Wiest, Elani F.

Interest in Current Good Manufacturing Practices (cGMP)-grade extracellular vesicles (EVs) is expanding. Some obstacles in this new but rapidly growing field include a lack of standardization and scalability. This review focuses on automated biomanufacturing of EVs in conditioned media collected from cultured mesenchymal stromal cells (MSCs). Different automated cell culture systems are discussed, including factors affecting EV quantity and quality, isolating EVs manufactured in an automated system, and validations needed. The ultimate goal when manufacturing cGMP-grade EVs is to identify a specific application and characterize the EV population in detail. This is achieved by validating every step of the process, choosing appropriate release criteria, and assuring batch-to-batch consistency. Due to the lack of standards in the field, it is critical to ensure that the cGMP-grade EVs meet FDA standards pertaining to identity, reproducibility, sterility, safety, purity, and potency. A closed-system automated bioreactor can be a valuable tool to generate cGMP-EVs in a scalable, economical, and reproducible manner.

Abstract Background Extracellular vesicles (EVs) have gained interest in the field of regenerative and transfusion medicine. Specifically, current Good Manufacturing Practice (cGMP)‐grade EVs produced by mesenchymal stromal cells (MSCs) are an intriguing option for cell‐free therapeutics. With the development of cGMP‐grade EV products, a simple and reliable method for batch‐to‐batch consistency is needed. Study Design and Methods The objective of this study was to validate a method to semiquantitatively assess the batch‐to‐batch consistency of isolated EVs. A multiplex bead‐based flow cytometric assay containing 37 surface markers and 2 assay control antibody‐coated capture beads was validated. Detection limits ( n = 10 buffer samples), repeatability ( n = 9 EV samples), and intra‐observer reproducibility over 2 days ( n = 10 EV batches) were assessed. A Spearman correlation matrix was used to evaluate the batch‐to‐batch consistency of independently isolated EV products ( n = 37 surface markers). Batches with a Spearman correlation coefficient ≥0.9 and p < 0.05 were considered statistically indistinguishable from previous batches. Results This assay demonstrated robust repeatability as well as intra‐ and inter‐assay reproducibility. In‐house batches of EVs were significantly correlated (r ≥ 0.90; p ≤ 1×10 −14 ). Compared with buffer, EV batches had correlation coefficients near zero (r ≤ −0.10; p ≥ 0.12). Commercially sourced EVs significantly correlated with in‐house EV batches, but fell below the 90% correlation cutoff (r ≤ 0.71; p ≤ 0.0004). Discussion This time‐efficient and technically simple assay offers a robust method of quality control for assessing the batch‐to‐batch reproducibility of cGMP‐grade EV products.