Forced Degradation Studies and CQA Determination for Process ...
Forced Degradation Studies and CQA Determination for Process Development Li Tao, Molecular and Analytical Development Biologics Development Bristol-Myers Squibb Company AAPS2015
1
Today’s Talk
Process development and process control for biologics manufacturing
Critical Quality Attributes (CQAs)
Forced degradation of biologics
Biologics Manufacturing Process Biologics are biosynthesized by living organisms through genetic engineering (upstream process)
Chinese Hamster Ovary (CHO), mouse, yeast, E Coli, etc
Transgenic animals (Goats, pigs, chickens, etc.) or plants
Biologics are purified from expressing media (downstream process)
Mainly chromatography steps
Output is drug substance (DS)
Drug substance are converted to the final formulation and presentation format to form drug product (DP)
Filter and fill, dilution, lyophilized, ready to use (RTU), etc
Process Development for Biologics Upstream Cell Line Development Protein Expression
Downstream Protein Purification/ Formulation
• Yield • Cell line purity • Titer • Molecular integrity • Titer • Molecular integrity • Genetic stability • PTMs (glycosylation, etc) • Impurities • Stability
What Constitutes a Well Developed Process? Capable of generating high quality DS or DP consistently with reasonable cost Developing a good biologics manufacturing process requires a deep understanding of:
Comprehensive product profile described by molecular attributes
Degradation pathway of the drug molecule-can be identified rapidly and systematically by forced degradation studies
Structure-function relationship
Product and degradation attributes are controlled during manufacturing and monitored during storage, the most critical ones are CQAs
Molecular Attributes of Biologics Higher order structures/oligomeric state
Monomer
High Molecular Weight (HMW) species – Dimer can be more or less active than monomer
Low Molecular Weight (LMW) species – Usually less or equal active than monomer
Particulates (micron and submicron sizes) – Conceptually loss of bioactivity – Higher risk related to immunogenecity
Molecular Attributes of Biologics Primary structure
From Post-translational modifications (PTMs)
– Glycosylations – N-terminal modifications (N-term pyroglutamation, glycation, carbamylation, gluconoylation, acetylation, etc) – C-terminal truncation or Lys removal – Sequence variants
From common degradation pathways – Oxidation – Deamidation/isomerization – Fragmentation – Disulfide bond scrambling and other forms (-S-, -S-S-S-)
Which Attributes are Critical to Product Quality? (CQA Determination) Particulates √ o
Light obscuration, light scattering, MFI, etc
HMW from aggregation √ o
SEC, AUC, CE-SDS, etc
LMW from fragmentation/truncation √ o
SEC, AUC, CE-SDS, etc
Secondary, tertiary, and quaternary structures (folding/unfolding, domain) √ o
CD, DSC, HDX-MS, etc
o
Controlled by process consistency and molecular platforms rather than by assay
Which Attributes are Critical to Product Quality? (CQA Determination) Glycosylation√ o
Required for effector function such as Antibody‐Dependent Cellular Cytotoxicity (ADCC)
o
May impact pharmacokinetic behavior
o
Rarely changes under normal storage condition
N- or C-terminal variants X o
Less likely to have an impact on bioactivity
Oxidation? o
May or may not impact bioactivity
Deamidation/isomerization? o
May or may not impact bioactivity
Typical Forced Degradation Study Panel Method Condition Thermal 40°C, 4 wk
SE-HPLC
CE-SDS/ SDS-PAGE
Aggre (s)
Aggre (s)
CEX
iCIEF/ Gel-IEF
TPM
Isomer/succi (s)
Isomer/succi (s)
Isomer/succi (s)
Profile change (m)
Profile change (m)
Profile change (m)
Profile change (m)
Deam (s)
Deam (s)
Binding
Bioassay
Loss of Loss of activity (m) activity (s)
(50 mg/mL) Freeze-thaw 5 freeze-thaw cycle (F: -80°D; T: 25°C) (50 mg/mL) Agitation
Aggre (m)
Aggre (m)
Aggre (m)
Aggre (m)
2 days (25°C) (10 mg/mL) Alkaline
Aggre (m)
pH 9, 7 days (25°C) (10 mg/mL) Acidic
Aggre (m)
SS bond cleaving (m)
pH 3.3, 2 days (25°C) (15 mg/mL) H2O2
Profile change (m)
Elo to H2O2 Molar ratio:1/430, 3 hr (25°C) (10 mg/mL) Irradiation ICH (25°C) (10 mg/mL)
Deam (s)
Aggre (s)
Aggre (s) peptide & SS bond cleaving (m)
Profile change (s)
Ox (s)
Profile change (s)
Ox (s)
s and m denote significant and minor changes under each degradation condition, respectively.
