for PS-20
Control Strategy for PS-80: Using Scientific Rationale AAPS National Biotechnology Conference Boston May 16th, 2016
Tween: Fatty Acid Esters of Polyoxyethylene Sorbitan The commercially available PS-20 and PS-80 are a structurally variable mixture of different fatty acid esters and head groups
w+x+y+z refers to the total number of oxyethylene subunits on each surfactant molecule. 2
Representative total ion current profiles and reconstructed ion chromatogram (RIC) for PS-20 >100 Features with m/z between 500-1000 in PS-80 Lauric acid (MW: 227) is the primary fatty acid in PS-20 Oleic Acid is the primary fatty acid in PS-80
Borisov, Oleg V., Junyan A. Ji, Y. John Wang, Felix Vega, and Victor T. Ling. Analytical chemistry 83, no. 10 (2011): 3934-3942.
Tween is a complex mixture of species, and its level of purity is important.
Why put Polysorbates into a Formulation? • Polysorbates are commonly used in biotherapeutic formulations to prevent any protein adsorption and subsequent loss at the product contacting surfaces such as filters, tubing, components of primary containers, glass vials, rubber stoppers, siliconized prefilled syringes, IV bags, etc. • Polysorbates are also commonly used to protect against interfacial stresses encountered during manufacturing processes such as filtration, pumping and filling, shaking and agitation.
• They may also prevent aggregation and proteinsilicone particle formation in prefilled syringes.
Polysorbate Degradation • Polysorbates are known to degrade by hydrolysis and auto-oxidation. M. Donbrow, E. Azaz, A. Pillersdorf, Autoxidation of polysorbates, J. Pharm. Sci. 67 (1978) 1676-1681.
• The rate of hydrolysis of Polysorbates in a formulation that was stored at 5oC and kept at a pH of 5.5 over the shelf-life of the DP was found to be very slow. R.S.K.Kishore, A.Pappenberger, I.B.Dauphin, A Ross, B. Buergi, A. Staempfli, H.C.Mahler, Degradation of polysorbate 20 and 80: Studies on thermal auto-oxidation in bulk and hydrolysis in formulations, J. Pharm. Sci. 100 (2011) 721-731.
• Proper storage and use of PS-80 can minimize oxidation
General Practices for PS-80 • To minimize auto-oxidation during storage, PS-80 should be stored with an overlay of an inert gas such as nitrogen each time a bottle is opened. • A good practice for manufacturing of clinical formulations is to purchase PS-80 in quantities that will be used once and any excess discarded.
•SECRP ELSD- no sample prep. - Genentech1-2 & Pfizer3 •Anion Exchange CAD and/or ELSD – BI & Pfizer •Degradation productsELSD - Janssen (J&J)6 •SPE: PS recovery, high prot. - Amgen8 & Janssen (J&J)6 •Technique ComparisonPfizer10
Detection Techniques
• PS degradation does not have a potential influence on the stability of DP - Roche9 • Hydrogen peroxide does not result in DP oxidation - Bayer4 • PS degradation product shows no impact on CQAs - Pfizer10
Impact of Polysorbate Degradation
Polysorbate (PS) in the Biotherapeutic Industry
• Lipoprotein lipase, and 120 other “problem HCPs” - Univ. of Delaware5 • Lipid-specific mechanism proposal - Janssen (J&J)6 • Impact from protein of PS adsorption observed for DP - Amgen7
Mechanism: Polysorbate Loss
•Janssen (J&J)6 •Genentech12
Particulate & polysorbate Loss
1J.
Chrom. A 1218 (2011) 2138. Chrom. A 1215 (2008) 156. 3J. Chrom. A 1262 (2012) 122. 4J. Pharm Sci., 91 (2002) 2252. 2J.
5SBE’s
4th International Conference on Accelerating Biopharmaceutical; Quality Analytical. Talks; Kelvin Lee .
6Pharm.
Biotech., (2014) 1. Pharm. Sci. 103 (2014) 1337. 8Anal. Chim 806 (2014) 144. 9Pharm Res. 28 (2011) 1194. 10AAPS NBC, June 2015 How Stable is the Stabilizer? 7J.
