Drugs research project

central drugs research institute and performance enhancing drugs research questions and research and development drugs
Dr.ThorasRyder Profile Pic
Dr.ThorasRyder,Hong Kong,Researcher
Published Date:07-07-2017
Your Website URL(Optional)
Comment
Public and Private Sector Contributions to the Research & Development of the Most Transformational Drugs of the Last 25 Years A Tufts Center for the Study of Drug Development White Paper Tufts University School of Medicine • Boston, Massachusetts, USA JANUARY 2015 AUTHORS: Ranjana Chakravarthy Kristina Cotter Joseph DiMasi Christopher-Paul Milne Nils Wendel Sponsored in part by a grant from Pfizer Inc.Public/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 1 Executive Summary Much has indeed changed in the paradigm by which biopharmaceutical R&D is conducted since the authors undertook their first analysis of the relative contributions of the public and private sectors to the discovery and development of new medicines nearly two decades ago. What has not changed is that when you examine the drugs that have contributed the most and are still contrib- uting significantly to the health and well-being of the US and even globally, the role of the biophar- maceutical industry is pivotal in the translation from theory to therapy. In fact, two decades of reli- able analyses by academia and government, assessed using a variety of methodological approaches, consistently demonstrate that 67 percent to 97 percent of drug development is conducted by the private sector. In the current study, the authors examine a diverse array of evidentiary materials on the history of 19 individual drugs, 6 drug classes and 1 drug combination identified as the most transforma- tive drugs in health care over the last 25 years by a survey of over 200 physicians conducted by two Harvard Medical School researchers. The results of the analysis show that drug discovery and development is anything but a direct linear process. Instead it is a complex ecosystem with a wide range of novel collaboration archetypes, involving industry-academic partnerships, venture capital, disease foundations, as well as public-private, pre-competitive consortia, so that learning is from many disciplines and the result of multiple feedback loops. In fact, only 4 individual drugs appear to have been almost completely researched and developed by one sector, however, one sector or the other did dominate particular phases of the R&D continuum. For example, 54% of basic science milestones were achieved predominantly by the public sector, and 27% by the private sector. For discovery milestones, it was 15% by the public sector, and 58% by the private sector. The private sector was again dominant in achieving the major milestones for both the chemistry/manufacturing/controls and drug development phases, in 81% and 73% of the drugs reviewed, respectively. For 19-27% of the case histories in all categories, dominance of one sector versus the other could not be determined. The research that was done was expansive in its scope, often spanning oceans in the geographic reach of the institutions involved, as well as extend- ing over decades, an average of 25 years from discovery to approval. These drugs are having a positive and profound impact on healthcare even today. Nearly 70% of these drugs are now available as generics in the US, and are making contributions to medical sys- tems worldwide. In fact, nearly 40% of the drugs appear on the World Health Organization (WHO) Model List of Essential Drugs – a list of some 300 drugs, which are therapeutically important, affordable and generally in ready supply in appropriate formulations. In addition, nearly 40% of the drugs have orphan indications, thereby providing therapy for rare disease populations, likely for unmet medical needs for which there were few, if any, treatment options. Almost all of these drugs have been launched on a worldwide basis and are still actively in development for additional indi- cations or formulation improvements. The results of our analysis confirm just how critical the private sector is to the time and resource consuming process of drug development. While the basic science underpinning the key disciplines Public/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 2 needed to discover and develop drugs is often initiated in academia, it is pharma firms, in particular, where these disciplines grow to give the necessary critical mass, expertise and experience needed for successful drug discovery. Disciplines like medicinal chemistry, process chemistry and formula- tion, drug metabolism and pharmacokinetics, and safety sciences are practiced at a scale and level of competence and integration in the industry that far outstretch academic applications. But some remain skeptical of the role of the private sector in this important enterprise, and assert that it could and should be exclusively within control (at least financially) of the government. This begs the question: How Much Government Funding Would be Needed to Replace Industry New Drug R&D? In response, we made an effort to conservatively assess what the additional cost to govern- ment and taxpayers would be if such a radical policy change were made. To keep the analysis con- sistent with the period covered by the particular innovative compounds that we study in detail in this report, we initially restricted attention to new drugs approved from 1987 to 2002. The estimates suggest that conservatively the NIH budget would have to nearly double to maintain just the flow of the most innovative drug approvals, and would have to increase nearly two-and-half times to main- tain the development of all new drugs. The relative inexperience of government in the latter stages of the R&D continuum would likely result in the government spending significantly more on devel- oping new drugs than does the industry. We found even higher relative costs for drugs approved from 2003 to 2011. Given the trends for industry R&D costs and NIH budget appropriations, it is also likely that an analysis of more recent and future approvals would show more substantial increases in the relative cost of private sector R&D. Far from being a bystander prior to marketing, industry’s scientific contributions go beyond drug development and include basic and applied science, discovery technologies, and manufacturing protocols. With available funding from the public sector decreasing while medical needs and scien- tific complexity increase, private sector collaborations with academia and government have become increasingly key in furthering medical advancement. Public/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 3 Figure 1: Government and BioPharma Industry Investments Are Highly Complementary Roles of NIH and Private Sector in Biomedical Research U.S. Private Sector: 51.1.B PRODUCT Clinical Research Clinical Research Translational Translational Research Research Basic Research Basic Research IDEA National Institutes of Health: 