Mammalian Cell Culture: Current Status, Future Prospects

Cell Culture and Upstream Processing, Berlin, September 2004

John BirchLonza

slide 2Author / Date

Cell Culture: an Industrial Process

Much of the early process development driven by vaccines ( human and veterinary ) – Polio vaccine 1954

FMD vaccine in deep tank suspension culture

Since late 1980s (tPA 1987), large scale use driven particularly by recombinant therapeutic proteins

slide 3Author / Date

Significance of Mammalian Cells

Used especially for large complex proteins especially those requiring posttranslational modifications – in practice a large proportion of the totalApprox. 73 licensed proteins ( USA ) of which ca. :35 produced in micro organisms : E.coli ( 28 ) , S.cerevisiae ( 7 )38 ( 52% ) produced in mammalian cells

CHO 27 ( 10 Mabs )NS0, SP2/0,hybridomas etc 8 ( 8 Mabs )BHK 2C127 1

slide 4Author / Date

Biopharmaceutical Proteins

Strong demand – sales of biopharmaceuticals represent 10-30% of all new pharmaceuticals in USA in recent years(1)Large number ( hundreds ) of proteins in developmentSales of ca.$ 30bn in 2003 (2)expected to grow to $59bn by 2010 ( Datamonitor )Mammalian cell products account for ca. 60% of market (1)Mabs are the fastest growing category ( from 1% of biopharmaceutical market in 1995 to 14% by 2001 )(1)

1. Polastro & Tulcinski, Scrip magazine Sept. 2002 2. Walsh, Nature Biotechnology 2003, 21, 865 – 870

slide 5Author / Date

Key Issues in Manufacturing (1)

Speed in development (especially for early phase clinical material)

Rapid cell line creation – avoid amplification and multiple rounds of single cell cloningUse of cell lines (CHO) preadapted to suspension cultureUse of generic technologyPredictive scale up systems

slide 6Author / Date

Key Issues in Manufacturing (2)

High dose requirements, particularly for antibodies, leads to large volume demand (10’s to 100’s kg/year )Estimated 2004 protein demand > 2000kg (mostly Mabs and fusion proteins )( UBS )This is a driver for –

Increased capacity Increased reactor size – up to 20000 litres (economy of scale )Improved technology to increase cost efficiency ( taking note of downstream implications )

slide 7Author / Date

Mammalian Cell Capacity

Long lead times (3-5 years ) and high costs ($200m - $500m)($3+ Million / m3)

510 ( 26%)19702006

190 ( 23%)8402002

Contract Manufacturers

Total

Estimated capacity

( thousands of litres )

slide 8Author / Date

20,000L Bioreactor & Add TanksPortsmouth, New Hampshire

slide 9Author / Date

Improved Upstream ProcessEfficiency

Parameters TargetedReactor throughputSpecific production rateMaximum viable cell concentrationProlongation of culture at high cell viabilityGrowth rate

Routes to optimisationDesign of gene vectorProperties of host cell lineCell line screeningOptimisation of fermentation process

slide 10Author / Date

Design of Gene Vector to Maximize Transcription

Strong promoter to drive expression of product gene(s)

Amplification of gene copy number

Vectors with elements that create genomic environment for high transcriptional activity (positional independence)

Targeting of expression vector to transcriptionally active site in genome by homologous recombination

slide 11Author / Date

Glutamine synthetase (GS) gene expressionsystem

Expression vector encoding product gene plus GS gene, allowing glutamine synthesisGS is inhibited by methionine sulphoximine (MSX)Selection in glutamine-free medium for GS minus cell types (e.g. NS0)Selection in the presence of MSX for GS positive cell types (e.g. CHO)Only cells with GS gene (and hence product gene) surviveIncrease selection stringency - use weak promoter on GS gene - selects for rare integration into transcriptionally efficient sites in genomeExpression of linked product gene, driven by strong promoter, enhanced by favourable integration site

slide 12Author / Date

Improved Upstream ProcessEfficiency

Parameters TargetedReactor throughputSpecific production rateMaximum viable cell concentrationProlongation of culture at high cell viabilityGrowth rate

Routes to optimisationDesign of gene vectorProperties of host cell lineCell line screeningOptimisation of fermentation process

slide 13Author / Date

Properties of The Host Cell Line

Significant opportunities to create cell lines with improved growth and productivity characteristics

