The U.S. Food and Drug Administration (FDA) issued a draft guidance, Benefit-Risk Assessment for New Drug and Biological Products, on September 30, 2021.1 The draft guidance is groundbreaking in its articulation of a vision of drug development that many entities involved in developing innovative new drugs and biological products should review it carefully — particularly precommercial companies whose approach emphasizes efficiency and permits some residual uncertainty with respect to clinical benefits and safety issues.
As reflected in the draft guidance, FDA expects sponsors to focus their development activities in such a way that the data and information generated by those activities will minimize uncertainty in the benefit-risk analysis conducted by FDA during its review of new drug applications (NDAs) and biologics license applications (BLAs). Because FDA anticipates that clinical development will be structured in this manner, the agency also signals a relatively low tolerance for uncertainty, even in the area of rare disease drug development where patients express a desire for access and willingness to accept risk. Notably, FDA also puts industry on notice that active-control trials will be required in some cases, representing a departure from the established principle that FDA cannot require comparative efficacy data as a condition of approval. Comments are due November 29.
The Draft Guidance — A Groundbreaking Articulation of the Benefit-Risk Paradigm
Expectations of Sponsors: Structured Benefit-Risk Planning
The draft describes the structured planning that sponsors should undertake to reduce uncertainty and demonstrate that the product has a favorable benefit-risk profile. FDA expects sponsors to actively and purposefully consider incorporating benefit-risk considerations into development activities. FDA also expects sponsors to engage in structured benefit-risk planning throughout the lifecycle of every drug product, including even those that do not present challenging benefit-risk issues.
Further, FDA expects sponsors to design their clinical trials and other studies to evaluate the most important potential benefits and risks of a drug early in development. For example, a sponsor might need to collect dose exposure response data prior to phase III to identify doses that can optimize benefit relative to risk and inform dosing recommendations. In general, FDA expects sponsors to focus their development activities in such a way that the data and information generated by those activities will facilitate the agency’s benefit-risk assessment.2
In the area of rare disease drug development, FDA indicates that sponsors should maximize the potential for clinical trials to provide interpretable scientific evidence about benefits and risks starting at the earliest stages. They should also seek to minimize bias and maximize precision with trial design features such as randomization, blinding, enrichment procedures, and adequate trial duration. The draft affirms that the approval standard remains the same — substantial evidence3 — for all drugs, including those for rare diseases.
FDA’s Approach to Benefit-Risk Assessment
Benefit-risk assessment requires an informed judgment as to whether the benefits (with their uncertainties) of the drug outweigh the risks (with their uncertainties and the potential for managing such risks) under the conditions of use described in the approved product labeling. According to the draft, FDA will conclude that benefit outweighs risk where (1) there is robust registration-quality data demonstrating that the product will, to a high degree of statistical confidence, have its intended, clinically significant effect, and (2) the safety of the drug has been fully analyzed with no serious risks identified. In this scenario, FDA’s benefit-risk assessment is straightforward.
But many scenarios are not straightforward. For example, a drug’s benefit may be supported by scientific evidence from well-designed, properly executed studies, but the results may not be statistically significant in FDA’s view, or there may be questions about the validity or clinical relevance of one or more endpoints. Likewise, a drug’s safety may not have been fully characterized — for example, the sponsor may have identified a signal of a serious risk without having confirmed the likely frequency of its occurrence in clinical practice.
Where benefits or risks are not sufficiently characterized, FDA cannot determine whether the benefits to the indicated population will outweigh the risks of the drug following approval. In such cases, the draft guidance explains, FDA considers the available evidence, data gaps, severity of the condition, patient population, and current treatment landscape. As described in the draft guidance, a structured approach to benefit-risk assessment facilitates the agency’s analysis of these considerations.
A key aspect of FDA’s review of an NDA or BLA involves regulatory options available to mitigate risks. FDA identifies two such options — labeling content and risk evaluation and mitigation strategy (REMS) — and distinguishes risk mitigation measures from regulatory options to reduce uncertainties. Examples of the latter include requiring additional clinical studies to improve the characterization of safety, efficacy, or dose response. FDA will use safety labeling or REMS techniques to achieve a favorable benefit-risk assessment only where the risk profile of the product has been adequately characterized. Accordingly, sponsors should not expect to address unresolved questions about a risk that is clinically significant by proposing labeling language or a REMS, as FDA would be more likely to require the sponsor to reduce the uncertainty through further studies.
FDA may require a sponsor to demonstrate that risk mitigation measures are effective in keeping risks to an acceptable level after approval. Accordingly, a sponsor may be expected to include REMS elements in the design of clinical investigations and to submit data to FDA supporting the usefulness of those measures in its application. Where a drug is associated with serious risks in a broader population, FDA may achieve a favorable benefit-risk ratio by limiting the labeled indication to a subpopulation for whom the benefits outweigh the risks. Such a subpopulation may be characterized by disease severity or by genetic, pathophysiologic, or historical factors.