31.B Batelle analysis for PhRMA: http://www.pharma.org/sites/default/files/pdf2014economicfutures-report.pdf National Institute of Health Office of Budget; http://report.nig.gov/categorical_spending.aspx I. Introduction In the 15 years since Tufts CSDD undertook its first in-depth review of the relative contribution of the public and private sectors to the research and development (R&D) of new medicines, much has changed. The funding available for biomedical R&D from the NIH has flat-lined, even decreased in relative terms, while the amount spent by the biopharmaceutical industry has doubled over the 1 same time period (see Figure 1). The biotech revolution and demographic evolution of major pharmaceutical firms set in motion a paradigm shift in company structure and behavior that have created a 21st century biopharmaceutical industry bearing little resemblance to last century’s big pharma hegemony. What has not changed is that when you consider the drugs that have contributed most significantly to the health and well-being of the US and even globally, the role of the biophar- maceutical industry is pivotal in the translation from theory to therapy. What has changed is that industry has implemented a wide range of novel collaboration archetypes in a unique approach combining inputs from industry-academic partnerships, venture capital, disease foundations, and public-private, pre-competitive consortia so that learning is from many disciplines and the result of multiple feedback loops. The cohort of drugs that we examined in the current study exhibited many instances of public-private interdependence, a feature of the evolving 1 A Research & Development Ecosystem for the 21st Century, Pfizer Presentation, May 2, 2014 INNOVATIONPublic/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 4 R&D paradigm that has become more pronounced over the last decade. The result was a medical armamentarium that was considered transformational for health care over the last quarter decade. Thus it is particularly ironic that one of the major criticisms of the biopharmaceutical industry voiced in certain quarters is that companies within the industry take credit for inventing a prod- uct without having actually contributed to its creation. Although this fiction endures, it has never been proven true and is at odds with two decades of reliable analyses by academia and government based on sponsorship, patent, project, and licensing data, as well as considerations of central scien- tific contribution to applied science, clinical improvement, and the development of manufacturing protocols that consistently demonstrate 67 percent to 97 percent of drug development is conducted 2,3,4,5,6,7 by the private sector. This fact in no way diminishes the public sector’s efforts in discover- ing innovative drugs and biologics; publicly funded research has been demonstrated to be vital for the advance of pharmaceutical science and improved medicines. Trends in R&D have been build- ing throughout the last several decades that serve only to increase the interdependence of the two sectors, making it the prominent feature of the landscape for biomedical innovation in the 21st Century. Yet, the criticism persists that industry only buys up the hard work of others, repackages it, and sells it at a premium to the public who funded the work to begin with, adding little value to the process or the products along the way. For years, critics have argued that important advances in medicine are a result of the efforts of public agencies. Publicly funded research is concentrated during the basic research phase. However, basic science is defined by the International Council of Science as “fundamental theoret- ical or experimental investigative research to advance knowledge without a specifically envisaged 8 or immediately practical application.” In the life sciences, basic research might encompass explo- ration of the biology of a disease, that can identify a protein, a receptor, or an enzyme (i.e., drug targets) implicated in the disease. This is a critical step but still a long way from becoming a new medicine. While the basic science underpinning the key disciplines needed to discover and develop drugs is often initiated in academia, it is pharma firms, in particular, where these disciplines grow to give the necessary critical mass, expertise and experience needed for successful drug discovery. 2 Department of Health and Human Services (DHHS), National Institutes of Health (NIH). (2001). Report to the United States Congress, NIH Response to the Conference Report Request for a Plan to Ensure Taxpayers’ Interests are Protected, July 2001. 3 Zycher B, DiMasi JA, Milne CP. Private sector contributions to pharmaceutical science: thirty-five summary case histo- ries. American Journal of Therapeutics. 2010; 17(1): 101-120. Retrieved from http://www.manhattan-institute.org/html/ mpr_06.htm. 4 Kneller R. The importance of new companies for drug discovery: origins of a decade of new drugs. Nat Rev Drug Discov. 2010; 9(11): 867-882 5 Sampat BN, Lichtenberg FR. What are the respective roles of the public and private sectors in pharmaceutical innovation? Health affairs (Project Hope). 2011; 30(2): 332-339. 6 Stevens AJ, Jensen JJ, Wyller K, Kilgore PC, Chatterjee S, Rohrbaugh ML. The role of public-sector research in the discov- ery of drugs and vaccines. The New England Journal of Medicine. 2011; 364(6): 535-541. 7 Lincker H, Ziogas C, Carr M, Porta N, Eichler H-G. Regulatory watch: Where do new medicines originate from in the EU? Nat Rev Drug Discov. 2014; 13(2): 92-93. 8 The value of basic scientific research. International Council for Science (ICSU). Dec, 2004. Available athttp://www.icsu. org/publications/icsu-position-statements/value-scientific-research. Accessed January 12, 2015Public/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 5 Disciplines like medicinal chemistry, process chemistry and formulation, drug metabolism and pharmacokinetics, and safety sciences are practiced at a scale and level of competence and integra- tion in the industry that far outstretch academic applications. Technology innovation also occurs mainly in pharma’s domain. High throughput screening, parallel chemistry, structure-based drug design, and the large-scale measurement of in vitro properties needed to design safe medicines with acceptable dosing frequency are capabilities not widely available in the academic setting, but which are de rigueur capabilities for private industry processes and practices. Several recent studies have supported the premise that while the public sector is often responsible for laying the basic science groundwork, the private sector provides the kind of applied research and development needed to get drugs approved for marketing. According to Sampat & Lichtenburg, “government funding has an indirect role in drug development – funding basic underlying research 9 that is then built upon.” In Stevens et al., the investigators make a distinction between the key