Inline Flocculation for Harvest and Perfusate Clarification
0 likes265 views
The document outlines a presentation on inline flocculation for harvest and perfusate clarification by Merck KGaA, discussing challenges faced with high cell density in bioprocessing. It details experimental plans and results showing that inline flocculation using pdadmac is feasible with a two-minute residence time, having no significant impact on yields or impurity levels. The findings suggest that this method could provide advantages over traditional batch flocculation in terms of efficiency and process reliability.
Inline Flocculation for Harvest and Perfusate Clarification
- 1. Merck KGaA Darmstadt, Germany Akshat Gupta, Sr Applications Engineer, MSAT Applications Engineering Elizabeth Goodrich, Director of MSAT Applications Engineering Harvest Industry Summit, 9 May 2018 Inline Flocculation for Harvest and Perfusate Clarification
- 2. This document provides an outline of a presentation and is incomplete without oral commentary and discussion. Results are intended as general examples and are not to be construed as product claims or specifications. The results indicated in this presentation summarize outcomes and observations obtained in application studies conducted with the specific model streams and detailed experimental conditions. Therefore, all test results should be confirmed by the end user using feed stream and process conditions representative of the specific application. Disclaimer
- 3. 01 02 03 04 05 Agenda 3 Introduction Inline Flocculation System Experimental Plan Results Summary
- 4. Increased product titers generally obtained with higher cell density 1. Higher cell density results in: • Increased cells and cell debris • Increased soluble impurities (DNA, HCP) 2. Results in less efficient primary clarification (centrifugation, primary depth filter) 3. Reduced throughput on downstream filters (secondary depth filters, sterile filters) Introduction Challenges with high cell density
- 5. Introduction Flocculation - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Electrostatic patching Dao, V.H., et al., Synthesis, properties and performance of organic polymers employed in flocculation applications. Polym. Chem. 2016, 7, 11. Czemierska, M., et al., Purification of wastewater by natural flocculants. BioTechnologia. 2015, 96, 272-278. - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + - - - - - - - - - - - - - - - + + + + + + + + + - - - - - - - - - - - - - - - + + + + + + + + + - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Charge neutralization Polymer bridging Flocculation Mechanism Traditional Clarification
- 6. • Advantages of flocculation over traditional clarification has been well established. - Increased throughput - Reduced turbidity - Enhanced impurity removal Introduction Flocculation vs. traditional clarification Filter Loading at 10 psi (L/m2) Pool turbidities (NTU) Millistak+® D0HC depth filter 54 8.56 Millistak+® X0HC depth filter 202 pDADMAC flocculation & Clarisolve® 40 MS depth filter 293 3.37
- 7. Introduction Batch flocculation • Flocculant is directly added to the bioreactor or an equivalent sized mixing vessel • Mixing is performed for predefined mixing time • Clarification is performed using Clarisolve® depth filter or other suitable clarification technology. Mixing / incubation (~30 minutes) Flocculant addition (15-30 minutes) Clarification
- 8. ▪ Prior art shows pDADMAC inline cell flocculation can be carried out using static mixer and tubular reactor with residence time 13.2 minutes. ▪ Successful depth filter switching based on differential pressure demonstrated. ▪ Advantages of inline cell flocculation ▪ Reduced reliance on large scale mixing systems ▪ Reduced cleaning validation requirements ▪ Reduced mixing time requirement Introduction Inline flocculation P Burgstaller, D., Krepper, W., Haas, J., Maszelin, M., Mohoric, J., Pajnic, K., Jungbauer, A. and Satzer, P. (2017), Continuous cell flocculation for recombinant antibody harvesting. J. Chem. Technol. Biotechnol.
