Read an Excerpt
Cancer on Trial
Oncology as a New Style of Practice
By PETER KEATING ALBERTO CAMBROSIO THE UNIVERSITY OF CHICAGO PRESS
Copyright © 2012 The University of Chicago
All right reserved. ISBN: 978-0-226-42891-8
Chapter One
From "Nonentity" to Global Network The Rise of a New Style of Biomedical Practice
C'est toujours au confl uent de rencontres, de hasards, au fil d'une histoire fragile, précaire, que ce sont formées les choses qui nous donnent l'impression d'être les plus évidentes. (Foucault, 1994b: 449)
An Early Career in Clinical Oncology
In the early 1960s a young Canadian physician nearing the end of his medical degree, Pierre Band, developed an interest in cancer treatment. In order to explore his options for further study in the field, he contacted the American Medical Association (AMA). The reply was somewhat disheartening: the AMA representative was "not sure that [he could] provide [Dr. Band] with exactly the sort of information" he sought. The AMA had discontinued approval of cancer residency-training programs because the few that existed had "varied so widely in nature and duration." He suggested instead that Dr. Band contact the American Cancer Society (ACS) and the US National Cancer Institute (NCI). Like the AMA, the ACS replied that "there really [wasn't] any recommended special training for oncology" and pointedly added that a "cancer specialist" was a "non-entity."
While there were no "cancer specialists" at the beginning of the 1960s, the ACS did admit the existence of two "outstanding" cancer institutions—namely, Buffalo's Roswell Park Memorial Institute and New York City's Memorial Cancer Center. In other words, the lack of cancer specialists did not signify a deficiency of treatments or treating institutions but rather that the traditional specialties such as internal medicine, surgery, or radiology continued to control cancer treatment. Doctors could become a "so-called clinical cancerologist" only by specializing in those fields. In response to Band's inquiry, the NCI thus counseled that he first train in one of the established medical specialties and then join a clinical center devoted the study and treatment of malignant diseases.
In spite of the discouraging news, Dr. Band continued his quest for training in cancer therapy. Funded by a National Institute of Cancer of Canada fellowship in 1964–65, he spent eighteen months in Georges Mathé's unit at the Gustave-Roussy Institute in Villejuif on the outskirts of Paris, arguably then, as today, the leading French cancer hospital. Subsequently, from 1966 to 1969, having availed himself of a National Institutes of Health (NIH) fellowship, he worked at the Roswell Park Memorial Institute. The two fellowships not only familiarized him with the routines of two premier cancer institutions, they also introduced him to a new organization for clinical research that embodied a new "style of practice," cooperative oncology groups.
Unlike previous cancer treatment and research programs that had functioned in single institutions, cooperative groups treated and studied patients across multiple cancer treatment centers using common protocols. In Europe, Band's mentor, Georges Mathé, was one of the originators of Europe's first cooperative oncology group: the Groupe Européen de Chimiothérapie Anticancéreuse (GECA), founded in 1961 and the forerunner of the European Organization for Research and Treatment of Cancer (EORTC). In America, Roswell Park acted as one of the principal sites of the Acute Leukemia Group B and of the Eastern Cooperative Oncology Group (ECOG), established in 1955 by members of the NCI under the name of Eastern Solid Tumor Group (figure 1.1). Dr. Band attended his first ECOG meeting in 1966 and when he later joined the University of Alberta in Edmonton, the university became the first Canadian ECOG affiliate. In the early 1970s Dr. Band was elected chairman of ECOG's breast cancer committee and designed a landmark clinical trial protocol that will be discussed later in this book. Through these organizations, Dr. Band ultimately pursued a successful career as a chemotherapist and a cancer clinical trial specialist, both as a member of Canadian clinical and academic centers and as a principal investigator for ECOG and GECA/EORTC.
We begin with this vignette about Dr. Band's career because it highlights the remarkable rise of cancer treatment research. Consider the following figures that contrast sharply with the idea of cancer treatment as a marginal pursuit. In the United States at the end of the first decade of the twenty-first century, the US National Cancer Institute, the largest funding source for clinical cancer trials, sponsored 14,000 investigators at over 3,000 institutions, enrolling 25,000 patients in clinical trials each year. In Europe, 300 hospitals and cancer centers in more than thirty countries and almost 3,000 scientists and clinicians participated in EORTC activities by 2010. The number of yearly articles related to cancer clinical trials grew from fifty-six in 1965 to over 6,000 in 2009 (figure 1.2). Similarly, the American Society of Clinical Oncology (ASCO) expanded from sixty-six members in 1964 to more than 27,000 in 2010, when it attracted 32,700 attendees (26,600 professional and 4,370 exhibitors) to its annual meeting (figure 1.3). Founded in 1975, the European Society for Medical Oncology (ESMO), the continental equivalent of ASCO, counted some 5,000 members in 2010 and drew 16,000 participants to its annual meeting. Last but not least, anticancer drugs have become a major pharmaceutical market, and cancer research is now pursued on a global basis (Eckhouse, Lewison & Sullivan, 2008; Chabner & Roberts, 2005) with an estimated global anticancer market of $70 billion in 2008. The exponential growth of clinical cancer research has produced concomitant changes in routine cancer treatment: the vast majority of (Western) cancer patients are treated according to treatment protocols established via multicenter clinical trials.
