Abstracts

SESSION I: Regulator's Perspective on Risk-Benefit Decisions

A United States Regulator’s Perspective of Risk-Benefit Considerations

John K. Jenkins, MD, U.S. Food and Drug Administration, Center for Drug Evaluation and Research, Silver Spring, MD

The U.S. statutory standard for establishing the benefit of a drug is substantial evidence of effectiveness for the drug’s intended use, as demonstrated by adequate and well-controlled clinical trials. In making decisions about marketing approval or withdrawal, this demonstrated benefit is weighed against the known risks, identified in the pre-market clinical trials and in post-market experience, and the uncertainty of unknown risks that have not yet been observed or identified. All drugs have risks, and depending on certain factors – such as the magnitude of the benefit, the severity of the disease being treated, and the availability and benefits and risks of other treatment options – the same risk may be tolerated for one disease, but not for another. For diseases where approved therapies exist, the question is often asked, ‘Why do we need another drug to treat the same disease?’, implying that at some point, society might determine that existing therapeutic options are sufficient or that new therapeutic options must be superior in some way to existing therapy. In the U.S. there is no statutory requirement that a new therapeutic be superior to existing options, however, regulatory decisions require an assessment of whether the benefits outweigh the risks, and the issue of choice is often an important factor.

The issue of choice has at least two facets. First, what is the perceived societal value of having additional choices available for treatment of a specific condition? Second, who should be making the choices? Should regulators make the assessments of whether the existing treatment options for a given disease are sufficient? Or should the regulator leave this decision to the medical community, and focus on ensuring that the new treatment is safe and effective for the indicated use? Given the rapidly growing science in areas such as pharmacogenomics, the issues underlying considerations of choice will evolve as we are better able to identify which patients will benefit from a therapy and which patients will be more susceptible to the risks. A regulator’s viewpoint on the issue of choice can play a significant role in decisions regarding the balance of benefits and risks of a product and its marketing status.

Science provides the data to inform our analyses of questions of benefit and risk, but it does not give us the answers. Regulators must make judgments on whether a drug’s benefits outweigh its risks when considering the population of patients for whom the drug will be indicated. This population-based decision must then be translated to the individual patient level where the prescriber and patient must make a well-informed decision. Even for an approved product, the benefits might not outweigh the risks for an individual patient. We rely on health care providers and patients to make individualized decisions about available medications and patient needs, and part of our responsibility as regulators is to communicate our knowledge of benefits and risks to help inform those decisions.

A European Regulator’s Perspective of Risk Benefit Considerations

Alasdair Breckenridge, MD, Medicines and Healthcare products Regulatory Agency, London UK

The regulation of medicines changed completely after the thalidomide episode. In the United Kingdom under the Medicines Act of 1968, new drugs had to show evidence of both efficacy and safety as well as of quality for the first time, but it was many years before a systematic approach to the analysis of risk and benefit even began to be considered. One of the main reasons for this was that the regulatory agenda was captured by issues of drug safety following major concerns about drugs such as antidepressants and non steroidal anti inflammatory agents. As part of this, the value of licensing clinical trials in predicting the safety and efficacy of medicines in general clinical use became a matter of debate leading eventually to the introduction of risk management plans in Europe. This gave additional impetus to the need for formalised assessment of risk and benefit. Further, as measurement of clinical and cost effectiveness became of increasing importance to payers of healthcare, the need for a framework in which risk and benefit could be measured has become more obvious, so that there are now proposals that this type of analysis be a mandatory part of regulatory submissions and post marketing evaluation of drug treatment.

Health Canada’s Proposal for Modernizing Drug Regulatory Decisions to Optimize Benefits, Minimize Risks and Manage Uncertainties

Robyn R. Lim, PhD, Office of Legislative and Regulatory Modernization, Health Products and Food Branch, Health Canada, Ottawa, Ontario, Canada

Legislating a benefit-risk-based evidence standard for product marketing consideration/reconsideration is proposed for a modernized, socially-aware approach to Canadian drug regulation. Required evidence will encompass foundational safety/efficacy/quality information and, where appropriate, additional elements providing circumstantial and practical meaning to drug performance metrics. Health Canada’s 2006 external consultations and internal examinations of best existing review practices seeded development of the framework that supports nuanced but explicit approaches to assessment and management of benefits, harms/risks and uncertainties. The re-focussed construct aims to improve judgement, decision and communications relevance and clarity to downstream health community partners, and so facilitate their own decision-making.

To contend with issue flux through time and with potential disparities in perceptions across health sub-communities, community participation in evidence generation and judgements is envisaged when important uncertainties and/or societal issues exist. Specific anticipated circumstances for broader involvement include: 1) “early” approvability of potential “breakthrough” treatments (with no available therapy/life-threatening conditions); 2) benefit-risk in fine balance and/or disputed by experts.

Perhaps assessments, judgements, decisions and communications can improve in the shorter term through regulatory community adoption of additional, non-legislative, qualitative approaches, such as: 1) a tailored safety management system, based on human factors research, to increase attention to potential cognitive errors, promote situational awareness and encourage reflection and discussion regarding personal and peer judgements; 2) training, modelled on legal paradigms, for preparation of narrative accounts of assessments and decisions; 3) visual representations development to support transparency in key issues identification and in interpretations of relative value of benefits, harms and uncertainties.

