Emerging Safety Science: Workshop Summary (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

1 Introduction I n recent years, the costs of new drug development have skyrocketed. The average cost of developing a new approved drug is now estimated to be $1.3 billion (DiMasi and Grabowski, 2007). At the same time, each year fewer new molecular entities (NMEs) are approved. DiMasi and Grabowski report that only 21.5 percent of the candidate drugs that enter phase I clinical testing actually make it to market. In 2007, just 17 novel drugs and 2 novel biologics were approved. In addition to the slowing rate of drug development and approval, recent years have seen a number of drugs withdrawn from the market for safety reasons. According to the Government Accountability Office (GAO), 10 drugs were withdrawn because of safety concerns between 2000 and March 2006 (GAO, 2006). Finding ways to select successful drug candidates earlier in development could save millions or even billions of dollars, reduce the costs of drugs on the market, and increase the number of new drugs with improved safety profiles that are available to patients. Emerging scientific knowledge and technologies hold the potential to enhance correct decision making for the advancement of candidate drugs. Identification of safety problems is a key reason that new drug develop- ment is stalled. Traditional methods for assessing a drug’s safety prior to The planning committee’s role was limited to planning the workshop, and the workshop summary has been prepared by the workshop rapporteurs as a factual summary of what occurred at the workshop.  

 EMERGING SAFETY SCIENCE approval are limited in their ability to detect rare safety problems. Prior to receiving U.S. Food and Drug Administration (FDA) approval, a drug will have been tested in hundreds to thousands of patients. Generally, drugs cannot confidently be linked to safety problems until they have been tested in tens of thousands to hundreds of thousands of people. With current methods, it is unlikely that rare safety problems will be identified prior to approval. There is, however, an emerging safety science that seeks to change this paradigm by attempting to understand a drug’s safety or toxicity ear- lier in its development. This emerging science is focused in two areas. One is the use of various basic sciences, including genomics, metabolomics, pharmacogenomics, and others, to understand the mechanisms underly- ing toxicity and to predict when a particular compound will have safety issues. The other is the use of new analytical tools for mining large data sets to identify signals that indicate safety problems (e.g., those associated with a class of drugs, those associated with particular molecular entities, or those associated with particular genetic profiles) and even to derive insights regarding a drug’s mechanism of toxicity. The application of emerging science to drug safety is one of the goals of the FDA’s Critical Path Initiative. A 2004 FDA white paper, Innovation or Stagnation: Challenge and Opportunity on the Critical Path to New Medical Products, describes this evolution as follows: Not enough applied scientific work has been done to create new tools to get fundamentally better answers about how the safety and effective- ness of new products can be demonstrated, in faster time frames, with more certainty, and at lower costs. In many cases, developers have no choice but to use the tools and concepts of the last century to assess this century’s candidates. As a result, the vast majority of investigational products that enter clinical trials fail. . . . A new product development toolkit—containing powerful new scientific and technical methods such as animal or computer-based predictive models, biomarkers for safety and effectiveness, and new clinical evaluation techniques—is urgently needed to improve predictability and efficiency along the critical path from laboratory concept to commercial product. (FDA, 2004:5) Since the publication of that report, significant progress has been made in the development of just such techniques. But the diffusion of these innovations in drug development and drug review has been lim- ited. To address this concern, the Institute of Medicine’s Forum on Drug Discovery, Development, and Translation sponsored a public workshop— Emerging Safety Science—with the goal of surveying new technologies that can be used to better understand and predict the safety and toxicity of new drugs. The workshop was held April 23–24, 2007, at the FDA’s White Oak Conference Center. 

INTRODUCTION  The workshop addressed two general approaches to safety science. Speakers on the first day discussed the use of basic-science approaches to understand the effects of various compounds on the body, with the ultimate goal of being able to predict which compounds will exhibit which safety problems in humans. During these sessions, speakers also considered the current and foreseeable difficulties/challenges involved in developing these approaches. These included • the current limitations of using animal models to predict human toxicity; • the likely complexity of underlying toxicity mechanisms and pre- disposing human factors, as well as challenges in defining and modeling their interaction; • the need for (and difficulty of) validating biomarkers, and the necessity of confirming potential toxicity biomarkers with human data; • the inherent difficulty of dealing with idiosyncratic (and rare) events; and • the need to maintain a balanced perspective so that drug candi- dates are not discarded prematurely based on the potential for toxicity alone. Speakers on the second day focused on new ways of obtaining and analyzing postmarket data to identify safety problems more rapidly once drugs are marketed. Discussion during these sessions focused on how deficiencies in the current systems available for detecting and evaluating adverse events could be improved and on the development of new meth- ods for monitoring postmarket drug safety. Proposals for fundamental changes in how adverse event data are collected, shared, and analyzed were presented during this part of the workshop. Throughout the workshop, participants emphasized that the ultimate goal of applying these new technologies in safety science is to create a continual, iterative process in which basic scientific data can help inform and predict clinical outcomes, and clinical outcomes can be used to inform and corroborate the basic science. This report summarizes presentations and discussions at the work- shop, which should serve as a useful survey of current and emerging tools in the drug safety armamentarium: • Chapter 2 sets the stage by describing the current state of the art in investigative toxicology, including innovative ways of using traditional methods. • Chapter 3 is the first of four chapters devoted to emerging screen- ing technologies. It describes cell-based screening methods (in vitro exper- 

 EMERGING SAFETY SCIENCE iments conducted using human cells) and their uses in lead identification and optimization (the process used by companies to identify and select the candidate[s] most likely to succeed throughout the development pro- cess), safety evaluation, and off-target activities, as well as in clinical prediction and exploration of putative biomarkers. • Chapter 4 reviews various uses of and methods for toxicogenomics (the conduct of gene expression analyses to help predict the toxic effects of compounds and provide insights into the mechanisms of toxicity). • Chapter 5 describes how metabolomics (the detection and quantifi- cation of small molecules, or metabolites) is being used to gather informa- tion on drug toxicities and their underlying mechanisms. • Chapter 6 considers drugs that are toxic in only a subset of patients. Using the case of the anti-HIV drug Abacavir, it describes how phar- macogenetics (the study of genetic variations that affect an individual’s response to a drug) can be used to identify these patients so as to prevent or at least anticipate toxicity. • Chapter 7 presents a case study involving the experiences of the Predictive Safety Testing Consortium, formed by the C-Path Institute to bring industry, academia, and the FDA together to investigate qualify- ing nephrotoxicity biomarkers (quantifiable biological responses that can provide information on disease states or drug responses) for use in safety testing. • Chapter 8 describes new approaches to pharmacovigilance (the process of collecting, monitoring, and evaluating adverse event data from patients and health care providers to identify drug safety issues). These approaches include an online signal management program, new methods for analyzing data from the FDA’s Adverse Event Reporting System, and proposals for a large-scale active surveillance network. • Chapter 9 considers how to integrate the various approaches to safety science and create feedback loops that will allow information to be shared throughout the system. Means of achieving such integra- tion include building interdisciplinary knowledge; creating databases that allow easier identification of associations between compounds and adverse events; understanding the relevance of animal models; and devel- oping “bridging” biomarkers that can bridge, or translate, early preclini- cal findings to clinical findings. • Finally, Chapter 10 addresses areas in which further work is needed and outlines possible next steps.