Extractable Study of Container Closure Pack Used to Store Oral Drug Product Using Dynamic Headspace Technique

Importance of the Topic

This application note addresses extractables and leachables (E&L) assessment for a container-closure pack (CCP) intended to store an oral Penicillin V formulation. E&L studies are regulatory expectations for drug product safety assessment because chemical species migrating from packaging can be toxic or impact drug performance. The study demonstrates how appropriate extraction design and targeted analytical workflows (dynamic headspace GC–MS with trap concentration and simultaneous SCAN/MRM acquisition) support sensitive identification and quantitation of potential migrants and inform risk assessment relative to calculated Analytical Evaluation Threshold (AET).

Objectives and Study Overview

• Calculate the AET for the Penicillin V oral product and define acceptance criteria for leachables.
• Compare extraction approaches that simulate worst-case and realistic use scenarios: acidic aqueous incubation (mimicking the slightly acidic drug matrix) and solvent reflux with polar (ethanol) and non-polar (hexane) solvents.
• Apply a dynamic headspace (HS-Trap) GC–MS/MS method combined with simultaneous SCAN/MRM acquisition (ASSP) to maximize qualitative identification and trace quantitation of extractables.
• Semi-quantify unknown extractables using calibration data from 15 representative standards commonly observed in E&L studies.

Calculation of AET

The Safety Concern Threshold (SCT) selected was 1.5 µg/day (PQRI guidance for relevant drug classes). Using product dose information (100 mL pack, 20 mL maximum daily dose equivalent, four 5 mL doses), the AET was calculated at 3.75 µg/pack and converted to a concentration of 0.0375 ppm for the liquid dosage form. This AET was used as the screening threshold for reporting and toxicological consideration.

Experimental Design and Sample Preparation

• Incubation (acidic aqueous): Two bottle caps incubated with 50 mL formic acid media (pH 2) at 85 °C for 72 hours. Extracted with ethyl acetate and concentrated for analysis.
• Reflux (polar and non-polar): Single cap cut into pieces and refluxed with 50 mL ethanol (polar) or hexane (non-polar) at ~80 °C for 2 hours. Aliquots were concentrated, reconstituted in methanol and analyzed.
• Headspace VOC screening: Direct static headspace (1 mL loop) and dynamic headspace trap modes were compared on the same sample to evaluate sensitivity and the number of detected compounds.
• Semi-quantitation: Fifteen standards prepared (0.1–2 ppm linear range; typical calibration levels 0.1–2 ppm or 0.1–2.5 ppm) and analyzed in combined SCAN/MRM acquisition. Unknowns were identified by library match and quantified by applying averaged slope/intercept from standards (MRM-based when available; SCAN used for non-target semiquantitation).

Used Instrumentation

• Shimadzu GCMS-TQ8050 NX triple quadrupole GC–MS with HS-20 NX headspace autosampler (Trap mode for dynamic headspace).
• Column: SH-I-5Sil MS, 0.25 mm I.D. × 30 m, 0.25 µm film.
• Ionization: Electron impact (EI).
• Acquisition: Simultaneous SCAN and MRM (ASSP) enabling compound identification and sensitive quantitation in a single run.
• Dynamic headspace trap conditions (key parameters): trap cooling ~ −20 °C, trap desorb 280 °C, multi-injection count up to 3 in this work (platform supports up to 10), trap equilibrium/desorb sequences, oven program from 40 °C to 280 °C; carrier helium; sample shaking and extended injection times to concentrate volatiles.

Analytical Method Performance

• Linearity and sensitivity: All 15 target standards demonstrated acceptable linearity (reported r2 values generally ≥ 0.99 in MRM mode) and signal-to-noise ratios at LOQ meeting acceptance (S/N ≥ 10 in MRM measurements).
• SCAN mode: Useful for non-target screening and library-based identification but showed higher baseline noise and lower sensitivity versus MRM; used for conservative semiquantitation of unknowns.
• Dynamic headspace vs static: Trap (dynamic) mode substantially increased the number and intensity of detected extractables compared with static headspace (1 mL loop). Concentration on the cold trap plus multiple extractions enabled trace-level detection not achievable in loop-mode static HS.

