REVIEW www.nature.com/clinicalpractice/onc

Dichotomies in cancer research: some suggestions for a new synthesis Michael B Sporn

S U M M A RY Continuing high cancer incidence and mortality raise concern about the prevailing overall approach to the control of this disease. The purpose of this article is to elaborate on fundamental dichotomies between traditional and revisionist viewpoints and then to attempt a synthesis of these contrasting perspectives. Topics considered include the importance of controlling carcinogenesis in its earliest stages; consideration of epigenetic, as well as genetic, factors in cancer; development of appropriate genetic animal models of carcinogenesis; the need for multifunctional agents to prevent and treat cancer; and the limits of reductionism. The need for development of new preventive and therapeutic measures that will maintain quality of life, not merely extend life, is stressed. Finally, the importance of context in cancer biology is emphasized, as epitomized in Walt Whitman’s famous quotation that “Nothing out of its place is good and nothing in its place is bad.” KEYWORDS carcinogenesis, chemoprevention, epigenetics, oncogenes, reductionism

REVIEW CRITERIA A formal literature search for this review was performed using Entrez for articles published to January 2006. This review includes a summary of the author’s own work and knowledge based on reading oncology literature over the past 30 years, which covers various fields relating to oncology and cancer prevention. Knowledge gained from regular attendance at conferences, workshops, and other national and international meetings was also included.

MB Sporn is Professor of Pharmacology and Medicine at Dartmouth Medical School, Hanover, NH, USA. Correspondence Department of Pharmacology, Dartmouth Medical School, Remsen 524, Hanover NH 03755, USA [email protected] Received 24 January 2006 Accepted 27 April 2006 www.nature.com/clinicalpractice doi:10.1038/ncponc0536

364 NATURE CLINICAL PRACTICE ONCOLOGY

INTRODUCTION

Rather than being a cause for unfounded optimism that “we are poised to make dramatic gains against cancer in the near future”, the latest Cancer Trends Progress Report—2005 Update from the US National Cancer Institute (NCI)1 should raise grave concerns that our entire perspective and approach to the overall problem of controlling cancer is inadequate. It is self-evident that cancer incidence drives cancer mortality (as well as total cancer-related costs to society), and the latest figures, which indicate that the overall rate of occurrence of new cases of cancer has not diminished,1,2 are not cause for optimism. The NCI Report touts a minuscule drop (only a little over 1%) in the overall cancer death rate, and while the four most common cancers (lung, prostate, breast, colorectal) show slightly better results (but no more than a 3% reduction), such incrementalism hardly generates optimism that we will “eliminate the suffering and death caused by cancer, and…do so by the year 2015”.1 On the contrary, devastating killers such as pancreatic cancer, melanoma, and lung cancer in women who do not smoke, show no signs of being eliminated.2 Most of all, these sobering facts indicate that we need to address the basic assumptions that underlie current cancer research and practice, and develop alternative approaches. This review is not intended to unleash a polemic, since there have indeed been spectacular recent advances in basic cancer research and clinical practice.1 Rather, it will elaborate on fundamental dichotomies in our outlook between traditional and revisionist viewpoints; these dichotomies can hopefully be bridged to create a new synthesis that builds on old strengths, while encouraging a fresh approach to the entire cancer problem. Seven such dichotomies are discussed, and an attempt at a synthesis for planning future action is made. DICHOTOMY 1

‘The disease is cancer’ versus ‘the disease is really carcinogenesis’ There is a widespread reluctance to recognize that cancer is fundamentally a chronic process JULY 2006 VOL 3 NO 7

REVIEW www.nature.com/clinicalpractice/onc

(carcinogenesis) that begins long before patients become symptomatic and seek medical advice. This reluctance is shared by the entire community that deals with cancer, including physicians, the pharmaceutical industry, the insurance industry, many of the philanthropies that fund cancer research, the US NCI (as reflected in its budgetary priorities), and finally the FDA, which regulates the introduction of new drugs into the clinic. Until cancer is recognized as a chronic disease process, the practical, economically feasible control of cancer will be elusive. This problem has been discussed in many reviews.3–6 DICHOTOMY 2

‘Emphasis on cure of end-stage disease’ versus ‘prevention of early disease progression’ Our failure to conceptualize cancer properly leads, inevitably, to our failure to combat it. In contrast to current practice in cardiovascular medicine, in which so much emphasis is now placed on prevention of progression of early atherosclerosis with the use of statins, antiplatelet agents, and antihypertensives, the oncology community has not embraced the concept of chemoprevention (Box 1) of carcinogenesis.3,5 It is difficult to obtain accurate figures relating to budgetary allowances, either private or public, for prevention of progression of early disease, compared with treatment of end-stage disease. The NCI’s past practice of combining all basic research in cancer into the ‘prevention’ category has complicated this analysis, but it is readily apparent that vastly greater resources are devoted to treatment of end-stage disease. A quick tour of the exhibit floor of any major national cancer meeting in the US illustrates how much effort Big Pharma devotes to treatment of end-stage disease, while almost dismissing the development and use of drugs for chemoprevention. Even when prevention is considered, this approach is often limited to efforts to alter lifestyle or the environment, as shown by the neglect of chemoprevention in the recent NCI Progress Report,1 which devotes over 50 pages to various efforts to evaluate or alter lifestyle or the environment, without ever mentioning the word ‘chemoprevention’. This omission occurs in spite of the fact that chemoprevention is now a proven modality for preventing breast cancer in women.7–9 Simply modifying lifestyles will not eliminate cancer. We must find new ways to deal with the alarming number of lung cancer deaths in women who do not smoke—what are JULY 2006 VOL 3 NO 7 SPORN

