1. Andres H. Gutirrez, Leonard Moise, Frances Terry, Kristen Dasilva, Chris Bailey-Kellogg, William Martin, Anne S. De Groot Assessing Unwanted Host Cell Protein Immunogenicity in Therapeutic Protein Formulations 2013 PDA Europe, 6th Monoclonal Antibodies Workshop September, 2013

2. How did we get to HCP/CHO/CHOPPI? 2002 Invitation to Predicting Biologic Protein Immunogenicity Conference at FDA 2011 CHO Genome Published 2006-2007 Immunogenicity scale Tregitopes, Collaboration With Gene Koren and others CHO genome immunogenicity analysis Plenary at ECI CCE conference HCP / CHO Cells Host Cell Proteins Parallels with Graves model 2004 Benchmarking Vaccine tools for Biologics: Clustered T cell epitopes EpiBars CHOPPI On line . . . 3. Why are we interested in the Impact of species- specific sequences on immunogenicity? Autoimmune Graves Disease Graves Disease Example 4. Autoimmune Graves Disease begins with a response to a single epitope that is mismatched and presented in the context of murine MHC hTSHR variant 1_NM_000369 and murine TSH-R mTSHR variant 1_NM_011648 alignment mTSHR_variant_1_NM_011648 PPSTQTLKLIETHLKTIPSLAFSSLPNISRIYLSIDATLQRLEPHSFYNL hTSHR_variant_1_NM_000369 PPSTQTLKLIETHLRTIPSHAFSNLPNISRIYVSIDVTLQQLESHSFYNL peptide 5-6 (78-94) (variant) Graves Disease Example 5. Epitope fully conserved in human and murine FVIII: Tolerated in FVIII-expressing HLA DR mice (have autologous FVIII) Immunogenic in FVIII KO mice (do not have any FVIII) Epitopes containing human/murine FVIII sequence mismatches: immunogenic in FVIII-expressing HLA DR mice (foreign) immunogenic in FVIII KO mice (still foreign) FVIII KO Not KO FVIII Example (murine) 6. Murine response to TSH-R Mouse Sequence same as Hu Mouse Sequence Different T cell Epitope Present Tolerance Immunogenicity T Cell Epitope Absent No Response Absent epitope, no response Human response to HCP Human Sequence Same as CHO Human Sequence Different T cell Epitope Present Tolerance Immunogenicity T cell Epitope Absent No Response Absent epitope, no response Mice immunized with human TSH-R Humans exposed to CHO or other HCP Important Parallels HCP effects 7. Genomics Transcriptomics A new technology for HCP evaluation 8. Pathogen Immune Response? Self/ Microbiome 8 Active area of research - EpiVax/URI 9. HCP Contamination cancels trial Immune response to HCP (CHO) led to recent cancellation of phase III clinical trials: Higher than expected rate of Anti- CHO antibody development (what is expected????). IB1001 hemophilia (Inspiration Biopharmaceuticals) 10. Danger signals of all sorts Aggregates how do they work? (probably dont work if no T cell epitopes) Immune complexes Complement T cell epitope content (absence of) Treg epitope content Pre-existing T cell response (Tolerance or heterologous immunity) What drives immunogenicity? 11. Factors (roof Immunogenicity) Immune effect Glycosylation () Increase presentation? Increase foreign- ness of protein, need T cell epitopes PEGylation () Slow antigen processing, mask T cell epitopes and B cell epitopes Host Cell-derived Protein () CPG DNA (if bacterial); CHO T cell epitopes Oxidized Form of the Product () Increase foreign-ness, modify presentation Excipients () Increase Danger signal, T cell epitopes Leachates () Increase Danger signal, T cell epitopes Characteristics of Patients (or) Missing Protein is foreign, T cell epitopes Frequency, Duration and Route of Administration (or) Administration like a vaccine, DAMPs, T cell epitopes Aggregates () Aggregation increases T cell epitope presentation In almost every case T cell epitope drives Immune response 12. Antigen Epitope Drug or Vaccine How it works 13. In the right context self proteins can be immunogenic. Take Epo, for example. T cell epitope content is unequally distributed throughout the human (and CHO) proteome.* Immune response depends on protein prevalence, function & previous exposure.