Main Results and Discussion

• Acidic aqueous incubation (pH 2, 85 °C, 72 h): Only a few volatiles were detected, predominantly low-level siloxanes and minor organics. All detected species were below the calculated AET (0.0375 ppm), indicating negligible risk for migration into the aqueous Penicillin V formulation under realistic conditions.
• Reflux with ethanol (polar): Produced more peaks than aqueous incubation but generally lower overall extractable load compared with hexane reflux.
• Reflux with hexane (non-polar, worst-case): Produced a large number of hydrocarbon-type extractables (alkanes, branched decanes/tridecanes), siloxanes, phenolic antioxidants (e.g., 2,4-di-tert-butylphenol), esters and other high‑boiling migrants. Several analytes were detected above the AET when using hexane reflux — illustrating that aggressive non-polar solvent extraction uncovers many potential migrants that would not necessarily appear under real storage/usage conditions for an aqueous oral product.
• Identification challenges and approach: Unknown peaks were identified via NIST library matching and semiquantified using averaged calibration parameters from the target compounds (MRM-based calibration where possible). SCAN spectra supported tentative IDs but have higher uncertainty; orthogonal confirmation would be recommended for any leachable flagged above AET.

Practical Benefits and Applications

• Dynamic HS-Trap coupled to a fast triple quadrupole with ASSP enables a combined screening/quantitation workflow in a single injection, improving laboratory throughput for E&L screening campaigns.
• Trap enrichment and multi-extraction capability markedly enhances sensitivity for volatile and semi-volatile extractables compared with static headspace loop injections, supporting detection at sub-AET levels without solvent exchange or large-volume sample handling.
• The combined SCAN/MRM acquisition allows qualitative identification (library matching in SCAN) and highly selective, low-noise quantitation (MRM) concurrently — a practical advantage when both non-target discovery and reliable quantitation are required for regulatory submissions.

Limitations and Considerations

• Solvent reflux with strong non-polar solvents provides a worst‑case extractable profile but can overestimate realistic leachable risk for aqueous drug products; interpretation must consider chemical compatibility and actual product contact chemistry.
• Semi-quantitation of unknowns using averaged calibration factors is approximate; confirmatory analysis with authentic standards or orthogonal techniques (e.g., accurate-mass MS, GC–MS with high-resolution) is recommended when levels approach or exceed AET.
• SCAN-based identifications can suffer elevated noise and matrix interferences; robust spectral library scores and retention index corroboration improve confidence.

Future Trends and Opportunities

• Broader adoption of dynamic headspace trap systems for routine E&L screening due to superior sensitivity and lower sample preparation burden for volatile/semi-volatile migrants.
• Integration of high-resolution mass spectrometry (HRMS) with dynamic HS for improved non-target identification and elemental composition determination of unknown leachables.
• Advanced data processing and machine-learning workflows to prioritize true leachables from background peaks and to automate tentative ID/flagging relative to toxicological thresholds.
• Greater harmonization between conservative worst‑case extraction approaches and more realistic matrix‑matched extractions to reduce unnecessary follow-up while maintaining patient safety.
• Improved standard libraries and retention-index databases for polymer additives, plasticizers, antioxidants and siloxanes to accelerate ID in complex extractable profiles.

Conclusion

The study shows that dynamic headspace trapping combined with a high-speed triple-quadrupole GC–MS with simultaneous SCAN/MRM acquisition is an effective screening and semi-quantitative platform for E&L assessment of a CCP intended for an aqueous, slightly acidic oral Penicillin V product. Realistic aqueous incubation produced extractables below the calculated AET, indicating the CCP is acceptable for the intended use. Aggressive hexane reflux exposed many additional hydrophobic extractables at concentrations above the AET, highlighting the distinction between worst-case solvent-based extraction and realistic migration under product-relevant conditions. For any analyte detected near or above the AET in practical extracts, orthogonal confirmation and source-tracing are recommended.

References

• USP General Chapter <1663> Assessment of Extractables Associated with Pharmaceutical Packaging/Delivery Systems.
• Product Quality Research Institute. Safety Thresholds and Best Demonstrated Practices for Extractables and Leachables in Parenteral Drug Products (Intravenous, Subcutaneous, and Intramuscular). PQRI recommendation document.