Synthetic Depth Filters as a Platform Approach for Virus Filtration
Size-exclusion-based parvovirus filtration is an essential unit operation in recombinant protein and monoclonal antibody (mAb) purification. Optimizing this step enhances viral safety and process economics. This page explores using a synthetic depth filter as a prefilter for the Viresolve® Pro Device, enabling rapid process development and enhanced filter capacity. Results were generated as part of a larger collaboration with KBI BioPharma, Durham, NC, United States.
Section Overview
- Optimizing Virus Retentive Filtration with Synthetic Depth Prefiltration
- Advantages of a Virus Reduction Prefilter Platform Approach
- Prefiltration Improves Virus Filter Throughput
- Prefiltration Improves Virus Filter Flux
- Prefiltration Reduces Filtration Area Requirements
- Synthetic Depth Filters Streamline Virus Filtration
- Materials and Method
Optimizing Virus Retentive Filtration with Synthetic Depth Prefiltration
Prefilters removes protein aggregates and improve virus filter capacity. However, size-based prefilters struggle to effectively separate ~10 nm drug product from ~20 nm virus particles.1,2,3 While adsorptive membrane prefilters enable easy scale-up and minimal extractables and leachables, they have limited binding sites, narrow operating ranges and require optimization.4 Multi-mode depth filters provide wider operating windows and abundant adsorptive binding sites, but may raise concerns with extractables and leachables Synthetic depth filters offer wide operating ranges and high binding capacity while minimizing risks to product quality from organically sourced media.
Advantages of a Virus Reduction Prefilter Platform Approach
Experimental studies with any product can identify the optimum combination of prefilter, pH, conductivity, drug product concentration, and other factors that maximize the productivity of the unit operation.4 However, this approach is time-consuming and may not be practical for some biomanufacturing scenarios.
A generalized platform approach to virus filtration enables more rapid process development and enables strategies such as modular validation that can further reduce time and expense for early phase projects.5
Prefiltration Improves Virus Filter Throughput
Throughout the filtration runs product recovery, flux decay, throughput, and processing time were monitored. For all process intermediates, recovery ranged from 94% to 100% (data not shown). Figure 1 shows the flux decay curves on Viresolve® Pro Device Micro formats. For all mAbs tested, the presence of a prefilter either improved the volumetric throughput or reduced filter fouling during the run.
Figure 1.Flux decay profile for various mAbs evaluated in the presence or absence of a prefilter. The black dashed line indicates the target throughput during the evaluation of 1000 L/m2.
Prefiltration Improves Virus Filter Flux
A productivity analysis compared the average flux at 200 L/m2 volumetric loading in the presence or absence of a prefilter (Figure 2). For all process intermediates, runs with the Millistak+® HC Pro X0SP depth filter upstream of Viresolve® Pro Device Micro format resulted in higher average flux as compared to those with the Viresolve® Prefilter devices.
Figure 2.Productivity analysis of Viresolve® Pro Device in the presence or absence of a prefilter.
Figure 3.Average flux for mAb5 and mAb6 at higher volumetric throughput of 900 L/m2.
These studies show how depth filters improve Viresolve® Pro Device throughput with a range of process intermediates. Filtration runs with Millistak+® HC Pro X0SP depth filter upstream of the Viresolve® Pro Micro Device resulted in higher average flux as compared to runs with Viresolve® Prefilter devices.
Prefiltration Reduces Filtration Area Requirements
The data from these studies were used to estimate filtration area requirements for a hypothetical virus filtration process: a batch size of 1000 L, a batch time of 2 hours, a maximum allowable throughput of 1000 L/m2, and a maximum allowable flow decay of 75%.6
Table 1 shows the minimum filtration area (Amin) and expected reductions in filtration area for a hypothetical process when the Viresolve® Pro Device is used alone or with a prefilter. Prefiltration reduced filtration area requirements for many of the mAb streams.
With the Millistak+® HC Pro X0SP device, the reduction in Viresolve® Pro Device filtration area ranged from 8% - 95%, depending on the mAb. Similar results were obtained for the Viresolve® Prefilter for process intermediates that reached 75% flow decay without a prefilter. However, for process intermediates that reached the 1000 L/m2 throughput target, the benefits of the Viresolve® Prefilter runs were less obvious, likely due to the lower permeability of these depth filters.
