Skip to Navigation
University of Pittsburgh
Print This Page Print this pages

October 11, 2012

SCIENCE 2012: Panel looks at pros, cons of screening tests

Among the presentations at the University’s annual science symposium was “To Screen or Not to Screen” in which a trio of medical experts discussed the pros and cons of certain medical testing.

Breast cancer screening

Mammography can detect breast cancers, but such testing is not without risks, said radiology faculty member Wendie Berg.

While it is true that early detection of breast cancer can alter the natural course of the disease, reducing mortality and allowing for more breast conservation and less harmful treatments, she said it’s important not to harm healthy women.

“We only expect, on average, five women in 1,000 who are screened with mammography will have breast cancer. We don’t want to be hurting the other 995 we’re screening,” she said.

The downsides of testing include the need for additional testing or biopsies (and the stress that accompanies the uncertainty), false negatives, overdiagnosis and radiation exposure.

Across multiple trials, almost all indicate a net benefit to mammography, showing a 20 percent reduction in mortality for women who were screened. As women increase in age, the benefit is more clear, Berg said. For women ages 60-69 there was a net gain of 32 percent reduction in mortality and a 22 percent reduction for women in their 50s. Younger women saw less of a benefit with a decreased mortality rate of 15 percent associated with screening.

How frequently to screen?

Berg said for women in their 40s, research has shown that screening every two years has an 18 percent benefit while screening annually resulted in a 36 percent benefit. The difference was most significant in that age group, partly due to more aggressive cancers, she said.

However, for women above age 50, the data showed little difference between screening every year and screening every two years. “If we have to consider limited resources, it’s a reasonable place to think about cutting back,” she said.

Another issue that needs to be considered is when to stop having mammograms. “It really doesn’t make much sense if a woman doesn’t have at least a seven-nine year life expectancy,” she said. For women in their 70s who have multiple comorbidities, she said, there is only borderline potential to benefit from a mammogram. Unless there is a clinical abnormality, there is little reason for screening in that age group and beyond, she said.

Downsides to mammography

Berg said about 10 percent of women who have their annual screening may need to be recalled for additional views, ultrasound or both. “Most of those — about 65 percent — will have nothing of concern except that they’ve now gone through a recall, extra testing, extra stress,” she said. Another 20 percent of those who are recalled will be recommended for short-term follow up — yet another stress-inducer, she noted.

The remaining 15 percent who are recalled will be recommended for biopsy. Of them, only 33 percent — or 2-5 women per 1,000 — actually have cancer, she said.

Over a 10-year period, research has shown that nearly one-quarter of women who were screened had at least one false positive result. “Many times it’s the same woman that’s recalled again and again,” Berg said.

Statistics show a decade of screening saves the lives of 250 women per 100,000 or one life saved for every 100 false positive results, she said. “Is that acceptable? It’s a good question,” Berg said.

False negatives are another risk. Mammography is less effective at finding cancers in women with denser breast tissue. And, Berg said, breast density itself is a strong risk factor for developing breast cancer. “For those with extremely dense breasts there is a three-fold to six-fold increased risk compared to women with fatty breasts. That’s as high as any other risk factor we know of and yet has not received much attention until recently,” Berg said.

She noted that bills have been introduced in many states, including Pennsylvania, that would require mammography reports to include information on a woman’s breast density. Advocacy groups (see areyoudense.com or areyoudenseadvocacy.org) are calling attention to the issue.

Other types of testing also risk false positives and unnecessary biopsies. Research has shown that 5 percent of women who have a screening ultrasound are biopsied and that 10 percent of biopsies prompted by ultrasound are cancer. Screening MRIs also result in a 3-4 percent biopsy rate with 40 percent of the biopsies discovering cancer.

Radiation risk increases the younger the woman is, Berg said, which is part of the reason women who are not at high risk are advised to wait until age 40 to start screenings.

Possible other tests to add, Berg said, include ultrasound, which finds an extra three or four cancers per 1,000; MRI, which finds an additional 15-20, and molecular breast imaging, which likewise finds an additional 15-20.

Clinical breast exams find an additional 0.3 cancers per 1,000.

Berg told women who fail to do regular breast self-exams to quit feeling guilty. The practice is no longer recommended (but considered an option) because it has been found to yield no benefit in reducing breast cancer mortality.

Better understanding

“I think it is a very complex situation we have in breast imaging and a lot of opportunities,” Berg said, noting that it’s not possible to conduct long-term studies to compare the benefits of each additional test as technology changes.

