Container Closure Integrity Testing Now Needs Two Files

FDA's August 14, 2026 draft guidance on container closure systems defers extractables and leachables to a separate companion document, which means a fill-finish partner's single "packaging validated" letter no longer covers both halves of the standard. Wholesale buyers now have to request two separately dated qualification packages, and an incoming-acceptance file holding one of them is missing the half FDA declined to bundle.

Two documents landed four days apart this month, and read together they reset what a purchasing file on vialed material has to contain. The first was a recall notice. The second was a draft guidance that names, in its own text, a second guidance still to come. The gap between those two documents is the whole story for anyone requalifying a fill-finish subcontractor this quarter.^10,1

A four-day gap: recall then guidance

On August 10, 2026, Fresenius Kabi issued a nationwide recall of one lot of Tyenne (tocilizumab-aazg) Injection, 400 mg/20 mL vials, to the user level after an internal investigation found glass particles present in the product. The finding was not a label error or a strength deviation. It was a container closure failure discovered inside the manufacturer's own quality system, in a filled and released lot.¹⁰

The recalled lot had already been distributed nationwide to wholesalers, distributors, and direct customers. Fresenius Kabi is notifying distributors and customers by letter and arranging for return of the recalled product. Once a lot has moved into a distribution channel, the closure defect stops being a manufacturing problem and becomes a reconciliation problem for every intermediate holder: pulling stock, matching lot numbers against shipped orders, and issuing notifications downstream.¹⁰

On or about August 14, 2026, FDA announced the draft guidance "Container Closure Systems for Human Drugs and Biological Products" under docket FDA-2026-D-7957, setting out guiding principles for evaluating the quality of container closure systems used to package human drugs and biological products. The sequence is worth stating plainly: a live glass-particulate recall, then a draft framework describing the documentation that supports closure quality. Buyers get a priced example of the failure before they get the paperwork standard.^1,10

What the docket actually requires: FDA-2026-D-7957

The draft defines a container closure system as the sum of packaging components that together contain and protect a drug, including protection during storage and use, without altering its safety, identity, strength, quality, or purity. That definition is deliberately whole-system: the vial, the stopper, the seal, and their interaction are one evaluated unit, not three separately sourced parts each with its own certificate. The draft's evaluation principles are written against the assembled system, so component certificates collected from three vendors do not, on the draft's own terms, amount to an evaluation of the closure.¹

The draft also places the obligation squarely on the manufacturer to document and maintain the supporting data in its own files. For a distributor or reseller, that single allocation of responsibility governs everything downstream. The evidence is not generated at the warehouse and cannot be reconstructed there. It is obtained, in writing, from the party that filled and sealed the vial — or it does not exist in the purchasing file at all.¹

• Container closure integrity testing is recommended in lieu of sterility testing for sterile products, which makes CCIT data a release document rather than a supplemental one.
• The container closure system is to be evaluated in the stability protocol as appropriate, so integrity is a time-dependent claim tied to a study, not a one-time pass.
• Postapproval container closure system changes may trigger new stability studies, meaning a mid-contract switch of vial or stopper supplier restarts data generation.
• Comments go to Dockets Management in Rockville, MD, identified by the docket number, before the stated close date if they are to be considered before work on the final version begins.¹

The 2026 draft is not the first FDA document on this subject. FDA already maintains guidance on container and closure system integrity testing in lieu of sterility testing within a stability protocol, and a separate document addressing container closure system and component changes for glass vials and stoppers. The 2026 draft consolidates the evaluation principles; it does not retire the reason those earlier documents were written, which was that vial and stopper changes are the changes most likely to go undocumented.^3,4,6

Two packages, not one: closure integrity and extractables/leachables are separate deliverables

Here is the provision most summaries of the fda container closure guidance 2026 will skip. The draft explicitly references a forthcoming, separate companion document, "Extractables and Leachables for Pharmaceuticals and Biologics." Extractables and leachables evaluation and toxicological risk assessment are being handled in a distinct guidance rather than folded into the container closure systems draft. FDA split the packaging-quality question into two documents on purpose.¹

That split has an immediate procurement consequence. A one-page letter from a fill-finish partner asserting that packaging is validated cannot satisfy both halves of a standard that FDA itself declined to bundle. Integrity data and extractables and leachables documentation are now separately dated, separately generated document sets, and an incoming-acceptance file that contains one of them contains half of what a reviewer would look for.¹

