Container-Closure Integrity for Prefilled Syringes: FDA Review Expectations in NDAs and BLAs Since 2018

Container-closure integrity for prefilled syringes: what FDA reviewers asked in NDA and BLA reviews since 2018

Since 2018, FDA product quality and device reviewers have approached prefilled syringe (PFS) container-closure integrity (CCI) with a consistent set of expectations, most visible in BLA reviews for biosimilars and new biologics. Reviewers seldom required a particular test technology. They focused on three questions. Is the method validated with syringe-relevant positive controls at a sensitivity tied to microbial ingress? Is CCI shown after device assembly and on stability through expiry? Does integrity survive real shipping, especially air transport, where pressure changes can move the plunger stopper across the sterile boundary?

Key takeaways

• No preferred technology. In the SIMLANDI (adalimumab-ryvk) review, FDA stated it had no preferred CCI method. It named dye testing (probabilistic) and helium leak testing (deterministic) as common approaches for autoinjector systems, provided the syringe can be evaluated in a way that prevents dye leakage artifacts ⁷⁴.
• Sensitivity anchored to about 20 µm. Across many BLAs, FDA asked for a CCI method validated to detect breaches that could allow microbial ingress, generally specified as ≤20 µm ^{32242840414344}.
• Positive controls must be syringes, not vials. FDA judged CCI validations inadequate when breach controls were vials rather than prefilled syringes. This happened for ZIEXTENZO ⁴⁸⁴⁹, and it contributed to a Complete Response for pegfilgrastim-unne (LUBT004) ¹⁰¹.
• CCI in lieu of sterility on stability. FDA's standard request was annual CCI testing in place of sterility testing on stability samples through expiry ^32244454. FDA did not accept a sponsor's proposal to rely on sterility testing instead of CCI ⁷⁴.
• Shipping validation must address plunger movement. FDA repeatedly asked sponsors to show that air-transport pressure changes do not move the plunger enough to breach the sterile boundary or affect sterility ^3322314240. FDA also refused to let real-time shipping results be deferred to a postmarketing commitment ⁴⁵.

The standard FDA request for PFS CCI

Many BLA reviews contain a near-identical block of product quality requests for PFS and PFS-in-autoinjector presentations. The fullest versions appear in the reviews for WYOST/JUBBONTI (denosumab-bbdz) ³², LYUMJEV (insulin lispro-aabc) ²⁴, PYZCHIVA (ustekinumab-ttwe) ⁴⁰, OMLYCLO (omalizumab-igec) ⁴¹, NYVEPRIA (pegfilgrastim-apgf) ⁴³, AHZANTIVE (aflibercept-mrbb) ⁴⁴ and IMJUDO (tremelimumab-actl) ²⁸. The common elements are:

The same framework applied to other primary containers used with pen injectors. For BOMYNTRA/CONEXXENCE (denosumab-bnht), FDA asked for initial and stability CCI after assembly of prefilled cartridges and pen devices, at ≤20 µm sensitivity, with annual CCI replacing sterility testing through expiry ⁵⁴. For ALHEMO (concizumab-mtci), FDA asked whether air-pressure changes could move an unsecured cartridge plunger and breach the sterile boundary ²⁰.

Deterministic versus probabilistic methods

The reviews contain little explicit debate over deterministic versus probabilistic methods. FDA's position, where stated, is performance-based. In the SIMLANDI review, FDA said there was no preferred CCI method and pointed to both dye testing and helium leak testing as common approaches for autoinjector systems ⁷⁴. Reviewers put more weight on how a method was validated than on which category it belonged to.

Applicants used a mix of technologies:

• Dye ingress (probabilistic). SKYRIZI (risankizumab-rzaa) used methylene-blue dye ingress under vacuum: five vacuum/ambient-pressure cycles with positive and negative controls, and visual assessment of the contents ⁴⁷. UDENYCA (pegfilgrastim-cbqv) specified "No Dye Ingress" as its CCIT acceptance criterion for the assembled PFS with needle-safety guard ⁴. Among NDAs, reviewers assessed dye-ingress CCI for Gvoke PFS (glucagon) ⁹³. For Vyleesi (bremelanotide), reviewers concluded the syringe was suitable to maintain sterility over shelf life ¹¹⁴¹¹⁷.
• Dye and microbial ingress in parallel. NIVESTYM (filgrastim-aafi) ran both microbial-ingress and UV-Vis dye-ingress testing against the same laser-drilled positive controls ³⁹. SKYRIZI's validation compared dye-ingress and microbial-ingress detection limits and concluded that dye ingress was equally sensitive ⁴⁷.
• Deterministic methods used alongside or in place of dye. SKYRIZI reported helium leak testing of all relevant PFS interfaces and stated that CCI met a maximum allowable leakage limit under USP <1207.1> ⁵. PALYNZIQ (pegvaliase-pqpz) reported vacuum-decay CCI results for its 2.5-mg and 20-mg PFS, with no leaks at 0, 3 and 6.75 months of accelerated aging. The reviewer did not comment on the choice of method ¹⁰⁸. AJOVY (fremanezumab-vfrm) reported a voltage-based CCI test ("No leak detected" at 1.0 to 4.0 VDC, n=299), alongside a separate liquid-leakage pressure test ⁷¹⁷⁷.

