I spent 6 months tracking my glucose response to different meal compositions. 47 breakfasts, continuous glucose monitor, spreadsheets my friends find genuinely alarming. You know what I never had a problem with? Recruiting enough test subjects. My n was 1, and that was the whole point.
Now imagine your n is the entire global population of patients with your disease, and it's 400 people. Scattered across dozens of countries. Many of them children. Some of them dying before your enrollment window closes. Someone at the FDA says: run a randomized controlled trial. Match your cohorts. Maintain your placebo arm. And I'm supposed to believe that's scientific rigor?
The Plausible Mechanism Framework, opened for public comment this year, is the most important regulatory correction I've seen in the health optimization space. It lets drugs for diseases affecting fewer than 1,000 U.S. patients reach approval based on mechanism of action, biomarker data, target engagement, and natural history controls. No impossible RCT required. This aligns regulation with biological reality for the first time.
The RCT Was Never the Right Tool Here
I love data. I am annoying about data. But I also know that the right measurement framework depends on the system you're measuring. My HRV protocol works because I can run it every morning for years and accumulate signal. A family with a child who has arginase 1 deficiency doesn't have years. There are maybe a few hundred patients in the entire country.
Loargys got approved this month for hyperargininemia, the first therapy targeting elevated plasma arginine in ARG1-D. It went through accelerated approval based on plasma arginine reduction. Confirmatory trials are required. That's the model working correctly: approve on the best available mechanistic evidence, then verify. Voxzogo followed the same path for achondroplasia, approved on annualized growth velocity with a confirmatory obligation attached.
These aren't rubber stamps. They're intelligent tradeoffs. The ROI calculation is straightforward: the cost of withholding a plausibly effective therapy from a child with a fatal condition vastly exceeds the cost of approving it with robust post-market monitoring requirements. Polaryx's regulatory affairs director, Minsu Kang, said developers now have "a much better idea as to what exactly the FDA is looking for" for ultra-rare diseases. Clarity is a compound gain. Every sponsor who understands the pathway is a sponsor more likely to invest in a rare disease program.
Where the Risk Actually Lives
I'll grant the strongest objection here: biomarkers fail. They do. The cardiac arrhythmia cases from the early 1990s are real and terrible. But the failure mode in that era was treating a biomarker as a validated endpoint without a confirmatory obligation. The Plausible Mechanism Framework explicitly requires confirmatory data. That's a different system.
The real risk isn't the framework's design. It's the FDA's capacity to execute it. Paul Melmeyer of the Muscular Dystrophy Association warned about "removal of staff with the regulatory expertise" needed for consistent implementation. He's right, and that should terrify everyone. A sophisticated approval pathway staffed by a gutted agency is a sports car with no brakes. The framework needs resources to function. Congress needs to fund the people who run the post-market surveillance, not just celebrate the policy announcement.
But the answer to "we might not monitor well enough" is not "so let's block the therapy entirely." That's like saying you shouldn't start a training protocol because you might not track your recovery perfectly. You optimize the tracking. You don't skip the protocol.
The MPS II advocacy community staged a funeral at FDA headquarters after 3 Hunter syndrome therapies hit rejections or delays in 8 months. Those families aren't asking for lower standards. They're asking for standards that match the biology of their children's diseases. The Plausible Mechanism Framework is exactly that: a regulatory toolkit calibrated to the actual constraints of ultra-rare conditions.
I've run enough failed self-experiments to know that not every promising protocol delivers. Some of these drugs will disappoint. But the framework gives them a chance to be tested in the patients who need them, with confirmatory obligations attached. That is how you optimize a system where the cost of inaction is measured in children's lives, not in spreadsheet rows.
