Substantial Equivalence and the Not Substantially Equivalent Decision: How FDA Clears and Rejects 510(k) Devices

The 510(k) clearance pathway turns on a single legal determination: whether a new device is substantially equivalent to a predicate already on the market. For regulatory and clinical teams, that determination shapes submission strategy, predicate selection, performance testing scope, and the risk of receiving a not substantially equivalent letter that can delay or derail a product's path to market.

The analysis below walks through the statutory definition of substantial equivalence, the four-part framework FDA applies, and the specific circumstances—intended use divergence, unresolved safety questions, and technological differences without adequate supporting data—under which FDA has issued not substantially equivalent decisions.

Want to ask Rhizome your own regulatory questions? Try it for free.

Substantial equivalence and the "not substantially equivalent" decision: how FDA clears (and rejects) 510(k) devices

Substantial equivalence (SE) is the legal test at the center of the 510(k) program. It is how FDA decides whether a new device is enough like a device already on the market to reach patients without a full premarket approval (PMA). Understanding SE, and its mirror image, the "not substantially equivalent" (NSE) determination, is essential for anyone building a 510(k) strategy: the same four-part analysis that clears a device is the analysis that sinks one.

What "substantial equivalence" means

A device is substantially equivalent when, compared to a legally marketed predicate device, it has the same intended use and either the same technological characteristics, or different technological characteristics that do not raise different questions of safety and effectiveness and that the submitted information shows are as safe and effective as the predicate 8216.

The statutory basis is section 513(i) of the Federal Food, Drug, and Cosmetic Act. Section 513(i)(1)(A) states that a device is substantially equivalent if it has the same intended use as the predicate and either the same technological characteristics, or different technological characteristics where the submitted information demonstrates the device is as safe and effective as a legally marketed device and does not raise different questions of safety and effectiveness 21110. Section 513(i)(1)(B) defines "different technological characteristics" as a significant change in the materials, design, energy source, or other features of the device relative to the predicate 85211.

Substantial equivalence does not mean identical. It means the new device is at least as safe and effective as its predicate for the same intended use, and that any differences do not open safety or effectiveness questions the predicate never had to answer.

The five-step SE decision framework

FDA works through a sequence of decision points. A device must pass every one to be found substantially equivalent:

  1. Same intended use. FDA first asks whether the new device and the predicate have the same intended use. Differences in indications for use do not automatically create a new intended use; they become a new intended use only when they affect, or may affect, safety or effectiveness in a way that cannot be adequately evaluated under the SE standard 8216.
  2. Same or different technological characteristics. If the devices do not share the same technological characteristics, FDA moves to whether the differences raise different questions of safety and effectiveness. "Different technological characteristics" means a significant change in materials, design, energy source, or other features 85211.
  3. No different questions of safety and effectiveness. If technological differences exist, FDA must find that they do not raise new or different safety and effectiveness questions. If they do, the device is found NSE 211.
  4. Performance data demonstrate the device is as safe and effective. Where differences exist but do not raise new questions, FDA evaluates whether the submitted information, including clinical or scientific data if needed, shows the device is as safe and effective as the predicate. FDA reviews the proposed scientific methods and the accompanying performance data to decide whether the methods are acceptable and whether the data support SE 871617.
  5. Adequate, complete data. When performance data are provided, FDA emphasizes that full test reports describing how testing was conducted are crucial to the SE assessment 28.

On predicate selection, FDA recommends starting from legally marketed devices with the same intended use whose technological differences do not raise different questions, then narrowing to the predicate(s) actually cited 22. Best practice is to pick predicates that use well-established methods, meet or exceed expected safety and performance, and are not associated with unmitigated safety issues or design-related recalls 373842. Multiple predicates are allowed in limited circumstances, such as combining features of two or more predicates with the same intended use, or supporting more than one indication under the same intended use 22. Reference devices are not predicates: they cannot be used to satisfy Decision Points 1 through 4 and may support only scientific methodology or standard reference values at Decision Point 5a 34. A new design feature or added component must still meet the SE standard against at least one predicate from the same classification regulation 34.

When FDA finds a device NOT substantially equivalent

An NSE determination is the failure of that same framework. FDA finds a device not substantially equivalent on four grounds:

1. Different intended use. FDA can only find a device SE if it has the same intended use as the predicate 121116. A change in indications amounts to a new intended use when it affects, or may affect, safety or effectiveness and cannot be evaluated under the SE standard 824. Changes in patient population, anatomical location, or clinical context can create a new intended use when they significantly affect safety or effectiveness 9.

2. Different technological characteristics that raise new questions of safety and effectiveness. Even with the same intended use, FDA finds NSE when the device has different technological characteristics and those differences raise different questions of safety and effectiveness, meaning questions not already considered for the predicate 28111614. If the differences do not raise new questions, FDA proceeds to evaluate the performance data instead 211.

3. Performance data do not demonstrate substantial equivalence. When intended use is the same and technological differences do not raise new questions, FDA still reviews the data to confirm the device is as safe and effective as the predicate. If the performance data, bench or clinical, fail to demonstrate SE, FDA can find the device NSE 27916.

4. Insufficient information. The SE determination depends entirely on the information submitted. If the submission does not contain enough data to support the required comparisons or to show the device is as safe and effective as the predicate, FDA cannot make an SE finding on the record as provided 281116.

