Return to Part 1
Popular Concepts Revisited
Today popular conceptions of breast cancer and its management are becoming outmoded as research exposes their lack of merit. Unfortunately, instead of this leading to the modification of common medical procedures, or to the understanding that these protocols are becoming obsolete, many persevere unchanged. Some of the tenets accepted as gospel need revisiting--not only the mastectomy, but also the classification of ductal carcinoma in situ as cancer, the misconception surrounding "early detection," axillary lymph node dissection, radiation following surgery, and intensive postsurgical follow-up (chest x-ray and bone scans), as well as one of the newer therapies--high-dose chemotherapy with bone marrow or stem-cell transplant.
There is not unanimous agreement about what constitutes breast cancer. Breast anatomy may be helpful in understanding how the disease is defined. Breasts house a series of milk-producing glands that empty into smaller and larger tubes called ductules and ducts. Put simply, breast cancer means uncontrolled growth of cells--tissue that, when removed and analyzed by a pathologist, shows that malignant cells have overrun the anatomical boundary of the duct and extend into the surrounding tissue.
The most common form of breast cancer, referred to as "infiltrating ductal carcinoma," comprises 70% to 80% of invasive tumors that arise within the mammary ducts and invade the surrounding fatty tissue (called the stroma). The other 20% to 30% are subtypes (invasive lobular, medullary, mutinous, tubular, adenocystic, papillary, carcinosarcoma, inflammatory). This scheme of classification is based on locale and behavior. Americans might be differentiated geographically--as New Yorkers, Southerners, and Californians--but there are plenty of variations within each regional type.
Breast cancer is not a single disease, but an umbrella term for a plethora of diseases. It is no more homogeneous than infectious diseases--mumps and malaria have as little in common as herpes and cholera. Similarly, breast cancers differ strikingly from one another. Within each tumor itself there is enormous heterogeneity. Tumors are as diverse biologically as Manhattan is socially. A tumor is not composed of a single type of cell. It is like a vegetable basket that contains bits of lettuce, carrots, beets, broccoli, and zucchini indiscriminately fused together. Each vegetable has a unique shape, texture, growing pattern, and chemical composition. A tumor is a biological entity unto itself--like the city of Manhattan--yet the inhabitants of the city neither look alike nor behave, eat, or recreate uniformly. Some tumor cells metastasize early; some never do; others do so slowly. Some are accelerated by estrogen; others are not. Some encourage blood vessel growth; some do not. These processes are determined by the genetic material within the myriad cells that comprise the tumor. Even though it may be reasonable to say that two women have breast cancer, when the disease in the two women is compared, there might be so many differences that one begins to question whether they truly do have the same disease. Rate of growth and infiltration may take 3 years in one woman and 40 years in another.
Ductal Carcinoma in Situ Classification
Some abnormalities look like cancer under the microscope but do not act like it, and therefore are not truly breast cancer. One of these discrepancies between anatomy and behavior is "ductal carcinoma in situ" (DCIS), which consists of an abnormal proliferation of ductal cells that do not invade the basement membrane of that duct (hence the term "in situ," meaning "confined to the site"). Because DCIS does not extend beyond the borders of the duct, it is noninvasive, and does not therefore constitute a true malignancy. In 1934 Halsted's former resident, Joseph Bloodgood,32 described DCIS as precancerous tissue--a depiction that still applies.
Richard Margolese, surgical oncologist at McGill, says, "The management of ductal carcinoma in situ ... is controversial. It is not clear whether all carcinomas are preceded by DCIS or if all DCIS leads inexorably to carcinoma.... A better understanding of the biology of DCIS would lead to better clinical management."33 Because of the confusion surrounding the sequelae to DCIS, many current breast cancer studies include women with DCIS, because they regard it as a malignancy; thus the outcomes of these studies are skewed. According to William Silen of Harvard, "Twenty-five or thirty years ago, it was taught that there was no such thing as noninvasive carcinoma of the breast. In pathology, I was taught that if you looked far enough, you'd always find invasion. I'm absolutely convinced that a lot of the so-called cures achieved with radical mastectomy were patients who actually had noninvasive carcinomas of the breast."33(p358) Unfortunately, the language does not change appropriately every time the explanatory model shifts. Understandably, this causes confusion. The term "carcinoma" is used both to define malignancy and to describe tumors that are not malignant by virtue of the fact that they neither invade nor spread.
Public misconception abounds concerning the concept of "early detection." It is perhaps the most mystifying oxymoron within the vocabulary of the breast cancer paradigm. What is early? Ideally, it is before the local malignancy has spread, or metastasized. Although it's a difficult notion to accept, there is no way of knowing whether malignant cells have spread by the time of detection. Frequently the term "early" is confused with the term "small." Generally a small lump is preferable to a larger one--but this is not always the case. The significant determinant is biological: whether the cancer has infiltrated beyond local boundaries, how fast it is growing, and where it is growing.
