These are the three main routes to market for a medical device in the United States. A 510(k) shows a device is substantially equivalent to one already on the market, a PMA is full approval for high-risk Class III devices, and De Novo classifies a novel lower-risk device that has no predicate.
Free sandbox · No credit card · 21 CFR Part 11 aligned
510(k), PMA and De Novo
noun · clinical research
Three FDA premarket routes for medical devices: a 510(k) notification demonstrating substantial equivalence to a predicate, a premarket approval (PMA) application for Class III devices, and De Novo classification for novel low-to-moderate risk devices.
Related terms
In short
Explained
US devices are sorted into three classes by risk: Class I (lowest, general controls), Class II (moderate, general plus special controls) and Class III (highest, typically life-supporting, life-sustaining or presenting a potential unreasonable risk). The class, and whether a comparable device already exists, steer which premarket route applies.
The words matter. FDA clears 510(k)s, approves PMAs and grants De Novo requests, and regulators and reviewers notice when manufacturers use them loosely. Describing a 510(k) device as FDA "approved" is a common error.
A 510(k) is a premarket notification. The manufacturer identifies a legally marketed predicate device and argues that the new device has the same intended use and either the same technological characteristics or different ones that do not raise new questions of safety and effectiveness. Evidence usually includes bench and performance testing; some 510(k)s also need clinical data, for example when a technology differs or a new indication is claimed. Check the current FDA guidance for the product code you are using.
De Novo (section 513(f)(2) of the Federal Food, Drug, and Cosmetic Act) is for devices that would otherwise default to Class III only because no predicate exists, but whose risks can be managed with general or special controls. Since a 2012 amendment a sponsor may request De Novo directly, without first receiving a "not substantially equivalent" decision on a 510(k). If granted, FDA establishes the classification and the controls, and later similar devices can use the new device as their predicate.
A PMA application is the route for Class III devices that need FDA approval before marketing. It contains the full evidence of safety and effectiveness: non-clinical laboratory data, manufacturing information and, in most cases, clinical study results. A device studied to support a PMA is often tested in a significant-risk study under an investigational device exemption (IDE), commonly in staged form such as an early feasibility study followed by a pivotal study. See early feasibility and pivotal device study EDC.
| 510(k) | De Novo | PMA | |
|---|---|---|---|
| Typical device class | II (some I) | I or II (new) | III |
| Predicate required | |||
| FDA outcome | Cleared | Granted | Approved |
| Clinical data | Sometimes | Often | Usually |
| Evidence standard | Substantial equivalence | Reasonable assurance (controls) | Valid scientific evidence |
Choosing
| Situation | Likely route | What drives the evidence |
|---|---|---|
| Similar to a marketed device, same intended use | 510(k) | Comparison to the predicate; bench testing; sometimes clinical data |
| New type of device, low-to-moderate risk, no predicate | De Novo | Risk analysis, special controls, usually clinical performance evidence |
| High-risk device or one that supports or sustains life | PMA | Pivotal clinical study plus manufacturing and non-clinical data |
| Class III device that is a modification of an approved PMA | PMA supplement | Scope of the change; may or may not need new clinical data |
| Software or AI function with a new intended use | Depends on risk | Intended use and risk class; check current FDA software guidance |
This is a simplified guide. Regulatory strategy for a specific device should be confirmed with FDA guidance and a regulatory professional.
Build your device study in the free Capture sandbox: eCRFs with edit checks, e-signatures and an audit trail, ready to test before enrolment.
Clinical evidence
Whichever route you choose, clinical data that goes into a submission has to be reliable and traceable. Reviewers and inspectors look for a protocol that was followed, consent that was obtained before procedures, an audit trail showing who changed what and why, and device-specific records such as device identifiers, deployment details, device deficiencies and serious adverse events. Good clinical practice applies; see Good Clinical Practice (GCP) and, for ethics review, the institutional review board.
The data system should therefore let you design device-specific forms (procedure details, implant or device logs, performance measures), capture adverse events with an assessment of device relationship, and export clean, documented datasets. For a pivotal study that supports a PMA or a De Novo request, consistent multi-site capture and monitoring are central. For the practicalities, see EDC for medical device clinical trials, EDC for diagnostic accuracy studies and EDC for AI medical device clinical validation.
After clearance, approval or grant, the work is not over. Post-market surveillance and follow-up studies continue to generate data, often in registries. Capture has a post-market page for that stage.
In Capture
Capture builds eCRFs without code, with skip logic, edit checks that raise auto-queries, a field-level audit trail and e-signatures with 21 CFR Part 11-aligned controls. Use the template library as a starting point for adverse events, vital signs and medical history, add your device-specific forms, and export CSV or Excel with a data dictionary, or SDTM datasets as SAS XPT with Define-XML. Run the whole study in the free sandbox before you go live.
Stage 1
DraftBuild and edit freely
Stage 2
ApprovedLocked for live use
Stage 3
In the casebookOnly approved forms
SDV requirement changed on Systolic BP
Data manager · Reason: critical safety value
Form approved: Vital Signs (VS) v2
Study owner · Locked for live use
Answer changed: Weight 68.0 to 68.5 kg
Site coordinator · Reason: transcription error
A 510(k) shows substantial equivalence to a legally marketed predicate device. A PMA is full premarket approval for high-risk Class III devices, backed by valid scientific evidence of safety and effectiveness. De Novo creates a new Class I or II classification for a novel device with no predicate and low-to-moderate risk.
No. FDA clears 510(k) devices, approves PMA devices and grants De Novo requests. Using "approved" for a 510(k) device is imprecise.
Often not; many 510(k)s rely on bench and performance testing. Some need clinical data, for example when the technology or indication differs from the predicate. Check current FDA guidance for your product code.
When the device is new, has no suitable predicate, and its risks are low to moderate so that general or special controls can provide reasonable assurance. A sponsor may submit De Novo directly without a prior 510(k).
Yes. Once granted, the device and its classification are available as a predicate for future 510(k) submissions of similar devices.
The same fundamentals as any regulated study: consent records, protocol-driven forms, device-specific logs, adverse event capture, an audit trail and clean exports. Capture provides these with 21 CFR Part 11-aligned controls and a free sandbox for testing.
Keep exploring
EDC for medical device trials
Device study data capture.
Early feasibility and pivotal device studies
Staged IDE studies.
EDC for diagnostic accuracy studies
Test performance studies.
AI medical device validation
Validating AI devices.
Good Clinical Practice
The conduct standard.
Biologics license application
The biologics equivalent.
Phases of clinical trials
The drug convention.
Free sandbox with every feature. No credit card.