Polysorbate Degradation and Impact on biopharmaceuticals Symposium. 12 Pharm. Res. (2015) 32, 2877-2888
Case Study
A Polysorbate Degradation Study: To Assess Protein Stability
mAb 1
mAb 1 Conjugate (ADC)
mAb 2
mAb 3
mAb 4
IgG2
IgG4
IgG1
IgG1
IgG2
Protein (mg/ml)
21
30
50
10
160
pH
5.5
5.5
5.8
5.8
5.3
Protein
9
Study Design
mAb
mAb & 0.1 mg/ml PS-80
mAb & 0.1 mg/ml PS-20
mAb & 0.01 mg/ml hydrogen peroxide
1. Prepare first mAb. 2. Repeat the set of additions with : mAb 1 conjugate and mAbs 2-4 3. Storage: 2-8 ̊C for ten months 4. Time 0, month 2, month 4 and month 10: SEC UPLC UV and appearance
mAb & 0.1 mg/ml PS80 & 0.01 mg/ml deg. products
hydrogen peroxide, 2dodecanone, ethanol, formic acid, lauric acid, oleic acid and acetic acid 5/18/2016
HMMS after 10 months
Filter Settings - compound: (mAb 1 conjugate, mAb 1: IgG2, mAb 2: IgG1, mAb 3: IgG1, mAb 4: IgG2)
LMMS after 10 months
Filter Settings - compound: (mAb 1 conjugate, mAb 1: IgG2, mAb 2: IgG1, mAb 3: IgG1, mAb 4: IgG2)
Full List of Project Particulate and Cloudiness Observations at Month 10 for the Degradant Study Sample
mAb 1
mAb 2
mAb 3
mAb 4
ADC 1
Particulate
Cloudy
Particulate
Cloudy
Particulate
Cloudy
Particulate
Cloudy
Particulate
Cloudy
No Addition
0
0
0
0
0
0
0
0
1
0
PS-20 PS-80 Peroxide
0 0 0
0 0 0
0 1 0
0 0 0
0 0 0
0 0 0
0 0 0
0 0 0
0 0 0
0 0 0
Dodec. Ethanol
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
1 1
0 0
Formic Acid Lauric Acid
0
0
0
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
1
0
Oleic acid
0
0
0
0
1
0
0
0
1
0
Acetic Acid
0
0
0
0
0
0
0
0
1
0
PS-80 and Deg.
0
0
0
0
0
0
0
0
0
0
Deg. Mix.
0
0
0
0
0
0
0
0
NA
0
Pfizer Confidential │ 13
Regulatory Scenarios for Polysorbate Based on End of Shelf-life Data for Early DP Lot(s) Scenario 1 No change is observed in polysorbate concentration over the shelf life of the DP
• Use the data to support in-process monitor/control of polysorbate without a release test for the commercial process
Scenario 2 • Use the data to support in-process If a change in polysorbate monitor/control of polysorbate and no concentration is seen over time but need for a test for the commercial process the critical quality attributes of the (see next slide) protein are not affected
• Polysorbate needs to be monitored at Scenario 3 release and on stability. If a change is seen over time and one or more critical quality attributes are affected due to this change
14
Scenario 2
• Use the data to support in-process monitor/control of polysorbate and no need for a test for the commercial process
If a change is seen over time and the critical quality attributes are not affected
• Use formulation development studies: • To confirm the polysorbate level added to the formulation is appropriate. • The polysorbate added to the formulation is well above the level required to provide protein protection from loss and aggregation. • Data from the excipient robustness studies should also be used to support that there is no impact on stability with polysorbate levels that are at lower concentrations than in the formulation at release.
15
Conclusion • Surveying the literature and using in-house experience a white paper will be prepared covering the knowledge collected to date. • Scientific rationale will be used to propose the systematic assessment, as detailed in this talk, of how PS-80 should be positioned in a regulatory filing.
Acknowledgements
Marguerite Arechederra – Waters (formally Pfizer); Ran the experiments and collated the data for the case studies presented. Sandeep Yadav – Genentech; Vikram Sadineni and Rajesh Gandhi and Mark Bolgar - BMS; Arup K Roy and Vincent John Corvari – Eli Lilly and Kishore Ravuri – Roche, Kevin Maloney – BiogenIdec; Ken Hinds and Wei Liu – Jansen; and Matt Fesinmeyer - Amgen for engaging discussions on the topic of how to position PS-80 in a regulatory filing.