Box 1 Definitions for common oncological terms. Chemoprevention The use of pharmacologic agents to arrest or reverse the process of carcinogenesis at its earliest stages, to prevent the development of invasive cancer Epigenetics Alterations in gene expression and function that occur without changing the nucleotide sequence of DNA. Examples include chemical modifications in DNA (e.g. DNA methylation) or histones, and alterations in chromatin structure (e.g. chromatin remodeling) Histone language or histone code Covalent modifications of histone molecules that are read by other proteins to bring about distinct downstream transcriptional events SERM (selective estrogen receptor modulator) A single SERM can have either pro-estrogenic or anti-estrogenic action in different organs, depending on context. Thus, raloxifene has desired antiestrogenic activity in the breast, but it does not have undesired pro-estrogenic activity in the uterus, as tamoxifen does. Therefore, unlike tamoxifen, raloxifene does not cause uterine carcinoma

the lifestyle changes that will stop these deaths? We need new, cost-effective technologies to identify early bronchial dysplasia, and safe drugs to control its progression.6 Similarly, in pancreatic cancer, we need new methods to identify the earliest lesions.10,11 It should be realized that “the very notion of early detection requires reassessment for pancreatic cancer...[and that] the only hope for cures lies in the detection and eradication of the preinvasive state”.12 In response to problems such as these, a new synthesis is required that will encourage not only the important Susan Komen Foundation ‘Race for the Cure’, but also a new ‘Race for Prevention’. We must define prevention as a joint approach that will emphasize both the use of pharmacologic agents for chemoprevention and a healthy, non-smoking, non-obese lifestyle. The entire American health-care system needs to revise its priorities regarding prevention, since dismissal of preventive measures is not a problem limited to cancer alone. For example, a recent report on type 2 diabetes (which is now of epidemic proportions) emphasizes that insurance providers pay for extremely expensive treatments of end-stage disease (such as diabetes-related amputations), but refuse to pay for preventive efforts that would eliminate the need for such drastic measures in the first place.13 Similarly, in oncology practice, NATURE CLINICAL PRACTICE ONCOLOGY 365

REVIEW www.nature.com/clinicalpractice/onc

most prevention activities devoted to stopping progression of disease in its earliest stages are not reimbursed by insurance providers. DICHOTOMY 3

‘Cancer is a genetic disease’ versus ‘cancer is also an epigenetic disease’ Cancer is genetic. Everything in biology is genetic! If cancer is merely a genetic disease, why is it that there are so many more heavy smokers, most of whom incur DNA damage, than there are cases of lung cancer? Why are there so many more cases of advanced dysplasia of the uterine cervix than cases of carcinoma of the cervix? Epigenetics in its broadest sense is now recognized as a discipline essential for understanding the process of carcinogenesis (Box 1).14–18 Recent emphasis on the critical role of inflammation in carcinogenesis is a reflection of the importance of epigenetics, since inflammation is an epigenetic process driven by context.19–22 As Wallace Clark has noted, carcinoma is the result of a prolonged series of failures in the reciprocal communication between epithelium and stroma.23 Carcinoma is the result of dysfunctional interaction between genetic and epigenetic factors. Epigenetics leads to the study of mesenchymal elements (the stroma) associated with all epithelia, because mesenchyme provides the context for epithelium. The original embryonic differentiation of most epithelial tissues is determined by signals from the underlying mesenchyme.24–27 It is now apparent from studies that build on Bissell’s original concept of ‘dynamic reciprocity’28,29 that stromal elements exert a powerful influence over the progression of carcinogenesis in their associated epithelia, and that mutational events in adjacent stromal cells may be a critical determinant of the potentially invasive properties of overlying epithelium.30,31 Perhaps ‘carcinomas’ (defined as tumors originating in epithelium) are truly ‘carcinosarcomas’ (tumors originating both in epithelium and stroma). Genetics, reductionistically, defines the material world at a specific locus or loci (sequence of bases in DNA), but carcinoma is a transactional disease. Cancer does not arise because something is wrong with one specific molecule; the problem is that the functional relationships between a set of critical molecules have been disrupted so that they no longer control cell differentiation, proliferation, and motility. To use a musical metaphor, the instruments in the orchestra are not playing 366 NATURE CLINICAL PRACTICE ONCOLOGY

together to make harmonious music, although the instruments themselves might not be defective. It is much more difficult to restore order to such a disorganized system than to fix an individual instrument. In cells and tissues, there is no master conductor; the system is self-assembling, starting from conception. It makes much more sense not to let the system become disorganized in the first place. Thus, carcinoma is a network disease, and the network is not stable—in other words, the entire regulatory network in a given tissue, comprised of both the epithelium and its numerous associated stromal elements (fibroblasts, endothelial cells, macrophages, lymphocytes, neutrophils, all derived from mesenchyme) is dysfunctional. Physiologically, tissue needs to respond to injury (e.g. infection or mechanical damage) by remodeling itself, but this must be done in a networked, transactional manner that is functional and stable. The overemphasis on genetics is dangerous; genetic mutation alone will not cause carcinoma. For example, why does the genetically homozygous Rb–/– mutation result only in cancer in the eye, and not in cancer in other tissues of the body, in spite of the fact that all cells in the body can carry this mutation? Carcinogenesis by plastic film is another classic example of cancer starting with an epigenetic change; tumors are induced in rodents by the mere disruption of intercellular communication in connective tissue by insertion of inert plastic films.32 Thus, we need to develop a whole new biology to understand local tissue networks. We already have a fairly thorough understanding of regulation at a distance by endocrine mediators (the so-called endocrine language), but the paracrine language of tissue networks is still being deciphered. The overall topic of networks is considered later in this review. DICHOTOMY 4

‘New emphasis on transgenic mouse models with single gene disruption’ versus ‘classical carcinogenesis models that damage multiple genes’ The importance of the realization that human cancer is not simply a genetic disease, involving malfunction of only a few genes, cannot be overemphasized. Although there have been great advances in developing transgenic mouse models for specific organ sites, the use of chemical carcinogens to study the process of carcinogenesis remains an important tool. Why did chemicalinduced or radiation-induced models fall out of SPORN JULY 2006 VOL 3 NO 7

