Issue link: https://www.e-digitaleditions.com/i/1499694
22 BioPharm International ® Partnerships for Outsourcing eBook May 2023 www.biopharminternational.com Manufacturing NFF can be operated at constant pressure or constant FLUX [LMH; liter/m2/hr] and potential shear stress on the product must always be considered. Vendor, mem- brane chemistry, and pore sizes should be kept constant when scaling. Membrane area and conditioning vol- umes should be scaled linearly to process feed volume, whereby contaminants are regarded as comparable between different process runs and production scales. Stackable (i.e., parallelizable) filter elements should also be considered, as it can offer more f lexibility in usable membrane area, covering larger scales than self-contained filter capsules. Cross/tangential-flow filtration. In cross/tan- gentia l-f low f i ltration (CFF/ TFF), t he feed mate- rial is fed in a horizontal f low over the membrane; thereby, generally reducing the propensity for filter membrane fouling. The key optimization parameter is the FLUX (LMH) as a function of TMP (bar). Com- monly used formats are either hollow-fiber modules or membrane cassettes. Molecular weight cut-off (MWCO) is an important pa- rameter and should be considered carefully regarding the molecular weight of contaminants to be depleted and product particle size itself. MWCO must be kept constant for all scales. CFF is either controlled using constant trans-mem- brane pressure or permeate f low FLUX. While filter load (volume per area) is kept steady during scale-up, the cross-f low rate is often scaled up by keeping the shear rate constant. This makes sense for shear-sensi- tive products (e.g., enveloped viruses). Chromatography. Most biological production pro- cesses comprise at least one chromatography step in a bind/wash/elute or flow-through modus to enrich prod- uct and deplete process-related residuals. The most used modalities are affinity, ion exchange, size exclusion, hy- drophobic interaction, and mixed-mode resins. When scaling, the critical, linear scaling parameters are either f low velocity (cm/h) or product residence time. To keep f low velocity constant between scales, column volume should be scaled-up linearly with the volume to be loaded by keeping the column height con- stant and increasing the column diameter. Volumes for all buffers should also be scaled linearly and flux rates kept constant (column volumes/hour). Linear scale-up of column volumes is not always straightforward because columns themselves, mem- brane adsorber or monoliths units are only available in discrete sizes. Options to mitigate this problem can be to run more than one product cycle or parallelization. Scaling up by constant column volumes/hour is used as an alternative approach and might become the in- dustry standard for certain modalities (e.g., affinity, ion-exchange), helping to avoid running processes be- yond column capacity. Scale-out versus scale-up Changing the scales of an existing facility, for exam- ple, ceiling height, autoclave capacities, f loor weight loads, gas supply capacities, liquid handling size units, and seed train changes can be associated with sig- nificant challenges. Consequently, scale-out is often seen as a more viable alternative to scale-up, provid- ing manufacturing flexibility and better adaptation to market fluctuations. SUTs are particularly well-suited for this strategy. TABLE I. Key constant and scaled parameters that must be considered when scaling suspension cell cultures (5). Type of parameter (8) Parameter Description Unit Constant Scaling k L a (Volumetric O2 mass transfer coefficient) The efficiency that oxygen is delivered to the bioreactor at the set operating conditions 1/h P/V Specific power input W/m³ Vessel volumes per minute (VVM) gassing rate Volume of air sparged per unit volume of growth medium per minute Gas flow/reactor volume/minute Tip speed The velocity of the tip of the rotor m/sec-¹ Scale-independent Temperature - °F or °C %DO Percentage of dissolved oxygen % pH - - Cell seed density Number of viable cells in suspension volume Viable cells/mL Harvest time point Time at which harvest occurs Hours or days Scaled Scale-dependent Gas-flow rates - LPM Stirrer speed - RPM