Tween: Fatty Acid Esters of Polyoxyethylene Sorbitan The commercially available PS-20 and PS-80 are a structurally variable mixture of different fatty acid esters and head groups
w+x+y+z refers to the total number of oxyethylene subunits on each surfactant molecule. 2
Representative total ion current profiles and reconstructed ion chromatogram (RIC) for PS-20 >100 Features with m/z between 500-1000 in PS-80 Lauric acid (MW: 227) is the primary fatty acid in PS-20 Oleic Acid is the primary fatty acid in PS-80
Borisov, Oleg V., Junyan A. Ji, Y. John Wang, Felix Vega, and Victor T. Ling. Analytical chemistry 83, no. 10 (2011): 3934-3942.
Tween is a complex mixture of species, and its level of purity is important.
Why put Polysorbates into a Formulation? • Polysorbates are commonly used in biotherapeutic formulations to prevent any protein adsorption and subsequent loss at the product contacting surfaces such as filters, tubing, components of primary containers, glass vials, rubber stoppers, siliconized prefilled syringes, IV bags, etc. • Polysorbates are also commonly used to protect against interfacial stresses encountered during manufacturing processes such as filtration, pumping and filling, shaking and agitation.
• They may also prevent aggregation and proteinsilicone particle formation in prefilled syringes.
Polysorbate Degradation • Polysorbates are known to degrade by hydrolysis and auto-oxidation. M. Donbrow, E. Azaz, A. Pillersdorf, Autoxidation of polysorbates, J. Pharm. Sci. 67 (1978) 1676-1681.
• The rate of hydrolysis of Polysorbates in a formulation that was stored at 5oC and kept at a pH of 5.5 over the shelf-life of the DP was found to be very slow. R.S.K.Kishore, A.Pappenberger, I.B.Dauphin, A Ross, B. Buergi, A. Staempfli, H.C.Mahler, Degradation of polysorbate 20 and 80: Studies on thermal auto-oxidation in bulk and hydrolysis in formulations, J. Pharm. Sci. 100 (2011) 721-731.
• Proper storage and use of PS-80 can minimize oxidation
General Practices for PS-80 • To minimize auto-oxidation during storage, PS-80 should be stored with an overlay of an inert gas such as nitrogen each time a bottle is opened. • A good practice for manufacturing of clinical formulations is to purchase PS-80 in quantities that will be used once and any excess discarded.
•SECRP ELSD- no sample prep. - Genentech1-2 & Pfizer3 •Anion Exchange CAD and/or ELSD – BI & Pfizer •Degradation productsELSD - Janssen (J&J)6 •SPE: PS recovery, high prot. - Amgen8 & Janssen (J&J)6 •Technique ComparisonPfizer10
Detection Techniques
• PS degradation does not have a potential influence on the stability of DP - Roche9 • Hydrogen peroxide does not result in DP oxidation - Bayer4 • PS degradation product shows no impact on CQAs - Pfizer10
Impact of Polysorbate Degradation
Polysorbate (PS) in the Biotherapeutic Industry
• Lipoprotein lipase, and 120 other “problem HCPs” - Univ. of Delaware5 • Lipid-specific mechanism proposal - Janssen (J&J)6 • Impact from protein of PS adsorption observed for DP - Amgen7
Mechanism: Polysorbate Loss
•Janssen (J&J)6 •Genentech12
Particulate & polysorbate Loss
1J.
Chrom. A 1218 (2011) 2138. Chrom. A 1215 (2008) 156. 3J. Chrom. A 1262 (2012) 122. 4J. Pharm Sci., 91 (2002) 2252. 2J.
5SBE’s
4th International Conference on Accelerating Biopharmaceutical; Quality Analytical. Talks; Kelvin Lee .
6Pharm.
Biotech., (2014) 1. Pharm. Sci. 103 (2014) 1337. 8Anal. Chim 806 (2014) 144. 9Pharm Res. 28 (2011) 1194. 10AAPS NBC, June 2015 How Stable is the Stabilizer? 7J.
