In lieu of an abstract, here is a brief excerpt of the content:

c h a p t e r s i x The Generalization of Screening  From Cervical to Breast Tumors The practical success of screening for cervical cancer validated the principle of physical elimination of premalignant lesions and legitimated the extension of this preventive approach to other tumors.1 The transformation of cervical cancer into a model malignancy was, however, made possible by selective emphasis on some aspects of this tumor. For example, experts brought to the foreground the possibility of displaying of all the stages of transition from a mild cervical dysplasia to a fully invasive carcinoma but did not dwell on the erratic and unpredictable behavior of cervical lesions or the difficulty of correlating cytological diagnoses with outcomes.2 Their descriptions seldom indicated that cervical lesions have many unusual traits: very slow growth, accessibility to the physician’s gaze, and great fragility. A view that accentuated the universality of the “cervical cancer” model and underplayed its particularity facilitated the initial transposition of this model to other tumors. It might have also favored underestimating the difficulties of such transposition. Many cancer experts were influenced by the debate on the “biological determinism ” of malignant tumors. This debate focused on early detection of breast cancer. Surgical pathologist Ian MacDonald, biometrician Neil McKinnon (both were Canadians, but MacDonald worked in at University of Southern California ),and their colleagues argued in the 1950s that the“curability”of a given breast tumor depended mainly its biological characteristics. Slow-growing tumors can be cured even if discovered late (on an absolute time-scale), while aggressive and rapidly growing ones cannot be cured even if discovered early. Advocates of this view strongly criticized the “false premises and false promises” that dominated the field of cancer prevention and cure.3 Their views impressed many practition- ers. For example, a young American radiotherapist Franz Bushke wrote in 1957 to his mentor, Maurice Lenz: I suppose that you have seen Ian Macdonald’s paper in the proceedings of the Third National Cancer Conference, 1957. I thought it was very interesting and it thoroughly fits my philosophy. If his observations stand the test of criticism, I think they should once and for all stop our insistence on proving statistically the superiority of one method over the other. If 25 percent have a favorable prognosis,regardless of what is done,and 50 percent have an unfavorable prognosis , regardless of the type of therapy, and only in 25 percent treatment is critical , I do not think that one can ever prove the superiority of one method over the other by statistical methods.4 Many doctors, especially in Europe, agreed with MacDonald and McKinnon’s views and argued that if a cancer of the breast is growing slowly, a delay in its detection will not make much difference in terms of survival, and if it grows fast, it had already spread at the moment of detection of a lump in the breast.5 One of the promoters of this view, British surgeon Lester Breslow, explained in 1959 that “the practice of actively looking for lumps in the breast brings to medical attention and treatment nowadays cases of slow-growing cancer that in former years would have been neglected. These should have perhaps never have been diagnosed in the past, and also might have never have advanced to the point of causing death. Tabulating such cases now as cured or as long term survivors may be illusory.”6 Robert Sutherland’s 1960 book, Cancer: The Significance of Delay, reflected a similar preoccupation.7 Sutherland, a senior lecturer in preventive medicine at the University of Leeds and director of British Empire Cancer Campaign (BECC) council in Yorkshire, opens his book with Cicero’s quotation, “There are two grand faults to be avoided: the first is an over-great hastiness and rashness in giving up our assent, presuming that we know things before we really do so.” A diagnosis of early tumors, Sutherland explained, often relies exclusively on a histological analysis, not on information on the tumor’s biological behavior.8 However , histologically identical tumors can have very different fates:“Surely it would be better to discard the words early and late with their time connotations and their ambiguity, and to replace them with such purely descriptive terms as nonin filtrating, non-infiltrating but metastasizing, infiltrating and infiltrating and metastasizing.”9 Not only was the biology of a tumor highly variable, but reactions of the person who...

Share