contributions of the public and private sector in the R&D process of new medicines. This study describes public research contribution as “upstream,” meaning publicly funded research often provides insight for basic research, illuminates the mechanisms of a disease, as well as identifies pathways for therapeutic intervention. The private sector also contributes the basic science, but these contributions emerge more often during the later discovery stages for a specific product and succeeding stages of development that are necessary to bring those drugs to launch (which the 10 researchers define as “downstream”). Stevens et al. explain how the industry has evolved over the last few decades due to the emergence of biotechnology companies and major policy changes in the 1980s. The passing of the Bayh-Dole Act in 1980 allowed universities, nonprofit research insti- tutes, and teaching hospitals to own intellectual property and then have the ability to license those findings to whomever they chose (including private companies). Thus, a new system emerged in which the two sectors worked together to translate scientific findings into real products that can be 11,12 marketed. By looking closely at the development of some of the most innovative drugs of the last fifty years, researchers have begun to understand just how interdependent the innovative process for new medicines has become. This was amply demonstrated by the current authors in their prior work in conjunction with economist Ben Zycher from the Center for Medical Progress. That paper discusses the relative contributions to the R&D of 35 important drugs by the public and private sector in three crucial stages: basic science; applied science; as well as clinical, delivery and manufacturing 13 improvement. It found that the central scientific contribution by the private sector was evident in all categories, but most significantly to applied science, followed closely by its contribution to enhancing clinical performance and improving commercial production. Nonetheless, the authors 9 See Sampat & Lichtenburg at 4 10 See Stevens et al at 5 11 Reichert JM, Milne CP. Public and private sector contributions to the discovery and development of “impact” drugs. American Journal of Therapeutics. 2002; 9(6): 543-555 12 See Stevens et al. (2011) 13 See Zycher, DiMasi, and Milne (2008) at 3Public/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 6 acknowledge that the importance of publicly funded research cannot be downplayed, but that both sectors are crucial for advances in pharmaceutical science. Another study further elucidates how the traditional borders separating the two sectors have blurred over the years because both sectors are “challenged to show returns on their invest- 14 ments.” This study also emphasizes that over the last few decades drug discovery has evolved into a system that is a “complex chain of interrelated events and it involves an incremental learning pro- 15 cess that takes place over time.” The primary aim of the current paper is to follow up the 2008 analysis of 35 important drugs by Zycher, DiMasi & Milne. We focus here on a recently identified cohort of “the most transforma- tive drugs of the past 25 years” as determined by a survey of medical practitioners conducted by 16 two physician-scientists from Harvard Medical School in an article published in mid-2013. By examining the publicly available scientific literature, the private collection of the Tufts University research libraries, as well as CSDD’s leased and proprietary databases, we ascertained the relative contributions of the public and private sectors to the basic research, discovery, development, and production for 19 individual drugs, 6 drug classes and 1 drug combination identified by Kesselheim and Avorn. The following section briefly summarizes the history of how these transformative med- icines reached the marketplace by means of journeys that were often expansive in scope (countries and institutions) and extensive in time (usually decades), but nearly always with one commonality – their paths crisscrossed between both the public and private sectors. II. Methods & Results As discussed earlier, the current paper is a follow-up of a 2008 analysis, which tracked the same theme, but differed somewhat in approach in terms of the study cohort selection and methodology. We will briefly summarize them as they relate to the current study. The cohort in Zycher et al. was selected by merging lists from the literature of important drugs both from the perspective of impact on medical practice as well as utilization (i.e., numbers of prescriptions). Appropriate cohort selec- tion is especially critical for informative work on this subject matter. Certain characteristics are desirable: medicines that serve an important role in healthcare; retain their socio-medico impor- tance currently or have done so until fairly recently; were developed over a period of time (i.e., not all in a quick burst of public health urgency and extraordinary resource allocation such as with AIDS drugs in the decade from the mid-1980s to the mid-1990s); are broad-based in terms of therapeutic areas; and finally, were selected on the basis of fulfilling these criteria without any pre-selection bias. For these reasons, the authors chose to examine a cohort of drugs from a recently published work – The Most Transformational Drugs of the Last 25 Years (Kesselheim & Avorn, 2013) – that, generally speaking, demonstrated these characteristics. Since a detailed discussion of 14 Gelijns AC, Tosenberg N, Moskowitz AJ. Capturing the unexpected benefits of medical research. New England Journal of Medicine 1998 Sep 3;339(10):695. 15 Ibid 16 Kesselheim AS, Avorn J. The most transformative drugs of the past 25 years: a survey of physicians. Nat Rev Drug Discov. 2013; 12(6): 425-431.Public/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 7 the selected cohort is outside the scope and intent of this paper, the readers can judge for them- selves by perusing the paper, which was published in a major scientific journal by authors who have often been critical of the pharmaceutical industry. There are several advantages to this cohort of drugs compared to the one analyzed in 2008: less risk of selection bias; more narrow focus mostly on individual drugs over a more concentrated period of R&D; and, the drugs have been judged to be important by a survey of nearly 200 expert physicians across 15 specialties from 30 leading aca- demic medical centers. In terms of the methodology, we were informed by our prior studies but relied on a wider variety of resources that had become available to us over time: case files on individual drugs previously studied by Tufts CSDD; two previous analyses of impact drugs conducted by Tufts CSDD; data extracted from Tufts CSDD proprietary databases and commercial databases to which CSDD leases access; drugsfda; the Merck index; Google searches; as well as background literature from pro- fessional journals, trade press, textbooks