- 9. ▪ Linear relationship between residence time and length of the reactor ▪ Design for 2000 L bioreactor Expected loading capacity 350 L/m2 Number of switching 1 (two banks of depth filter) Area installed in each bank 3 m2* Feed flux 50 LMH Feed flowrate 2.5 lpm Introduction Impact of residence time P Switching *Area round up from 2.86 m2 Tube diameter (inches) 2 minutes 13 minutes 1 inch 32 ft 210 ft 1.5 inch 14 ft 94 ft
- 10. ▪ Perfusion cell culture is being adopted for semi continuous and continuous manufacturing ▪ Common perfusion technologies include centrifugation, alternating tangential flow (ATF) filtration, tangential flow filtration ▪ All methods rely on mechanical retention methods ▪ General trend indicates adoption of more open membranes to improve product passage ▪ No absorptive clearance method in place for impurity removal. Introduction Perfusate clarification P
- 11. ▪ Multicolumn cycling approach for protein A being adopted for continuous capture process ▪ Due to the nature of continuous cycling system design, the resin is in contact with significantly more foulants. ▪ Literature indicates steeper loss in dynamic binding capacity in case of continuous chromatography versus batch process. ▪ Premature loss in dynamic binding capacity results in ▪ Reduced productivity ▪ Operational challenges ▪ System challenges ▪ Poor process economics Introduction Perfusate clarification James Pollock, Glen Bolton, Jon Coffman, Sa V. Ho, Daniel G. Bracewell, Suzanne S. Farid, (2013), “Optimising the design and operation of semi-continuous affinity chromatography for clinical and commercial manufacture” Journal of Chromatography A Continuous Batch
- 12. 01 02 03 04 05 Agenda 12 Introduction Inline Flocculation System Experimental Plan Results Summary
- 13. ▪ Feed Harvest: Internal CHO cell culture harvest 17e6 cell/ml Flocculating agent: pDADMAC 10% solution dosed at 30pg/cell ▪ Mixing System Top mount impeller centered 0º inclination - Three blade marine impeller - Servodyne mixer controller - 3L Polypropylene vessel ▪ Particle size distribution - Mettler Toledo Particle track G400 Inline flocculation system Batch experiments
- 14. ▪ At 20 W/m3 power to volume ratio uniform particle distributions are achieved within 2 minutes. ▪ If mixing is stopped, aggregation of flocculates continue resulting in generation of larger flocculates. These large flocculates may result in inconsistent depth filter capacities. Inline flocculation system Batch experiments - results ▪ Under optimized mixing condition, two minutes residence time is sufficient for flocculation ▪ Critical to impart mixing during residence time
- 15. ▪ Experimental setup - Peristaltic pump: Raw harvest - Syringe pump: Flocculant - Static mixer 16 elements - Helical coil reactor for residence time - Clarification with Clarisolve® depth filter - Data acquisition DAQ 2.0 / turbidity Inline harvest flocculation system Experimental Setup P1 Static mixer Helical coil retention chamber Flocculant Clarisolve® 20 MS/ 40 MS depth filter
- 16. ▪ Experimental setup - Peristaltic pump: Raw harvest - Peristaltic pump: Flocculant - Static mixer 16 elements - Helical coil reactor for residence time - Clarification with Millistak+® HC Pro X0SP/ C0SP - Data acquisition DAQ 2.0 / turbidity Inline perfusate flocculation system Experimental Setup P1 Static mixer Helical coil retention chamber Flocculant Millistak+® HC Pro X0SP/ C0SP depth filter
- 17. ▪ Flowrates and dimensions of helical coil retention chamber were kept constant except for length to keep Reynolds number and Dean’s number constant. ▪ Depth filter feed flux was kept constant at 150 LMH ▪ Reynold’s number ▪ Dean number Inline harvest flocculation system Helical coil retention chamber Stephan Klutz, Safa Kutup Kurt, Martin Lobedann, Norbert Kockmann, Narrow residence time distribution in tubular reactor concept for Reynolds number range of 10–100, Chemical Engineering Research and Design, Volume 95, 2015, Pages 22-33, Static mixer Helical coil retention chamber Inner Tube Diameter (di) 0.31cm Coil Diameter (dc) 3.175 cm Pitch distance 0.635cm Re= 𝑑𝑖.𝑣.𝜌 𝜇 = 32 D𝑛 = 𝑅𝑒 𝑑𝑖 𝑑𝑐 = 9.51 Dynamic viscosity and density of cell culture media containing 10 % Fetal Bovine Serum were used ρ =1006 Kg/m3, µ= 0.00113 Pa.s
