In the late 1940s, after attending Yale (Halsted's alma mater) as an undergraduate and completing medical school, my father, Malcolm Beinfield, did a surgical residency at Harlem Hospital. Harlem housed several prodigious masters of surgery at the time. However, unlike their counterparts at the Mayo
Clinic, Memorial Sloan-Kettering, or Presbyterian hospitals--all
of whom were part of the grand establishment of medicine, replete with highly endowed funding for the best and most advanced research--Harlem depended on old-fashioned empirical observation and pragmatic experience. The surgeons at Harlem questioned the logic of mastectomy for their patients with breast cancer. It was not until 1948 that Harlem's Louis Wright became the first black surgeon admitted to the American College of Surgeons. Perhaps the forming of independent clinical judgments was facilitated by his status as an outsider.
Joining the clinical faculty at Yale in the 1950s, my father witnessed Drs Ira Goldenberg and Leonard Prosnitz in the 1960s perform lumpectomies followed by radiation therapy. In 1964 he heard George Crile describe animal experiments that refuted the teachings of Virchow and Halsted: cancer cells did not spread predictably, Iymph nodes did not act as filters, and access to vital organs occurred via the bloodstream as well as the Iymph. In a 1955 article called "Common Sense in Cancer," Crile21 warned against super-radical attempts to accomplish the impossible. He noted that for many surgeons, the presence of cancer justified anything that they elected to do: "They do not admit that attempts to cure incurable cancers usually do harm. Fear of cancer should not be exploited. Surgeons should not subject patients to useless operations in cancer's name.... This is not the solution of a problem, it is the definition of one.... When we cannot cure, we must be careful that at least we do no harm."
Yet at the same time that Crile was rethinking the model and suggesting a less drastic surgical intervention, Owen Wangensteen, himself a surgeon of great distinction at the University of Minnesota, submitted that the reason the Halsted mastectomy did not produce better results was that it was not radical enough. Wangensteen proposed what he called a super-radical mastectomy, removing not only the pectoral muscles and Iymph nodes of the breast and underarm, but the nodes adjacent to the sternum as well as a portion of the first rib and collar bone. It was necessary for him to saw through and split the sternum to excise the Iymph nodes in the space around the heart. This brutal surgery required at least several weeks of hospitalization, and a number of women did not survive. To Wangensteen's credit, he noted his rather poor results, reported the operative deaths, and terminated the use of this procedure. He erroneously thought, however, that his "failure may have been in the execution of the concept rather than in the concept itself."22
In the early 1950s, Wangensteen's contemporary, Jerome Urban at Memorial Sloan-Kettering, excised a sizable portion of the chest wall in order to reach the internal mammary Iymph nodes. Bypassing mortality, Urban performed a comprehensive calisthenic surgery (removing more tissue than anybody else) without any proven gain. Wangensteen and Urban were both clinical investigators whose approach to medicine appeared to regard patients primarily as experimental subjects.
In contrast to the super-radical mastectomy, Crile's arguments began to be echoed by brothers Bernard and Edwin Fisher, who in 1958 began studies that were to culminate in the genesis of the school of "biological determinism"--meaning that the outcome of treatment was predetermined by the biology of a systemic disease process. Unlike many of his predecessors, Bernard Fisher was a pioneer in the application of clinical research methodology, establishing the importance of prospective randomized studies, which have now become the standard. Prospective means preplanned and randomized means selected by chance (such as every other chart). Through 23 clinical trials with thousands of women over decades, Fishery23 clearly established that mastectomy had no survival advantage over lumpectomy with radiation in women with a tumor size that conformed to the criteria of the study: 4 cm or less.
It was my father's medical school roommate, Nathaniel Berlin, clinical director of the National Cancer Institute (NCI) through the 1960s and chairman of the NCI Breast Cancer Task Force until 1975, who secured funding for Fisher's studies after Congress passed the National Cancer Act of 1971. The climate was such that Fisher was unable to recruit enough American surgeons into the study--they were unwilling to venture beyond the conformity of ideas and established standards of practice, though Canadian physicians were willing. The atmosphere surrounding the clinical selection of lumpectomy over mastectomy remained charged well into the '80s.
My father performed his first lumpectomy in 1978, but not without derision from his colleagues. On occasion, the women he treated would request a second opinion from another surgeon. If a woman had metastatic disease--sometimes years following a lumpectomy--one colleague of my father's insinuated that had the woman come to him (rather than my father), he would have done the proper operation (mastectomy) and cured her, thus proving he was able to "get it all."
By modern standards, Halsted's studies were sloppy and unkempt. This is not completely incomprehensible, though, because his landmark paper proclaiming "operations for the cure of cancer of the breast" was based on research between 1889 and 1894, the same period that his addiction plagued him so heavily. For the bulk of 1889 he was even hospitalized in Providence. Although Halsted's study covered the period between June 1889 and January 1894, he mistakenly included women in his report from March 1894, three months after the study was closed. Halsted3 stated: "Local recurrence is a return of the disease in the field of operation in the apparent or buried scar." Yet under the heading of women without local recurrence, he included those who recurred on their scar, contradicting himself. He focused on local recurrence, not survival, and tracked the women he saw for 3 years or less. Out of 50 cases, only 3 women were followed and found to be alive 3 years later. Eighteen were followed for less than 2 years, and 43 were followed for less than 3 years. If lumpectomy studies showed anything less than a 5-year survival, they would have been regarded as statistically laughable. But due to Halsted's authority and the ideological loyalty he inspired, his research methodology and results, though poor, never seemed to deter multitudes of followers.