Metabolic engineering

Variant SelectionCholesterol independent NS0 variantSuspension variants of CHO

slide 14Author / Date

Improved Upstream ProcessEfficiency

Parameters TargetedReactor throughputSpecific production rateMaximum viable cell concentrationProlongation of culture at high cell viabilityGrowth rate

Routes to optimisationDesign of gene vectorProperties of host cell lineCell line screeningOptimisation of fermentation process

slide 15Author / Date

Cell Line Screening

Highly productive transfectants are rare even with a good selection system

Various approaches to improve screening process to find these rare events

Increase transfection efficiency ( larger pool to select from )Improve stringency of selectable marker to eliminate lowproducersHigh throughput methods (FACS + cell surface product capture)Early screening needs to predict cell specific production capability AND growth characterisitics at production scale

slide 16Author / Date

Cell line selection

Transfection and selection conditions for GS-CHO cell lines expressing cB72.3 antibody

1973

1242

681

Numbers of stable

transfectants

Electroporation condition

2.5 x 106 cells electroporated

slide 17Author / Date

Cell Line Selection

Influence of Selection Conditions for GS-CHO Cell Lines making cB72.3 Antibody

Cell lines have not been amplified.

0

2

4

6

8

10

12

14

16

40 80 120 160 200 240 280 320

Antibody (mg/L)

25 µM MSX

50 µM MSX

slide 18Author / Date

secreted antibody

fluorochrome-labelled detection antibody

neutravidinbridge

biotinylated Protein A

biotinylated-cell surface

Affinity-matrix surface capture

slide 19Author / Date

Flow cytometric analysis of AMSC-labelled GS-CHO cells

slide 20Author / Date

Clonal Variation – GS-NS0 (non-optimised culture )

60

31

24

19

19

17

Specific prod.rate (pg/cell/day)

320

573

518

212

209

379

Cumulative cell time

(109 cell·h/L)

79014

75019

53522

17019

1658

26017

Product concn.(mg/l)

Max.viable cell concn. (105 /ml)

slide 21Author / Date

Selecting high producing cell lines

Transfect 107 host cells with vector

30-60 transfectants

30-60 transfectants

productivity assessment(quantitative)

Preliminary productivity assessment (quantitative)

STATIC CULTURE

SUSPENSION(shake flask culture)

Select 3 cell lines for further analysis

200-300 transfectants

Adapt to protein-free medium

6 week

4 weeks

8 weeks

3 weeks

5-10 transfectantsFed-batch assessment of

growth, productivity, product quality3 weeks

Single colonies per well

slide 22Author / Date

Prediction of bioreactor behaviour from shake-flask model

0.0

0.4

0.8

1.2

1.6

2.0

antibody Qp

Parameter

Valu

e in

reac

tor r

elat

ive

to s

hake

-flas

k

slide 23Author / Date

Improved Upstream ProcessEfficiency

Parameters TargetedReactor throughputSpecific production rateMaximum viable cell concentrationProlongation of culture at high cell viabilityGrowth rate

Routes to optimisationDesign of gene vectorProperties of host cell lineCell line screeningOptimisation of fermentation process

slide 24Author / Date

Improving the Fermentation Process

Significant potential to improve processes

Physicochemical environment

Medium design and feeding strategies ( most processes are fed-batch, some perfusion culture )

Use of chemically defined media

slide 25Author / Date

The Physicochemical Environment

Control pH, temperature, dissolved oxygen concentrationSmall changes in pH can have profound effect on cell growth and productivity

Responses are cell line specific and can impact:

Maximum cell concentrationIntegral viable cell hoursSpecific production rateMetabolism: lactate accumulation

slide 26Author / Date

1

10

100

0 50 100 150 200 250 300 350 400

Time (hours)

Viab

le c

ell c

once

ntra

tion

(105 /m

L)