Loss of activity (s)
Loss of activity (s)
Forced Degradation Data-Aggregation 6.0
% HMW by CE-SDS
5.0 4.0 3.0
2.0 1.0 0.0
0
12
% HMW by SEC
5.0 4.0 3.0 2.0 1.0 0.0
Ref
Ctr
40C
F-T
Agi
pH9
pH3.3 H2O2
Tbl H2O2
UV
SiOil
Fe
Forced Degradation Data-Bioassay 120
100
% ADCC Activity
80
60
40
20
0
Common Degradation Pathways-Oxidation Surface exposed Met residues are most susceptible and can be used as oxidation markers
Impact of Met oxidation can be evaluated
Generating oxidized samples by H2O2
Measuring oxidation levels on each Met residue by peptide mapping
Assessing bioactivities
Trp residues are usually more resistant to oxidation except for UV-vis and metal mediated reactions
An indicator for light exposure and other reactive species
Root cause of many discoloration
Less of a process control issue, since biologics are by default protected from light and heavy metals
Met Oxidation Measured by Peptide Mapping and LC/MS
Forced Degradation Data-Bioassay 120
100
% ADCC Activity
80
60
40
20
0
Forced Oxidation– Case Study 1
Samples Control 0.1% H2O2 3hr 0.1% H2O2 6hr 0.1% H2O2 24hr
%H3ox (Met a)
%H9ox (Met b)
%H19ox (Met c)
%H39ox (Met d)
%L1ox Bioactivity (Met e) (%)
1.0
0.7
1.6
0.6
0.8
107
1.2
1.0
37.5
22.2
1.1
93
1.1
1.3
65.4
40.4
1.2
88
1.1
1.6
97.3
88.9
1.2
73
Summary of Forced Oxidation – Case Study 1 Met a (on HC-CDR1) is not susceptible to oxidation Bioactivity is not affected by oxidation that is relevant to normal MFG and storage conditions Extreme oxidative conditions result in nearcomplete oxidation of Met c, but bioactivity is largely retained Met oxidation is not a CQA for this mAb
Bioassay Potency vs Met Oxidation (Case Study 2)
Common Degradation PathwaysDeamidation Mostly involves Asn residues going through succinimide formation at close to neutral pH for biologics Conformational restriction is a main factor affecting site specific deamidation rate Is it a CQA? Asn
IsoAsp
Succinimide
Asp
Common Degradation PathwaysDeamidation
Analysis of the Forced Degradation Samples by iCIEF 40 35 30
Area %
25 20 15 10 5 0
Acidic Group
Forced Degradation Data-Bioassay 120
100
% ADCC Activity
80
60
40
20
0
iCIEF is More Sensitive Than Peptide Mapping for Deamidation Measurement SampleName: 584309-3
0 Week, 45 °C
Absorbance
0.30
0.20
0.10
0.00 SampleName: 584310-4
2 Week, 45 °C
0.30
Absorbance
0.25
Peptide Mapping Results
0.20 0.15 0.10
Sample
Deam Marker
T0
2.6%
45C_2m
2.6%, 2.7%
45C_3m
2.6%, 2.6%
control
2.7%
0.05 0.00 SampleName: 584311-5
4 Week, 45 °C
0.20
Absorbance
0.15
0.10
0.05
0.00 SampleName: 589475-6 0.14
6 Week, 45 °C
Absorbance
0.12 0.10 0.08 0.06 0.04 0.02 0.00 5.00
5.50
6.00
6.50
7.00
7.50
8.00
8.50
9.00
9.50
10.00
10.50
11.00
11.50
12.00
12.50 Minutes
13.00
13.50
14.00
1
Cumulative Effect of Charge Variants Number of deamidation sites: N Level of deamidation at each site: di
% of molecules without deamidation: = (1-d1)(1-d2)∙∙∙(1-dN) d
N
non-deam
1%
10
90%
2%
10
82%
5%
10
60%
1%
20
82%
2%
20
67%
5%
20
36%
Impact of HMW has to be Evaluated in the Forced Degradation Studies
Conclusions • CQAs identification is critical for biologics process development and manufacturing process control
• Many CQAs are universal for biologics • Some require molecule/mechanism specific studies to identify • Those related to degradation require even more comprehensive studies to determine • Some of the common degradation pathways such as oxidation and deamidation are not CQAs for monoclonal antibodies. • Bioactivity assays may not be sufficient for assessing impact of degradation
Acknowledgment
Rong Yang, Jinmei Fu, Jacob Bongers, Yunping Huang Liji Zhu, Ryman Navoa, Yingchen Cheng, Yemin Xu Wenjun Zhang, Ming Zeng, Wei Wu, Richard Ludwig Tapan Das, Reb Russell, Morrey Atkinson
ASMS'13 Tao
27
Thank You!