One hundred years later, a double standard still remains. Lumpectomies are held to rigorous standards of efficacy, whereas mastectomies have never been subjected to anything close to the same requirements. A recent scandal has also clouded clear thinking. In 1994 Bernard Fisher, professor of surgery at the University of Pittsburgh, was ousted from his chair of the National Surgical Adjuvant Breast and Bowel Project (NSABP) because an investigator from Montreal, Roger Poisson, committed acts of scientific misconduct on Fisher's watch. Poisson altered surgical biopsy dates for 6 patients so they would be eligible within the Protocol B-06 requirements. His actions, irresponsible because of the deceit involved, did not, however, affect the end results. All 354 patients at his hospital were eliminated from the total group of 2163 women by subsequent auditors, and adequate numbers remained to assure overall credibility for the study, which covered the period between 1976 and 1984.24,25
There were, however, public alarm and breech of trust over this incident. Even though no patient's welfare was compromised, and no research outcomes were altered, the safety of lumpectomies was thrown into question by newspaper headlines that did not fully explain the nature of the error, possibly setting back use of this breast-conserving procedure. Now extensive reviews of Fisher's data have been published, confirming the original conclusions-namely, that mastectomy, lumpectomy, and lumpectomy with radiation provide comparable survival advantage.26
Outmoded Ideas and Practices
It is becoming clear that the Halsted mastectomy was based on an outdated model of breast cancer. Fisher27 revised the model after years of clinical trials, concluding that
"cancer is a systemic disease involving a complex spectrum of host-tumor interrelations and that variations in local-regional therapy are unlikely to substantially affect survival. All of the findings ... did not conform to the concepts that served as the basis for the principles of the Halstedian hypothesis but, rather, provided a matrix for the formulation of an alternative thesis, which is biologic, rather than anatomic and mechanistic, in concept. Its components are completely antithetical to those of the Halstedian thesis."
Fisher further clarified some misconceptions regarding u ho is eligible for lumpectomy. Tumor size or location does not preclude saving the breast by use of lumpectomy. Large tumors can often be shrunk by preoperative chemotherapy. Women with lymph nodes that are found to have (positive) or not have (negative) cancerous cells are equally eligible. Age is also not a factor--lumpectomy is equally appropriate for older and younger women. Finally, there is the issue of patient choice, and a woman's preference for mastectomy. To this Fisher27 says, "Patient autonomy will not be compromised and paternalism will not be resurrected if physicians firmly inform patients that, in almost all cases based on current knowledge, mastectomy is no longer justifiable, and lumpectomy followed by breast irradiation will not put them at greater risk of developing systemic disease or of dying than mastectomy would." Fisher's reanalysis and results were published in a 1995 report. He found that upon evaluation of three treatments (simple mastectomy, lumpectomy with irradiation, and lumpectomy alone), an average of 60% of patients were alive after 12 years and about 50% had no tangible signs of disease.26
To account for the discrepancy between the research supporting lumpectomy and the persistence of its l
lack of use, Harvard professor of surgery William Silen28 laments the replacement of data by dogma. "One of the best examples of this," Silen says, "is the use of the Halsted radical mastectomy for breast cancer.~ He identifies several problems, beginning with residency training when the young doctor is indoctrinated into managing situations in the "usual manner because that's the way we've always done it. Such normative behavior is expected to occur automatically and without question." He continues: "Beyond the period of training, surgical practice is strongly influenced by the leaders of the profession who are not always meticulously scrupulous in attention to the validity of the material they publish." He chastises the profession to more accurately assess the outcomes of what it does.28
Although remuneration for mastectomy is more than triple that of lumpectomy, financial motives do not account for the hegemony of this procedure. Habits and tradition assume an authority of their own. Is it reasonable to liken surgeons, men or women, to the tribal Africans who perform clitorectomies with the unshakable conviction that they are acting in the best interest of the woman? In both instances, what is best for the woman is associated with maintaining conformity with an outmoded belief. It is neither the women nor the doctors who are to blame; both come to the matter with honorable intentions. Cultural forces conspire: professional recommendations conflict, an irrational fear of keeping the breast is planted in women, and mastectomy constitutes a conclusive sacrificial act that permits women to feel as though they are doing everything they can.
Mastectomy itself is not difficult, nor does it constitute a serious risk. Perhaps it even serves as a form of penance for women who unconsciously feel that they have been bad enough, or foolish enough, to have contracted the disease in the first place. It appears to be the very least they can do to neutralize the offending body part, to cast it, along with some small measure of their fear, aside. Upon encountering the dreaded words of the doctor, "I'm sorry, the mass is malignant," a woman can be overcome by waves of shock, succeeded by an avalanche of terror, followed by the resolve to beat this disease. It is not uncommon for a woman to respond with offensive resolve, asserting, "l want it out."