Main

Dedicated cancer centres now form an important part of the cancer research landscape worldwide, and many of them are recognized as centres of excellence — not only by researchers, but also by those patients seeking state-of-the-art treatment and access to clinical trials. In the United States, the Cancer Centres Programme of the National Cancer Institute (NCI) is now fully accepted as an integral component of the nation's cancer research effort. In the United Kingdom, the newly formed National Cancer Research Network will probably develop a similar programme. Last year, the NCI Cancer Centre Programme used $169 million or 7.7% of the extramural NCI budget. These funds provided partial support for 60 NCI-designated cancer centres in 31 states, of which 40 were deemed 'comprehensive'. Although this commitment is small in comparison to the budget for grants to individual investigators (RO1 grants — $899 million or 41% of the extramural budget), it represents a vital force in cancer research, treatment and prevention, and is firmly based in NCI and National Institutes of Health (NIH) history.

Comprehensive Cancer Centres are designed to join the forces of basic, translational and population cancer research into ever-improving clinical trials in adult and paediatric oncology. Additional aims are to provide effective cancer education and prevention methods to the surrounding community and wider region, and to offer the highest quality surgical, radiotherapeutic, medical and paediatric treatment for cancer. To accomplish these aims, the centres must invest heavily in basic and population science, as well as in clinical science. How did this multidisciplinary approach to cancer research and treatment arise? (see TIMELINE.)

Timeline | The evolution of cancer centres in the United States

History of the designated cancer centre

The concept of the NCI-designated cancer centre has its roots in the period immediately after the Second World War1,2,3 when Vannevar Bush (Box 1) published Science, the Endless Frontier4 — his tribute to the future of science. Scientific achievements had contributed significantly to victory, convincing Bush and the leadership of the United States that a further investment in basic scientific research by government — in both the public and the private sectors — would greatly enhance the nation's health and welfare, as well as its economic strength. Within the sphere of health research, the then relatively tiny NIH and the Office of Naval Research began to fund basic research grants to a few universities and their medical schools. But from the very beginning of the NIH effort, certain congressmen, particularly Senator Lister Hill of Alabama and Representative John Fogarty of Rhode Island, recognized that the American public would not continue to support basic biomedical research unless it was directly and visibly linked to an expansion of the conquest of disease5. Their astute sense of congressional mood was moulded into broad national policy by the efforts of individuals such as Mary Lasker and Sidney Farber, organizations such as the American Cancer Society, and others of a similar stance who supported political and fundraising efforts on behalf of devastating diseases such as cancer, heart disease and stroke.

The list of categorical NIH institutes has grown continuously since then. Within almost all of these institutes — and particularly in the NCI — there was a strong sense that Congress and the tax payers wanted every effort made to close the gap between what could be learned at the bench and the application of that knowledge at the bedside. In the early 1950s, the intramural programme of NIH on its campus in Bethesda, Maryland, responded to that vision and created Building Ten, a 500-bed research hospital that was later named the Warren Magnusen Clinical Centre (Fig. 1). Within the Magnusen Centre, the laboratories of investigators in the various NIH institutes were in close proximity to the beds of what became the largest, best-equipped, and best-funded clinical research centre in the world. NCI rapidly became the largest user of the centre as trials using combinations of chemotherapeutic agents showed increasing promise, particularly in the childhood malignancies and lymphomas6,7,8,9 and, later, in breast cancer10.

Figure 1: The Warren G. Magnesen Clinical Centre's first patient: Charles C. Meredith.
figure 1

Charles Meridith was enrolled in a trial that studied hormone treatments for prostate cancer. (Image courtesy of the Warren G. Magnesen Clinical Centre, Bethesda, Maryland, USA.)

