Opportunities and Challenges of Chemistry in Cancer Research A

Opportunities and Challenges of Chemistry in Cancer Research: A Call to Action From Leaders in the Field __________________________________ Thursday, February 10 and Friday, February 11, 2005 Philadelphia, PA __________________________________ Executive Summary Chemistry offers great opportunities for the detection, prevention, and treatment of cancer and will continue to be a critical component of biomedical research in the post-genomic era. The sequencing and annotation of the human genome and those of many model organisms represents a revolutionary advance in research and translation. Increasingly, attention will be focused on individual macromolecules and on small molecules, both naturally occurring and synthetic, and their functions in a cellular context. Chemistry will provide tools for determining molecular structure, defining molecular function, detecting alterations associated with cancer, and modulating function with therapeutic or preventive strategies. It is likely that the impact of chemistry in cancer research will be as great as that of genetics and molecular biology in the pre-genomic era. The Chemistry in Cancer Research Task Force was created in 2001 to provide a focus for the growth of chemistry within AACR; it evolved into a Working Group in 2004. An important early step in the evolution of the Chemistry in Cancer Research Working Group (CICR) was to survey the landscape of opportunities and the challenges to increasing the profile of chemistry within the AACR and to provide a report to the Board of Directors with recommendations for action. This was the focus of a Think Tank held in February 2005 in Philadelphia that brought together members of the CICR Steering Committee and other experts for a one-and-a-half day meeting. Chemists working in industry, academia, and government were broadly represented at the Think Tank. Separate sessions were held in Proteomics, Drug Discovery in Therapy and Prevention, Computational Biology, Structural Biology, Chemical Biology of Cancer, and Biomarkers/Analytical Chemistry. The full report of the Think Tank, with specific recommendations for each scientific area discussed, is attached. The action items for consideration by the Board are highlighted below: A key role of any society is to provide accurate, up-to-date information to its members. AACR has a critical role in educating its membership about the opportunities for chemistry in cancer research. Through symposia, educational sessions, special conferences, publications, etc., AACR can not only highlight the importance of chemistry and chemical approaches to its members but also provide a forum where chemists can assimilate the key concepts of complementary disciplines such as cancer biology, genetics, and clinical research. This should stimulate rapid and exciting translational advances and attract new members to the AACR. 1 Collaboration is critical to contemporary research and researchers are rapidly and productively adapting to this new model. It is time for professional societies to do the same. AACR is urged to pursue collaboration with organizations such as the American Chemical Society and the American Society of Mass Spectrometry to create exciting interdisciplinary programs of education, training, and research. AACR’s flagship journal, Cancer Research, does not have adequate representation at the level of a senior editor or members of the editorial board in chemical biology although the journal lists this area among its sub-headings. This results in excellent papers with chemical content being inadequately reviewed or labeled as too specialized for publication in the journal. A Senior Editor in Chemical Biology should be named and additional Associate Editors appointed with expertise in various areas of chemistry. Distinguished AACR members with strong research programs in chemistry should be nominated for election to the Board of Directors. Recruiting young investigators is critical to the future of cancer research. The Annual Meeting is an exciting, if overwhelming, scientific venue at which to expose young scientists to opportunities in cancer research. Participation in the Annual Meeting can be supplemented with participation in Special Conferences, which offer a more intimate, and focused scientific experience. Scholar-in-Training Awards can provide travel grants to young investigators and mentoring sessions (e.g., funding opportunities for chemists) can provide critical advice that can help advance their careers. CICR and AACR should consider adopting mechanisms for recruitment and mentoring developed by other successful AACR groups such as Women in Cancer Research and the Molecular Epidemiology Working Group. CICR should consider partnering with these groups in joint mentoring sessions where appropriate. Scientific rigor is the sine qua non for advancing cancer research. Several areas of chemistry present challenges to this maxim when applied to cancer because of the complexity of information generated and the absence of standards by which data sets can be compared and related. Such fields include bioinformatics, proteomics, high throughput screening, and biomarker research. AACR can play an important role in emphasizing rigor and accuracy in its programs and publications and in educating its membership about the pitfalls of research in areas in which a true understanding of the methodology is beyond the capabilities of most scientists. A related issue is full disclosure of information, in particular chemical structure, at AACRsponsored meetings and in AACR journals. A disturbing trend is evolving in which papers describing exciting activity of new compounds are presented or published without disclosure of their structure. Without structure, research cannot be reproduced so it is not up to the standards of contemporary science. This is a problem that AACR should address aggressively. The CICR Steering Committee has assembled a subcommittee to make recommendations to the Board of Directors on this important issue. Improving the image of chemistry in the public arena was echoed as a theme throughout the Think Tank. This is critical to understanding the implications of research presented to the public and to building consensus for funding. Chemists must find effective ways to describe their work and develop new nomenclature for emerging areas of research in cancer. This may help them to be better understood and appreciated by public audiences as well as by colleagues in other scientific disciplines. 2 Introduction A Think Tank on Chemistry in Cancer Research was held February 10-11, 2005 in Philadelphia, hosted by the Chemistry in Cancer Research (CICR) Working Group of the American Association for Cancer Research. In his introductory remarks, CICR Chairperson and Vanderbilt University Professor Lawrence J. Marnett, Ph.D., noted that the goal of the Working Group is to develop within AACR a dynamic intellectual atmosphere for chemists working in cancer research to enhance progress in the fight against cancer by interacting with colleagues, learning of advances in related fields, reporting the results of their research, generating new research ideas, creating funding opportunities, and impacting funding agencies. Marnett, whose own research focuses on DNA damage by endogenously generated DNA-damaging agents and the structure and function of COX-2 in therapy and prevention, set the stage for discussion by reviewing recent trends in chemistry that create the opportunities and challenges that must be addressed by the Working Group: Technological advances have changed the face of research for all scientists, but especially chemists. It is now possible to rapidly synthesize large numbers of molecules to use as libraries to interrogate biological function. High throughput methodologies, which were invented in industry, have migrated to academic research centers. This provides academic scientists the infrastructure for high throughput screening to identify molecules with novel biological function, high throughput crystallography to determine molecular structure, and high throughput sample preparation and analysis for proteomics. Complementary advances in the strategies, reagents, and instrumentation for organic synthesis, NMR spectroscopy, mass spectrometry, X-ray crystallography, molecular imaging, and computation have created unprecedented opportunities for discovery and translation. Increasingly, these techniques are being applied in multi-investigator efforts that provide a mechanism to rapidly generate and test hypotheses in a fashion that addresses critical clinical issues. These multi-disciplinary activities are at the heart of the NIH Roadmap, which is an effort to encourage scientists to consider the translational implications of their research discoveries. Chemistry is a central component of this effort. The pharmaceutical sector, a dynamic, powerful, and competitive force in modern medicine, has experienced a decline in new drug introductions despite increased expenditures for research. The safety barrier for new drugs continues to increase, which forces companies to evaluate efficacy and adverse effects