Use of DoE in Biopharmaceutical Industry

ProtAffin Biotechnologie AG

3d European DoE user Meeting, Lucerne, 1st of June 2010

2PROTAFFIN

Alain Poncin (aponcin@protaffin.com)

2010 - : ProtAffin, Graz, Austria : Head of Manufacturing

Development of a new class of proteins that binds specifically to GAG’s

Lead candidate : PA401, a decoy IL8 to treat chronic inflammation such as COPD.

2008 – 2010 : LFB Biotechnologies, Paris, France : Head of DSP Unit

Development of Coagulation Factors from Milk of Transgenic Animals

1988 – 2008 : Eurogentec, Liege, Belgium : from lab to Executive Management

CMO, Development of Production Process for more than 70 various proteins

mailto:aponcin@protaffin.com

3PROTAFFIN

22 years of experience with DoE

Ion exchange chromatography process parameters characterization by DoE

Societe de BioChromatographie et Nanoseparation, 19-22 October 2010, Lyon, France.

P. Paolantonacci, B. Claudel, D. Gachelin, A. Poncin, M. Ollivier

Implementation of QbD in Down Stream Processes for plasma derived and recombinant proteins

products.

Webinar PDA on Quality by Design, 3 March 2010. A. Poncin, P. Paolantonacci, M. Ollivier

Study of the anion exchange chromatographic step operating conditions of the new IgG

manufacturing process - Characterisation by Design of Experiment

Plasma Product Biotechnology Meeting (PPB09), 11-15 May 2009, Menorca, Spain.

P. Paolantonacci*, D. Gachelin, S. Nakache, A. Poncin, A. Sauger, M. Ollivier

Risk Assesment and DoE must be used in synergy for Quality by Design success

PDA Conference on Quality by Design, 7-8 October 2008, Frankfurt, Germany. A. Poncin

Eurogentec current validation strategy.

Interlaken, 5th international conference on HIC/RP chromatography, 2007. A. Poncin

4PROTAFFIN

Biopharmaceutical Industry

– Giants (Pfizer, Genentech, Abbot, GSK, Amgen, Lilly,...) but also a lot of SMEs (> 5000 companies in 2009)

– Since 1982 (approval of recombinant Insulin), 114 products on the Market, 1011 $/year

– Not only production of effective and safe Drugs by Recombinant DNA Technology but also Agriculture, Bio Fuels, Clean Techs...

– Use of living cells • Microbes (bacteria and yeast)

• Eukaryotes (CHO, NSO, BHK,...)

• Organism (plants and animals)

– To produce proteins (enzymes, antibodies, hormones,...), DNA, sugars and living (attenuated) cells (stem cells)

5PROTAFFIN

Pharmaceutical Industry

Highly regulated : FDA, EMA, ICH,...to protect Public Health

• 20th Century : mainly reactive, validation, lack of flexibility

• 21th Century : Science based

•Process Analytical Technology (FDA, 2004)

•Quality by Design (ICH Q8R2, Q9, Q10, Q11,...)

Methodological experiments based on statistical principles of

orthogonality, reference distribution, and randomization, provide effective

means for identifying and studying the effect and interaction of product

and process variables. Traditional one-factor-at-a-time experiments do

not address interactions among product and process variables.

Formal Experimental Design:

A structured, organized method for determining the relationship between

factors affecting a process and the output of that process. Also known as

“Design of Experiments”.

6PROTAFFIN

Production Process for Recombinant Proteins

– Proteins are not simple amino acid sequences

– Unfortunately, bacteria produces usually mammalian

Proteins as Inclusion Bodies (inactive aggregates)

Requiring In vitro solubilisation and refolding Folded PA401 (3D structure)

7PROTAFFIN

Case Study : Refolding of Proteins

• 14 factors potentially critical

– Protein concentration

– pH

– Solubilisation agent

– Residual Chaotropic

– Reshuffling buffer

– Salt type

– Salt concentration

– Sugar

– Organic solvant

– Cofactor

– ...