** Marc H.V. van Regenmortel, Ph.D., Katia Boven, M.D., Fred Bader, Ph.D. Immunogenicity of Biopharmaceuticals: An Example from Erythropoietin: Protein structure, contaminants, formulation, container, and closure all can affect the immunogenicity of the product. BioPharm International 2005. http://www.biopharminternational.com/biopharm/article/articleDetail.jsp?id=174494&sk=&date=&pageID=5 *A.S. De Groot, J. Rayner, W. Martin. Modeling the immunogenicity of therapeutic proteins using T cell epitope mapping. In: Immunogenicity of Therapeutic Biological Products. Developments in Biologicals. Fred Brown, Anthony Mire Suis, editors. Basel, Karger, 2003. Vol 112:71-80. **Clute, S. C., L. B. Watkin, M. Cornberg, Y. N. Naumov, J. L. Sullivan, K. Luzuriaga, R. M. Welsh, and L. K. Selin. 2005. Cross-reactive influenza virus-specific CD8+ T cells contribute to lymphoproliferation in Epstein-Barr virus-associated infectious mononucleosis. The Journal of clinical investigation 115:3602-3612. CHO are mammalian proteins How can self proteins be immunogenic? 14. T Cell Epitope Content - Predicted Potential for Immunogenicity of Selected Proteins -80 -60 -40 -20 0 20 40 60 80 100 HumanFSH beta HumanIgACD HumanIgGCD Human Albumin Human Amylase De-immunized INF-beta Human Transferrin *Human Gonadotropin Random Expectation Influenza Hemagglutinin *Human GHRH *Human Gonadotropin w/signal TetanusToxin Human Erythropoietin BrazilNut Antigen *Human GHRHw/signal **HumanINF- beta Less Immunogenic Proteins (based on clinical experience) Have Fewer T cell Epitopes De Groot, As, Goldberg M, Moise L, Martin W. Evolutionary deimmunization: An ancillary mechanism for self-tolerance. Cell Immunol. 2007 Apr 17; Pages 148-153. http://dx.doi.org/10.1016/j.cellimm.2007.02.006 Are self proteins immunogenic? 15. EpiVax Immunogenicity Hypothesis: Immune Response = Sum of Epitopes T cell response depends on: T cell epitope content + HLA of subject Protein Immunogenicity can be Ranked epitope Protein Therapeutic 1 + 1 + 1 = Response epitopeepitope De Groot A.S. and L. Moise. Prediction of immunogenicity for therapeutic proteins: State of the art. Current Opinions in Drug Development and Discovery. May 2007. 10(3):332-40. In biologics, immunogenicity is related to T cell epitope content 16. EpiVax - Immunogenicity Scale Low Neutral High Albumin Tetanus ToxinProtein X or mAb Y Proteins ranked by T- Epitope content per Amino Acid De Groot A.S., Drug Discovery Today - 2006; De Groot A.S., Mire-Sluis, A. Ed.. Dev. Biol. Basel, Karger, 2005. vol 122. pp 137-160. Antigen A Antigen B Aggregate immunogenicity drives Immune response 17. EpiMatrix predicted excess/shortfall in aggregate immunogenicity relative to a random peptide standard. - 80 - - 70 - - 60 - - 50 - - 40 - - 30 - - 20 - - 10 - - 00 - - - 10 - - - 20 - - - 30 - - - 40 - - - 50 - - - 60 - - - 70 - - - 80 - Thrombopoietin Human EPO Tetanus Toxin Influenza-HA Albumin IgG FC Region EBV -BKRF3 Follitropin -Beta A protein score > 20 indicates a significant immunogenic potential. Proteins that have previously been demonstrated to be immunogenic have higher potential immunogenicity on the scale. Those that have rarely been demonstrated to be immunogenicity have lower T cell epitope content. Immunogenicity scale 18. Some Vaccine Antigens High Scores (work done for NMRC, Dept. of Defense) 19. - 80 - - 70 - - 60 - - 50 - - 40 - - 30 - - 20 - - 10 - - 00 - - -10 - - -20 - - -30 - - -40 - - -50 - Human EPO Immunogenic Antibodies* Tetanus Toxin Influenza-HA Albumin IgG FC Region EBV-BKRF3 Fibrinogen-Alpha Non-immunogenic Antibodies Follitropin-Beta Hirudin(-90.41) See my Blog Thinking out Loud for a discussion of Leech proteins and Tick Saliva proteins-Tick saliva proteins also have low immunogenicity potential. Other Antigens Extremely Low Scores (Hirudin, Tick Saliva, Some Parasites) 