WHY MAMMOGRAMS ARE NOT THE ANSWER. Search and Destroy by Shannon Brownlee In 1986 the Japanese health department launched a campaign to screen infants for neuroblastoma, the second most common form of early-childhood cancer, after leukemia. The test was easy to administer: Parents simply pressed a piece of filter paper to their baby's wet diaper, allowed the paper to dry, and then mailed it to a laboratory. Doctors had known since the 1950s that neuroblastoma tumors cause the body to excrete an unmistakable chemical signature in the urine; in the 1980s analyzing that signature became simple enough to perform on a large scale. With the advent of this simple test, Japanese doctors reasoned that they could detect tumors in their earliest stages, before they began producing symptoms, and thus improve the odds of babies surviving the disease. Parents complied with the public health campaign in droves. By the mid-1990s 80 percent of Japanese infants, nearly one million per year, were being tested. Not surprisingly, diagnoses of neuroblastoma shot up, as much as tenfold in some areas. The cure rate also rose dramatically. Before screening began, neuroblastoma--which usually appears between six months and six years of age--killed 80 percent of its young victims in Japan. By 1995 practically every Japanese child whose tumor was detected through screening was successfully treated with surgery, radiation, and chemotherapy. There was just one problem: While the cure rate for neuroblastoma rose, the overall death rate--that is, the number of Japanese children per 100,000 who died of the disease--scarcely budged. The reason for this apparent contradiction, it later turned out, is that there are two different forms of neuroblastoma, both of which look identical under a microscope, but only one of which is picked up by the screening test. That type, moreover, often disappears spontaneously without treatment. Meanwhile the other, potentially deadly form of neuroblastoma cannot be detected by the test early enough to affect survival rates because it either does not put out the telltale chemicals until it has already progressed to an advanced stage or the tumor is so aggressive it resists most treatment. In other words, Japanese doctors have been using an array of intrusive, potentially dangerous treatments on infants to "cure" harmless neuroblastoma tumors that would disappear of their own accord, while doing little or nothing to reduce the deaths associated with the disease's deadlier variant. By now the astute reader will be wondering if there might be a connection between neuroblastoma screening and the ongoing--and increasingly heated--debate over the merits of mammography in combating breast cancer. There is. As with neuroblastoma, there are several kinds of breast cancer, not all of which will prove deadly. And while mammography--in contrast with the neuroblastoma test--does detect some potentially deadly cancers and has thus saved some women's lives, it also picks up many times more cancers that would never have become symptomatic during the patient's lifetime or that could have been treated just as easily had they not been detected until the woman could feel the lump herself. Thus for every woman saved by early diagnosis, many others receive costly, painful, and potentially dangerous treatment to destroy tumors that pose little or no threat--tumors that they might die with, not of. As Barnett Kramer, director of medical applications of research at the National Institutes of Health, puts it, "Not everything that looks like cancer acts like cancer." That notion flies in the face of what practically every woman in America has come to believe about breast cancer. Most of us accept the premise that all breast tumors will eventually kill us if left untreated and that catching them early is our only hope. That belief is reinforced by doctors who feel compelled to treat the vast majority of breast tumors as if they were potentially deadly, because they lack the tools to determine which ones really are. And so the battle lines are drawn: On the one side is most of the American medical establishment--doctors, hospitals, major advocacy groups such as the American Cancer Society (ACS)--which pushes mammograms as a necessity for women over 40 and spends tens of millions of dollars per year on public health campaigns to this effect. On the other side is a smaller but increasingly vocal contingent of epidemiologists, doctors, and advocates who suggest that mammography screening has led thousands of women each year to receive risky, costly treatment they don't need, while doing little or nothing to cut the overall breast cancer death rate. What's the right answer? For individual women, there is none--it depends on the risk of breast cancer a woman is willing to tolerate and on the cost and risk she's willing to tolerate in order to reduce that risk. But for society, it's becoming clear that mammography is no longer the best focus of our financial and medical resources. Whether or not individual women continue to be screened, doctors, hospitals, scientists, and politicians need to turn their focus--and their money--toward the emerging technologies that could help doctors more accurately predict which tumors require aggressive treatment and which can safely be left alone. The United States needs to enter the post-mammography era--and the sooner the better. arly diagnosis is the fundamental premise behind all cancer screening. A good screening test picks up tumors before they have a chance to spread, allowing a surgeon to remove all vestiges of cancerous cells or allowing a radiation therapist to destroy them with X-rays. Ideally a screening test detects errant cells when they are still in a precancerous stage. That's what the Pap smear does. Invented in the 