REVIEW www.nature.com/clinicalpractice/onc

fashion? This shift occurred concomitantly with the new emphasis on genetics. Chemical carcinogens or promoting agents, however, can mimic epigenetic damage in human carcinogenesis. The carcinogen aflatoxin causes both protein and DNA damage, both of which are important for carcinogenesis.33,34 Promoting agents (such as phorbol esters) are mechanistically linked to the epigenetic inflammatory process. Promotion and progression factors are of great importance in human carcinogenesis, and studies using animal model systems must take this into account. Because human carcinogenesis in most cases involves multigene damage, it is preferable to utilize animal models with multigene damage. Perhaps studies can begin with genetic models that target key genes, but then superimpose further random genetic and epigenetic damage caused by chemical carcinogens or radiation; this process would mimic the oxidative damage to DNA that is known to occur during human carcinogenesis.35 New approaches that employ both transgenic mice and the use of either chemical carcinogens or promoting agents are needed. We should develop animal models in which individual genetic susceptibility factors are modeled transgenically, while chemical carcinogens or radiation are used to add further multi-gene damage. The use of nonmutagenic promoting agents—whether they be hormonal or inflammatory— allows the study of epigenetic changes to be introduced into the experimental design. Such multifactorial designs provide a more representative model of human carcinogenesis. DICHOTOMY 5

‘New emphasis on monofunctional agents’ versus ‘need for multifunctional agents’ Monofunctional ‘magic bullets’ are the latest fashion, as Big Pharma spends more and more money on research in this area. Unquestionably, there have been great advances made in the development of new agents that target specific signal transduction pathways.36,37 Carcinomas are usually characterized by a multiplicity of genetic defects, however,35,38,39 and it is therefore unlikely that monofunctional agents will be curative for most advanced carcinomas. Because leukemias are genetically simpler than carcinomas,40 leukemogenesis is a poor model for the genetic complexity of carcinogenesis,3 and successes in the treatment of leukemias have not translated into equivalent treatment JULY 2006 VOL 3 NO 7 SPORN

successes for common carcinomas. In the treatment of leukemia the problem of resistance to new targeted agents has already been observed;41 unless monofunctional agents are used in combinations, especially for the treatment of carcinoma, they are unlikely to have major impact on the totality of cancer deaths. By contrast, the use of multifunctional agents for chemoprevention and as adjuncts to chemotherapy has been shown to be of major importance. The SERMs (selective estrogen receptor modulators) are probably the best example,7–9,42,43 (Box 1) but other agents such as retinoids and rexinoids that target nuclear receptors, are also of major significance.44–46 Likewise, there is increasing interest in using agents that target the STATs, Nrf2, NFκB, β-catenin, or c-Jun transcription factors, all of which are clearly multifunctional, for both prevention and therapy.47–54 Ligands that control the function of transcription factors are ideal multifunctional agents. Chromatin modifiers, such as histone deacetylase inhibitors55 and DNA-demethylating agents,56 are other examples of important new multifunctional agents that act epigenetically. With respect to the use of combinations of agents, major obstacles still remain, particularly with the US FDA and with Big Pharma. Why should the FDA forbid the evaluation of combinations of two or more new experimental drugs? Scientific wisdom flies in the face of such an intellectually indefensible position. If you were a nutritional biochemist 200 years ago and had discovered a new essential amino acid, such as leucine, it would not be possible to show that leucine was essential for life unless this new substance was assayed in the presence of all the other essential amino acids. Hence, the ideal design for evaluating both preventive and therapeutic agents could be one in which both monofunctional and multifunctional agents are used. DICHOTOMY 6

‘Reductionism’ versus ‘the whole can be greater than the sum of its parts’ Because reductionism has no context, it creates animate personae of its own. We have created our own homunculi, such as ‘oncogenes’ or ‘suppressor genes’. As Mintz and Fleischman noted many years ago,57 such genes are not intrinsically ‘onco’ or ‘suppressor’. Almost all of these genes have opposite actions in different contexts—they are just molecules that do what they do. The NATURE CLINICAL PRACTICE ONCOLOGY 367

REVIEW www.nature.com/clinicalpractice/onc

recent elucidation of the oncogenic activity of the so-called suppressor, p53, is an excellent example.58,59 Similar paradoxes are known for many other critical regulatory molecules, such as Myc,60 NFκB,61 and TGF-β.62,63 As emphasized in a recent review, many regulatory genes, including p21 (CDKN1A), RUNX1, NOTCH1, E2F1, RAS, VHL, and KLF4, behave as ‘Jekyll and Hydes’, as both tumor suppressors and oncogenes.64 The multiple actions of all of these regulators are all dependent on context. Perhaps our current terminology is a consequence of the limitations of the outlook at the time when these genes were being identified (i.e. in the context of the discovery of these genes, they were designated as ‘oncogenes’ or ‘tumor suppressor genes’, as a convenient way to describe their function and to form a consistency in terms of ‘scientific’ language). At that time, such simplicity might have been appropriate. Perhaps the time is now ripe to consider a more flexible and accurate nomenclature and language to define the multifunctional nature of many genes. Reductionist thinking is at the very core of the current approach to cancer and dominates present research efforts, most notably in the present emphasis on cancer as a genetic disease, and in the efforts to define specific genes as causes of cancer or as targets for cancer therapy. Although this approach has produced a huge amount of important new knowledge and even important new drugs, such as trastuzumab, it has failed to control the present epidemic of cancer deaths. There was great optimism 10 years ago that the discovery of ‘the breast cancer gene’ would have important implications for the control of this disease, but the therapeutic impact of the discovery of BRCA1 and BRCA2 mutations has been limited, other than to identify women who might be protected by prophylactic mastectomy. Why has reductionism been such a failure for control of cancer? Perhaps reductionism, as a way of thinking, cannot encompass the dynamic nature of the biology of cells. Cells represent an ever-changing, fluid interaction of regulatory networks, mechanistically coupled to changes in their immediate environment. Reductionism is an ideal way to analyze static states, or even simple linear events, but the reductionist approach is inadequate for conceptualization of networks that are highly interactive and that change markedly and rapidly in response to environmental perturbation. In short, reductionism cannot deal with context, especially with rapid changes in context; 368 NATURE CLINICAL PRACTICE ONCOLOGY