In practice, a probabilistic dye-ingress method was acceptable when validated against syringe-format defects of relevant size. A deterministic method did not exempt a sponsor from showing that it covered the relevant sealing interfaces.

Method sensitivity and positive controls

Sensitivity was the most common reason FDA found a CCI package deficient.

• ZIEXTENZO (pegfilgrastim-bmez). FDA found the dye-ingress validation inadequate because the positive control was a vial, not a prefilled syringe. FDA asked for a new qualification study with compromised PFS positive controls, sensitive to breaches of less than 20 µm. It also asked that routine dye-ingress testing include at least one compromised PFS as a system-suitability control ⁴⁸⁴⁹.
• LUBT004 (pegfilgrastim-unne). In a Complete Response, FDA found the PFS CCI method inadequately validated. Breach controls used only vials, so the applicant had not shown that the detection limit could be evaluated in the PFS or that breached-vial and breached-syringe results were comparable. FDA also found the repeatability acceptance criterion for positive controls insufficiently robust, and asked for repeat validation ¹⁰¹.
• NIVESTYM (filgrastim-aafi). Positive controls were syringes with 5-µm effective-diameter laser-drilled holes. Both microbial-ingress and dye-ingress methods detected them, and intact samples passed. The reviewer first flagged missing test reports and no sample-size justification. After an information request, the reviewer found design verification and validation sufficient ³⁹.
• SKYRIZI (risankizumab-rzaa). FDA asked how many syringes were tested per lot, whether that number was statistically relevant, and whether any deviations occurred. The sponsor reported 20 samples per lot for stability CCIT, based on the USP <71> sterility-test sample number, with no deviations. FDA found this adequate ³⁴.
• ETICOVO (etanercept-ykro). FDA asked for method-validation information showing enough sensitivity to detect breaches that could permit microbial ingress ⁵⁶.
• Brekiya (dihydroergotamine mesylate). The reviewer noted that the integrity specification was higher than the level set in the original validation studies ⁹⁸. This is a reminder that a routine limit has to be consistent with the validated detection capability.

Integrity across shipping and distribution

Shipping was the area where FDA asked the most, and where it most often turned down sponsor proposals to defer data.

Plunger movement under reduced pressure. The SKYRIZI review explains the concern. Reduced external air pressure can move the stopper outward. If the stopper touches a nonsterile syringe surface and then returns into the sterile area, the product could be contaminated ³⁵. SKYRIZI's four batches passed CCIT after air shipment, ground transport and drop/shake/vibration. Maximum stopper movement was 0.23 to 0.40 mm, and FDA concluded that movement of ≤0.4 mm met the acceptance criterion without moving the stopper beyond the sterile area ³⁰. FDA also accepted post-shipping methylene-blue testing plus minimal stopper movement as justification for not running post-shipping dose-accuracy testing ³⁷.

Recurring shipping requests. FDA asked for shipping validation covering air pressure, plunger movement and sterile-boundary breaches in the reviews for ABRILADA ³³, IDACIO ²², OMVOH ³¹, TEZSPIRE ⁴², SIMLANDI ³⁸, LYUMJEV ²⁴, WYOST/JUBBONTI ³², PYZCHIVA ⁴⁰, OMLYCLO ⁴¹ and IMJUDO ²⁸. For ABRILADA, FDA stated explicitly that syringe shipping validation should include CCI testing ³³.

Real-time versus simulated shipping. For STIMUFEND (pegfilgrastim-fpgk), FDA rejected the proposal to submit real-time shipping results as a postmarketing commitment. FDA said shipping can affect PFS CCI and sterility, that simulated studies alone were insufficient, and that real-time results had to be in the BLA at submission ⁴⁵. For NYPOZI (filgrastim-txid), FDA noted that air-transport pressure changes and g-forces may affect CCI more than ground transport. Leaving air-shipping validation out of the initial filing would be unacceptable if the product would ship by air, unless earlier studies adequately represented those stresses ⁴⁶.

