Women urged to consider mammograms' hidden risks Cause for concern? By Judy Peres Tribune staff reporter February 17, 2002 The roiling debate over the validity of mammogram studies is bringing renewed attention to the common medical procedure, raising new, unsettling questions about its usefulness. Most medical experts remain convinced that mammograms save lives and that women should continue having them. But doctors and ethicists also say women need to weigh that benefit against some unadvertised costs before deciding when--and whether--to be screened. Women should know there are breast cancers so virulent that they are likely to be lethal no matter how early they are detected, these experts say, as well as others so indolent they would not become life-threatening even if never detected. And mammograms are more likely to find the indolent ones. Because doctors cannot yet distinguish the slow-growing tumors from the ones that can spread and kill, some women are being treated with surgery, radiation and toxic drugs for cancers that would never hurt them. And because mammography remains an extremely imprecise test, many more endure the fear that they might have breast cancer until additional tests rule it out. Mammography flags a huge number of spots that turn out not to be cancer. It also misses 1 in 5 cancerous tumors, and even the most experienced radiologists can disagree on what a test shows. "Most people think it's just common sense that detecting something early is going to be beneficial," said Dr. Steven Woolf, a professor of family practice at Virginia Commonwealth University and a member of the U.S. Preventive Services Task Force. "They assume there's little to lose and much to gain. "If they had a better sense of the downsides, they might be a little more careful." Mammograms, specialized X-rays that can detect breast tumors months or even years before they can be felt, are considered the best defense against a potentially lethal disease. Because smaller tumors typically are more treatable, women in the U.S. are urged to have regular mammograms, generally starting at age 40. Women have been told, based on eight large clinical trials, that mammograms can reduce their risk of dying of breast cancer by some 30 percent. But last fall two Danish researchers published a new analysis of those trials, casting doubt on the earlier conclusions. The two researchers said some of the trials were unacceptably flawed and others were of very poor quality. The only studies that were methodologically reliable, they found, showed no difference in death rates between women who got the test and those who did not. While authorities are taking a closer look at the trials, they say current guidelines should continue to be followed. The National Cancer Institute recommends that women 40 and older have mammograms every year or two--although both that agency and the U.S. Preventive Services Task Force have discussed revising their guidelines. The debate over official mammography recommendations arises partly because screening mammograms are done on women who appear to be healthy, unlike diagnostic mammograms, performed on women with symptoms that might indicate breast cancer. "You have to be more careful with screening because you're aiming at a healthy public," said Maryann Napoli of the New York-based Center for Medical Consumers. "You don't want to give people a problem where they didn't have one." Overdiagnosis Thirty years ago, the theory behind mammography made sense. Like a more powerful telescope that allows an astronomer to see stars never before detected, mammography enabled doctors to see things that previously had been invisible. Experts believed they would be able to improve survival by treating relatively early tumors before the cancer had spread to vital organs. It was assumed that, given time, all breast tumors would grow, spread and kill. But that assumption was based on the treatment of patients in whom doctors were able to see or feel golf ball-size tumors that likely had already spread. As it turned out, all cancers are not equally threatening. Mammograms detect not only micro-tumors, as expected, but also tumor look-alikes that grow very slowly or not at all and that will never spread or even become physically evident during a woman's lifetime. These "pseudo-tumors" look like the real thing under the microscope, but they don't behave the same. Detection of such non-lethal malignancies is known as overdiagnosis. Largely because of overdiagnosis, an estimated 203,500 cases of breast cancer will be reported this year. Cancers are born when a mutation occurs in a normal cell, causing it to start dividing and multiplying out of control. It then grows, over months or years, into a pulpy mass from which cells break off and circulate throughout the body, looking for a place to colonize. Breast cancer patients don't die from the original tumor; they are killed by cell colonies that take over vital organs such as the liver, lungs or brain. What scientists have only recently begun to appreciate is that many cancers don't have the ability to go through all the steps necessary to become life-threatening. Some clusters of cancerous cells might sit there, innocuously, for decades. Others actually disappear on their own. Tumor cells might circulate through the bloodstream but never find a place to colonize. Or they could form a beachhead on a distant organ but not have the capacity to sprout new blood vessels so they can continue growing. Experts are quick to point out that overdiagnosis only rarely reflects incompetence on the part of those who analyze tumors. "They're not misdiagnosing benign