www.cookpharmica.com
CONFIDENTIAL—COOK PHARMICA LLC
QbD Based Process Development Strategies
for Antibodies
Presented by Kumar Dhanasekharan, PhD
Director of Process Development
ABOUT COOK PHARMICA
CONFIDENTIAL—COOK PHARMICA LLC 2
• CONTRACT DEVELOPMENT AND MANUFACTURING ORGANIZATION
• WHOLLY-OWNED SUBSIDIARY OF COOK MEDICAL
• LEGACY OF LIFE SCIENCES INNOVATION SINCE 1963
• 900,000 FT2 (83,600 M2) FACILITY IN BLOOMINGTON, INDIANA - USA
CLINICAL AND COMMERCIAL
DEVELOPMENT
PROCESS DEVELOPMENT
ANALYTICAL DEVELOPMENT
FORMULATION DEVELOPMENT
DRUG SUBSTANCE
CELL CULTURE MANFACTURING
CAPACITY TO 250 L, 600 L, (2) 2,500 L
DRUG PRODUCT
BARRIER ISOLATOR TECHNOLOGY
VIAL FILLING
LYOPHILIZATION
SYRINGE FILLING
SAFETY DEVICE AND AUTOINJECTOR ASSEMBLY
BLISTER THERMOFORMING
KITTING AND CARTONING
3
THE ONE SOURCE, ONE LOCATION MODEL
CONFIDENTIAL—COOK PHARMICA LLC
GROWTH IN DIVERSITY OF PROJECTS AND CLIENTS
4 CONFIDENTIAL—COOK PHARMICA LLC
Client Partners
• Early Stage and Evolving Biotech
• Large Pharmaceutical Companies
• Biosimilar Ventures and Established Subsidiaries
• Sub-contractor for US Government (BARDA)
Molecule/ Product Classes
• Mabs and Fusion Proteins (Innovator, Biosimilar, Biobetter)
• Recombinant Therapeutics Proteins (Enzymes, Interferons)
• Peptides, Oligonucleotides, and Proteoliposomes
• Vaccines (drug product only)
Lifecycle Phase
• Preclinical Development
• Phase I/ II/ III Development, Reg./ Tox. and cGMP Manufacture
• PV and Continued verification
• Commercial Manufacturing
• mAb - Molecule to Market Lifecycle
• Product Understanding
– Criticality Assessments
– Preliminary Analytical Control
Strategy
• Process Understanding
– Risk Management
– Process Characterization
• Product-Process Mapping
– Design Space
– CPP-CQA Linkages
• Control Strategy – PV Readiness
with well-defined/ well-
characterized process
PRESENTATION OUTLINE
5
Product Understanding
Process Understanding
Product-Process
Mapping
Control Strategy
Validation Lifecycle
CONFIDENTIAL—COOK PHARMICA LLC
Molecule Target Product
Profile (TPP)
Process Definition
Clinical Manufacture
(Phase I, Phase II)
Process Optimization (Process Lock,
Phase III)
Process Installation
and Startup
Process Validation Campaign
Commercial Manufacturing and Continued
Verification
• Process Definition (Phase I, Phase II) – In-Depth Product Characterization but little or no process characterization
– Preliminary Control Strategy and Process Definition for meeting TPP
– Initial scale-up for GMP Manufacture
• Process Optimization (Phase III) – In-depth process characterization
– DOE Driven Studies and establish Design Space
– Final Scale-up and process lock
– Establish Control Strategy
MOLECULE TO MARKET
6
Preliminary Control Strategy
Process Characterization
CONFIDENTIAL—COOK PHARMICA LLC
Prior to early development, in-depth molecule
characterization supports criticality risk assessment of
product quality attributes and insight into stability
PRODUCT UNDERSTANDING: IN-DEPTH CHARACTERIZATION
7
• Antigen Binding
• Cell Based Binding Functional Analysis
• N-Glycan Analysis/ Glycosylation • N-Terminal Sequencing • C-Terminal Lysine • Post translational modification(PTM) analytics • Peptide Mapping (LC-MS)
Structural Analysis
CONFIDENTIAL—COOK PHARMICA LLC
Based on a criticality risk assessment, a testing plan based on criticality of product attributes is established (One or more analytical methods associated with each CQA)
PRELIMINARY CONTROL STRATEGY
8 CONFIDENTIAL—COOK PHARMICA LLC
• A key strategy for success and quick turnaround is an established development platform
– Standardized processes with established design space
– Well-characterized standard equipment, resins etc. with established scale-up criteria and engineering design space
• An established technology platform allows for significantly reduced parameter screening to develop a process for a class of molecules that meets product TPP and provides assurance of quality
– No loss of flexibility with appropriate technology platform
PROCESS DEFINITION – A TECHNOLOGY PLATFORM APPROACH
9
Concept in harmony with QbD – A Systematic Approach to Leveraging Prior Knowledge…
Operating Space
Design Space
Characterization Space
mAb - A mAb - B mAb - C mAb - D
CONFIDENTIAL—COOK PHARMICA LLC
Cell Culture Platform
– Standard cell expansion processes (e.g. WAVE® bags)
– Standard Off-the-shelf media options with baseline feed strategy
– 2L and 20L platform with established design charts for scale-up
– Standardized processes: Feed Strategy, Inoculation Density, Temperature shift, pH/ pCO2 control, Gassing Strategy etc.
