(624f) Fully Automated Platforms for Self-Optimization of Continuous Flow Synthesis of Active Pharmaceutical Ingredients

Continuous flow synthesis of active pharmaceutical ingredients (APIs) is an ideal technology for an automated self-optimization optimization due to its inherent benefits, such as precise control of reaction times, temperatures, and composition. We present the development and demonstration of fully automated platforms for self-optimization of continuous flow synthesis of APIs at Pfizer as a part of the Flexible API Supply Technology (FAST) initiative. These lab-scale continuous fully automated platforms include in-house designed LabVIEW™ Virtual Instrument (VI) automation software integrated with lab equipment and iterative experimental design based on artificial intelligence (AI) and active machine learning (ML) optimization algorithms. We implemented automated controls of lab equipment including Uniqsis Polar Bear Plus Flow^TM reactor, feed pumps, temperature controllers, thermocouples, stirrers, and Coriolis mass flowmeters. In addition, we enabled integration of a variety of in-line process analytical technology (PAT) tools via OPCUA, including Mettler ReactIR™ spectroscopy, and UPLC for feedback optimization, data visualization, and real time process understanding. We integrated advanced optimization algorithms including mixed-integer nonlinear programming (MINLP, MATLAB® & SIMULINK®) algorithms, multi-objective Bayesian (Python^TM), and multi-dimensional sinusoidal dynamic experimentation. Thus, integration of equipment, PAT, and automation control software produces closed-loop systems that, when paired with an optimization protocol, enables automated design of experiments (DoEs) with automated execution of the DoE, ultimately leading to self-optimization of continuous flow synthesis of API. We present multiple case studies employing this autonomous self-optimization platform to enable the identification of optimal conditions for flow synthesis of APIs, while reducing the amounts of raw materials consumed, compared to a one factor at a time (OFAT) approach. This automated platform requires minimal human intervention, relieving expert scientists of manual tasks so that they may focus on new ideas.