A Solution for Aggregate Removal in Engineered mAbs

Results—DoE study Finding optimal conditions A Design of Experiment (DoE) study was used to evaluate the optimum mobile phase pH and conductivity to achieve monomer yield > 80% and reduction of aggregate levels to < 2%. The design space: pH range 4.5–6.0, [NaCl] from 0–500mM. Load density was kept constant at 100mg / mL resin. Abstract With advances in engineered antibody designs, treatment performance improves, but higher aggregate levels are often produced in the cell culture creating new purification challenges. Current solutions for aggregate removal include bind and elute strategies with cation exchange or hydrophobic interaction chromatography resins which, whilst effective, often result in poor process economics and low recoveries. Alternatively, caprylic acid has been successfully used as a flocculant for antibody aggregates but requires a filtration step resulting in a more labor intensive and complicated process. The work in this poster describes the performance of a resin-based approach using immobilized Caprylic acid. It effectively removes high levels of aggregates as well as leached ligand from protein A affinity resin and 301 out of 380 HCPs identified using HPLC-MS/MS2 including those that are either high-risk and/or challenging to remove. Conclusions Simulated high aggregate levels in our mAb test solution has shown that POROS Caprylate Mixed-Mode resin operated in flowthrough mode, is very promising for  Effective removal of high (10%) aggregate levels in mAbs using flow-through mode  Delivering high monomer yields (> 80%) with low aggregate impurity levels (< 2%)  Improved mAb purification process designs, were flow through can be used for the cation exchange step and the anion exchange-based final polishing step  The economics of a such intensified process design can be highly advantageously for existing and new modalities Ying Chen, Al de Leon, Kelly Flook, Thermo Fisher Scientific, Bedford, MA 01730 USA An innovative approach to addressing high aggregate challenges in engineered monoclonal antibodies % Aggregate Loading density (g/L resin) Monomer recovery (%) 1% 85.6 80.4 2% 181.9 96.3 3% 256.8 99.2 Learn more at thermofisher.com/purification-contact Materials and methods Sample Preparation A IgG1 type mAb was produced in-house and purified using Thermo Scientific MabCaptureA affinity resin. In order to mimic high aggregate levels, the mAb was then stressed through multiple exposures to high and low pH adjustments, until the aggregate level reached approximately 10%. [2] Hydrophobic Weak Cation Exchange O OH O Figure 1: POROS beads and Caprylic Acid form a mixedmode, hydrophobic weak cation exchange resin—POROS Caprylate Mixed-Mode Cation Exchange resin Purified mAb was then applied to 1mL POROS Caprylate Mixed-Mode resin packed into OmniFit glass column (6.6mmID x 30 mmL). HPLC-SEC was performed with a Thermo Scientific MabPac SEC-1 on Thermo Scientific UltiMate 3000. Buffer: 50mM Sodium Phosphate, 300 mM NaCl, pH 6.5; flow rate: 0.2 mL/min; detection: UV at 280nm. HCP and Protein A ligand leach was performed with Cygnus CHO Host Cell Protein ELISA-kit and Repligen Protein A ELISA-Kit, respectively. Materials and methods (continued) To simulate an antibody with high levels of aggregate, a biosimilar version of Herceptin was produced and subjected to pH cycling to induce aggregate formation. Aggregate levels and monomer purity were determined using size exclusion chromatography (SEC) POROS Caprylate resin— Aggregates, Impurity Clearance (HCP, LPrA) Monoclonal Antibody (Herceptin ) Affinity Capture Select Resin (MabCapture A ) Low pH hold, Depth Filtration Viral inactivation Aggregation production Generated ~10% HMW aggregates Clarified cell culture fluid Protein A resin selectively interact with Fc region of antibody Virus inactivation Aggregate induction by pH cycling, monitored by HPLC analysis using Thermo Scientific MAbPac SEC-1 Evaluated over a broad range of pH and conductivities to maximize impurity removal . Figure 2: Schematic of sample generation, aggregate induction and resin performance test. HPLC – SEC used for aggregate level determination on mock-up feed solution Retention time, mins 0 50 100 150 200 250 Absorbance, 280nm mAU 1 2 3 4 0.00 2.50 7.50 10.00 12.50 15.00 17.50 20.00 22.50 25.00 0 40 6.5 18 5.00 Figure 4: SEC chromatograph of mAb feed prior to purification using POROS Caprylate resin (blue) and after (black). Inset is an expanded section of high molecular weight species. The DoE study centered around conditions found favorable in previous flow-through experiments and qualitative, wellplate based HTS screening. The pH range was chosen from 4.5 to 6.0, the NaCl concentration from 0 to 500 mM. Figure 5: Design Space, [NaCl] and pH vs. monomer and aggregate response Even with lower conductivity conditions, POROS Caprylate MixedMode resin is able to reduce aggregate