and historical reviews of drug origins. The initial review was guided by certain criteria and categorical determinations provided by the senior authors and informed by prior CSDD research. It consisted of extracting a plethora of information on con- tributions to various milestones in the R&D history of the targeted drugs: disease process, drug target, mechanism of action, drug concept, isolation and purification, synthesis and early test- ing, patenting, lead optimization, pre-clinical studies, formulation and manufacturing protocols, clinical development, approval and launch. Upon second review, the data was condensed into a manageable quantum relegated to four categories that appear in Table 1 (basic research, discovery, chemistry/manufacturing & formulation/controls CMC, and development). Gaps in the available data were identified, and a preliminary assessment of which sector provided the dominant contri- bution for each R&D phase of each drug was undertaken. Sometimes this could not be determined because of data gaps or the complexity of assigning a dominant contributor to highly inter-related work. Research team members who conducted the first review of a particular medicine switched with other team members for second reviews, with a final review by the senior researchers, so that all drugs were “touched” by more than one researcher. Public/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 8 Table 1: Major Contribution (e.g., target identification/validation, patents, proof-of-concept, FDA approval) to R&D of Study Drugs with Breakdown by Sector and Phase Basic Research Discovery CMC Development Drug Name Alglucerase Private Public Private Public Clozapine Public/Private Private Private Private Epoetin Alfa Public Public/Private Private Private Epoprostenol Private Private Private Private Fluoxetine Private Private Private Private Imatinib Public Public/Private Private Private Latanoprost Public Public Private Private Lovastatin Public/Private Private Private Private Metformin Public Public Private Private Nitisinone Private Private Private Public/Private Omeprazole Private Private Private Private OnabotulinumtoxinA Public Private Private Public/Private Propofol Private Private Private Public/Private Remifentanil Public/Private Private Private Private Rituximab Public Private Private Public/Private Sildenafil Private Private Private Private Sumatriptan Public Private Private Private Tamsulosin Public Private Public/Private Private Zidovudine Public Public/Private Public/Private Public/Private Classes ACE inhibitors Public Public/Private Public/Private Private Anti-VEGF agents Public Private Private Private Biophosphonates Public Public Private Private HIV Protease Inhibitors Public/Private Public/Private Private Private Interferons beta-1b, 1a Public Public/Private Private Private TNF blockers Public/Private Public/Private Public/Private Public/Private Combinations Combined fluticasone and salmeterol Public Private Public/Private Private The results as seen in Table 1 basically represent a graphic retelling of the story line from the sum- mary case histories (see Appendix) – very few drugs went from theory to therapy without work being done on them by both the private and public sectors. In fact, only 4 individual drugs appear to have had been almost completely researched and developed by one sector – in the current study, it was the private sector (although not necessarily without any contribution by the public sector at all (epoprostenol) and sometimes because of a significant element of serendipity (sildenafil). Public/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 9 On the other hand, as seen in Table 2, all of the four phases of R&D are dominated by one or the other sector. For example, 54% of basic science milestones were achieved predominantly by the public sector, and 27% by the private sector. For discovery milestones, it was 15% by the public sector, and 58% by the private sector. The private sector was again dominant in achieving the major milestones for both the CMC phase and development phase, in 81% and 73% of the drugs reviewed, respectively. From 19 to 27% of the case histories in all four phases, dominance of one sector versus the other could not be determined. Table 2: Percent Contribution in 4 Phases of R&D by Public and Private Sectors Basic Discovery CMC Development 54% Public 58% Private 81% Private 73% Private Discovery and CMC comprise a series of complex and iterative processes, now more commonly referred to as the translational phase which include the activities described below Translational Phase Lead identification – to screen for lead compounds using structural and computational biology processes on various classes of targets such as receptors, proteins/enzymes, DNA and RNA/ribosomal targets Lead validation – to rapidly assess the therapeutic value (i.e., effect on disease-specific molecular cascade) of a large number of compounds on a given target from a combinatorial library or other compound collection, often by running parallel assays with high-throughput screening (HTS) Lead optimization – to determine the relationship of the physio-chemical properties and biological activity of the mol- ecule as well its PK/PD actions at the target site for modifying the molecule to maximize its desirable properties (efficacy) and minimize undesirable ones (e.g., side effects) Pre-clinical studies – to establish pharmacological profile, best route of administration, drug interactions; to understand the effects of a novel chemical entity in a complex organism; to better predict the new drug’s behavior in humans utilizing a large and diverse number of parameters such as absorption, distribution, metabolism and excretion (ADME), bioavailability, protein binding, stability and half-life, maximum serum concentration, as well as multicompartmental analysis of blood, liver, and other tissues Among the many important steps that occur during the translational phase (represented graphically in Table 2) is target validation – one of the greatest challenges in drug discovery. Basic scientific research carried out by academia, government researchers and industry explores the complex biology and causes of diseases and in doing so may identify a disease protein, a receptor or an enzyme (drug targets), that are implicated in the disease. In private sector research, efforts are largely focused on developing new medicines that act upon these receptors or enzymes in order to create improvements in the disease condition. However, not all discoveries of potential targets are Public/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 10 directly applicable to the development of new medicines. In fact, the vast majority of potential tar- gets discovered in basic science research must still be re-validated as the first step in the discovery research process. Thus the meaning of the term ‘translational’ phase becomes clear – the transfer of knowledge of underlying disease biology into a research hypothesis