Comparative analysis of empty and full adeno-associated viruses under stress conditions by anion-exchange chromatography, analytical ultracentrifugation, and mass photometry

Adeno-associated viruses (AAVs) have emerged as a promising gene delivery vehicle for the treatment of diseases. As AAVs are a complex therapeutic modality, new analytical techniques are needed to thoroughly characterize the critical quality attributes (CQAs) to support drug development. Empty and full ratio is one of the CQAs of AAVs that may impact drug safety and efficacy. While the empty and full ratio of AAV therapeutics in untreated conditions can be well characterized by different analytical methods, limited studies have demonstrated whether these analytical methods can be used for characterizing stressed AAVs, which can help with assessing the stability of the molecule and identify potential degradation pathways. Here, we employ three orthogonal analytical techniques - (1) anion-exchange chromatography (AEX), (2) analytical ultracentrifugation (AUC), and (3) recently introduced mass photometry (MP) - to investigate their ability to characterize AAV samples subjected to various stress conditions, including freeze/thaw, physical agitation, and thermal stress. Based on our observations, AEX is a high-throughput technique. However, it falls short in quantifying the amount of partially filled AAVs, and the quantification is significantly impacted by post- translational modifications, which often occur to AAVs under stress conditions. AUC provides the best resolution for stressed AAV samples but is limited by its throughput and high sample consumption. MP combines the strength of being high-throughput and requires the least amount of samples, albeit at the expense of lower resolution compared to AUC. Our study suggests that each of these three techniques, AEX, AUC, and the emerging MP method, is suitable for characterizing empty and full AAV particles at various stages throughout the lifecycle of drug development. (c) 2025 American Pharmacists Association. Published by Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.