Establishment of a Comparability Strategy to Support a Cell Line Change During Clinical Development of a Monoclonal Antibody Bryan J. Harmon

Disclaimer The views and opinions expressed in the following PowerPoint slides are those of the individual presenter and should not be attributed to Drug Information Association, Inc. (“DIA”), its directors, officers, employees, volunteers, members, chapters, councils, Special Interest Area Communities or affiliates, or any organization with which the presenter is employed or affiliated.   These PowerPoint slides are the intellectual property of the individual presenter and are protected under the copyright laws of the United States of America and other countries. Used by permission. All rights reserved. Drug Information Association, DIA and DIA logo are registered trademarks or trademarks of Drug Information Association Inc. All other trademarks are the property of their respective owners.

Outline Drivers for Cell Line Changes Elements of Comparability Strategy Case Studies Conclusions

Cell Line Changes During Clinical Development Driver Examples Quality risk with initial cell line Genetic splicing or mutation identified ASM exposure during cell line generation Lack of assurance of clonality Initial cell line is not commercially viable Insufficient titer Insufficient cell line stability Not consistent with manufacturing platform Intellectual property issues

Types of Cell Line Changes Additional round of cloning Different clone from same host cell line Different host cell line Cell line changes: Are considered the biggest risk among process changes Have been practiced very conservatively in the industry

Elements of Integrated Comparability Strategy Host cell line & clone selection criteria Analytical comparability testing strategy In vitro biological testing Nonclinical PK, PD & immunogenicity assessments Clinical assessments Need for & extent of each element driven by risk assessments

Risk Rating = Severity x Occurrence x Detection Risk Assessments – FMEA Analysis Severity – impact on toxicity, safety, efficacy or PK/PD Occurrence – likelihood of being outside preclinical & clinical experience (process capability, control & robustness) Detection – capability of analytical methods to detect occurrence Risk Rating = Severity x Occurrence x Detection Risk assessments must be cross-functional (toxicology, medical, analytical, process scientists)

Host Cell Line & Clone Selection Criteria In evaluating risk of cell line change, must consider: Post-translational modification capabilities of potential new host cell line Clonal variability of chosen host cell line in product quality attributes Capability to mitigate comparability risks through process development/optimization

Impact of Host Cell Line on Glycosylation

Impact of Host Cell Line on Glycosylation CE-LIF Oligosaccharide Profiling

Risks of Cell Line Changes Different host cell line Different clone from same host cell line Additional round of cloning Increasing risk to CQAs of molecule

Host Cell Line & Clone Selection Criteria In evaluating risk of cell line change, must consider: Post-translational modification capabilities of potential new host cell line Clonal variability of chosen host cell line in product quality attributes Capability to mitigate comparability risks through process development/optimization

Clonal Variability in Glycosylation Fc Glycosylation of CHO-derived IgG1

Clonal Variability in Glycosylation Fab Glycosylation of CHO-derived IgG1 with 2 Glycosylation Sites

Comparability Risk Mitigation During Clone Selection Greater emphasis on product quality parameters that are: Enzymatic processes that are likely to be clone specific: e.g., glycosylation, proteolytic clipping Genetic issues: e.g., mutations, frame shifts, splices Critical to the biological activity of the mAb: e.g., ADCC → Fucosylation CDC → Galactosylation Lesser emphasis on product quality parameters that are: Optimized through purification process development; e.g., host cell protein, aggregation Caveat: aggregation could be an indicator of other issues (e.g., splicing, disulfide reduction) Chemical mechanisms that are less likely to be clone specific; e.g., oxidation, deamidation, glycation

Host Cell Line & Clone Selection Criteria In evaluating risk of cell line change, must consider: Post-translational modification capabilities of potential new host cell line Clonal variability of chosen host cell line in product quality attributes Capability to mitigate comparability risks through process development/optimization

Physico-Chemical Comparability Testing Assess impact of cell line change on CQAs of MAb based upon risk assessment of quality attributes Additional testing to satisfy regulatory concerns; e.g., Glycosylation analysis for MAb whose MOA is not dependent upon effector function Co-mixture analysis of representative lots where appropriate (e.g., LC-MS peptide mapping, SEC, CEX, CE-SDS) Assessment of impact on degradation mechanisms (e.g., stressed or accelerated stability study) Pre-defined acceptance criteria: At early stages of development: Insufficient data to establish statistical limits tighter than specifications at early stages of development Qualitative criteria for characterization assays Allowance for investigative testing (e.g., source of differences in charge heterogeneity)