– Minimal experimentation to establish high titer process and meet critical product attributes and target product profile
PROCESS DEFINITION – UPSTREAM DEVELOPMENT
10
2L Platform 20L Platform
Expansion
WAVE Bags
CONFIDENTIAL—COOK PHARMICA LLC
Scale-up
Downstream Platform
• Standard 2 or 3 column process
• Well-characterized resin library and established vendors
• Established column packing and operation and scale-up criteria
• Established viral filtration technologies
• Minimal screening experiments to establish baseline process to meet in-process controls and product attributes
PROCESS DEFINITION – DOWNSTREAM DEVELOPMENT
11
• Protein A • Low pH Viral
Inactivation
• Purification • (CEX)
• Polishing • (AEX, MMC, HIC)
• Viral
Filtration
• UF/DF
CONFIDENTIAL—COOK PHARMICA LLC
TYPICAL DEVELOPMENT PROGRAM FOR EARLY PHASE MOLECULES
12 CONFIDENTIAL—COOK PHARMICA LLC
(Months)
Select base media Determine Feed Strategy Optimize 2L Process
Confirmation at 20L Scale
Tox. Material Production
Optimize downstream process Select Optimal Formulation
Base Media Selection (12x500mL Shake
Flask Studies) Feed Strategy
(36x500mL Shake Flask Studies)
Cell Culture Process Development
(18x12L Bioreactor runs)
Downstream Development (2- or 3-column process)
Pre-formulation material production
(1x20L)
Confirmatiory Runs
(4x2L runs)
Cell Culture Scale-Up (4x20L runs)
Downstream Confirmatory
runs
Downstream Scale-up
Tox. Material (1x250L SUB or 3x20L runs)
Viral Clearance Studies
Formulation Development and Stability (Screen 4-6 formulations)
Analytical Package for mAb Characterization and Release (Total # of Samples)
1 2 3 4 5 6 7 8 9 10
generates cell line and identifies top
clones
Clone-screening; basal and feed
media screening
(Shake flask studies)
Cook 2L Process Shakedown:
4 x 2L runs
Downstream 3-column platform
Cook 2L Process Definition:
2 x 2L runs
1 x 20L run
Cook 20L Process Scale-
up:
4 x 2L runs
4 x 20L runs
• Background – Collaboration with SELEXIS for a biosimilar trastuzumab (model molecule), using a
client-owned SELEXIS cell line for demonstrating proof-of-concept.
