What maximum shelf-life should be assigned to sterile injectable preparations? A risk-based hierarchical classification approach

Introduction
Despite demonstrated physical and microbiological integrity of primary containers, validation of the physicochemical stability of preparations, qualification of operators and equipment, and compliance with the 2023 Good Preparation Practices, the assignment of maximum microbiological shelf-life for our preparations remains uncertain. The GERPAC consensus conference [1] proposes a risk-based approach. The aim of this work was to explore the assignment of microbiological beyond-use dates using a classification of preparations according to their level of risk.

Materials and methods
A multidisciplinary team (pharmacists and pharmacy technicians) analyzed 134 distinct compounding procedures, covering 119 molecules, prepared in our unit (>65,000 preparations/year). All steps considered as risk-prone were identified: number of vials to be reconstituted, sampling steps, transfers, dilutions, withdrawals or additions of solvent, and filtration. A classification grid was developed based on the number of steps performed, the risk associated with each step, the nature of the drug substance, and the route of administration. A final score categorized preparations into 3 risk levels: low, intermediate, and high.

Results/Discussion
Low-risk preparations mainly consisted of ready-to-use products, including monoclonal antibodies, involving 1 to 3 preparation steps and administered intravenously. High-risk preparations involved reconstitution of multiple vials, transfer operations and/or additions or withdrawals of volumes, exceeding 5 preparation steps and/or involving a high-risk route of administration (e.g., intrathecal). These results could lead to adjustments in shelf-life according to the complexity score and raise questions regarding the relevance of anticipatory compounding for complex preparations. Conversely, an extension of shelf-life could be considered for low-risk preparations. This reflection could be further developed, particularly in settings involving extensive anticipatory and serial production, by considering evolving factors throughout the preparation life cycle: return of unadministered preparations from wards, multiple reassignments, temporary breaks in cold-chain, etc.

Conclusion
The assignment of microbiological stability remains the responsibility of the pharmacist and relies on risk-based judgement, leading to inter-institutional variability. The proposed approach aims to help objectify some determinants of microbiological risk, particularly by integrating the complexity of compounding procedures into its assessment. It also highlights the importance of considering extrinsic factors beyond the compounding step throughout the entire preparation life cycle. Further perspectives include validation through appropriately designed MFTs with the aim of broader implementation in other institutions.

See the poster