Regulatory Classification & Pathway Strategy
Classification is the decision every other regulated activity inherits. Whether a product is a drug, a biologic, a device, a combination, a food or a cosmetic determines the evidence it must generate, the quality system it is built under, the regulator that reviews it, and the route by which it reaches the market. Get it wrong and the studies run to the wrong standard — an error that surfaces at submission, where it is most expensive to correct.
What an explainer is not
A topic explainer is SPEQ’s synthesis of what a practice involves, cited to the standards that govern it. It does not reproduce their text, and it does not determine which of them apply to your product or process.
[ POSITION IN THE FRAMEWORK ]
7 DIMENSIONS · 22 LINKSClassification decides which body of law applies, which evidence is needed and which authority reviews it — so it is the one regulatory decision that is cheap while the product is an idea and ruinous once the design is fixed.
06 · QUALITY MATURITY — REGULATORY CLASSIFICATION & PATHWAY STRATEGY, REACTIVE TO ADAPTIVE
Classification is assumed from what the product resembles commercially. Nobody has written down why it is that and not something adjacent.
A rationale exists and was written for the first market, then carried to the others on the assumption that the answer travels.
The determination is argued from primary mode of action and intended use, made per market, and recorded early enough to shape the development plan.
A change in intended use, claim or component triggers reassessment, so the classification tracks the product rather than the plan it was filed under.
Classification is a design constraint: the product is deliberately shaped toward a pathway the organisation can actually evidence, rather than a pathway being sought for a product already built.
SPEQ’s shared five-stage progression, labelled synthesis — not the FDA QMM rating scale. Where does your organization sit? Score your quality system →
07 · REGULATORY & EVIDENCE
GOVERNING STANDARDS · 5
Derived from the 5 standards SPEQ maps to this subject, across 3 regulatory bodies: EC, IMDRF, FDA.
RECORDS & OBJECTIVE EVIDENCE
- The classification rationale, stating primary mode of action and intended use
- Per-market determinations, with the differences between them explained
- Any formal designation request and the authority’s response
- Reassessment records triggered by a change of claim, component or intended use
- The development and evidence plan traceable to the classification it assumes
COMMON INSPECTION FINDINGS
- A classification carried from one market to another with no jurisdiction-specific assessment
- Intended use in the marketing materials broader than the intended use the classification rests on
- A component or claim change with no reassessment of the pathway
- The rationale written after the evidence programme was designed rather than before it
- A combination product classified on its most convenient constituent rather than on mode of action
Classification follows the claim, not the object
The instinct is to classify by what the product physically is. Regulators classify by what it is intended to do and how it achieves that effect. The same physical article can be a device in one configuration and a drug in another, because the intended use and the mechanism by which the principal effect is achieved are different. A wound dressing that acts by covering is a device; the same dressing claiming to kill bacteria pharmacologically is arguing itself into a different regime.
In the US, 21 CFR Part 3 formalises this as the primary mode of action — the single mode of action expected to make the greatest contribution to the overall intended therapeutic effect — and assigns the FDA Center with primary jurisdiction accordingly. Where the answer is genuinely uncertain, Part 3 provides the Request for Designation: a formal route to a binding determination, which is far cheaper than discovering the Center after a development programme was built to the wrong evidence standard.
A classification is per market, not global
Classification does not travel. The EU classifies through Regulation (EU) 2017/745 for devices and Directive 2001/83/EC for medicinal products, and the boundary between them is drawn differently from the US line in several places — software, substance-based devices and borderline products most obviously. A product developed only to drug expectations because that is its home-market classification will arrive in a device market without design controls, a technical documentation file or a notified-body route.
For software the divergence is sharpest and the vocabulary matters: IMDRF N10 defines Software as a Medical Device and N12 sets out the risk-categorisation framework the device world reasons in, while the same functionality embedded in a drug-delivery system may be assessed as part of the drug product instead. Confirm classification jurisdiction by jurisdiction rather than extrapolating from the lead market — this is the single most common source of late-stage regulatory rework.
Combination products: classified once, regulated twice
A combination product is assigned to one Center under 21 CFR Part 3, but that assignment does not collapse the manufacturing obligations of its constituent parts. 21 CFR Part 4 sets out how the cGMP requirements combine: a streamlined approach in which a facility operating under one system demonstrates compliance with specified provisions of the other, rather than running two complete quality systems in parallel.
Sponsors routinely conflate the two Parts. Part 3 answers "who reviews this"; Part 4 answers "which manufacturing rules apply". A prefilled syringe assigned to a drug Center still has design-control and risk-management obligations flowing from its device constituent, and a quality system built as though the assignment settled everything will be short of exactly those elements at inspection.
Write the rationale, and re-open it when the product moves
The classification rationale is a document, not a conclusion someone reached. It should state the intended use, the mechanism by which the principal effect is achieved, the claims to be made, and the regulatory basis for the route chosen — because when a regulator challenges the route, the organisation has to be able to reconstruct why it was selected, often years later and with different people in the room.
It also has to be re-opened. Intended use, claims and formulation move during development, and each of those can move the classification with them. The practical control is a trigger in change control: a change to indication, mechanism, claim or route of administration prompts a classification re-assessment rather than being absorbed silently into the development plan.
SPEQ interpretation — classification is a quality-system input
Classification is usually owned by regulatory affairs and treated as a regulatory deliverable. In practice it is an input to the quality system, because it decides which regulation the quality system is built against, which records must exist, and what an inspector will cite. When it changes late, the remediation is not a document update — it is design history files that were never opened, risk management to a standard nobody applied, and validation scoped to the wrong regime.
The defensible arrangement is that the classification rationale is a controlled document with a named owner, referenced by the quality manual, and re-assessed on defined triggers. That costs almost nothing while the answer is stable, and it is the only thing that makes a late change survivable.
FREQUENTLY ASKED
What is the primary mode of action, and who decides it?
It is the single mode of action of a combination product expected to make the greatest contribution to the overall intended therapeutic effect, defined in 21 CFR Part 3. The sponsor reasons it first, but FDA decides — and a sponsor who wants that decision in advance can file a Request for Designation under §3.7 rather than proceeding on its own reading.
Can the same product be classified differently in the US and the EU?
Yes, and it frequently is. The US applies the primary mode of action test under 21 CFR Part 3; the EU draws the device/medicinal boundary through Regulation (EU) 2017/745 and Directive 2001/83/EC, with different treatment of substance-based devices and of software. Confirm classification per jurisdiction rather than extrapolating from the lead market.
Does a combination product need two quality systems?
No. 21 CFR Part 4 provides a streamlined approach: a facility already operating under one system demonstrates compliance with specified provisions of the other rather than duplicating both in full. What it does not permit is ignoring the constituent part that is not the assigned Center’s — design controls and risk management still apply to a device constituent inside a drug-led combination.
When should a classification be re-assessed?
On any change to intended use, claims, mechanism of action, route of administration or formulation. These are the inputs the classification was reasoned from, so a change to any of them can move the answer — and the change-control system is the only place that will reliably notice.