earlier in the drug discovery process. This requires an increasingly sophisticated understanding of biology and molecular mechanism, with expertise and resource commitment that is often beyond the capabilities of most individual companies or institutions. This has provided the driving force for collaborative interactions between pharmaceutical laboratories, with strengths in compound discovery, optimization, and scaleup, and academic laboratories, with expertise in molecular target identification, cancer cell biology, and human experimentation. Funding for chemistry research is changing at the National Institutes of Health. The National Cancer Institute is currently funding to the 16th percentile with projections dropping to the 12th percentile in 2006. The composition of NIH study sections also is changing, with the vast majority focused on diseases and translational research. This provides an opportunity to focus chemistry in a disease-based context but increases the difficulty of funding the basic advances in chemistry necessary to provide novel conceptual and technological approaches in the future. AACR faces increased competition for members and programmatic themes from other associations dedicated to cancer. Building a strong program in chemistry would provide a unique focus of activity that would not only attract new members but would distinguish AACR activities. 3 The Think Tank participants were charged with reviewing the state-of-the-art of chemistry as it applies to cancer research as a first step toward developing recommendations for AACR. Marnett asked that each speaker and discussant go beyond the science to consider programmatic areas AACR can pursue in educational activities, lobbying issues, training for the profession, and support. Specifically, speakers were asked to address the following questions: What are the exciting scientific opportunities in chemistry? What areas of chemistry will drive progress in cancer research? What are the opportunities for the application of chemistry to cancer biology? How can AACR best advance and make visible chemistry in cancer research? How can AACR most effectively interact with other organizations that have members who are interested in chemistry in cancer research? AACR Chief Executive Officer Margaret Foti, Ph.D., M.D. (h.c.), in welcoming the participants, urged them to seize the day and make an impact, thinking broadly about fostering new areas of cancer research. She said that AACR has worked vigorously and progressively to identify and advance important scientific areas in cancer research, and noted that prior think tanks and focus groups have resulted in a cascade of new programs, activities and policies. She asked that participants consider partners with whom the association might advance its mission, and noted that the results of this meeting would be conveyed in a report to the AACR Board of Directors as well as shared with members in other ways. Participants in the Think Tank came from many sectors and specialties, yet their perspectives – and recommendations for action – were remarkably consistent and forward-looking. Chemists at the meeting described important advances and future hopes in cancer research, as well as a strong interest in expanding the role of the AACR in providing intellectual, political, and public support for chemistry in cancer research in a variety of programs and venues. 4 Scientific Opportunities and Challenges Proteomics Chairperson: John A. Secrist III, Ph.D., Southern Research Institute Discussants: Catherine C. Fenselau, Ph.D, University of Maryland Daniel C. Liebler, Ph.D., Vanderbilt University Steven R. Tannenbaum, Ph.D., Massachusetts Institute of Technology Chairperson John A. Secrist III, Ph.D., a medicinal chemist who directs drug discovery at the Southern Research Institute, introduced the session on proteomics. Proteomes comprise collections of proteins in living systems. Unlike genomes, which are static, proteomes constantly change in response to disease and environmental stimuli. The genetic dysregulation that is a hallmark of many cancers is reflected by proteomic changes. Understanding these proteomic changes will provide new opportunities for cancer detection and therapy. Application of analytical proteomics approaches in cancer research at the molecular and cellular levels has proven invaluable to dissect protein signaling networks and characterize molecular phenotypes. Application of proteomics in clinical studies has also generated intense interest and considerable controversy, the latter due largely to uncertainties about the limits of current technologies and the best ways to apply them. Advances in chemistry will continue to be needed to drive advances in proteomics in cancer research. Tools and techniques were a major theme throughout this session. It was noted that educational workshops are a growth area for AACR and funds for additional workshops are being discussed with the National Cancer Institute. The recent AACR Special Conference on Proteomics in Cancer Research serves as an example of a multidisciplinary meeting to highlight both analytical chemistry and its application in basic and clinical cancer research. The task of studying many proteins simultaneously in a cell or organism poses unique scientific challenges, reported Catherine C. Fenselau, Ph.D, a professor at the University of Maryland and a member of its cancer center. Her work has involved the application of mass spectrometry to characterize changes in proteomes related to acquired drug resistance. The specific aim of her research is to develop new methods of comparing complex proteomes, and she described recent work done in drug-sensitive and resistant estrogen-positive breast cancer cells. Dr. Fenselau’s group has developed isotopic labeling methods to enable quantitative comparisons of proteins. She noted several challenges in the characterization of complex protein samples, including 1) more efficient sample preparation techniques, 2) reproducible and rapid fractionation techniques, and 3) affinity techniques to enrich the low abundance proteins in complex mixtures. Dr. Fenselau identified several opportunities for chemistry to contribute to proteomics. The first is the development of faster and more sensitive methods for quantitative analysis of proteins. Second is the development and adaptation of new approaches to labeling proteins and peptides for selective analyses. A third opportunity is in collaboration with information and computer scientists to develop enhanced analysis capabilities for data, including tools for quantitation and metrics for data reliability. Finally, Dr. Fenselau noted that development of approaches to find and identify proteins in low abundance is critical to the field. Many drugs, carcinogenic chemicals and inflammatory processes damage proteins through oxidation and covalent modifications. Characterizing protein damage and its cellular significance is the challenge for Daniel C. Liebler, Ph.D., a professor of biochemistry and pharmacology at Vanderbilt University who 5 directs the university’s proteomics laboratory. His laboratory is studying the protein targets of reactive electrophiles, a problem long considered “analytically intractable,” but now can be approached through mass spectrometry-based methods for high throughput characterization of proteins. The objectives of these studies are to elucidate mechanisms of injury and adaptation and to identify biomarkers for toxicity and disease. He described the identification of protein targets of model biotin-tagged electrophiles, in which the biotin tag enables selective capture and analysis of protein adducts. A surprising finding is that electrophiles display a high degree of selectivity in protein modification and that most proteins are targeted at one or two sites. This raises the question of whether selective protein adduction has functional consequences. Protein damage is a characteristic of cancer and other chronic, degenerative diseases, as well as drug and chemical toxicity. Selective modification of proteins and affinity capture may also provide a means of identifying the protein targets of drugs that act by unknown mechanisms. He noted that, in many cases, researchers don’t know the precise protein targets with which a promising drug interacts, and that this is another area where chemists can contribute to cancer research. He indicated that chemistry can provide essential tools for this research, including novel affinity probes and electrophile-based protein capture chemistries, as well as new protein digestion methods. Research in carcinogen-protein adducts also is the focus of Steven R. Tannenbaum, Ph.D., professor of chemistry and of biological engineering at the Massachusetts Institute of Technology. His research focuses on bioanalytical chemistry and mass spectrometry to analyze adducts on both proteins and nucleic acids. Dr. Tannenbaum discussed major changes in approaches to analyze protein adducts with mass spectrometry. Whereas previous approaches were focused on release of adducts and subsequent analysis, new proteomics approaches enable analysis of the modified peptides to map adducts to specific