20. Handled on a case-by-case basis Consider Source Maximum dose (mg biologics/kg body weight) Route of administration Frequency of dosing Pre-clinical and clinical data Detection process in evolution Regulators prefer Leech-like Proteins And HCPs - Regulatory Perspective 21. Detection Protein staining Immunoblotting Identification 2D-PAGE/MS 2D-LC/MS Quantitation ELISA using anti-HCP antibodies May need to develop internal processes Some kits are available Risk assessment Cytokine release assays New Approach Immunogenicity Screening in silico Analytical Tests for HCP 22. MHC binding is a prerequisite for immunogenicity Epitopes are linear and directly derived from antigen sequence Binding is determined by amino acid side chains Matrix-based predictor MHCII Mature APC Immunogenicity prediction 23. EpiMatrix EpiVax uses EpiMatrix to predict epitopes matrix based prediction algorithm Can predict either class I or class II MHC binding MHC binding is a prerequisite for immunogenicity MHC II Pocket Peptide Epitope Mature APC MHCII T cell epitopes are linear and directly derived from antigen sequence Binding is determined by amino acid side chains (R groups) and encoded in single letter code 23 8/26/2013 Confidential 24. Clusters of MHC binding drive T cells DRB1*0101 DRB1*0301 DRB1*0401 DRB1*0701 DRB1*0801 DRB1*1101 DRB1*1301 DRB1*1501 T cell epitopes are not randomly distributed but instead tend to cluster in specific regions. These clusters can be very powerful, enabling significant immune responses to low scoring proteins. ClustiMer recognizes T-cell epitope clusters as polypeptides predicted to bind to an unusually large number of HLA alleles. 8/26/2013 Confidential 25. What Makes Proteins Really immunogenic? Sequences that Contain EpiBars Confidential Roberts CGP, Meister GE, Jesdale BM, Lieberman J, Berzofsky JA, A.S. De Groot, Prediction of HIV peptide epitopes by a novel algorithm, AIDS Research and Human Retroviruses, 1996, Vol. 12, No. 7, pp. 593-610. ClustiMer - Locates highly immunogenic regions EpiBar : A common feature of highly immunogenic clusters EpiBar 26. EpiVax Immunogenicity Scale Confidential - 80 - - 70 - - 60 - - 50 - - 40 - - 30 - - 20 - - 10 - - 00 - - -10 - - -20 - - -30 - - -40 - - -50 - - -60 - - -70 - - -80 - Thrombopoietin Human EPO Immunogenic Antibodies* Tetanus Toxin Influenza-HA Albumin IgG FC Region EBV-BKRF3 Fibrinogen-Alpha Non-immunogenic Antibodies Follitropin-Beta PROTEIN_001 (35.13) Protein Immunogenicity Scale Proteins Scoring above +20 are considered to be potentially immunogenic. On the left of the scale we include some well-known proteins for comparison - 80 - - 70 - - 60 - - 50 - - 40 - - 30 - - 20 - - 10 - - 00 - - -10 - - -20 - - -30 - - -40 - - -50 - - -60 - - -70 - - -80 - Thrombopoietin Human EPO Immunogenic Antibodies* Tetanus Toxin Influenza-HA Albumin IgG FC Region EBV-BKRF3 Non-immunogenic Antibodies Follitropin-Beta 27. EpiMatrix mAb Immunogenicity Scale - 80 - - 70 - - 60 - - 50 - - 40 - - 30 - - 20 - - 10 - - 00 - - -10 - - -20 - - -30 - - -40 - - -50 - - -60 - - -70 - - -80 - IgG FC Region Nuvion (0%) Avastin (0%) AB01 (EPX Adjusted Score: -46.98) AB02 (EPX Adjusted Score: -44.48) AB03 (EPX Adjusted Score: -44.81) AB04 (EPX Adjusted Score: -45.81) AB05 (EPX Adjusted Score: -45.88) AB06 (EPX Adjusted Score: -47.85) AB07 (EPX Adjusted Score: -46.99) AB08 (EPX Adjusted Score: -46.30) AB09 (EPX Adjusted Score: -47.40) AB10 (EPX Adjusted Score: -45.88) AB11 (EPX Adjusted Score: -47.40) Synagis (1%) Simulect (1.4%) Humira (12%) Bivatuzumab (6.7%) Remicade (26%) Rituxan (27%) Campath (45%) Humicade (7%) Reopro (5.8%) Tysabri (7%) LeukArrest (0%) Herceptin (0.1%) Compare with: 27 8/26/2013 Confidential Due to the presence of Tregitopes, antibodies tend to fall lower on the immunogenicity scale. We have developed a refined method using regression analysis to predict the