The appropriateness of applying one or more of these techniques to address a serious risk is determined by FDA, not the sponsor — though the burden of demonstrating the adequacy of risk mitigation measures is the sponsor’s.
Uncertainty in the Assessment of Benefit and Risk and Rare Diseases
As the draft guidance explains, some uncertainty in the assessment of a drug’s benefit — for example, effectiveness in long-term use — is unavoidable. But FDA expects sponsors to minimize such uncertainty through careful study design and conduct. Uncertainty can also be reduced through regulatory tools (e.g., requirements for additional studies conducted prior to or after marketing to further characterize efficacy). FDA, not the sponsor, determines whether benefit has been adequately demonstrated in the NDA or BLA.
Uncertainty in the assessment of benefit may be more acceptable with respect to a drug that is intended to treat a serious disease with unmet needs, for instance, a drug being reviewed through the accelerated approval pathway. Likewise, FDA may tolerate greater uncertainty of benefit and risk for a drug that is being developed to treat a rare disease. Uncertain benefit may be acceptable for rare diseases if there are few or no approved therapies. In any case, however, the statutory standard of substantial evidence of efficacy must be met.
Where serious risks are anticipated with use of a drug intended for a serious or life-threatening disease or condition, FDA will grant approval if the drug has a clearly demonstrated, direct, and meaningful benefit on the most important clinical outcomes and the drug meets an unmet need.4 Alternatively, FDA could determine that the drug provides a specific important advantage over currently available therapies. To demonstrate such an advantage, a sponsor would have to conduct active-control trials that provide statistically significant evidence of clinical superiority to at least one currently available therapeutic option.
In addition, the draft guidance refers to a long list of topics that FDA expects to be assessed to a high degree of certainty, beyond what many sponsors typically think of with respect to efficacy and safety, such as primary endpoint data and adverse event reports. As one example, the document notes that targeted collection of specific safety information may be required in some cases. As additional examples, it asserts that benefit-risk assessment will often require an in-depth understanding of the relevant patient population, natural history of the condition, effects of the drug that may be used in combination with existing therapies, and impacts of novel manufacturing technology or control strategy. The guidance also points to the need for robust information on dose exposure response and the adequacy of the primary efficacy endpoint.
The draft guidance describes a limited role for patient perspectives. FDA explains that the benefit-risk assessment is ultimately made at the population level, by the agency, with patient perspectives balanced against the agency’s regulatory judgment. The draft cites the example of a drug that is associated with significant risk and limited benefit, with no way to identify in advance (e.g., using predictive biomarkers) those individuals who might benefit. In such a case, FDA would not approve the drug, as it would lead to more harm in the indicated patient population overall, even if individual patients may derive benefit and express the desire for access. The draft emphasizes that FDA weighs patient perspective carefully and that the agency’s benefit-risk assessment is informed by evidence that a benefit is important to patients — but responsibility for making the regulatory judgment belongs solely to FDA.
FDA explains that interested parties should submit comments on the draft guidance by November 29, 2021, “to ensure that the Agency considers your comment ... before it begins work on the final version of the guidance.”5 FDA notes, however, that comments may be submitted on any guidance at any time, as provided by the Good Guidance Practice regulations.6
1Center for Biologics Evaluation and Research & Center for Drug Evaluation and Research, Guidance for Industry: Benefit-Risk Assessment for New Drug and Biological Products (Sept. 2021), https://www.fda.gov/media/152544/download; see also 86 Fed. Reg. 54,211 (Notice of Availability).
2This suggests that FDA remains committed to requesting dose exploration and in some instances requiring the sponsor to reduce dose as a condition of approval. See, e.g., Badrul A. Chowdhury, M.D., Ph.D., et al., The Risks and Benefits of Indacaterol — The FDA’s Review, N Engl J Med 2011; 365:2247-2249.
3See 21 U.S.C. § 355(d); 21 C.F.R. § 314.126.
4The draft guidance explains that FDA would consider factors such as the severity of the condition the drug is intended to address, the patient population, and “the current treatment landscape,” that is, the safety and efficacy of “other interventions used for the intended population,” including drug and nondrug options and off-label uses. FDA will also “have a lower tolerance for potential serious risks or toxicities ... when it evaluates preventative medicines, where the target population may be healthy people.”
586 Fed. Reg. at 54,211.
6See 21 C.F.R. § 10.115(g)(5).
Sidley Austin LLP provides this information as a service to clients and other friends for educational purposes only. It should not be construed or relied on as legal advice or to create a lawyer-client relationship. Readers should not act upon this information without seeking advice from professional advisers.
Attorney Advertising—Sidley Austin LLP, One South Dearborn, Chicago, IL 60603. +1 312 853 7000. Sidley and Sidley Austin refer to Sidley Austin LLP and affiliated partnerships, as explained at www.sidley.com/disclaimer.
© Sidley Austin LLP