that is further explored and eventually confirmed by 10 to 15 years of discovery research, preclinical research, and clinical 17 development that hopefully leads to an important new medicine. As shown in Table 3, no particular therapeutic area dominates the cohort and a dozen or so of the major therapeutic areas are represented among the individual drugs, indicating that this cohort likely broadly reflects trends in drug development as a whole. Nearly 70% of these drugs are avail- able as generics in the US, and one is available as an OTC product as well, while another is available in Japan as a biosimilar. As valuable as these drugs are to medical practice, they are also making contributions to the value of the healthcare dollars spent in medical systems worldwide. In fact, nearly 40% of the drugs are listed on the World Health Organization Model List of Essential Drugs, which comprises the list of some 300 drugs considered essential for any particular country to be able to provide its population because they are therapeutically important, affordable and generally in ready supply in appropriate formulations. In addition, nearly 40% of the drugs have orphan indi- cations, thereby providing therapy for rare disease populations, likely for unmet medical needs for which there were few if any treatment options. Almost all of these drugs have been launched on a worldwide basis and are still actively in development for additional indications or formulation improvements. Although Tables 1-3 suggest the complexity and diversity of the research origins of our drug cohort, there are other spheres in which these defining characteristics are also evident – length of time and breadth of geography. While the headquarters of the sponsoring companies are located in just five countries, research took place in nearly two dozen countries. On the whole, R&D occurred from the 1950s through the 1980s, with approvals for initial indications occurring for the most part during the late 1980s through the early 2000s. On average, the time from initial discovery efforts to approval took 25 years. The basic research often had been going on for a decade or two prior to discovery, while further development on new indications or formulations continue to this day on many of these transformative drugs. Our compilation of case summaries on the drugs and drug combination/classes in our study revealed seven underlying themes that characterize each of their histories (and some cases evidence more than one theme): drug rescue; technical fix; screening programs; serendipity; spin-offs; drug champions; and sector-sharing. Drug rescue – A salient aspect of the story of clozapine is that of a drug abandoned for a time due to safety risks but rescued to address an unmet medical need owing to advocacy and incentive pro- grams provided by the public sector coupled with the perseverance, expertise, and resources of the private sector. Metformin highlights another example of the private sector continuing to pursue a 17 A Research & Development Ecosystem for the 21st Century, Pfizer Presentation, May 2, 2014Public/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 11 Table 3: Therapeutic Area (TA), Generic, WHO Essential Drugs List (EDL, April 2013), and Orphan Approval TA Generic WHO EDL Orphan Drug Name Alglucerase Metabolic N Y Clozapine Mental Health Y Y N Epoetin Alfa Blood Disorder N Y Epoprostenol Inflammation Y Y Fluoxetine Mental Health Y Y N Imatinib Oncology Y Y Latanoprost Ophthalmology Y Y N Lovastatin CV Y N Metformin Endocrine Y Y N Nitisinone Metabolic N Y Omeprazole GI Y/OTC Y N OnabotulinumtoxinA Nerve Disorder N N Propofol Anesthesia Y Y N Remifentanil Anesthesia N N Rituximab Oncology N N Sildenafil Uro/Gen Y N Sumatriptan Pain Y N Tamsulosin Uro/Gen Y N Zidovudine HIV/AIDS Y Y Y Classes ACE inhibitors CV Y Y N Anti-VEGF agents Oncology N N Biophosphonates Musculo-skeletal Y N HIV Protease Inhibitors HIV/AIDS N Y N Interferons beta-1b Immune Y- ex US Y Interferons beta-1a Immune Y- ex US Y TNF blockers Inflammation N N Combinations Combined fluticasone and salmeterol Respiratory N NPublic/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 12 drug with some challenges at the prompting of talented and visionary researchers in the broader medical research community. In similar vein, an example of perseverance when confronted with risk, albeit acceptable ones (due to its potential for meeting an unmet medical need) was shown on the part of the private sector by continuing to address medical concerns over a number of years until a seminal clinical study put the issue to rest for lovastatin. Technical fix – Although not as common or compelling a theme for this cohort of drugs as with the 2008 cohort, the history of epoetin alfa shows how the decades long advance of basic research and early discovery towards clinical application was stymied until the technical problem of producing it in sufficient quantity was worked out by the private sector. Similarly, while interferon too owed its discovery and development to a significant overlap between the two sectors, working out the chal- lenges of commercial production was the province of the private sector. Screening programs – A common theme that emerges from our reviews is that drug companies during this time period had ongoing screening programs that were proactively on the lookout by various means for candidate compounds to address unmet medical needs with commercial potential. This is how epoprostenol came about, and omeprazole as well. The latter, however, is a somewhat atypical story of primarily being engendered from the efforts of one sector, but exemplifies the productivity of pharma company screening programs of candidate compounds for common conditions that nonetheless qualify as unmet medical needs. Propofol, reminfentanil, and tamsulosin continue the theme of drug companies pursuing screening programs for candidate drugs to address apparent areas of unmet need, but in their case providing the resources in terms of funding and/or investigative compounds to clinical researchers at academic medical centers. While much of the basic research and discovery work was done in the public sector, GSK’s screen- ing program certainly accelerated the pace of development of zidovudine. The development of ACE inhibitors and sumatriptan appear to be examples of company screening programs picking up some promising leads from the basic research available in the public sector on population-wide health problems and facilitating public sector research until a candidate drug emerged. Serendipity – Somewhat in contrast to the role of screening programs in which companies are actively on the lookout for compounds to address identified needs, the ubiquitous scientific in terloper of serendipity makes an appearance in a couple of our case histories. The fact that nitisinone was originally developed as an herbicide shows how serendipity can play a role in the origin of drugs. However, as the saying goes, luck favors the