0

0.3

0.6

0.9

1.2

1.5

Lact

ate

conc

entr

atio

n (g

/L)

pH 7.3 pH 7.0

slide 27Author / Date

Chemically Defined Media

Increasing use of chemically defined media free of animal derived raw materials

Reduced risk of introducing adventitious agentsImproved process consistency and robustness (avoids potential variability of raw materials such as serum proteins and hydrolysates)Chemical definition assists process optimisationBenefits purification (reduced contaminant load)

slide 28Author / Date

Downstream Benefits of Chemically Defined Medium for GS-NS0 Cell Line

Purity of MAb at harvest

Optimised Protein containing <30%culture

Optimised protein free culture 62%-76%

slide 29Author / Date

Medium Design and Feeding Strategies

Optimise basal medium

Optimise feeds

Maintain nutrient sufficiency

Minimise waste metabolite formation

Use of GS system avoids accumulation of ammonium ions from metabolism of glutamine

slide 30Author / Date

Optimisation of a GS-CHO Process

Culture conditions for a GS-CHO making cB72.3 antibody were optimisedSuspension variant of CHO-K1 isolated:

grows in chemically defined medium without need for adaptation (can take several months)

Efficiency and stringency of transfection conditions increased to improve selection of highly productive clonesGrowth conditions further optimised

slide 31Author / Date

Process optimisation for a GS-CHO cell line

4301

3560

2829

1917

585

334

139

Antibody(mg/L)

26Iteration 4

31

20

14

4

2

Fold increase

Iteration 5

Iteration 3

Original cell line

New cell line

Iteration 2

Iteration 1

Process

slide 32Author / Date

GS-CHO: antibody production

P rocess developm ent s tage22H 11 orig22H 11 v1 22H 11 v2 LB01 v2 LB01 v3 LB01 v4 LB01 v5

Antib

ody

(mg/

L)

0

1000

2000

3000

4000

5000

22H 11 orig22H 11 v1 22H 11 v2 LB01 v2 LB 01 v3 LB01 v4 LB01 v5Spec

ific

prod

uctio

n ra

te (p

g /(c

ell.h

))

0.0

0.5

1.0

1.5

2.0

slide 33Author / Date

GS-CHO growth

P r o c e s s d e v e lo p m e n t s t a g e2 2 H 1 1 o r ig 2 2 H 1 1 v 1 2 2 H 1 1 v 2 L B 0 1 v 2 L B 0 1 v 3 L B 0 1 v 4 L B 0 1 v 5

Viab

le c

ell c

once

ntra

tion

(106 /m

L)

0

2

4

6

8

1 0

1 2

1 4

1 62 2 H 1 1 o r ig 2 2 H 1 1 v 1 2 2 H 1 1 v 2 L B 0 1 v 2 L B 0 1 v 3 L B 0 1 v 4 L B 0 1 v 5

IVC

(106 c

ells

.h/m

L)

0

1 0 0 0

2 0 0 0

3 0 0 0

4 0 0 02 2 H 1 1 o r ig 2 2 H 1 1 v 1 2 2 H 1 1 v 2 L B 0 1 v 2 L B 0 1 v 3 L B 0 1 v 4 L B 0 1 v 5

Proc

ess

dura

tion

(day

)0

5

1 0

1 5

2 0

2 5

slide 34Author / Date

Growth comparison: “old” vs. “new” GS-CHO cell lines

0

40

80

120

0 100 200 300 400

Time (h)

Viab

le c

ell c

once

ntra

tion

(10

5 /mL)

22H11LB01

slide 35Author / Date

Process optimisation for a model GS-NS0

1405

1427

1239

1026

772

640

Cumulative cell time

(109 cell·h/L)

1422

1035

807

589

293

476

cB72.3 antibody(mg/L)

0.97Iteration 4

0.71Iteration 3

0.64Iteration 2

0.60Iteration 1

0.36Chemically defined

0.74Serum-free

Qp

pg/(cell·h)Process

slide 36Author / Date

Future Opportunities ( 1 )

Continued process improvements ( 10g/l achievable ?)

Continued improvements in selection procedures for finding highly productive clones

Improved cell lines e.g. engineered for improved growth, product synthesis, energy metabolism , glycosylation characteristics

Use of genomic and proteomic approachesto increase our knowledge of cell physiology and to inform design of cells and processes

slide 37Author / Date

Future Opportunities (2)

Improved product potency – impact on quantities required

Alternative manufacturing routes – microbial production of e.g. antibody fragments or whole antibodies, transgenic production for very large quantities ?

Mammalian cell culture likely to be key technology for foreseeable future

slide 38Author / Date

Summary

Increasing number of mammalian cell products in development is a driver for improved and faster process development technologiesHigh volume demands are driving both capacity and improvements in process efficiencyGreat potential to use advances in basic science to inform process improvements