Automated self-optimization of continuous crystallization of nirmatrelvir API

Abstract

Continuous flow crystallization is an attractive mode of operation, due to its ability to generate consistent product quality while requiring a smaller footprint and lower production costs than its batch counterpart. We present a novel combination of a custom/in-house automated continuous crystallization platform integrated with self-optimization algorithms. We demonstrate the automated optimization of continuous crystallization of nirmatrelvir (PF-07321332), one of the active ingredients in Paxlovid™, a potent, selective, and orally bioavailable inhibitor of SARS-CoV-2 M^pro. The continuous crystallization platform consists of three mixed suspension mixed product removal (MSMPR) crystallizers in series and includes an in-house designed automation user interface integrated with lab equipment. The platform also has an iterative design of experiments (DoE) based on mixed-integer nonlinear programming (MINLP) self-optimization algorithms. We implement automated controls for the lab equipment, including a flow sonication cell as the nucleation device, feed pumps, temperature controller units (TCUs), thermocouples, pressure sensors, stirrers, and coriolis mass flow meters. We enable integration of variety of in situ process analytical technologies (PATs) via open platform communications unified architecture (OPC UA), including Mettler Toledo ParticleTrack™ with FBRM® (Focused Beam Reflectance Measurement) technology and Blaze™ Metrics imaging probe for the data visualization and real time process understanding.