- 18. 25ft 12.5ft 5ft Inline harvest flocculation system Experimental setup
- 19. 01 02 03 04 05 Agenda 19 Introduction Inline Flocculation System Experimental Plan Results Summary
- 20. ▪ Cell culture fluid (CCF) – Chinese Hamster Ovary (CHO) cell culture fluid TCD 14.84e6 cell/ml Viability 60-75% ▪ Flocculant: 1% v/v pDADMAC solution in water 10% pDADMAC Stock solution Cat No # 1.37069.0100 prediluted to 1% ▪ Depth Filter: Clarisolve® 40MS ▪ Experimental Matrix Experimental Plan Inline harvest pDADMAC flocculation Process/ addition Residence time (minutes) pDADMAC dosing Batch (- Control) NA 0 pg/cell Batch (+ control) 30 minutes 30 pg/cell Inline 10 minutes 30 pg/cell Inline 5 minutes 30 pg/cell Inline 2 minutes 30 pg/cell
- 21. ▪ Cell culture fluid (CCF) – Chinese Hamster Ovary (CHO) cell culture fluid TCD 14.84e6 cell/ml Viability 70.38% ▪ Flocculant: 2.42M Acetic acid stock solution ▪ Target pH 4.5 ▪ Depth Filter: Clarisolve® 20MS ▪ Experimental Matrix Experimental Plan Inline harvest acid precipitation Process/ addition Residence time (minutes) 2.42M Acetic acid dosing Batch (+ control) 30 minutes 4.4% Inline 10 minutes 4.4% Inline 5 minutes 4.4% Inline 2 minutes 4.4% Additional experiment was performed to compare the filtration performance of acid precipitation With 8.7M Acetic acid and 2.42M acetic acid
- 22. ▪ Cell culture perfusate fluid (CCF) – CHO cell culture EX-CELL® Advanced, TCD 40e6 cell/ml Viability 60-80% ▪ Perfusion: Alternating tangential flow 0.22µm, 0.11 m2 ▪ Flocculant: 2.5M Acetic acid stock solution ▪ Target pH 4.5 ▪ Depth filter: Millistak+® HC Pro X0SP and C0SP ▪ Residence time : Inline 2 minutes Batch: 30 minutes ▪ Experimental Matrix Experimental Plan Inline perfusion acid precipitation Process/ addition Depth Filter 2.5M Acetic acid dosing Batch Millistak+® HC Pro X0SP 0% Batch ( control) Millistak+® HC Pro X0SP 2.2% Batch (control) Millistak+® HC Pro C0SP 2.2% Inline Millistak+® HC Pro X0SP 2.2% Inline Millistak+® HC Pro C0SP 2.2%
- 23. 01 02 03 04 05 Agenda 23 Introduction Inline Flocculation System Experimental Plan Results Summary
- 24. Inline harvest pDADMAC flocculation & clarification using Clarisolve® 40MS depth filter Results ▪ No impact of residence time on differential pressure and instantaneous turbidity ▪ Slight continued precipitation observed in inline flocculated streams turbidity increased to ~15 NTU after 2-hour hold. Batch control remained at 4.7 NTU.
- 25. ▪ No impact of residence time on titer, % aggregate and host cell protein levels. ▪ Overall no impact observed on clarification performance by reducing residence time from 10 to 2 minutes for inline pDADMAC flocculation. Results Inline harvest pDADMAC flocculation & clarification using Clarisolve® 40MS depth filter
- 26. Results Inline harvest acid precipitation & clarification using Clarisolve® 20MS depth filter ▪ No impact of residence time on differential pressure and instantaneous turbidity ▪ No continued precipitation observed but pool turbidities slightly higher in case of inline flocculation streams (Batch 3.42 NTU, 10 min RT 7.73 NTU, 5 min RT 11.1 NTU 2 min RT 9.87 NTU) ▪ pH within +/- 0.1 pH unit
- 27. ▪ No impact of residence time on titer, % Aggregate and host cell protein levels. ▪ Overall no impact observed on clarification performance by reducing residence time from 10 to 2 minutes for inline acid precipitation. Results Inline harvest acid precipitation & clarification using Clarisolve® 20MS depth filter