Full size image

The Korean War had a profound influence on what followed. Young doctors were drafted into the armed services, and many young residents from leading academic health centres, seeking an opportunity to meet their service obligations, applied to the commissioned corps of the United States Public Health Service, and to the NIH, for positions as Clinical Associates in the Magnusen Centre, where they were needed to provide medical care for the increased number of patients admitted to the new hospital. In return for their service, they were trained in the principles of clinical research, particularly translational cancer research. Many of these young trainees were bitten by the bug of clinical research and, having completed their terms of service, returned to academic health centres throughout the country where they were funded to continue clinical research in their field of interest by a growing NIH budget. To further assist them and their mentors and to expand clinical research throughout the country, the NIH developed the General Clinical Research Centres; these relatively small centres are now located in 70 academic health centres throughout the United States. As well as construction costs, funds to defray some of the infrastructural costs of clinical research — such as nursing and dietary care, physician and administrative leadership, specialized laboratories and hospital admissions — were provided.

In the early 1960s, in response to perceived opportunities in clinical research, the extramural programme of the NCI announced three new initiatives that presaged the present cancer centres. Cancer Research Facilities Grants permitted construction of buildings solely devoted to cancer research, whereas Cancer Research Programme Project Grants, and Cancer Clinical Research Centres Grants offered funds for collaborative research and for construction of clinical cancer research units similar to the General Clinical Research Centres. These initiatives were greeted with enthusiasm by cancer researchers throughout the country — particularly those who were already working in dedicated cancer research institutes that were loosely affiliated with medical schools, such as Roswell Park, Memorial Sloan Kettering and Fox Chase. But it did arouse suspicion and outright opposition by some leaders of broadly based academic health centres. Opponents held that cancer research — whether basic or clinical — should not be isolated from the mainstream of biology or clinical medicine. Cancer, it was stated, is a vast biological problem — best understood at a fundamental level if individual investigators work in a multidisciplinary setting where they can receive intellectual and technical input from colleagues in a wide variety of disciplines. Clinical cancer research — which, at the time, was narrowly defined by many as clinical trials of chemotherapeutic agents — would be far better pursued in a general medical setting in the same way that antibacterial chemotherapy was managed.

Behind the shield of this intellectual reasoning was, and still is, a more fundamental concern — money. Cancer represents a huge fraction of clinical medicine and brings a proportionately large amount of funds — donated by patients, paid by health insurers or acquired through grants — into academic health centres. This revenue could be shared with other less fundable departments, and could therefore be important in maintaining the entire infrastructure of these complex academic enterprises. What would happen if these funds became locked away in separate cancer centres? Although practical opposition to the centre concept was — and can still be — vociferous, supporters have successfully argued that neither paediatric nor adult oncology can exist in isolation from other disciplines, such as imaging science and radiation, and surgical oncology. Nor can they advance without the assistance of epidemiologists and biostatisticians. Furthermore, clinical progress can only be made by teams of clinicians, clinical investigators and basic scientists who are working together to translate information gained at the cellular and molecular level into new therapeutics and diagnostics. Moreover, cancer cannot be considered a single disease. Each type of cancer presents unique scientific and clinical challenges that require intensive subspecialization that is far beyond the capability of a single oncology division in a department of medicine, surgery, radiology or paediatrics. Effective progress in cancer care and cancer research demands specialized multidisciplinary centres of excellence. Hence the inevitable, if controversial, movement towards a cancer centre programme within the NCI.

By 1966, the die was cast. A single large grant was made by NCI to the Memorial Sloan Kettering Institute. The grant was intended to provide partial support for the total research programme at Sloan Kettering, replacing more than 40 smaller, individual grants. This gave NCI an opportunity to review the Institute's entire programme in a single site visit and allowed Sloan Kettering to provide infrastructure support to an integrated programme of clinical, population and basic cancer research that would gradually replace the uncoordinated efforts that had gone on before. Similar grants were soon awarded to the M. D. Anderson Cancer Centre and the Roswell Park Cancer Centre. The Cancer Centres Program had been launched.