reductionism doesn’t address the ‘big picture’. Moreover, as we learn more about the anatomy and physiology of the transcriptional apparatus, it becomes increasingly apparent that the activity of the ‘transcriptome’ is determined not just by DNA, but by an exceptionally complex set of repressors, co-repressors, activators, and co-activators,65 as well as by the complex epigenetic ‘histone language’ or ‘histone code’,66,67 (Box 1) all of which are in dynamic interaction with what happens metabolically in the cell. The whole is indeed much greater than its component parts. We need to abandon concepts that describe regulatory molecules as either ‘oncogenes’ or ‘suppressor genes’ and realize that they are bifunctional switches in regulatory networks. We need to consider that the same molecule, such as IκB kinase (a key regulator of the activity of the transcription factor, NFκB), can convey either proaptotic or anti-apoptotic messages, depending on the timing and mechanism of activation,68 and that NFκB is part of an oscillatory network.69 This network approach to studying cell regulation is now a powerful new tool for the study of carcinogenesis and its prevention. Significant technical and conceptual advances have already been made, not only for study of NFκB, but also in the elucidation of the complexities of many other regulatory networks, such as TGF-β/Smad signaling,70,71 or Wnt signaling.72 The numbers of interacting molecules in these networks are far beyond any simplistic reductionism; for example, 221 molecular associations were identified in the TNF-α/NFκB signal transduction pathway,73 755 interactions in a TGF-β/Smad study,71 and 3,186 mostly novel interactions among 1,705 proteins (including the Wnt pathway) in a study using yeast two-hybrid interaction mating.72 The new term, ‘interactome’, has been coined to conceptualize this extremely interactive nature of the proteome.74 Studies such as these will have important implications for understanding the cause and prevention of many other chronic degenerative diseases beyond cancer.19 The ‘histone code’ that regulates metabolic modulation of the transcriptional activity of specific genes is yet another manifestation of this interactive, epigenetic nature of the gene–environment relationships.66,67 For therapy of cancer, beyond this interactive nature of the proteome, we now have the knowledge that carcinoma in humans is rarely the result of single gene damage; indeed, there SPORN JULY 2006 VOL 3 NO 7

REVIEW www.nature.com/clinicalpractice/onc

may be billions of mutations in the totality of the cells that comprise a tumor.39 In a reductionistic analysis, which gene should we single out as a unique molecular target? A recent study of protein kinase genes in a set of 26 primary human lung cancers has identified 188 distinct somatic mutations in 141 genes, and suggests that more than 40 of these mutations (in genes that encode protein kinases alone) may have contributed to oncogenesis.75 The authors stress the need for developing multiple kinase inhibitors, if practical therapy is to be achieved. For the ‘blockbuster’ mentality in Big Pharma, this suggestion has disastrous implications. DICHOTOMY 7

‘Hypothesis-driven research’ versus ‘the need for observational research’ Why is this topic even considered in this article? One answer is the way we look at things, and the way in which we conceptualize our general ideas, ultimately, determine what we do in the laboratory, or in clinical research. At the present time, the necessity of doing hypothesis-based research is almost a sacred canon in biomedical research, both basic and clinical. Although hypothesisbased research has produced many useful results, there are also many examples of extremely important developments in biology and medicine that have not resulted from such an approach. For example, what was Darwin’s hypothesis when he started his voyage on the HMS Beagle? Indeed, the captain of the HMS Beagle, Robert Fitzroy, reputedly recruited Darwin as the ship’s biologist to collect enough exotic data to prove the existence of a Creator.76 Instead, Darwin collected thousands of specimens, eventually to formulate a new hypothesis of the Origin of Species. Darwin was a very astute observer of natural phenomena, and we still need to train young scientists to have respect for this approach to research. In the US, grant applications that are not hypothesis-based are generally dismissed without further review, even if they might have significant potential to make important new observations. Observational research is held in very low repute. Until we have totally solved the cancer problem, we should encourage both approaches. AN ATTEMPT AT A SYNTHESIS OF THE PRECEDING DICHOTOMIES

The magnitude of human suffering from cancer and the immense cost of this disease to society as a whole compel us to try to formulate a new JULY 2006 VOL 3 NO 7 SPORN

synthesis that will build on important discoveries and approaches from the past, but will also allow novel and unorthodox viewpoints to be added. It is clear that more of the same will not do. Cancer will never be controlled by an exclusively genomic outlook, no matter how important and essential this outlook may be. We should view cancer from the perspective of carcinogenesis,77 and emphasize the importance of understanding the genesis of cancer from its very onset. This approach does not focus on the single cell as the fundamental unit of ‘transformation’ and the inception of cancer. Instead, it focuses on a defective transactional relationship between the various cells that comprise the tissues in which carcinoma arises. A set of unifying themes then arise from the answers to the following questions: How do cells and tissues respond to injury? What are common processes involved in injury of cells and tissues? What are the intracellular and extracellular networks that mediate these responses? What are common responses to tissue injury? When are these responses maladaptive? What can we do to control these maladaptive responses? Carcinogenesis as a maladaptive response to tissue injury is an old idea, particularly the idea that carcinogenesis is an aberration of the wound healing response.78,79 With modern genomic and proteomic techniques, we can begin to specify the multiple defects in DNA and proteins that are involved in aberrant responses to tissue injury. Moreover, we now have a better understanding of the complex repair networks, whether they are DNA or protein repair mechanisms that respond to injury. The ultimate way to deal with carcinogenesis, however, is to suppress this disease in its earliest stages, before extensive DNA and protein damage occur. The activation of the transcription factor Nrf2 and its induction of ‘phase 2’ enzymes, which are responsible for ‘electrophile counterattack’ (a concept first proposed by Talalay and co-workers80), protects cells from oxidative or other mutagenic or epigenetic damage. This pathway provides a powerful focus for devising new chemopreventive strategies.47,54,81,82 It has taken millions of years of evolution to organize the components of cells into adaptive units in functional networks. Indeed, it has been estimated that it took more evolutionary time for prokaryotic, non-nucleated life to evolve into eukaryotic, nucleated life than it took for the development of life itself. There is tremendous NATURE CLINICAL PRACTICE ONCOLOGY 369