1
Today’s Talk
Process development and process control for biologics manufacturing
Critical Quality Attributes (CQAs)
Forced degradation of biologics
Biologics Manufacturing Process Biologics are biosynthesized by living organisms through genetic engineering (upstream process)
Chinese Hamster Ovary (CHO), mouse, yeast, E Coli, etc
Transgenic animals (Goats, pigs, chickens, etc.) or plants
Biologics are purified from expressing media (downstream process)
Mainly chromatography steps
Output is drug substance (DS)
Drug substance are converted to the final formulation and presentation format to form drug product (DP)
Filter and fill, dilution, lyophilized, ready to use (RTU), etc
Process Development for Biologics Upstream Cell Line Development Protein Expression
Downstream Protein Purification/ Formulation
• Yield • Cell line purity • Titer • Molecular integrity • Titer • Molecular integrity • Genetic stability • PTMs (glycosylation, etc) • Impurities • Stability
What Constitutes a Well Developed Process? Capable of generating high quality DS or DP consistently with reasonable cost Developing a good biologics manufacturing process requires a deep understanding of:
Comprehensive product profile described by molecular attributes
Degradation pathway of the drug molecule-can be identified rapidly and systematically by forced degradation studies
Structure-function relationship
Product and degradation attributes are controlled during manufacturing and monitored during storage, the most critical ones are CQAs
Molecular Attributes of Biologics Higher order structures/oligomeric state
Monomer
High Molecular Weight (HMW) species – Dimer can be more or less active than monomer
Low Molecular Weight (LMW) species – Usually less or equal active than monomer
Particulates (micron and submicron sizes) – Conceptually loss of bioactivity – Higher risk related to immunogenecity
Molecular Attributes of Biologics Primary structure
From Post-translational modifications (PTMs)
– Glycosylations – N-terminal modifications (N-term pyroglutamation, glycation, carbamylation, gluconoylation, acetylation, etc) – C-terminal truncation or Lys removal – Sequence variants
From common degradation pathways – Oxidation – Deamidation/isomerization – Fragmentation – Disulfide bond scrambling and other forms (-S-, -S-S-S-)
Which Attributes are Critical to Product Quality? (CQA Determination) Particulates √ o
Light obscuration, light scattering, MFI, etc
HMW from aggregation √ o
SEC, AUC, CE-SDS, etc
LMW from fragmentation/truncation √ o
SEC, AUC, CE-SDS, etc
Secondary, tertiary, and quaternary structures (folding/unfolding, domain) √ o
CD, DSC, HDX-MS, etc
o
Controlled by process consistency and molecular platforms rather than by assay
Which Attributes are Critical to Product Quality? (CQA Determination) Glycosylation√ o
Required for effector function such as Antibody‐Dependent Cellular Cytotoxicity (ADCC)
o
May impact pharmacokinetic behavior
o
Rarely changes under normal storage condition
N- or C-terminal variants X o
Less likely to have an impact on bioactivity
Oxidation? o
May or may not impact bioactivity
Deamidation/isomerization? o
May or may not impact bioactivity
Typical Forced Degradation Study Panel Method Condition Thermal 40°C, 4 wk
SE-HPLC
CE-SDS/ SDS-PAGE
Aggre (s)
Aggre (s)
CEX
iCIEF/ Gel-IEF
TPM
Isomer/succi (s)
Isomer/succi (s)
Isomer/succi (s)
Profile change (m)
Profile change (m)
Profile change (m)
Profile change (m)
Deam (s)
Deam (s)
Binding
Bioassay
Loss of Loss of activity (m) activity (s)
(50 mg/mL) Freeze-thaw 5 freeze-thaw cycle (F: -80°D; T: 25°C) (50 mg/mL) Agitation
Aggre (m)
Aggre (m)
Aggre (m)
Aggre (m)
2 days (25°C) (10 mg/mL) Alkaline
Aggre (m)
pH 9, 7 days (25°C) (10 mg/mL) Acidic
Aggre (m)
SS bond cleaving (m)
pH 3.3, 2 days (25°C) (15 mg/mL) H2O2
Profile change (m)
Elo to H2O2 Molar ratio:1/430, 3 hr (25°C) (10 mg/mL) Irradiation ICH (25°C) (10 mg/mL)
Deam (s)
Aggre (s)
Aggre (s) peptide & SS bond cleaving (m)
Profile change (s)
Ox (s)
Profile change (s)
Ox (s)
s and m denote significant and minor changes under each degradation condition, respectively.