1940s by George Papanicolaou, a Greek-born American doctor, the Pap smear allows doctors to sample cells from the cervix and look for precancerous lesions. If they discover such lesions, doctors can then remove them, or sometimes the cervix or the uterus, before the tumor develops into full-blown cancer. By the early 1970s widespread use of the Pap smear had cut U.S. mortality from cervical cancer in half, the first demonstration that early cancer detection could save lives. The Pap smear, it turns out, was uncommonly effective. By contrast, the merits of mammography have never been so clear-cut. The early-diagnosis approach to breast cancer emerged in the first part of the twentieth century--when women often didn't come to the doctor until their tumors were the size of oranges--and was firmly entrenched by the time physicians began putting soft-tissue X-rays, or mammograms, to use in the 1960s. The ACS had already been pushing breast self-exams for two decades; the purpose of mammograms was to detect breast tumors still earlier, before they could be felt manually. By 1989 at least eleven national health organizations, including the American Medical Association and the National Cancer Institute (NCI), were recommending that women 40 years of age and older get regular mammograms. But for almost as long as mammography has been in widespread use, it has had its critics, particularly when it comes to younger women. The NCI has flip-flopped three times since it first recommended in the 1980s that all women over 40 get regular mammograms. Most recently, the NCI took up the question in 1997, convening an independent panel that concluded there was insufficient evidence either for or against mammography for women in their forties. Under pressure from Congress, however, the NCI reversed itself later that year, recommending that all women over 40 get screened. The current round of debate began last fall with the publication of a paper in The Lancet, a British scientific journal, arguing that the main studies purporting to show the benefits of mammography were so methodologically flawed that their results were meaningless. The issue boiled over in January, when an independent panel of scientific advisers to the NCI agreed with the paper's finding. It concluded that the evidence that mammography saves lives was insufficient to say one way or the other whether women of any age should get a yearly mammogram. The counterattack began almost immediately. In January ten medical groups, led by the ACS, took out full-page newspaper ads stating that "lives will be lost" if women are discouraged from getting mammograms. The following month Tommy Thompson, secretary of Health and Human Services, pronounced, "The federal government's recommendation remains clear: Women in their forties and older should be screened every one to two years with mammography." At issue is the meaning of seven large clinical breast cancer trials conducted in North America and Europe since the early '60s. These trials tracked the survival rates of women who received regular mammograms and compared them to those of women who had regular clinical breast exams but very few, if any, mammograms. Of the seven trials, five appeared to show an improvement in survival among the group receiving regular mammograms. The improvement was as much as 30 percent, at least for women 50 years of age and older. The Lancet article reanalyzed the data from the seven clinical trials and concluded that the five that showed a benefit suffered from methodological errors that invalidated their conclusions. In one study, for example, which began in the '60s, 153 out of 30,131 women who received mammograms died of breast cancer after 18 years of study, compared with 196 out of 30,131 women who did not receive screening and died of the disease. That's a 30 percent difference, but it's based on a difference of 43 women out of more than 60,000. For their part, mammography supporters counter that the authors of the Lancet article inflated the importance of tiny errors in the clinical trials and failed to acknowledge statistical adjustments intended to correct methodological flaws. But a compelling critique of the effectiveness of mammography can be made even without entering into the details of this arcane debate over the meaning of the clinical trials. That's because when one looks at the broader picture, mammography resembles the neuroblastoma screening conducted on Japanese children. Regardless of any benefit the clinical trials might (or might not) suggest, the overall death rate from breast cancer--again, the number of women per 100,000 who die of the disease--has hardly budged over the last three decades, even as more and more women have been getting mammograms. As Fran Visco, president of the National Breast Cancer Coalition, a patient advocate group, put it in testimony before Congress earlier this year, "Mammography should be accepted for what it is ... it may extend the lives of some women who have breast cancer, but it does not prevent or cure breast cancer, and it has many limitations." Today approximately 60 percent of women over age 50 are screened regularly, up from 13 percent in 1980. Yet the mortality rate from the disease has only recently begun to drop noticeably, and it's not at all clear that the drop is due to mammography. Age-adjusted mortality from breast cancer held steady at about 27 deaths per 100,000 women from 1973--when the U.S. government began keeping statistics--until 1992, when the numbers began to dip slightly. By 1998, the last year for which statistics are available, the death rate from breast cancer was down to 22.7 per 100,000--a significant reduction, but still modest when one considers the vast increase in the number of women who are now being screened. Some cancer