sequences. He discussed the analysis of proteins modified by reactive nitrogen species involved in inflammation and cancer. A challenge is detection of low abundance modifications, such as nitrotyrosine. He described chemical derivatization methods to capture nitrotyrosine-modified peptides for analysis and demonstrated that protein nitration displays selectivity. Tannenbaum noted that a better understanding of protein-ligand chemistry is needed to better understand protein adduction, calling it a fertile area for future research. Application of these approaches to complex proteomes, such as serum remains a challenge, in part because of the difficulty in selectively capturing modified proteins and in part because of interference from highly abundant serum proteins. He further noted that many membrane proteins or low-abundant soluble proteins were difficult to analyze by current methods. Global proteome analyses can be laborious and very slow. Finally, he pointed out that a major need in the field is for the availability of better informatics tools to facilitate the highthroughput analysis of data generated from complex proteomes. Recommendations for Proteomics Discussion following the presentations highlighted several broad areas in which advances were necessary to advance proteomics in cancer research. Because proteomics represents an integration of chemistry with cell and cancer biology, computer science and biostatistics, some areas of need lie outside the realm of chemistry per se. A major such area is bioinformatics, which nevertheless must be informed by chemistry to develop tools that are useful for proteomics. However, there was general consensus among the participants that advances in chemistry related to protein-molecule interactions, mass spectrometry and quantitative methods could advance the application of proteomics in cancer research. It was suggested that an AACR Special Conference on Chemistry in Proteomics would help to highlight key issues and advance the field. Major chemistry-related recommendations are: 6 1. Despite rapid advances in technologies, proteomics approaches are still unable to comprehensively characterize complex protein mixtures with high sensitivity and with high throughput. Chemists can contribute to the development of analytical proteomics methods, including new methods for protein and peptide fractionation and separations, new mass spectrometry instrumentation and novel applications of existing instrumentation, reagents and approaches for quantitative proteomic analyses and streamlined and high-throughput methods for processing samples for mass spectrometry analyses. 2. A major task in proteomics is the selective analysis of key protein targets of drugs or proteins that play key roles in cancer. The fundamental problem is detection and analysis of specific proteins in complex mixtures. Chemists can fundamentally advance the field by developing novel protein labeling reagents, specific protein capture chemistries and new methods to identify the protein targets of small molecules. 3. The relatively poor representation of proteomics in the AACR flagship journal Cancer Research reflects a lack of proteomics expertise on the Editorial Board. Lack of Editorial Board expertise in this area diminishes the quality of peer review and effectively diverts cancer-related proteomics work to other journals. Appointment of a senior editor in the area of proteomics would help the journal identify and attract high quality submissions and would improve the representation of proteomics in the field. 7 Drug Discovery in Prevention and Therapeutics Chairperson: Steven K. Davidsen, Ph.D., Abbott Laboratories Discussants: Malcolm F.G. Stevens, Ph.D., University of Nottingham Peter Wipf, Ph.D., University of Pittsburgh Kenneth Bair, Ph.D., Chiron Corporation Chairperson Steven K. Davidsen, Ph.D., of Abbott Laboratories, whose principal interest is in small molecule cancer targets, introduced the session, which highlighted the interdisciplinary nature of modern cancer drug discovery. Whether preventive or therapeutic, drugs under development require a “village” of researchers to come to market. Yet, despite the need for teamwork across disciplines, many chemists reported a sense that their contributions to drug discovery are invisible to others in the field – a predicament credited in part to chemistry serving as the “lingua franca” of the medical and biological sciences, so integrated that it may be taken for granted. Similarly, tensions were noted in the wider perception that drug discovery has not succeeded as well as hoped in bringing forward successful new therapies, despite some critical successes. It was felt that National Cancer Institute support for chemists is diminishing, for example, due in part to a perception that drug discovery is not aiding the war on cancer. The mix of industry and academic approaches to drug discovery, well represented in this topic area at the meeting, also create challenges and opportunities for the field. Among the major challenges: the proprietary nature of industrial research and the resulting trend toward omitting chemical structures from submitted papers and posters, a topic that gave rise to strong feelings about scientific integrity. Financing for research in the private and academic sectors were contrasted, and, in areas like natural products, the pharmaceutical industry’s decision to deemphasize pursuit of this approach offers academics the chance to take up the charge. Malcolm F.G. Stevens, Ph.D., of the University of Nottingham, is a professor of experimental chemotherapy, with a background in pharmacy and chemistry; currently, his specialty area is small molecules. He calls chemists and developmental pharmacologists those who “actually do drug discovery.” He believes that interesting chemistry drives interesting biology, and espoused an approach to cancer drug discovery that involves starting with interesting molecules (i.e. with novel functionalities) that are then submitted for the widest possible screening in vitro against panels of tumor cell lines. When a compound is shown to elicit a unique phenotypic response this becomes a starting point for chemists to optimize activity and pharmacologists to define the mechanism of action. The two disciplines working in harness then develop the project from lead to clinical candidate to clinical trial – hopefully. Such an approach emphasizes the merits of intelligent medicinal chemistry as the primary driver of anticancer drug discovery and necessitates interdisciplinary bridge-building. Stevens sparked a general discussion of the AACR journal Cancer Research by noting he had submitted a paper and received the comment “this paper contains chemistry and thus is not relevant to Cancer Research.” Participants noted that chemists had been removed from the editorial board, and, as with meeting submissions, papers submitted to the journal should require chemical structures. Participants felt that the journal was emblematic of the field in general: that chemistry is seen as secondary to biology in the world of drug discovery. It was noted that the best drug developers embrace all the disciplines in their search for cancer drugs, and that doing otherwise limits the likelihood of curing cancer. 8 In contrast to the phenotype-based approach to cancer drug discovery described by Dr. Stevens, Kenneth Bair, Ph.D., summarized the target-based strategy that he has taken at Chiron Corporation focused on the identification of novel kinase inhibitors. He echoed earlier comments that medicinal chemists require functional expertise in a variety of subdisciplines – in structural chemistry, computational chemistry, chemical informatics, and analytical chemistry – to create new drugs. Dr. Bair emphasized that chemical informatics affects how information is used, allowing for better decision-making based on complex relational data. From his viewpoint, the power of translational medicine lies in discovering whether a drug is performing as expected in humans. This information is vital to engineering the optimal drug molecule for a given target and commonly necessitates an iterative clinical program in which chemists improve upon the properties of first-generation compounds based on clinical data. With a large information base, including both general and kinase-specific screening libraries, the company works with one-half million compounds. These compounds are selected from multiple sources and are filtered to eliminate non-drug-like compounds through such screens as molecular weight, overreactivity, and more. The result of kinase selectivity profiles is an enhanced therapeutic index, a development that is exciting to biologists. Cross-disciplinary teams are important in reaching a better discovery starting point; chemists review the profile of all hits, then work with biologists to discover more relevant profiles. Regarding the reluctance of some researchers to publish or disclose structures in their journal articles or meeting posters, particularly those conducting proprietary research for industry, it was noted that other societies are now requiring disclosure, including the International Association for Antiviral