• Objective – Create scalable process via a technology platform approach for commercial-ready
titers with minimal development
• Approach
CASE STUDY – RAPID DEVELOPMENT OF A BIOSIMILAR
13 CONFIDENTIAL—COOK PHARMICA LLC
MAMMALIAN CELL LINE DEVELOPMENT PLATFORM
Candidate Clones
Productivity
assay
Functional
assay
Manufacturing
Preclinic/tox
Clonal Cell Lines
Fed batch process
cGMP manufacturing
Clinical trial supply
Market supply
Research cell bank
Generation of high performance and
stable cell lines using the
SUREtechnology PlatformTM for
cGMP manufacturing
SGE high-productivity expression vector
Single-cell cloning in chemically-defined media
Suspension growth in chemically-defined basal media (commercial media)
Optimized feed strategy (commercial feed)
Robust growth to high cell densities
1000 cells embedded in semi-
solid medium
SUREtech vectors CMO Pool Single Cell Clone
Round 1 Single Cell Clone
Round 2
7-8 weeks 7-8 weeks
ClonePix 100 clones Transfection
CONFIDENTIAL—COOK PHARMICA LLC
INITIAL SHAKE FLASK STUDIES: FEED STRATEGY SCREENING WITH FULL FACTORIAL DOE
15
Variables
Feed Concentration 1% 2% NA
Feed Frequency Daily Bi-daily NA
Feed Timing D3-7 D3-12/13 D3-18
Non-linear relation between total feed amount and production 16-20% best range for good titer with reduction in wasted feed (cost)
JMP analysis indicates daily feeds D3-18 best
16%
12 flask DoE full factorial
Concentration Frequency Timing
1 Bi-daily D3-7
1 Bi-daily D3-12/13
2 Daily D3-7
1 Bi-daily D3-18
2 Daily D3-12/13
1 Daily D3-7
2 Bi-daily D3-7
1 Daily D3-18
2 Bi-daily D3-18
2 Daily D3-18
1 Daily D3-12/13
2 Bi-daily D3-12/13
CONFIDENTIAL—COOK PHARMICA LLC
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Tite
r (g
/L)
Time (Days)
Flask 1, Bi-daily, D3-7, 1%
Flask 2, Bi-daily, D3-13, 1%
Flask 3, Daily, D3-7, 2%
Flask 4, Bi-daily, D3-18, 1%
Flask 5, Daily, D3-12, 2%
Flask 6, Daily, D3-7, 1%
Flask 7, Bi-daily, D3-7, 2%
Flask 8, Daily, D3-18, 1%
Flask 9, Bi-Daily, D3-18, 2%
Flask 10, Daily, D3-18, 2%
Flask 11, Daily, D3-12, 1%
Flask 12, Bi-daily, D3-13, 2%
Flask 13, FS2 F10 condition
RESULTS – TITER DRIVEN OPTIMIZATION OF FEED STRATEGY
16 CONFIDENTIAL—COOK PHARMICA LLC
Daily Feed of 1% and 2% are top performers for titer
2L PROCESS SHAKEDOWN – UNIVARIATE STUDY WITH TEMPERATURE SHIFT
17
2.8
2.85
2.9
2.95
3
3.05
3.1
3.15
3.2
3.25
31.5 32 32.5 33 33.5 34 34.5
Tite
r (g
/L)
Temperature (C)
Peak titer vs. Shift temp
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0.0E+00
5.0E+06
1.0E+07
1.5E+07
2.0E+07
2.5E+07
3.0E+07
3.5E+07
0 5 10 15 20 25
Via
bili
ty (
%)
Via
ble
Ce
ll D
en
sity
(C
ells
/mL)
Time (Days)
A temperature shift of 32C seems more favorable for titer.
CONFIDENTIAL—COOK PHARMICA LLC
2L PROCESS DEVELOPMENT: DEFINE PH AND GLUCOSE LEVELS
18
Results: Maintaining pH above 6.8 early benefits cell growth, but
shortens culture longevity; Standard glucose levels gives better titer
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0.0E+00
2.0E+06
4.0E+06
6.0E+06
8.0E+06
1.0E+07
1.2E+07
1.4E+07
1.6E+07
1.8E+07
0 2 4 6 8 10 12 14 16 18 20 22 24
Via
bili
ty (
%)
Via
ble
Ce
ll D
en
sity
(C
ells
/mL)
Time (Days)
0
0.5
1
1.5
2
2.5
3
3.5
4
0 2 4 6 8 10 12 14 16 18 20 22 24Ti
ter
(g/L
)
Time (Days)
2L-07: Std. Gluc; pH 6.6-7.2
2L-08: Std. Gluc, pH 6.8-7.2
2L-09: Lo Gluc; pH 6.6-7.2
2L-10: Lo gluc; pH 6.8-7.2
Previous data: S3 F8
Factors Tested:
- Low Glucose vs Std. Glucose Levels
- pH range 6.6 – 7.2 vs 6.8 – 7.2
CONFIDENTIAL—COOK PHARMICA LLC
