Subcutaneous (SC) administration of monoclonal antibodies (mAbs) has several benefits over intravenous (IV) infusion including being less invasive, requiring shorter administration times, and the possibility for self-administration. However, to achieve the dosage needed within the volume limitations of SC administration, the formulation must be highly concentrated, often over 100 g/L and sometimes upwards of 200 g/L. Here we demonstrate, using only our products, how you can achieve a final mAb concentration of over 200 g/L whilst maintaining key quality attributes.
Introduction
Production of a highly concentrated monoclonal antibody requires many process steps to work in harmony to deliver a suitable quality bulk drug substance. It also may require many products and pieces of equipment often from various vendors. However, this can cause difficulties due to different lead times for equipment and materials.
In this study, we ran a complete mAb purification process using only our products at a 50 L scale as outlined in Figure 1. A few steps were investigated to a greater extent, including:
- Midstream clarification using two different depth filter trains for comparison.
- Running a viral filtration step using Pegasus™ Protect and Pegasus™ Prime virus filters in combination with ÄKTA pilot™ 600 chromatography system.
- Formulation of the mAb up to 200 g/L on traditional tangential flow filtration (TFF) in combination with a single-pass tangential flow filtration (SPTFF) step.
- Final 0.2 µm filtration on Supor™ Prime sterilizing grade filters.
Fig 1. An overview of the mid- and downstream purification process.
Materials and methods
Cell culture
A fed batch culture using HyClone™ ActiPro™ cell culture media, HyClone™ Cell Boost™ 7a supplement, and HyClone™ Cell Boost™ 7b supplement was run at a 50 L scale in an Xcellerex™ XDR-50 bioreactor to produce a monoclonal antibody.
Clarification using depth filters
Prior to harvest, the Supracap™ 50, Supracap™ 100, and Stax™ depth filter capsules were rinsed with DI water and primed with PBS. They were then drained to minimize product dilution. At the time of harvesting the 50 L fed batch cell culture, a 2.5 kg portion was removed for bench-scale evaluation (Fig 2).
Fig 2. Depth filtration setup showing pre-study and main bulk processing trains for primary and secondary clarification steps.
Train one consisted of two PDP8 Supracap™ 50 depth filter capsules run in parallel for primary filtration (44 cm2) followed by a PDE2 Supracap™ 50 depth filter capsule for secondary filtration (22 cm2). Train two mimicked the capsule staging of train one but used PDK7 and PDCX media for primary and secondary filtration, respectively.
The remainder of the 50 L cell culture was split and processed using larger capsules. Train one consisted of a PDP8 Stax™ M depth filter capsule (0.5 m2) followed by a PDE2 Stax™ S capsule (0.25 m2). Train two consisted of a PDK7 Stax™ M capsule followed by two PDCX 20 in. and one PDCX 10 in. Supracap™ 100 filter capsules run in parallel (0.25 m2). The selected flux was 50 LMH for the primary depth filters and 100 LMH for the secondary depth filters. The primary and secondary clarifications were run separately to allow for sampling and turbidity measurement in between.
After the depth filtration, the product was 0.2 µm filtered through Mini Kleenpak™ 20 capsules with Supor™ EKV membrane (20 cm2) for the depth filtrate from the Supracap™ 50 depth filters and Kleenpak™ capsules with Supor™ EKV membrane (1500 cm2) for the depth filtrates from the main bulk.
Affinity capture chromatography and viral inactivation
The clarified depth filtrates were pooled and captured on a 0.8 L ReadyToProcess™ MabSelect PrismA™ chromatography column on an ÄKTA pilot™ 600 chromatography system over four cycles. The load was set to 50 g/L. The chromatography method used is shown in Table 1 below.
Table 1. Capture MabSelect PrismA™ chromatography method
Step | Column volume (CV) | Buffer/material | Flowrate (residence time) |
---|---|---|---|
Equilibration | 3 | 20 mM sodium phosphate (NaP), 500 mM NaCl, pH 7.0 | 4 min |
Load | N/A | Clarified harvest mAb | 6 min |
Wash 1 | 1.5 | 20 mM NaP, 500 mM NaCl, pH 7.0 | 6 min |
Wash 1 | 3.5 | 20 mM NaP, 500 mM NaCl, pH 7.0 | 4 min |
Wash 2 | 1 | 50 mM sodium acetate, pH 5.5 | 4 min |
Elution | 3 | 100 mM sodium acetate, pH 3.5 | 6 min |
Strip | 1 | 100 mM acetic acid (HAc) | 4 min |
Clean-in-place (CIP) | 3 | 500 mM NaOH | 5 min |
Re-equilibration 1 | 1.5 | 20 mM NaP, 500 mM NaCl , pH 7.0 | 6 min |
Re-equilibration 2 | 1.5 | 20 mM NaP, 500 mM NaCl, pH 7.0 | 4 min |
The eluate pool (at a titer of 34 g/L and 3.9L) was viral inactivated for 60 min at pH 3.7 using 4 M acetic acid. The pH was then adjusted to 5.0 with 2 M Tris base and the product solution was 0.2 µm filtered on Supor™ EKV sterilizing grade filter 1500 cm2. The product volume was 4.5 kg and had a titer of 27.9 g/L after viral inactivation and filtration.
