DISPOSABLE FT-AEX MEMBRANE CHROMATOGRAPHY FOR POLISHING
AEX resins typically are used as a polishing step in a FT mode for contaminant removal. Because the AEX resins remove only
trace level of impurities, in principle, a small column should be significant to meet the impurity clearance requirements.
However, the columns usually are over-sized at the manufacturing scale to meet process throughput requirements. AEX membrane
chromatography provides an attractive alternative because of its convective mode of mass transport, which allows operation
at a significantly higher linear flow-rate or smaller residence time than columns. The convective mode of mass transport coupled
with the need for trace levels of impurity clearance enables the use of a membrane chromatography adsorber that is significantly
smaller in size than a conventional column, which reduces buffer requirements.
Zhou and his collaborators at Amgen have extensively studied the membrane chromatography technology using Sartobind membrane
capsules.13,14 The authors estimated water and buffer consumption comparing a stainless steel column to disposable membrane chromatography
during polishing steps for contaminant removal and virus clearance in MAb purification processes. The authors observed a 95%
buffer savings with the FT membrane chromatography as compared to traditional column chromatography, regardless of the production
scale.15 The buffer and water consumption did not include buffers for column packing and validation incurred with resin reuse. At
the 2,000-L scale, a 35-L column was necessary to accommodate the throughput, whereas a 0.18-L membrane volume was sufficient
to clear contaminants and pathogens and achieve the required product purity. The smaller membrane chromatography device provides
product of the same quality while reducing floor space requirements, hardware equipment, processing time, and labor use.
Figure 1 shows BioSolve (Biopharm Services, Ltd.), the Excel-based model that was used in this study. Models are useful tools
to define process performance and gain insight into process fit in the confines of an existing plant. Of course, they only
provide an approximation—a model of the real situation limited by the information available for input. The use of the model
provides an estimation of CoG and CoG breakdown by cost categories. In the model, the CoG takes into consideration the fixed
overheads of the facility and the variable operations costs of the process. The fixed overheads include capital charges, taxes,
and insurance, whereas the variable costs include materials, consumables, labor, and waste management.12,16
Figure 1. BioSolve model used to estimate total CoG and CoG breakdown by cost categories. Data provided by Biopharm Services.
Figure 2 depicts the framework of BioSolve, which comprises the user interface, process definition, productivity levels, cost
calculations, and outputs. The interface provides a list of predefined key input parameters (e.g., process scale, expression
level, solution preparation basis, single-user options) that enable the user to perform quick what-if analyses for different
scenarios. The process information is defined in the cost model, which consists of the sequence of unit operations, mass and
volume balances, equipment sizing, process operating conditions, and resource allocation. The model computes the facility
throughput, equipment list, materials and consumables usage, and labor requirements. The model outputs include plant productivity,
bill of materials and consumables, capital expenditure, and CoG.
Figure 2. Framework of the BioSolve model. The model comprises the user interface, process definition, productivity, cost
calculations, and output.