prepared mind. The discovery of sildenafil was the result of both a good data monitoring program and an astute researcher not- ing an unexpected side effect that resulted in the melding of a commercial opportunity with an unmet medical need. Spin-offs – Crucial to the R&D efforts of alglucerase and representative of the interdependence of public and private sectors was Henry E. Blair of the New England Enzyme Center, based at Tufts University, who later co-founded a private company to focus on orphan drugs that became Genzyme Corporation, now one of the largest biotech companies in the world, which in turn Public/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 13 played no small part in precipitating the biotech revolution. The development of onabotulinum- toxin A is another example of research started in the public sector by the original inventors, who transformed their efforts into private sector enterprises, such as Miotech and Oculinum (later acquired by Allergan). Drug champions – Fluoxetine exemplifies that the decades-long journey of a drug from lab bench to market shelf requires many champions along the way both in the private and public sector, who oen k ft eep a program going by maintaining progress in one sector when it encounters roadblocks in another. While the initial discoveries relating to the causes of CML were publicly funded, the actual development of imatinib was the result of a fruitful collaboration between the private and public sectors, and another instance of drug champions moving between sectors to keep research from dying on the vine. Sector-sharing – The last and most prevalent of our themes is sector-sharing, or the tendency of drug histories to reveal that the primary role for moving a particular medical innovation forward oen in ft volves bi-directional feedback between the sectors with “sharing” occurring along paths that were both planned and unplanned. For example, while research performed in the public sector (at Columbia University) was absolutely critical for the development of latanoprost, a collaboration with the private sector allowed it to be developed into the blockbuster drug it became. This was the case with rituximab as well, which was primarily worked on in the public sector through the upstream R&D phases of basic research and discovery, and only later collabo- ratively in downstream studies with the private sector. This was the mirror image of the way the combination drug of fluticasone and salmeterol came about, with upstream studies demonstrat- ing that they were effective when used in combination being largely performed or funded by the private sector with Glaxo playing the dominant role. For downstream development, Glaxo worked collaboratively by funding clinical trials at various academic medical centers while it addressed drug delivery. Somewhat departing from this scheme were the Anti-VEGF agents with publicly-funded studies establishing the concept that angiogenesis is a critical aspect of tumor growth and a potential anti-cancer target, but private research early on led to the discovery of both the pro-angiogenic factor and a method to target its activity in a manner suitable for the clinic. The bisphosphonates were similar in that the concept that they could be used as a therapy for bone disorders began in publicly-funded laboratories in Switzerland, but collaborations with industry were absolutely critical for the development of the drugs that were eventually approved. HIV protease inhibitors were the mirror image of bisphosphonates, being developed primarily by the private sector, but dependent upon initial discoveries made by a combination of publicly- and privately-funded research. Lastly, both the identification of TNF-alpha as a target for autoimmune disease and the subsequent development of therapies that targeted it were the result of a number of collaborations between the private and public sectors. Public/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 14 III. Policy Implications The results of our analysis of case histories for the most transformative drugs over the last 25 years confirms just how critical the private sector is to the time and resource consuming process of drug development. But some remain skeptical of the role of the private sector in this important enter- prise, and assert that it could and should be exclusively within control (at least financially) of the government. This begs the question. How Much Government Funding Would be Needed to Replace Industry New Drug R&D? As an alternative to the current system of private sector biopharmaceutical R&D supported by intellectual property protection, some have advocated for replacing industry conducted and funded biopharmaceutical R&D with government funding the R&D process in full by either conducting 18 the R&D itself or directly contracting for it. Legislation to that effect had been proposed in the 19 United States Congress. Such a system would have to be supported by additional taxes. There are a number of serious drawbacks to such an approach. These include adverse selection in the disease categories emphasized if politics intrude on decision-making and inefficiencies resulting from the difficulty of administrators to judge the effectiveness of R&D activities and to align R&D objectives 20 with consumer demand. The nature of the current government grants process in life sciences also suggests that discovery and development would proceed less efficiently and more conservatively 21 than it does currently in the private sector. Discussions of the costs and benefits of replacing industry R&D with direct government control can be better informed with data. Consequently, we made an effort to conservatively assess what the additional cost to government and taxpayers would be if such a radical policy change were adopted. To have an analysis that is consistent with the period covered by the particular innovative com- pounds that we study in detail in this report, we first restricted attention to new drugs approved from 1987 to 2002. This period for approvals also dovetails nicely with a published study of R&D 22 costs per approved new compound incurred by industry. Given the number of new drug approvals over this period, we can then estimate the aggregate industrial expenditures incurred to obtain those approvals (inclusive of the costs of research failures) and compare the ongoing costs of sus- taining that level of output with the amount spent on life sciences research by the U.S. federal agen- cy that is overwhelmingly responsible for funding that research (the National Institutes of Health NIH). We also then apply this technique using a more recent R&D cost analysis and a more recent approval period (2003 to 2011). 18 Wright BD. The economics of investment incentives: patents, prizes, and research contracts. Am Econ Rev. 1983;93(4):691-707. 19 The Free Market Drug Act of 2004. H.R. 5155, 108th Congress, 2004. 20 DiMasi JA, Grabowski HG. Should the patent system for new medicines be abolished? Clin Pharmacol Ther 2007;82(5):488-490. 