The Mammogram War
A decision not to recommend breast X-rays for younger women has sparked controversy, fear and a new look at this imperfect medical tool.
CATE BELLANCA AND BARBARA BRENNER ARE both in their 40s, and both have had breast scares. But that's where the resemblance ends. Bellanca had a mammogram in 1991. The X-ray showed some suspicious spots. A biopsy found a malignancy. After a mastectomy, Bellanca had no doubt that, as she put it, ""if I hadn't had that mammogram, I could be dead now.'' Brenner, executive director of the advocacy group Breast Cancer Action, got a mammogram when she was 41; her breasts looked healthy. But a few months later she found a lump. It was a tumor; a lumpectomy followed by radiation and chemotherapy got rid of it. Brenner had mammograms every six months afterward; all looked normal. Then last October she noticed an odd change in her breast. It, too, turned out to be cancer. She had a mastectomy. Brenner's view of mammograms? ""They give you a false sense of security,'' which, if it lulls women into slacking off on self-exams, can kill.
The two women also have very different views of the decision by an expert panel of the National Institutes of Health (NIH) last month not to recommend regular mammograms for women in their 40s. After hearing 32 expert presentations, the 13-member NIH panel deliberated until 2:30 one morning. Its verdict: ""breast cancer mortality is no lower in women who [received mammograms in their 40s] than in controls.'' ""I felt it was irresponsible,'' says Bellanca, a photo editor in New York. ""The 40s are critical.'' But Brenner says, ""Mammograms don't work for a lot of younger women.''
At breast clinics and cancer support groups and around office water coolers, women are voicing approval, concern and, most of all, confusion about what's come to be known as ""that mammogram decision.'' The panel's verdict was astonishing enough to women, who have had drummed into their heads that early detection of breast tumors saves lives. But the aftermath is even more perplexing. Dr. Richard Klausner, director of NIH's National Cancer Institute and the man who convened the panel, trampled its conclusion. Pronouncing himself ""shocked,'' Klausner asked NCI's outside advisers to rule on the question, hoping, like a canny lawyer shopping for a sympathetic jury, that they'd reach a different verdict. One panel member, Jeanne Petrek of Memorial Sloan-Kettering Cancer Center in New York, resigned, writing in a letter that the panel's draft was ""unacceptable to me and I cannot have my name associated with it.'' And many of the panel members, asked to give their personal rather than official judgment, said that they would recommend regular mammograms for fortysomethings, giving the appearance that experts have one kind of advice for the public but another for their patients--and their daughters, mothers or wives. How angry are some women? Statistician Donald Berry of Duke University, and a member of the panel, received ""hateful voice-mail messages suggesting I was trying to kill women,'' he said. The Senate, drawing on what an op-ed piece last week in The Washington Post called ""some mysteriously acquired [medical] insight,'' voted 98-0 for a nonbinding resolution praising the value of mammograms for women in their 40s. But Dr. Susan Love, author of the definitive book on breast cancer, ""Dr. Susan Love's Breast Book,'' lauded the panel's recommendations (story, page 60). ""So it makes doctors feel somewhat impotent to say, "We don't know','' she says. ""But it's about time we started telling women the truth.''
No one denies that mammograms, which can detect a lump only one eighth of an inch across, can find tumors sooner than manual exams, by the woman or her doctor, who can't feel anything much smaller than half an inch across. Depending on how rapidly a tumor grows, mammography can detect it as much as two years before a manual exam. For that reason, women 50 and older who have annual mammograms have a mortality rate 30 percent lower than women of this age who do not. But in the last few years researchers have begun seeing that early detection doesn't help younger women as much. Says Fran Visco of the National Breast Cancer Coalition, ""Women have been led to believe there's this tool out there that's going to save our lives. But it's just not true.''