Polishing chromatography on Capto™ S ImpAct resin
The viral inactivated (VI) mAb product was polished using a 0.8 L ReadyToProcess™ Capto™ S ImpAct chromatography column on an ÄKTA pilot™ 600 chromatography system over 3 cycles. The load was set to 70 g/L. The chromatography method used is shown in Table 2.
Table 2. Polishing method using Capto™ S ImpAct chromatography columnStep | Column volume (CV) | Buffer/material | Flowrate (residence time) |
---|---|---|---|
Equilibration | 5 | 50 mM sodium acetate, 50 mM NaCl, pH 5.2 | 5.4 min |
Load | N/A | Adjusted VI pool, pH 5.2 ± 0.2 | 5.4 min |
Wash | 5 | 50 mM sodium acetate, 50mM NaCl, pH 5.2 | 5.4 min |
Elution | 7 | 50 mM sodium acetate, 200 mM NaCl, pH 5.2 | 5.4 min |
Strip | 3 | 1 M NaCl | 5.4 min |
CIP | 3 | 0.5 M NaOH | 5.4 min |
Re-equilibration | 5 | 50 mM sodium acetate, 50 mM NaCl, pH 5.2 | 5.4 min |
Product conditioning and polishing chromatography on Mustang™ Q XT chromatography membrane capsules
The Capto™ S ImpAct chromatography resin eluate pool (3.6 kg at a titer of 33.4 g/L) was conditioned to pH 6.0 and a conductivity below 11 mS/cm using a 100 mM sodium acetate, pH 8.3. The product weight was 15.4 kg with a titer of 7.8 g/L after conditioning.
A second polishing step was conducted on Mustang™ XT 5 chromatography membrane using an ÄKTA pure™ 150 chromatography system over 12 cycles in flowthrough mode. The load was set to 2000 g mAb/L. The chromatography method used is shown in Table 3.
Table 3. Polishing method using Mustang™ Q XT chromatography membraneStep | Column volume (CV) | Buffer/material | Flowrate (MV/min) |
---|---|---|---|
Equilibration | 20 | 50 mM sodium acetate, pH 6.0 | 10 |
Load | N/A | Adjusted Capto™ S ImpAct eluate | 10 |
Wash | 10 | 50 mM sodium acetate, pH 6.0 | 10 |
Strip | 5 | 1 M NaCl | 10 |
CIP | 5 + 30 min hold | 1 M NaOH | 10 |
Re-equilibration | 10 | 50 mM sodium acetate, pH 6.0 | 10 |
Virus filtration
The flowthrough eluate pool from Mustang™ Q XT chromatography membrane was 0.2 µm filtered using Supor™ EKV sterilizing grade filter 1500 cm2 (total weight of 15.5 kg and a titer of 7.8 g/L), and then applied to a filter train consisting of Pegasus™ Protect prefilter and Pegasus™ Prime virus filter connected to an ÄKTA pilot™ 600 chromatography system. The filter train was primed with the Mustang™ Q XT chromatography membrane equilibration buffer prior filtering the product.
The filtration was performed as a manual run at an initial flow of 50 mL/min. After observing a steady delta filter pressure, the filtration was switched over to pressure/flow control and the pressure limit was set to 2.5 bar (0.25 MPa, 36 psi). UNICORN™ software will not go over 80% of the pressure limit meaning that the maximum allowed pressure was 2.0 bar (0.20 MPa, 29 psi). The flow was set to 110 mL/min and the system maintained a steady delta filter pressure, while the flow decreased over time.
After the virus filtration, the product weight was 15.6 kg and the titer was determined to be 7.5 g/L.
