CCITContainer Closure Integrity Testing (USP <1207>)

Before 2016, container-closure integrity was typically tested by dye-ingress (immerse the package in coloured dye under vacuum, look for ingress) or microbial-challenge (immerse in a microbial broth, incubate, test for growth). Both are probabilistic: they rely on chance ingress through any defect during the test window, and both have notoriously low sensitivity to small defects (~10 µm and above). The 2016 revision of USP <1207> recategorised these as 'probabilistic' methods and elevated 'deterministic' methods — vacuum decay, mass extraction, pressure decay, high-voltage leak detection (HVLD), helium leak — as preferred for both development qualification and routine release.

• Vacuum decay — package under controlled vacuum; pressure rise over time indicates leakage. Sensitive to ~5 µm orifice equivalent. Non-destructive, widely automatable, the most common 100 % inspection method for sterile vials.
• Pressure decay — package pressurised; decay over time indicates leakage. Used for flexible packs (bags, pouches).
• Mass extraction (Mocon) — trace gas extraction from sealed package under vacuum. High sensitivity (sub-µm), non-destructive.
• High-voltage leak detection (HVLD) — applies high-voltage potential across the package; defects allow current flow detected by electrode pairs. Excellent for liquid-filled vials and ampoules. Concerns about effect on sensitive biologics resolved with modern low-current HVLD.
• Helium leak (mass spectrometry) — package backfilled or surrounded with He tracer; He leak rate measured by mass spectrometer. Highest sensitivity (10⁻⁹ to 10⁻¹¹ mbar·L/s), typically destructive; reserved for development and qualification.
• Headspace gas analysis (laser-based) — measures O₂, CO₂, moisture inside sealed package non-destructively; useful for both integrity and gas-environment verification.

The MALL is the maximum leak rate at which the package still functions as a sterile/contamination-free barrier across the shelf life. It depends on the product (liquid vs solid, sterile vs non-sterile), the storage conditions, the shelf-life duration, and the patient risk profile. USP <1207.1> describes how to derive a product-specific MALL using mechanistic models or empirical data (microbial ingress, gas exchange, moisture exchange). Method qualification then demonstrates that the chosen CCIT method can reliably detect leakage at or below the MALL — typically by challenging the method with reference standards (micropipettes, laser-drilled defects) at known leak rates above and below the MALL.

Qualification deliverables: limit of detection (LOD), probability of detection (POD) curve, robustness to package variability, false-positive rate, false-negative rate. Routine CCIT then runs against the qualified MALL with an AQL-based sampling plan (or 100 % inspection for high-risk products like cell/gene therapy).

The FDA's 2008 guidance permits CCIT to replace the USP <71> sterility test at stability time-points if the CCIT method is validated to demonstrate integrity is maintained, the integrity-loss probability is appropriately conservative, and the sterility-test result history supports the substitution. This saves the destructive, slow and notoriously variable USP <71> test at each stability pull and is now standard practice for sterile drug products in modern stability programmes.

1. Still relying on dye-ingress as the primary integrity method post-2016 with no <1207> gap analysis on file.
2. MALL not product-specific — defaulted to a generic value with no mechanistic or empirical justification.
3. Method qualification done once at install and never re-qualified after package or filling-line changes.
4. POD curve not generated — only a binary 'detected / not detected' record at a single leak rate.
5. Routine CCIT excursion handled as a process aside rather than as a sterility-assurance investigation with quarantine implications.
6. Stability sterility-test replacement claimed without the validation pack to support it.
7. HVLD applied to a sensitive biologic without product-impact assessment (oxidation, aggregation).

• Method file holds the qualified CCIT technique per package family — MALL, qualification report, POD curve, reference-standard schedule, frequency of re-qualification.
• Filling-line CCIT instrument streams every inspection record to the batch record automatically — units passed/failed/rejected, time-stamped, e-signed by the system.
• Routine excursion (any unit failing CCIT) triggers a controlled-deviation workflow with batch-quarantine logic — the product is not released until the investigation closes.
• Stability protocol references the CCIT-in-lieu-of-sterility validation; the stability pull schedule shows CCIT instead of <71> where qualified.
• Change control: any package change (vial glass, stopper, filling-line, capping torque setting) auto-routes to the CCIT owner for re-qualification assessment.