When your device has no ancestor on the market
The FDA De Novo pathway is a request for the agency to classify a novel device into Class I or Class II when it is genuinely new — that is, when no legally marketed predicate device exists to support a substantial equivalence claim — and its risk profile is low to moderate. Without it, a device with no predicate would automatically fall into Class III and face the most burdensome premarket approval route, regardless of how modest its actual risk is. De Novo exists precisely so that novel-but-low-risk technology is not blocked by the absence of history.
A note on scope: Projects House is a product engineering firm, not a regulatory consultancy or a law firm. This article is educational background to help you plan an engineering and testing program. Regulatory strategy and submissions should be developed with qualified regulatory professionals.
How it differs from the familiar route
The conventional clearance route asks you to point at an existing cleared device and demonstrate that yours is substantially equivalent in intended use and technological characteristics. With no predicate, there is nothing to anchor that argument to. De Novo inverts the logic: instead of arguing similarity, you ask the agency to establish a classification for your device type and to define the special controls — standards, performance testing, labeling requirements, and sometimes postmarket requirements — that will assure its safety and effectiveness going forward.
We will not re-explain equivalence here; it is covered in the FDA approval process for medical devices, and the search itself in how to choose a predicate device. The classification tiers and what each implies are in FDA medical device classes.
When this pathway is the right fit
- A thorough predicate search came back empty. "We didn't find one" is not enough. You need a documented, systematic search of the classification database and cleared-device records, with a reasoned explanation of why each near-candidate fails as a predicate.
- The risk is not high. Life-supporting devices, life-sustaining devices, and most long-term implants will not fit — those are steered toward the premarket approval route.
- Special controls are definable. If you can point to recognized consensus standards, bench and animal testing, labeling, and performance requirements that credibly address the identified risks, you have the raw material for a proposal.
- A new intended use breaks equivalence. Sometimes the hardware is familiar but the new indication creates a category that did not exist before. That is a classic De Novo situation.
The pathway has become especially common for clinical software and algorithm-driven products, where the output is a new kind of clinical insight rather than a new physical mechanism — the territory covered in software as a medical device.
What the submission has to contain
A De Novo request is heavier than a conventional submission, because there is no prior device whose data you can lean on. Expect to assemble a complete description of the device and its technology, a detailed risk analysis identifying each hazard and demonstrating how it is controlled, bench performance data and often animal or clinical data, a reasoned proposal for the special controls and the classification you are requesting, and draft labeling and instructions for use.
The risk analysis is doing more work here than in any other submission type, because the special controls you propose have to be traceable to identified hazards. That makes a rigorous risk file the spine of the whole package — see ISO 14971 risk management. Your design and testing records need to hold up as evidence too, which is why a compliant design control process should have been running properly from the start of development rather than reconstructed at submission time. If human data is required, the study design considerations are in medical device clinical trials.
Two ways in — and one is clearly better
You can submit a direct De Novo request, or you can arrive at the pathway after a conventional submission is returned with a finding that no suitable predicate exists. The direct route is almost always preferable: it saves an entire submission cycle, avoids creating a record of a negative determination, and lets you build the file from the beginning around the argument that actually matters — why the risk is low and how it is controlled — rather than around an equivalence argument that was doomed from the outset.
The practical implication is that the pathway decision belongs in your regulatory strategy phase, before you commit to a test plan, not after the device is finished. Choosing the wrong route sends you back to redo testing that was scoped for a different evidentiary standard.
Time, cost, and effort
Set expectations accordingly: a De Novo program is substantially longer and more expensive than a conventional clearance — more documentation, more rounds of agency questions, and usually more data. For context on the scale of the wider budget, see FDA approval cost for a medical device.
One step is nearly always worth taking: use the agency's presubmission mechanism to get feedback before you file. It is comparatively inexpensive and lets you validate the pathway choice, the scope of the data package, and the proposed special controls before you spend real money on testing. The most expensive mistakes in this area are filing a De Novo when another route fit, and burning months trying to force an equivalence argument that does not exist.
The strategic upside most people miss
A device classified through this pathway creates a new category — and from that moment it can serve as the predicate for everyone who follows. Yes, that means later competitors get an easier route. But it also means you wrote the rules: the special controls defined for the category were derived from your device, your testing, and your labeling, and competitors must meet requirements shaped around your design. Combined with the lead time you accumulate, that is a durable commercial advantage. More on planning the regulatory and engineering work together in our medical device development hub.
If you are developing something with no obvious predicate and need the engineering, verification, and risk documentation built to support a submission, contact Projects House through our form and we will map the development program with you.