Once a tumor has been found, whether by mammography or palpation, it must be treated. The basis for the contemporary approach dates back to the 1890s, when William Halsted, a professor of surgery at Johns Hopkins University, developed radical mastectomy—the technique for removing a breast, the underlying chest muscles, and the lymph nodes in the armpit. A giant in medical history, Halsted was held in such respect that surgeons after him took an uncommonly long time to admit that his ideas about cancer were off base. He believed that breast cancer oozed slowly through the lymphatic system of the body the way foul water from a brackish pond travels through ditches to other parts of the landscape. (The lymphatic system is the collection of ducts, lymph nodes, and other organs that drain the tissue fluid called lymph into the bloodstream.) To Halsted's way of thinking, breast cancer could be cured by carving a wide swath around the initial tumor and its draining sites, leaving clear margins of healthy tissue—a scorched-earth approach.
Seventy-five years later a number of North American researchers began questioning Halsted's ideas. Instead of seeping slowly throughout the body from its initial site, breast cancer seemed to hop from place to place as cells sloughed off by the main tumor mass floated through the bloodstream. If this was true, removing a woman's entire breast would often be pointless, because most of the tissue surrounding the tumor would be healthy. After this heretical notion provoked much controversy among cancer specialists, Bernard Fisher, of the University of Pittsburgh, put together a series of large clinical trials that were intended to resolve the dispute. These trials demonstrated that a much less drastic operation called a lumpectomy was appropriate in many cases; when combined with radiation therapy it provided a chance of survival indistinguishable from that provided by Halsted-style mastectomy. After the results were published, in the mid-1980s, surgical practices changed. Remarkably, though, they didn't change very fast. Even today only a third of the women who undergo breast surgery have lumpectomies, whereas at least two thirds probably could, in light of the size of their tumors. The rest have modified radical mastectomies, a somewhat less severe version of Halsted's operation.
In the same period researchers questioned another tacit assumption about breast cancer: that lumps in the breast appear suddenly and must be treated quickly. Instead we have come to think that most breast tumors take years to develop to detectable size; some need decades. In its initial stages the tumor would be visible under a microscope—if one could somehow scan the entire breast with such precision. Only after months or years, however, can the malignancy be detected on a mammogram or by palpation. Yet standard practice still involves a rush to operate after diagnosis. The patient wants to know the details of her condition right away, and surgical exploration is the only conclusive means of providing this information. But surgeons also operate partly in the name of "getting this thing out before it spreads." I cannot count the number of times I have seen a surgeon solemnly tell anxious family members in the hospital waiting room, "Well, we got it all." Although surgeons today target their work more carefully and humanely than they did in the past, they still ignore a crucial fact: we cannot know whether we "got it all." Although our treatments can eliminate the primary tumor (not a small consideration, given its potential danger), there is little evidence that we actually eradicate breast cancer in any more patients than we did decades ago.
As I said, switching when appropriate from mastectomy to lumpectomy leaves survival rates unaffected—the less-severe operation simply does the same job. And we have, in my view, less than compelling evidence that routine mammography benefits otherwise healthy women. Any improvement in our treatment of breast cancer must therefore come from radiation or chemotherapy, both of which are by now standard practice. After a lumpectomy, according to a meta-analysis published last November in the New England Journal of Medicine, radiation decreases the risk of recurrence in the breast region (local-regional recurrence, as it is known) to that associated with full mastectomy—five to 10 percent during the next ten years. Without radiation the risk is about 20 percent. But radiation has no known effect on distant metastases from breast cancer (nonlocal recurrence), which are the chief sources of danger. As a result, post-surgical radiation has no demonstrable effect on survival—a point seldom made clear to patients. Reducing local recurrence is a reasonable goal, because any tumor in the breast could be lethal. But it will not diminish the danger from metastases.
Nor should we think that the modest benefit from radiation has no price. Radiation therapy is rough on the body. It kills cells, inducing inflammation in the breast. Responding to the inflammation, cells called fibroblasts proliferate and lay down scar tissue. Fifteen to twenty-five years after radiation therapy tissues in the irradiated area often feel thick, hard, even wooden—a strange, unpleasant sensation. Meanwhile, the blast of radiation may have created genetic damage that will lead to other types of cancer. For instance, worrisome evidence suggests that radiation may promote lung cancer in the irradiated side of the body. The risk seems slight, but its mere possibility should remind physicians that radiation therapy is not without its price.
Chemotherapy is a more complex issue, with better-established benefits and fewer side effects than it had in the past. Chemotherapeutic compounds kill cancer cells; the hope is that they will destroy any metastases lurking in the body. Thirty years ago chemotherapy was a treatment of last resort, administered only to patients with extensive metastatic disease. I can recall flying overnight to the University of Wisconsin in 1959 to pick up some precious vials of 5-fluorouracil, one of the first, then-experimental chemotherapeutic treatments, for a woman whose cancer was overwhelming her liver. I was dismayed when the drug made her extremely sick, though I thought that it might have postponed her death a little. In the early 1970s researchers discovered that chemotherapy was also effective when several agents were administered right after surgery—as an adjuvant treatment, in the jargon. Today chemotherapy is usually given in the form of a pharmaceutical cocktail, one of the most common being cyclophosphamide (sold principally under the trade name Cytoxan), doxorubicin (Adriamycin), and the original 5-fluorouracil (sold under several names, but usually called 5-FU). Although we have discovered how to moderate the side effects of chemotherapy, it is still heavy going. Common complaints include fatigue, nausea, hair loss, mouth sores, diarrhea, and premature menopause.
