Understanding the Manufacturing Process of Lipid Nanoparticles for mRNA Delivery Using Machine Learning.

Lipid nanoparticles (LNPs), used for mRNA vaccines against severe acute respiratory syndrome coronavirus 2, protect mRNA and deliver it into cells, making them an essential delivery technology for RNA medicine. The LNPs manufacturing process consists of two steps, the upstream process of preparing LNPs and the downstream process of removing ethyl alcohol (EtOH) and exchanging buffers. Generally, a microfluidic device is used in the upstream process, and a dialysis membrane is used in the downstream process. However, there are many parameters in the upstream and downstream processes, and it is difficult to determine the effects of variations in the manufacturing parameters on the quality of the LNPs and establish a manufacturing process to obtain high-quality LNPs. This study focused on manufacturing mRNA-LNPs using a microfluidic device. Extreme gradient boosting (XGBoost), which is a machine learning technique, identified EtOH concentration (flow rate ratio), buffer pH, and total flow rate as the process parameters that significantly affected the particle size and encapsulation efficiency. Based on these results, we derived the manufacturing conditions for different particle sizes (approximately 80 and 200 nm) of LNPs using Bayesian optimization. In addition, the particle size of the LNPs significantly affected the protein expression level of mRNA in cells. The findings of this study are expected to provide useful information that will enable the rapid and efficient development of mRNA-LNPs manufacturing processes using microfluidic devices.

Cryo-milling with spherical crystalline cellulose beads: A contamination-free and safety conscious technology.

Crystalline cellulose is a common inactive pharmaceutical additive. If this material can also be used to construct beads for the wet milling of pharmaceutical compounds, it could possibly address issues related to wear and contamination associated with zirconia and polyethylene beads. In this study, the model drug phenytoin was milled with spherical crystalline cellulose (SCC) in liquid nitrogen. The particle size of the milled product was found to be comparable to that obtained using zirconia beads, verifying the feasibility of using SCC beads for this purpose. Using a design of experiment approach, the bead amount, agitation speed, and milling time were all determined to have a significant effect on the milled particle size, giving a D50 value as low as 0.3 µm. No breakage of the SCC beads was observed during the milling process in durability tests under conditions that will degrade spherical D-mannitol beads, showing that this material exhibits sufficient durability. In addition, the variation in elastic modulus between beads was minimal. Because SCC is commercially available and easy to handle, the present wet milling technique is considered to have potential applications to the manufacture of pharmaceuticals on an industrial scale, as it shows sufficient milling capability and durability.

Cryo-milling using a spherical sugar: Contamination-free media milling technology.

Milling beads experience wear upon repeated use. And milling beads made of material that is safe when ingested have not yet been developed. The present report describes the development and characteristics of spherical d-mannitol (SDM) beads, which would be safe when ingested. The model drug phenytoin was dispersed in liquid nitrogen along with SDM and the materials were agitated at high speed. The effects of the amount of beads, agitation speed, and milling time on phenytoin particle size, yield, and bead fractures were investigated using a central composite experimental design. The diameter of milled phenytoin particles decreased significantly as the amount of SDM beads and agitation speed increased. In contrast, no difference was found in the diameter with milling time. Although the fractured SDM ratio increased slightly at higher agitation speeds, the SDM was not broken and was durable enough for milling. This milling technique was applicable not only to phenytoin but also to other drug substances. Bead durability and applicability indicated that SDM can be used as wet milling beads that are considered safe for use if ingested.

Ultra Cryo-Milling with Liquid Nitrogen and Dry Ice Beads: Characterization of Dry Ice as Milling Beads for Application to Various Drug Compounds.

Ultra cryo-milling using liquid nitrogen (LN2) and dry ice beads has been proposed as a contamination-free milling technique. The morphological change of dry ice beads was visually monitored in LN2 to clarify their production process and cryo-milling process. We found that dry ice pellets, which are starting material of beads and available on the market, immediately disintegrate in LN2, resulting in the spontaneous production of dry ice beads. In addition, the resultant beads maintain their size and shape even under vigorous agitation in LN2, demonstrating that they could play a role of milling media in the milling process. The driving conditions of this cryogenic milling process including beads size were optimized to enhance the milling efficiency. Dry ice beads provided superior milling efficiency compared to original pellets. The milling efficiency increased as the size of the dry ice beads decreased; furthermore, the larger the amount of beads used, the finer the milled particles. Any crystals of three drug compounds were effectively pulverized to the sub- or single-micron range. Cryo-milling with dry ice beads is valuable on pharmaceutical field because it does not contaminate the product with fractured and/or eroded beads.

New approach to evaluate the lubrication process in various granule filling levels and rotating mixer sizes using a thermal effusivity sensor.

The principles of thermal effusivity are applied to an understanding of the detailed mechanisms of the lubrication process in a rotating mixer. The relationships and impact of the lubrication process by the pattern of powder flow, the filling level, and the rotating mixer size were investigated. Thermal effusivity profiles of the lubrication process, as obtained, indicate that lubrication is a two-phase process. The intersection point of the first and second phases (IPFS) is influenced by changing the filling level, thus changing the resulting number of avalanche flows created. The slope of the second phase (SSP) is influenced by the relationship between the number and the length of avalanche flows. Understanding this difference between the first and second phases is important to successfully evaluate the impact of proposed changes in the lubrication process. From this knowledge, a predictive model of the lubrication profile can be generated to allow an evaluation of proposed changes to the lubrication process. This model allows estimation of the lubrication profile at different filling levels and in different rotating mixer sizes. In this study, the actual lubrication profile almost coincides with the model predicted lubrication profile. Based on these findings, it is assumed that lubrication profiles at a commercial scale can be predicted from data generated at the laboratory scale. Further, it is assumed that changes in the filling level can also be estimated from the laboratory or current data.

Evaluation of risk and benefit in thermal effusivity sensor for monitoring lubrication process in pharmaceutical product manufacturing.

In the process design of tablet manufacturing, understanding and control of the lubrication process is important from various viewpoints. A detailed analysis of thermal effusivity data in the lubrication process was conducted in this study. In addition, we evaluated the risk and benefit in the lubrication process by a detailed investigation. It was found that monitoring of thermal effusivity detected mainly the physical change of bulk density, which was changed by dispersal of the lubricant and the coating powder particle by the lubricant. The monitoring of thermal effusivity was almost the monitoring of bulk density, thermal effusivity could have a high correlation with tablet hardness. Moreover, as thermal effusivity sensor could detect not only the change of the conventional bulk density but also the fractional change of thermal conductivity and thermal capacity, two-phase progress of lubrication process could be revealed. However, each contribution of density, thermal conductivity, or heat capacity to thermal effusivity has the risk of fluctuation by formulation. After carefully considering the change factor with the risk to be changed by formulation, thermal effusivity sensor can be a useful tool for monitoring as process analytical technology, estimating tablet hardness and investigating the detailed mechanism of the lubrication process.