21 Leaf C. The Truth in Small Doses: Why We’re Losing the War on Cancer – And How to Win It. Chapter 10, Simon and Shuster, New York, NY, 2013. 22 DiMasi JA, Hansen RW Grabowski HG. The price of innovation: new estimates of drug development costs. J Health Econ 2003;22(3):151-185.Public/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 15 The DiMasi et al. (2003) study of industry R&D costs included estimates of the time costs of new Investigator panel Patient panel 12 drug development, along with estimates of the cash outlays (out-of-pocket costs). Since cash out- lays, in theory, can be capitalized at different rates depending on whether industry or government 10 is conducting the R&D, we restrict our attention to cash outlays. For a list of approvals, we utilized 8 some of the data for a study conducted by Food and Drug Administration (FDA) researchers that 6 23 grouped approvals of therapeutic new molecular entities (NMEs) into innovation categories. The 4 compounds included in the study were new drugs and biologics evaluated by the FDA’s Center for 2 Drug Evaluation and Research (CDER) from 1987 to 2011, excluding diagnostics, drugs approved 0 under the 505b(2) regulatory pathway, and compounds used only for military personnel. Elibility Frequency Study Volunteer Target Patient Criteria of Procedure Volunteer Receptivity Population Administration Compliance to Protocol The Lanthier et al. (2013) study grouped the new drug and biologic approvals into three innovation Procedures categories. The three categories are first-in-class, advance-in-class, and addition-to-class. First-in- class drugs represent pharmacological innovation, as drugs with new mechanisms of action, or other novel pharmacologic properties, are brought to market for the first time. Compounds in the advance-in-class category were not first-in-class approvals, but had received a priority review rating from the FDA (potential significant gain over existing therapy). The addition-to-class category includes all other approvals. The annual averages for the number of approvals in each of these cate- gories for our initial period of analysis (1987-2002) are shown in Figure 2. Figure 2. Therapeutic New Molecular Entity Approvals (1987-2002) by Innovation Category 30 27.9 25 20 13.9 15 10 7.4 6.2 5 0 First-in-Class Advance-in-Class Addition-to-Class All Sources: Lanthier et al., Health Aff, 2013;32(8):1433-1439; author calculations The combination of the first-in-class and advance-in-class categories accounts for nearly half of all the new drug approvals. For the purposes of our analysis, we combine first-in-class and advance-in- class compounds into what we shall call a “most innovative” category. We developed aggregate cost 23 Lanthier M, Miller KL, Nardinelli C, Woodcock J. An improved approach to measuring drug innovation finds steady rates of first-in-class pharmaceuticals, 1987-2011. Health Aff, 2013;32(8):1433-1439. Approvals per YearPublic/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 16 estimates from the published literature for both the most innovative category and for all approvals from 1987 to 2002. DiMasi et al. (2003) provides average cost estimates in year 2000 dollars for out-of-pocket costs for a period that corresponds closely with the approval period we are using. Thus, we use the figures in that study adjusted for inflation to year 2013 dollars by applying the same price index used for the study (GDP Implicit Price Deflator). Doing so yields an out-of-pocket cost per approved compound, inclusive of the cost of failures, of 526 million. The clinical period out-of-pocket cost estimate across all compounds is 368 million, while the pre-human cost estimate is 158 million. The DiMasi et al. (2003) study results allow for some differentiation based on the FDA review rat- ings of the approved products. The results for the period covered had average clinical period costs for approved drugs with a priority rating that were 33.5% higher than for drugs with a standard rating. We will assume that the higher relative costs for priority drugs carries over to failures. Then, utilizing the distribution of the drugs approved from 1987 to 2002 by FDA therapeutic rating 24 (43.3% priority and 56.7% standard), we can decompose the 368 million clinical period cost per approved drug over all drugs into 429 million for priority drugs and 321 million for drugs with standard ratings. Absent evidence about differentiation on the pre-human side, we assume that the overall average applies to both priority and standard drugs. This then yields estimates of total out- of-pocket costs per approved compound of 587 million for drugs with priority ratings and 479 for drugs with standard ratings. Given the above cost estimates for priority and standard drugs, and the numbers of approvals from 1987 to 2002 in the most innovative category and for all drugs, we calculated the cost of developing the most innovative drugs approved from 1987 to 2002 in aggregate to be 128 billion, and the cost of developing all of the drugs approved from 1987 to 2002 to be 234 billion. When considered on an average annual basis, these results amount to 8.0 billion per year for priority approvals and 14.7 billion per year for all new therapeutic approvals. R&D efforts typically continue after original new drug approval to test new dosage strengths and regimens, new formulations, new indications, and to meet regulatory post-marketing commitments. Thus, we can define lifecycle R&D costs as the sum of R&D expenditures prior to and post original approval. DiMasi et al. (2003) found that post-approval R&D cost per approved compound to be 34.8% that of pre-approval R&D cost. This implies a post-approval R&D cost per approved com- pound of 183 million. We have no evidentiary basis to distinguish between priority and standard drugs for post-approval costs, although it may well be the case that more post-approval R&D is done for drugs with priority ratings. However, we conservatively assume that post-approval cost per approved compound is the same whether the drug received a priority or a standard rating. Thus, the implied lifecycle cost per approved compound is 770 million for drugs with priority ratings and 662 million for drugs with standard ratings. These figures translate to lifecycle R&D costs of 24 Ten of the 446 approvals were biologics approved early in the period that did not receive FDA therapeutic significance ratings. We made judgment calls and assigned ratings for these compounds based on the Lanthier et al. (2013) innovation categorization and other considerations.2012 2010 2008 2006 2004 2002 2000 1998 1996 1994 1992 1990 1988 1986 1984 1982 1980 1978 1976 Public/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 17 169 billion for the most innovative compounds and 316 billion for all compounds. On an average lifecycle R&D cost per year basis, we then have 10.6 billion for most innovative compounds and 19.8 billion for all compounds. We can now compare industry costs to total NIH budget expenditures. We ask what would be the cost to government of assuming industry R&D expenditures so as to maintain a steady-state level of approvals consistent with what we have seen for the 1987 to 2002 period in comparison to NIH budget levels. First, we observe what total appropriations have been for the NIH by fiscal year from 25 1976 to 2013. Figure 3 shows those values in constant (year 2013) dollars. Figure 3. Inflation-Adjusted NIH Budget (2013 ), 1976-2013 35 30 25 20 15 10 5 0 Fiscal Year Sources: http://www.nih.gov/about/almanac/appropriations/part2.htm ; author calculations using the GDP Implicit Price Deflator 30 27.9 25 The initial year is significant for us as DiMasi et al. (2003) found a representative time profile for new 20 drug development of approximately 12 years. Thus, initial work on 1987 approvals would have begun, on average, in 1976. The average annual total NIH budget between 1976 and 2002 was 13.4 billion. 