However, she said it is important to examine the outcomes of women who are diagnosed with cancer. She advocated collecting information on their breast density, their screening history (including what types of screening they had and how often), the molecular subtype of their cancer, their treatment history and outcomes.

Prostate cancer screening

Joel B. Nelson, chair of the School of Medicine’s Department of Urology, said that since 1975 there has been a rapid increase in the diagnosis of prostate cancer using a prostate specific antigen (PSA) test.

And while death rates due to prostate cancer have declined, 29,000 men will die from the disease this year, he said.

But has broad screening for prostate cancer done more harm than good?

“There’s been a real change in how men first present for prostate cancer,” he said, noting that before PSA testing became widespread, slightly fewer than half of men diagnosed had clinically localized disease. “Fully a quarter of them had metastatic prostate cancer, which is a lethal form of prostate cancer,” he said.

“It’s very different now: About 85 percent of the men who we diagnose at the time they present with this disease have clinically localized disease and only 2 percent present with metastatic disease,” he said.

Studies have shown that screening for prostate cancer reduces deaths due to prostate cancer. PSA tests “will clearly detect this at the time where the cancer is localized, at a more early stage when it is more curable,” he said.

However, research shows doctors must screen 1,000 men to save 1.07 men from prostate cancer death. And most men who have prostate cancer will die of something else even if they are not treated for the disease.

“There were a lot of men walking around not knowing they had prostate cancer until we applied this test,” he said.

“The majority of men alive today with prostate cancer don’t know they have it and frankly they don’t need to know they have it because it will never harm them,” he said. PSA testing is effective in detecting prostate cancer, “but 66 percent — conservatively — of men with elevated levels of PSA in fact don’t have the disease,” he said.

“So you’re applying a tool that’s relatively crude to pick up a disease you want to pick up very specifically,” Nelson said, adding that increased diagnosis comes at the cost of potentially unnecessary treatment. “There’s no question that the treatments we apply are harmful.”

Nelson noted that the American Cancer Society has stopped urging doctors to offer PSA tests. He finds a compelling case for not diagnosing and treating prostate cancer “because on balance we’re harming many more men than we benefit.” However, a decrease in screening likely will result in a higher rate of metastatic disease.  “One of the major challenges we have is: Who has a cancer that is in fact significant vs. who has one that would only be diagnosed at the time of autopsy?”

Cardiovascular screening

Emma Barinas-Mitchell, an epidemiology faculty member in the Graduate School of Public Health, discussed the use of carotid intima-media thickness (CIMT) tests to screen for the risk of cardiovascular disease.

Over the past 50 years, deaths from cardiovascular disease have been decreasing. Still, cardiovascular disease accounts for one-third of all deaths in the United States, with heart disease being the No. 1 cause of death and stroke being the 4th-highest cause.

Cardiovascular disease also is expensive, accounting for one dollar of every six spent on health care, or a total of $44 billion annually.

Barinas-Mitchell said the atherosclerotic process underlies a majority of cardiovascular disease events. A leading cause of heart disease and stroke, atherosclerosis is caused by a chronic immune inflammatory process that begins with plaque development and moves on to stenosis, calcification and eventually rupture, leading to heart attacks or strokes.

Because the disease begins early in life and develops over decades, it lends itself to screening. If detected early, the process can be averted, saving lives.

CIMT uses ultrasound to measure two layers of the arterial wall: the inner layer, or intima, and the second layer, or media. The thickness can be used as a measure of atherosclerotic potential and a predictor of heart attack or stroke.

However, many observational studies show thickness is related to risk factors such as age, gender, race, smoking or cholesterol levels. She said the Framingham risk score, which calculates a person’s 10-year cardiovascular risk based on such factors, is an accurate tool.

“We know that CIMT predicts disease but does it predict disease better than what we already use clinically?” she said.

Some research has shown coronary artery calcification scores were a better measure than CIMT.

For those at low risk, “The incremental value with CIMT wasn’t enough to recommend using it widespread,” she said. However, there may be some benefit in using it in asymptomatic adults at intermediate risk.

Barinas-Mitchell said that CIMT is valuable research tool for understanding how and why atherosclerosis develops, but before implementing it clinically on a widespread level, better guidelines need to be established.

—Kimberly K. Barlow

Filed under: Feature,Volume 45 Issue 4

Leave a Reply