use of doe in biopharmaceutical industry protaffin ......3 protaffin 22 years of experience with doe...
TRANSCRIPT
-
Use of DoE in Biopharmaceutical Industry
ProtAffin Biotechnologie AG
3d European DoE user Meeting, Lucerne, 1st of June 2010
-
2PROTAFFIN
Alain Poncin ([email protected])
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:[email protected]
-
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