Development of high-yielding chemically defined and animal component-free generic processes using GS cell lines. David Mainwaring, Lonza Biologics

Transcription

1 Development of high-yielding chemically defined and animal component-free generic processes using GS cell lines. David Mainwaring, Lonza Biologics

2 Introduction Generic processes Process optimisation GS-CHO example Summary File Name 07/10/2004 / 2

3 Generic processes Why? Contract manufacture of biopharmaceuticals Deal with many different cell lines and cell types Not possible or desirable to optimise every process Takes time for early phase trials need material quickly Costs money Generic process allows rapid generation of material for the clinic File Name 07/10/2004 / 3

4 Generic processes Basis for generic processes is chemically defined media No hydrolysates are used They are potential source of variability Costs are associated with screening to find acceptable batches of hydrolysates Effect on metabolism may differ between cell lines Aids understanding of process Animal component-free taken as granted Regulatory reasons Simplifies purification File Name 07/10/2004 / 4

5 Generic processes Generic processes are designed to be adaptable Slight modifications to feeds make them suitable for other cell types GS-NS0 process suitable for other selection systems with NS0 and hybridoma Add glutamine, reduce glutamate concentration and changes in some other amino acid components GS-CHO process suitable for dhfr-cho with similar changes and removal of HT from inoculum process Reactor conditions modified depending on cell type File Name 07/10/2004 / 5

6 Process optimisation There are several routes to improving process productivity, falls into two main categories Increase the time integral of viable cell concentration (IVC) Increase the specific productivity of the cells Alternatively big is better approach Scale-up to larger reactors File Name 07/10/2004 / 6

7 Process optimisation Increasing IVC can be achieved in many ways Increase maximum viable cell concentration Prevent cell death Increase process duration Fed-batch processes Anti-apoptotic engineering Temperature shifts ph shifts Make each cell more productive Productivity enhancers Select a new cell line File Name 07/10/2004 / 7

8 Scale-down model Impractical to do all work at production scale Need a scale-down model as representative of final production bioreactor as possible Ideally use small-scale fully controlled bioreactors Expensive and limited number of treatments possible Shake-flasks are a good compromise Large numbers can be used in parallel Differences in ph and DOT control Need to regularly test advances in a representative bioreactor File Name 07/10/2004 / 8

9 Fed-batch processes Fed-batch processes Analysis of spent culture media Re-supplement depleted components Initial rounds of optimisation lead to highest increases Typically several fold increase over batch process Large increases in product concentration become more difficult to achieve as process develops Example using GS-NS0 cell line 6A1(100)3 File Name 07/10/2004 / 9

10 Process optimisation GS-NS0 cell growth 10 Viable Cell Concentration (10 6 cells/ml) Elapsed Time (h) Original process Improved process v1 Improved process v2 Improved process v3 File Name 07/10/2004 / 10

11 Process optimisation product concentration 1200 Product Concentration (mg/l) Time Integral of Viable Cell Concentration (10 9 cell h/l) Original process Improved process v1 Improved process v2 Improved process v3 File Name 07/10/2004 / 11

12 Is the fed-batch process improvement generic? 2000 Product Concentration (mg/l) Time Integral of Viable Cell Concentration (10 9 cell h/l) Old Process New Generic Process File Name 07/10/2004 / 12

13 Further process optimisation Fed-batch process is applicable to multiple cell lines Generic Similar results with GS-CHO process Other factors are considered during optimisation Temperature shifts GS-NS0 example Productivity enhancers Effect on two different GS-NS0 cell lines ph shifts GS-CHO File Name 07/10/2004 / 13

14 Temperature shifts GS-NS Viable Cell Concentration (10 6 cells/ml) Temperature shift Product Concentration (mg/l) Elapsed Time (h) 0 Control Temperature decrease File Name 07/10/2004 / 14

15 Increasing specific productivity Ideally should be selecting high producers during cell line selection Can modify culture conditions to increase specific productivity ph, temperature Productivity enhancers Rely on increasing the cell specific rate of product synthesis Often do not know how they function Are they generic? File Name 07/10/2004 / 15

16 Acetate in bioreactors GS-NS0 800 Product Concentration (mg/l) Time Integral of Viable Cell Concentration (10 9 cell h/l) Control Acetate 1.6-fold increase in product concentration No change in IVCC, direct result of increased q P File Name 07/10/2004 / 16

17 Productivity enhancers flask 600 Product Concentration (mg/l) Control Acetate Concentration (mm) Product concentration increases with acetate concentration, a result of increased q P File Name 07/10/2004 / 17

18 Productivity enhancers - bioreactor 1200 Product Concentration (mg/l) Time Integral of Viable Cell Concentration (10 9 cell h/l) Control Acetate 1.4 fold increase in q P, concomitant decrease in IVC of similar magnitude No increase in product concentration File Name 07/10/2004 / 18

19 Productivity enhancers Productivity enhancers can be highly effective at increasing product concentration Several downsides Can be cytotoxic Cell type and cell line specific Need optimising for each cell line and process time consuming Not suited to generic processes File Name 07/10/2004 / 19

20 GS-CHO process optimisation Principles used for development and optimisation of the GS-NS0 process applied to GS-CHO Multiple modifications to process Change of cell line from CHOK1 to CHOK1SV host Iterative feed modifications Basal medium change Increased feed volume Extended culture duration Modification of ph control strategy Cloning of cell line File Name 07/10/2004 / 20

21 Process optimisation - productivity 6000 Product Concentration (mg/l) Time Integral of Viable Cell Concentration (10 9 cell h/l) Iteration 2 Iteration 3 Iteration 4 Iteration 5 Iteration 5 CY01 File Name 07/10/2004 / 21

22 Maximum viable cell concentration 20 Viable cell concentration (10 6 /ml) H11 orig 22H11 v1 22H11 v2 LB01 v2 LB01 v3 LB01 v4 LB01 v5 CY01 v5 File Name 07/10/2004 / 22

23 Time integral of viable cell concentration IVC (10 9 cell h/l) H11 orig 22H11 v1 22H11 v2 LB01 v2 LB01 v3 LB01 v4 LB01 v5 CY01 v5 File Name 07/10/2004 / 23

24 Specific productivity 2.0 Specifc production rate (pg/cell/h) H11 orig 22H11 v1 22H11 v2 LB01 v2 LB01 v3 LB01 v4 LB01 v5 CY01 v5 File Name 07/10/2004 / 24

25 Process duration Process duration (days) H11 orig 22H11 v1 22H11 v2 LB01 v2 LB01 v3 LB01 v4 LB01 v5 CY01 v5 File Name 07/10/2004 / 25

26 Product concentration 6000 Antibody Concentration (mg/l) H11 orig 22H11 v1 22H11 v2 LB01 v2 LB01 v3 LB01 v4 LB01 v5 CY01 v5 File Name 07/10/2004 / 26

27 Summary Generic processes enable rapid generation of clinical material Typically yield 1 2 g/l Processes can be optimised if more material is required Process optimisation can achieved by increases in IVC and cell specific productivity Chemically defined, optimised processes yielding over 5 g/l can be achieved File Name 07/10/2004 / 27

28 Acknowledgements Cell culture process development Assay development group for support File Name 07/10/2004 / 28