Research. Biologists also face the issue of structure disclosure, and while it is possible to gain value from a poster or paper without the structure, many consider it an issue of scientific rigor. It was noted that the AACR board is awaiting guidelines for how to legislate the issue, and what punitive action would result from failure to disclose a structure in journal papers, invited presentations or meeting posters. A subcommittee of the Working Group will be formed to provide insight and advisement. Drawing on the field of natural products is an exciting and inspiring way to expand the arsenal of anticancer therapies, according to Peter Wipf, Ph.D., of the University of Pittsburgh. A synthetic organic chemist, Wipf’s research studies natural products, and one of his recent projects targets the disorazoles, which are highly potent agents that interfere with tubulin dynamics. He sees several exciting opportunities for chemists in cancer research, including asymmetric catalysis and the synthesis and structure-activity relationships of natural products. The ability of chemists to design and synthesize compounds with tailored properties will ultimately lead to new active pharmaceutical intermediates. Such research provides the tools necessary to elucidate signal transduction pathways in cancer cells via chemical genetic experiments. This approach necessitates collaboration with biologists, thereby building bridges between the disciplines. Wipf dubbed peptidomimetics the “holy grail” in pharmaceutical R & D, noting there are many examples in nature. Similar to natural products research, the study of peptidomimetics has largely been abandoned by the pharmaceutical industry – a trend Wipf sees as creating opportunities for academics and small startup companies. Other participants agreed, noting that natural scaffolds influence how proteins interact, and that plants provide superior diversity of chemical structures. Pharmacognosy, considered by many a dead science, may be a discipline which should be revived. Additionally, the National Cancer Institute’s collection program and repository for natural products is slowly being shut down, and cessation of collection would stop the pipeline for research in this area. 9 Recommendations for Drug Discovery in Prevention and Therapeutics 1. The discovery of drugs that benefit cancer patients requires the collaboration of scientists from multiple backgrounds. It is essential that chemists working in cancer drug discovery have a functional understanding of aligned disciplines and that researchers in these aligned disciplines have a basic understanding of the role of chemistry in drug discovery. Training that emphasizes cross-functional expertise is important and should be supported by the AACR. 2. The use of small molecules, particularly natural products, for the elucidation of signaling pathways in cancer cells is vital to the future of cancer research and is a field well suited for academic & government chemists. Emphasis should be placed on the collection and characterization of natural products and the elucidation of the interesting biology of natural products, especially as revealed by chemical genetic experiments. 3. As a means of ensuring scientific rigor, the AACR should revise its guidelines so that chemical structures are included in all publications. This policy should also apply to meeting posters and presentations, requiring that structures be disclosed by the time of the meeting. 10 Structural Biology Chairperson: Christopher J. Michejda, Ph.D., National Cancer Institute-Frederick Discussants: Angela Gronenborn, Ph.D, NIDDK/NIH and University of Pittsburgh Vivian Cody, Ph.D. Hauptman-Woodward Research Institute, Buffalo Stephen Neidle, Ph.D., University of London, Great Britain Christopher J. Michejda, Ph.D, who leads a drug discovery laboratory at the National Cancer Institute and also chairs the Institute’s Chemistry and Structural Biology Faculty, introduced the session. Structural biology seeks to understand the three-dimensional structure of macromolecules and to learn how structure affects function. The knowledge of structure allows the design of small molecules that inhibit the function of macromolecules and thereby provides an indispensable starting-point for rational drug design. Structural biologists express and purify proteins and nucleic acids, crystallize the macromolecules, collect the data that allows them to build and refine structural models of the target molecules. The principal tools of structural biology are X-ray crystallography, high resolution nuclear magnetic resonance spectroscopy and ultra high resolution cryo-electron microscopy. The participants of the Think Tank were asked to consider the value of thinking in 3D as this approach allows researchers to better understand the functional and dynamic properties of macromolecules. It was noted that a detailed knowledge of structure allows for the possibility of targeting various sites on a single protein, which may allow the design of “multiplex drugs” that will be less sensitive to inactivation by mutation Angela Gronenborn, Ph.D., who led a macromolecular NMR group at NIDDK but who has recently been appointed to be the Director of Structural Biology Program at the University of Pittsburgh, stated that structural biology belongs “nowhere and everywhere.” She noted that structure based drug design is a very powerful paradigm because it allows researchers to determine precisely how a small molecule inhibitor interacts with a target protein, thereby providing a rational basis for improvement. The postgenomic era, where opportunities in drug discovery have turned out to be target-rich but lead-poor, has forced us to consider concepts such as structural genomics. However, critics question whether structure based design is sufficiently efficient for today’s requirements. Gronenborn feels strongly that structural studies coupled with combinatorial refinement of hits is a very powerful emerging tool for new drug discovery. While much of the structural data comes from crystallographic studies, NMR is a powerful tool for detecting weak interactions between small molecules and macromolecules and can be utilized in numerous innovative experimental approaches. Technology development is constantly changing the landscape of structural biology and continued technical innovation should be encouraged. It is critical that structural biologists as well as chemists build bridges with other disciplines involved in cancer research. Vivian Cody, Ph.D., who is a Professor of Structural Biology at the Hauptman-Woodward Research Institute and the University of Buffalo, provided a strong argument for structural genomics. She was especially enthusiastic about the initiative in structural genomics supported by the National Institute of General Medical Sciences. Started in 2000 as a pilot project for protein structure determination on a large scale, the initiative has awarded five-year funding to nine centers to develop technology and lower costs for structure determinations. Dr. Cody strongly recommended that the AACR supports this initiative. She further noted that structural biology will experience major changes in high through-put crystallization methods, improved methods for data collection, improvements in direct methods for structure solution and more rapid and automated methods for structure refinement and modeling. She echoed Dr. Gronenborn in noting that there are tremendous opportunities in close collaboration with chemists and 11 biologists, including in areas such as structure fragment screening, structure based screening, and target validation. Stephen Neidle, Ph.D., who is a Professor of chemical biology and leads a program in structural biology at the University of London’s School of Pharmacy, focused his remarks on the structural biology of nucleic acids, calling them among the most important targets in cancer research. He noted that structural complexity of nucleic acids has increased considerably since the early days of the double helix. DNA complexes with proteins are an active area of study although the problem is complicated by inadequate algorithms for modeling of nucleic acids and their protein complexes. Structural studies, especially by NMR, of covalent DNA adducts of small molecules has been very successful. He cited the structure of the tamoxifen-DNA adduct, which provided very powerful information directly relevant to cancer, since the formation of these adducts is probably responsible for the mutagenic effects of the drug. As an example of structural studies having an impact on drug design involving DNA telomerase function, including the drugs that stabilize quadruplex structures, Dr. Neidle mentioned siRNA structures, backbone-modified RNA’s, DNA aptamers, drug-DNA-protein complexes and quadruplex DNA/RNA structures. Recommendations for Structural Biology 1. Structural biology and chemistry are natural partners in drug discovery and development. Although some methods are already in place for rapid structure-based hit to lead drug development, new technology and methods development initiatives should be supported. Computational chemistry should also be an important component of the mix. 2. Structural biologists should be encouraged to develop methods for tackling difficult structural problems such as membrane proteins and assemblies of macromolecules. The potential of NMR spectroscopy in solving critical dynamic problems in protein-protein, protein-nucleic acid and protein-small molecule interactions is only beginning to be realized. The development of “metabolomics” has been slow but has the potential of rapid profiling of metabolism of xenobiotics. 