• Regulatory science encompasses a wide range of subjects, including not only disciplines traditionally associated with regulation, such as statistics and clinical research, but also disciplines outside the biomedical sciences such as economics, risk communication, and sociology.
• Regulatory science could be a methodological means of determining the im-pact and value of the rules, principles, and laws governing FDA-regulated research.
• A strong relationship between regulatory science and translational science could provide a path to creating a well-rounded discipline.
• Defining a regulatory science workforce includes defining, and making promising scientists aware of, regulatory science as an attractive, respected career option.
The discussions in the next session of the workshop recognized and were built from the premise that developing a discipline of regulatory science calls for defining what is meant by regulatory science and then building a workforce equipped with a set of core competencies to fit that definition.
In a dialogue, Barry Coller, The Rockefeller University, and Rob Califf, Professor of Medicine, Vice Chancellor for Clinical and Translational Research, Duke University Medical Center, described regulatory science
as seen through the lens of translational science. In a panel discussion, panelists provided observations about the core competencies needed for an effective regulatory science workforce and offered perspectives on the role of regulatory science in their respective sectors and agencies.
A Regulatory Science Taxonomy
In Coller’s view, regulatory science is a subset of translational science. He provided the following definitions as a basis for the discussion:
• Translational science is the application of the scientific method to address a health need.
• Regulatory science is the application of the scientific method to improve the development, review, and oversight of new drugs, biologics, and devices that require regulatory approval prior to dissemination.
Translational science traditionally has been broken down into four phases:
• T1: Discovery to candidate health application
• T2: Health application to evidence–based practice guidelines
• T3: Practice guidelines to health practice
• T4: Practice to population health impact
The taxonomy of regulatory science can be aligned with the translational science taxonomy through four analogous phases, as follows:
• RS1: Preclinical evaluation of safety and efficacy
• RS2: Clinical trial design and analysis
• RS3: Postmarketing review of safety and optimal utilization
• RS4: Health policies, including social aspects of regulatory science
From this taxonomy, it is possible to develop a list of the multidisciplinary research expertise needed in regulatory science. (See Box 3-3
1This section is based on the presentations by Barry Coller, Vice President for Medical Affairs, Physician-in-Chief, and David Rockefeller Professor, The Rockefeller University, and Rob Califf, Professor of Medicine and Vice Chancellor for Clinical and Translational Research, Duke University Medical Center.
for a list of regulatory science competencies identified during the course of the workshop.)
Coller commented that the eight priority areas in FDA’s 2011 strategic plan for regulatory science can be conceptualized within the regulatory science nomenclature of phases RS1 through RS4. For example, the FDA goals to modernize toxicology to enhance product safety, support new approaches to improve product manufacturing and quality, and ensure FDA readiness to evaluate innovative emerging technologies all fit within “RS1,” the first phase of the regulatory science taxonomy. Others, such as implementation of a new prevention-focused food safety system to protect public health and facilitate development of MCMs to protect against threats to U.S. and global health and security, are crosscutting issues that integrate several phases of regulatory science.
Califf articulated FDA’s definition of regulatory science (“regulatory science is the science of developing new tools, standards and approaches to assess the safety, efficacy, quality and performance of FDA-regulated products” [FDA, 2010]), which in his view is complementary to the definition offered by Coller but contemplates the conduct of activities that do not necessarily entail the application of the scientific method, such as policy development, and disciplines such as decision science, sociology, cognitive psychology, and behavioral economics. He also noted that regulatory science has multiple layers, and not every layer has the same goals.
Regulatory Science Training
It may be possible to create a training regimen for the regulatory sciences that builds on the existing translational science training programs, Coller said. Existing translational science meetings also may provide opportunities for disseminating new knowledge in regulatory science. Culturally, regulatory science could find an academic home within existing translational science centers and institutes. Similarly, existing journals focused on translational science may provide opportunities for publishing regulatory science scholarship. Building relationships between the regulatory and translational sciences may provide a path to creating a well-rounded discipline and earning the respect needed for any new field to succeed.
Califf applauded FDA for increasing the size of its workforce but added that the training systems for FDA scientists need to be improved. More thorough training should occur before FDA scientists start their duties at the agency so that they do not have to be trained so extensively on the job. FDA regulators need lifelong education, he added.
Many clinical researchers are ill at ease with regulatory science, said Califf. Much of this unease arises from the fact that regulations are not
well integrated with an understanding of what is needed to conduct first-rate clinical research. Researchers from the growing number of disciplines involved in clinical research, such as informatics specialists, data managers, psychologists, sociologists, economists, and even Institutional Review Board (IRB) members, must be given a better grounding in the relationship between regulation and clinical research, he said.