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0 2 4 6 8 10 12 14 16 18 20
Tite
r, (
g/L)
Time (Days)
2L-08 Golden Batch20L-05 1% D3-1820L-06 2% D3-12, 1% D13-1820L-07 2% D3-12, 1% D13-1820L-08 1% D3-18
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0.0E+00
2.0E+06
4.0E+06
6.0E+06
8.0E+06
1.0E+07
1.2E+07
1.4E+07
1.6E+07
1.8E+07
2.0E+07
0 2 4 6 8 10 12 14 16 18 20V
iab
ility
, %
Via
ble
Ce
ll D
en
sity
(C
ells
/mL)
Time (Days)
• 2L-08 baseline condition – “Golden Batch”
• Scale up to 20L
PROCESS CONFIRMATION AND SCALE-UP
19 CONFIDENTIAL—COOK PHARMICA LLC
0.0
2.0
4.0
6.0
8.0
10.0
0 2 4 6 8 10 12 14 16 18 20
Glu
cose
, (g/
L)
Time (Days)
2L-08 Golden Batch20L-05 1% D3-1820L-06 2% D3-12, 1% D13-1820L-07 2% D3-12, 1% D13-1820L-08 1% D3-18
0.0
1.0
2.0
3.0
4.0
5.0
6.0
0 2 4 6 8 10 12 14 16 18 20
Lact
ate
, (g/
L)
Time (Days)
PROCESS SCALE-UP: METABOLITE PROFILES
20
6.0
6.2
6.4
6.6
6.8
7.0
7.2
7.4
7.6
7.8
8.0
0 2 4 6 8 10 12 14 16 18 20
pH
Time (Days)
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
0 2 4 6 8 10 12 14 16 18 20
pC
O2
, (m
mH
g)
Time (Days)
CONFIDENTIAL—COOK PHARMICA LLC
• Optimized seed train in 2L
• Diluted vs concentrated feeds
• Additional work required for optimization of 20L process
HIGH TITER PROCESS DEVELOPMENT – WORK IN PROGRESS
21
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
0 2 4 6 8 10 12 14 16 18 20 22 24Ti
ter,
(g/
L)
Time (Days)
2L-08 Golden Batch2L-15 2% D3-12, 1% D13-182L-16 1% D3-182L-17 1% D3-182L-18 2% D3-12, 1% D13-18
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0.0E+00
5.0E+06
1.0E+07
1.5E+07
2.0E+07
2.5E+07
0 2 4 6 8 10 12 14 16 18 20 22 24
Via
bili
ty ,
%
Via
ble
Ce
ll D
en
sity
(C
ells
/mL)
Time (Days)
CONFIDENTIAL—COOK PHARMICA LLC
Standard 3-column process
DOWNSTREAM PROCESS – DEVELOPMENT AND SCALE-UP RESULTS
22
Process Step Pool Conc. (g/L) Pool Vol. (L) Total g Step Yield
(%) SEC-HPLC
Monomer (%) HCP by ELISA
(ng/mg)
Clarified Harvest 1.59 18.91 30.61 100 NA NA
MabSelect Sure/Viral Inactivation Cycle 1 4.29 3.4 14.57 93.8 96.4 1075.4
MabSelect Sure/Viral Inactivation Cycle 2 4.22 3.25 13.72 94.4 96.6 NA
Combined Viral Inactivation Pool* 4.26 5.67 24.15 NA 95.1 962.8
POROS XS Pool** 6.09 3.4 20.73 87.5 99.1 3.9
POROS Q Pool 1.02 17.76 18.12 99.7 99.9 < 1
Viral Filtration Pool 0.96 19 18.24 100.7 99.8 < 1
UFDF Pool 21.89 0.82 18.06 99.0 99.1 < 0.05
Results from a pilot scale downstream process for a 20L scale bioreactor run:
CONFIDENTIAL—COOK PHARMICA LLC
RS – Innovator 20L Run
Monomer
RESULTS: SIZE VARIANTS BY SEC-HPLC
CONFIDENTIAL—COOK PHARMICA LLC
RESULTS: CHARGE VARIANTS BY CIEF
24 CONFIDENTIAL—COOK PHARMICA LLC
RESULTS: GLYCOSYLATION ANALYSIS (N-GLYCANS WITH FL-HPLC)
25 - fucose - mannose - galactose - N-acetyl-D-glucosamine - Sialic acid -
15.00 17.50 20.00 22.50 25.00 27.50 30.00 32.50 35.00 38.00
-2.0
0.0
2.5
5.0
7.5
10.0
12.5
15.0
17.5
20.0
22.5
25.0
G 0 - G
l c N
A c
G 0
G 0 F
M 5
G 0 F
+ G
l c N
A c
G 1
G 1 F
a
G 1 F
b
G 2
G 2 F
G 2 F
+ S
A
G 2 F
+ 2
S A
min
LU
CONFIDENTIAL—COOK PHARMICA LLC
Innovator – Black 20L Run - Blue
Molecule Target Product
Profile (TPP)
Process Definition
Clinical Manufacture
(Phase I, Phase II)
Process Optimization (Process Lock,
Phase III)
Process Installation
and Startup
Process Validation Campaign
Commercial Manufacturing and Continued
Verification
• Process Definition (Phase I, Phase II) – Little or no process characterization
– Preliminary Control Strategy and Process Definition for meeting TPP
