April 27, 2026
Re: [Docket No. FDA-2026-D-1256] “Considerations for the use of the Plausible Mechanism Framework to Develop Individualized Therapies that Target Specific Genetic Conditions with Known Biological Cause” draft guidance.
The National Center for Health Research (NCHR) appreciates the opportunity to comment on this draft guidance. NCHR is a nonprofit, nonpartisan think tank that evaluates the safety and effectiveness of medical products and health policies to ensure they are supported by strong science and benefit patients and public health.We understand the importance of addressing the urgent need for therapies for patients with rare, severe, or life-threatening genetic conditions. Progress has been made in recent years, but more is needed. However, the flexibility proposed in this framework raises important concerns regarding evidentiary standards, safety, and long-term clinical benefit that are well documented in medical literature.
1. Strengthen Evidence Requirements for Small, Nonrandomized Studies
The proposed guidance allows approval based on small, nonrandomized studies, including the possibility that a first-in-human trial may serve as the primary evidence for approval. While we understand the challenges of well-designed studies of treatments for very rare diseases, not all rare diseases are so rare that would justify the important limitations of such poorly designed research. Without a comparison group, it is often impossible to determine whether observed outcomes are due to the treatment or to variance in the disease’s natural course. In small studies, differences between patients, such as disease severity, timing of treatment, or overall health, can have a greater influence on results than the treatment itself. The guidance does not clearly define what constitutes a sufficiently small or rare patient population to justify the proposed evidentiary flexibilities. Without clear criteria, there is a risk that therapies for conditions that are not extremely rare could inappropriately rely on less rigorous study designs. Analyses of FDA approvals further show increasing reliance on smaller and less rigorous trials (Downing et al., 2014), along with variability in the strength of supporting evidence (Hwang et al., 2019). Together, these factors inevitably lead to greater uncertainty about how well these therapies work and how safe they are.
Recommendations
- Require use of external controls or natural history data when relevant.
- Clearly define the criteria or thresholds for what constitutes a sufficiently small or rare patient population to justify evidentiary flexibility. This should be based on international patient populations when appropriate.
- Clearly define how adequate sample size should be determined for approval, including the factors used when statistical power is not feasible (e.g., disease rarity, expected effect size, variability in outcomes, and strength of supporting evidence).
- Pre-specify endpoints, analysis plans, and bias-mitigation strategies.
- Appropriately use within-patient comparisons (baseline vs post-treatment).
2. Improve Post-Market Evidence Generation
The framework relies heavily on post-marketing studies to confirm benefit and further characterize safety, because individualized therapies will be approved based on limited evidence from very small patient populations. However, generating robust post-market evidence can be inherently challenging after a treatment has FDA approval. The same constraints that limit pre-market studies, such as small, highly specific patient populations and lack of feasible comparators, usually will persist after approval, with the added difficulty of finding a comparison/control sample that is not using the treatment. Is it feasible for researchers to generate sufficiently powered real-world evidence, and if so, explicit plants to do so should be part of the application to the FDA and its response to the application. In addition, variability in clinical presentation and fragmented data collection may further limit the ability to interpret outcomes consistently across patients. An important goal of FDA approval based on such limited evidence should be a plan to ensure persuasive confirmatory evidence that is timely, complete, and scientifically solid. The FDA must develop and enforce incentives for such confirmatory evidence, including meaningful penalties when those criteria are not met.
Recommendations
- When approval is based on limited or early evidence, a follow-up study to confirm real patient benefit should already be underway at the time of approval.
- Establish mandatory patient registries for all treated patients with standardized demographic data (age, race, ethnicity).
- Set clear timelines for post-market studies and enforce them in a timely manner through actions such as restricted use or withdrawal of approval if requirements are not met. Labeling that specifies the uncertainty of evidence should be required upon approval but is not an adequate safeguard in the absence of confirmatory evidence.
- Promote data sharing across programs to pool evidence in small populations.
3. Limit Extrapolation Across Genetic Variants Without Direct Evidence
The guidance allows expansion of therapies to additional genetic variants based on shared mechanisms. This has the potential to create an enormous loophole that will result in prescriptions to exponentially more patients than are likely to benefit. This would be unfair to patients and physicians, who would be unable to make informed medical decisions, and create confusion and additional uncertainty for the types of patients who were initially studied and therefore more likely to benefit. Biological and clinical responses may vary across mutations, even within the same gene, which can affect both effectiveness and safety. Evidence supporting approvals has shown variability across studied populations and conditions, highlighting limits to generalizability (Downing et al., 2014). As a result, extrapolating across variants without strong, direct supporting data would extend FDA approval and patient use beyond what the available evidence can reliably support, increasing uncertainty about how these therapies will perform in various patient subgroups.
Recommendations
- Require variant-specific clinical or mechanistic evidence before extending approval to additional genetic variants, and permit extrapolation only when there is strong scientific justification supported by empirical data.
- Limit extrapolation to variants with well-characterized and similar biological effects, supported by evidence demonstrating consistent treatment response across variants.
- Use a stepwise, evidence-based expansion approach, beginning with closely related variants and expanding only as additional supporting data become available.
