Choosing an Imaging Test

Ed Epstein has lived a dozen years with non-Hodgkin follicular lymphoma, which was first diagnosed as stage 4 when two lumps emerged on his head. It has since recurred twice in his kidneys, both times picked up by imaging tests.

The 75-year-old retired healthcare administrator, who has controlled the slow-growing form of lymphoma with a series of drugs and an experimental vaccine, knows that his cancer-free status could change at any time. “I look at what I have as a chronic disease that I have to treat as it pops up,” says Epstein, a suburban Chicago resident.

Needless to say, numerous imaging scans have played a major role in Epstein’s diagnosis and follow-up: magnetic resonance imaging (MRI) scan, more than a dozen computed tomography (CT) scans and a handful of kidney scans.

Cancer diagnosis and treatment, as any patient quickly learns, involves not only a new world of therapies, including drugs, surgery and radiation, but also a battery of imaging tests to ensure the right therapies are selected. Never has imaging technology been more sophisticated. But the options are not always created equal, in the sense that different types of scans might be ordered depending upon a patient’s medical circumstances.

What type of scan is ordered, for example, may vary significantly depending upon where an individual falls along the treatment spectrum: cancer screening, disease staging or long-term monitoring? The importance of so-called false-positives—a test result implying a condition exists when, in fact, it doesn’t—can vary significantly depending upon the answer to that question.

“False-positives can lead to unnecessary biopsies or therapies in addition to the significant anxiety it causes patients and their loved ones,” says Charles L. Smith, PhD, vice president of Cancer Care Services at US Oncology. “Conversely, false-negatives (not seeing the problem on the imaging results) can also lead to serious consequences because a cancer could go untreated until detected further down the road.”

Physicians must also consider the type of cancer involved. An MRI might be an ideal test for looking at soft tissue, bone or the brain, says John Barstis, MD, a clinical professor of medicine in the division of hematology/oncology at the University of California, Los Angeles. But in some cases, there might be no better technology than an ultrasound, which uses sound waves to determine whether an ovarian cyst is filled with fluid or denser, potentially cancerous tissue, he says.

“It’s important to understand that there isn’t a simple [imaging] hierarchy,” Barstis says. “It’s not that CAT scans [also known as CT scans] are better than ultrasounds and MRIs are better than CAT scans. They all have strengths and weaknesses depending on the situation.”

Routine cancer screening in people without any symptoms is usually only recommended for malignancies in which earlier detection has been shown to boost survival, says Stephen Balter, PhD, a medical physicist and clinical associate pro-fessor of radiology and medicine at Columbia University Medical Center. For instance, mammograms have been shown to identify some breast malignancies at a more curable stage, he says.

When recommending screening tests, doctors also weigh other factors, such as the person’s age, he says. “At a younger age, something that is suspicious is more likely to be a false-positive, which could lead to un-necessary procedures and needless anxiety.”

To describe the relative accuracy of a particular imaging test, physicians will sometimes talk about its sensitivity (ability to detect disease) and its specificity (accuracy). One type of sensitive test, according to Barstis, is a bone scan, in which a small amount of radioactive material injected into a blood vessel gets absorbed by the bones and is detectable via scanning.

But a bone scan is not highly specific, in terms of distinguishing cancer from a myriad other nonmalignant causes. Bone damage from a prior fall might appear worrisome, Barstis says. Elsewhere in the body, a suspicious spot on an imaging scan might be a prior infection or a benign lump, among other findings.

The issue of sensitivity versus specificity has figured prominently in the debate over mammography screening for women without symptoms. (There’s no debate if the woman or her physician has felt a lump.) To illustrate the risk-benefit equation, researchers will sometimes break down the number of mammograms needed to prevent cancer mortality and weigh that number against other risks, like false-positives.

In a 2009 Cochrane Review analysis of seven randomized studies, researchers estimated that 2,000 women aged 40 to 49 would have to undergo annual mammography screenings over a 10-year period to extend the life of one woman through early detection. During that same span of time, 10 women would be unnecessarily treated for breast cancer and more than 200 would suffer anxiety related to false-positive findings.

It’s not that CAT scans are better than ultrasounds and MRIs are better than CAT scans. They all have strengths and weaknesses.

Understandably, the risk-benefit debate essentially vanishes once someone has received a cancer diagnosis. But the imaging test that’s recommended—to stage the cancer and sometimes to monitor treatment response—varies depending upon its location and type, whether a biopsy is needed and the costs involved, Barstis says.

An MRI, which uses radio waves in a powerful magnetic field to create an image, doesn’t work well on areas of the body that involve motion, such as the lungs, he says. Even the slightest movement can blur the picture.

Some scans create an anatomical image, such as through a CT scan, which uses X-rays to take a series of cross-sectional pictures. Others are considered more of a functional test, such as a bone or a PET (positron emission tomography) scan, showing the metabolic activity of the cancer.

In a PET scan, a small quantity of radioactive deoxyglucose (a modified sugar) is injected into a vein, where it travels throughout the body. Since cancer cells often use more deoxyglucose than normal cells, the goal is to highlight potentially malignant areas. But that increased glucose uptake could be for one of many nonmalignant reasons, including an infection.

