510(k) vs De Novo vs PMA: Which FDA Pathway Do You Need?

The Three FDA Premarket Pathways at a Glance

The FDA classifies medical devices into three regulatory classes based on risk. The FDA has a risk-based framework for medical device classification that sorts devices into class I, class II, and class III. Class I devices are subject to general controls only, and most are exempt from premarket review. Class II devices require special controls in addition to general controls. Class III devices require premarket approval pma for safety because they sustain human life, are implanted, or present potential unreasonable risk. The device class your product falls into largely determines which submission pathway you need.

Most class I devices skip premarket review entirely. Most class II medical devices go through 510(k) clearance or de novo classification. Most class III medical devices require PMA.

The FDA medical device classification system groups devices into three regulatory classes based on the level of regulatory controls needed.

What Is a 510(k) Clearance and When Does Your Device Qualify?

A 510(k) is a premarket notification showing the FDA that your device is substantially equivalent to a legally marketed predicate device. The FDA does not approve a 510(k). It grants FDA clearance. You are not proving your device is better than a predicate. You are proving it is at least as safe and effective for the same intended use.

Devices classified under 510(k) are low to moderate risk, mostly class ii. Some class i devices that are not exempt and a small number of reclassified class iii devices can also use the 510 k pathway.

Key elements of a 510 k submission include:

• Intended use and indications for use
• Device description and technological characteristics
• Performance data (bench testing, biocompatibility, electromagnetic compatibility, software verification)
• Labeling review
• Quality system alignment

About 85% of 510(k) applications receive a Substantially Equivalent decision. 510(k) submissions typically take 30 to 90 FDA review days, though calendar time from submission to clearance often runs 140 to 175 days when you factor in interactive review cycles.

There are three types of 510(k):

• Traditional 510(k): Full comparison to predicate with all supporting data. Used for most submissions.
• Abbreviated 510(k): Relies on a recognized guidance document or consensus standard. Works when your device conforms to published FDA expectations.
• Special 510(k): For modifications to your own previously cleared device. Focuses on design control documentation rather than full testing.

What Counts as a Predicate Device?

A predicate device is a legally marketed device that your 510(k) compares against. It can be a pre-amendments device (marketed before May 28, 1976), a reclassified device, a previously 510(k)-cleared device, or a De Novo-granted device. The predicate must share the same intended use and have similar technological characteristics.

How to find predicates:

• Search the FDA’s product classification database by product code, regulation number, or device name
• Review the 510(k) Premarket Notification database for cleared devices in your category
• Compare indications for use statements, not just device names

Avoid combining indications or technologies from multiple predicates (“split predicates”). The FDA pushes back on these strategies.

What the FDA Reviews in a 510(k) Submission

Think of FDA review as a checklist:

• Indications for Use statement: Must match the predicate’s intended use or justify any differences
• Device description: Materials, components, how the device incorporates its technology
• Technological characteristics comparison: Side-by-side comparison with the existing device
• Performance testing: Bench testing, biocompatibility per ISO 10993, electromagnetic compatibility, software validation, sterilization validation
• Labeling: Instructions for use, warnings, contraindications

FDA uses these pieces to make a substantial equivalence determination. The question is simple: does your device have the same intended use, and do any differences raise new questions about safety and effectiveness?

Since 2023, most 510(k) submissions must use the eSTAR (Electronic Submission Template And Resource) format. Submissions that do not follow eSTAR requirements may be placed on hold or refused.

What Disqualifies a Device from the 510(k) Route?

You cannot use 510(k) when:

• No legally marketed device shares the same intended use
• Your device has a new indication that shifts the risk profile
• Novel technology raises new safety or effectiveness questions that existing performance data cannot answer
• Your device falls into class III without reclassification or a special rule

Real-world triggers: AI-enabled diagnostic software with no comparable predicate, or a new implant using a material with unknown long-term biocompatibility.

After an NSE (Non-Substantial Equivalence) decision, your options are to pursue a de novo request or submit a premarket approval application. Both cost additional time and user fees. If you have any doubt about 510(k) eligibility, validate it through a Pre-Submission meeting before committing.

What Is De Novo Classification and When Is It the Right Move?

A de novo classification request is for novel devices that are low- to moderate-risk and have no legally marketed predicate. Instead of proving equivalence to something that already exists, you ask the FDA to create a new classification regulation for your device. If granted, your device becomes the first of its kind cleared to market, and the De Novo decision itself can serve as a predicate for future 510(k) submissions.

Congress created the de novo process through FDAMA in 1997. It has become common for digital health products, software as a medical device (SaMD), clinical decision support software, and connected devices.