The interim benchmark for the second half is already published. The ICH "Q3E Guideline for Extractables and Leachables," dated December 1, 2025, presents a framework for the assessment and control of extractables and leachables for pharmaceutical products, and stands as the closest currently available primary reference while FDA's companion draft is pending. A requalification request that names Q3E by title and date is asking for something that exists. A request that asks vaguely for "E&L data" invites whatever the subcontractor happens to have.¹

Where 211.22(a) and 211.192 put the investigation burden

The draft states that when a container closure integrity failure is detected, the manufacturer must investigate the root cause and, on the basis of a quality risk assessment, implement corrective actions — citing 21 CFR 211.22(a) and 211.192. Those two citations are the reason a closure deviation cannot be closed out with a retest and a rejected lot. The regulation asks what caused it and what changed as a result.¹

For a wholesale buyer, that converts a vague quality question into a document request with a definite answer. Both Fresenius Kabi actions on the record here began inside the manufacturer's quality system — one from an internal investigation, one from retained samples examined during a quality investigation — and each produced exactly the artifact the regulation contemplates. The distinguishing fact in a requalification file is whether the investigation record, the root-cause determination, and the risk-assessment-driven corrective action exist as retrievable documents attached to a specific date and lot.^1,9,10

Two Fresenius Kabi recalls, two years apart, same particulate finding

The Tyenne recall is not an isolated finding at that manufacturer. An earlier, separate Fresenius Kabi action recalled two lots of fluorouracil injection due to potential glass particulate identified in retained samples during a quality investigation. The second lot was included as a precaution because it had been produced in the same filling campaign. Different product, different molecule class, same defect mode, same investigative trigger.⁹

The fluorouracil recall establishes the scoping rule that matters most in a purchasing file: the campaign, not the batch record alone, defines exposure. A defect found in retained samples of one lot pulled a second lot that had shown nothing, because they shared a filling line and a time window. Any buyer holding stock from a shared campaign inherits that scope whether or not their specific lot ever failed a test.⁹

• Both recalls originated in the manufacturer's own quality investigation rather than a field complaint, which is the system working — and still costs the channel a return cycle.
• Both involved glass particulate, the failure mode that fill-and-finish qualification records and vial stopper compatibility data are meant to bound.
• The Tyenne lot had already been distributed nationwide to wholesalers, distributors, and direct customers before the recall was announced.
• The fluorouracil action shows that a precautionary lot can be pulled on campaign association alone, with no independent finding against it.^9,10

Both findings surfaced after release — one in a distributed lot, one in retained samples — which is where the practical purchasing question sits. A release certificate describes the lot as tested on the day it shipped; it says nothing about what a later retained-sample examination will turn up in the same filling campaign. That is why campaign identifiers, and not certificate numbers alone, belong on incoming documentation.^9,10

Writing the two-package demand into the next quote

The comment period on docket FDA-2026-D-7957 is open, which means the requalification language written now will be in place before the final version issues. The practical move is to stop asking a fill-finish partner whether packaging is qualified and start naming the two document sets, each with its own date, in the incoming-acceptance clause of the requalification request.¹

1. Container closure integrity test data for the assembled vial-and-stopper system, identified by method and by the lot or campaign it supports, with the CCIT-in-lieu-of-sterility position stated explicitly.
2. The stability protocol section showing how the container closure system is evaluated over the claimed shelf life, not merely at release.
3. A written statement of any postapproval or in-contract change to vial, stopper, or seal component, with the stability work that change triggered.
4. Extractables and leachables documentation referenced to the ICH Q3E Guideline for Extractables and Leachables dated December 1, 2025, including the toxicological risk assessment, supplied as a separate dated package.
5. Closed investigation records for any container closure integrity failure, showing the root-cause determination and the corrective action taken on the basis of a quality risk assessment, per 21 CFR 211.22(a) and 211.192.
6. Fill and finish qualification records for the line and campaign, sufficient to scope which distributed lots a future particulate finding would reach.¹

Items one through three are the closure package. Items four onward are the extractables and leachables package plus the deviation history that makes both auditable. A subcontractor that returns a single bundled packaging file has answered the question FDA asked before August 14, 2026, and the gap is a documentation gap rather than a pricing one — it closes when the second dated package arrives, and not before.¹

The recall four days earlier priced the alternative. One lot, distributed nationwide, notified by letter, returned. Whoever held that stock absorbed the reconciliation regardless of who filled the vial. Two document sets requested in advance cost a week of correspondence; a channel-wide return does not.^10,1