In short, while "cancer specialists" or "clinical oncologists" have become common figures in the medical landscape, they did not exist a few decades ago. In the early 1960s a cancer specialist was not only a nonentity, many practitioners also regarded the treatment of terminally ill cancer patients with heroic courses of chemotherapy as highly questionable. The randomized clinical trials that today form the basis of medical oncology were relatively rare and aroused outright opposition from physicians loath to assign patients randomly to competing treatments. Barth Hoogstraten, a Dutch-born physician who immigrated to the United States after the war to become one of the pioneers of cancer chemotherapy, has recalled how in the mid-1950s patients admitted to the NCI's Department of Medicine were treated by physicians "who had no formal training in oncology for the simple reason that the medical schools did not yet offer such programs, and the word oncology did not even exist" (Hoogstraten, 2005: 82). Emil Frei III, another founding father of US chemotherapy, recalled that the "greatest underlying controversy" of those early years was whether one should treat cancer patients with chemotherapy since the treatment was often worse than the disease. In developing cancer chemotherapy, he and his colleagues in Europe thus confronted a "cynical" or "skeptical" old guard. Maurice Tubiana, a leading postwar French oncologist, has described how, during the introduction of clinical trials at the Gustave-Roussy Institute in the early 1960s, "several physicians were scandalized and reacted with hostility.... The controversy soon turned into a quasi-religious war; some physicians, including the head of the surgery department, a very distinguished and prestigious person, decided to resign from a hospital where the fate of patients was decided by tossing a coin" (1995: 301–2, our translation). How then did such a spectacular change in the medical landscape occur? How did medical oncology move from a nonentity and in some regards a reviled practice to the central position it now occupies in modern medicine? This book attempts to answer these questions.
About This Book
Writing the history of a laboratory or organizations such as the Roswell Park Memorial Institute, the NCI, ASCO, or ECOG is a relatively straightforward task; writing the biography of a few key individuals in cancer research or an intellectual overview of the development of the field would also be a relatively clear-cut undertaking. Examples of these historical genres abound, and they are valuable additions to the literature. None of these forms of historical inquiry seemed to capture the research we wished to pursue. We had decided from the outset not to write a standalone institutional history of cancer organizations, a conceptual history of the evolving understanding of cancer, or a political or economic account of the governmental and industrial shaping of the cancer market. Rather, our research sought to capture all those aspects of cancer research and treatment within a single frame. We hoped to avoid the snare of writing a "total" history of the meteoric rise of cancer treatment and research. Rather than the elusive global picture, we sought more modestly to describe the evolving configurations of institutional, organizational, conceptual, human, and material resources that, through a remarkably dense network, have produced the now-taken-for-granted pillars of modern oncology: cancer clinical trials, treatment protocols, and the like.
The organizing principle of this book is simple: in addition to the present introductory section, there are three main parts, each corresponding to a major period in the development of cancer clinical trials. Each part begins with the description of a clinical trial that is paradigmatic of the corresponding period. Subsequent chapters focus on the novel aspects of the trial, such as the kinds of drugs used; the material, organizational, and statistical infrastructure of the trial; and its goal(s) and nature. Of course, no single trial captures all the subtleties and complexities of any given period, and each part thus describes other trials and related events. In spite of its shortcomings, introducing each part with a concrete undertaking—a specific clinical trial—has a clear and distinct advantage: it allows us to discuss practices—to "approach our topic by analyzing what people did" (Foucault, 1994b: 634–35, our translation)—rather than focusing on structures such as professions, specialties, or organizations.
Practices do involve such institutions (macroactors, some sociologists would say) and we discuss them in the course of this book. But—pace the social sciences with their neat disciplinary borders—practices, especially in emerging areas such as clinical oncology in the post–World War II era, do not respect the settled boundaries between social, economic, cultural, and organizational domains; they reconfigure existing structures in often unexpected ways and they mobilize resources assigned to other tasks. In short, they generate fluid arrangements that sometimes congeal into stable innovations (at least temporarily) and sometimes do not. Given that our interest lies in the establishment of cancer clinical trials and related practices, we have opted for an empirical analysis of the processes that led to this outcome.
