FDA Premarket Approval (PMA) is the most stringent FDA pathway, reserved for Class III devices that support or sustain human life, are of substantial importance in preventing impairment, or present a potentially unreasonable risk. A PMA requires valid scientific evidence — usually pivotal clinical data — and a full manufacturing inspection, typically taking two to four years and costing significantly more than a 510(k).

By Tibor Zechmeister and Felix Lenhard.

TL;DR

  • PMA is FDA's most rigorous device submission pathway, required for most Class III devices where general and special controls are insufficient to provide reasonable assurance of safety and effectiveness.
  • The legal basis is the Federal Food, Drug, and Cosmetic Act Section 515; the implementing regulations sit in 21 CFR Part 814 .
  • A PMA must contain valid scientific evidence, usually including at least one pivotal clinical investigation conducted under an Investigational Device Exemption (IDE).
  • FDA user fees for a standard PMA submission are in the high six-figure USD range for standard applicants, with reduced fees for qualifying small businesses .
  • EU startups entering the US through PMA should expect 24–48 months from IDE to approval, not including pre-submission iterations.
  • The closest MDR analogue is Class III conformity assessment under Annex IX (Chapter I + Chapter II) plus Annex X type examination, but the two systems differ structurally.

Why this matters

Most EU MedTech founders enter the US thinking "510(k)" — find a predicate, demonstrate substantial equivalence, submit. For roughly 99% of devices, that instinct is right. But a small, high-stakes subset of products cannot take that path. Implantable cardiac devices, novel life-supporting ventilators, new-chemistry hemostatic agents, and first-of-their-kind diagnostic systems used in critical care all land in PMA territory. If your device is one of them, the US market is not a six-month project you bolt on after CE marking. It is a parallel multi-year regulatory program with its own clinical strategy, its own budget, and its own team.

Tibor has sat on both sides of the table on devices that straddled Class IIb/III in Europe and Class II/III in the US. The mistake he sees most often: founders assume the MDR dossier they built for a Notified Body will carry them through PMA with translation and formatting. It does not. Some artifacts transfer. The core clinical strategy usually has to be rebuilt.

What FDA actually says (and what MDR says in parallel)

The FDA side

PMA is defined in Section 515 of the Federal Food, Drug, and Cosmetic Act. 21 CFR Part 814 sets out the content, format, and procedures for PMA applications . The statutory standard is "reasonable assurance of safety and effectiveness" — a higher bar than 510(k)'s "substantial equivalence to a predicate." FDA must have valid scientific evidence that the device, for its intended uses and conditions of use, will provide clinically significant results.

A PMA submission typically contains: - Device description, principles of operation, and manufacturing information - Non-clinical laboratory studies (bench, animal, biocompatibility, sterilization, software, cybersecurity) - Clinical investigations — usually at least one pivotal study conducted under IDE - A proposed labeling package - Financial certification and disclosure statements for clinical investigators - An environmental assessment or claim of categorical exclusion

FDA then conducts a pre-approval inspection of the manufacturing site against the Quality System Regulation, which is transitioning to the Quality Management System Regulation (QMSR) aligned with ISO 13485 .

The MDR parallel

Under MDR (EU) 2017/745, Class III devices follow Article 52 and are routed through Annex IX (full QMS assessment plus technical documentation assessment) or the combination of Annex X (type examination) with Annex XI (production quality assurance). Clinical evidence requirements sit in Article 61 and Annex XIV Part A. For implantable and Class III devices, Article 61(4) sets the default expectation that clinical investigations are conducted unless specific exemptions apply, and MDCG 2023-7 clarifies the exemption pathways.

The regimes converge on one idea — the highest-risk devices need real clinical data generated on the device itself — but they diverge on everything else. FDA wants a single federal submission reviewed by one agency against one statutory standard. MDR distributes the work across a Notified Body (conformity assessment), competent authorities (clinical investigation authorization under Article 70), and in some cases an expert panel consultation under Article 54 for certain Class III implantables.

A worked example

Consider a Vienna-based startup developing a novel implantable neurostimulator for treatment-resistant epilepsy. Under MDR, the classification rules (likely Rule 8 as a long-term active implantable) push it to Class III. Annex IX applies. A pivotal clinical investigation under Article 62 and EN ISO 14155:2020+A11:2024 is non-negotiable. Expert panel consultation under Article 54 is likely triggered.

In the US, the same device lands in Class III by rule — active implantable neurological stimulators are historically PMA devices. The founders must:

  1. Schedule a Pre-Submission (Q-Sub) meeting with FDA to align on study design, endpoints, and statistical plan before spending a single euro on US clinical work.
  2. Submit an IDE application under 21 CFR Part 812 , including the investigational plan, prior investigations, manufacturing info, and IRB documentation.
  3. Conduct the pivotal trial under the approved IDE. For a neurostim device, expect 150–250 subjects, primary endpoint at 12 months, long-term follow-up of several years.
  4. Compile the PMA, including the full clinical study report, manufacturing section, and labeling.
  5. Pay the PMA user fee — which for FY2026 sits in the high six-figure USD range for standard applicants, with a reduced small-business fee if the company qualifies .
  6. Host the FDA pre-approval inspection of the manufacturing facility.