experts, moreover, argue that this drop is mostly the resultof improved treatments--especially the hormone blocker tamoxifen, which can cut a woman's risk of cancer recurrence by 25 percent. (A study is now underway to determine whether mammography, or new treatments, or both, are responsible for the declining death rate.) arly detection can fail to reduce the cancer death rate for two reasons. Either the test is not sufficiently sensitive to pick up cancers early enough to make a difference, or it largely detects so-called "indolent" tumors--tumors that might never pose a danger to the patient, even if left undetected. Mammography suffers from both flaws. Earlier and earlier diagnosis has undoubtedly saved some women's lives, but it can do little to improve a woman's odds against the most deadly breast cancers--which are extremely fast-growing and invasive--no matter how early they are caught. At the same time, many of the tumors mammography detects, particularly in older women, might never have caused symptoms serious enough to merit treatment. "Mammograms are very good at finding indolent breast cancer but not at finding the aggressive sort," says Visco. The problem is that doctors often can't tell the difference between aggressive and indolent breast tumors, particularly when they are small. So as improved mammography picks up smaller and smaller lumps, doctors are treating more and more of them as if they were potentially deadly, though in fact most will turn out not to be. "If you take ten women with [early] tumors and treat them all with surgery, radiation, and chemo, one will die anyway," says Susan Love, a prominent breast cancer surgeon and frequent critic of mammography. "Two of them will be cured, and seven would have gotten better with local treatment alone. Out of those seven who would have survived without chemo or radiation, an unknown number will suffer significant, permanent side effects from the treatment." These lasting side effects can include organ and nerve damage, pain, sterility, depressed immunity, cognitive impairment, decreased bone density, and the risk of future leukemia--to say nothing of the psychological toll on women who are left to worry about a recurrence of breast cancer for the rest of their lives. o the individual woman facing the possibility that she is among the unlucky percentage with an aggressive tumor, the prospect of unnecessary chemo and radiation may seem a small price to pay for peace of mind. But as a matter of public health policy, the overtreatment caused by mammography is a real problem. The average size of tumors picked up by mammography has steadily decreased over the past 20 years. Now approximately 60 percent of breast cancers--about 120,000 cases per year--are "localized" (less than two centimeters in length and not yet spread to the lymph nodes), according to data from the NCI. Of those 120,000 cases, 70 percent, or about 84,000 women, would be cured by surgery alone. (An unknown number would have been fine--i.e., their tumors would have disappeared on their own or never become symptomatic--even without surgery.) Yet because doctors cannot tell which tumors require additional treatment, they give radiation to nearly all breast cancer patients with localized tumors. On top of the radiation, doctors gave additional chemo or hormone therapy to 62 percent of those patients in 1995, the most recent year for which statistics are available. The most troublesome area of overtreatment is a condition known as ductal carcinoma in situ, or DCIS. Thirty years ago these miniscule abnormalities, often no bigger than a pencil point, were diagnosed in only 6 percent of patients, usually during a biopsy or surgery for breast cancer. Today, thanks to mammography, DCIS represents nearly 20 percent of breast cancer diagnoses--and doctors have no idea what to do about it. They disagree over whether DCIS is a precancerous condition that bears watching but requires no treatment, or a small tumor that simply has not yet become invasive and needs to come out. "We know the majority of women with DCIS will not get cancer, but we don't know which ones. So of course our hand is forced," says Love. A DCIS diagnosis is now one of the most common reasons a woman undergoes a mastectomy, because the tiny lesions are often found scattered throughout the breast and cannot be removed individually. "We've gone from seeing women come in with fruit-sized tumors to detecting tumors that are no bigger than a needle. Yet there's this persistent sense that if you just make the mammograms better, or do digital mammograms, this wonderful, life-saving bounty will occur," says Barron Lerner, a physician and author of The Breast Cancer Wars. "We have reached the limits of early detection." Nor is the dilemma confined to breast cancer. The prostate-specific antigen test, or PSA, for example, picks up large numbers of harmless prostate tumors (see "Unkind Cuts," by Shannon Brownlee, tnr, March 22, 1999), and leads to thousands of unnecessary surgeries. As new screening technologies emerge, and more and more Americans are screened, says Kirsten Boyd Goldberg, editor of The Cancer Letter, a newsletter that covers the cancer industry, "This problem is going to be repeated over and over in nearly every single cancer." one of this is to suggest that women and doctors should abandon mammography entirely. But barring a breakthrough cure for breast cancer--the Holy Grail of cancer research, and about as likely in the near future--there are two ways the public health community can respond to this overtesting dilemma. One is to continue pushing women to get mammograms, while simultaneously conducting multimillion-dollar studies to hopefully, someday, prove that their benefits outweigh their costs. This, unfortunately, seems to be the