Nixon's war on cancer

In 1971, NCI's resources and independence were massively increased by legislation that followed Richard Nixon's controversial 'war on cancer'. This was inspired by growing victory in the childhood cancers, but at the time there was no evidence that the common adult cancers would respond similarly. Popular theories at the time, which lacked firm scientific bases, included the following: first, cancer is a virally mediated disease, such as polio or smallpox, that can be eliminated by vaccination; second, the immune system can be taught to eliminate enough cancer cells to cure patients; and third, combinations of new chemotherapeutic agents will destroy cancer cells without crippling toxicity. Accordingly, the Cancer Centres Programme and all NCI programmes received increased funding. Three types of cancer centre were envisaged. The largest type, the Comprehensive Cancer Centre, was defined as an institution devoted to clinical, basic, epidemiological and biostatistical cancer research and to advanced diagnostic, treatment and preventive methods in most, if not all, of the cancer subspecialties. Two additional types of cancer centre, clinical and basic, were classified as Specialized Cancer Centres, designed to take advantage of unique skills in a restricted area of cancer research. Of course, it was predictable, if unappreciated at the time, that as the number of centres grew, the budgets for each would become constrained. To this day,budget limitations and perceived lack of budget flexibility remain the main complaints of cancer centre directors11,12.

It is easy to look back over 30 years and scoff at the naivety of the planners of the war on cancer and the early history of the cancer centres programme. But those planners did foresee the future. There is now a much firmer understanding of the role of viruses in cancer13,14,15, and approaches to the immune system that might turn it against cancer are being actively pursued — at the bench and in the clinic16. New anticancer agents with great specificity, such as trastuzumab (Herceptin) and STI-571 (Glivec), are now in the clinic (see below). This is because we have invested heavily in cancer biology and have also begun to invest in its translation to clinical medicine.

Cancer centres today

Regional centres of excellence. Of the 60 cancer centres that exist in 2001 in the United States, there are 9 basic cancer centres, 11 clinical cancer centres and 40 Comprehensive Cancer Centres. Most of the states in which cancer centres are located have only one Comprehensive Cancer Centre, and this can act as a regional centre for the surrounding, less-populated states. However, some highly populated states have several Comprehensive Cancer Centres (up to five). Whether the regional cancer centres have created uniform excellence of therapy throughout the areas that they serve is open to debate. Competition, rather than collaboration, dominates academic medicine, and private oncologists or surgeons in community hospitals are often unwilling to refer patients to cancer centres for fear of losing them. But in an increasingly sophisticated information-based age, patients are self-referring to designated cancer centres because they have reason to believe that more advanced care might be available to them in such settings. Nearly all paediatric cancer patients are managed in cancer centres, but it would be impossible to manage all adult oncology patients in such a small group of institutions. Clearly, collaborations between cancer centres and community physicians and hospitals are required if cancer care is to be uniformly excellent in the United States and the designated cancer centres are to serve the regional purpose for which they were created. Such collaborations are difficult to achieve and irregular in performance.

Clinical trials, education and prevention. A Comprehensive Cancer Centre is designed to organize cancer care and research so as to optimize the delivery of ever-more-effective early diagnosis, therapy and prevention. This is accomplished by joining the forces of basic cancer research, translational research and population science into new clinical trials and preventive strategies. All of this must be accomplished at the same time as serving the cancer education and prevention needs of the community that surrounds the centre. To meet these ambitious goals, all of the members of a cancer centre should be brought together as an interdisciplinary team of colleagues who function in a cross-fertilizing fashion that supports interaction. This matrix approach5 (Fig. 2) is gaining recognition as the optimal way to organize a cancer centre. Better approaches are definitely needed as the latest cancer statistics show that adult cancer prevention, early diagnosis and treatment results are improving only slowly17,18. But although progress in cancer prevention and early diagnosis is slow, there have been some important advances made in cancer centres. For example, the nurses' and physicians' study within the Dana-Farber/Harvard Cancer Centre is the largest analysis of the incidence of cancer and its links to lifestyle choices, such as diet, in any population worldwide19,20,21. It provides important data that have exposed the mysticism that, for many years, has surrounded the influence of lifestyle on cancer. The M. D. Anderson Cancer Centre has advanced the notion of chemoprevention of cancer, and several chemopreventive agents, such as carotenoids, are now in clinical trials22,23,24,25. Finally, The Johns Hopkins Cancer Centre has been a leader in the clinical application of cancer genetics, particularly in colorectal cancer. This has led to improved genetic testing for individuals with hereditary colorectal cancer syndromes26,27,28.