Total calendar time from first Q-Sub to PMA approval: 36–60 months is realistic for a genuinely novel device. Total direct regulatory and clinical spend: easily eight figures USD when you include the pivotal trial. This is not a 510(k). This is not CE marking plus translation. This is a second, parallel program.

The key insight: the MDR pivotal study and the FDA pivotal study can often be designed as one global study if you plan early. If you plan late, you run two studies. That difference alone can be the difference between raising one Series B and raising two.

The Subtract to Ship playbook

PMA is where "minimum viable regulatory" runs into a wall. You cannot subtract your way out of a pivotal trial for a life-sustaining device. But you can subtract everything that is not the critical path.

1. Confirm the pathway before you build the company. If your device class and intended use push you to PMA in the US and Class III under MDR, that is the single most important fact in your business plan. It drives funding, team, timelines, and exit strategy. Use FDA's product classification database and the MDR classification rules to triangulate early.

2. Use Pre-Submission aggressively. Q-Subs are free, non-binding, and the single highest-leverage interaction with FDA. Bring a focused question, a clear proposal, and accept that the meeting's value is in what FDA tells you not to do.

3. Design one global pivotal trial. Get your statistician, your MDR Notified Body contact, and your FDA reviewer aligned on endpoints, inclusion criteria, and follow-up duration before the protocol is locked. Align to EN ISO 14155:2020+A11:2024 for EU acceptance; align to 21 CFR Part 812 for FDA acceptance. The overlap is larger than most founders believe.

4. Build the QMS once, label it twice. A QMS built to EN ISO 13485:2016+A11:2021 can be mapped to QMSR with discipline. Do not build two quality systems.

5. Budget realistically. The true cost of CE marking is already six to seven figures for a Class III device. Layer PMA on top and you are in the tens of millions. If your round does not support that, either pick one market first or reconsider the intended purpose.

6. Sequence markets deliberately. Most EU PMA-bound startups ship CE first, then follow with PMA using the MDR clinical data as supporting evidence. Some do the reverse. Both work; neither is cheap.

Reality Check

Answer honestly. Each "no" is a risk you need to resolve.

  1. Do you know with certainty whether your device is Class III under MDR Annex VIII and whether it falls under an existing PMA regulation in the US?
  2. Have you had a Pre-Submission meeting with FDA, or mapped out what you would ask in one?
  3. Is your clinical investigation plan designed to satisfy both EN ISO 14155:2020+A11:2024 and FDA IDE expectations?
  4. Have you budgeted separately for (a) CE marking, (b) IDE clinical trial, (c) PMA submission fee, (d) FDA pre-approval inspection readiness?
  5. Does your fundraising narrative account for a 36–60 month PMA timeline?
  6. Is your QMS under EN ISO 13485:2016+A11:2021 designed in a way that maps cleanly to QMSR?
  7. Do you know which FDA review division owns your device type, and have you read their recent guidance documents?
  8. Have you decided which market ships first, and is that decision driven by data rather than preference?

Frequently Asked Questions

Can I use my MDR clinical investigation data in a PMA submission? Often yes, but rarely without gaps. FDA will review the protocol, endpoints, statistical analysis, and site monitoring against its own expectations. If the study was designed with FDA acceptance in mind from day one, the overlap can be 80%+. If not, expect to run supplementary work.

How much does a PMA user fee actually cost? For FY2026, the standard PMA fee is in the high six-figure USD range, with a reduced small business rate for qualifying applicants . Check the FDA user fees page for the current fiscal year — FDA updates these annually.

Is PMA always required for Class III US devices? For devices in PMA-regulated categories, yes. Some Class III devices were grandfathered or reclassified over the years, so a small number of Class III devices still clear via 510(k). The product classification database is the source of truth.

How does PMA compare to MDR expert panel consultation under Article 54? They are not equivalent but conceptually parallel. Article 54 triggers expert panel scientific opinion for certain Class III implantables and Class IIb active devices intended to administer or remove medicinal products. PMA triggers FDA advisory panel meetings for novel or controversial devices. Both are expert review layers on top of the primary regulatory review.

Can a startup realistically afford a PMA pathway? Yes, but only if the business model and funding match the regulatory burden. PMA-bound startups typically raise larger Series A/B rounds, partner with strategics earlier, or pursue non-dilutive grants in parallel. See our piece on pre-seed and seed funding for MedTech.

Is De Novo a backdoor to avoid PMA? Only for genuinely novel, low-to-moderate risk devices without a predicate. De Novo is a reclassification pathway to Class I or II, not a way to escape PMA for high-risk devices.

Sources

  1. Regulation (EU) 2017/745 on medical devices, consolidated text. Article 52, Article 54, Article 61, Annex IX, Annex X, Annex XIV Part A.
  2. Federal Food, Drug, and Cosmetic Act Section 515. 21 CFR Part 814 (Premarket Approval of Medical Devices) .
  3. 21 CFR Part 812 (Investigational Device Exemptions) .
  4. EN ISO 14155:2020+A11:2024 — Clinical investigation of medical devices for human subjects.
  5. MDCG 2023-7 (December 2023) — Clinical investigation exemptions under Article 61(4)-(6).