How to read a PMA Refuse-to-File decision

A PMA can fail before FDA ever reaches the science. Not because the device doesn’t work — because the application itself isn’t built well enough for a reviewer to start. That’s what a Refuse-to-File decision is, and reading one clearly is a different skill from reading a deficiency in the data.

What FDA is actually screening for

Filing review happens under 21 CFR 814.42, and it’s worth holding onto what the regulation is actually asking: not whether the device is safe and effective, but whether the application contains enough of what’s required to let a reviewer start evaluating that question at all. FDA has 45 days from receipt to notify the applicant whether the PMA has been filed or refused, and a refusal at this stage means the file itself is missing something structural — a required administrative element, an incomplete section, data that can’t yet be evaluated because a piece it depends on isn’t there. This is upstream of the Summary of Safety and Effectiveness Data that eventually documents an approval; there’s no SSED to argue about yet, because the file hasn’t reached the point where FDA is evaluating the evidence. Anyone who has worked 510(k) submissions will recognize the shape of this from the Refuse-to-Accept checklist, but the two aren’t the same mechanism wearing different names. RTA runs on a published checklist against 21 CFR 807, with its own review window; RTF runs on 814’s filing requirements, with its own 45-day clock and its own, less standardized set of grounds. Assuming PMA filing works like 510(k) filing because the underlying instinct — screen at the door before spending review resources — is the same, is a mistake worth naming early.

Reading the letter like a checklist, not a rejection

An RTF letter names specific deficiencies, tied to specific filing requirements, and the most useful way to read it is as a literal list of items to close rather than a verdict to argue with. The common failure mode isn’t missing a deficiency — it’s treating the letter as if it were a scientific objection to the device, and responding by re-explaining the original position instead of supplying whatever was actually missing. Unlike a deficiency raised later in substantive review, where FDA has looked at your evidence and found it insufficient, a filing-stage refusal is almost always fixable by adding the specific thing that’s absent — a missing certification, an incomplete section, a piece of data the rest of the file depends on. Before resubmitting, go through the letter item by item and confirm each one is actually closed, not just addressed in spirit. A resubmission that restates your original reasoning without adding the cited element reads, to the reviewer checking it against the same checklist, exactly like the first one did.

None of this changes once a PMA clears filing and moves into substantive review, where the questions get harder and the answers stop being about completeness. What changes is the nature of what you’re responding to — a filing refusal is a list of what’s missing, and closing that list is a more mechanical problem than the one you’ll be solving a few months later.