Continuous manufacturing requires new approaches to ensure product quality. Building on quality by design (QbD) and quality by control (QbC), the aim is to dynamically adjust critical process parameters (CPPs) based on real-time measurements of critical quality attributes (CQAs). Although control of individual unit operations gained attention over recent years, comprehensive control of integrated manufacturing lines remains a challenging topic with few studies available. One major consideration is the modularity of such integrated control concepts, allowing for flexibility and resilience in a real-world production environment. This study examines an integrated control concept for a drug substance synthesis and purification manufacturing line. It consists of a continuous flow reactor, a decanter centrifuge for solid-liquid separation, and an agitated counter-current extraction column for liquid-liquid extraction. Mechanistic models based on balance equations and well-established relations for dynamic behavior were used as a virtual non-linear plant. A modular multi-layer control structure, including model-predictive control (MPC), successfully managed to control multiple objectives, including impurity limits and production rate. Maintaining all impurities within specified limits, changes in the demanded production rate could be managed in less than one hour within a 1 g/h-tolerance. It showed robustness against failures in single unit operations and under varying production rate demands.