The base matrices used in modern chromatographic materials are made of either natural (e.g. agarose) or synthetic (e.g. polystyrene or methacrylate) polymers. By good design, chromatographic beads of such polymers can be made with controlled diameters and rigidity and therefore resistant to high fluid velocities without giving too much back pressure. As a result modern chromatographic resins can be operated at fluid velocities in the range of 500 to 800 cm/h in manufacturing scale columns. This will improve productivity 2 to 4 times compared to earlier generations of resins. However, rigidity is often a result of an increased matrix volume to get the mechanical strength. This will decrease the size and number of pores which will lead to less surface area and restricted mass transfer and hence lower dynamic binding capacity. Thus, there is a trade-off between the mechanical rigidity and the dynamic binding capacity. Recently chromatographic materials based on membranes have been introduced. These have low back pressures and allow high fluid velocities with good mass transfer properties. On the other hand they have low accessible surface area and therefore low binding capacity, poor selectivity and limited plate numbers.
The most commonly used chromatographic techniques are ion exchange, affinity chromatography and hydrophobic interaction. Ion exchange ligands such as sulphopropyl (SP) or quaternary amines (Q) are commonly used in most downstream processes. Some newly developed ion exchange ligands do not only exhibit electrostatic interactions but also hydrophobic interactions and hydrogen bounding. This so-called multi-modal functionality allows protein binding at increased conductivity and introduces novel selectivity compared to traditional SP or Q ion exchangers. Affinity ligands offer the highest selectivity and are the preferred choice for the main purification step. Unfortunately it is difficult to find suitable affinity ligands with the desired chromatographic properties (specificity, stability, binding constants etc.). One exception is Protein A which is commonly used for purification monoclonal antibodies. Developments in combinatorial chemistry and protein engineering have opened new opportunities for creation of novel ligands. In particular the latter has enabled development of new affinity ligands by modification of stable protein scaffolds. Protein engineering has also successfully been used to adapt the properties of existing protein based ligands, for example in GE Healthcare's MabSelect SuRe™ which uses an alkali-stable Protein A. Combinatorial chemistry, on the other hand, is resource demanding and often requires a detailed knowledge of the structure of the target protein. Very few affinity ligands based on this technology have, therefore, reached the market. On the other hand the combinatorial chemistry approach was used to develop GE Healthcare's multi modal ligands such as Capto™ MMC.
Efficient chromatographic processes also demand well designed equipment. Most important is the chromatographic column. Such columns should be easy to pack and operate, scalable and hygienic. There are several different procedures to pack columns mainly depending on the properties of the chromatographic resin and the scale of operation. The most frequently used columns at large scale are so called “pack-in-place” columns which allow packing by pumping in resin slurry. Recently, columns for axial compression packing have been developed that allow easier automation. At GE Healthcare we have automated the complete process including filling, packing and evaluation of the packed bed using software control. Using such an automated process achieves reproducible packing with little operator dependence.
The ability to scale-up chromatographic processes is crucial for biopharmaceutical manufacturing. There are two different aspects on scale-up. The first is related to chromatographic performance which is dependent on constant residence time for the component of interest. To achieve that fluid velocity and column bed height are normally kept constant while bed diameter and volumetric flow are increased to handle the increased sample volume. The second aspect is related to pressure/flow properties in chromatographic beds. Small diameter columns exhibit significant wall support for the chromatographic resin and such columns can, therefore, be operated at much higher fluid velocities compared to wide production scale columns. The wall support does not exist in these large diameter columns and pressure/flow properties are, therefore, not linearly scalable. In the past this aspect of scale-up has been treated in an empirical manner or by performing large scale experiments. However, during recent years models have been developed which can predict pressure/flow properties over a range of column diameters.
Despite the significant development during the last decade there are still technical challenges to overcome. These mainly relate to increased titers and volumes, in particular for monoclonal antibodies, and to the introduction of new therapeutic agents like DNA plasmids and viruses. Furthermore, the biopharmaceutical industry is highly regulated by government organizations and is increasingly under pressure to reduce cost-of-goods. Chromatographic processes that can handle 10-100 kg of antibodies produced in 15 to 25 m3 cell culture fluid within 1-2 days are needed. In many cases the technical solutions exist but due to complexity, cost and/or risk of failure they are not always appealing to the biopharmaceutical community. To solve these issues developments over the entire field of downstream purification are needed, including the introduction of continuous processes. Regarding large entities such as plasmids or viruses there is a need to overcome the diffusion limitation for large molecules in existing chromatographic resins. Improved surfaces and /or new materials from the nanotechnology field might be possible solutions.
Chromatography has shown to be extremely valuable for purification of biopharmaceuticals and will remain the preferred technique even for the future. The biomass and process economy challenges from future manufacturing scenarios will drive the development of technology in the coming years. By improved understanding of the chromatographic process and by exploitation of new technology in material sciences and chemical engineering it is likely that the required improvements are reachable.