International Experiences in Quantitative Benefit-Risk Analysis to Support Regulatory Decisions

Abstract

Themed sections in Value in Health provide readers an opportunity to learn more about a contemporary issue in health care, research, or policy from several perspectives. This issue of Value in Health includes a themed section on patient-centered benefit-risk assessment to support regulatory decision making. Regulatory agencies and biomedical companies around the world largely have embraced patient-centeredness, the basic idea that addressing patients’ needs is of central importance to their activities. Articles by Klein et al. [], Mühlbacher et al. [], and Johnson and Zhou [] describe the context and the current procedures for considering the patient perspective in regulatory decisions in Canada, Europe, and the United States. As outlined in these articles, agencies in these regions are increasingly seeking information directly from patients about their experiences with medical conditions and how they weigh risks and benefits of new medical technologies to inform regulatory approval processes.
Health Canada, the European Medicines Agency (EMA), and the US Food and Drug Administration (FDA) have all initiated pilot studies, partnerships, and other initiatives to engage patients and caregivers and to improve transparency in regulatory decision making. Each of these regulatory agencies offers opportunities for patients or patient advocates to participate on advisory panels on an ad hoc basis. These agencies, however, differ in the extent to which they are formally assessing methods or inviting results from efforts to communicate acceptable risk-benefit trade-offs on behalf of various stakeholders.
Activities at the Canadian regulatory agency, Health Canada, have been somewhat limited. In 2014, Health Canada initiated a patient involvement pilot project to simulate how input from various stakeholders could be gathered and included in the drug submission review process. The project collected qualitative data from patients, caregivers, physicians, and patient groups, but was limited to an orphan drug context. As summarized by Klein et al. [], Health Canada is “assessing the overall suitability and feasibility of adopting, modifying or collaborating with other existing models such as those used by the FDA and EMA, and Health Technology Assessment (HTA) bodies.”
Mühlbacher et al. [] report that activities within Europe are primarily research-focused, aimed at assessing methods for evaluating benefit-risk trade-offs. The EMA produced a benefit-risk methodology report in 2010 presenting possible approaches to quantitative benefit-risk assessment followed by a pilot study in patients with multiple sclerosis. The EMA also coordinates the Innovative Medicines Initiative (IMI) in Europe, a private-public collaboration involving 34 partners. As part of IMI’s activities, the Pharmacoepidemiological Research on Outcomes of Therapeutics by a European Consortium [], active from 2009 to 2014, encompassed many aspects of benefit-risk assessment including reviews of alternative methods and visual representations for benefit-risk assessment. This work continued with several extension studies examining the use of various frameworks and methods to elicit preference weights to facilitate benefit-risk assessments. The IMI is continuing its exploration of methods with the Patient Preferences in Benefit-Risk Assessments During the Drug Life Cycle (PREFER) project. The 5-year initiative is expected to get underway in October 2016 with the aim to conduct studies on elicitation of patients’ perspectives on benefits and risks of medicinal products through a product’s life cycle to inform decisions by regulatory and health technology assessment bodies.
As part of the fifth authorization of the Prescription Drug User Fee Act (PDUFA V), under the goal of “[e]nhancing benefit-risk assessment in regulatory decision making,” the FDA agreed to initiate a public process to collect patient perspectives across a range of diseases. Activities by the FDA’s Center for Drug Evaluation and Research (CDER) appear to bridge those of the “wait-and-see” approach of Health Canada and the “study-the-methods” approach of the EMA. Johnson and Zhou [] report that CDER’s approach has thus far conducted more than 20 patient-focused drug development (PFDD) public meetings in which patients, caregivers, and other stakeholders are invited to provide testimony on the impact of their disease and treatments and what benefits matter most to patients. Although this initiative has been favorably received, the large number of organizations that wish to interact with the FDA exceeds CDER’s capacity. In response, CDER has invited patient groups to host external meetings and to generate “patient-initiated” guidance to inform the CDER’s regulatory decisions [].
The CDER has satisfied the PDUFA V goals by conducting public meetings and releasing documents to inform the community about each topic separately. The PFDD meetings brought patients and regulators closer together and provided a platform for patients and caregivers to convey how their disease impacts their lives in more depth than traditionally requested during application reviews for new drugs. An expectations gap, however, appears to have emerged. The patient community and advocacy organizations had hoped that the FDA would act on the knowledge gained during the meetings, but CDER had planned to participate only as a passive recipient of information from the community.
Although small in size in comparison with the CDER, the FDA’s Center for Devices and Radiological Health (CDRH) is not taking a backseat to the CDER with regard to its inclusion of patient preferences to inform regulatory decision making for medical devices. In May 2015, the CDRH kicked off a seismic shift in the use of patient preferences to inform regulatory decisions when it and the Center for Biologics Evaluation and Research issued draft guidance on using patient preferences to inform benefit-risk determinations for medical devices []. This guidance represents the first time a regulatory agency has explicitly invited sponsors to develop and submit quantitative evidence on patient preferences for consideration in benefit-risk assessments. The article by Ho et al. [] reports on CDRH’s collaboration with the Medical Device Innovation Consortium (MDIC) in a public-private partnership to develop a framework and to catalog methods that can be used to generate evidence on patient preferences. CDRH’s work with the MDIC signals a high level of commitment in incorporating patient perspectives in regulatory decision making.
Both the centers at the FDA will continue to strengthen their commitments to patient engagement initiatives through the next 5-year cycle covered by the user fee agreements: PDUFA for drugs and Medical Device User Fee Amendments (MDUFA) for devices. The CDRH has set forth aggressive patient engagement goals in its 2016 to 2017 strategic objectives, with 90% of its employees to be interacting with patients as part of their job responsibilities by the end of 2017 []. At the CDER, we expect to see initiatives that bridge from initial PFDD meetings to more systematic collection of patients’ input to inform drug development and benefit-risk assessment. We anticipate the CDER to take a more active role in exploring how various tools for patient input can be “fit for purpose” in the drug development and regulatory context.

