By Kilian Kobl - Project Manager at Ypso-Facto
Current oligo manufacturing processes are still somewhat larger scale versions of the initial processes conceived in the lab. Oftentimes, during development of an API, the choice of speed had to be made at the expense of conceiving a process that produces less waste and makes best use of the raw materials.
Besides that, traditional process development has relied heavily on trial-and-error experimentation which is extremely time-consuming. For example, if we take a look at oligonucleotide purification by ion exchange, there are many parameters you may need to investigate: the impact of buffer type, pH, gradient, flow rate, etc. to name just a few. In fact, there are so many parameters that it is nearly impossible to achieve a full process characterization within a reasonable time frame using trial and error or even Design of Experiment approaches. Consequently, in a fast-moving environment, at some point, one remains stuck with a non-optimal separation.
Predictive simulation tools can help greatly speeding up process development while leaving some time for optimization. With the help of a few well-targeted experiments, the process is characterized and represented digitally by a mechanistic model. The effect of operating conditions like crude concentration, loading, number and duration of washing steps, elution strategy, fraction collection strategy etc. on product purity and solvent consumption can then be predicted in silico. With such a tool, an optimum is quickly found while the results are backed up by a solid mechanistic description of the process...
Discover the MARCH/APRIL 2022 issue of Chimica Oggi – Chemistry Today, the peer reviewed, bimonthly journal, of the TKS TeknoScienze Publisher.
Ypso-Facto is a service company helping industrial firms to develop, optimize and secure their chemical processes and bioprocesses.