Numbers tell the story. When researchers pool results from the eight major studies that have been conducted on mammograms worldwide, with about 180,000 women of all ages, they find that women who have a mammogram every year or two between 40 and 49 have a decreased mortality from breast cancer of about 15 percent. And even that small reduction ""did not kick in until 10 to 12 years out,'' says ob-gyn Dr. Ruthann Zern, a member of the NCI panel. Thus it is possible that it was mammograms in the women's 50s, not screenings when they were 40 to 49, that really helped, says Dr. Suzanne Fletcher of Harvard University. In a 1995 reanalysis of the studies, Karla Kerlikowske of the University of California, San Francisco, and colleagues concluded that mammography had no effect on whether a fortysomething woman was alive or dead seven to nine years after her cancer was detected.
Why are the benefits of mammograms for younger women not greater? One possibility, suggests Dr. David Dershaw of Sloan-Kettering, is that the studies are flawed. Some go back to the 1960s, when mammography was much cruder. ""The most recent studies,'' says Dershaw, ""do show that mammograms for women in their 40s can decrease the risk of dying of breast cancer substantially.''
Another reason mammograms aren't saving more women 40 to 49 is that breast cancer in these women is rare: 16 cases per 1,000 women. Since relatively few die of the disease in the first place, it would take an awfully powerful detection technique, not to mention effective treatment, to make much of a dent in the death rate. And mammography in younger women just isn't that accurate, partly because it does not see well through the dense breast tissue of premenopausal women. According to the NCI panel, mammograms miss as many as 25 percent of invasive cancers in 40- to 49-year-old women, compared with 10 percent in older women. Also, some tumors detected ""early'' are not necessarily curable. As a result, explains Russell Harris of the University of North Carolina, even ""if you screen 1,000 women in their 40s for 10 years, you prolong one or two lives.''
New imaging techniques might do a better job of detecting tumors in dense breasts (page 59). But there may be fundamental biological reasons that mammograms will never extend the lives of more women who start annual screenings at 40. These have to do with the different kinds of breast tumors (chart, page 57):
Slow-growing. ""Indolent'' tumors can take years to reach a palpable size. They are thus more likely to be detected by X-ray than by manual exam: a woman's next mammogram appointment rolls around before the tumor is large enough to feel. But indolent tumors are less dangerous than aggressive tumors. As Dr. Barnett Kramer of NCI says, ""With or without mammography, those are usually curable,'' typically by lumpectomy. So women skipping regular mammograms in favor of monthly self-exams would have just as high a survival rate as wom- en having the X-rays. UNC's Harris estimates that indolent tumors make up half of all breast tumors.
Ductal carcinoma in situ. Confined to the ducts of the breast, this kind of tumor accounts for about 10 percent in all women, and from 15 to 60 percent in women in their 40s. Nancy Reagan had DCIS. ""It may stay there a woman's whole life and never invade surrounding tissue,'' says Dr. Michael Cohen of Sloan-Kettering, ""but we don't know how to tell the one that won't spread from one that will.'' Women with DCIS are almost always operated on, but because a DCIS may never kill, the advantage to catching it with mammography, rather than waiting and catching it by manual exam, may be small.
Aggressive. This kind of tumor is the most likely to be fatal. Even by the time a mammogram, let alone palpation, finds it, a few cells will have likely broken off to seed another organ with cancer. In other words, says UNC's Harris, ""early detection is still too late.'' A 1996 study by UCSF's Kerlikowske found that women in their 40s had much faster-growing tumors than older women, possibly because cancer-causing genes kick in at a younger age or because higher levels of estrogen feed these tumors. To muddy the waters even more, the risk of aggressive tumors suggests that women in their 40s should have mammograms as frequently as every six months: done only every year or two, as in the worldwide studies, they offer little benefit. Somewhere between fatally aggressive and very indolent cancers are the kinds where early detection makes the difference between life and death. ""These are the women whose lives we can extend with mammograms,'' says Harris.