Final formulation through TFF and extended final concentration using SPTFF
The virus filtered mAb product solution was concentrated using a T-series cassette with 30 000 Mr Omega™ PES membrane 0.3 m2 on an ÄKTA flux™ 6 filtration system. The recirculation flow rate was set to 360 LMH and initially the TMP was set to 0.7 bar (0.07 MPa, 10 psi). The initial concentration was performed until the mAb titer was approximately 40 to 50 g/L. Then diafiltration (DF) was started using 20 mM sodium acetate at pH 5.0 as the final formulation buffer. The DF was performed for 6 turn-over volumes (TOV). The entire UF/DF process took several hours. At the end of the process the TMP had increased to 1.5 bar (0.15 MPa, 22 psi). During the concentration after the DF, we aimed for as high a concentration of the mAb as possible, but not exceeding the volumetric limit of the recirculation vessel of the ÄKTA flux™ 6 system with the risk of generating foam. When the volume was close to 500 mL in the ÄKTA flux™ 6 system product vessel, we decided to end the TFF step, to prevent the risk of product foaming due to a too low volume for recirculation. The retentate and the recovery flush had a titer of 135 g/L and 22.6 g/L, respectively.
The TFF product pool at a titer of 81.5 g/L was run on an 8-in-series SPTFF using 30 000 Mr Delta membranes. The average feed flow rate was 2.7 mL/min and the average feed pressure was about 3.3 bar (0.33 MPa, 48 psi). After the SPTFF step, the product solution was concentrated 3 times. The SPTFF unit was flushed out with 32 mL of formulation buffer for product recovery. The flush volume was collected and analyzed separately. The titer of the concentrated mAb was 218 g/L after the SPTFF step and the step yield was above 100%.
Filtration using Supor™ Prime membrane of the final high-concentration mAb product
The final mAb product with a concentration of 218 g/L was 0.2 µm filtered on Supor™ Prime sterilizing grade filters. The volume of the final product was approximately 360 mL. The filter capacity was determined using a 3 cm2 Supor™ Prime filter. The capacity was determined to be 390 L/m2 when filtering at a flux of 400 LMH at a max pressure setting of 2.0 bar (0.20 MPa, 29 psi). The remaining bulk was filtered on a 240 cm2 Supor™ Prime filter. The final product had a titer of 216.7 g/L.
Results and discussion
Cell culture
The cell culture used for this study was a fed batch CHO cell culture. At the point of harvest, the viability was at 90%, the mAb titer was 3.8 mg/mL, the maximum cell density reached 36 × 106 cells/mL, and the turbidity at harvest was 2100 FNU.
Clarification
Two different depth filter trains were evaluated for this clarification. The first was a Stax™ mAx approach consisting of a primary PDP8 filter and secondary PDE2 filter, designed specifically for mAb harvest from CHO culture. The second train consisted of a primary PDK7 filter and secondary PDCX filter. Both of these grades are slightly tighter than their Stax™ mAx counterparts and were included to look for improved fine particle removal from this challenging high cell density feed stream.
Bench-scale capsules were used for both filter trains to determine capacity. Pressure-driven capacity was defined here as throughput at a differential pressure of 1.0 bar (0.1 MPa, 15 psi). The depth filtration throughputs from the pre-study are shown in Figure 3. As expected, the PDP8 plus PDE2 combination demonstrated a higher throughput (approximately 90 L/m2 and > 180 L/m2, respectively) compared to the PDK7 plus PDCX (approximately 85 L/m2 for both).
The turbidity from the depth filtrations, both the smaller pre-study and the main bulk is shown in Figure 4. As expected, the tighter depth filter train of PDK7 plus PDCX generated a filtrate pool with a lower turbidity (3 to 4 NTU) compared to that of the PDP8 plus PDE2 train (9 to 13 NTU). We did observe lower stage one filtrate turbidities from the Stax™ capsules compared to the Supracap™ 50 capsules, however, this was not surprising as the Stax™ capsules were loaded substantially lower.
a)
c)
b)
d)
Fig 3. Depth filtration throughput for pre-study using a) PDP8 plus b) PDE2 or c) PDK7 plus d) PDCX.
Fig 4a. Pre-study depth filtrations on Supracap™ 50 format for train 1 (green) and train 2 (orange).
Fig 4b. Main bulk depth filtration on Stax™ and Supracap™ 100 format for train 1 (green) and train 2 (orange) and Supor™ EKV filtrations.
Affinity capture chromatography, viral inactivation, and polishing chromatography
The capture chromatography using MabSelect PrismA™ resin, as well as, both polishing steps on Capto™ S ImpAct resin, and Mustang™ Q XT membrane performed as expected and demonstrated the suitability of running Mustang™ Q XT membranes on an ÄKTA pure™ 150 system during process development. Example of the chromatograms are shown in Figure 5.