“Interpretation of the scans by a radiologist highly experienced with the modality is an important factor in making sure the imaging study yields the right information,” Smith says. “Because the results of imaging studies play a crucial role in making treatment decisions for cancer patients, having the study performed with the right technology and interpreted by an experienced, accredited radiologist is an important facet of the care cycle.”

More recently, PET scans are often combined with a CT scan, Barstis says. The combined image allows the physician to see not only if there is intense uptake that points to potential cancer, but also its precise location.

The blending of imaging information from PET (measuring metabolic activity) and CT (providing anatomic detail) scans has become a powerful and broadly accepted tool in imaging cancer patients, Smith says.

PET scans have been helpful in determining treatment response in some types of lymphoma, such as Hodgkin, says Koen van Besien, MD, a lymphoma specialist at the University of Chicago Medical Center. While fibrous tissue may be left behind after treatment, the PET scan can detect remaining cancer activity in those areas, he says. “It helps in Hodgkin disease to change treatment mid-course, if necessary.”

Also, a PET scan can help project survival odds, van Besien says. He cites one study, published in 2008 in the journal Blood, which found that 94 percent of those patients who completed treatment with no visible signs of cancer on the PET scan hadn’t relapsed one year later.

There is so much angst, understandably, when an individual has had cancer. You want to know yesterday, today and tomorrow that everything is OK.

One of the biggest imaging dilemmas for physicians and patients alike can develop after treatment ends, says Albert Blumberg, MD, who chairs the American College of Radiology’s Commission on Radiation Oncology. Some patients prefer repeated imaging tests over many years, he says. “There is much angst, understandably, when an individual has had cancer. You want to know yesterday, today and tomorrow that everything is OK.”

Unfortunately for some cancers, there are few effective drugs or other treatment options available, if it does recur, Blumberg says. One example is lung cancer, he says, another is a type of brain tumor, called glioblastoma. “There is no reason to be aggressively participating in surveillance, if once you have that information, you can’t do anything curative with it.”

In other cancers, such as breast, routine monitoring via mammogram can potentially make a difference, Blumberg says. Even with breast cancer, though, more intensive imaging isn’t necessarily better, according to a 2009 Cochrane Review analysis following women with early-stage disease. Women who got additional tests, such as a bone or liver scan, didn’t enjoy any survival advantage over women who followed up with a regular physician’s exam and annual mammogram only.

Monitoring also can sometimes be excessive in patients with Hodgkin lymphoma, van Besien says. The University of Chicago physician says he and his colleagues typically only recommend CT scans in the first three to five years after treatment wraps up because that’s when the chance of recurrence is highest.

Van Besien also worries about PET scan monitoring, particularly in slow-growing types of lymphoma, such as Epstein’s. A slow-growing malignancy might not uptake sufficient glucose to appear on the PET image, he says. “No test comes without side effects. I think the most important side effect of overuse of PET scans is all kinds of false positives and extra biopsies that can cause anxiety.”

To date, Epstein’s oncologist hasn’t advised a PET scan, and his recurrences have been identified by routine CT scans. In 2004, a CT scan identified a tumor wrapped around his left kidney. Five years later, long after the first tumor was eliminated with drug treatment, another CT scan found a tumor on his right kidney.

So Epstein visits his oncologist every three months for a checkup, including a physical exam to look for lumps. And if Epstein’s lymphoma does recur, it will most likely be caught, not by a scan, but by him or his physician. In fact, according to a review article published in the Journal of Clinical Oncology, 80 percent of lymphoma relapses are identified by symptoms caught by the patient or physician.

“Being informed about the benefits and risks associated with imaging for cancer staging and monitoring can help patients understand the management strategy they plan with their oncologist,” Smith says.

But in recent years, Epstein says he’s become more concerned about limiting his scans as he adjusts to a lifetime of cancer treatment and monitoring. One of his primary concerns: subjecting himself to false cancer alarms and related biopsies, something shown to be a problem in monitoring scans. He might get another CT scan this fall, his first in two years, if his oncologist insists. “My attitude is the less imaging, the better, unless it’s really needed,” he says.

With more imaging options than ever, it’s important that patients and survivors gain a better understanding of some of the variables that can influence which types of scans are suggested and why, physicians say.

For individual women, though, breast screening is a highly personal decision, says Phil Evans, MD, a clinical professor of radiology and director of the University of Texas Southwestern Center for Breast Care at Dallas. “Most women feel like finding breast cancer is more important than any false-positive that they might have.”

A woman’s personal cancer risk also can influence screening decisions. In 2007, the American Cancer Society updated its guidelines to recommend an MRI, along with mammography, in women at high risk—defined as those with at least a 20 percent or higher lifetime risk of developing a breast malignancy. Although MRI carries a higher risk of false-positives (meaning it’s less specific) compared with a mammogram, it’s also more sensitive and thus more likely to catch small malignancies in these more vulnerable women, Evans says.