Applications of Stated-Preference Methods

The MDIC Methods Catalog outlines a number of methods to quantify patients’ preferences regarding benefits and risks for medical technologies, provides examples of their use, and presents numerous issues that should be considered when selecting a method []. Although the MDIC framework stops short of recommending a specific preference elicitation method, stated-preference methods are prominently featured. These methods include discrete choice experiments (DCEs), best-worst scaling, and threshold techniques. In the context of benefit-risk assessment, stated-preference methods are used to generate metrics that can be used directly or as part of broader frameworks to compare benefits and risks of alternative treatments for collective groups of stakeholders. Nevertheless, the diverse applicability of these methods in health and medical care delivery is demonstrated in several additional articles published in the Value in Health themed section.
Articles by dosReis et al. [] and Hollin et al. [] report on stakeholder engagement processes that incorporate qualitative methods to identify and describe attributes to be included in stated-preference studies. These processes are crucial in conducting high-quality stated-preference studies, particularly when applied to a novel decision context. Nevertheless, for applications to support benefit-risk assessments, bottom-up approaches to identifying attributes on the basis of stakeholder engagement should be combined with top-down identification of benefits and risks for specific treatments that are of most relevance for regulatory decisions.
The article by Mühlbacher and Bethge [] also has implications for informing benefit-risk assessments. They applied a DCE to elicit patient preferences about the relative importance of achieving sustained virological response with hepatitis C treatment options relative to associated side effects and modes of administration. They demonstrated that patients considered the probability of side effects in addition to their severity when completing the DCE choice questions. In another article relying on DCE survey data, Dong et al. [] applied latent class analysis to examine preferences and willingness to pay for a pharmacogenetic test that could reduce an individual’s risk of a life-threatening adverse drug reaction. The authors identified two groups of patients with distinct preference patterns: those who were cost-conscious and those who were risk-averse. Finkelstein et al. [] also applied latent class models to data from a DCE to account for preference heterogeneity with respect to end-of-life care options among patients with cancer. These two studies demonstrate the potential for ecological fallacy in studies that have not explored heterogeneity in preferences across patients.
Recognition that preferences vary across individuals may prompt one to ask how these methods can be used to make regulatory decisions that apply to large groups of patients. In CDRH’s draft guidance, and as reflected in the MDIC framework, there is explicit recognition that risk tolerance varies greatly across individuals and that information will be important for product labeling and risk communication strategies. It will be fascinating to observe activities at the CDRH over the next couple of years as it interacts with medical device companies submitting data on patient preferences. We expect that other regulatory agencies will also be watching closely.

Authors

Shelby D. Reed Gabriela Lavezzari

Your browser is out-of-date

ISPOR recommends that you update your browser for more security, speed and the best experience on ispor.org. Update my browser now

×