510(k) vs PMA vs De Novo: Choosing Your FDA Pathway

2026-09-03 · Caduvo Team

A practical side-by-side of 510(k), De Novo, and PMA pathways—timelines, evidence burden, user fees, and reimbursement trade-offs—to help MedTech teams choose the right premarket route.

Your regulatory path isn’t a choice—it’s a function of device risk and whether a predicate exists. But guessing wrong costs 12–18 months and six figures in rework. Here’s how to pick between 510(k), FDA De Novo, and PMA before you lock your evidence plan.

The Three Pathways at a Glance

• 510(k): For Class I, II, or exempt devices with a legally marketed predicate. You prove substantial equivalence. FDA “clearance,” not approval. Standard FY2024 user fee: $21,760. FDA target review: 90 days, but real-world 3–6 months.
• De Novo: For novel Class I or II devices without a predicate. You demonstrate reasonable assurance of safety and effectiveness. FDA grants a classification and creates a new predicate. Standard FY2024 user fee: $145,068. FDA target review: 150 days, but often 12–18 months in practice.
• PMA: For Class III devices that support or sustain life, are implanted, or present unreasonable risk. Requires clinical data demonstrating safety and effectiveness. FDA “approval,” not clearance. Standard FY2024 user fee: $483,560. FDA target review: 180 days, but including clinical trials, total timeline 2–4 years.

510(k): Substantial Equivalence and Clearance

Three words define the 510(k): substantial equivalence. If your device has the same intended use and similar technological characteristics as a legally marketed predicate, you can clear it through a 510(k). The standard is lower than PMA: you don’t need to prove safety and effectiveness from scratch—you show your device is as safe and effective as the predicate. That’s why the distinction “510k clearance vs approval” matters. Clearance means FDA found substantial equivalence; approval means FDA found your device safe and effective on its own.

Evidence burden is typically bench testing, biocompatibility, and sometimes animal data. Clinical data is rare—only when a material difference raises safety questions. In FY2024, the standard 510(k) user fee is $21,760. FDA’s goal is 90 days, but a poorly chosen predicate or a hold letter adds 30–60 days per response. Pick your predicate before design freeze. Choosing the wrong predicate is the most common avoidable delay. A hold letter for predicate mismatch can cost 60–90 days. Use FDA’s Product Classification database to confirm predicate eligibility before submission.

De Novo: For Novel Low-to-Moderate Risk Devices

No predicate? Your device might need the de novo pathway. The FDA De Novo request is for novel devices with low-to-moderate risk that don’t fit 510(k) because no legally marketed device exists. You must provide enough data for FDA to determine that general controls (and possibly special controls) provide reasonable assurance of safety and effectiveness.

Think of it as a hybrid: you’re asking FDA to classify your device as Class I or II and establish it as a new predicate for future competitors. The review standard is higher than 510(k) but lower than PMA. FDA’s target is 150 days, but real-world timelines stretch to 12–18 months because FDA often asks for iterative data clarifications. The FY2024 De Novo user fee is $145,068. De Novo submissions should include a detailed benefit-risk analysis and proposed special controls.

One strategic note: if you submit a 510(k) and FDA issues an NSE (not substantially equivalent) because no suitable predicate exists, you can request a De Novo classification. Also, devices with Breakthrough Device Designation often use the De Novo pathway with expedited review.

PMA: The Clinical Evidence Bar for Class III

PMA is the heavyweight. If your device is Class III—implanted, life-sustaining, or presents potential unreasonable risk—you must prove safety and effectiveness with clinical data. The “510k vs pma” debate is really about evidence burden: PMA requires prospective clinical trials under an IDE, often with a control group. The FDA review clock is 180 days, but the total time from first patient enrolled to approval typically runs 2–4 years.

You’ll submit a traditional PMA (one large application) or modular PMA (shell plus modules submitted as data becomes available). Both require manufacturing information, clinical data, and labeling. FDA may convene an advisory panel. The FY2024 PMA user fee is $483,560, but that’s a fraction of clinical trial costs. Small business user fees are reduced, but total PMA cost including clinical trials can exceed $1 million.

Downstream, PMA devices are often eligible for a new technology add-on payment under Medicare, but coverage is not automatic. You still need coding, coverage, and payment.

Decision Framework and Reimbursement Trade-offs

Use this sequence:

1. Is there a legally marketed predicate with the same intended use?

• Yes → 510(k), unless your device has different technological characteristics that raise new safety questions (then 510(k) with clinical data or De Novo).
• No → go to step 2.

1. Is the device low-to-moderate risk (no life-support, no implant, not Class III)?

• Yes → De Novo.
• No → PMA.

Reimbursement implications vary by pathway. A 510(k) device may be coded under existing HCPCS codes without new coverage. A De Novo classification creates a new regulation and may require a new HCPCS code, but Medicare often covers De Novo devices under existing benefit categories faster than PMA devices. PMA devices face more scrutiny but can command premium pricing if they meet unmet needs.

Actionable Takeaways

• Lock classification early: wrong pathway triples your timeline.
• If a predicate exists, use it—510(k) is the fastest, cheapest route, but pick your predicate before design freeze.
• No predicate? De Novo is often better than forcing a 510(k) that gets an NSE.
• Reserve PMA for true Class III devices, and plan for clinical trial timelines.
• Reimbursement strategy should inform your pathway choice, not vice versa.