The viral inactivation step at low pH also performed as expected. It is not uncommon to see a slight increase in product cloudiness when increasing the pH after the low pH viral inactivation. We observed a moderate increase in cloudiness. It is not uncommon to include a depth filtration step after the viral inactivation to remove the precipitated impurities. In this case we managed to filter directly using a 0.2 µm Supor™ prime filter 1500 cm2 and the differential pressure during the filtration increased to 0.6 bar (0.06 MPa, 8.7 psi).
(A)
(B)
(C)
Fig 5. Chromatograms from (A) MabSelect Prisma™ chromatography resin, (B) Capto™ S ImpAct chromatography resin, and (C) Mustang™ Q XT chromatography membrane capsules.
Virus Filtration
A virus filtration step is traditionally performed at constant pressure, that is, through pressurizing a stainless-steel vessel with the product solution being transferred through tubing into the virus filter. However, we have shown that virus filtration can be performed using an ÄKTA pilot™ 600 chromatography system with a filter train consisting of Pegasus™ protect prefilter and Pegasus™ prime virus filter.
The chromatogram from the virus filtration is shown in Figure 5. It shows how the ÄKTA pilot™ 600 chromatography system with a UNICORN™ software method can maintain a steady delta "column" pressure throughout virus filtration and adjust the flow rate accordingly. Some air was introduced after 12 L had been applied to the filter and had to be removed manually through the vent ports of the filters. After removal of the air filtration continued as expected.
Fig 6. Chromatograms from the virus filtration on ÄKTA pilot™ 600. The bumps in the curves were due to trapped air.
Final TFF formulation and final SPTFF concentration
During the final formulation we showed that a mAb solution can be formulated on T-series cassettes with Omega™ PES membrane of 30 kDa up to about 135 g/L using an ÄKTA flux™ 6 chromatography system. Following the TFF concentration we made a final concentration up to 216.7 g/L using a stand-alone SPTFF operation with a Cadence™ in-line concentrator module with Delta membrane of 30 000 Mr. However, we believe that the SPTFF process can be further optimized to improve flow and reduce feed pressure compared to this first-time experiment with this mAb on SPTFF. A unit with only 4-in-series Delta membranes may have been sufficient.
The final concentrated product was filtered through a 0.2 µm Supor™ prime filter with a high throughput capacity of 390 L/m2 as seen in Figure 6. A dip in pressure was observed due to a short opening of the pump head. In total, including after the pressure drop, the product was filtered at an average flux of 369 LMH. The peristaltic pump and tubing used precluded filtration at higher pressures. The remaining solution not filtered with the small capsule was filtered with the medium capsule.
Fig 6. Supor™ Prime filtration throughput versus pressure.
Analytical Results
The mAb production process was monitored using several different analytical methods. Analytical results are listed in Table 4. Charge variants were monitored over the process, but the distribution of charge variants did not change significantly over the process.
Table 4. Analytical results for the final bulk productAssay | Result | Acceptance criteria |
---|---|---|
HCP (ng/mg drug product) | 2.5 | < 100 |
DNA (pg/mL) | < 0.03 | < 10 ng/dose for mAbs US FDA |
Aggregates (%) | 3.6 | mAb dependent |
Charge variants (% main) | 40 | Consistent profile |
Concentration (g/L) | 218 | Not available |
Protein A (ng/mg drug product) | 5 | Not available |
Conclusions
In conclusion, we have shown a complete mAb production process was run demonstrating the interconnection of our products. In addition, we have shown:- Two different depth filter trains were tested as primary and secondary clarification. The PDP8 plus PDE2 option showed higher throughput and the PDK7 plus PDCX option gave the lowest filtrate turbidity. Both options work well for a high density fed batch cell culture.
- Successful demonstration of running Mustang™ Q XT chromatography membrane capsules on an ÄKTA pure™ 150 chromatography system.
- Successful demonstration of running virus filtration using Pegasus™ Protect and Pegasus™ Prime as a filter train connected on an ÄKTA pilot™ 600 chromatography system. The system managed to keep a constant pressure by decreasing the flow rate as the filter flow decayed over time.
- Successful demonstration of concentrating a mAb up to 135 g/L using an ÄKTA flux™ 6 system in combination with three 0.1 m2 T-series cassettes with 30 000 Mr Omega™ polyethersulfone (PES) membrane.
- Further concentration of mAb to above 200 g/L on an SPTFF unit.
- The aggregate level of 3.6% was high for a mAb solution but expected given that no optimization of the formulation buffer was performed.
- The final filtration on Supor™ prime membrane was successful and showed a high capacity for high-concentration mAb solutions at 390 L/m2.
CY46012-xxXXX24-AN