Messenger RNA (mRNA) therapeutics and vaccines require a specialized formulation and encapsulation in lipid nanoparticles (LNPs) to facilitate drug product stability and delivery into target cells. Traditional batch manufacturing of biopharmaceuticals, particularly in the LNP formulation steps, is time-consuming and inefficient due to the intensive resource and labor requirements to complete the individual steps of the process development. Continuous manufacturing, incorporating process analytical technologies, automation, and real-time predictive analytics with process adjustments, has emerged as an alternative that would potentially reduce costs while improving production efficiency and product quality. Specifically in the final formulation processing steps, adequately monitoring the production of mRNA-LNPs allows for greater control of nanoparticle formation. For example, LNP models can ensure maintenance of particle composition stability despite shear forces that may result from processing steps. Raman spectroscopy is a powerful tool that has been increasingly implemented for the in-line analysis of biochemical content and concentration. Raman spectral data acquisition with offline critical quality attribute calibration can be used to develop machine learning models used for predictive analytics, as well as understanding and optimizing process parameters. We developed a novel model from off-line Raman spectral analysis to monitor total lipid concentration in mRNA-LNP formulation production unit operations. Further, we have identified unique Raman spectral signatures for individual lipid components towards the development of an LNP compositional analysis model. The orchestration of Raman spectroscopy as well as other process analytical technologies at specific downstream LNP formulation unit operations has demonstrated an increased understanding of these process development steps. However, significant challenges remain in accurately developing and implementing these technologies to enable real-time monitoring and control of LNP production steps.