REVIEW www.nature.com/clinicalpractice/onc

adaptive redundancy in the manner in which normal or premalignant cells respond to environmental stress. Metastatic, invasive carcinoma, on the other hand, is a ‘maladaptive box’, and it would appear to be much more difficult to exit from this box than never to have allowed entry in the first place. Our present priorities in cancer research are distorted, with not enough effort devoted to prevention, rather than treatment, although a strong emphasis on treatment is still essential for those cancers that are not preventable. When prevention of early disease fails, we must adopt new approaches to prevent metastasis, which is usually the ultimate killer. A constructive synthesis of all the viewpoints discussed here is needed, with particular emphasis on strategies presently considered unorthodox. We need new emphasis on epigenetics.18 We need further development of important new multifunctional drugs, such as the rexinoids, the deltanoids, and chromatin modifiers such as histone deacetylase inhibitors and DNA-demethylating agents.83 Development of newer agents that have unique mechanisms of action are required, such as synthetic triterpenoids that can target entire regulatory networks, including the thiol networks that control the oxidation/reduction potential of premalignant or malignant cells.84–86 We need the addition of not-for-profit pharmaceutical efforts to provide the new drugs that Big Pharma might not wish to make.87,88 We need a more enlightened effort on the part of regulatory agencies to allow clinical trials of combinations of new drugs. In particular, we need to stop the regulatory and legalistic nonsense that prohibits simultaneous testing of two new experimental drugs in combination. A less rigid approach in our outlook as to how creativity gets translated into important practical new preventive or therapeutic interventions should be used. We need to accept the limits of reductionism, which can be intensely useful and practical for analysis of short-term changes in linear systems that are not subject to rapid environmental perturbation, and realize the need to develop new circular, systemsoriented approaches to more complex problems. Importantly, prevention activities should be encouraged by providers of health insurance. Society can no longer afford the terrible costs of treating only end-stage disease. The ultimate question we need to ask is: When will we become truly serious about preventing cancer, and when will we allocate the resources necessary 370 NATURE CLINICAL PRACTICE ONCOLOGY

to prevent cancer in a practical manner? Finally, we must assure that we create a more nurturing and financially supportive scientific environment for young investigators who are the future of cancer research. At present we are strangling our young; the median age for a first NIH R01 grant has risen to above 40 years. Most of all, we must foster a new outlook that will nurture the creativity necessary for the needed new synthesis in cancer research. Perhaps we can obtain some clues by examining the genesis of the careers of creative thinkers outside of science. One example is Walt Whitman, who had a formative experience early in his career as a poet, when he served during the American Civil War as a hospital volunteer, caring for the hundreds, even thousands, of mortally wounded soldiers who were taken to Washington, DC, for their terminal care. Writing about Whitman’s experiences, the critic Paul Zweig notes, “As a poet, [Whitman] had the ability to lose himself in what he saw, to see it from his own perspective. Now as a hospital nurse, he seemed to feel the boys’ needs as if they were his own.”89 Whitman had earlier expressed such thoughts in Leaves of Grass, as follows: “I do not ask the wounded person how he feels, I myself become the wounded person.” Ultimately, our current problem may be our flawed perspective. The cancer problem should be viewed from the perspective of the needs and feelings of the prospective or actual cancer patient. There is an overriding need for scientists, physicians, the hospital industry, the pharmaceutical industry, and the regulatory agencies all to connect with the needs of actual or prospective patients as if they were their own personal needs. This is a challenge, but if it can be accomplished, it will focus not just on cold statistics such as 5-year survival rates, but on the quality of life of the actual or prospective patient. In such an analysis, a preventive approach provides an ideal synthesis of all the divergent dichotomies discussed here. But how long must we wait, how much will our patience continue to be taxed, how much suffering must patients and their families endure, before a true spirit of cooperativity between all interested parties provides the new synthesis that is so desperately needed? A special change in cooperativity among the companies in Big Pharma, and between Big Pharma and Big University Research, is drastically needed. As noted recently by Clifton Leaf, “[As] universities SPORN JULY 2006 VOL 3 NO 7

REVIEW www.nature.com/clinicalpractice/onc

Safety Utility Economy Esthetics

Figure 1 Target illustrating the goals of the eminent Swiss engineer, Christian Menn, in his design of a new bridge. Esthetic considerations for Menn are nonscientific and largely subjective. “Economy and elegance are achieved through nonscientific means. They depend almost entirely on the creativity of the engineer.” In biomedical terms, we may equate esthetics with Quality of Life. Adapted from Reference 94, which provides citations to the original writings and drawings of Christian Menn.

have evolved from public trusts into something closer to venture capital firms, what used to be a scientific community of free and open debate now often seems like a litigious scrum of datahoarding and suspicion.”90 Intellectual property issues are strangling cooperation that would allow testing of important combinations of new experimental drugs. CONCLUSION

We clearly are not winning the war on cancer, as the incidence figures have shown no significant decrease.2 It is time to stop focusing on ‘The War on Cancer’ and start focusing on the Quality of Life. Too many people still must endure the ordeal of cytotoxic chemotherapy (admittedly often lifesaving) for their end-stage disease. Although we have achieved spectacular cures in certain situations, incidence ultimately drives mortality. We must take a hard look at the “byzantine world of American health care, in which the real profit is made not by controlling chronic diseases... but by treating their many complications”.13 Although this quote comes from a recent article on diabetes, the current practice of oncology is a perfect example of this tragic situation, which leads to so much misery and ultimately to a poor

quality of life for those who suffer. We need a new, holistic approach to prevention that will include both chemoprevention and a better lifestyle, with cooperative efforts from people in both areas, directed toward a better quality of life for the patient. The exceptionally high costs of heroic end-stage care cannot be sustained forever by our health-care systems.91–93 Bevacizumab at $100,000 per patient per year is both unconscionable and economically unsustainable.92 The old paradigms no longer suffice. We must do something new and imaginative. Ultimately, we need a fusion of Art and Science for the benefit of both the prospective patient and the real, suffering, cancer victim. The necessity of fusing an esthetic and artistic orientation with our most materialistic scientific activities, to provide these activities with ultimate meaning and human connectivity, is a recurrent theme in the works of even the most applied practitioners of technology (see Figure 1).94,95 We need a new synthesis. KEY POINTS ■