- 28. Results Inline perfusate acid precipitation & clarification using Millistak+® HC Pro X0SP depth filter ▪ pH trended 0.1 unit lower in case of inline as compared to batch process ▪ Greater than 250L/m2 capacity in both batch & inline flocculation ▪ Turbidity levels below 1 NTU in all cases Non-flocculated feed turbidity 1 NTU Flocculated feed turbidity 389 NTU Millistak+® HC Pro X0SP Millistak+® HC Pro X0SP Millistak+® HC Pro X0SP Millistak+® HC Pro X0SP Millistak+® HC Pro X0SP Millistak+® HC Pro X0SP
- 29. Results Inline perfusate acid precipitation & clarification using Millistak+® HC Pro C0SP depth filter Non-flocculated feed turbidity 1 NTU Flocculated feed turbidity 389 NTU ▪ pH trended 0.1 unit lower in case of inline as compared to batch process ▪ Greater than 450L/m2 capacity in all cases ▪ Turbidity levels below 1 NTU in all cases Millistak+® HC Pro C0SP Millistak+® HC Pro C0SP Millistak+® HC Pro C0SP Millistak+® HC Pro C0SP
- 30. ▪ Yields ranged from 76 -81% * Not adjusted for ~5% dilution due to acid and base addition ▪ HCP reduction ranged from 61% for non flocculated clarified using Millistak+® HC Pro X0SP to 75% in case in inline & batch flocculated feed streams clarified using Millistak+® HC Pro X0SP /C0SP. Results Inline perfusate clarification using Millistak+® HC Pro X0SP and C0SP Millistak+® HC Pro Millistak+® HC Pro Millistak+® HC Pro Millistak+® HC Pro Millistak+® HC Pro
- 31. Feed Streams Titer (mg/ml) HCP (ng/ml) DNA (µg/ml) Turbidity (NTU) Normalized HCP (ng/mg) Normalized DNA(µg/mg) Fed Batch Harvest 1.78 1853929 69.19 >1000 1.04E+06 38.87 Acid precipitated clarified with Clarisolve® 20MS 1.66 1318839 0.29 3.42 7.94E+05 0.17 Perfusion process Alternating Tangential Flow (0.22um) 0.37 68613 9.69 1.04 1.85E+05 26.19 Millistak+® HC Pro X0SP clarified perfusate 0.3 26591 ND 0.5 8.86E+04 NA In line acid precipitation Millistak+® HC Pro X0SP clarified perfusate 0.28 16190 ND 0.5 5.78E+04 NA Results Comparison of impurity profiles between a typical fed batch and perfusion process ▪ Normalized host cell protein levels in perfusate ~ 80 % lower than a typical fed batch process. ▪ DNA levels ~30% lower in perfusate than a typical fed batch process. ▪ Acid precipitation harvest clarification removes ~25% HCP and 2.3 logs of DNA ▪ Millistak+® HC Pro X0SP clarification helped remove ~50% of HCP ▪ Inline perfusate flocculation subsequent clarification with Millistak+® HC Pro X0SP helped remove 70% of HCP
- 32. 01 02 03 04 05 Agenda 32 Introduction Inline Flocculation System Experimental Plan Results Summary
- 33. ▪ Inline flocculation of CHO cell culture using pDADMAC and acid precipitation are feasible with 2-minute residence time. ▪ Harvest Flocculation ▪ pDADMAC Flocculation ▪ Similar pressure profiles & turbidity trends for all residence times & batch flocculation ▪ Slight continued precipitation observed in case of inline flocculated feed streams ▪ Inline flocculation had no impact on yield, % aggregate and HCP levels. ▪ Acid precipitation ▪ Similar pressure profiles & turbidity trends for all residence times ▪ No Continued precipitation observed but pool turbidites higher in case of in case of inline flocculated feed streams ▪ Inline flocculation had no impact on yield, % aggregate and HCP levels. ▪ Perfusate flocculation offers a continuous clarification method to address soluble impurities. ▪ Inline flocculation can be an interesting alterative for batch flocculation and can be helpful in addressing certain facility constraints in a GMP Manufacturing Inline Flocculation Summary
- 34. ▪ Marine Maszelin ▪ John Amara ▪ Kara Pizzelli ▪ Krista Cunningham ▪ Dana Kinzlmaier ▪ Mike Bruce ▪ Derek Silva ▪ Brandon Medeiros ▪ Mike Cunningham ▪ Mike Susienka Acknowledgement
- 35. Schedule an in-person or remote visit today. EMDMillipore.com/mlab
- 36. The vibrant M, M Lab, Clarisolve, EX-CELL, and Millistak+ are trademarks of Merck KGaA, Darmstadt, Germany or its affiliates. All other trademarks are the property of their respective owners. Detailed information on trademarks is available via publicly accessible resources. © 2021 Merck KGaA, Darmstadt, Germany and/or its affiliates. All Rights Reserved.