Key features of De Novo:

• De Novo classification is for novel medical devices without predicates
• De Novo requests can lead to class I or class II designations
• A granted De Novo downclassifies a device from an initial class III default
• De Novo requests require a thorough risk assessment and testing
• De Novo devices can serve as predicates for future 510(k) submissions

How De Novo Differs from 510(k) in Practice

The day-to-day project differences between 510(k) and De Novo are real:

• Legal standard: 510(k) asks for substantial equivalence to a marketed device. De Novo asks for reasonable assurance of safety and effectiveness through general and special controls.
• Data expectations: De Novo often needs more extensive nonclinical testing, formal risk analysis (ISO 14971), and sometimes clinical data.
• FDA interaction: De Novo involves more back-and-forth with FDA reviewers, especially around proposed special controls and risk mitigations.
• Review time: De Novo is slower than 510(k) but faster and lighter than PMA.
• User fees: De Novo fees ($173,782 standard) are roughly 6.7x higher than 510(k) fees.

Does De Novo Require Clinical Data?

Clinical data are not always required for a De Novo, but they are common for certain device types.

FDA is more likely to request clinical data when:

• The device makes new diagnostic claims
• Novel implant materials are involved
• Algorithms directly influence treatment decisions

Bench and simulated-use testing may be enough when:

• The device is an accessory or simple monitoring tool
• Materials and functions are well understood
• Risk can be controlled with performance testing and labeling

The Apple Watch ECG feature received De Novo authorization with clinical data supporting its algorithm. Some simpler connected monitoring accessories have gone through with bench data only. Check specific De Novo decision summaries on the FDA database for devices similar to yours.

How Long Does De Novo Review Actually Take?

The Food and Drug Administration aims to decide on de novo requests within 150 FDA review days. Real-world performance tells a different story.

A study of 180 De Novo requests from 2017 to 2023 found a median decision time of 309 calendar days and a mean of 338 days. The range was wide: Apple Watch features were decided in 28 to 33 days, while complex imaging or AI devices took over 30 months.

Typical timeline stages:

• Pre-Submission and study planning: 2 to 4 months
• Submission preparation: 2 to 6 months
• Acceptance review: 15 days
• Substantive review and AI responses: 6 to 12+ months
• Final decision

Complex SaMD or AI/ML de novo requests often sit on the longer end because of questions about training data, algorithm transparency, and change protocols.

What Is Premarket Approval (PMA) and Which Devices Require It?

Premarket approval is the most rigorous pathway. It applies to class iii medical devices that sustain or support life, are implanted, or present significant risk. The PMA process is the most rigorous, necessitating valid scientific evidence for Class III devices. That almost always means human clinical trial data gathered under an Investigational Device Exemption (IDE).

PMA includes a deep review of manufacturing processes, design controls, labeling, and postmarket surveillance plans. It goes far beyond performance data alone.

PMA applications require extensive clinical data for high-risk devices. Even small modifications after PMA approval can require a supplement, compared with the more flexible change rules available to 510(k)-cleared devices.

What Devices Typically Require PMA?

Common PMA device types include:

• Implantable cardioverter defibrillators and pacemakers
• Drug-eluting stents
• Cochlear implants
• Mechanical heart valves
• High-risk in vitro diagnostics (HIV and hepatitis screening tests)

These are devices where incorrect performance can directly cause serious injury or death. The stringent regulatory requirements match the risk.

Some legacy class III devices avoid PMA because they were marketed before May 28, 1976 (pre-amendments device) or have been downclassified through panel review or reclassification orders. You can confirm your device’s status in the FDA product classification database. If unclear, a 513(g) request gives you formal FDA classification feedback.

How Long Does PMA Take and What Does It Cost?

PMA review times average around 180 FDA days for high-risk devices. But the statutory review clock is just one piece.

Realistic timeline phases:

• Early feasibility and device design: 1 to 2 years
• IDE clinical trial execution: 2 to 5 years
• PMA preparation and submission: 6 to 12 months
• FDA review cycles: 12 to 24+ months

Total development and fda approval timelines for complex class iii devices run 3 to 7 years. Some programs run longer.

Cost bands for a full PMA program range from low millions to $10M to $100M or more. Clinical trials account for roughly 40 to 60% of total spending. FY 2026 PMA user fees alone are $579,272 for standard applicants and $144,818 for small businesses.

Any company considering PMA should get FDA feedback on key endpoints and study design through Pre-Submission meetings before committing to large pivotal studies. Extensive scientific evidence requirements make early alignment with FDA non-negotiable.

How to Choose the Right Pathway

Three questions can direct most innovative devices to 510(k), De Novo, or PMA when asked in the right order.

Question 1: Is your device high-risk? Does it sustain human life, is it implantable, or does the FDA already classify similar devices as class III? If yes, PMA is likely required. Confirm through a 513(g) request or a pre-submission meeting. If no, move to Question 2.