disease and calling it cancer," said Dr. Craig Henderson, a medical oncologist at the University of California-San Francisco. "But we define cancer, as we've done for 50 years, by how it looks under the microscope--not by how it behaves. A good percentage of the things pathologists call malignant will never kill you." Given the current limited understanding of cancer biology, when doctors find a small cancer through a mammogram, they can't tell for sure whether it will grow and spread or remain static. Scientists are working to identify genetic markers that will permit pathologists to distinguish deadly breast tumors from the indolent variety. And a few genes are already helping clinicians predict which breast cancers have greater potential to do harm. But for now, no one can say with certainty which ones can be left untreated. Overtreatment Because deciding to wait and monitor a malignant tumor might endanger a woman's life, virtually all women diagnosed with breast cancer have the tumor, or the entire breast, surgically removed. Most also get radiation, chemotherapy and hormonal treatments. Cancer treatment has become more refined, inflicting less collateral damage than it did in the past. But chemotherapy still carries side effects, from hair loss to leukemia; radiation can damage blood vessels; and hormonal treatments increase the risk of blood clots and uterine cancer. Oncologists readily acknowledge they have been overtreating most women--"curing patients who didn't need to be cured," in the words of Dr. Barnett Kramer, director of the Office of Medical Applications of Research in the National Institutes of Health. But, for now, many believe there is no choice. "You have to err on the side of caution," said Craig Jordan, director of breast cancer research at Northwestern University's medical school. Women whose cancers are found on screening mammograms generally have a better prognosis--not just because of earlier detection but also because many of those tumors are inherently less aggressive. Yale University researchers followed two groups of women with small tumors that had not spread beyond the breast. In one group, these early-stage cancers were found through routine mammograms; the others were diagnosed only after symptoms appeared. Seven years later, all but 2 percent of the women whose cancers were detected on mammograms were alive and well. But 13 percent of those with symptomatic tumors of the same size had relapsed or died. "Our interpretation is that screening mammography is finding mostly non-lethal tumors," said Dr. Sandra Moody-Ayers, the study's primary author. Direct evidence that not all breast cancer kills comes from autopsies showing that many women die with breast cancer but not from it. In several studies, post-mortem exams of women who died of something other than breast cancer found that 2 percent to 16 percent of them had invasive breast tumors that never got big enough to be detected while they were alive. While epidemiologists have come to recognize that the more than 30 million mammograms performed on U.S. women every year are finding a significant number of harmless tumors, many clinicians are reluctant to talk about overdiagnosis. Dr. Timothy Cote, medical director of the cancer statistics branch of the NCI, is among those who fear that women will distill the wrong message from a public discussion of overdiagnosis. "Regular screening is important," Cote said. "Diagnoses of breast cancer are believable, and appropriate treatment by qualified physicians should follow. Most women diagnosed with breast cancer, if left untreated, will have very serious consequences." But others say understanding overdiagnosis can be useful to women. Just knowing that not all breast cancers are potentially fatal can be an eye-opener and reduce panic in the event of a positive test result. In addition, before deciding when to start screening and how often to do it, women might want to know that mammograms are likely to detect some very slow-growing cancers that might never cause symptoms. That information could also be helpful to women deciding what treatment to seek once a cancer is confirmed. Finding `precancer' The most dramatic consequence of widespread mammography screening has been the huge increase in the incidence of tiny breast cancers known as DCIS, for ductal carcinoma in situ. DCIS is entirely confined within a milk duct and hasn't invaded the surrounding breast tissue. Some experts question whether DCIS should even be considered breast cancer, sometimes describing it as "precancer" even though it is malignant. And the NCI's breast cancer database does not include non-invasive tumors in its overall breast cancer rates. But doctors generally think of DCIS as breast cancer because some DCIS has the potential to turn into invasive cancer. Nevertheless, in many cases, those tiny cancers will never become life-threatening because they're so slow-growing, research shows. Some may even disappear on their own. Because doctors can't tell for sure which cases of DCIS may eventually become dangerous, they all get treated. That almost always means surgical removal and, in about half the cases, additional treatment such as radiation or hormonal therapy, both of which can have harmful side effects. According to the American Cancer Society, nearly 48,000 new cases of DCIS will be diagnosed in the U.S. this year. That's an increase of about 900 percent over 1983, when there were only 4,900 cases. An epidemic? Such an explosion might look like an epidemic, except that nearly all of the increase is due to mammograms picking up abnormalities that were probably there all along. Autopsy