3. Structural biology should be a key component of the team approach to the solution of major problems in cancer research, including drug discovery and development, prevention of cancer and carcinogenesis. Structural data on key target macromolecules, particularly with small molecule ligands in place, has a proven history of informing further development, especially when coupled with appropriate modeling methods. Suitable steps for the inclusion of structural biology in the team approach to the solution of major human health problems should be encouraged. 4. AACR can play a major role in highlighting the importance of structural biology in the team approach to cancer research by featuring critical structural contributions to the overall paradigm. Sophisticated structural biology is a difficult science. Consequently, the recognition of structural biology as a co-equal partner rather than just a tool will greatly improve the enthusiasm of structural biologists to become committed participants in the team approach. 12 Computational Chemistry/Informatics Chairperson: John T. Hunt, Ph.D., Bristol-Myers Squibb Discussants: Billy W. Day, Ph.D., University of Pittsburgh John N. Weinstein, M.D., Ph.D., National Cancer Institute John T. Hunt, Ph.D., of Bristol-Myers Squibb’s cancer biology group and a former head of the company’s cancer chemistry group, introduced the Computational Chemistry/Informatics session. This session built upon nearly all other sessions of the think tank meeting because a fundamental challenge – and opportunity – for chemists in cancer research lies in the vast array of information available for exploration and analysis. Researchers speaking on nearly every topic agreed that new technologies are delivering an overwhelming amount of data, whether the subject area was genomics, transcriptomics, proteomics, metabonomics, biochemical activity data, systems biology analysis, structural biology, molecular modeling or structure activity relationships. Those working in computational chemistry and informatics have the twin challenge and opportunity of distilling, managing and making sense of that data overload. Other key issues include maintaining data quality, sharing data and informatics across intramural multidisciplinary teams, as well as data sharing with the broader scientific community. Billy W. Day, Ph.D., a medicinal chemist from the University of Pittsburgh and director of the proteomics core lab for the university, sees potential challenges for drug discovery and development in the areas of nontraditional SAR and QSAR and disease simulation. He pointed out that academic access to the rich array of industry data would facilitate model building and might even allow computational researchers to build disease and drug action simulations at an organismal level. In biological mass spectrometry, he believes that a strong need exists for better software and approaches, including: • • • • • new approaches to biomarker discovery three-dimensional structure reconstruction from designed cross-linking studies software for de novo top-down analysis of macromolecules >10,000 Da software for rare and/or “unnatural” post-translational modifications development of sequences and three-dimensional structures for analyte-capturing and –reporting aptamers, an area also ripe for new computational tools Day described a disease modeling strategy that could one day, perhaps sooner than expected, be applied to cancer. The modeling platform was the result of a team effort at a small company involving an immunologist, a clinician with an engineering background and a mathematician, to build models that would aid in treating and predicting the innate immune response in sepsis, trauma and hemorrhagic shock. A system with fewer than 200 ordinary differential equations (with variables in the equations including lab measures such as cytokine levels and nitric oxide enzymes and metabolites, and clinical measures such as cell counts, blood pressure and body temperature) that simulates the first 24 hours of the innate immune response in rodents and in man provided the basis for building a computational system with a graphical user interface that is friendly for medicinal chemists, biologists and immunologists for examination of potential interventions. The modeling platform allows continual updating of the model, experimental prediction in preclinical studies, data output, and building of a database that can be used for better future modeling and clinical trial simulations/design. The one approved compound for treatment of septic shock requires treatment of 16 patients to show one good response, and is an expensive drug. Therefore, effective computational simulations could maximize the efficacy of the trials of such agents and aid in cost containment. 13 To build these types of models, very dense and accurate kinetic maps are needed. Day noted that future computational research in cancer research would benefit from kinetic maps to build models for such things as how xenobiotics interact with the host, what immune cells experience and do, and genotypic factors. With these types of models, chemoprevention researchers can start with healthy people and predict what preventive agents will do in terms of perturbing the biological systems. John N. Weinstein, M.D., Ph.D., who heads the bioinformatics, biostatistics and computational biology faculty at the National Cancer Institute, takes the opposite perspective. He asks not what biology or AACR can do for chemistry, but what chemists can do for biologists, as chemists’ insights are needed for large data sets. Today’s availability of larger networks of data require chemists’ input to determine the quality and quantity of correlative molecular data needed to advance clinical research, for example in the fields of defining molecular markers of drug efficacy and toxicity. While early cancer therapies were developed in the absence of most of the data available today, current researchers anticipate that the large amount of currently obtainable data will lead to improved treatment options in the future. As the complexity of the data is evolutionary, dynamic and changing over time, it poses particular challenges and opportunities and requires a systems perspective. He noted that biologists don’t tend to think in terms of three-dimensional structures, and urged AACR to help broaden and enrich the perspectives of molecular biologists, so that they will learn to think in chemical and structural terms. Weinstein, whose research takes place in the genomics and bioinformatics group at NCI, aims to integrate information from the NCI drug screen with information on molecular targets, to develop both new drugs and new targets. He characterized previous research as a “cottage industry,” looking at one gene at a time. Today, it is being industrialized and carried out on a massive scale. Weinstein expressed frustration in his work with the NCI cell screen and the lack of chemical input. He summarized the kinds of data that his group developed, including pharmacology data, information on hundreds of targets from the literature, 2D protein gels, cDNA arrays and protein level lysate arrays, to illustrate the challenge of integrating those data in a meaningful way. Weinstein calls the effort to synergistically integrate this wide array of data as “integromics.” The issue of “too much data” can be seen as a high-dimensional data analysis problem, Weinstein said, which can be addressed using a range of techniques, from classical statistical methods and computerintensive resampling techniques, to signal processing and artificial intelligence. Cautioning against the notion of a “one-size-fits-all” solution to the challenge, since any technique can fail if an inappropriate data set is used, and the results of an analysis can depend on the question posed. Linear approaches tend to be better for looking at broad aspects of structure in a high-dimensional space, he said, while non-linear or non-parametric approaches are better for finding small elements of structure. Biological interpretation follows data analysis, and Weinstein described a rich array of freely available bioinformatic resources from the NCI genomics and bioinformatics group, which, taken together, exemplify the possibilities and advances that allow faster results, including: 1. Translating gene identifier types for lists of hundreds or thousands of genes 2. Batch-processing and organizing lists of thousands or tens of thousands of genes and providing visualizations of the genes in the framework of the gene ontology hierarchy 3. Generating color-coded Clustered Image Maps (CIMs) or heat maps to represent high-dimensional data sets such as gene expression profiles for data on drug activity, target expression, gene expression, and proteomic profiles 4. Searching and organizing the biomedical literature on genes, gene-gene relationships, and gene-drug relationships in order to speed up the capture and organization of literature from PubMed searches 14 5. Searching a relational database of commercially available antibodies and qualitycontrol information on them 6. Linking molecular markers and the drug discovery process He noted examples of translational chemistry in which databases have yielded medical uses, including identifying markers to distinguish colon and ovarian cancers by verifying sequences and corroborating expression; and spurring clinical development of oxaliplatin for colon cancer, via a heat map database identifying critical evidence leading to drug development and a clinical trial. Recommendations for Computational Chemistry/Informatics 1. AACR should encourage more computational chemists, biological modelers and theorists to enter the field of cancer research, as well as to educate their colleagues about what they are doing and assist in bringing new methods to the table. Computation is seen as a field impenetrable to many in cancer research. 