8PROTAFFIN

Effect of protein concentration

Unfolded Folded : decreasing yield as Protein concentration increases

PA401 : active form is a dimer

• DoE required to find a (Hyper) space where Yield and Quality of refolding is acceptable

00.10.20.30.40.50.60.70.80.91

00.10.20.30.40.50.60.70.80.9

1

0 0.5 1 1.5 2 2.5

mu

lim

eri

sati

on

yie

ld

Yie

ld o

f re

fold

ing

protein concentration

Effect of protein concentration

9PROTAFFIN

Refolding of Proteins by DoE

Unfolded Folded Difficult to analyse

Aggregates Easy : 96 wells plate - turbidity

10PROTAFFIN

14 Factors

Building of DoE by using only

12 factors (res IV + cp) : 34 runs

Followed by 2 duplications

- temperature

- duration

144 data

11PROTAFFIN

Results

Design-Expert® Software

Norm Turbidity

Error from replicates

Shapiro-Wilk test

W-value = 0.922

p-value = 0.000

A: Protein

B: pH

C: GSH/GSSG

D: Dilution

E: Arginine

F: NaCl

G: KCl

H: Salt

J: Ethanol

K: Tween

L: Glycerol

M: Heparin

N: Temperature

O: Duration

Positive Effects

Negative Effects

Half-Normal PlotH

alf-N

orm

al %

Pro

ba

bility

|Standardized Effect|

0.00 0.03 0.05 0.08 0.11

0.010.020.030.0

50.0

70.0

80.0

90.0

95.0

99.0

99.9

A

C

D

E

F

K

AC

ABCF

Out of the 14 potentital

critical factors, only 7

(+interactions) really

critical for PA401

12PROTAFFIN

DoE 1 (and 6 s)

Define : which are the critical factors implied in refolding of PA401

Mesure : current refolding yield

Analyse : DoE1 (turbidity)

Improve : optimised conditions for refolding of PA401

Control : quality of refolded PA401 in optimised conditions

scale up and testing : no activity

Second DoE : RSM using the 7 critical factors identified

analysis : potency assay by chromatography

13PROTAFFIN

PA401 – DoE2

– 7 critical factors by RSM : 86 runs (RSM for 14 factors would have required 550 runs)

Factor DoE1 DoE2

Protein 0.5 – 2.5 0.5 – 2.5

pH 6.0 – 8.0 8.0 – 9.5

GSH/GSSG 0.2 – 5.0 0.2 – 5.0

Dilution 5 – 20 5 – 20

Arginine 0 – 0.2 0 – 0.5

NaCl 0 – 0.5 0 – 0.5

Tween 0 - 1 0 - 1

14PROTAFFIN

Results :

Design-Expert® Software

Protein bound

2168.75

-2.4352

X1 = A: Protein

X2 = D: Dilution

Actual Factors

B: pH = 8.75

C: GSH/GSSG = 2.60

E: Arginine = 0.25

F: Tween = 0.50

0.50

1.00

1.50

2.00

2.50

5.00

8.75

12.50

16.25

20.00

100

550

1000

1450

1900

P

rote

in b

ou

nd

A: Protein D: Dilution

- Factor B no more significant

- Quadratic Model using the 5 critical Factor and optimisation of refolding conditions

15PROTAFFIN

Optimised refolding

Constraints

Lower Upper Lower Upper

Name Goal Limit Limit Weight WeightImportance

Protein maximize 0,5 2,5 1 1 3

pH is in range 8 9,5 1 1 3

GSH/GSSG is in range 0,2 5 1 1 3

Dilution minimize 5 20 1 1 3

Arginine is in range 0 0,5 1 1 3

Tween is in range 0 1 1 1 3

Protein bound maximize -2,4352 2168,75 0,1 1 3

Solutions

Number Protein pHGSH/GSSG Dilution Arginine Tween Protein bound

Desirability

1 2,5 9,5 5 5 0,5 1 442 1

Design-Expert® Software

Desirability

1

0

X1 = A: Protein

X2 = D: Dilution

Actual Factors

B: pH = 9.50

C: GSH/GSSG = 5.00

E: Arginine = 0.50

F: Tween = 1.00

0.50

1.00

1.50

2.00

2.50

5.00

8.75

12.50

16.25

20.00

0.000

0.238

0.475

0.713

0.950

D

es

ira

bil

ity

A: Protein D: Dilution

16PROTAFFIN

Control of Optimised refolding

2010 05 12 HeparinDispl refold run010 1to10 500microL001:10_UV 2010 05 12 HeparinDispl refold run010 1to10 500microL001:10_Cond 2010 05 12 HeparinDispl refold run010 1to10 500microL001:10_Conc 2010 05 12 HeparinDispl refold run010 1to10 500microL001:10_Flow 2010 05 12 HeparinDispl refold run010 1to10 500microL001:10_Fractions 2010 05 12 HeparinDispl refold run010 1to10 500microL001:10_Inject 2010 05 12 HeparinDispl refold run010 1to10 500microL001:10_UV@01,BASEM1