Polysorbate Degradation and Impact on biopharmaceuticals Symposium. 12 Pharm. Res. (2015) 32, 2877-2888
Case Study
A Polysorbate Degradation Study: To Assess Protein Stability
mAb 1
mAb 1 Conjugate (ADC)
mAb 2
mAb 3
mAb 4
IgG2
IgG4
IgG1
IgG1
IgG2
Protein (mg/ml)
21
30
50
10
160
pH
5.5
5.5
5.8
5.8
5.3
Protein
9
Study Design
mAb
mAb & 0.1 mg/ml PS-80
mAb & 0.1 mg/ml PS-20
mAb & 0.01 mg/ml hydrogen peroxide
1. Prepare first mAb. 2. Repeat the set of additions with : mAb 1 conjugate and mAbs 2-4 3. Storage: 2-8 ̊C for ten months 4. Time 0, month 2, month 4 and month 10: SEC UPLC UV and appearance
mAb & 0.1 mg/ml PS80 & 0.01 mg/ml deg. products
hydrogen peroxide, 2dodecanone, ethanol, formic acid, lauric acid, oleic acid and acetic acid 5/18/2016
HMMS after 10 months
Filter Settings - compound: (mAb 1 conjugate, mAb 1: IgG2, mAb 2: IgG1, mAb 3: IgG1, mAb 4: IgG2)
LMMS after 10 months
Filter Settings - compound: (mAb 1 conjugate, mAb 1: IgG2, mAb 2: IgG1, mAb 3: IgG1, mAb 4: IgG2)
Full List of Project Particulate and Cloudiness Observations at Month 10 for the Degradant Study Sample
mAb 1
mAb 2
mAb 3
mAb 4
ADC 1
Particulate
Cloudy
Particulate
Cloudy
Particulate
Cloudy
Particulate
Cloudy
Particulate
Cloudy
No Addition
0
0
0
0
0
0
0
0
1
0
PS-20 PS-80 Peroxide
0 0 0
0 0 0
0 1 0
0 0 0
0 0 0
0 0 0
0 0 0
0 0 0
0 0 0
0 0 0
Dodec. Ethanol
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
1 1
0 0
Formic Acid Lauric Acid
0
0
0
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
1
0
Oleic acid
0
0
0
0
1
0
0
0
1
0
Acetic Acid
0
0
0
0
0
0
0
0
1
0
PS-80 and Deg.
0
0
0
0
0
0
0
0
0
0
Deg. Mix.
0
0
0
0
0
0
0
0
NA
0
Pfizer Confidential │ 13
Regulatory Scenarios for Polysorbate Based on End of Shelf-life Data for Early DP Lot(s) Scenario 1 No change is observed in polysorbate concentration over the shelf life of the DP
• Use the data to support in-process monitor/control of polysorbate without a release test for the commercial process
Scenario 2 • Use the data to support in-process If a change in polysorbate monitor/control of polysorbate and no concentration is seen over time but need for a test for the commercial process the critical quality attributes of the (see next slide) protein are not affected
• Polysorbate needs to be monitored at Scenario 3 release and on stability. If a change is seen over time and one or more critical quality attributes are affected due to this change
14
Scenario 2
• Use the data to support in-process monitor/control of polysorbate and no need for a test for the commercial process
If a change is seen over time and the critical quality attributes are not affected
• Use formulation development studies: • To confirm the polysorbate level added to the formulation is appropriate. • The polysorbate added to the formulation is well above the level required to provide protein protection from loss and aggregation. • Data from the excipient robustness studies should also be used to support that there is no impact on stability with polysorbate levels that are at lower concentrations than in the formulation at release.
15
Conclusion • Surveying the literature and using in-house experience a white paper will be prepared covering the knowledge collected to date. • Scientific rationale will be used to propose the systematic assessment, as detailed in this talk, of how PS-80 should be positioned in a regulatory filing.
Acknowledgements
Marguerite Arechederra – Waters (formally Pfizer); Ran the experiments and collated the data for the case studies presented. Sandeep Yadav – Genentech; Vikram Sadineni and Rajesh Gandhi and Mark Bolgar - BMS; Arup K Roy and Vincent John Corvari – Eli Lilly and Kishore Ravuri – Roche, Kevin Maloney – BiogenIdec; Ken Hinds and Wei Liu – Jansen; and Matt Fesinmeyer - Amgen for engaging discussions on the topic of how to position PS-80 in a regulatory filing.
