FDA 510(k) Clearance: The Complete Guide (2026)

The 510(k) is the most-used FDA route to the US medical device market. Thousands of devices clear through it every year — from orthopedic implants to diagnostic software. The logic is simple to state and easy to get wrong: show the FDA that your device is substantially equivalent to a device already on the market, and back it with the right testing. This guide walks through what 510(k) actually is, the three submission flavors, the step-by-step process, the 2026 changes (QMSR, cybersecurity, eSTAR), realistic timelines and costs, and when 510(k) is the wrong route entirely.

This is planning intelligence, not legal or regulatory advice. It summarizes FDA requirements, guidance, and industry reporting as of September 2026. Submission requirements, user fees, and review timelines change — confirm every detail with qualified regulatory counsel before committing a calendar or a budget. Nothing here guarantees clearance.

What 510(k) actually is — and what it is not

A 510(k) is a premarket notification: you notify the FDA that you intend to market a device and demonstrate that it is substantially equivalent to a predicate — a device that is already legally marketed in the US. Substantial equivalence has two parts: the device has the same intended use as the predicate, and it has the same technological characteristics, or any differences do not raise new questions of safety and effectiveness (supported by data).

Two distinctions matter. First, the FDA clears 510(k)s; it approves PMAs. "Cleared" versus "approved" is not marketing language — it reflects the legal basis of the decision, and using the wrong word in promotional material is itself a compliance risk. Second, 510(k) is not a lighter version of a PMA for the same device class: it is a different legal route. Most cleared devices are Class II; many Class I devices are exempt from 510(k) entirely, and most Class III devices require a PMA.

The three flavors: Traditional, Abbreviated, Special

There is no single 510(k). The FDA runs three submission types, and choosing the right one is one of the earliest decisions in the process.

Step by step: from classification to clearance

• Classify the device: confirm the regulation number, product code, and device class — this decides whether 510(k) is even the right route
• Find a predicate: identify a legally marketed device with the same intended use and comparable technology; a weak predicate is the single most common reason 510(k)s fail
• Plan the testing: bench testing, biocompatibility, software documentation, and electromagnetic compatibility as the device demands — test against the predicate's gaps, not in a vacuum
• Build the quality system: the QMSR is in effect since February 2026 (see below) — the FDA expects design controls to be in place before the submission
• Prepare in eSTAR: the electronic template is mandatory for 510(k)s and acts as a completeness guide — most RTA (Refuse to Accept) rejections trace back to incomplete eSTAR sections
• Submit and pass RTA: the FDA screens for administrative completeness before substantive review begins
• Answer the interactive review: the FDA will almost always ask questions; response speed and quality decide the real-world timeline

eSTAR and RTA: the mechanics that trip people up

Since October 2023, 510(k) submissions must use the eSTAR electronic template — PDFs assembled by hand are no longer accepted. eSTAR is more than a format: it walks the submitter through every required element, and the FDA uses it as the screening checklist. A submission that fails the Refuse to Accept (RTA) screen never reaches a reviewer; it bounces back for correction, and the clock never starts. The typical RTA causes are unglamorous: missing sections, inconsistent indications-for-use statements, and testing summaries that do not connect back to the substantial-equivalence claim.

What is new in 2026: QMSR, cybersecurity, software

The Quality Management System Regulation (QMSR) took effect on 2 February 2026, replacing the old Quality System Regulation's structure with ISO 13485:2016 incorporated by reference. For 510(k) submitters this changes the quality-system language the FDA expects: procedures, terminology, and records should align with ISO 13485 rather than the old QSR part numbers. Companies holding both an FDA clearance and a CE mark benefit from the convergence; companies with a purely FDA-oriented quality system needed to remap their documentation.

Cybersecurity is now a submission element, not a nice-to-have. Under FD&C Act §524B, cyber devices — including devices going through 510(k) — need a software bill of materials (SBOM) and a plan for monitoring and remediating vulnerabilities. Software documentation follows the FDA's risk-based expectations: the higher the software's role in the device's safety, the deeper the documentation. For AI/ML functions, the FDA's PCCP guidance (December 2024, reissued August 2025) adds a pre-authorized plan for future algorithm changes — covered in detail in our AI device guide.

Timelines and costs: what to actually budget

The MDUFA V performance goal for a 510(k) substantive decision is 90 FDA days — note "FDA days": time the FDA spends waiting on the submitter does not count, and interactive review rounds add industry days on top. In practice, a clean Traditional 510(k) runs roughly six to nine months from submission to clearance; anything with major questions takes longer. The real schedule risk is not the FDA goal — it is the RTA bounce and the quality of your answers during interactive review.

User fees follow the published annual MDUFA schedule and change every fiscal year, with a reduced fee for qualified small businesses — so this guide states no dollar figures. Budget the fee from the current schedule, and budget separately for what actually drives cost: testing, clinical data if the predicate argument needs it, and regulatory support. A submission that needs new clinical data is rarely a budget 510(k) anymore.

When 510(k) is the wrong route

Three situations rule out 510(k). No predicate exists and the risk is low-to-moderate: that is De Novo, which creates the classification — and the predicate — for everyone who follows. The device is high-risk (most Class III): that is usually a PMA, with full clinical evidence. Or the device is exempt: many Class I devices need only establishment registration and device listing, no 510(k) at all. Forcing a device into 510(k) when De Novo fits is a classic expensive mistake — the equivalence argument collapses, and the submission dies after months of review.

For the structured pathway view — regulator, conformity route, and document logic — see the GHMAP regulatory pathways/united-states/class-ii page for the FDA route and the EU Class III regulatory pathway page for the EU's highest-risk route on ghmap.io.

Mistakes that kill 510(k)s

• A predicate with a different intended use — the equivalence argument never recovers
• Technological differences hand-waved instead of tested — new questions of safety need new data
• Indications-for-use statements that drift between eSTAR sections
• Submitting before the quality system and design controls are real — QMSR made this check sharper in 2026
• Ignoring cybersecurity documentation for a connected device until the FDA asks
• Treating interactive review as negotiation instead of answering the question asked, with data

The 510(k) remains the fastest credible route to the world's largest device market — but only for devices that genuinely fit it. Predicate strategy decides the outcome before the submission is written; everything after that is execution.