A practical note on procedure: FDA treats "not substantially equivalent" as a final decision, not a hold or deficiency letter that keeps the submission open 268269.

Worked examples of the NSE grounds

FDA's SE guidance illustrates these grounds with concrete scenarios. These are the clearest documented picture of the NSE analysis in practice, because FDA publishes detailed decision summaries for cleared (SE) devices but does not routinely post standalone NSE letters for individual submissions.

New intended use. A change in indications can be a new intended use when it raises different safety or effectiveness questions 249:

  • A general surgery device cleared only to treat external injuries, relabeled with instructions to use it inside a body cavity: the predicate comparison may not address the infection risk of internal use, so the new indication may be a new intended use requiring De Novo or PMA 249.
  • A surgical ablation device cleared for ablation of cardiac tissue, resubmitted for treatment of atrial fibrillation: the specific atrial fibrillation indication raises safety and effectiveness questions the predicate did not, making it a new intended use 249.
  • An in vitro diagnostic HER2 test moving from a breast cancer population to a lung cancer population to guide trastuzumab use is a new intended use 266. In the PMA context, FDA treats a ventricular assist device changing from bridge-to-transplantation to destination therapy as a new patient population 251.
  • A device indicated near critical organs, expanded to a different anatomic location that poses no additional or different risk, may not be a new intended use; but expanding to a location with greater proximity to critical organs, or that is technically more complex, may require clinical data 6.

Different questions of safety and effectiveness from technology. Technological differences raise different questions when the new technology introduces issues not previously considered 14:

  • A biological indicator using natural bacterial spores versus one using recombinant/genetically engineered technology, where the fluorescent signal reflects plasmid enzyme expression rather than spore viability. The change could produce false negatives if viable bacteria lack sufficient plasmid, raising new questions about plasmid, host spore, and indicator performance 14.
  • A mechanical embryo dissection device versus an electrical one: the change in energy source raises new safety questions about heating effects on the embryo 14.
  • An internal oral airway device that mechanically repositions pharyngeal soft tissue versus an external device on the mandible and neck that applies continuous vacuum. The external negative pressure on vascular, respiratory, and nerve structures raises new safety questions, including nerve stimulation risk 14.
  • A non-resorbable implanted support device changed to a resorbable material, or an IVD switching to a different monoclonal antibody clone, may require clinical data because prior performance data may not transfer 27.

What happens after an NSE: automatic class III and the De Novo route

An NSE determination has a hard consequence. The device remains in class III and may not be legally marketed unless it is later found SE to an existing class I, II, or preamendments device, is reclassified under section 513(f)(3), obtains an approved PMA, or is granted a De Novo request 58.

This is the "automatic class III" problem. A genuinely new type of device that FDA has not previously classified is statutorily placed in class III by operation of section 513(f)(1), regardless of its actual risk, simply because no predicate exists 60. The De Novo pathway is Congress's fix: a requester can ask FDA to classify a device that has no legally marketed predicate, or that received an NSE determination in certain circumstances 110. If the requester shows that general controls, or general and special controls, provide a reasonable assurance of safety and effectiveness under section 513(a)(1)(A) or (B), FDA grants the request and classifies the device into class I or class II 1519. That down-classification lets the device be marketed, creates a new classification regulation, and makes the device available as a predicate for future 510(k)s 16811. If the request is declined, the device stays in class III and cannot be marketed 18.

One eligibility limit matters for strategy: devices found NSE solely because their performance data were inadequate to demonstrate substantial equivalence are generally not eligible for De Novo 67. De Novo is a route for novel device types that lack a predicate, not a second chance to cure a failed data package against an existing predicate.

Reading NSE through De Novo grants

Because standalone NSE letters are not published in the same public form as SE clearance summaries, the most visible real-world footprint of the NSE-to-De Novo path is the set of granted De Novo classifications. Each represents a device type that could not reach the market on substantial equivalence, either because no adequate predicate existed or because the device otherwise fell into automatic class III. Recent examples of such novel device types classified through De Novo include the DeepView AI System (DEN250028) 186, the Aurie System (DEN250023) 187, Essilor Stellest (DEN250016) 188, the Neurolyser XR (DEN250015) 189, LifeVac (DEN250012) 192, the VitaSmart Hypothermic Oxygenated Perfusion System (DEN250009) 193, the Spur Peripheral Retrievable Stent System (DEN240048) 211, the Teal Wand (DEN240045) 213, and the Visby Medical Women's Sexual Health Test (DEN240020) 222. De Novo decision summaries such as those for Kerasave (DEN200063) 223, the AMStent Tracheobronchial Covered Stent System (DEN230087) 224, the Parsortix PC1 Device (DEN200062) 225, and the MISHA Knee System (DEN220033) 240 describe the same statutory background: the De Novo pathway is available for devices without a predicate or after an NSE determination.

Bottom line for RA strategy

Substantial equivalence is not a similarity comparison, it is a structured, sequential test: same intended use, then technological characteristics, then new questions of safety and effectiveness, then performance data, then completeness. An NSE determination is what happens when a device fails any step, and it is a final decision that drops the device into class III 58268269. The four grounds, different intended use, technology that raises new questions, data that do not demonstrate SE, and insufficient information, map exactly onto the same decision points 1278111416. For a novel device type with no adequate predicate, the De Novo pathway is the intended off-ramp, with the important caveat that an NSE driven purely by an inadequate data package against an existing predicate generally does not qualify 67.