SESSION II: Making the Tough Calls: Current Approach and Case Studies

Making the Tough Calls – When a Drug Cause Rare and Serious Adverse Events, But Offers Only Symptomatic Benefits

Robert Temple, MD, Food and Drug Administration, Silver Spring, MD

Weighing a useful symptomatic benefit against a rare, but serious or adverse effect is actually quite a common problem. Only a few drugs (mainly in oncology, cardiology, infectious disease) actually influence survival and other outcomes, so that most drugs treat symptoms. When a rare serious adverse effect emerges, what determines what we do: withdraw, label (second line, Box) or lesser labeling.

1. Actual severity and rate (1/1000 is a bigger problem than 1/50,000)
2. Adequacy of alternatives, including:

• Other members of class
• Pharmacologically different products
• “Perfect” substitutes leads to WD – cerivastatin, hepatotoxic NSAIDs, SJS NSAIDs, troglitazone

Alternatives with liabilities tend to allow problem drugs to persist.
Pain is good example: APAP, NSAIDs, narcotics all have major, lethal liabilities and “only” treat pain. Yet they survive because of different problems.

An Industry Perspective: Case Studies of Benefit-Risk Assessment

Paul Coplan, DSc, Executive Director, Risk Management and Epidemiology, Purdue Pharma L.P., Stamford, CT

This presentation will address three topics: 1) An assessment of why structured, prospective approaches to benefit-risk assessment of medicinal products are important for drug developers; 2) A report on the development of a framework for evaluating whether the benefits of medicines outweigh the risks of medicines for regulatory decisions by a group of industry scientists from several companies under the auspices of PhRMA (the PhRMA Benefit-Risk Assessment Team Framework); and 3) A description of case study examples in benefit-risk assessment.

Can Modeling Assist Drug Regulators?

Lawrence D. Phillips, PhD, Department of Management, London School of Economics & Political Science, UK

The answer is “No” until agreement is reached about what constitutes “risks” and “benefits.” This is the conclusion from visits by the Benefit-Risk Project team of the European Medicines Agency to six European drug regulators, where we interviewed over 50 people and observed regulators in various meetings. This talk will explain how the EMA’s 80-Day Guidance Document for Assessment Reports was subsequently changed to explain what is meant by risks and benefits, and the initial response to that change. I will answer the question with a “yes” by demonstrating a model of when to approve the H1N1 vaccines in Europe. This exercise showed the potential for models to align decision makers, to create transparency in the balancing of benefits against risks, and to improve communication with physicians and the public.

SESSION IV: Decision and Risk Communication

Informed Clinical Decision Making: Need for Minimal Risk Reporting Criteria

Nananda Col, MD, MPP, MPH, FACP, Maine Medical Center, Portland, ME

Risk communication research has typically focused on how to communicate the risks of treatments, with relatively little attention given to the selection of which risks and benefits to communicate. Poor selection of risks to communicate has profound implications for informed decision making. Informed decisions require that patients understand their treatment options and likely outcomes so that they can determine which best fit their values and preferences. When medications can affect a wide range of clinical endpoints, communicating all of the risks and benefits of a single treatment, let alone all treatment options, would create confusion and impede decision making. Narrowing the discussion to a subset of the most important risks and benefits is necessary, but there are no widely accepted criteria for selecting or prioritizing risks and benefits. Risks can differ greatly in their clinical significance, frequency, attributable risk, and statistical significance. Presently, the measurement and reporting of risks is highly variable in clinical trials, systematic reviews, and decision aids, none of which typically report the criteria used to select the risks reported. Lack of criteria for reporting risks result in outcome reporting bias which can affect the conclusions of systematic reviews, the balance of risks and benefits for an individual treatment, and the choice of treatment. Developing minimal risk reporting criteria, analogous to Nutrition Facts, enables decision-makers to understand the most important risks of a particular treatment, to directly compare different treatments, and to synthesize benefits and risks. The FDA should support the development of risk reporting criteria and a platform for disseminating this information on the comparative effectiveness of treatments to the public.

Closing Keynote

Constructing Preferences

Baruch Fischhoff, PhD, Howard Heinz University Professor, Department of Social and Decision Sciences, Department of Engineering and Public Policy, Carnegie Mellon University, Pittsburgh, PA

With novel decisions, people sometimes do not know what they want, but must infer their preferences from basic values related to the choice at hand. When that happens, choices are vulnerable to manipulation by subtle aspects of how decisions are posed and responses elicited. Such threats to informed choices lurk in how terms are defined, options are displayed, language is used, and frames of reference are evoked. Decision-making methods should be evaluated empirically for their success in producing choices that are individually robust and collectively consistent. They should be aids to good judgment, rather than replacements for it. Decision-making methods should bear the burden of proof, for demonstrating the transparency of their assumptions and conclusions.

*Additional abstracts coming soon.