Time-dependent particle size increase during lipid nanoparticle purification by cross-flow filtration

Although cross-flow filtration (CFF) is a time- and resource-efficient purification method, CFF has rarely been explored for lipid nanoparticle (LNP) purification as a scalable alternative to dialysis. CFF-based processing allows for buffer exchange by diafiltration (DF) and setting a target product concentration by ultrafiltration (UF). Herein, we investigate the effect of CFF-based processing on LNP characteristics and process performance by performing a parameter study through the variation of selected membrane-related and operational parameters. We used a pre-dilution approach prior to LNP purification to reduce the ethanol content while maintaining LNP characteristics. Taking advantage of the integration potential of CFF for process analytical technology (PAT), we successfully established size monitoring for LNPs by integrating at-line dynamic light scattering (DLS), providing near real-time process insights. During processing, we observed an increase in LNP size and a change in their size distribution, dependent on processing time but not on the varied process parameters. Following comprehensive off-line analyses, all other LNP characteristics remained constant and final lipid recoveries were achieved in the range of 86-89% for all CFF processes. Long-term, the CFF-purified LNPs showed a lower increase in size compared to the dialyzed LNPs during storage of 14 days. Lastly, examination of purified LNP behavior during sterile filtration revealed changes in particle size in the upper size range. In general, we provide comprehensive insights into CFF-based LNP processing and its impact on LNP characteristics and process performance. Such studies are expected to contribute to the understanding of CFF-based LNP processing and their future application for size-controlled LNP production.