A surprising number of device programs lose months before the submission is even drafted. The problem is not poor engineering or weak clinical thinking. It starts earlier, when teams are still asking how to choose 510(k) or De Novo and make the wrong call based on assumptions about novelty, intended use, or what the FDA will consider a valid predicate.
For med tech companies, this is not a paperwork question. It is a business decision with direct impact on timeline, evidence strategy, cost, reimbursement planning, and investor confidence. If you choose a 510(k) path when no legally marketed predicate truly supports substantial equivalence, you can end up in repeated FDA questions or a dead end. If you default to De Novo too quickly, you may overbuild your evidence package and extend time to market unnecessarily. The right pathway depends on a disciplined reading of the device, the claims, the risk profile, and the current regulatory landscape.
How to choose 510(k) or De Novo without guessing
The core distinction is straightforward. A 510(k) relies on substantial equivalence to a legally marketed predicate device. A De Novo request is generally used for novel devices with no suitable predicate, where the risk can still be reasonably controlled through general controls or general and special controls.
In practice, the decision is rarely that simple. Many devices look similar to products already on the market, but the similarities may not be enough once you compare intended use, indications for use, technological characteristics, and risk. Teams often focus too heavily on product design and not enough on regulatory fit. Two devices can share hardware features and still fall into different pathways if the clinical context or performance questions are materially different.
A useful starting point is this: if you can identify a predicate with the same intended use and comparable technology, and any differences do not raise new questions of safety or effectiveness, a 510(k) may be viable. If no suitable predicate exists, or if your differences create new regulatory questions that cannot be bridged cleanly, De Novo becomes the more realistic option.
Start with intended use, not product similarity
The fastest way to distort pathway strategy is to begin with a list of devices that look like yours. FDA review does not start with visual similarity. It starts with what the device is intended to do, for whom, and under what conditions.
A company developing a digital therapeutic accessory, AI-enabled diagnostic aid, or a modified interventional device may find products with overlapping features in the market. That does not mean those products can serve as predicates. Small shifts in the user population, anatomical site, disease state, or clinical decision role can change the regulatory analysis substantially.
When evaluating 510(k) feasibility, compare your draft indications for use against potential predicates line by line. If you find yourself rewriting your intended use mainly to force alignment with an available predicate, that is a warning sign. The submission strategy should support the product you plan to commercialize, not a narrower version invented to fit a pathway.
The predicate question is stricter than many teams expect
A valid 510(k) strategy requires more than finding something in the same product category. You need a predicate that is legally marketed and close enough to support substantial equivalence. If the technological differences affect performance, biocompatibility, software behavior, sterility, electromagnetic compatibility, or clinical function in a meaningful way, FDA may conclude the differences raise new questions of safety or effectiveness.
That does not automatically rule out 510(k). Many cleared devices differ from predicates in important ways. The issue is whether those differences can be resolved through testing and a coherent equivalence argument. If they cannot, a De Novo pathway may be more appropriate even if the product appears adjacent to existing technologies.
Risk classification matters more than novelty alone
Novel does not always mean De Novo, and familiar does not always mean 510(k). The more useful lens is risk and controllability.
De Novo is designed for novel devices that are low to moderate risk and do not have a predicate. If your device introduces a new mechanism, a new clinical role, or a new combination of features, but the risks can be addressed through labeling, bench testing, software validation, usability work, and possibly clinical evidence, De Novo may be the right route. It can also establish a new classification regulation that later devices may use as predicates.
That strategic benefit matters. For some companies, De Novo is not just a fallback when 510(k) is unavailable. It is the clearest path to market access and a chance to define a category. The trade-off is that De Novo typically demands more upfront regulatory work and a more explicit discussion of benefit-risk, classification, and special controls.
When a 510(k) is attractive but fragile
A 510(k) is often appealing because it can be faster and more predictable when the predicate logic is strong. But many borderline 510(k) strategies are fragile. They depend on generous assumptions about equivalence, narrow claim wording, or optimistic views of what testing will resolve.
If your regulatory argument works only if FDA interprets the technology in the most favorable possible way, it is probably not a durable 510(k) strategy. That is where a commercially strategic approach matters. A submission is not successful because it was theoretically possible. It is successful because it survives FDA scrutiny without avoidable cycles of deficiency questions, rework, or pathway collapse.
Evidence planning often reveals the right answer
One of the most practical ways to determine how to choose 510(k) or De Novo is to map the evidence package each pathway would require.
For a 510(k), the focus is usually on demonstrating substantial equivalence through bench, software, electrical safety, EMC, biocompatibility, sterilization, packaging, shelf life, and other performance testing as applicable. Clinical data may not be required, although some device types or technology differences make it necessary.
For De Novo, the evidence burden can be broader. FDA may expect more detailed discussion of risk controls, human factors, analytical validation, clinical performance, and post-market considerations. That does not mean De Novo always requires a large clinical study, but it often requires a more comprehensive justification for why the device is safe and effective in the absence of a predicate framework.
If your team’s evidence plan already looks more like a first-of-kind justification than an equivalence comparison, that is useful information. It suggests the program may be better positioned for De Novo. On the other hand, if the key performance questions can be answered through established nonclinical testing against a well-matched predicate, 510(k) may be the more efficient route.
FDA interaction can reduce expensive assumptions
Companies sometimes treat pathway selection as an internal decision to finalize before engaging FDA. That can be costly. For many devices, especially those with novel features, a pre-submission interaction is one of the most efficient ways to test the regulatory logic before significant resources are committed.
A well-prepared Q-Submission can help clarify whether FDA agrees with the proposed predicate strategy, whether additional data are likely to be expected, and whether the technology raises questions that point away from 510(k). It will not remove all uncertainty, but it can replace speculation with a more informed planning basis.
This is particularly valuable for startups and growth-stage companies managing capital carefully. The pathway decision affects not only submission timing but also design controls, verification and validation planning, clinical budgeting, and commercialization sequencing. Regulatory clarity early in the program protects more than the submission. It protects the operating plan.
Common situations where the answer depends
Some devices sit in the gray zone. Software-enabled devices are a common example. You may find predicates with similar workflows, but if your algorithm influences clinical decisions in a materially different way, FDA may view the risk and effectiveness questions differently. Combination-like features, novel materials, home-use adaptations, and expanded patient populations can create similar issues.
In these cases, there is rarely a shortcut. The right answer depends on the exact claims, the device architecture, the hazard analysis, and the maturity of the comparison set in the market. A disciplined regulatory assessment often saves time by showing where the real fault line is. Sometimes that means confirming a workable 510(k). Sometimes it means recognizing early that De Novo is the cleaner path and planning accordingly.
A strong strategy does not chase the shortest pathway on paper. It chooses the pathway with the highest probability of regulatory success on a commercially acceptable timeline. That is the standard sophisticated med tech teams should use.
If your program is still debating 510(k) versus De Novo, the most useful next step is not to force a decision. It is to pressure-test the intended use, predicate landscape, risk profile, and evidence plan until the right path becomes clear enough to execute with confidence. That is where experienced regulatory judgment creates real value, because getting the pathway right early tends to make everything after it move faster.