Synthetic Depth Filters Streamline Virus Filtration
Virus filters in traditional batch processing must deliver high flux and low flux decay to achieve high throughput within minimal processing time. While the Viresolve® Pro Device alone provides high flux, with some process intermediates it may premature foul, limiting capacity. Using a depth filter, such as the Millistak+® HC Pro X0SP device, as a prefiltrer improved performance across a wide operating window of pH, conductivity, and product concentration.
Across all test conditions, depth filtration enabled the Viresolve® Pro Device to reliably reach the 1000 L/m² throughput target with low flow decay (~10–40%), greater process efficiency, and higher average flux compared to runs without prefiltration. The Millistak+® HC Pro X0SP filter’s high permeability resulted in an 8–95% reduction in virus filtration area.
These results confirm the viability of fully synthetic depth filters as broadly applicable prefilters for virus filtration. A universal synthetic prefilter offers efficiency and productivity gains across diverse molecules, shortens process development timelines, reduces costs, and minimizes risks in viral clearance studies—a benefit in CDMO settings where programs must be advanced under tight timelines. To learn more about how prefiltration technologies can provide significant increases in the performance of your viral clearance filter watch our webinar
Feedstreams
The six mAb process intermediates (PIs) used in this study were purified by capture affinity chromatography and two polishing steps. These PIs represented a wide range of conditions typical for virus filtration: 1–10 g/L, pH 5.0–8.0, conductivity 9.5–36 mS/cm, and buffer systems including Tris, sodium phosphate, sodium chloride, acetate, and citric acid and included IgG1, IgG2, IgG4, and bispecific mAbs with isoelectric points (pIs) of 7.3–8.4. All intermediates were stored below –65 °C, then thawed and filtered through a 0.2 µm bottle-top vacuum filter before processing.
Filtration
The following filters were used for these studies:
- Viresolve® Pro Devices Micro formats (effective filtration area (EFA): 3.1 cm2)
- Viresolve® Prefilter OptiScale®40 devices (EFA: 5 cm2)
- Millistak+® HC Pro X0SP NanoPod devices (EFA: 5 cm2)
Experimental runs were set up as shown in Figure 4. Viresolve® Pro Devices Micro format were flushed with water at 30 psi (15.5 mL, or 50 L/m2) with an average flow rate of 2.58 mL/min (permeability: 15.27 to 17.42 LMH/psi). A water flush was then performed on the prefilter (if applicable) at 30 psi with a minimum throughput of 100 L/m2 for Viresolve® Prefilter OptiScale® 40 devices and Milistak+® HC Pro devices (50 mL). After water flushing, flushed Viresolve® Pro Device Micro formats were connected to the prefilter outlet and a buffer flush was performed with the coupled filters (minimum throughput of 15.5 mL, or 50 L/m2 for the Viresolve® Pro Micro Device) with a permeability range of 11.83 to 20.22 L/m2/h/psi (LMH/psi). This buffer permeability measurement was used as the initial flow (J0) calculations; all filters were within the 9-25 LMH/psi recommendation for water permeability.
Figure 4.Viresolve® Pro Device experimental setup
Feed material was thawed, equilibrated to room temperature, and filtered through a 0.2 µm filter before loading into the feed vessels. Filtration was performed at 30 psi, with cumulative filtrate volume measured on a balance throughout the run. Runs were stopped at 90% flow decay, a target throughput of 1000 L/m², or depletion of feed material. At the end of each run (except select runs without a prefilter), the system was depressurized, residual feed was replaced with process buffer, and a 50 L/m² recovery flush (15.5 mL for the Viresolve® Pro Device Micro formats) was collected under re-pressurization at 30 psi. All post-use flushes were pooled with bulk filtrate. Protein concentrations of feed and filtrates were determined by A280 absorbance to calculate step yield for each run.
Cumulative filtrate and time measurements were used to calculate filter flux (LMH) over each run. Results were analyzed through normalized flux decay (J/J0), using buffer permeability as the basis for initial flux. In a manufacturing setting, batch time and process efficiency are also key metrics; therefore, average flux (LMH) was calculated for each run at 200 L/m² volumetric loading.
Related Products
References
To continue reading please sign in or create an account.
Don't Have An Account?