Prevention of early-stage disease should be emphasized rather than directing efforts only at treating end-stage disease



Epigenetics is equally important as genetics, and a greater appreciation of epigenetics is pivotal to our understanding the process of carcinogenesis



Multifactorial animal model experiments that utilize not only transgenic mouse models, but also classical carcinogenesis models that disrupt multiple genes, are more representative models of human carcinogenesis



The combination of monofunctional and multifunctional agents is needed for developing both preventive and therapeutic strategies



A reductionist approach does not encompass the complexity of carcinogenesis or its context; instead it should be recognized that cellular networks are highly interactive systems that respond rapidly to environmental perturbation



Observational and hypothesis-driven research are both required to encompass a holistic approach to our understanding of cancer

References 1 (online December 2005) Cancer Trends Progress Report—2005 Update [http://progressreport.cancer. gov] (accessed 17 May 2006) 2 Jemal A et al. (2006) Cancer statistics, 2006. CA Cancer J Clin 56: 106–130 3 Sporn MB (1996) The war on cancer. Lancet 347: 1377–1381

ncponc_2005_305f1.eps JULY 2006 VOL 3 NO 7 SPORN

NATURE CLINICAL PRACTICE ONCOLOGY 371

REVIEW www.nature.com/clinicalpractice/onc

4

5 6

7

8

9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

27

28

[No authors listed] (1999) Prevention of cancer in the next millennium: Report of the Chemoprevention Working Group to the American Association for Cancer Research. Cancer Res 59: 4743–4758 Sporn MB and Liby KT (2005) Cancer chemoprevention: scientific promise, clinical uncertainty. Nat Clin Pract Oncol 2: 518–525 O’Shaughnessy JA et al. (2002) Treatment and prevention of intraepithelial neoplasia: an important target for accelerated new agent development. Clin Cancer Res 8: 314–346 Fisher B et al. (1998) Tamoxifen for prevention of breast cancer: report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. J Natl Cancer Inst 90: 1371–1388 Martino S et al. (2004) Continuing outcomes relevant to Evista: breast cancer incidence in postmenopausal osteoporotic women in a randomized trial of raloxifene. J Natl Cancer Inst 96: 1751–1761 Fisher B et al. (2005) Tamoxifen for the prevention of breast cancer: current status of the National Surgical Adjuvant Breast and Bowel Project P-1 study. J Natl Cancer Inst 97: 1652–1662 Hruban RH et al. (2005) Identification and analysis of precursors to invasive pancreatic cancer. Methods Mol Med 103: 1–13 Maitra A et al. (2005) Precursors to invasive pancreatic cancer. Adv Anat Pathol 12: 81–91 Hingorani SR et al. (2003) Preinvasive and invasive ductal pancreatic cancer and its early detection in the mouse. Cancer Cell 4: 437–450 Urbina I (11 January 2006) In the treatment of diabetes, success often does not pay. The New York Times 155: A1, A26–A27 Feinberg AP and Tycko B (2004) The history of cancer epigenetics. Nat Rev Cancer 4: 143–153 Lund AH and van Lohuizen M (2004) Epigenetics and cancer. Genes Dev 18: 2315–2335 Laird PW (2005) Cancer epigenetics. Hum Mol Genet 14 (Spec No 1): R65–R76 Jones PA (2005) Overview of cancer epigenetics. Semin Hematol 42 (Suppl 2): S3–S8 Rauscher FJ III (2005) It is time for a Human Epigenome Project. Cancer Res 65: 11229 Nathan C (2002) Points of control in inflammation. Nature 420: 846–852 Coussens LM and Werb Z (2002) Inflammation and cancer. Nature 420: 860–867 Balkwill F et al. (2005) Smoldering and polarized inflammation in the initiation and promotion of malignant disease. Cancer Cell 7: 211–217 de Visser KE et al. (2006) Paradoxical roles of the immune system during cancer development. Nat Rev Cancer 6: 24–37 Clark WH Jr (1995) The nature of cancer: morphogenesis and progressive (self)-disorganization in neoplastic development and progression. Acta Oncol 34: 3–21 Grobstein C (1956) Inductive tissue interaction in development. Adv Cancer Res 4: 187–236 Grobstein C (1953) Inductive epitheliomesenchymal interaction in cultured organ rudiments of the mouse. Science 118: 52–55 Cunha GR et al. (1980) Induction of nuclear androgenbinding sites in epithelium of the embryonic urinary bladder by mesenchyme of the urogenital sinus of embryonic mice. Endocrinology 107: 1767–1770 Sakakura T et al. (1976) Mesenchyme-dependent morphogenesis and epithelium-specific cytodifferentiation in mouse mammary gland. Science 194: 1439–1441 Bissell MJ and Barcellos-Hoff MH (1987) The influence of extracellular matrix on gene expression: is structure the message? J Cell Sci Suppl 8: S327–S343