Chemotherapy is more effective in younger, pre-menopausal women than in older, postmenopausal women. Many researchers believe that premenopausal women receive this extra benefit because the drugs chemically destroy the function of their ovaries, frequently stopping their monthly estrogen-progesterone cycles. The decline in hormone production affects the two out of three tumors that are estrogen- or progesterone-receptor-positive. Deprived of the chemical catalyst they need to grow, the metastases often shrink, lengthening the time until they become dangerous.
Before chemotherapy, doctors sought to create a similar effect by directly removing patients' ovaries. If the cancer recurred, they tried to amplify the effect by removing women's adrenal glands. These small, triangular organs sit atop the kidneys and secrete hormones, one of which is DHEA, a still-mysterious substance that is a favorite preoccupation of health faddists. Most women convert some of the DHEA into estrogen, thus producing small amounts of estrogen after menopause. Even cutting out the adrenal glands—a procedure involving major abdominal surgery—wasn't the final step. In their zeal to combat estrogen, doctors also injected women with testosterone or prescribed synthetic male hormones.
Although these treatments sometimes slowed the progress of breast cancer, their cumulative impact was horrific and even barbarous. After having the breast and underlying tissue on one side of her chest cut away and covered with a thin skin graft, a woman could be subjected to intense radiation tharapy, primitive forms of chemotherapy, the removal of her ovaries, and, eventually, the removal of her adrenal glands. If this failed, she might be injected with male hormones that made her skin oily, pimply, and hairy.
Things are better today. Surgery is often less severe; chemotherapy can be more tolerable; and, perhaps most important, there is a substitute for removing the adrenal glands and injecting women with testosterone. In the 1970s researchers discovered that the actions of estrogen and other hormones can be blocked chemically by drugs called hormone antagonists, of which the most noteworthy is tamoxifen, which is sold under the trade name Nolvadex. Circulating in the body, tamoxifen molecules prevent breast cells from linking up to estrogen or progesterone molecules by attaching themselves to the cells in their place, blocking the hormones completely. Because hormones can stimulate the growth of the majority of breast tumors that are estrogen- or progesterone-receptor-positive, the potential of tamoxifen for treating this type of cancer is clear. (Tumors that do not respond to hormones grow faster and are less treatable.) Tamoxifen can cause irregular periods or aggravate hot flashes; it also apparently increases the risk of uterine cancer. But it is a vast improvement over removing women's adrenal glands.
Many studies have shown that the unwanted side effects of chemotherapy and antihormonal therapy are offset by the benefits. A compelling example is the Early Breast Cancer Trialists' Collaborative Group, an Oxford-based meta-analysis of 133 randomized clinical trials around the world which examined the effects of chemotherapy and antihormonal tamoxifen therapy on 74,652 women with breast cancer. In the trials breast-cancer patients were usually given mastectomies or lumpectomies and then divided randomly into two groups: one that received some type of chemotherapy or antihormonal therapy (the test group) and one that did not (the control group). The Oxford group reported in 1992 that women who underwent either treatment after surgery had a 12 to 14 percent greater chance of surviving for ten years than those who did not.
Note, though, exactly how this good news was phrased. The improved survival prospect for women who underwent these treatments was in comparison with the prospect for breast-cancer patients who received no systemic therapy—a relative, rather than absolute, increase in survival. Without chemotherapy, a woman with breast cancer faces a 55 percent chance of dying within ten years, according to the Oxford team. Chemotherapy cuts the figure to 48.7 percent, an absolute difference of 6.3 percent. Similar figures appeared in the tamoxifen trials. After surgery 41.2 percent of the women receiving tamoxifen died within ten years, whereas 47.4 percent of the control group died—an absolute difference of 6.2 percent. The treatments, then, buy some time for some women, and this is a very good thing; any woman told that she has even a small chance of surviving longer with adjuvant treatment will likely choose it. Nevertheless, we should not be extolling these therapies as major breakthroughs.
"Wait a minute," some readers may be saying. "Are you being too quick to dismiss these results? Don't these decreases in risk mean that we are eliminating this disease in some people? It may be a small number, but aren't we curing a few more than we did in the past?" These questions are difficult to answer, not least because they involve coming up with a good definition of "cure." For instance, if a woman of sixty-five is treated for breast cancer and then dies five years later of a heart attack, was she "cured" of cancer? Her family, friends, and even her doctor might think so, because the disease never troubled her again. Indeed, a common clinical definition of "cure" is survival for five or ten years. Many researchers would be less quick to claim a cure, because the cancer might have been on its way to recurring when the heart attack intervened.
Acknowledging these complexities, statisticians evolved a precise definition for "cure" in the 1950s. A true cure of a lethal disease like breast cancer is achieved only when people with that disease face the same chance of death as others in the population of the same age and sex. Such true cures are possible with many diseases. Most pneumonia sufferers, for instance, after recovering with the aid of antibiotics have exactly the same prospects for survival as people who have never had pneumonia. Like pneumonia, some cancers, including cervical cancer, childhood leukemia, superficial melanomas, and Hodgkin's disease, can be truly cured.
Breast cancer, unfortunately, is not among this select group. As far as we know, a woman found to have invasive breast cancer is always at higher risk of dying prematurely than women without breast cancer. Even thirty years after her diagnosis she is up to sixteen times as likely to die of the disease as a woman in the general population. That is why responsible researchers in this field avoid the word "cure." Even as they report advances, they must acknowledge the reality: Postsurgical chemotherapy and antihormonal therapy do buy time—an important advance. The slowed progress of the disease can give a woman additional years of life and even allow her to die of other, less traumatic, causes. But breast cancer is every bit as incurable as it was in Halsted's day.