Soft sensor based on pH for real-time monitoring of mRNA medicines production

Abstract

Real-time monitoring of in vitro transcription (IVT) reactions is critical for the continuous manufacturing of high-quality mRNA vaccines and therapeutics against a wide spectrum of diseases. We report the first real-time, non-invasive IVT monitoring approach, based on H + release during nucleotide triphosphate (NTP) incorporation. This soft sensor processes fiber-optic pH measurements to infer the concentrations of up to 40 IVT species that cannot be otherwise measured in real-time. Validated against orthogonal offline measurements, it delivers updates every 25 milliseconds via two complementary models. The first couples a mechanistic IVT model with an Unscented Kalman Filter (UKF) to dynamically estimate ≈ 40 key indicators, including mRNA yield (R 2 = 0.95) and NTP depletion (R 2 = 0.84). The second applies the Henderson-Hasselbalch correlation to reconstruct mRNA yield (R 2 = 0.93) and NTP depletion (R 2 = 0.76) from buffer capacity and pH change alone. This soft sensor enables real-time process monitoring by generating ≈1,600 concentration estimates per second, supporting Quality-by-Digital-Design and advanced control for continuous, disease-agnostic mRNA medicines manufacturing.