Postmarketing follow-up. IMULDOSA (ustekinumab-srlf) was recommended for approval with a postmarketing commitment to run a worst-case headspace study. The study had to confirm that stopper-placement-height specifications would not allow stopper movement beyond the sterile boundary under low pressure ¹⁰⁰. For OTULFI (ustekinumab-aauz), FDA identified a need to confirm package integrity of the PFS during real-world commercial shipping ⁹⁹.

CCI as evidence of no product loss. For Fulvestrant Injection (NDA 210326), FDA asked for post-shipping verification of performance. The sponsor submitted an ASTM D4169-16 simulated-shipping study. Extractable volume failed because of a fill-volume error, but CCIT passed. FDA agreed that maintained CCI showed shipping would not cause product loss ⁶⁷. For Lanreotide Acetate Injection (NDA 215395), the applicant committed to repeat CCIT after ASTM D4169-16 simulated transportation for a modified safety needle ⁶⁸.

Integrity over storage and shelf life

CCI in the stability program. FDA's usual expectation was CCI testing on stability, replacing sterility testing annually and at expiry ⁵⁶⁴³⁴⁴. For SIMLANDI, FDA did not accept sterility testing in place of CCI and required CCI annually and at expiry ⁷⁴. SKYRIZI's program ran CCI testing on stability rather than at release, in addition to sterility monitoring at 2 to 8 °C. The program covered accelerated and long-term studies, transport simulation, and confirmation of CCI after recommended storage and shipment ⁵³⁴.

End-of-shelf-life testing. For PERSERIS KIT (risperidone), FDA required a revised post-approval stability protocol with sterility and endotoxin at 0 and 18 months and CCIT at 18 months (end of shelf life). The specification also had to identify the CCIT method, acceptance criteria and method reference ⁵². The applicant clarified that CCIT applies to each sealed pouch containing a filled syringe, and FDA found the arrangement adequate ⁵³⁵¹. For UDENYCA, the device requirement that the PFS maintain CCI through end of shelf life was reported as passed, and CCI was part of the device-functionality stability program ¹³³⁶.

NDA microbiology conclusions. In NDA reviews, CCI was usually handled as one element of the sterility-assurance package, with less method detail in the review text. Reviewers found CCI validation and controls adequate for Mounjaro (tirzepatide) ¹²¹, Zegalogue (dasiglucagon) ¹¹³¹²⁶, the midazolam autoinjector (NDA 216359) ¹¹² and Leqvio (inclisiran) ¹¹⁶¹²⁸. For Epinephrine Injection in the Abboject syringe system, reviewers concluded that CCI validation under worst-case stopper conditions was adequate and that stability data supported sterility assurance through the 15-month shelf life ¹¹⁰¹²⁰.

Device interfaces: needle shields, stoppers and Luer connections

Reviewers treated the PFS sealing interfaces as distinct CCI barrier points. For SKYRIZI, FDA asked which device components were supplied sterile. The PFS/rigid needle shield and PFS/plunger stopper interfaces were identified as the integrity barriers, and the sponsor showed that assembly into the needle-stick protection device breached neither ¹⁵. The SKYRIZI CCI assessment covered potential leaks at the needle shield/needle, needle/glue/syringe cone, and plunger/barrel interfaces ⁵.

For Luer-lock syringes, FDA extended the review to connector performance. For ENFLONSIA (clesrovimab-cfor), FDA noted that baseline and stability CCI data had been provided, but asked for baseline, stability and preconditioned data on Luer-lock performance under ISO 80369. The sponsor supplied leakage, stress-cracking and separation testing plus simulated shipping, drop and vibration reports, and FDA found the data acceptable ⁸⁸. For EYDENZELT (aflibercept-boav), FDA accepted the sponsor's argument that a manufacturing-site change did not affect Luer performance, supported by CCI testing on process-validation and stability lots ⁸⁹.

Implications for PFS submissions

The pattern across these reviews points to a few practical priorities for sponsors filing PFS NDAs and BLAs:

1. Choose a CCI technology based on performance. FDA has accepted dye ingress, helium leak, vacuum decay and electrical methods, and has said it has no preferred method ^7447510871.
2. Validate with compromised prefilled syringes as positive controls, at a sensitivity justified against microbial ingress (≤20 µm is FDA's usual benchmark). Include a syringe-format system-suitability control in routine testing ^4849101.
3. Justify the number of samples per lot and make routine limits consistent with the validated detection capability ³⁴⁹⁸.
4. Show CCI after final device assembly, and plan annual and expiry CCI testing in lieu of sterility on stability ²⁴³²⁷⁴.
5. Include real-time shipping data, with air transport where relevant, in the original submission. Quantify stopper movement against the sterile boundary rather than relying on simulation alone or a postmarketing commitment ^454630100.