FDA sheds light on benefit-risk assessments to inform decision-making throughout product lifecycle

Patient experience data tend to play a central role in FDA decisions especially when Patient-Reported Outcomes (PROs) or other types of Clinical Outcome Assessments (COAs) are used as primary endpoints, and patients can provide meaningful input when conditions are not well-characterized, as with some rare diseases. The guidance provides key insight around data collection, analyses, and presentation in a marketing application, as well as timepoints where discussions with FDA during drug development can be most impactful. While the draft guidance is an important step forward, we expect that industry stakeholders to request more specific details around collection methods and fit-for-purpose data collection tools to inform strategic planning for drug development program and lifecycle management. FDA is accepting comments through November 29.

Benefit-risk assessment guidance

FDA uses the benefit-risk assessment to make an informed decision as to whether the benefits (with their uncertainties) of the drug outweigh the risks (with their uncertainties and approaches to managing risks) under the conditions of use described in the product labeling. The benefit-risk assessment becomes challenging in cases where there is greater uncertainty surrounding benefit or known or potential serious safety risks are identified, e.g., risks that are life-threatening or associated with significant morbidity.

The draft guidance offers clarification on several aspects of the benefit-risk assessment, including:

• The therapeutic context in which the drug will be used;

• The evidence submitted in the premarket application and/or generated in the postmarket setting that informs FDA’s understanding of the benefits and risks of the drug;

• Remaining uncertainties about the drug’s benefits and risks; and,

• Regulatory options to reduce uncertainties and manage risks.

Benefit-risk planning in the premarket setting

The draft guidance provides an extensive table listing examples of considerations that may be included in the benefit-risk assessment for a given drug, noting that the relevance and relative importance of any consideration depends on the specific details of each application. Notable examples from this table (“Table 1”) include:

• Aspects of the indicated condition that are most relevant to, or have the greatest impact on, the intended population (e.g., incidence, duration, morbidity, mortality, health related quality of life, important differences in outcome or severity in subpopulations)
• Current approved treatments and standard of care, including their efficacy, safety, tolerability, and other limitations (e.g., subpopulations who do not respond to or do not tolerate treatment, curative versus palliative intent)
• Efficacy and safety of other interventions used for the intended population, such as drugs used off-label or other non-drug interventions
• Clinical relevance of the study endpoints, and the ability to measure or predict clinical outcomes of importance to patients
• Potential impact of product quality issue(s) that could negatively impact the drug’s safety or effectiveness
• Whether certain labeling (e.g., boxed warnings) and/or REMS is necessary to support favorable benefit-risk assessment
• Whether a postmarketing study or clinical trial is necessary to assess a known serious risk or a signal of a serious risk

FDA acknowledges in the draft guidance that its benefit-risk assessment must take uncertainties into account, offering the following key examples of limitations that may affect benefit-risk assessments:

• Aspects of the program or study design, such as the population, choice of controls, endpoints, duration, and data sources, as well as any differences between the clinical study and real-world use.
• Reliability of the estimates of benefit or risk based upon variability in estimated effects due to sampling (statistical uncertainty) or issues with trial conduct such as missing data, poor protocol compliance, etc.
• Proposed risk management strategies, such as patient monitoring, which have not been studied in clinical trials, or that have been studied in clinical trials but would be potentially difficult to implement in practice.
• Introduction of a novel technology or control strategy in the drug’s manufacturing process, or other potential issues regarding the product formulation or manufacturing.

The draft guidance notes that a higher degree of uncertainty is common in drug development programs for rare diseases, where the prevalence of disease, and consequent limitations of study size, can limit the precision of safety and efficacy characterizations. Thus, as noted in the guidance, FDA “often exercises greater regulatory flexibility in these cases." Separately, HHS and FDA have also offered drug sponsors greater flexibility on human subjects protection regulations during the COVID-19 pandemic, which we have summarized online here.

Activities in premarket development to inform the benefit-risk assessment

That draft guidance states that in its benefit-risk assessment, FDA provides significant deference to the decisions and activities undertaken by sponsors in the development of their drugs, and the evidence generated to support their marketing applications. Accordingly, the agency urges sponsors to conduct “structured benefit-risk planning”: a purposeful activity carried out by the sponsor to incorporate consideration of the product’s benefit-risk assessment throughout the drug development lifecycle. In addition to supporting premarketing development and evaluation, planning for postmarket benefit-risk assessment during the premarket stage can inform approaches to collecting additional information in the postmarket setting to further reduce uncertainties, as discussed in greater detail below.

FDA provides the following examples to emphasize the importance of structured benefit-risk planning:

• Identification of patients (e.g., utilizing a predictive biomarker) who are more likely to experience greater expected benefit or less likely to experience serious adverse events of the drug may enable determination of a population for whom the drug may have a more favorable benefit-risk profile.
• Use of an active control arm may be beneficial in circumstances when it may be critical to ensure that the drug does not have an unacceptable benefit-risk profile compared to an approved, alternative therapy, or to show that the drug is more effective than available therapy.
• Consider enriching a trial to enable the demonstration of benefit in a specific subpopulation.
• Trial sponsors may benefit from targeted case report form prompts and/or independent adjudication to actively ascertain the occurrence and nature of an adverse event of interest.