Even within the medical products industry, regulatory science training often occurs on the job. Califf suggested that organizations such as the Regulatory Affairs Professional Society (RAPS) become more involved in establishing a more formal mechanism for regulatory science training.
Several speakers called for a component of regulatory science to include the evaluation of regulations. The concept was discussed in depth by Clifford Lane, Deputy Director for Clinical Research and Special Projects, National Institute of Allergy and Infectious Diseases (NIAID), NIH. As defined by Lane, regulatory science is “the intellectual and practical activity encompassing the systematic study of the structure and behavior of the regulatory world through observation and experiment to determine the impact of the rules, principles, and laws governing FDA-regulated research.” This definition gets to the notion that regulatory science should address the value that regulations provide. Regulatory science first would look at the purpose of the original regulation and then generate a testable hypothesis about the impact of the regulation. Research then would examine how successful the regulation has been at achieving its original purpose, determine if the regulation produced any unintended or unanticipated consequences, and quantify the broadly defined cost of implementing the regulation. The analysis of the data would in turn provoke a discussion on the overall value of the regulation and lead to a conclusion about whether the regulation should be modified, eliminated, or left unchanged.
Lane highlighted one law and one regulation that could be tested using this research strategy:
• The FDA Amendments Act of 2007, Title VII, called for the expansion of ClinicalTrials.gov with the aim of enhancing patient enrollment and providing a mechanism to track subsequent progress of clinical trials. A testable hypothesis could be that the expansion
2This section is based on the presentation by Clifford Lane, Deputy Director for Clinical Research and Special Projects, NIAID, NIH.
of ClinicalTrials.gov to include a results database has enhanced patient enrollment and provided a way to track progress of clinical trials without generating excessive costs. Studies could compare the rates of enrollment by several metrics, such as the total number of patients in clinical studies, the percentage of studies filled within a specified timeframe, the number of published papers using data from ClinicalTrials.gov, and website utilization statistics. The impact of additional regulation could be measured by looking at new informed-consent language and additional staff needed to comply with the regulation.
• Title 21, Chapter 2, Subchapter D, Part 314, Subpart I addresses the approval of new drugs when human efficacy studies are not ethical or feasible, also known as the Animal Rule. The purpose of this regulation was to enable the licensure of products that have been studied for their safety and efficacy in ameliorating or preventing serious or life-threatening conditions caused by exposure to lethal or permanently disabling toxic biological, chemical, radiological, or nuclear substances; for which definitive human efficacy studies cannot be conducted because it would be unethical to deliberately expose healthy human volunteers; and for which field trials have not been feasible. A testable hypothesis could be that products in this category that would not have been licensed have been licensed since the regulation was developed. Research could identify potential products that fall into this category and assess the impact of the regulation on the ability of those products to be licensed.
In the subsequent discussion session, Califf commented that regulators need a well-developed understanding of why the regulations exist and what they are supposed to accomplish, particularly emphasizing the application of regulations to real clinical trials.
Coller noted that the regulatory process is part of the political process and that regulatory science should include efforts to better understand the relationship between the two. This has direct bearing on the research scheme that Lane proposed because it has an impact on how the end effect of regulation can differ from the intended effect. Mary Dwight, Vice President for Government Affairs, Cystic Fibrosis Foundation, amplified this comment by noting that political drivers are overwhelming regulatory concerns today. Patient education must be part of the solution to this problem so that patients can speak out about their needs for more effective, efficient therapeutic development models based on good regulatory science. Lane noted that policy makers are largely driven by data; provide them with good data and they will make scientifically sound decisions, but in the absence of data, they will make decisions driven by politics and
opinion, he said. Regulatory science can tilt this process in favor of good decision making by generating good data.
Two speakers presented case studies that could serve as examples of the practice of regulatory science. Dwight presented a case study involving the development of a therapeutic agent (see Box 3-1). Munir Pirmohamed, Deputy Director, Medical Research Council Centre for Drug Safety Science (CDSS), University of Liverpool, discussed a case study involving an issue of drug safety (see Box 3-2).
As part of a panel discussion of the core competencies that a regulatory science workforce should have, Steven Galson, Vice President for Global Regulatory Affairs, Amgen Inc., listed certain core competencies that would be helpful in addressing the types of research questions relevant to the impact of regulation on clinical research. (See Box 3-3 for a list of regulatory science competencies identified during the course of the workshop.) He added that FDA has long taken advantage of training opportunities at NIH by sending staff to work in clinics and laboratories there, but if FDA does expand its regulatory research, there may be a need to create a specialized division at FDA that funds and conducts this research.
Several panelists from federal agencies provided comments on regulatory science workforce capacity needs to carry out their agency missions.