4. Limit Reliance on Surrogate Endpoints
The framework emphasizes mechanistic rationale, such as evidence that a therapy targets a specific biological pathway or genetic mutation thought to cause disease, and biomarker-based evidence, which may show biological effects but do not necessarily translate into meaningful clinical outcomes for patients. As CBER Director Vinay Prasad has pointed out, research shows that many surrogate endpoints have weak correlations with outcomes such as survival, and improvements in these markers do not consistently translate into real clinical benefit (Prasad et al., 2015). Unfortunately, even longer-term follow-up indicates they often do not predict improved quality of life or overall survival (Rupp & Zuckerman, 2017). Analyses of FDA approval decisions also show increasing reliance on endpoints that may not directly measure how patients feel, function, or survive (Downing et al., 2014). The evidence is clear: FDA’s track record of relying on biomarkers or other surrogate endpoints without convincing validation has led to approval of therapies where the actual benefits for patients remain uncertain or ultimately are found to be outweighed by serious risks in real-world use.
Recommendations
- Prioritize clinical outcomes (how patients feel, function, survive) as critical metrics during the study rather than post-market metric assessment.
- Allow surrogate endpoints only when there is clear evidence linking them to clinical benefit in the short-term and long-term.
- Require post-approval confirmation of clinical outcomes when surrogates are used.
5. Strengthen Pre-Market Safety Evaluation
The guidance applies to individualized therapies developed for patient populations where sample sizes may be fewer than a dozen in some cases, rather than hundreds, and studies may be relatively short-term. Under those circumstances, pre-market studies are unlikely to identify rare, long-term, or delayed adverse effects. This is particularly relevant for genome editing and RNA-based therapies, as well as other individualized approaches, where risks such as off-target effects, immune responses, and delayed toxicities may not be fully captured in small, short-duration studies (Bainbridge, 2024). Consequently, there will be substantial uncertainty about the safety profile of these therapies at the time of approval.
Recommendations
- Require targeted safety assessments (e.g., off-target effects, immunogenicity) based on therapy type.
- Ensure minimum follow-up duration appropriate to expected risks.
- Standardize core safety endpoints across similar therapies.
Conclusions
This framework is intended to advance therapies for rare, severe, or life-threatening genetic conditions. However, the inherent severe limitations of what the FDA suggests might be extremely small patient populations and reduced evidentiary thresholds introduce challenges to providing any meaningful evidence of safety and effectiveness. As written, it is unclear whether the Framework can ever ensure that these therapies provide meaningful clinical benefit while maintaining patient safety.
When therapies are approved based on limited or uncertain evidence of meaningful clinical benefit, patients and clinicians may face difficult decisions without clear information about likely benefits or risks. In such cases, patients may be exposed not only to potential medical harm but also to substantial personal and financial burdens, including out-of-pocket costs, for treatments whose effectiveness has not been adequately established. The concept of “financial toxicity” describes the harmful patient-level financial burden associated with treatment, including copayments and other out-of-pocket costs that can affect treatment decisions and well-being (Carrera et al., 2018; Tran & Zafar, 2018; Zafar, 2016). Even insured patients often experience stressful financial distress, sometimes leading them to alter or forgo recommended treatments due to cost (Carrera et al., 2018; Tran & Zafar, 2018). The uncertainty of choosing an unproven treatment might be acceptable as a free or at-cost humanitarian treatment, but it is unfair for patients and families to pay full price for unproven treatments that the FDA approves, given the that new products for rare diseases often cost millions of dollars per patient. In addition to the harmful stress and financial burden on individual patients and their families, that could be extremely harmful to the affordability of health insurance and the long-term solvency of Medicare and Medicaid. Traditionally, the FDA does not consider the cost of treatments during approval decisions, but when the FDA approves treatments based on flexibility rather than scientific evidence, the agency should balance risks and benefits that include financial toxicity. In addition, the FDA’s flexibility standards should be balanaced with clear, enforceable standards for evidence generation, safety evaluation, and post-market accountability. Strengthening these safeguards will be critical to maintaining scientific rigor, public trust, and the long-term success of individualized therapeutic development.
Respectfully submitted,
National Center for Health Research,
Washington, D.C.
References:
- Downing, N. S., Aminawung, J. A., Shah, N. D., Krumholz, H. M., & Ross, J. S. (2014). Clinical trial evidence supporting FDA approval of novel therapeutic agents, 2005-2012. Jama, 311(4).
- Hwang et al. 2019 Association between FDA approval and evidence strength. Association between FDA and EMA expedited approval programs and therapeutic value of new medicines. JAMA Intern Med. 2019;179(5):582-590. doi:10.1001/jamainternmed.2018.8461
- Prasad, V., Kim, C., Burotto, M., & Vandross, A. (2015). The strength of association between surrogate end points and survival in oncology: a systematic review of trial-level meta-analyses. JAMA internal medicine, 175(8), 1389-1398.
- Bainbridge, J. W. (2024). Success in sight for gene editing. New England Journal of Medicine, 390(21), 2025-2027.
- Rupp, T., & Zuckerman, D. (2017). Quality of life, overall survival, and costs of cancer drugs approved based on surrogate endpoints. JAMA Internal Medicine, 177(2), 276-277.
- Carrera, P. M., Kantarjian, H. M., & Blinder, V. S. (2018). The financial burden and distress of patients with cancer: understanding and stepping‐up action on the financial toxicity of cancer treatment. CA: a cancer journal for clinicians, 68(2), 153-165.
- Yousuf Zafar, S. (2016). Financial toxicity of cancer care: it’s time to intervene. Journal of the National Cancer Institute, 108(5), djv370.
- Tran, G., & Zafar, S. Y. (2018). Financial toxicity and implications for cancer care in the era of molecular and immune therapies. Annals of translational medicine, 6(9), 166.