The Regulatory Landscape Is Finally Ready
Frameworks for small-population trials, patient-generated evidence, and AI-driven pattern recognition now exist. The infrastructure to feed them does not.
For decades, the regulatory path to drug approval assumed large populations. Randomized controlled trials with thousands of patients. Phase 1, 2, and 3, each requiring more data, more patients, more time. For diseases affecting a few hundred or a few thousand people, that pipeline was never going to arrive.
That is changing, not because regulators suddenly care more, but because the science and the policy have matured in parallel.
What exists now
- Plausible Mechanism Framework (Feb 2026): If you can demonstrate the disease-causing molecular abnormality and show your therapy targets the root cause, combined with well-characterized natural history data, you can pursue approval with a single study. One hundred mutations of the same gene no longer need one hundred trials.
- Rare Disease Evidence Pathway (RDEP): For ultra-rare conditions affecting fewer than 1,000 US patients with known genetic defects and no existing therapies, the FDA now accepts a single pivotal trial plus confirmatory evidence.
- Innovative Trial Designs for Small Populations: Single-arm trials where patients serve as their own controls. Bayesian designs. Adaptive enrichment. Disease progression modeling. The toolkit for small populations has expanded dramatically.
- Real-World Evidence: The 21st Century Cures Act formalized patient-generated real-world data as a valid basis for regulatory submissions. The ISTAND pilot created pathways for novel biomarkers.
What is still missing
Every one of these frameworks relies on something that barely exists: structured, longitudinal patient data at clinical granularity. The Plausible Mechanism Framework requires "well-characterized natural history data" but provides no standard for how to collect, version, or share it across programs. Guidance documents "encourage" data sharing but specify no privacy-preserving infrastructure.
The regulatory door is open. The question is whether the patient-side infrastructure exists to walk through it. That is the gap Cureledger is built to close. We did not wait for regulators to ask; we have been building toward this moment for over a decade.
Key context
- The FDA default is now one trial, not two for all drugs (Feb 2026)
- Master protocols allow one application to cover multiple mutations of the same gene
- Natural history data can serve as the missing comparison group
- The federal advisory committee for newborn screening (ACHDNC) was terminated in April 2025, leaving a gap in oversight that patient-driven data infrastructure can help fill