Question 2: Does a legally marketed predicate device exist? Can you find an existing device with the same intended use and similar technological characteristics? If yes, a 510(k) is likely your pathway. Focus on predicate selection and substantial equivalence. If no, move to Question 3.

Question 3: Is your device novel but of moderate risk? Is such a device new to the market, but controllable with general and special controls? If yes, a de novo classification may be appropriate. Expect a longer review and more FDA interaction than with a 510(k). If uncertain, a Pre-Submission (Q-Sub) meeting is the right next step.

Write down your answers before reading further. This turns the logic into a practical worksheet for your team.

When the Answer Is Not Clear-Cut

Some devices sit between pathways. Grey areas include:

• Software as a medical device with clinical decision support software claims
• AI/ML systems that learn and change over time
• Connected devices that blur device and service boundaries
• Combination products (device plus drug or biologic)

A weak or partial predicate can tempt teams to pursue a risky 510(k) strategy when a De Novo may be a better fit. If your team spends weeks building a stretched substantial equivalence argument, that is a signal.

One common scenario: a startup planned a 510(k) for a new device using an AI-based diagnostic algorithm. During a Q-Sub, the FDA indicated the algorithm raised new safety questions with no valid predicate. The team pivoted to De Novo early, avoiding an NSE determination that would have cost six months and a wasted user fee.

A Pre-Submission meeting is the single best investment when the pathway is genuinely ambiguous.

Common Mistakes When Choosing the Wrong FDA Pathway

Choosing the wrong pathway does not just delay your submission. It can result in an NSE determination, a failed De Novo, or years spent in PMA clinical trials for a new device that could have cleared via 510(k). Many of these errors come from skipping early classification work or overestimating predicate strength.

Mistake 1: Selecting a Weak or Mismatched Predicate

A weak predicate has a different intended use, a different user population, or a different core technology, raising new safety questions. FDA responds with detailed requests for justification or an NSE.

One company chose a predicate based only on a similar form factor. It ignored differences in indications and risk. The result: an NSE after four months and a restart under De Novo.

Rule of thumb: Start predicate searches from intended use and risk profile, not from superficial design similarities.

Mistake 2: Filing a 510(k) When De Novo Was Required

This happens often with digital health, SaMD, and AI tools. The company files a 510(k), FDA determines the device raises new safety or effectiveness questions, and issues an NSE. The premarket notification process ends, and the company must pursue the de novo process or PMA.

This can add 6 to 12 months and require a fresh user fee plus updated testing. If you are stretching the substantial equivalence argument, De Novo may already be the more realistic option.

Mistake 3: Underestimating What De Novo Actually Requires

Some teams treat De Novo as a “big 510(k)” and underprepare. Common gaps include:

• Incomplete risk management file
• Missing justification for proposed special controls
• Limited performance data that does not cover worst-case conditions

These gaps trigger Additional Information letters and long pauses. Treat De Novo planning closer to a small PMA: formal clinical strategy, thorough bench testing, and clear proposals for general and special controls class iii default devices need.

Mistake 4: Starting a PMA Without FDA Alignment First

Some medical device manufacturers design and run expensive clinical trials before confirming endpoints, patient populations, and comparators with FDA. This can produce clinical data that do not answer FDA’s actual questions, forcing new studies.

Use Pre-Submission and IDE interactions to lock in expectations. Investors and boards should expect to see FDA feedback letters before funding long PMA programs under the fd c act requirements.

Before You Submit: Why a Pre-Submission Meeting Changes Everything

A Pre-Submission meeting (Q-Sub) is a free FDA feedback mechanism. You present your device, proposed indications, and regulatory strategy. FDA provides written responses to your questions. It is not legally binding, but FDA’s written record means the agency cannot later claim it was unaware of your approach.

Benefits of a Q-Sub:

• Confirms fda device classification and likely pathway
• Reduces risk of NSE or declined De Novo
• Clarifies clinical and testing expectations
• Aligns eSTAR content requirements early

To prepare an effective Q-Sub, you need a clear device description, draft indications for use, a preliminary testing plan, and specific questions about classification and the approval process. A guidance document from FDA outlines the Q-Sub format and timeline.

Confirm Your FDA Pathway Before You Commit

Pathway selection drives total time, cost, and data needs for marketing authorization. Getting it right before submission saves months. Getting it wrong is one of the most common and preventable reasons submissions fail.

• 510(k): For moderate risk devices with a strong predicate and clear substantial equivalence
• De Novo: For novel moderate risk devices without a predicate, where regulatory controls can manage risk
• PMA: For most class iii devices where only a premarket approval application with extensive scientific evidence will satisfy FDA

Early classification work and Q-Sub discussions are cheaper than recovering from an NSE or failed De Novo.

Book a Free Consultation

Book a free pathway consultation with a regulatory strategist who can review your device, intended use, and predicate landscape and confirm the right path forward.