Establishment of optimized methodology by GC/MS for hazard evaluation of content-containing interaction during pharmaceutical development
3 October 2019
TOUSSAINT Balthazar1,2, TOLOSANA Bastien1, LUTON Anaïs1, ROSA Frédéric1, LEGRAND Pauline1,3, BOUDY Vincent1,3, GOULAY-DUFAŸ Sophie1 1 : Département de Recherche et Développement Pharmaceutique, Agence Générale des Equipements et Produits de Santé, 7 rue du Fer à moulin, 75005 Paris, France2 : Institut Galien Paris-Sud, CNRS, Univ. Paris-Sud, Univ. Paris-Saclay, 92290 Chatenay-Malabry, France
3 : Unité de Technologie Chimique et Biologique pour la santé, UMR8258 Inserm, U1022, Université Paris Descartes, Faculté de Pharmacie, 4 avenue de l’Observatoire, Paris, France
Primary plastic packaging is more and more used. However, its advantages are counterbalanced by toxicity hazard of plastic additives. It appears important to select the safer material upstream the pharmaceutical development. In that context, an optimized methodology using gas chromatography-mass spectrometry (electronic impact) was set for the evaluation of extractables (EX) and leachables (LC) for a new product development.
The cyclic olefin copolymer (COC) composes the vial, and bromobutyl (BBT) constitute the vial stoppers of the new product. Evaluation of EX was realised with small pieces of COC and BBT, which were exposed to 5 solvents or mixed solvents for 7 days at 35 °C. Potentials EX from COC and BBT were also searched by thermo-desorption at 120°C. All samples were analysed by GC-MS. The developed method
was initially tested with the analyse of 17 plastics additives minimum. Two injections modes were used for the evaluation of EX: liquid injection (LI) for research in organic solvents (column Rxi-5HT) and headspace injection (HS) for thermo-desorption (column Rtx-1301). One experimental batch with final formula (aqueous media), stressed at 40 °C during 6 months in COC/BBT packaging, was analysed by LI and HS. For stressed samples, a salting-out with sodium sulphate was performed for HS analysis. Mass spectrums observed were compared to the NIST spectrums library (for spectral similarity index > 90 %, identifications were retained).
Numerous molecules were observed with LI due to the different extraction conditions (85 molecules from COC and BBT with an identification > 90 %, including butylated hydroxytoluene, acid phthalic derivatives, norbornene and derivatives, decalin). With HS, 18 molecules were observed (including ethanol, acetone,
alkane and cyclic alkene and pentanal). During analysis on the stressed product, 13 molecules with an identification > 90 % were found during HS analysis, and 2 with LI. Ethanol, acetone, formic acid, acetic acid, pentanal, methylcyclopentane, and cyclohexane seems result from BBT rubber stoppers.
The establishment of this method was performed on 2 weeks. This approach allowed a rapid and large screening of potentials leachables in final product. Leachables found in the stressed product were mostly
identified during EX research by thermo-desorption. For this product, the use of thermo-desorption for the research of EX seems more predictive for the evaluation of potentials leachables in comparison to solvents organic tests, more aggressive. For each identified molecule, samples with enough purity will be used to confirm the NIST identification and perform quantification. Thus, it will be possible to check their concentration in the stressed product, considering toxicological threshold.