CQA Risk Assessments

Physico-Chemical Comparability Testing Assess impact of cell line change on CQAs of MAb based upon risk assessment of quality attributes Additional testing to satisfy regulatory concerns; e.g., Glycosylation analysis for MAb whose MOA is not dependent upon effector function Co-mixture analysis of representative lots where appropriate (e.g., LC-MS peptide mapping, SEC, CEX, CE-SDS) Assessment of impact on degradation mechanisms (e.g., stressed or accelerated stability study) Pre-defined acceptance criteria: At early stages of development: Insufficient data to establish statistical limits tighter than specifications at early stages of development Qualitative criteria for characterization assays Allowance for investigative testing (e.g., source of differences in charge heterogeneity)

Characterization Tests Typical Physico-Chemical Comparability Tests Release Tests Characterization Tests Potency/Biological Activity Bioassay Surface plasmon resonance Structural Integrity (Primary, Secondary & Tertiary) Intact LC-MS Partial reduction LC-MS LC-MS peptide mapping* Far & near UV circular dichroism Free thiol analysis Calorimetry** Molecular Heterogeneity Cation-exchange chromatography* Oligosaccharide profiling Product-Related Impurities Size-exclusion chromatography* Analytical ultracentrifugation** Non-reduced CE-SDS* Reduced CE-SDS* Process-Related Impurities Host cell protein Triton X-100 DNA Insulin Protein A MSX * Include co-mixture analysis of representative lots ** Added based upon regulatory feedback

Physico-Chemical Comparability Testing Assess impact of cell line change on CQAs of MAb based upon risk assessment of quality attributes Additional testing to satisfy regulatory concerns; e.g., Glycosylation analysis for MAb whose MOA is not dependent upon effector function Co-mixture analysis of representative lots where appropriate (e.g., LC-MS peptide mapping, SEC, CEX, CE-SDS) Assessment of impact on degradation mechanisms (e.g., stressed or accelerated stability study) Pre-defined acceptance criteria: At early stages of development: Insufficient data to establish statistical limits tighter than specifications at early stages of development Qualitative criteria for characterization assays Allowance for investigative testing (e.g., source of differences in charge heterogeneity)

Case Study #1 Property MAb1 Isotype IgG4 Phase of Development Pre-Phase 2 Cell Line Change GS-NS0 to GS-CHO-K1SV MOA Dependent upon Effector Function? No Drivers for Cell Line Change Elimination of non-human glycoforms Alignment with platform

Case Study #1 Prior Knowledge: Risk Assessment: Experience in GS-NS0 to GS-CHO-K1SV cell line changes suggested risk of: Changes in glycosylation profile Changes in charge heterogeneity resulting from differences in proportions of charge variants Risk Assessment: Expected differences presented low risk to the safety and efficacy of molecule Comparability Strategy: Extraordinary efforts would not be made in clone selection and process development to eliminate these differences Demonstrate comparability through: Physico-chemical testing In vitro biological assays Non-clinical in vivo PK, PD and immunogenicity studies

Case Study #1 Prior Knowledge: GS-NS0 to GS-CHO-K1SV Cell Line Change Cation-Exchange Chromatography

Case Study #1 Prior Knowledge: Risk Assessment: Experience in GS-NS0 to GS-CHO-K1SV cell line changes suggested risk of: Changes in glycosylation profile Changes in charge heterogeneity resulting from differences in proportions of charge variants Risk Assessment: Expected differences presented low risk to the safety and efficacy of molecule Comparability Strategy: Extraordinary efforts would not be made in clone selection and process development to eliminate these differences Demonstrate comparability through: Physico-chemical testing In vitro biological assays Non-clinical in vivo PK, PD and immunogenicity studies

GS-CHO-K1SV-Derived MAb1 Case Study #1 Differences in glycosylation profiles were observed: Glycoforms GS-NS0-Derived MAb1 GS-CHO-K1SV-Derived MAb1 Batch 1 Batch 2 Non-human glycoforms a-Gal-containing 2.0% 2.4% Not observed NeuGc-containing 2.8% Human glycoforms b-Gal-containing 40.8% 40.9% 26.9% 27.2%