13.9 15 We consider the annual costs noted above as the amounts needed to maintain a steady-state of 10 7.4 6.2 approvals and compare them to the average annual NIH budget over the period analyzed. Figure 5 4 shows how much more taxpayers would have to pay to have government replace industry as a 0 funder of new drug development in relation to what it already pays to fund the NIH. Advance-in-Class Addition-to-Class All 25 For consistency, and taking the perspective that dollars spent by industry and taxpayers have alternative uses, we applied the same general economy-wide price index as was used for the industry R&D cost data. The Biomedical Research and Development Price Index (BRDPI) was developed using NIH inputs to reflect how much the NIH budget must change to maintain purchasing power for its activities. It generally shows more price inflation than does the GDP Implicit Price Deflator. Applying the BRDPI to the NIH budget data would show the same general pattern, but a lower rate of increase over time (for the portion of the period where real expenditures were rising) than is suggested by Figure 2. BillionsPublic/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 18 Figure 4. Additions to the NIH Budget Needed to Replace Industry R&D Funding of Therapeutic New Molecular Entities (1987-2002 approvals) 147.0% 150% Most Innovative All Compounds 120% 109.0% 90% 78.6% 60.0% 60% 30% 0% Pre-Approval R&D Cost Lifecycle R&D Cost Sources: Lanthier et al., Health Aff, 2013;32(8):1433-1439; author calculations The estimates suggest that the NIH budget would have to nearly double to maintain just the flow of the most innovative drug approvals, and would have to increase nearly two-and-half times to maintain the development of all new drugs. Note, though, that these are likely very conserva- tive estimates of how much extra it would cost government to replace what industry does. NIH 30 research-funded endeavors are generally not set up to meet the rigorous demands of regulatory approval authorities. This relative inexperience would likely result in the government spending 25 significantly more on developing new drugs than does the industry (at least for a significant period of time). Furthermore, as can be seen from Figure 3, in real terms subsequent to our period of analysis 20 NIH funding has been relatively flat, and even declining in some years. In contrast, drug develop- ment costs have increased significantly in real terms for decades, and, given data on increasing 15 clinical trial complexity and declining clinical approval success rates for recent years, it is likely that an analysis that covered a more recent period than we have analyzed would show substantially 10 7.4 6.2 26 27 higher industry R&D expenditures in relation to government-supported life sciences research. , 5 We turn now to an analysis that utilizes more recent development cost estimates and data on more 0 recent approvals. The Lanthier et al. (2013) data run to 2011 approvals. Thus, we focus on 2003 to First-in-Class Advance-in-Class Addition-to-Class All 2011 U.S. new drug approvals. A new study of private sector R&D costs (http://csdd.tufts.edu/files/ 26 Getz KA, Wenger J, Campo RA, Seguine ES, Kaitin KI. Assessing the impact of protocol design changes on clinical trial performance. Amer. J Ther. 2008;15:450-457. 27 DiMasi JA, Feldman L, Seckler A, Wilson A. Trends in risks associated with new drug development: success rates for investigational drugs. Clin. Pharmacol Ther. 2010;87(3):272–277. Relative to Average NIH BudgetPublic/Private Sector R&D Contributions to the Most Transformational Drugs of the Last 25 Years page 19 uploads/Tufts_CSDD_briefing_on_RD_cost_study_-_Nov_18,_2014..pdf ) corresponds approximately to this period. In 2013 dollars, the out-of-pocket cost per approved new compound is 1.395 billion. Of that amount, 965 million is associated with the clinical period while the pre-human cost esti- mate is 430 million. For the new analysis, standard rated drugs had higher average costs, and the share of approvals over the 2003 to 2011 period that received priority ratings was somewhat higher than for the prior period (48.2% priority and 51.7% standard). Applying the same methodological approach used for the 1987 to 2002 approvals, we decompose the 965 million clinical period cost per approved com- pound into 787 million for priority drugs and 1.132 billion for standard drugs. Under the same assumptions applied above, we apply estimates of total cost per approved compound of 1.217 billion for drugs with priority ratings and 1.562 billion for drugs with standard ratings. Given these estimates, we calculated that the cost of developing the most innovative drugs approved from 2003 to 2011 to be 156 billion, and the cost of developing all drugs to be 278 billion. On an average annual basis, the results are 17.3 billion for priority approvals and 30.9 billion per year for all new therapeutic approvals. The new R&D cost study also includes estimates of post-approval R&D. The post-approval out- of-pocket cost per approved compound is 466 million. Using the same assumptions noted above regarding the allocation of post-approval costs for priority and standard approvals, we estimate the lifecycle R&D cost per approved compound to be 1.683 billion for drugs with priority ratings and 2.028 billion for drugs with standard ratings. These values translate to lifecycle R&D costs of 212 billion for the most innovative compounds and 370 billion for all compounds. On an average annual basis, this implies that lifecycle R&D cost is 23.6 billion for the most innovative compounds and 41.2 billion for all compounds. Figure 5. Additions to the NIH Budget Needed to Replace Industry R&D Funding of Therapeutic New Molecular Entities (2003-2011 approvals) 200% Most Innovative All Compounds 158.7% 150% 90.0% 119.0% 100% 66.7% 50% 0% Pre-Approval R&D Cost Lifecycle R&D Cost Sources: Lanthier et al., Health Aff, 2013;32(8):1433-1439; author calculations Relative to Average NIH Budget