Continuous manufacturing of lipid nanoparticles: Influence of device designs and operating parameters

Lipid nanoparticles (LNPs) have gained tremendous attention as drug delivery vehicles in recent years. Considering the rapidly growing number of applications and demand for LNPs, it is imperative to develop equipment platforms for their continuous manufacture. In this work, we experimentally investigated the particle size, polydispersity index (PDI) and zeta potential (the critical quality attributes, CQAs) of Imwitor 742 LNPs synthesized using antisolvent precipitation. Three fluidic devices, namely a fluidic oscillator, a pinched-tube and a vortex-based cavitation device, were used for the synthesis. The continuous production of LNPs was carried out using a loop configuration. The influence of micro-mixing time (t_micro, 20–80 ms) of the devices, the flow rate ratio (ϕ) of antisolvent to solvent (2–10), lipid concentrations (1–50 mg/mL) and surfactant concentration (0.1–2 % w/v) on the CQAs of the LNPs was experimentally investigated. Suitable correlations describing the influence of these key process variables on LNP size were developed. The data were also used to build an artificial neural network (ANN) model to predict the particle size. The model showed high predictive accuracy capturing the non-linear behaviour of several of the process parameters. Among the studied devices, the pinched-tube demonstrated the best performance in terms of small (<200 nm) and monodisperse (PDI < 0.1) LNPs at the lowest rate of energy dissipation. The presented data and analysis will be useful for selecting appropriate devices and operating parameters for continuous manufacturing of LNPs with desired quality attributes.