2. A consensus conference may be needed on the issue of data quality. There is a need to standardize vocabulary, to select and apply quality metrics to describe the data to be used for modeling and databases, and to introduce new computational tools to cancer research. Experiments need to be anchored with appropriate standards, and the lack of recognition that these should be inserted at the initiation of the research process is still seen as a problem. The need for high quality data, particularly in large data sets, is essential if informatics tools are to yield useful and reliable results. Related challenges are that imprecise data may still be useful, and different results can be achieved by analysis of the same data. 3. Access to industry’s large and generally well-controlled data sets would be a great asset to computational approaches. This presents an intellectual property challenge that may be insurmountable. However, data-sharing with academic computational laboratories is likely to be much more achievable and should be more easily enabled. 15 Chemical Biology of Carcinogenesis Chairperson: Richard N. Loeppky, Ph.D., University of Missouri Discussants: Judy L. Bolton, Ph.D., University of Illinois Lisa A. Peterson, Ph.D., University of Minnesota Carmelo Rizzo, Ph.D., Vanderbilt University Chairperson Richard N. Loeppky, Ph.D., an organic chemist and University of Missouri professor working at the interface of the chemistry of nitrosation and nitrosamine carcinogenesis, introduced the session. Chemists have and will continue to make strong contributions to the prevention of cancer through the elimination and reduction of carcinogen exposure and the exogenous and endogenous processes which generate them. Chemist’s ability to elucidate mechanisms and the protein and DNA damaging reactive intermediates responsible for the initiation of the disease process underpins research in the chemical biology of carcinogenesis. Chemical research, and opportunities, in this field span vast interdisciplinary boundaries reaching from epidemiology to the molecular biology of mutation, promotion and progression. Covalent modification of DNA by electrophiles is an initial step in chemical carcinogenesis. If these modifications are not repaired, they compromise the fidelity of DNA replication, leading to mutations and possibly cancer. From chronic toxicity and mutagenicity to intervention and the discovery and application of chemopreventive strategies, the field is yielding exciting new information that will help prevent cancer. An illustrative current example comes from the elucidation of some of the metabolic and chemical processes which give rise to the unintended effects long-term estrogen therapy in postmenopausal women. In all research in the chemical biology of carcinogenesis, the development of new analytical instruments and methodologies, the elucidation of new mechanisms of endogenous chemical processing, and the discovery and synthesis of new DNA and protein adducts are hallmarks. Judy L. Bolton, Ph.D., a physical organic chemist and a professor of medicinal chemistry at the University of Illinois at Chicago, discussed her research on hormonal carcinogenesis. She showed how epidemiological data from the Women’s Health Initiative Study of 16,000 healthy postmenopausal women treated with either a placebo or Prempro motivated her work and led her to explore the chemistry of metabolically produced reactive intermediates generated from the equine estrogens, such as those present in Premarin and Prempro, the most widely prescribed estrogen therapy. Women on Prempro had a 26% increase in breast cancer incidence. Much of Bolton’s research focuses on highly redox active/electrophilic o-quinones that are generated from the enzyme mediated oxidation of electron rich aromatic rings. She studies reactive intermediates through synthesis and characterization, generates them with model systems in vitro, and tracks them with biomarkers in vivo. Among these are GSH and DNA base adducts that her group has independently synthesized. These substances also offer the potential of being used as biomarkers in humans. This process allows her to predict the potential of reactive intermediates to cause cytotoxicity, mutagenesis. An important result of Bolton’s research, which is so illustrative of the value of chemist’s and chemistry in cancer prevention, is her recognition that the same processes that produce reactive quinones from some of the equine estrogens, can also become manifest in the metabolic transformations of related drugs such as Tamoxifen and Ralaxofene. Her research, through traditional medicinal chemistry, may lead to the development of better chemopreventive agents with a much lower genotoxic risk. Lisa A. Peterson, Ph.D., of the University of Minnesota’s Cancer Center and Division of Environmental Health Sciences focused her discussion on two of her research projects to illustrate the importance of 16 multidisciplinary interactions between chemistry and other fields when exploring significant questions in carcinogenesis. She began by discussing the differential repair of DNA adducts arising from the tobacco nitrosamine carcinogen NNK. Molecular epidemiology studies have shown that humans with the I143V/K178R genotype of the alkyl guanine transferase (AGT) DNA repair protein have a two-fold increase in lung cancer risk. NNK generates two different types of reactive fragments on metabolism. One of these is a methylating agent and the other incorporates the more bulky part of the NNK molecule into the O-6 position of guanine DNA bases. Her group synthesized the bulky adduct as a nucleotide and incorporated it into several DNA oligomers at specific sites. The kinetics of AGT mediated repair of O-methyl-dG versus the bulky adduct were measured for three human mutations of AGT vs. wild type. While differential repair rates were a function of both the mutant AGT and the DNA oligomer sequence, rates of removal of the bulky adduct were lower for the mutant with the enhanced lung cancer phenotype while demethylation rates, mutant to mutant, did not change significantly. This research not only documents at the molecular level how genotypic differences in repair proteins can lead to higher lung cancer incidence but shows the importance of the bulky adduct in this pathogenesis. Further elaboration of this work could lead to the identification of individuals who are at risk for smoking-derived lung cancer. She noted that it is possible to distinguish between proteins that can and cannot repair damage, as well as rates of repair in vitro; however, she stressed importance of chemists and biologist collaboration to develop a model for predicting the likelihood of mutagenesis. Peterson also discussed her research on furan, a simple compound that is widely used in industrial chemistry but is also a constituent of tobacco smoke and car exhaust. In rodents, furan is a liver toxicant and carcinogen, and has been listed as a possible human carcinogen. Furan toxicity requires metabolism, but, on the basis of research to date, has been classified as an epigenetic carcinogen rather than a genotoxin. The Peterson group’s approach is directed at testing this assumption. They use many of the same methods described by Bolton, metabolism, model studies utilizing the putative reactive intermediate cis butenedial, the structural characterization of DNA base adducts, and their synthesis. Exposure to epigenetic and genotoxic carcinogens is regulated by separate protocols. Peterson’s work could produce biomarkers valuable to not only understanding its biological mode of action, but substances of use in molecular epidemiology. In collaboration with biologists, her group is also evaluating the potential of this compound’s metabolites to interfere with mitosis and mitochondrial activity. Carmelo Rizzo, Ph.D., of Vanderbilt University’s Department of Chemistry used his work within a program project on the chemical biology of carcinogen DNA