The challenges of continuous multicolumn protein A chromatography

by

Dr Hani El-Sabbahy, a biopharmaceutical application engineering specialist at science-based technology company 3M, outlines some of the challenges associated with continuous multicolumn protein A chromatography.

The advantages of continuous bioprocessing have been well covered in the literature. They include reduced equipment footprint, reduced capital cost, higher productivity, greater flexibility to increase capacity by operating parallel production lines, and, in some respects, enable a steady state operation. There is, therefore, significant interest in continuous processing in the biopharmaceutical industry.1

Whilst a number of companies currently use continuous upstream processes, there remain a number of challenges and barriers to implementing continuous downstream processes and consequently overall integrated continuous processes for the production of biopharmaceuticals. These challenges include less flexibility of each production line and greater risk of contamination, both of which result from long run times. Additionally, process understanding remains relatively poor due to the complex nature of biological systems and, at times, the empirical approach to process development that is still prevalent today.

Furthermore, the industry is only really at the beginning of addressing the need for robust online measurement of key parameters required to enable effective process control.2 With these in mind, one of the key barriers is system complexity or perceived complexity.3 Of course, the more complex the system, the more difficult it is to understand, model and control.

The cost of protein A is one of the most significant costs in a mAb process. Decreasing the cost of protein A should have a significant impact on the overall cost of the downstream process. Continuous multicolumn chromatography is able to achieve this by using more of the capacity of the resin, and for clinical manufacture, more of the resin lifetime.4

Work specifically on the impact of multicolumn continuous protein A mAb capture chromatography on cost of goods has shown the biggest impact is at clinical, rather than commercial, scale due to the greater resin reuse at commercial scale.4,5 However, there was a key difference between these two studies, namely, the residence times investigated. The Xenopoulos’ study conducted the continuous multicolumn protein A chromatography using incompressible beads at 0.5 min residence times showed savings of 16% to 24% at commercial scale. In contrast, Pollock’s study was conducted at residence times more characteristic of batch operations (greater than or equal to three minutes) and showed only a five percent reduction in cost of goods at commercial scale. Therefore, it appears that selecting most productive conditions can make the difference between modest and significant cost savings at commercial scale.

Loading and regeneration are necessary steps for each of the columns in a continuous multicolumn chromatography unit operation. The regeneration, as in a batch system, consists of the equilibration, wash, elution and cleaning steps. Matching the loading and regeneration steps as closely as possible is necessary to achieve maximum productivity.4 Optimisation of the column regeneration steps is required to ensure they are as short as possible without impacting product purity and quality. Once this has been achieved, the regeneration time can effectively limit the maximum achievable productivity unless more columns are introduced.

Another important point to consider is the length of the load step which is a function of the loading residence time and the product titre. Higher product titres will give rise to shorter load times whilst lower product titres will give rise to longer load times at constant residence time.

It has been proposed that residence time is the most important parameter affecting the productivity of a periodic countercurrent chromatography protein A capture step.6 However, particularly at higher feed stream product titres, shorter residence times can be impractical due to the length of the regeneration time.

In the literature, some have suggested that this can be mitigated by running those steps faster than loading or for a shorter time or by compromising on residence time.5 The other approach to deal with this, is to increase the number of columns to maintain the balance between loading and regeneration. This increases system complexity which, as discussed above, makes it more difficult to control and model.

Alternatively, it is hypothesised that feeding the continuous multicolumn protein A chromatography unit operation with cleaner feed material may allow these regeneration steps to be further shortened, thereby allowing shorter residence times or higher titre feed material to be used whilst minimising system complexity.

To explore some of these issues, 3M has commissioned experts at UCL (University College London) to conduct in-depth research into the performance of the Emphaze AEX Hybrid Purifier. The goal of the project is to quantify the value of purifying the cell culture fluid with the purifier before the continuous chromatography operation. This product may potentially reduce the time required for non-loading steps by reducing soluble impurities upstream of capture chromatography. The results are expected to be published in a scientific journal next year.

References:

  1. Konstantinov, K.B., and Cooney, C.L., ‘White Paper on Continuous Bioprocessing. May 20–21, 2014 Continuous Manufacturing Symposium’, J. Pharm. Sci., 2015;104(3):813–820.
  2. Schmidt, S.R., ‘Drivers, Opportunities, and Limits of Continuous Processing’, BioProcess International, 20 Mar 2017. [Online]. Available: http://www.bioprocessintl.com/manufacturing/continuous-bioprocessing/drivers-opportunities-limits-continuous-processing/
  3. Munk, M., and Langer, E.S., ‘What is holding Industry back from Implementing Continuous Processing: Can Asia Adopt More Quickly? — Biopharma Asia’, Biopharma Asia, 28 Feb 2017.
  4. Pollock, J., et al., J. Chromatogr. A, 2013;1284:17–27.
  5. Xenopoulos, A., J. Biotechnol., 2015;213, no. Supplement C:42–53.
  6. El-Sabbahy, H., et al., ‘Factors affecting the productivity of 4-Column Periodic Counter Current Chromatography (4C-PCC)’, Integr. Contin. Biomanufacturing II, Nov 2015.