How FDA decides which center reviews your product

FDA is organized around product type — drugs at CDER, devices at CDRH, biologics at CBER — and most products file into one of those lanes without anyone having to think about it. A growing number don’t: a prefilled autoinjector is a drug and a delivery device in one unit; a drug-eluting stent is a device whose performance depends on a coating that is itself a drug. Before a sponsor can pick a submission type, it has to know which center is actually going to review the thing, and for a real combination product that answer is not always obvious from the outside.

Why a product needs an assignment at all

FDA’s three product centers grew up around three different statutory schemes — drugs, devices, and biologics each have their own approval standards, their own review staff, and their own institutional expertise. That structure works cleanly for a tablet or a surgical clamp. It works less cleanly for a product built from more than one of those categories at once, which is exactly what Congress anticipated when it added Section 503(g) to the Federal Food, Drug, and Cosmetic Act (21 U.S.C. § 353(g)): a statutory instruction that FDA assign every combination product to a lead center rather than leave sponsors to guess or, worse, to shop for the center they’d prefer. The Office of Combination Products coordinates that assignment and resolves disputes between centers, but the lead center still runs the actual review — the same reviewers, the same submission mechanics, as any other product in that center’s queue.

How the assignment actually gets made

The operative test is primary mode of action: of all the ways the product could plausibly achieve its intended effect, which single mode is expected to provide the most important therapeutic action. For a drug-eluting stent, that means asking whether the stent’s job is mainly mechanical — holding a vessel open — with the drug coating as a secondary refinement, or whether the drug’s biological effect is doing the real work and the stent is mainly a delivery scaffold. Two products that look similar from the outside can land with different centers if their intended use, their clinical data, and their own PMOA analysis point different directions — which is why a competitor’s past assignment is a data point, not a precedent you can rely on for your own product.

Getting a binding answer before you need one

A sponsor doesn’t have to wait for FDA to raise a jurisdiction question during review. The Request for Designation process at 21 CFR § 3.7 lets a sponsor describe the product, its components, and its own PMOA analysis, and ask FDA for a formal determination in advance — the kind of answer that’s binding for that product rather than an informal read from a pre-submission conversation. The value of filing early is mostly about sequencing: a jurisdiction correction that arrives after a pivotal study is designed around the wrong center’s evidence expectations is a far more expensive problem than the extra weeks an RFD costs up front. It’s the same logic that makes an early predicate search worth doing before building a 510(k) around the wrong reference device — get the foundational classification question answered while it’s still cheap to be wrong about it.

Getting the assignment right is foundational the same way choosing a submission pathway is: it’s a judgment call made early that shapes everything downstream, similar to how a De Novo request commits a novel device to a specific regulatory framework before a single unit reaches a patient. Read the assignment as infrastructure, not paperwork — it determines which set of rules your evidence has to satisfy for the rest of the product’s life.