studies show that up to one-third of women who die from causes other than breast cancer have non-invasive tumors that went undetected. The vast majority of DCIS cannot be felt, which explains the relatively small number of cases recorded before mammography became widespread. "The concern these data raise should not be over what is causing the `epidemic,'" said Virginia Ernster, professor of epidemiology at UC-San Francisco, "but over whether it's important." Ernster, a nationally recognized expert on DCIS, points out that the incidence of in situ (Latin for "in place") breast cancers rose relatively slowly between 1973 and 1982, but then accelerated from 1983 to 1992. "The marked increase coincides with the increasingly widespread use of screening mammography," she said. While only 134 mammography machines were installed in the U.S. in 1982, there were about 10,000 by 1990. Many doctors consider detection of DCIS a benefit of mammography because it allows a potentially lethal disease to be treated early. But no one knows what proportion of DCIS would become dangerous if left untreated. Overall, about one-fifth of women newly diagnosed with breast cancer have DCIS, and their prognosis is excellent: More than 98 percent are alive 10 years later. But it's unclear whether that's because the treatment is so effective, because DCIS is so benign to begin with--or both. "It has yet to be demonstrated that detection of DCIS actually will extend [women's] lives," Ernster said. False negatives, false positives Interpreting mammograms is an inexact science, and some radiologists see threats where others don't. When 10 radiologists were asked in a 1994 study to read the same 150 mammograms, there was substantial disagreement in 25 percent of the cases over what the mammogram showed. When a mammogram seemingly finds nothing, that information should be kept in perspective. "When I have a mammogram and they say it's OK, I feel it's a guarantee I won't get breast cancer for the next 12 months," said Brooklyn marketing executive Sharyn Miller. "Clearly, that's not true. But emotionally, that's my reaction. It's almost like it gave me a preventive ray--zap!--you're OK for another year." In fact, screening mammograms miss about 20 percent of all breast cancers, according to the NCI. The danger of false negatives is that they can instill a sense of complacency, causing patients not to report suspicious symptoms and physicians not to follow up. That could end up delaying necessary treatment. Several studies have shown there are factors that can affect the rate of false negatives: the size and location of the lesion, the density of the woman's breast tissue, the quality of the X-ray images and the skill of the radiologist interpreting them. Premenopausal women, whose bodies are still producing high levels of female hormones, are most likely to have dense breasts. Experts believe that's one reason mammograms miss nearly half of all breast cancers in women younger than 40. On the other hand, for women older than 65, whose breasts contain more fatty tissue, the false-negative rate is only about 10 percent. An even bigger cost of mammography screening is the high risk of a false positive. Each year, hundreds of thousands of women are told their mammograms are suspicious and their breasts must be biopsied, only to learn--after days or weeks of anxious waiting--that cancer is not present. An abnormal mammogram can trigger a cascade of painful, expensive and time-consuming procedures. Every questionable mammogram has to be followed up, even though less that one-half of 1 percent of those who get mammograms will turn out to have breast cancer. The follow-up usually entails additional X-rays or ultrasound images, but it frequently includes a biopsy--removing a sample of the suspicious tissue so pathologists can study it under the microscope. Radiologists estimate that 300,000 to 600,000 biopsies a year are performed because of screening mammograms. A 1998 study published in the New England Journal of Medicine estimated that by the time a woman has had 10 mammograms, her risk of having at least one false positive is 49 percent, and her risk of having a benign biopsy is 19 percent. Miller, 52, has had four biopsies in the past eight years. Although they all proved benign, she said: "I go in for my annual mammogram with my heart in my mouth." Wait and watch or biopsy? Radiologists know certain abnormal mammograms are highly unlikely to indicate cancer, and they are advised to just "wait and watch" in those cases rather than order an immediate biopsy. But that doesn't always happen. "Some doctors are so concerned about malpractice suits that they recommend biopsy," said Dr. Stephen Feig, director of breast imaging at Mt. Sinai Hospital in New York. "And some patients can't live with the uncertainty." With scientific confidence in the lifesaving benefit of mammography at least slightly shaken, many experts say it's more important than ever to consider the potential risks of the screening test. Dr. Michael Baum, a breast cancer expert at University College London, believes it's misleading for health officials to continue telling women "screening can save your life" without also discussing the possibility of overdiagnosis, false alarms and the other downsides of testing. "We really must inform women of these risks," Baum said. Copyright (c) 2002, Chicago Tribune Remember we are NOT Doctors and have NO medical training. This site is like an Encylopedia - there are many pages, many links on many topics. Support our work with any size DONATION - see left side of any page - for how to donate. You can help raise awareness of CAM.