– Initial scale-up for GMP Manufacture
• Process Optimization (Phase III) – In-depth process characterization
– DOE Driven Studies and establish Design Space
– Final Scale-up and process lock
– Establish Control Strategy
MOLECULE TO MARKET
26
Process Characterization
CONFIDENTIAL—COOK PHARMICA LLC
MONOCLONAL ANTIBODY CHARACTERIZATION APPROACH
27 CONFIDENTIAL—COOK PHARMICA LLC
Process Conformance
(2-4 2L runs, 2-4 20L runs)
Cell Culture Scale-up Verification runs
(6x20L runs)
Cell Culture Process Characterization (36x2L Bioreactor runs)
Downstream Scale-up Verification runs
Downstream Process
Conformance
Protein A Resin-Reuse Studies
Downstream Characterization (12-15 studies/step)
Column 1
Viral Clearance Studies
Analytical Package for mAb characterization and release (Total # of Samples)
Manufacturing Readiness
Column 2
Column 3
UF/DF
(Months)
1 2 3 4 5 6 7 8 9 10
PV Campaign Execution
Process Characterization (4 months) Demonstrated Pilot Scale Runs
MONOCLONAL ANTIBODY CHARACTERIZATION APPROACH
EXAMPLE PROCESS CHARACTERIZATION DOE FOR A CEX STEP FOR TRASTUZUMAB
• Cation Exchange Chromatography – POROS XS ® Resin in Bio-Rad small-
scale column (1.2 – 1.3 mL CV)
• Method – Equilibration: 10 mL of 50 mM Sodium
Acetate
– Load: pH and conductivity adjusted
material
– Equilibration Wash: 10 mL of 50 mM
Sodium Acetate
– Salt Strip: 10 mL of 50 mM Sodium
Acetate, 500 mM NaCl
Column # Pattern pH Conductivity Load Rate
1 −−− 4.5 3 40
2 000 5 4.5 60
3 +++ 5.5 6 80
4 +−+ 5.5 3 80
5 +−− 5.5 3 40
6 −++ 4.5 6 80
7 −−+ 4.5 3 80
8 −+− 4.5 6 40
9 ++− 5.5 6 40
Full Factorial Screening Design: • Factors: pH, Conductivity, Load Rate • Response: Yield, CHO-HCP Level
CONFIDENTIAL—COOK PHARMICA LLC
Target Operating Range: pH operating Space based on maximum desirability. Acceptable range defined depending on desired HCP clearance
(Additional Optimization DOE studies required)
STATISTICAL ANALYSIS: PREDICTION PROFILER – MAXIMIZING DESIRABILITY
CONFIDENTIAL—COOK PHARMICA LLC
DEFINE CONTROL STRATEGY - MANUFACTURING READINESS FOR PV
30
CQAs
In-Process Controls
Critical process Parameters
Key Parameters
nKPPs, Process Indicators, etc.
Control Strategy
Design Space Process
Characterization
Risk Assessment of facility, equipment,
process etc. feeds into control strategy
• Upon completion of characterization studies, a Control Strategy is defined suitable for process validation – Define Acceptable Ranges for critical
process parameters (CPPs) and Key Process Parameters (KPPs)
– Define specifications and in-process controls and limits for all CQAs
CONFIDENTIAL—COOK PHARMICA LLC
• A flexible technology platform allows for rapid
development for early phase clinical manufacturing
– Leverage prior knowledge and experience
– Reduced parameter screening
– Well-characterized equipment, resin library and methods
• Phase-appropriate QbD as a critical enabler for
successful validation and robust process – molecule to
market
– Process characterization as part of late stage development
SUMMARY
31 CONFIDENTIAL—COOK PHARMICA LLC
• Cell Culture Team – Claudia Berdugo
– Xiaoming Liu
• Purification Team – Carl Richey
– Leon Xu
– Ben Kester
• Analytical Team – Todd Stone
– Zaneer Segu
– Andrew McKee
– Spencer Beard
– Alex Rostovtsev
ACKNOWLEDGMENTS
32 CONFIDENTIAL—COOK PHARMICA LLC
• Sponsors – Victor Vinci – CSO and
VP, Cook Pharmica
– Igor Fisch – CEO, SELEXIS
• Collaborators at SELEXIS – David Calabrese
– Valerie LeFourn
– Pierre-Alain Girod