levels down to 1–2% This option is favorable for a directly following low salt anion exchange process step in the overall polishing process. As the AEX polishing is also often run in flow-through mode, the suggested savings pull through then at that step as well (lower buffer consumption, lower COGS, smaller column sizes, fast high yield break through). Figure 6: Design space for monomer vs. aggregate percentage Results—Load density study Conditions used for load density study Feed: Max loading: 325 g/L resin Monomer Purity: 89.4% % Aggregate: 10.6% Buffer & Residence Time: Sodium Acetate pH 5.25 275mM NaCl (28.62 mS/cm) Residence Time: 3 min Figure 7: Monomer recovery (dark blue) vs aggregate accumulation (orange), with aggregate levels marked for 1%, 2% and 3% 0 1 2 3 4 5 6 7 8 9 10 0 20 40 60 80 100 0 50 100 150 200 250 300 350 % Aggregate Cumulative Monomer Recovery (%) Loading Density (g/L resin) Result show very favorable monomer yield for the given aggregate impurity levels. Table 1: Loading density and monomer recovery at assigned aggregate impurity levels Results—Reduction of other HWMS POROS Caprylate Mixed-Mode resin is also effective in removing other high molecular weight species (HMWS), like host cell proteins (HCP) or leached Protein A resin ligand. Parameter Unit Loading density experiment R&D Batch A Loading density Experiment R&D Batch B Production Validation Batch MMCEX-001 Total load [mg] 160 175 100 Buffer conditions 25mM sodium acetate 275mM NaCl, pH 5.25 25mM sodium acetate, 75mM NaCl, pH 5.30 25mM sodium acetate, 12mM NaCl, pH 4.5 Host cell protein in load [ppm] 555 450 648 Host cell protein after column [ppm] 24 14 36 Leached protein A in load [ppm] 60.3 67.5 78.5 Leached protein A after column [ppm] 3.1 4.7 1.3 Text System 1mL CV Omnifit column 6.6mm ID x 300mmL residence time 3 minutes Table 2: HCP & Leached Protein A ligand reduction, 3 different experiments/conditions Trademarks/licensing © 2024 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher Scientific and its subsidiaries unless otherwise specified. This information is not intended to encourage use of these products in any manner that might infringe the intellectual property rights of others. EXT6169 1223 References 1.Brodsky Y, Zhang C, Yigzaw Y, Vedantham G. CaprylicVan, Biotechnol Bioeng. 2012 Oct 2.Stress-Induced Antibody Aggregates, Ajish SR Potty and Alex Xenopoulos, p44 ff, BioProcess International 11(3) March 2013, BioProcess Intended use statement POROS resins: Pharmaceutical Grade Reagent. For Manufacturing and Laboratory Use Only. Introduction With the need of designing therapeutics with higher efficacy, more engineered monoclonal antibody derivatives are actively pursued for the next generation of mAb-based drugs. With the more complex structures, like symmetric, asymmetric or fragmentbased bispecifics, the downstream process developer is challenged by mis-paired products, undesired fragments and higher levels of aggregates. Alternative new mAb designs are equally challenging. The use of caprylic acid as a flocculant for aggregate removal and high molecular weight species has been earlier suggested by Brodsky et al.[1] in 2012. The precipitation step though requires to introduce additional filtration and sedimentation steps. By chemically attaching caprylic acid (octanoic acid) to large pore POROS divinylbenzene polymeric beads, a chromatography resin with excellent aggregate removal capabilities was developed. The work described here tests the final design of the resin on loading, aggregate elimination and also best operational conditions (for our simulated mAb high aggregate test solution). Results—DoE study (continued) The 2-dimensional representation of the design space below show relatively large design conditions for high yield and purity expectations.. HCP Classification Identified HCP IgG-1 after Protein A purification IgG-1 after POROS Caprylate Purification High Risk 8 kDa glucose regulated protein(GRP78, BiP) 6.64E+05 n.d. Alpha-enolase (2-phospho-Dglycerate hydro-lyase) 2.43E+04 n.d. Cathepsin B (CatB) 1.00E+06 n.d. Cathepsin L (CatL) 4.16E+04 n.d. Cathepsin Z (CatZ) 7.52E+04 n.d. Glutathione S-transferase P 1 (GSTP1) 4.06E+05 n.d. Lysosomal Acid Lipase (LAL) 2.67E+05 n.d. Matrix metalloproteinase-19 (MMP19) 2.08E+05 n.d. Phospholipase B-like 2 (PLBL2) 1.67E+05 n.d. Monocyte Chemoattractant Protein1 (MCP-1) 1.72E+06 1.02E+05 Peroxiredoxin-1 (PRDX1) 4.20E+05 1.12E+05 Challenging to Remove Cathepsin D 8.43E+04 n.d. Insulin-like growth factor-binding protein 4 7.46E+04 n.d. metalloproteinase inhibitor 2.08E+05 n.d. galectin-3-binding protein 2.15E+05 3.31E+04 lipoprotein lipase 2.72E+06 7.42E+05 High Risk and Challenging Clusterin (CLU) 2.24E+07 1.56E+06 Table 3: Identification of HCP and relative quantification (total ion count) before and after POROS Caprylate purification Figure 3: HCP characterization and relative quantification using HPLC MS-MS/MS method