adducts to show how chemists work together with biologists to determine the factors which affect the repair of DNA containing specific adducts and how these adducts give rise to mutation when repair fails or is overwhelmed. The project contains three interdisciplinary components that represent a strongly coordinated chemistry-biology approach. A synthetic group produces oligonucleotides that are modified by potential mutagens in a structurally homogeneous way. The biology component then evaluates the effect of adducts on DNA polymerase rate, function and fidelity as well as their rate and susceptibility to repair as is related to oligonucleotide sequence and structure. This information is evaluated in active collaboration with another group that has used NMR spectroscopy to determine how the adduct changes the three dimensional structure of the oligomer. An important aspect is that all projects benefit from having access to sitespecifically adducted oligonucleotides; the structural work could not be accomplished without this access, and the biological work can be achieved in a more rigorous fashion with the access, for example. For structurally complex adducts, the synthetic chemistry is often rate limiting. Rizzo pointed out that Vanderbilt scientists have and are using this approach to focus on DNA adducts that arise from the endogenous oxidative degradation of lipids. Some recent accomplishments in this area 17 are the syntheses of oligonucleotides site-specifically adducted with acrolein, crotonaldehyde, and 4hydroxynonenal. These DNA adducts have also been shown to form DNA-DNA cross-links, a serious form of DNA damage, as well as DNA-peptide cross-links. The group has also synthesized oligonucleotides with spectroscopic probes to study the cross-linking chemistry. A similar approach is being utilized to examine deoxyguanosine adducts of aryl amines such as IQ, which is found in cooked meats. The general group discussion following the presentations of these chemists emphasized the conviction that research in the chemical biology of cancer plays a large role in the prevention of disease. Radiochemical DNA-based assays over two decades ago led to the discovery of numerous unknown adducts in humans. Structural identification of these “footprints” of exogenous exposure and endogenous processes is now becoming possible and offers the exciting potential of a new era in the discovery of new paths and mechanisms of pathogenesis. Yet, research in this field of carcinogenesis and its potential remains significantly unrecognized in both the chemical and cancer research communities. There is concern that the term “carcinogenesis” carries unwanted connotation and is jaded. The invention of new terminology is desirable and will be catalytic in the stimulation of research. There is an abundance of exciting scientific and practical research problems in the chemical biology of cancer, but there are too few new researchers entering the field, which has and continues to see the retirement of many its most productive and knowledgeable practitioners. Research in this area is also strongly linked to research in proteomics, biomarkers, structural biology, molecular epidemiology, and the discovery of anticarcinogenic agents. Its importance and potential for significant chemical research must be stressed. Recommendations for Chemical Biology of Carcinogenesis: 1. It is imperative to recruit new researchers to the chemical biology of cancer, and to ensure that adequate resources are available to support this research. Focused scientific meetings jointly sponsored by the AACR and the ACS are seen as important step in achieving this goal. Selective programming within both organizations will also be important. Young professors and other researchers should be invited to participate in smaller interdisciplinary meetings in the chemical biology of cancer in order to engage them as active researchers in the field. 2. Programming efforts within the AACR, perhaps involving the participation of the ACS when appropriate, should be made to educate both chemists and biologists in the chemical biology of cancer. These presentations should emphasize the importance of this area to the frontline prevention of cancer, its potential for interdisciplinary development, and its past accomplishments as indicators of still valid approaches. 3. To facilitate the research, to provide for more funding, and to enhance the recruitment of scientists to this area of research, effort and attention should be given to finding an exciting new name for the field which is devoid of the negative connotations of “carcinogenesis.” 18 Biomarkers/Analytical Chemistry Chairperson: Homer L. Pearce, Ph.D., Eli Lilly and Company Discussants: Stephen S. Hecht, Ph.D., University of Minnesota Cancer Center Ian A. Blair, Ph.D., University of Pennsylvania John D. Groopman, Ph.D., Johns Hopkins University Chairperson Homer L. Pearce, Ph.D., of Eli Lilly and Company, trained as a synthetic organic chemist and has been involved in cancer drug discovery and development at Eli Lilly and Co. for the past 26 years. This area of chemistry places the researcher at the intersection of the laboratory and the clinical trials process. Analysis and interpretation of analytical data provides insight into both the disease process (e.g. mechanisms of carcinogenesis) as well as the mechanism of drug action. Recent advances in this field have been made as a result of increasingly sensitive spectroscopic techniques (mass spectrometry and nuclear magnetic resonance). Determining whether an investigational cancer drug is actually working in early clinical trials can be very difficult. The development of molecularly targeted drugs relies heavily on the analytical measurement of surrogate biomarkers. Biomarkers were pioneered in the 1970s to monitor exposure in a work environment. They were first developed for metals and organic solvents, then gradually expanding to a larger, more varied spectrum of exposures. These biomarkers established causality by allowing more direct and more accurate measurement of exposure to exogenous chemicals and outcome. Similar approaches are being employed in the pharmaceutical industry where decisions about which molecular targets to pursue in the drug discovery setting are frequently based on the availability of biomarkers and validated methods to study the drug’s activity in early trials. This is also true for biomarkers of toxicity. Stephen S. Hecht, Ph.D., a professor and chemist from the University of Minnesota Cancer Center, focuses his work on tobacco and cancer. He reviewed the World Health Organization’s 2003 data on causes of cancer in the world population: Diet and nutrition, and tobacco, each account for 30 percent; Chronic infection accounts for 18 percent; Genetic susceptibility accounts for 4 percent; Alcohol consumption accounts for 2 to 4 percent; Environmental exposure accounts for 1 to 4 percent; and Occupational exposure accounts for 1 to 2 percent. His own work focuses on linking byproducts of tobacco – specifically the nitrosation of nicotine to NNK, NNA and NNN. Last year, NNK and NNN were elevated from probable human carcinogens to "Group 1A – known human carcinogens", by the International Agency for Research on Cancer (IARC). Hecht noted that linking environmental and endogenous sources of DNA adducts and other genetic changes, as well as cancer, can be tracked by carcinogen biomarkers. However, he stressed the need for sophisticated, robust analytical chemistry in this area, as DNA adducts occur at very low levels in humans. Despite work in this area since the 1980s, he said, there are few examples of characterized DNA adducts in humans. Citing the large number of questionable measurements generated in the past 20 years, Hecht called for more quality control and measurement standards, echoing a common theme at the meeting. Mass spectrometry methods, he said, have become more sensitive in just the past five years and are likely the preferred measurement method in the future. 