Figure 2: The interactions of clinical programmes and supporting disciplines in a matrix cancer centre.
figure 2

Ten disciplines that are staffed by either basic scientists, population scientists or clinical scientists, or sometimes by combinations of all three, interact with six clinical programmes that are led by clinical scientists but are staffed, in part, by basic scientists and/or population scientists. Where productive interactions occur, a red spot is shown. Note that the clinical programmes are heavily dependent on the supporting disciplines.

Full size image

A matrix cancer centre. The Dana-Farber/Harvard Cancer Centre (DF/HCC) in Boston is organized as a matrix (Fig. 2). DF/HCC is led by the President of the Dana-Farber Cancer Institute and the Dean of Harvard Medical School, with the assistance of the Dean of the Harvard School of Public Health and the Chief Executive Officers of the five main Harvard Medical School affiliated hospitals. DF/HCC draws its 800 faculty members from the entire Harvard Medical School and School of Public Health, and currently emphasizes six clinical programmes: breast oncology, gynaecological oncology, leukaemia, lymphoma, neuro-oncology and prostate oncology. Paediatric oncology is woven into the leukaemia, lymphoma and neuro-oncology programmes. Clinicians, clinical investigators, basic scientists and population scientists are members of each of the clinical programmes. The clinical programmes, in turn, interact in a matrix with a series of basic and population science disciplines ranging from biostatistics (which interacts with all of them) to viral oncology, which supports the gynaecological oncology and lymphoma programmes.

A small fraction of the funds awarded by the NCI to DF/HCC — or to any NCI-supported cancer centre — through a Cancer Centre Support Grant, supports the leadership of the clinical programmes and the disciplines that form the matrix, but none of the awarded funds actually supports the research projects carried out by the matrix. Such funding is derived from programme project grants, individual grants, other federal and non-federal grants and private donors whose generosity makes the entire enterprise possible. The bulk of the NCI Cancer Centre Support Grant supports the essential infrastructure of the centre, such as the biostatistics department, shared core laboratories, population science and animal facilities that are used by most, if not all, of the programmes and disciplines. Support is also provided for the heavy costs of protection of human subjects, the proper conduct of clinical trials and the necessary costs of watchful administration. Though none of the cancer centre awards are directed towards the support of projects, the centre is evaluated every 5 years on the basis of the quality of its projects, its contributions to its community and the strength of its organization and leadership.

The development of a matrix cancer centre involving an entire academic health centre as complex as the Harvard system is a major task that requires enormous efforts on the part of devoted faculty members and administrators to achieve success. One of the main challenges is balancing the goal of collaboration against the needs of individuals to support their own research and achieve academic recognition. Biomedical research in the United States has a strong centrifugal tradition: the vast bulk of NIH funds is devoted to the support of individual investigators operating in their own laboratories and fully occupied with their own projects. It is firmly held by many that progress will best be made if we rely on the creativity of individual scientists who are driven by their own curiosity. Cancer centres try to respect the individual scientist's creative ideas, as well as encouraging interdisciplinary collaboration that can lead to more effective clinical trials. Has this balancing act been successful? The examples in Box 2 illustrate the power of the collaborative approach.

Towards the future

Will cancer centres that emphasize collaboration provide the most important weapon against cancer? Or will a less organized approach through individual investigators have a more important role? Are organized research programmes wise investments or should we rely entirely on individual investigators ploughing their own furrows? No one can know the answers to these questions, although the arguments are repeated regularly in the scientific literature29, and even in the lay press30. The NCI has wisely supported several approaches, but with a much larger investment in individuals. We are persuaded that interdisciplinary collaboration — with more emphasis on recruitment of information scientists, chemists and engineers into existing cancer centres — is a very promising path to future success. Armed with the new weapons of genetics and strengthened by a multidisciplinary approach, the 'war on cancer', in the past derided as a propaganda tool, might well be won.