Identification and Comparison of Extractables in Drug Container Closure Systems

Significance of the Topic

The release of extractable compounds from pharmaceutical container closure systems can compromise drug safety and efficacy. Regulatory guidelines recommend a risk-based evaluation of semivolatile and nonvolatile extractables to ensure container materials do not introduce toxicologically relevant impurities into drug products.

Objectives and Study Overview

This study compared semivolatile extractables from four different drug container closures sourced from multiple manufacturers. The main goals were to (1) develop comprehensive extractable profiles, (2) compare container chemistries, and (3) demonstrate the enhanced identification power of combined electron ionization (EI) and chemical ionization (CI) on a high-resolution GC/Q-TOF platform.

Methodology

Empty containers (two low-density polyethylene, two polyethylene) were extracted with HPLC-grade n-hexane under sonication for 1.5 hours. Extracts were analyzed by GC/Q-TOF in splitless mode using both EI (70 eV) and CI (methane reagent gas) ionization. Automated deconvolution and library matching against NIST 14 in EI mode identified high-confidence hits (score ≥ 80). Low-score EI hits (50–79) were exported to a custom library to mine CI accurate-mass data. Semiquantitative concentrations were estimated using triphenyl phosphate as an internal standard.

Used Instrumentation

• Agilent 7890A GC with multimode inlet and DB-5ms column
• Agilent 7200 GC/Q-TOF mass spectrometer
• Agilent MassHunter Acquisition, Qualitative Analysis, Unknowns Analysis and Mass Profiler Professional software

Main Results and Discussion

Approximately 170 extractable compounds were detected per container. The union of EI and CI data increased identifications by ~14 % compared to EI alone. Venn analysis revealed 22 compounds common to all containers and container-specific profiles ranging from 141 to 207 entities. Container 4 exhibited the highest number of extractables, including one high-risk (Cramer Class III) compound at > 20 µg/mL, making it less favorable for sensitive formulations.

Benefits and Practical Applications

The combined EI/CI workflow on a high-resolution GC/Q-TOF maximizes coverage of semivolatiles, including labile species often missed by EI alone. Automated data mining and custom libraries enable rapid, reproducible extractables profiling, supporting risk assessments for packaging selection in pharmaceutical development.

Future Trends and Opportunities

Advances in ionization techniques, integration of in silico toxicity prediction, enhanced chemometric visualization, and machine learning-driven compound annotation will further streamline extractables and leachables studies, reducing development timelines and improving container qualification.

Conclusion

The study demonstrates that high-resolution GC/Q-TOF with orthogonal EI and CI acquisition significantly enhances semivolatile extractable identification. Chemometric tools visualize container differences, guiding material selection based on extractable burden and toxicological risk.

Reference

• Mire-Sluis A. et al. Extractables and Leachables: Challenges and Strategies in Biopharmaceutical Development. BioProcess Int. 2011, Feb.
• U.S. FDA. Guidance for Industry: Container Closure Systems for Packaging Human Drugs and Biologics. May 1999.
• Jenke D. Development and Justification of a Risk Evaluation Matrix to Guide Chemical Testing of Plastic Components. PDA J. Pharm. Sci. Technol. 2015, 69, 677–712.
• Jenke D. Creating a Holistic Extractables and Leachables Program for Biotechnology Products. PDA J. Pharm. Sci. Technol. 2015, 69, 590–619.
• IdeaConsult Ltd. Toxtree v.2.6.13. http://toxtree.sourceforge.net/.
• Lateef SS et al. Differential Analysis in Extractables and Leachables Study Using Agilent 7200 GC/Q-TOF and Data Mining Software. Agilent Tech. App. Note 2016, 5991-6688EN.
• Jenke D. et al. Utilization of Internal Standard Response Factors to Estimate Leached Compound Concentrations. J. Chromatogr. Sci. 2012, 50, 206–212.