19 He offered two examples of what chemists can contribute to biomarker research, in the areas of urinary NNAL, and phenanthrene metabolites. After following the nitrosation of nicotine into NNK, NNA, and NNN, Hecht reported that NNK was found to be a tobacco-specific lung carcinogen, very potent in rodents. That basic chemistry research led to other studies analyzing tobacco and tobacco smoke, the decrease in NNK in tobacco products, carcinogenicity studies in animals, chemoprevention, metabolism, DNA and protein adduction identification, biomarkers, and tobacco control – ultimately leading to the designation of NNK and NNN as group one carcinogens in 2004. NNK uptake is easily determined through blood and urine samples of smokers, smokeless tobacco users, and recipients of second-hand smoke. By applying an NNAL assay to tobacco users, Hecht was able to identify ethnic differences in metabolism, the effect of diet on NNK metabolism, and the relationship of dose to lung cancer. The NNAL assay also is useful in non-smokers – including smokers’ spouses and children, infants, casino patrons and others -- as a very powerful marker. Identifying the transplacental uptake of NNK is a strong example of chemistry’s real impact on tobacco control measures in society. Another example of high impact lies with phenanthrene metabolites. Here, Hecht looked for variants in genes and a link with lung cancer, without promising results, as it’s not possible to capture the complexity with gene variants. Carcinogen metabolite phenotyping, however, can integrate the genetic and environmental effects on a carcinogen. While it is more difficult to phenotype, the process has more potential to yield the needed information. By measuring phenanthrene tetrol, or PT in a long-term study of smokers and non-smokers, Hecht found that the ratio of PT to phenols was significantly higher in smokers – double the ratio found in non-smokers. Hecht strongly advocated attracting younger chemists to the field of cancer research, suggesting that they be recruited after achieving either a B.S. or Ph.D.; via specialty meetings; with incentives such as travel stipends and awards. He also noted that a good, regular meeting has the potential to galvanize the field. Observing that “chemicals, not genes, cause cancers or modify them,” Ian A. Blair, Ph.D., a Professor of Pharmacology and Chemistry at the University of Pennsylvania noted that research priorities should reflect this axiom. In Blair’s view, the “holy grail” is biomonitoring of biomarkers in humans. As such, he sees several exciting opportunities involving biomarkers: Application of sophisticated instrumentation; Development of new instrumental techniques; Use of mechanistic-based chemical biology; Application of new advances in analytical pharmacology; Synthesis of reactive intermediates to elucidate mechanisms of carcinogenesis; and Development of biomarkers of cancer in human populations - the ultimate goal. Blair noted that several areas of biomarker chemistry will drive cancer biology research, including: major recent advances in analytical technology; the resulting potential for developing specific biomarkers for cancer etiology, diagnosis, prevention and treatment efficacy studies; and the macromolecular diversity of biomarker targets, which may include lipids, proteins, nucleic acids, and carbohydrates as well as changes in mRNA and protein expression. He also posed the question: What advances in biology will drive biomarker research? 20 Cancer biomarkers should be developed within the context of human disease and appropriate animal or other experimental models, and should take account of advances in human systems biology. Rigorous methodology with appropriate controls (such as stable isotope labeled standards) should be applied to the discovery of protein biomarkers in plasma or serum. Cancer biomarker discovery and validation will occur, preferably through close collaborations between researchers from diverse fields and applied to human population. Collaborative studies should be initiated through mechanisms such as the SPORE, SCCOR programs and large epidemiological cohort studies. Blair offered his university’s organization of departments related to this work as a paradigm for the future of academic analytical chemistry/biomarker research in cancer, noting that it includes the departments of chemistry and pharmacology, the cancer center, an institute for translational medicine and therapeutics, and a multidisciplinary training program in cancer pharmacology. He suggested that development of new biomarkers would be critical for many facets of future cancer drug discovery. This offers opportunities for the development of unique collaborations between academia and the pharmaceutical industry. It will be important to discover and validate biomarkers so that the FDA and all pharmaceutical companies accept their utility. This can potentially transcend the need for need for secrecy and should allow relationships to flourish between these two different cultures. John D. Groopman, Ph.D., Professor and Chair of the Department of Environmental Health Sciences at Johns Hopkins University, also works on DNA and protein adducts and how they can be used as molecular biomarkers in patients with liver cancer. He noted that liver cancer accounts for 16,600 deaths annually in the United States, making it this nation’s most rapidly increasing solid-tumor cancer. This trend, he said, is largely driven by the four million hepatitis C patients in the U.S., and has prompted a resurgent interest at the National Institutes of Health – an emerging trend that will result in one million deaths due to liver cancer over the next 20 years. Currently, the median age of onset for liver cancer is 45, and it is possible this cancer’s prevalence is driven by environmental exposure, as well as by hepatitis B and C. Noting that many human cancers will turn out to be interactions between chemicals and biological agents, a trend Groopman sees emerging in a number of endpoints, he cautioned against allowing high throughput assays to drive concepts of the etiology, without using analytical chemistry to examine chemical exposures. High throughput assays can bias etiological concepts, perhaps leading to ineffective therapeutic approaches, he said. He noted that, for many of the researchers at the meeting, the research bottleneck occurs not with mass spectrometry, nor with analysis, but with cleaning up all the “junk” ions before that point; he said solving this particular challenge will require partnerships with industry, as it cannot be accomplished in academe. Two environmental factors contributing to liver cancer incidence are aflatoxin in foods and alcohol consumption. Groopman took the group through his early research on aflatoxin to illustrate, in part, the analytical challenges in the field. Cross-sectional studies in human populations in China established a dose response for the first time, measuring the secretion of aflatoxin N7-guanine in urine; the study required 287 urine samples done in duplicate to achieve the finding, requiring two years of HPLC in the mid-1980s. Today, his team has generated 400 samples over the past eight months using mass spectrometry for a clinical trial. While analytic technologies have improved, he feels that throughput has not advanced far enough. Relating biomarkers to disease outcome in experimental models offers another enormous opportunity for sharing data between experimental models and human clinical investigations. In his work, this process 21 involved simple questions about whether a relationship exists between DNA damage measured in urine samples and residual amounts of DNA damage in the target organ. Modulation of biomarkers and tracking them to disease modifications will give a more powerful validation of whether a risk marker exists. Surrogacy, or how well biomarkers reflect what’s going on in a target tissue, is an important issue to discuss going forward, Groopman said. Singular biomarkers often under-represent the efficacy of the intervention, and overestimate the underlying risk of disease; in addition, they cannot track with outcomes in multistage carcinogenesis. Biomarkers work well assessing pharmacodynamic action in agents, but have little if any predictive value for risk modulation in an individual, a critical consideration in designing studies for individual versus population risk. Groopman noted that these biomarkers will be the efficacy endpoints that can be used in the future to find out whether people can be modulated for risk at the different stages of intervention. Waiting to use cancer as an endpoint is waiting too late for these patients, and chemists can deliver that intermediate interface through biomarker research. Groopman sees the exciting scientific opportunities in chemistry as: Relating host, genetic and environmental susceptibility factors to individual cancer risk by using specific and validated biomarkers, and Developing higher throughput strategies for biomarkers by developing better preparative clean-up methods, higher throughput analytical instrumentation methods, and partnering with biostatisticians to develop new methods for validation and application. Opportunities for the application of chemistry to cancer biology include interactions between geneenvironment; gene-gene-environment; vector-environment; and vector-gene-host-environment, he noted. Recommendations for Biomarkers/Analytical Chemistry 1. 2. Development of community-wide support for quality control and analytical standards. Foster support for the development of higher throughput strategies for biomarkers, including: improved preparative clean-up methods, higher throughput analytical instrumentation methods, and partnership with biostatisticians to develop better methods for the validation and application of biomarkers. Foster collaborations between major sectors (regulatory, academic, government, and industry) to identify the most promising disease relevant pathways requiring biomarker development through special congresses and workshops. These interactions could lead to collaborative studies initiated through mechanisms such as the SPORE, SCCOR programs and ultimately employ large epidemiological cohort studies. Specialized incentives to attract more chemists to the field during their post-graduate careers. ### 3. 4. 22