Because of these and other problems with mammograms, the NIH panel advised women in their 40s to make up their own minds about regular screenings. And while custom-tailored medical advice based on a person's unique genetic makeup and history is replacing one-size-fits-all advice--avoid cholesterol, cut out salt, take hormones after menopause--mammograms may not be the best candidate for a ""decide for yourself'' approach. ""That is punting in the most preposterous way I can imagine,'' says Dr. Miguel Sanchez of Englewood Hospital in New Jersey. ""Here you have an expert panel that couldn't decide conclusively, and they're asking women, who have even less information, to decide? Come on.''
How, then, can women make the decision? One factor to weigh is what harm mammograms can do. The radiation dose is so low that it poses virtually no risk of triggering a cancer in healthy cells. ""Virtually'' because there is a small percentage of women who carry a genetic mutation that makes them supersensitive to radiation. For these women, says Sloan-Kettering's Cohen, ""we would be very cautious.'' The greater risk is of false positives--that is, detecting something that is not cancer. In fortysomething women, calculates Harris, 95 percent of ""abnormal'' mammograms do not turn out to be malignancies. Women 40 to 49 with a family history of breast cancer have fewer false alarms than other women, reports the NCI panel; Hispanics have more.
Another crucial consideration is whether a woman and her doctor understand the limitations of mammograms. Even having an annual screening doesn't mean a woman can skip monthly self-examination of her breasts. That's especially true for women under 50, says Englewood's Sanchez, ""because in the dense breast tissue of premenopausal women a tumor may not show up on an X-ray, but can still be detected manually.'' Manual exams pick up about three quarters of tumors in fortysomething women, says Julia Scott of the National Black Women's Health Project and a member of the NCI panel; in older women it's about half. A clean mammogram does not mean a woman is cancer-free.
Cost is another consideration for many women. Advocates for women's health worry that, emboldened by the NCI panel's decision, insurers will balk at paying for biannual, let alone annual, mammograms for women in their 40s. The panel did recommend that, if a fortysomething woman chooses annual mammograms, insurance should pay for it. But Medicare, which covers a population for whom annual mammograms have been proven to extend lives, covers no more than one screening every two years.
This year breast cancer will kill 5,700 women between 40 and 49. It is bad enough that early detection is not saving more of these women. But as the debate over early mammograms rages, there is a danger of losing sight of the real tragedy: the ability to cure aggressive breast cancer is hardly greater than it was in the 1930s. As Fran Visco says, ""Let's put the outrage where it belongs: we still don't know how to prevent, treat or cure this disease.''
Mammograms can detect tumors that are one eighth of an inch in diameter, while manual examinations can't detect tumors smaller than a half-inch around. Since tumors vary greatly in their aggressiveness, early mammographic detection may not always be helpful.
Aggressive tumor: So fast-growing that by the time it can be detected with mammography, it is already likely to have seeded another organ with cancer.
Ductal carcinoma in situ: Found in the ducts of the breast, it is an earlier and usually a more treatable form of cancer. Between 15 and 60 percent of the tumors detected by mammograms of women in their 40s are of this type. Treatment is lumpectomy, followed by radiation.
Indolent tumor: This kind is very slow-growing and may take years to develop. It is less dangerous than the faster-growing aggressive tumor. This means that even if the tumor is not detected until it can be felt by a manual exam, the patient still has an excellent chance of surviving. A lumpectomy usally cures this kid of breast cancer, which may make up more than half of all breat tumors.
Women with positive mammograms requring follow-up*
Of women 40-49 years old, 2.5% actually had cancer
Of women 50-70 years old, 6% actually had cancer
Funding for medical research has grown dramatically, and money for breast-cancer research leads the field.
Researchers have identified numerous factors that can increase your chances of getting breast cancer. Some can be controlled.
Hormonal: Early menstruation, late menopause advanced age at the birth of your first child, not bearing children at all.
Genetic: Genes that can mutate to increase risk, including BRCA1, BRCA2 and P53.
Lifestyle: Lack of exercise and high-fat diet.