By focusing on practices and their outcomes, we have resisted the temptation, felt by many social scientists, to use the past to explain the present and, in particular, to account for innovations by reducing them to preexisting practices or structures: while novelty necessarily emerges from existing structures—fully de novo creations do not exist—such novelty cannot be reduced to elements of those structures or to any deterministic interaction thereof. According to Paul Rabinow (2000: 44), a social science bent on explaining away the emergence of new forms by condensing them into a set of historical or social determinants is unable to acknowledge singularities—that is, new assemblages that "make things work in a different manner." Accounts of innovation should similarly avoid privileging, a priori, single factors or structures, whether social, technical, cultural or economic. As in our previous work on biomedical platforms (Keating & Cambrosio, 2003), we consider technological determinism—the explanation of history through technical or scientific breakthroughs—and sociological reductionism—positing social or political or cultural factors as the ultimate explanation of medical and techno-scientific developments—as mirror images of the same mistake. In short, we look for singularities rather than regularities, and we do so by recognizing that "there are no fundamental phenomena. There are only reciprocal relations, and perpetual gaps between them" (Foucault, 1994b: 277, our translation).
One of the consequences of our approach is that we do not deploy one of the standard sociohistorical tricks of the trade—namely, comparisons: between countries, patterns of research and education, different kinds of cancer, and different kinds of hospitals. To understand why, recall that in our initial vignette we saw how Dr. Band's early training spanned two continents and how clinicians and researchers established institutions such as ECOG and the EORTC on both sides of the Atlantic. While this would normally prompt comparisons between America and Europe in order to uncover sociocultural differences between them, our take here is quite different: the vignette foregrounds a network that links European and North American oncologists. In other words, the emergence and development of this network attracts our attention. Comparative analysis, in contrast, reworks the recurrent theme of the constraints and determinants of practice by showing how perceptions and actions are molded through vaguely defined local "institutions." We have instead tried to understand how practitioners produced increasingly similar perceptions, actions, and practices through the constitution of clinical research networks that span countries and continents or, to use a more abstract terminology, through the establishment of a new style of practice.
The establishment of oncology networks linking North America, Europe, and Japan has been the explicit goal of organizations such as the NCI. Most of the founders of clinical oncology in Europe transited through US institutions, and the NCI established a European outpost (a liaison office) in Brussels in 1972 to foster the integration of Western oncology and to gain greater access to European anticancer agents. Thus instead of looking for differences corresponding to preset categories (nations, hospitals, professions, insurance schemes, etc.), we suggest that one may profitably investigate the work implicated in the production of similarities without overlooking manifest differences. More often than not the constitution of a common practice is antecedent to the production of differences in unexpected quarters (e.g., larger differences within, rather than between nations). The history of cancer clinical trials amounts to an attempt to create a network spanning Europe and North America, articulated through a few major hubs (e.g., the NCI) and through which all sorts of entities (oncologists, protocols, drugs, etc.) circulate.
Contrary to the traditional medical specialties that developed locally and nationally before joining transnational initiatives, medical oncology and cancer clinical trials have pursued an international or, at least, a multicenter project (maps such as the one on the background of figure 1.1 are often used in the trialists' reports) from the outset. The alignment of clinical and research interfaces across numerous national and international sites has constantly preoccupied practitioners in this field, as has the production of clinically meaningful differences across clinical sites. During its 1966 congress in Tokyo, the Union for International Cancer Control (UICC) launched a Project on Controlled Therapeutic Trials to list ongoing and unpublished trials and to "keep under review" the methodology of the trials. In 1974 the UICC redefined the scheme to focus on the difficulties ("whether methodological, ethical or practical") raised by clinical trials. Reoriented yet again in 1978, the project now concentrated on the methodology of trials "with a view to rendering them easier to conduct and to have them yield more valid conclusions" (Flamant & Fohanno, 1982: 1–3). As part of the project, members published international surveys of randomization practices in cancer trials. The surveys had an explicitly normative edge ("to make recommendations regarding how randomization should be carried out") (Pocock & Lagakos, 1982: 5). Similar studies investigated alternative trial designs and the rationale for designing large trials (Grage & Zelen, 1982; Peto, 1982). Explicit recommendations for large trials included "Make complexity optional," "'Flag' all patients," "Keep a log," and "Telephone randomization" (no e-mail back then). As in the field of international health where American bilateral agreements competed with international projects, the NCI sponsored collaborative programs with Europe and Japan that promoted the free fl ow of drugs, scientists, and information and that led therefore to the harmonization of practices (Wittes & Yoder, 1998; Newell & Sugano, 1977; Goldin, Muggia & Rozencweig, 1979a).
(Continues...)
Excerpted from Cancer on Trial by PETER KEATING ALBERTO CAMBROSIO Copyright © 2012 by The University of Chicago. Excerpted by permission of THE UNIVERSITY OF CHICAGO PRESS. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