FDA Center for Biologics Evaluation and Research (CBER). Carolyn Wilson, Associate Director for Research, CBER, FDA, commented that CBER regulates a broad spectrum of therapeutic biologics, including complex entities such as gene therapies, cell therapies, and xenotransplants. Many of these therapeutics cannot be terminally sterilized and may not even be subjected to methods that might remove or inactivate infectious agents, raising issues such as how to ensure the safety of these entities, determine appropriate preclinical animal models, and ensure that there are not species-specific toxicities or therapeutic responses. To deal with these issues, CBER needs scientists who are trained in a variety of broad, scientific disciplines, including immunology, biochemistry, cell biology, developmental biology, microbiology, genetics, and the new “omics” sciences. According to Wilson, this workforce needs excellent analytical skills, the ability to adapt to new technologies and research paradigms, and the expertise to apply findings in a way that is not “checkbox regulation.” Solid training in the scientific method is critical, along with experience doing team science.
Dwight described the Cystic Fibrosis Foundation’s role in a successful collaborative effort to develop new therapeutics to treat cystic fibrosis (CF). The key to this effort was that the partners stayed focused on the desired goal but were flexible enough to adapt to changing circumstances experienced during the drug development process. She recounted how the mission of the foundation has changed since its founding in 1955, from caring for patients to finding treatments for the disease. This change reflects the tremendous advances in understanding the molecular basis of disease that have occurred since the discovery of the CF gene in 1989.
Collaboration has always been a core tenet of the CF community of patients and their parents; physicians, nurses, nutritionists, respiratory therapists, and social workers; and researchers. The tremendous advances in life expectancy that have occurred can be traced in part to a team approach to patient care.
Collaboration among three teams of researchers played a key role in the discovery of the CF gene. More recently, the development of promising therapeutics has been a result of collaboration among scientists in academia, industry, and FDA. Each of these collaborations required a cultural change. In the case of drug development, research teams needed to learn to share data among themselves and with pharmaceutical companies. At FDA, regulators needed to adapt their concepts of risk to recognize that risk has a different definition for patients suffering a certain early death from their disease without new treatments. The Cystic Fibrosis Foundation, Dwight explained, played the role of facilitator, coordinating and encouraging communications among all of the groups participating in this endeavor.
Dwight noted that FDA was very responsive to the particular needs of this work. In particular, large multicenter clinical trials would have been difficult to conduct given the patient population. FDA also strengthened its staff expertise and facilitated communications with trial sponsors.
However, the process is still too slow for patients living with chronic, life-threatening diseases, Dwight said. Direct communication between patient groups and regulators needs to be enhanced to inform how FDA balances risk and reward when it approves the design of even the earliest stages of the clinical trials processes.
aBased on the presentation by Mary Dwight, Vice President for Government Affairs, Cystic Fibrosis Foundation.
FDA Center for Tobacco Products (CTP). Regulatory science sits at the core of what the CTP is now charged to do to protect the nation’s health, but until now the country has never attempted to create science-based regulations for tobacco products. Doing so requires a pool of professionals in the biological and chemical sciences, toxicology, pharmacology, and product engineering, said Laurence Deyton, Director of CTP, FDA. It also
Adverse drug reactions impose a tremendous burden on human health. They account for some 2.5 percent of emergency room visits and 6.5 percent of hospital admissions in the United Kingdom and are a major problem for the pharmaceutical industry. Between 1990 and 2005, FDA and the European Medicines Agency (EMA) ordered 24 drugs to be withdrawn from the market because of adverse drug reactions. Most of these withdrawals occurred not long after the drugs reached the market, long before the costs of developing those drugs were recovered.
To better understand the fundamental biochemical mechanisms underlying adverse drug reactions, the Medical Research Council established the Centre for Drug Safety Science (CDSS). In addition to developing better methods for predicting adverse drug responses, CDSS aims to train researchers in the science of drug safety. Toward this end, the CDSS has established both research and clinical pharmacology training fellowships that focus on drug safety and personalized medicine. The center also offers master’s and Ph.D. degrees in drug safety science.
An important piece of the center’s training mission focuses on the collaborative relationships that CDSS has formed with regulators, academia, industry, and public advocacy groups. When CDSS identifies a research question, staff identify the appropriate clinical networks in the United Kingdom and develop collaborative hypothesis-testing research programs. The center then holds workshops involving academics, regulators, industry scientists, and health care officials to disseminate the results of those programs and develop recommendations to guide regulators.