path the American medical establishment is taking. The latest round of skeptical reports has not shaken the ACS's position. The Society expects to spend an estimated $20 million this year promoting mammography, particularly among minority and poor women. The NCI, for its part, is considering reanalyzing the existing clinical mammography trials one more time. The Institute has also not ruled out launching a new trial altogether, an effort that would cost several million dollars and would take close to a decade to complete. And conducting it would not be easy, since researchers estimate they would need to recruit nearly half a million women willing to be told by the flip of a coin whether or not to have regular mammograms. Indeed, if we continue on the current path, it's likely that ten years from now we will be in very much the same position we are in now--and that we were in ten years ago. There will be studies that purport to show that mammography saves women's lives and studies that purport to show that it doesn't. More than 100,000 American women per year will have aggressive treatment for tumors discovered through mammography--many unnecessarily. It will, in short, be a colossal waste of money, time, and women's lives. "We are not going to get anywhere if we keep arguing about mammograms and devoting millions of dollars to new studies," says Visco. "The end goal should be saving lives, not saving mammography." There is an alternative. Public health organizations could acknowledge that early detection through mammography has proven of limited use in the war against breast cancer and could devote their energies--and research dollars--to developing better diagnostic tools. For nearly a century now, doctors have used the size of a tumor, whether it has begun to invade surrounding tissue, and the presence of cancer cells in the lymph nodes to diagnose breast cancer. But these crude surgeon's measures do not take into account our emerging understanding of cancer genetics. Over the last two decades basic research has shown that every tumor contains its own unique set of genetic mutations. Some mutations allow cells simply to reproduce but not invade surrounding tissue. Others give cancer cells the power to break free from the mother tumor and travel through the bloodstream to distant locations in the body. Researchers can now identify thousands of the mangled bits of DNA that drive cancer cells. And over the last two years they have begun to figure out which mutations can serve as accurate markers for a patient's prognosis. These efforts have been made possible by a new tool known as "microarrays," or gene chips: tiny silicon wafers that detect genetic mutations and deliver that information to a computer. Clinical trials are now underway to sort out which of several thousand mutations are most likely to indicate how aggressive or indolent a tumor is and whether or not it has already metastasized. Using a breast tumor's genetic fingerprint would represent a paradigm shift in how doctors think about breast cancer--and transform the way they treat it. A new technology called ductal lavage is enabling researchers to flush cells from the lining of the milk ducts, where all breast cancers begin, in order to look for the earliest stages of cancer and to monitor DCIS. In as little as five years, microarrays could allow doctors to determine which cancers require aggressive treatment--and which don't. "It will be two or three years before we have confidence that microarrays are really as powerful a tool as early studies suggest," says Carlos Caldas, a researcher at Cambridge. But the effect of using them to help doctors stop overtreating breast cancer--and perhaps even to leave some breast cancers alone--he has said, is "potentially staggering." t's a potentially staggering development you've probably never heard of, however, in part because policymakers, the medical establishment, and patients still consider cancer an enemy that can only be fought by catching it early and attacking it with everything we've got. And over the past decades a vast network of entrenched interests has arisen to serve that intellectual framework. Last year alone, the NCI poured $36.7 million into mammography research. The radiologists who interpret the results of mammograms practice one of the most highly paid specialties. The manufacture of mammography equipment is a $300 million annual industry; cancer drugs are worth another several billion dollars more per year. This enterprise will have to be reconfigured in order for doctors to stop overtreating breast cancer. In Japan, doctors and health officials have long known that the death rate from neuroblastoma was not going down, even as the number of infants screened was going up. Yet they have remained reluctant to abandon the compelling logic of early diagnosis. Earlier this month the New England Journal of Medicine published the results of two large-scale, multimillion-dollar studies that should finally put to rest the debate over neuroblastoma screening. "We are really hoping this will be the nail in the coffin for neuroblastoma screening in Japan, and they will stop," says William Woods, leader of one of the clinical trials and director of the aflac Cancer Center in Atlanta, Georgia. His study, which was conducted in Canada and the United States, required 13 years to complete, the participation of half a million Canadian infants, a team of 31 researchers, and well over $6 million. What will it take for American policymakers to recognize mammography's limitations and choose a wiser course instead? New Republic Magazine, Issue date: 04.22.02 Shannon Brownlee is a Markle Senior Fellow at the New America Foundation. LINK to magazine of "centrist liberal opinion" 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.