Further, FDA says sponsors may want to conduct additional benefit-risk analyses to help inform the overall assessment in some circumstances, such as decisions involving complex tradeoffs between the drug’s expected benefit and risks, or significant or novel uncertainties regarding the drug’s benefits and risks.

Interactions between FDA and sponsors

FDA can provide insight and regulatory perspective that can inform a sponsor’s benefit-risk planning appropriate to the issues identified at a particular stage of development. Specifically, the guidance advises that FDA can provide insight in discussions with sponsors at the End of Phase 2 (EOP2) to help identify a potential safety issue that would require greater certainty about the drug’s benefits and/or risks to support approval. FDA says these discussions can help enhance the characterization of the drug’s benefits and risks, including decisions on study design, selection of appropriate patient populations, enrichment strategies, clinically meaningful endpoints, trial duration, dose-response assessment, and trial sizes.

In addition to EOP2 discussion with FDA, communications with the agency could involve deliberations regarding the clinical meaningfulness of a purported benefit or concern for non-clinical safety signals at the pre-IND phase for first-in-human studies. They could also involve considerations on the best design to characterize benefits and risks where the population is limited or vulnerable, such as for rare or serious diseases or pediatric populations, the draft guidance notes. Sponsors may want to consider adding “benefit-risk considerations” as a proposed question and/or agenda item and provide relevant supplementary information in the FDA meeting package.

Patient experience data

FDA says that “patients are experts in the experience of their disease or condition, and they are the ultimate stakeholders in the outcomes of medical treatment.” Therefore, the draft guidance explains, patient experience data can inform nearly every aspect of FDA’s benefit-risk assessment throughout the drug lifecycle. To that end, FDA encourages sponsors who are considering collecting and utilizing patient experience data as part of their evaluation of effectiveness or safety to have early interactions with FDA during the design phase of such studies and obtain feedback from the relevant FDA review division on appropriate research design and any applicable regulatory requirements.

Patient preference information (PPI) is called out by the draft guidance as one important type of patient experience data, and is defined by FDA as “assessments of the relative desirability or acceptability to patients of specified alternatives or choices among outcomes or other attributes that differ among alternative health interventions.” The guidance advises that FDA is most likely to utilize PPI to inform regulatory decision-making when:

1. significant risks of treatment or uncertainty about risks exist relative to the expected benefits;
2. patients’ views about the most important benefits and risks vary considerably within a population; and/or
3. when patients’ views as to the most important benefits are expected to differ from those of health care professionals.

Presenting benefit-risk information

The draft guidance cautions that effective communication by sponsors of the drug’s benefits, risks, and uncertainties is important to informing FDA on the benefit-risk assessment that supports regulatory decision-making, particularly when serious risks are involved. As part of an NDA submission, for example, the sponsor must provide “[a]n integrated summary of the benefits and risks of the drug, including a discussion of why the benefits exceed the risks under the conditions stated in labeling,” according to 21 CFR 314.50(c)(5)(viii).

In addition, FDA advises that description of the clinical importance of key benefits and risks may facilitate the agency’s benefit-risk assessment, including:

• Discussion of the magnitudes of effects and treatment effects (difference between drug and comparator).
• Exploration of the nature of effects (e.g., consideration of time course and durability of the drug’s effect, the clinical importance of benefit of a particular magnitude, and patient input on importance).
• Estimates of the statistical uncertainty around the magnitudes of the most important benefits and potential risks (e.g., with confidence intervals).
• Presentation of a graphical or tabular summary of results for the most important benefits side by side or juxtaposed with important potential risks.
• Discussion of additional sources of uncertainty about benefits and/or risks, untested risk management strategies, or potential differences between aspects of the clinical trial and expected real-world use (e.g., population, adherence, safety monitoring).

These recommendations are in line with FDA’s October 2018 guidance “Presenting Quantitative Efficacy and Risk Information in Direct-to-Consumer Promotional Labeling and Advertisements,” which we analyzed that guidance online here.

Benefit-risk planning in the postmarket setting

FDA also considers a drug’s benefits and risks and uncertainties in the postmarket setting, where it uses a diverse range of postmarket evidence to inform its risk-benefit assessments, including medical literature, postmarketing studies, adverse event reports, medication error reports, product quality reports, and in some cases, new data obtained from drugs of the same class. Examples of regulatory decisions that may be informed by such assessments include addition, modification, or removal of a REMS, initiation or release of postmarketing study requirements, labeling changes (e.g., addition, revision, or removal of a boxed warning), and marketing withdrawal.

Authored by Lowell Zeta and Lynn Mehler