UK +44 (0)1865 849841
Malaysia +60 3 2117 5193

What are the main bottlenecks for realising continuous manufacturing?

Bioprocesses are traditionally performed in (fed-) batch processes for the production of various recombinant proteins and therapeutics. Batch processing comprises of processes where the material flow has hold steps in between two unit operations where the product is recovered from each unit operation step. However, such batch processes are attributed to longer process time and costs, and inconsistent process performance.

Owing to the hold steps in between unit operation the overall space-time yield (STY) of the process is drastically reduced. Therefore, in order to overcome these bottlenecks alternative modes of operations were considered. Single-use systems combined with continuous processing offers an attractive alternative to batch processing. Continuous processing can be defined as processes where there is a continuous flow of material at a steady state with controlled productivity. The main advantages of continuous processing are low production costs, smaller plant footprints (Lean manufacturing) and increased STY. Food and chemical industries mainly use continuous processing, however, due to stringent regulatory implications has not yet been realised in biopharmaceutical industry.

With the release of Code of Federal Regulations (CFR) from the Food and Drug Administration (FDA), a ‘batch’ is defined in terms of quantity of the material rather the mode of operation, which in turn could accelerate implementation of continuous processing. Although continuous processing offers a superior alternative to batch processing, it is characterised by some drawbacks. The major drawbacks are processed variability and long process development times. Even though various approaches exist to overcome the drawbacks, continuous processing has not yet been realised in biopharmaceutical industry. Therefore, we shortlisted questions to identify the main bottlenecks for realising continuous processing,

  1. Models are available to identify process variability, but rarely seen in industrial bioprocesses. Why are modeling approaches still not transferable for industrial bioprocesses?
  • Process variability arises from critical process parameters (CPPs) and critical material attributes (CMAs) which in turn affect the critical quality attributes (CQAs) and overall process performance.
  • Mechanistic models can be used to identify and characterize variations in CPPs and CQAs.
  • However, such mechanistic models are not transferable. This is mainly because model development is done in a process development stage, which necessitates model optimization and validation prior to implementation in an industrial process.
  • Recently, a self-iterative model development workflow was proposed to aid bottlenecks in technology transfer of process models.

Modeling workflows can be used as an enabler to develop process models which can be directly transferred and validated in industrial processes.

  1. Continuous bioprocessing requires real-time monitoring of process variables to ensure productivity. What is the main bottleneck to achieve real-time process monitoring?
  • Real-time process monitoring requires necessary process analytical technology (PAT) tools. Commercial PAT tools (e.g. online sampling devices or microscopes) can be used to measure CPPs and CQAs in real time.
  • Model-based methods in combination with multivariate data analysis (MVDA) tools can be used for monitoring state variables (e.g. concentrations) or performance indicators (e.g. rates) in real-time.
  • Real-time architecture for measuring and modeling the process variables online have been made available with the latest computational power.
  • PAT tools (measure) in combination with real-time architecture (model) can be used to monitor process performance in real time.

 Advanced PAT tools and real-time architecture can be used as enablers for real-time process monitoring.

  1. Biological variations and time effects should be controlled for continuous processing. How can we effectively counteract biological time effects and variations?
  • Biological variations cause the utmost variability inside the process design space.
  • With advances in molecular biology, promoter systems with the ability to tune productivity have been developed and implemented.
  • Such systems can be used to control productivity at a steady state over long periods of time, effectively removing biological time effects.
  • Furthermore using model-based approaches, control strategies can be implemented to subdue process variability.

Tunable promoter systems can be seen as enablers to effectively counteract biological time effects, thereby controlling process variability.

Summarising, bioprocesses are complex and dynamic in nature with high process variability. Realizing continuous processing in biopharmaceutical industry necessitates identifying, monitoring and controlling process variability. Utilization of state-of-the-art modeling workflows and PAT tools renders consistent identification and monitoring of process variability. Tunable promoter systems can be used to curb biological time effects, thereby curbing process variability. Recently using continuous bioprocessing, Orkambi, a cystic fibrosis drug, from Vertex has been produced. It is evident that the biopharmaceutical industry is interested in continuous processing due to its vast advantages. Owing to the huge wealth of tools and models available, we envision the intense use of continuous processing in biopharmaceutical industry in the near future.

 
Vignesh Rajamanickan is a Post Doctorate Researcher, Research Division Biochemical Engineering, Institute for Chemical, Environmental and Biological Engineering, TU Wien, Austria.

 

He will be speaking at the Global Bioprocessing & Bioanalytics CongressDownload the agenda for the full list of speakers.

 

 

Leave a Reply

Subscribe to Our Newsletter

Get free reports and resources from our world class speakers.
  • This field is for validation purposes and should be left unchanged.

Life Sciences Twitter Feed

Archive