaller and smaller sections in the future, down to a few hundred microns, he says.
Beam me up, Scotty
“I can see very clearly that the whole approach now to the way screening tests are done is going to move more and more toward image processing techniques,” says Yankelevitz. These techniques include computer-aided detection and analysis that will not only detect and characterize abnormalities, but also give a percent probability of malignancy and make recommendations based on large databases of information.
“We’ll be able to make rational scientific decisions about the frequency of screening and the populations that should be screened,” he says.
Robert Smith, Ph.D., director of cancer screening at the American Cancer Society, says cancer screening, particularly for lung cancer, could save many lives if it pans out. “It’s important to keep in mind that when you’re screening for cancer you’re testing the many to find the few,” he says. “Typically, one of the underlying criteria is that the test is fairly simple and fairly cheap.”
Individual tests, such as lung cancer screening, may reach that criteria in the near future, but Smith is not ready to say full body screening, ? la Star Trek wizardry, is on the horizon.