Case Study #1 CHO-derived MAb1 Differences in charge heterogeneity profiles were observed: Co-mixture NS0-derived MAb1 LC-MS characterization of isolated CEX fractions identified small differences in proportions of typical sources of MAb charge variants: Heavy chain N-terminal pyroglutamate Heavy chain C-terminal lysine Heavy chain C-terminal desGly/amidation Glycation

Case Study #1 - Summary Physico-Chemical Testing: No apparent adverse impact observed in structural integrity, product-related impurities or process-related impurities Minor differences observed in molecular heterogeneity Glycosylation Charge heterogeneity In vitro Biological Assays No apparent differences observed in potency Nonclinical PK, PD & Immunogenicity Assessment No apparent differences observed The cell line change presents low risk to the safety or efficacy of MAb1

Case Study #2 Property MAb2 Isotype IgG1 Phase of Development Pre-Phase 2 Cell Line Change DHFR-CHO-DG44 to GS-CHO-K1SV MOA Dependent upon Effector Function? Yes Drivers for Cell Line Change Alignment with platform

Case Study #2 Prior Knowledge: Risk Assessment: No experience in DHFR-CHO-DG44 to GS-CHO-K1SV cell line changes Knowledge of clonal variability suggested risk of changes in glycosylation profile: Core fucosylation → impact ADCC activity Terminal b-galactose → impact CDC activity Risk Assessment: Changes in glycosylation could present significant risk to the safety and efficacy of molecule Comparability Strategy: Glycosylation as criterion for clone selection to mitigate comparability risk Fucosylation prioritized based upon proposed MOA Demonstrate comparability through: Physico-chemical testing In vitro biological assays (including ADCC & CDC) Non-clinical in vivo PK, PD & immunogenicity studies

Case Study #2 Impact of Fucosylation on ADCC Activity of MAb2

Case Study #2 Prior Knowledge: Risk Assessment: No experience in DHFR-CHO-DG44 to GS-CHO-K1SV cell line changes Knowledge of clonal variability suggested risk of changes in glycosylation profile: Core fucosylation → impact ADCC activity Terminal b-galactose → impact CDC activity Risk Assessment: Changes in glycosylation could present significant risk to the safety and efficacy of molecule Comparability Strategy: Glycosylation as criterion for clone selection to mitigate comparability risk Fucosylation prioritized based upon proposed MOA Demonstrate comparability through: Physico-chemical testing In vitro biological assays (including ADCC & CDC) Non-clinical in vivo PK, PD & immunogenicity studies

Clonal Variability in Fucosylation of GS-CHO-K1SV-Derived MAb2 Case Study #2 Clonal Variability in Fucosylation of GS-CHO-K1SV-Derived MAb2

GS-CHO-K1SV-Derived MAb2 Case Study #2 Similar fucosylation has been observed due to clone selection strategy & subsequent cell culture development: Glycoforms DG44-CHO-Derived MAb2 GS-CHO-K1SV-Derived MAb2 Batch 1 Batch 2 Batch 3 Fucose/oligosaccharide 0.93 0.94 0.95 0.96 b-Galactose/oligosaccharide 0.55 0.51 0.54 0.40 In vitro biological assays indicate comparable ADCC activity.

(manufacture of clinical trial lots is ongoing) Case Study #2 - Summary Physico-Chemical Testing: No apparent adverse impact on structural integrity, product-related impurities or process-related impurities Minor differences observed in molecular heterogeneity Lower b-galactosylation levels In vitro Biological Assays No apparent differences observed in ADCC activity Nonclinical PK, PD & Immunogenicity Assessment No apparent differences observed Thus far, cell line change presents low risk to the safety or efficacy of MAb2 (manufacture of clinical trial lots is ongoing)

Conclusions Characterization during clone selection can mitigate risks associated with a cell line change Integrated comparability strategy for a cell line change should start prior to clone selection MAb’s MOA & clonal variability in CQAs should drive clone selection strategy Cross-functional risk assessments play a critical role throughout; e.g., Defining CQAs for MAb Defining clone selection strategy Defining physico-chemical testing protocol & acceptance criteria Defining need for & extent of nonclinical PK, PD & immunogenicity assessments Assessing potential impact of observed differences When possible, comparability plan/protocol should be shared with FDA prior to execution (e.g., briefing document, IND amendment)