The success of COVID-19 mRNA vaccines showcased the transformative potential of this adaptable, plug-and-play, and rapid response platform. Although these vaccines offer faster development and production compared to those traditionally produced in cell culture or eggs, critical manufacturing challenges remain. Notably, costly raw materials are required for in vitro transcription (IVT) and current methods for real-time monitoring of IVT are lacking. Herein, we present simultaneous inline monitoring of ATP, CTP, GTP, UTP, and mRNA concentrations during IVT using Raman spectroscopy and partial-least squares regression (PLS) data analysis. Model prediction performance resulted in R2 of 0.82-0.99 and relative errors of 4%-13%, comparable to errors from reference offline assays (10%-12%). Furthermore, we analyze Raman spectral features associated with total mRNA concentration and sequence-specific variations, demonstrating sequence-independent prediction capability that eliminates the need for model recalibration across different products. This approach advances real-time monitoring for mRNA manufacturing and supports the future transition toward continuous processing, automated digital twins, and Pharma 4.0 paradigms.
Fulber et al. (Mon,) studied this question.