372 NATURE CLINICAL PRACTICE ONCOLOGY

29 Bissell MJ et al. (1982) How does the extracellular matrix direct gene expression? J Theor Biol 99: 31–68 30 Bhowmick NA et al. (2004) Stromal fibroblasts in cancer initiation and progression. Nature 432: 332–337 31 Bissell MJ et al. (2005) Microenvironmental regulators of tissue structure and function also regulate tumor induction and progression: the role of extracellular matrix and its degrading enzymes. Cold Spring Harb Symp Quant Biol 70: 1–14 32 Bates RR and Klein M (1966) Importance of a smooth surface in carcinogenesis by plastic film. J Natl Cancer Inst 37: 145–151 33 Smela ME et al. (2001) The chemistry and biology of aflatoxin B1: from mutational spectrometry to carcinogenesis. Carcinogenesis 22: 535–545 34 Eaton DL and Gallagher EP (1994) Mechanisms of aflatoxin carcinogenesis. Annu Rev Pharmacol Toxicol 34: 135–172 35 Marnett LJ et al. (2003) Endogenous generation of reactive oxidants and electrophiles and their reactions with DNA and protein. J Clin Invest 111: 583–593 36 Deininger M et al. (2005) The development of imatinib as a therapeutic agent for chronic myeloid leukemia. Blood 105: 2640–2653 37 Tripathy D (2005) Targeted therapies in breast cancer. Breast J 11 (Suppl 1): S30–S35 38 Mitelman F et al. (1994) Catalog of Chromosome Aberrations in Cancer, edn 5. New York: Wiley-Liss 39 Loeb LA et al. (2003) Multiple mutations and cancer. Proc Natl Acad Sci USA 100: 776–781 40 Rabbitts TH (1994) Chromosomal translocations in human cancer. Nature 372: 143–149 41 Gorre ME and Sawyers CL (2002) Molecular mechanisms of resistance to STI571 in chronic myeloid leukemia. Curr Opin Hematol 9: 303–307 42 Jordan VC (2006) Tamoxifen (ICI46,474) as a targeted therapy to treat and prevent breast cancer. Br J Pharmacol 147 (Suppl 1): S269–S276 43 Suh N et al. (2001) Arzoxifene, a new selective estrogen receptor modulator for chemoprevention of experimental breast cancer. Cancer Res 61: 8412–8415 44 Freemantle SJ et al. (2003) Retinoids in cancer therapy and chemoprevention: promise meets resistance. Oncogene 22: 7305–7315 45 Wu K et al. (2002) The retinoid X receptor-selective retinoid, LGD1069, prevents the development of estrogen receptor-negative mammary tumors in transgenic mice. Cancer Res 62: 6376–6380 46 Suh N et al. (2002) Prevention and treatment of experimental breast cancer with the combination of a new selective estrogen receptor modulator, arzoxifene, and a new rexinoid, LG 100268. Clin Cancer Res 8: 3270–3275 47 Yu X and Kensler T (2005) Nrf2 as a target for cancer chemoprevention. Mutat Res 591: 93–102 48 Darnell JE (2005) Validating Stat3 in cancer therapy. Nat Med 11: 595–596 49 Turkson J (2004) STAT proteins as novel targets for cancer drug discovery. Expert Opin Ther Targets 8: 409–422 50 Darnell JE Jr (2002) Transcription factors as targets for cancer therapy. Nat Rev Cancer 2: 740–749 51 Karin M and Greten FR (2005) NF-kappaB: linking inflammation and immunity to cancer development and progression. Nat Rev Immunol 5: 749–759 52 Yates MS et al. (2006) Potent protection against aflatoxin-induced tumorigenesis through induction of nrf2-regulated pathways by the triterpenoid 1-[2-cyano-3-,12-dioxooleana-1,9(11)-dien-28oyl]imidazole. Cancer Res 66: 2488–2494 53 Giudice A and Montella M (2006) Activation of the Nrf2-ARE signaling pathway: a promising strategy in cancer prevention. Bioessays 28: 169–181

SPORN JULY 2006 VOL 3 NO 7

REVIEW www.nature.com/clinicalpractice/onc

54 Jeong WS et al. (2006) Nrf2: a potential molecular target for cancer chemoprevention by natural compounds. Antioxid Redox Signal 8: 99–106 55 Kelly WK and Marks PA (2005) Drug insight: Histone deacetylase inhibitors—development of the new targeted anticancer agent suberoylanilide hydroxamic acid. Nat Clin Pract Oncol 2: 150–157 56 Marquez VE et al. (2005) Zebularine: a unique molecule for an epigenetically based strategy in cancer chemotherapy. The magic of its chemistry and biology. Nucleosides Nucleotides Nucleic Acids 24: 305–318 57 Mintz B and Fleischman RA (1981) Teratocarcinomas and other neoplasms as developmental defects in gene expression. Adv Cancer Res 34: 211–278 58 Olive KP et al. (2004) Mutant p53 gain of function in two mouse models of Li-Fraumeni syndrome. Cell 119: 847–860 59 Lang GA et al. (2004) Gain of function of a p53 hot spot mutation in a mouse model of Li-Fraumeni syndrome. Cell 119: 861–872 60 Nilsson JA and Cleveland JL (2003) Myc pathways provoking cell suicide and cancer. Oncogene 22: 9007–9021 61 Aggarwal BB and Takada Y (2005) Pro-apototic and anti-apoptotic effects of tumor necrosis factor in tumor cells. Role of nuclear transcription factor NF-κB. Cancer Treat Res 126: 103–127 62 Wakefield LM and Roberts AB (2002) TGF-β signaling: positive and negative effects on tumorigenesis. Curr Opin Genet Dev 12: 22–29 63 Siegel PM and Massague J (2003) Cytostatic and apoptotic actions of TGF-β in homeostasis and cancer. Nat Rev Cancer 3: 807–821 64 Rowland BD and Peeper DS (2006) KLF4, p21 and context-dependent opposing forces in cancer. Nat Rev Cancer 6: 11–23 65 Smith CL and O’Malley BW (2004) Coregulator function: a key to understanding tissue specificity of selective receptor modulators. Endocr Rev 25: 45–71 66 Strahl BD and Allis CD (2000) The language of covalent histone modifications. Nature 403: 41–45 67 Fischle W et al. (2003) Binary switches and modification cassettes in histone biology and beyond. Nature 425: 475–479 68 Janes KA et al. (2005) A systems model of signaling identifies a molecular basis set for cytokine-induced apoptosis. Science 310: 1646–1653 69 Nelson DE et al. (2004) Oscillations in NF-κB signaling control the dynamics of gene expression. Science 306: 704–708 70 Barrios-Rodiles M et al. (2005) High-throughput mapping of a dynamic signaling network in mammalian cells. Science 307: 1621–1625 71 Colland F et al. (2004) Functional proteomics mapping of a human signaling pathway. Genome Res 14: 1324–1332 72 Stelzl U et al. (2005) A human protein-protein interaction network: a resource for annotating the proteome. Cell 122: 957–968 73 Bouwmeester T et al. (2004) A physical and functional map of the human TNF-α/NF-κB signal transduction pathway. Nat Cell Biol 6: 97–105 74 Sanchez C et al. (1999) Grasping at molecular interactions and genetic networks in Drosophila

