#3 Boosting mAB Manufacture

Protein A resins are already advanced – why is it important to continue to innovate?

Monoclonal antibodies (mAbs) continue to be the largest and fastest growing class of biotherapeutics, so ensuring continued efficient process development and good manufacturability is important. Protein A resin, with its high specificity and affinity for the Fc region of antibodies, enable a plug-and-play approach where up to 99 percent purity can be achieved in one, almost generic, purification step. It’s relatively easy and cost effective to set up a purification platform that can accelerate process development and enable fast turnover of mAb projects.

The bioindustry has seen a more than a 100-fold increase in the productivity of cells in the upstream part of the biomanufacturing process, which means downstream purification technology must continue to evolve and meet these productivity gains. During the last few years, the protein A chromatography resin step has started to become a rate-limiting step in many processes; if the efficiency of protein A resin is determining the efficiency of the entire manufacturing facility, it’s important to continue to innovate. And I think there is still room for improvement. For example, one key concern for both regulatory agencies as well as the biopharma industry is bioburden control. The protein A step is loaded with all the cell culture nutrients cells are grown in, which makes the load fraction coming onto the protein A resins highly likely to promote bacterial growth. In addition, as protein A resins have historically been less chemically resistant to sodium hydroxide, the most commonly used cleaning and sanitization agent, there is an elevated risk for microbial contamination. Such issues can be fixed with new technologies.

How did the development project for MabSelect PrismA come about?

We knew from speaking to biopharmaceutical producers working with mAb purification that increased binding capacity and increased productivity of the protein A step were highly desirable. We gathered a lot of feedback from biopharmaceutical producers before starting the project, and we realized there was increasing focus on bioburden control. Regulatory agencies have started to ask more questions about the sources of bioburden, and what manufacturers are doing to control them; the cost can be significant – from tens of thousands of dollars for QA investigations, to several million if chromatography resins or drug batches need to be scrapped. To address these issues, we gave our R&D scientists the challenging task of developing a protein A resin with both higher capacity and increased alkaline stability, which would improve bioburden control and process economy in mAb processing.

How did the team meet this challenge?

To increase alkaline stability, our researchers started from the alkaline stabilized and engineered protein A ligands used in our MabSelect SuRe and MabSelect SuRe LX resins. They further engineered the ligand by substituting alkaline sensitive amino acids for more stable ones and then evaluating the outcome. However, we soon realized there could be almost an infinite number of combinations and variants. Instead, we turned to directed evolution, and developed a high-throughput screening assay. Several libraries were made with gene variant synthesis, where key positions were randomly substituted from a subset of amino acids. The gene libraries were then transformed into E. coli and we screened clones with unique sequences. All substitutions that had a positive impact on the stability were saved and combined with other substitutions. The best combinations were then cultivated on a larger scale to achieve more protein and verify the final application. In total, we created and evaluated more than 400 variants of the protein A ligand.

At the same time, efforts were made to increase the binding capacity of the resin. The number of binding domains of the protein A ligand were engineered, as well as the ligand density on the resin. The team experienced limitations in the existing available resin base matrices, so a decision was made to develop a new base matrix, optimized for the new protein A ligand. Both the protein A ligand and resin base matrix were then co-developed to find an optimal balance between binding capacity, alkaline stability and pressure flow properties to get the highest productivity. All in all, more than 240 base matrix variants were evaluated for optimized performance in combination with the new ligand.

It was a lengthy process and there were plenty of challenges along the way, but when we finally realized the design worked, and we could achieve the desired goals of the prototypes, we knew we had something the industry would be excited about. The development project included more than 50 people, lasted almost three years and resulted in a new ligand, a new base matrix and setting up MabSelect PrismA protein A ligand production in the Uppsala facility located in Sweden.

What makes the product so important?

Monoclonal antibodies will continue to be the most important class of biotherapeutics for the foreseeable future, but we are also seeing an increased number of bispecific antibody variants and antibody drug conjugates. Many of these have the Fc region intact, which makes protein A resins an attractive technology for capture. With MabSelect PrismA, the improved binding capacity allows biomanufacturers to deliver more mass throughput in their existing equipment and facilities, opening a bigger window of operation for prepacked columns. Today, the largest ReadyToProcess column we have can harvest and purify a high-titer 2,000 L bioreactor – that’s something we couldn’t do with the previous generation of resins in the portfolio. As you start to increase capacity and thereby shrink the size of columns in operations, you also get a lot of secondary effects, such as smaller buffer tanks and lower consumption of buffer, which can keep plant efficiency high.

The increased alkaline stability also means a protein A resin can, for the first time, use the same efficient sodium hydroxide solutions as other chromatography techniques, such as ion exchange, multimodal and hydrophobic interaction chromatography, which makes cleaning and sanitization more efficient, and improves process robustness.

How have customers reacted?

A number of biopharmaceutical producers have evaluated samples and, so far, the results reflect the data the GE R&D teams have seen – in some cases, much better than anticipated. Some report they would use the improved binding capacity to improve mass throughput of target mAbs from existing equipment, while others said they would use the improved capacity to shorten process time. We have also heard evaluators express interest in using the improved alkaline stability to clean and sanitize columns more efficiently.

How would you like to see protein A resins improve in the future?

With an increase in diversification of antibody structures, protein A will have to adapt to new target molecules. There is potential for protein A to be offered in several tailored versions based on antibody structure or other types of ligands that address a bigger portion of the non-Fc region containing antibody library. During the development of this product, we gained a lot of knowledge on how to engineer affinity ligands to achieve different characteristics – and through close collaboration between suppliers and end-users, new solutions can be developed.

The enhanced alkaline stability properties of MabSelect PrismA protein A ligand has the potential to be used on other chromatography platforms – and I predict there are applications where other chromatography base matrices can be used as anchors for the MabSelect PrismA ligand to reach specific purification objectives.

David Westman is Product Marketing Manager, BioProcess Chromatography Resins, at GE Healthcare Life Sciences, Sweden.