It is known that it takes an average of 10 years for a tumor of the breast to grow to 1 cm (a little less than half an inch) in diameter. It is hypothesized by Judah Folkman34 of Harvard that as the number of blood vessels supplying the tumor increases, so does the likelihood of metastatic disease. It is not known precisely how long it takes for tumors to acquire an adequate blood supply. It takes approximately 5 years from the time a cell becomes malignant (de, shows evidence of uncontrolled growth) to the time that it develops enough vasculature for tumor cells to enter the bloodstream. One cubic centimeter of breast cancer tissue contains roughly one billion cells. Based on the doubling rate of cells, it takes 30 replications for one cell to become one billion. If the time of replication is 120 days, tnen there are 3 replications per annum, so over a 10-year period there are 30 replications. It is thought that in the first 5 years (half of the hypothesized 10-year period), the mass is not sufficiently vascularized (does not have an adequate blood supply) to be able to metastasize. But a palpable mass--or one visualized on mammography that is 1 cm in diameter or more--may have been growing for 10 years. l By this stage it has likely become bloodborne and widely disseminated. Local treatment--mastectomy, lumpectomy, or radiation--will not have any impact on survival if malignant cells have been seeded elsewhere (note 11).
Breast surgeon Susan Love,35 testifying before the Senate in 1991, stated that
[w]e have spent a lot of time, energy, and money touting early detection and preserving it as if it were the answer. Unfortunately, we have misrepresented the situation through wishful thinking or just an attempt at simplification. We have acted as if all tumors go through progression from one centimeter to two centimeter[sl and on and on as if all tumors have the potential to be detected at a small size and therefore could be cured. Would that were true. What we are dealing with is a combination of a tumor and an immune system. Some tumors are very aggressive and will have spread before they are palpable. Thirty percent of [the women with] nonpalpable tumors are found to have positive Iymph nodes. Some tumors are very slow growing andwill not have spread even though they have reached a large size (note 12).
The value of"early detection" is complicated by a factor called "lead--time bias." Namely, women appear to live longer when the disease has been identified earlier, but mortality has not necessarily been affected. There is a widespread collective misunderstanding that if only the lump is found "early," the problem can be either aborted or "fixed." This has led to false guilt on the part of women who feel that through their negligence they are responsible for their misfortune, false blame toward doctors even though they could not have discovered the lump sooner (and even if they had, it would not have mattered), and anguish at a cost of millions of dollars in litigation without sound medical foundation.
Not only is the notion of"early" muddled, but the question of what is being detected is also difficult to grasp.36 Nearly one third of the women with tumors undetected by mammogram have positive Iymph nodes--a sign that the disease is already systemic.37 Mammography fails to detect one fifth of all cancers; in women under 50, it misses as much as 40%.38 Unfortunately, having a clear mammogram does not mean that a woman is cancer-free. But because many cases are visualized by mammography, the current recommendation is that, especially for women over 50, it is a useful tool--particularly when a qualified technician uses a reliable mammography machine with a skilled radiologist interpreting the results. As counterintuitive as it sounds, radiologists Samuel Hellman and Jay Harris39 assert that
"[d]etection of cancer at an earlier stage does not necessarily imply an improved cure rate."
Axillary Node Dissection
Axillary node dissection is another procedure that is no longer routinely justified, yet remains firmly entrenched. Halsted was wrong: cancer does not spread in an orderly fashion via the Iymph system, node by node. Whether nodes are positive or negative does not necessarily foretell whether an individual woman will have a survival advantage. An early hypothesis posited that the presence of malignancy in the Iymph nodes served as a marker for who should receive chemotherapy. But new studies have shown that it is not an accurate prognostic measure. In 1986, Hellman and Harris39 reported the following: "Twenty-five percent of patients without axillary Iymph-node involvement develop metastases while 25 percent of those with axillary Iymph-node metastases never develop distant metastases." Thirty-eight percent of women with negative Iymph nodes die of the disease, which demonstrates that the positive or negative status of these nodes does not provide reliable prognostic information.
Harvard surgeon Blake Cady urges that "[w]e need to move beyond the latest dogma and convention regarding routine axillary dissection for established functionally equivalent goals" (note 13). In a book called Important Advances in Oncology 1996, Cady writes a chapter titled "Is Axillary Node Dissection Necessary in Routine Management of Breast Cancer? No." Surgeon Peter Deckers suggests that "[w]ithin the next decade, axillary dissection will be extinct."33(p363) Again, it is the cellular biology that is most crucial in determining prognosis and treatment, and this is now the focus of current research. But there is a lag time between the incorporation of new information and the dispatch of old habits.
Fisher's Protocol B-04 study established that axillary node dissection does not provide survival benefit. When further treatment was dependent on whether the nodes show malignancy, then node dissection was perceived to be a useful procedure. Today, however, we have many biological markers that provide information equivalent to positive or negative node status, rendering this procedure obsolete. If these markers suggest that a tumor is aggressive, women will receive chemotherapy regardless of the status of their axillary nodes. The medical school dictum applies: "If the results of a test do not change what you do, do not do the test." So why does it continue as routine procedure? Again, one suspects a lag between habit and the adoption of the newer logical thinking. When queried, many oncologists say, "I just feel more comfortable knowing about the nodes." But unless there is good justification for axillary node dissection, it should be questioned because it does harm.