JULY 2006 VOL 3 NO 7 SPORN

75 76 77 78 79 80 81

82 83 84

85

86 87 88 89 90 91 92 93 94 95

melanogaster using FlyNets, an Internet database. Nucleic Acids Res 27: 89–94 Davies H et al. (2005) Somatic mutations of the protein kinase gene family in human lung cancer. Cancer Res 65: 7591–7595 Moorehead A (1969) Darwin and the Beagle. New York: Harper & Row Sporn MB (1991) Carcinogenesis and cancer: different perspectives on the same disease. Cancer Res 51: 6215–6218 Haddow A (1972) Molecular repair, wound healing, and carcinogenesis: tumor production a possible overhealing? Adv Cancer Res 16: 181–234 Dvorak HF (1986) Tumors: wounds that do not heal. Similarities between tumor stroma generation and wound healing. N Engl J Med 315: 1650–1659 Prestera T et al. (1993) The electrophile counterattack response: protection against neoplasia and toxicity. Adv Enzyme Regul 33: 281–296 Holtzclaw WD et al. (2004) Protection against electrophile and oxidative stress by induction of phase 2 genes: the quest for the elusive sensor that responds to inducers. Adv Enzyme Regul 44: 335–367 Surh YJ (2003) Cancer chemoprevention with dietary phytochemicals. Nat Rev Cancer 3: 768–780 Sporn MB and Suh N (2002) Chemoprevention: an essential approach to controlling cancer. Nat Rev Cancer 2: 537–543 Dinkova-Kostova AT et al. (2005) Extremely potent triterpenoid inducers of the phase 2 response: correlations of protection against oxidant and inflammatory stress. Proc Natl Acad Sci USA 102: 4584–4589 Liby K et al. (2005) The synthetic triterpenoids, CDDO and CDDO-imidazolide, are potent inducers of heme oxygenase-1 and Nrf2/ARE signaling. Cancer Res 65: 4789–4798 Doroshow JH (2006) Redox modulation of chemotherapy-induced tumor cell killing and normal tissue toxicity. J Natl Cancer Inst 98: 223–225 Nathan C and Goldberg FM (2005) Outlook: the profit problem in antibiotic R&D. Nat Rev Drug Discov 4: 887–891 Nathan C (2004) Antibiotics at the crossroads. Nature 431: 899–902 Zweig P (1985) The wound-dresser. In Walt Whitman, 143–157 (Ed Bloom H) New York: Chelsea House Publishers Leaf C (2005) The law of unintended consequences. In Fortune, September 19 2005, 250–268 Hutchinson L and DeVita VT Jr (2005) Herceptin: HERalding a new era in breast cancer care but at what cost? Nat Clin Pract Oncol 2: 595 Berenson A (15 February 2006) A cancer drug shows promise, at a price that many can’t pay. The New York Times 155: A1, C2 Lyall S (16 February 2006) British clinic is allowed to deny medicine. The New York Times 155: A6 Billington DP (2003) The Art of Structural Design: A Swiss Legacy. New Haven: Princeton University Art Museum and Yale University Press. McCarter R (2005) Louis Kahn. London and New York: Phaidon Press.

Acknowledgments I thank K Liby, L Wakefield, A Roberts, C Nathan, C Leaf, L Hutchinson, K McGaughy, and CDS for helpful comments and M Padgett for expert editorial assistance. This article is dedicated to C Everett Koop, true hero and outspoken champion of the cause of prevention of disease. Work supported by NIH grants, National Foundation for Cancer Research, members of the Dartmouth College Class of 1934, and Reata Pharmaceuticals.

Competing interests The author declared he has no competing interests.

NATURE CLINICAL PRACTICE ONCOLOGY 373

Dichotomies in cancer research: some suggestions for ...

add further multi-gene damage. The use of non- .... unifying themes then arise from the answers to the following .... is a recurrent theme in the works of even the.

243KB Sizes 0 Downloads 162 Views

Recommend Documents

Some Potential Areas for Future Research - Research at Google
Proportion of costs for energy will continue to grow, since. Moore's law keeps ... Challenge: Are there alternative designs that would .... semi-structured sources.

[RAED] PDF Current Strategies in Cancer Gene Therapy (Recent Results in Cancer Research)
[RAED] PDF Current Strategies in Cancer Gene Therapy (Recent Results in Cancer Research)

Most C6 Cells Are Cancer Stem Cells: Evidence ... - Cancer Research
Apr 15, 2007 - cell line using clonal and population analyses, rather than isolating .... Phone: 86-571-87784606; Fax: 86-571-87783757; E-mail: [email protected] 163.com. ..... multipotentiality or a cell's ability to give rise to multiple cell types.

Automatic Annotation Suggestions for Audiovisual ...
visualize semantically related documents. 1 Context. The Netherlands ... analogue towards digital data. This effectively ... adaptation for 'audiovisual' catalogue data of the FRBR data model 1 which has ...... Chapter Designing Adaptive Infor-.

Induction of Autophagic Cell Death in Malignant ... - Cancer Research
May 1, 2003 - analyzed by Cell Quest software (Becton Dickinson). Dead cells ... Cells were ana- ... from (Becton Dickinson) using CellQuest software. EM.

Suggestions for Parenting Grown-Up Kids
Committed to Excellence in Communicating Biblical Truth and Its Application ... The theme of harmony winds through Ephesians 4 and flows into six practical.

Hugo Franco_Guidelines and Suggestions for Balloon Gondola ...
There was a problem previewing this document. Retrying. ... to open or edit this item. Hugo Franco_Guidelines and Suggestions for Balloon Gondola Design.pdf.

[RAED] PDF Frontiers in Cancer Research: Evolutionary Foundations
[RAED] PDF Frontiers in Cancer Research: Evolutionary Foundations