0.0

5.0

10.0

15.0

20.0

25.0

mAU

12.0 14.0 16.0 18.0 20.0 22.0 24.0 26.0 ml

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 2829

18.14

20.64

21.75

2010 05 14 Heparin Displ refold run 2 1to10 500microL001:10_UV 2010 05 14 Heparin Displ refold run 2 1to10 500microL001:10_Cond 2010 05 14 Heparin Displ refold run 2 1to10 500microL001:10_Conc 2010 05 14 Heparin Displ refold run 2 1to10 500microL001:10_Flow 2010 05 14 Heparin Displ refold run 2 1to10 500microL001:10_Fractions 2010 05 14 Heparin Displ refold run 2 1to10 500microL001:10_Inject 2010 05 14 Heparin Displ refold run 2 1to10 500microL001:10_UV@01,BASEM

-1.0

0.0

1.0

2.0

3.0

4.0

mAU

15.0 20.0 25.0 ml

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

17.44

18.46

20.50

21.40

17PROTAFFIN

QbD and Design Space

ICH Q8R2 : Design Space:

The multidimensional combination and interaction of input variables (e.g., material

attributes) and process parameters that have been demonstrated to provide assurance

of quality.

Baysian statistics and Monte Carlo simulation

18PROTAFFIN

Refolding analysis : Monte Carlo Simulation

Process Understanding

19PROTAFFIN

Quality of results depends of analytics

20PROTAFFIN

DoE in Production Process for BioPharmaceuticals

WCB

LB

Acid

Base

Antifoam

Inoculum

Medium

Supernatant

Cell paste

Shake Flask

SF-101

Fermentor

FER-102

Centrifuge

CC-103&104

High Pressure

HP-104

Centrifuge

CC-104

Pellet

Storage

Storage

Acid precipitat ion

Centrifuge

CC-104

Acid

High Pressure

HP-104Centrifuge

CC-104

PelletPellet

Soluble proteins

Filtration

DE-106

CIP buffer

Buffer A

Buffer B

Waste

CIP buffer

Dilut ion buffer

Waste

CIP buffer

Buffer A

Buffer B

Tank

DP-105

Colum,

Col-201

Diafiltration

DF-202

Colum,

Col-203

Colum,

Col-205

Concentration

DF-204

CIP bufferCIP buffer

Buffer A

StorageCIP buffer

Buffer A

Diafiltration

DF-205Dilution

DF-206

Buffer A

Filtration

DE-207

Waste

Waste

Waste Waste

21PROTAFFIN

DoE in BioPharmaceutical Life Cycle

Discovery Preclinical Phase I Phase II Phase III Commercial

R IV Factorial Design : identification of Critical Factor

RSM : Optimisation

R III Factorial Design

(Design Space and

Proven Acceptable Range)

Multivariate Analysis

Process Design Process Continued Process

Qualification Verification

CTD

22PROTAFFIN

Quality by Design : DoE and Risk Assessment

Synergy between Risk Assessment and DoE

Product/Process Risk assessment

Critical questions to be solved

DoE and multivariate statistical analysis

What is really critical ?

How to master the criticality ?

23PROTAFFIN

Quality by Design

– Regulatory : to offer flexibility in the Design Space but...

– New for Biopharmaceutical Industry and still only poorly used

24PROTAFFIN

Design Expert

– Friendly but powerful DoE Software

• Friendly : Training of Colleagues, Customers, Students

Used by Lab Scientists and Technical Staff

• Powerful : Allowed me to Draw, Analyse, Optimise and Qualify

successfully more than 100 Process/Process steps

Possible Improvements :

• Addition of Adjusted R Square / Mallows’ Cp statistics

• Use of other regressions than OLS (Ridge,...)

– Many Thanks to Stat Ease Team for its Responsiveness and Flexibility