Fluid flow dynamics simulation of a UV-C irradiation process for the inactivation of viruses in protein solutions. Todd Nikolof

Size: px

Start display at page:

Download "Fluid flow dynamics simulation of a UV-C irradiation process for the inactivation of viruses in protein solutions. Todd Nikolof"

Silvia Richardson
5 years ago
Views:

1 Fluid flow dynamics simulation of a UV-C irradiation process for the inactivation of viruses in protein solutions Todd Nikolof

2 Current dedicated viral inactivation/removal procedures Dry heat treatment Pasteurisation Low ph incubation Solvent detergent Virus filtration 2

3 Shortfalls of current viral inactivation/removal methods Less effective against small, heat resistant, nonenveloped viruses e.g. hepatitis A virus & human parvovirus B19 Need for inactivation techniques for non-enveloped viruses still remains 3

4 virus inactivation Effect of UVC on viral inactivation: relationship to genome size Fluency (J/m 2 ) required for 4 log Genome size (kb) SV40 PPV HAV Sinabis Reo Adeno 4

5 5 UVC light and viral and protein damage

6 Continuous-Flow reactor Novel hydraulic spiral flow along an irradiation source Dean vortices 6

7 Mixing in the Continuous Flow Reactor Highly efficient mixing which is controlled in a fluid stream UVC treatment is controllable 7

8 % Aggregate % Fragment Stability data of protein composition of irradiated IgG at 2-8 o C Aggregate Fragment Time (months) Time (months) Pasteurised 200 J Pasteurised 200 J 8

9 Hepatitis A titre (IU/mg) Hepatitis B titre (IU/mg) Stability data of hepatitis titres of IgG at 2-8 o C Hepatitis A Hepatitis B Time (months) 400 J 200 J 100 J Pasteurised Control Time (months) Control Pasteurised 100 J 200 J 400 J 9

10 Characterisation of the effects of UVC irradiation on monoclonal IgG Glycosylation Mannose Galactose Glucose Glucosamine N-acetylglucosamine Sialic acid Peptide Mass Fingerprinting Bio-Activity Weak cation exchange chromatography 10

11 Scaling from Lab scale to Process Scale Issue: Change in Diameter of irradiation chamber Need to establish Comparability of UV exposure If there are changes in flow patterns Use computer modelling to demonstrate fluid dynamics Apply modelling to demonstrate scalability between lab and process scales of the UVivatec systems A convenient chemical actinometry method for measuring fluency 11

12 Computational Fluid dynamics in bioprocessing research International BioPharm, (2010) 23, Pharmaceutical Engineering, (2010) 30,

13 13 Room flooded in the Titanic

14 14 St Francis dam break near Santa Clarita, California (1928)

15 Modelling of chamber Modelling: Used CAD Based on four turns Volume of Chamber Lab Process 24 ml 100 ml Flow Rate 2-20 L/h L/h 15

16 Assumptions and Input Parameters Total number of particles used Lab: 350,000 Process: 1,000,000 Uses Navier-Stokes equations to describe motion of fluids Fluid viscosity and density used are those of water Flow rates simulated Lab: 2 and 20 L/h Process: 5, 25, and 40 L/h 16

17 Axial Direction of Forces within the Chamber Radial 17

18 Flow components: Lab scale Radial Axial Azimuthal 2 L/h 20 L/h 18

19 Net flow pattern: Lab scale 2 L/hr 20 L/hr 19

20 Residence time distribution of fluid: Lab Scale System 2 L/hr. 20 L/hr. Flow rate (L/hr) Theoretical peak retention time (s) Calculated average retention time (s)

21 Shear stress generated at different flow rates: Lab scale 2 L/hr. 20 L/hr. Shear stress increases 200 fold for 20 L/h as compared to 2 L/hr. 21

22 Flow components: Process scale 5 L/hr 25 L/hr 40 l/hr Radial Axial Azimuthal 22 Average radial, axial, and azimuthal velocities different flow rates.

23 Net Flow Pattern: Process scale 5 L/hr. 25 L/hr. 40 L/hr. Velocity vectors showing secondary flow pattern at sample plane 1 for the helix for Process scale at specified flow rates. 23

24 Residence time distribution of fluid: Process Scale 5 L/hr. 25 L/hr. 40 L/hr. 24

25 Shear stress generated at different flow rates: Process scale 5 L/hr. 25 L/hr. 40 L/hr. Shear stress in Pa for flow rates for Process scale. 25

26 Net Flow Patterns of the Lab scale and Process scale Lab system 2 L/h Process system 5 L/h 20 L/h 40 L/h 26

27 The effect of surface area to flow ratio on Dean vortices 2 L/h 2 L/h 20 L/h 20 L/h 27

28 Chemical Actinometry and Fluence Actinometry (iodide-iodate) is useful for assessing the dosage delivered to a fluid UV chamber 8KI + KIO 3 + 3H 2 0 hv 3I 3 + 6OH - + 9K + Fluence can be determined Fluence (mj/cm 2 ) = k.abs.vol Area Objectives Demonstrate that actinometry can be used to compare fluencies between laboratory and process scale 28

29 Expected fluence rates Flow rate (L/hr) Fluence (J/m 2 ) Time (seconds) Fluence Rate ((J/m 2 )/s) Fluence rate Flow rate 29

30 Fluence ((mw/cm2)/s) The effects of mixing on fluence rate Static Mixing Time (seconds) 30

31 Fluence rate ((mw/cm2)/s) Fluence rate for multiple passes n = 3 2 L/h 5 L/h 10 L/h 15 L/h 20 L/h Number of passes 31

32 Fluence rate ((mw/cm2)/s) Average Fluence rate at different flow rates Flow rate (L/h) 32

33 Conclusions UVC viral inactivation has minimal effect on protein integrity and stability Scale up an important consideration for the adoption of this technology Fluid dynamic modelling was used to study the generation of Dean vortices and the effect of flow rate on mixing and may be used to demonstrate scalability between laboratory and process UVivatec systems Chemical actinometry was shown to be a means of validating and monitoring UVC irradiation and can be used to monitor UVC system performance Actinometric results reflected the theoretical model 33

34 Acknowledgements Joseph Bertolini Mahesh Prakash Chor Sing Tan David Eakins Michelle Burns 34

High-throughput protein aggregate analysis of monoclonal antibodies using a novel dual-channel UHPLC instrument

APPLICATION NOTE 72598 High-throughput protein aggregate analysis of monoclonal antibodies using a novel dual-channel UHPLC instrument Authors Nicola McGillicuddy, 1 Amy Farrell, 1 Sara Carillo, 1 Martin