Loss of activity (s)
Loss of activity (s)
Forced Degradation Data-Aggregation 6.0
% HMW by CE-SDS
5.0 4.0 3.0
2.0 1.0 0.0
0
12
% HMW by SEC
5.0 4.0 3.0 2.0 1.0 0.0
Ref
Ctr
40C
F-T
Agi
pH9
pH3.3 H2O2
Tbl H2O2
UV
SiOil
Fe
Forced Degradation Data-Bioassay 120
100
% ADCC Activity
80
60
40
20
0
Common Degradation Pathways-Oxidation Surface exposed Met residues are most susceptible and can be used as oxidation markers
Impact of Met oxidation can be evaluated
Generating oxidized samples by H2O2
Measuring oxidation levels on each Met residue by peptide mapping
Assessing bioactivities
Trp residues are usually more resistant to oxidation except for UV-vis and metal mediated reactions
An indicator for light exposure and other reactive species
Root cause of many discoloration
Less of a process control issue, since biologics are by default protected from light and heavy metals
Met Oxidation Measured by Peptide Mapping and LC/MS
Forced Degradation Data-Bioassay 120
100
% ADCC Activity
80
60
40
20
0
Forced Oxidation– Case Study 1
Samples Control 0.1% H2O2 3hr 0.1% H2O2 6hr 0.1% H2O2 24hr
%H3ox (Met a)
%H9ox (Met b)
%H19ox (Met c)
%H39ox (Met d)
%L1ox Bioactivity (Met e) (%)
1.0
0.7
1.6
0.6
0.8
107
1.2
1.0
37.5
22.2
1.1
93
1.1
1.3
65.4
40.4
1.2
88
1.1
1.6
97.3
88.9
1.2
73
Summary of Forced Oxidation – Case Study 1 Met a (on HC-CDR1) is not susceptible to oxidation Bioactivity is not affected by oxidation that is relevant to normal MFG and storage conditions Extreme oxidative conditions result in nearcomplete oxidation of Met c, but bioactivity is largely retained Met oxidation is not a CQA for this mAb
Bioassay Potency vs Met Oxidation (Case Study 2)
Common Degradation PathwaysDeamidation Mostly involves Asn residues going through succinimide formation at close to neutral pH for biologics Conformational restriction is a main factor affecting site specific deamidation rate Is it a CQA? Asn
IsoAsp
Succinimide
Asp
Common Degradation PathwaysDeamidation
Analysis of the Forced Degradation Samples by iCIEF 40 35 30
Area %
25 20 15 10 5 0
Acidic Group
Forced Degradation Data-Bioassay 120
100
% ADCC Activity
80
60
40
20
0
iCIEF is More Sensitive Than Peptide Mapping for Deamidation Measurement SampleName: 584309-3
0 Week, 45 °C
Absorbance
0.30
0.20
0.10
0.00 SampleName: 584310-4
2 Week, 45 °C
0.30
Absorbance
0.25
Peptide Mapping Results
0.20 0.15 0.10
Sample
Deam Marker
T0
2.6%
45C_2m
2.6%, 2.7%
45C_3m
2.6%, 2.6%
control
2.7%
0.05 0.00 SampleName: 584311-5
4 Week, 45 °C
0.20
Absorbance
0.15
0.10
0.05
0.00 SampleName: 589475-6 0.14
6 Week, 45 °C
Absorbance
0.12 0.10 0.08 0.06 0.04 0.02 0.00 5.00
5.50
6.00
6.50
7.00
7.50
8.00
8.50
9.00
9.50
10.00
10.50
11.00
11.50
12.00
12.50 Minutes
13.00
13.50
14.00
1
Cumulative Effect of Charge Variants Number of deamidation sites: N Level of deamidation at each site: di
% of molecules without deamidation: = (1-d1)(1-d2)∙∙∙(1-dN) d
N
non-deam
1%
10
90%
2%
10
82%
5%
10
60%
1%
20
82%
2%
20
67%
5%
20
36%
Impact of HMW has to be Evaluated in the Forced Degradation Studies
Conclusions • CQAs identification is critical for biologics process development and manufacturing process control
• Many CQAs are universal for biologics • Some require molecule/mechanism specific studies to identify • Those related to degradation require even more comprehensive studies to determine • Some of the common degradation pathways such as oxidation and deamidation are not CQAs for monoclonal antibodies. • Bioactivity assays may not be sufficient for assessing impact of degradation
Acknowledgment
Rong Yang, Jinmei Fu, Jacob Bongers, Yunping Huang Liji Zhu, Ryman Navoa, Yingchen Cheng, Yemin Xu Wenjun Zhang, Ming Zeng, Wei Wu, Richard Ludwig Tapan Das, Reb Russell, Morrey Atkinson
ASMS'13 Tao
27
Thank You!
