Stakeholders responding to a U.S. Food and Drug Administration docket urged the agency to adopt flexible statistical standards for rare disease trials but warned that its September workshop overemphasized randomized controlled trials, RAPS reported.
The comments, submitted after Rare Disease Innovation, Science, and Exploration (RISE) workshop in September, reveal a widening gap between regulatory tradition and the practical limits of studying ultra-small patient populations. Nearly all feedback addressed drug and biologic development, where sponsors argue that conventional significance thresholds and concurrent control arms are often ethically or logistically impossible.
“In rare disease settings with small or heterogeneous populations, reliance on a single conventional significance threshold may not fully characterize treatment benefit and, in some circumstances, may make adequately powered trials infeasible,” the Biotechnology Innovation Organization (BIO) told FDA.
Rigid Trial Rules Delay Cures for Rare Diseases
- Rigid trial rules delay cures: Insisting on large, randomized placebo-controlled studies for diseases affecting few people often makes research impossible, leaving patients without options.
Success Criteria: Moving Beyond the P-Value
BIO advocated for a “totality-of-evidence” approach, asking the agency to provide case studies demonstrating fit-for-purpose statistical applications across diverse disease contexts.
Otsuka Pharmaceutical Development & Commercialization called on FDA to weigh disease severity, unmet medical need, and feasibility constraints when evaluating alternative evidentiary standards, noting that not all factors warrant equal consideration in determining acceptable levels of statistical uncertainty. UCB requested a framework for prospectively discussing alternative success criteria with sponsors, arguing that pre-specification reduces regulatory surprise.
Randomized Designs: Feasibility vs. Gold Standard
While commenters acknowledged randomization minimizes bias, they stressed it is frequently unworkable. Otsuka supported unequal randomization, adaptive designs, crossover or randomized withdrawal designs, early stopping rules, and combined Phase 2/3 studies.
“BIO encourages FDA to apply a fit-for-purpose, risk-based approach that supports innovative alternatives to randomized controls when appropriate, particularly where randomization is burdensome, ethically challenging, or infeasible in small populations, and to promote consistent, predictable evidentiary expectations for these approaches across FDA Centers,” BIO told FDA.
UCB emphasized that patient and caregiver engagement must occur early to assess the acceptability of placebo or sham arms, particularly where delaying treatment carries irreversible consequences.
“Patient and caregiver engagement should occur early in trial design to understand the acceptability of randomization, placebo, sham, or other control-arm exposure,” UCB wrote in its docket submission.
PhRMA argued the agency should support randomization only in scenarios where there is an increased likelihood a participant receives effective treatment, rather than designs that make an untreated control arm more tolerable.
The National Organization for Rare Disorders (NORD) raised concerns that the agency and outside speakers during the workshop expressed a “strong preference” for randomized controlled trials, which they say contradicts FDA’s recent white paper on a fit-for-purpose framework. Lilly echoed this, stating workshop discussions “repeatedly returned to reasons a randomized concurrent control should be included rather than exploring how other innovative approaches could be made reliable.”
Endpoint Selection: Capturing Heterogeneous Benefit
BIO asked FDA to promote endpoints centered on patient experience and disease-contextualized principles supported by natural history data and biomarkers. Otsuka encouraged consideration of hierarchical composite endpoints, which may better capture clinically meaningful benefit across heterogeneous disease manifestations while maximizing information from small cohorts.
UCB warned that requiring all participants to demonstrate benefit on the same manifestation may fail to capture meaningful treatment effect. PhRMA flagged inconsistency across FDA Centers, noting the agency sometimes rejects a proposed endpoint “without offering clear rationale or providing meaningful direction on a route forward, short of conducting a pivotal trial at risk.”
Commenters Endorse Bayesian Borrowing and Adaptive Designs for Rare Disease Trials
Commenters broadly endorsed advanced analytical methods to squeeze maximum evidence from minimal patients. BIO listed Bayesian borrowing, external and hybrid controls, adaptive and sequential designs, covariate adjustment, longitudinal modeling, and master protocols as promising strategies. UCB agreed these designs enable more information from limited populations. PhRMA suggested the agency draw on lessons from regulatory science pilots, improve sponsor communication, and increase staff training to evaluate these novel approaches consistently.
| Stakeholder | Primary Recommendation | Key Concern |
|---|---|---|
| BIO | Totality-of-evidence framework; case studies for fit-for-purpose stats | Single significance thresholds make powered trials infeasible |
| PhRMA | Fit-for-purpose randomization; consistent Center-level endpoint feedback | Workshop over-indexed on RCT superiority |
| Otsuka | Weigh severity/unmet need; unequal randomization & seamless designs | Need clarity on evidentiary expectations for external controls |
| UCB | Prospective framework for success criteria; early patient engagement | Single endpoints miss heterogeneous benefit |
| NORD / Lilly | Start with evidence needed, not RCT feasibility; nonrandomized designs not inherently inferior | Workshop preference for RCTs contradicts FDA white paper |
FDA Urged to Adopt EMA and CDRH Flexible Approval Models
The tension mirrors debates at the European Medicines Agency (EMA), where the PRIME scheme and conditional marketing authorizations already allow approval on less comprehensive data for severe unmet needs. FDA’s Center for Devices and Radiological Health (CDRH) employs a risk-based framework for device trials where randomized controls are infeasible; Lilly and others urged the drug centers to adopt a similar model.
Founded in 1993 from a merger of two earlier groups, the Biotechnology Innovation Organization (BIO) represents more than 1,100 firms, research schools and related associations across the United States and over 30 other countries. It is the world’s largest advocacy association for the biotechnology industry and spent $9.87 million lobbying the U.S. government in 2018, advocating for a more efficient and predictable regulatory process for new food and drug products among other issues.
Contraindications & When to Consult a Specialist
The docket remains open for public review. FDA has not announced a timeline for revised draft guidance, but the volume of commentary indicates that the next iteration must reconcile the agency’s statistical conservatism with the clinical reality of rare disease drug development.
References
- FDA. Rare Disease Innovation, Science, and Exploration (RISE) Workshop Series
- Regulations.gov. BIO Comment on Statistical Considerations for Rare Disease Trials
- Regulations.gov. PhRMA Comment on Rare Disease Trial Design
- Regulations.gov. Otsuka Comment on Alternative Evidentiary Standards
- RAPS. Docket Commenters Call for Flexibility in Rare Disease Trial Design