As an illustration of CDSS’s approach, Pirmohamed discussed the development of a new genetic biomarker for carbamazepine hypersensitivity in Caucasians. He described some of the research that identified this marker and then discussed the implications of these findings. Regulators, for example, need to consider that this biomarker was validated in at least three populations, but all from case-control analyses, not prospective clinical trials. Regulators can consider changing the prescribing label for this drug to require that all Caucasian patients be tested for the biomarker before the drug is prescribed, or they might simply provide this information to physicians. Only through the application of good science, said Pirmohamed, can regulators make good decisions about such issues.
aBased on the presentation by Munir Pirmohammed, Deputy Director, Medical Research Council CDSS, University of Liverpool.
requires public health experts, medical professionals, lawyers, educators, communications specialists, and behavioral scientists, all of whom understand and appreciate the role of regulatory science as it applies to tobacco product regulation, said Deyton.
Centers for Disease Control and Prevention (CDC). Melinda Wharton, Deputy Director, National Center for Immunization and Respiratory
• Basic investigation
• Clinical investigation and clinical trial design
• Clinical pharmacology
• Clinical research operations
• Decision theory
• Drug/device discovery and development
• Drug disposition and metabolism
• Information technology
• IRB experience
• Medical informatics
• Monitoring and quality assurance
• Pharmacology (whole animal)
• Protection of human subjects
• Public health
• Regulatory knowledge
• Research pharmacy
• Risk assessment and communication
• Systems analysis
• Systems biology
• Technology transfer
aThis box provides an integrated list of disciplinary components of regulatory science offered throughout the workshop by speakers and audience members.
Diseases, CDC, discussed the type of workforce CDC needs to conduct regulatory science, particularly in assessing risks and benefits. CDC’s list of required disciplinary expertise encompasses epidemiologists, biostatisticians, and laboratory scientists such as microbiologists, chemists, and toxicologists who focus on the identification and quantification of disease burden. CDC also needs health economists who can study cost effectiveness, she said, along with risk communications experts. She emphasized the need for all staff to be comfortable working in a collaborative environment across disciplines and with external investigators.
Galson remarked that an understanding of how payers make reimbursement decisions is also a critical competency, given that many decisions about how new therapies will be used are being made not by regulators or physicians but by those who pay for these therapies. Increasing the pool of scientists trained in regulatory science who can conduct comparative effectiveness studies would benefit the entire field. A participant commented that FDA commissioned Duke University’s business school to teach a course that included modules on funding drug development, pricing, and reimbursement.
A member of the audience with prior experience at FDA noted that a critical skill needed by FDA’s workforce is an understanding of the difference between predictability and probability, noting that FDA makes probability-based decisions. He noted that ability to conduct quantitative analyses is a core competency, emphasizing bioinformatics, statistics, and other quantitative sciences.
A number of definitions of the term—and discipline—of regulatory science were submitted by various speakers throughout the course of the workshop. As described earlier, Coller conceptualized regulatory science as falling along a set of phases analogous to those recognized in translational science. A participant observed that Lane’s definition to include the evaluation of regulations could be seen to fall within the “RS4” stage in Coller’s taxonomy. Other definitions offered related, complementary perspectives on the definition and components of regulatory science.
Alastair Wood, Partner and Managing Director, Symphony Capital LLC, remarked that, although science includes them both, innovation and discovery are different things. Implementation, including adopting, understanding, using, and modifying knowledge that already exists, is also distinct from the process of discovery and from innovation. Wood suggested that defining the science—and developing and training the workforce to practice the science—should acknowledge and focus on the difference in the domains involved, which demand different styles,
and these skills, in turn, could be acquired in different settings (e.g., a focus on discovery in academics; a focus on innovation, implementation, and adoption in industry).
Carl Peck, Professor of Pharmacology and Medicine, University of California, San Francisco (UCSF), distinguished regulatory science from regulatory research. Regulatory science, he said, is the entire body of knowledge practiced by FDA and by those regulated by FDA, including law, economics, and an overriding ethic of protecting the public health. Regulatory research is the development of that body of knowledge as well as new tools, standards, and approaches to assess the safety, efficacy, quality, and performance of FDA-regulated products.
Several participants noted that it may not be necessary or desirable to conceptualize regulatory science as a single, stand-alone discipline. It was suggested by a participant that all facets of therapeutics development are subject to science-based regulation. On this basis it was suggested that regulatory science could be viewed not as a freestanding discipline but rather as a subspecialty within every core discipline forming the basis of drug development science. Others conceptualized regulatory science as a multidisciplinary effort, and several workshop participants called for the establishment of academic “homes” that would centralize and support the workforce engaging in the practice of the regulatory sciences.
It was also noted that the fact that there is not a commonly agreed definition of regulatory science should not necessarily be seen as a barrier for advancing the field. Rather, it is an opportunity to create a multi-component discipline that is adaptive and responsive to the needs of the field.
This page intentionally left blank.