Research aims to predict scaled up process performance

Research has been initiated into the prediction of process performance before scale up to help increase yield and efficiency while cutting costs.

The research uses ultra-scale down methods, which give data on manufacturing performance but use minimal materials, and a programming framework. By combining these elements the researchers believe they can make predictions about how processes will work on a large scale.

Biopharm Services and University College London’s Innovative Manufacturing Research Centre (IMRC) for Bioprocessing are collaborating on the research having been awarded a knowledge transfer secondment by the UK Engineering and Physical Sciences Research Council (EPSRC).

Nigel Titchener-Hooker, who leads the IMRC, told in-PharmaTechnologist that at the end of the six month research project the team hopes to have demonstrated proof-of-concept. The methodology will then be incorporated into teaching at UCL.

Commercial application of the method is also being worked towards. Titchener-Hooker said that for individual operations commercial use could occur in 12 months. Work on all operations will take a little longer and reach commercial clients in 12 to 24 months.

EPSRC works with major biopharm companies and consequently the industry is aware of the research. Titchener-Hooker said that the researchers report their progress to EPSRC.

The research

The first step of the research focuses on how to measure engineering properties of biological materials used in bioprocessing by IMRC ultra scale-down methods. For instance, the research will look at how an antibody breaks up when certain manufacturing processes are performed.

Gaining an understanding of the properties of a molecule allows the researchers to feed parameters, such as sensitivity to shearing, into the programming framework developed by Biopharm Services.

By linking the data generated using ultra scale-down methods to the programming framework, called BioSolve, the researchers believe they can predict full process sequence scenarios.