immunogenicity of antibody sequences based on observed clinical responses (next slide). We have found that a balance in favor of Tregitope (regulatory) content over neo-epitope (effector) content is correlated with reduced clinical immunogenicity. NeoEpitopeContent Tregitope Content High Low Low Avastin (0%) Herceptin (0%) Mylotarg (3%) Simulect (1%) Synagis (1%) High Campath (45%) Remicade (26%) Rituxan (27%) 28. CHO genome Immune Response? Self/ Microbiome 28 Logical Next Step measure CHO/Self Conservation 29. Databases available Putatively Secreted (signal peptide) Mouse secreted 165 proteins Transcriptome 32,801 contigs Validated HCP contaminants 25 proteins CHO genome 24,383 predicted genes 30. Key Datasets Genome and transcriptome 31. Protein databases (UniProtKB/Swiss- Prot, Locate) BLAST SignalP EpiMatrix BlastiMer - JanusMatrix Tools used for this analysis 32. Identify secreted CHO proteins Collect published HCP from CHO Evaluate potential immunogenicity Evaluate sequence homology Identify clustered regions compare to CHO; Are human/CHO different at the cluster? Count as possible immunogenicity trigger. Approach 33. Immunogenicity Scores distribution 34. Immunogenicity Scale Validated HCP CHO contaminants 35. SL cytokine (84) Lysosomal protective protein (35) 36. But are human-like proteins immunogenic? CHO okay? peptides 37. Putatively Secreted (signal peptide) Mouse secreted 165 proteins Transcriptome 32,801 contigs Validated HCP contaminants 25 proteins CHO genome 24,383 predicted genes Human proteome 20,238 proteins Approach to conservation with Human 38. Identify secreted CHO proteins Evaluate potential immunogenicity Evaluate sequence homology Identify clustered regions compare to CHO; Are human/CHO different at the cluster? Count as possible immunogenicity trigger. Approach 39. T cell Receptor Face (epitope) MHC-binding Face (agretope) T cell epitopes are two-faced 40. Identifies cross-reactive peptides: Identical T cell-facing residues Same HLA allele but . . OK if different MHC-facing residues The God of Two Faces: JanusMatrix 41. TCR face vs. MHC binding face MHC/HLA TCR The most conservative approach: Identical T cell-facing residues Same HLA allele and minimally different MHC-facing residues 42. EpiMatrix adjusted immunogenicity score 43. Determination of conservation with self: JanusMatrix results 44. Cross-reactivity visualization Predicted 9-mer epitope from a source protein Human protein where cross-reactive epitopes are present 9-mer from human prevalent proteome, 100% TCR face identical to source epitope Source protein HCV_G1_NS2_794 45. CEFT Peptides (immunogenic) 46. Flu and Tet tox epitopes SNF2 histone linker PHD RING helicase ETAA16 protein Ankyrin repeat domain 18A Flu HA308-318 Ubiquitin specific protease 1 Poly ADP ribose polymerase family, member 9 Poly ADP ribose polymerase family, member 9 Tetanus Toxin830-844 Olfactory receptor, family 5, subfamily D, member 14 47. hTregitope-IGGC-167 hTregitope-IGGC-289 HTREG_IGGC-289 HTREG_IGGC-167 48. CHO: lysosomal protective protein Lysosomal protective Lysosomal protective 49. SL cytokine CHO: SL cytokine SL cytokine 50. Identify secreted CHO proteins Evaluate potential immunogenicity Evaluate sequence homology Identify clustered regions compare to CHO; Are human/CHO different at the cluster? Count as possible immunogenicity trigger. New Approach for CHO 51. Immune Response = Sum of Epitopes Sum includes + (T effectors) and (Tregs) scores Protein Therapeutic Host Cell Protein Contaminant HCP Epitope New Approach 52. For an individual, T cell response depends on: T cell epitope content x HLA Treg Epitope content x HLA Vaccine or Foreign Protein = TeffPT1+ TeffPT2 . . . ) = Response CHO = TeffPT + TeffPT + TeffHCP TregPT) = Treg Adjusted Response Immune response depends on Foreign-ness Potential Tregs Adjuvant (Danger signal) Proposed adjustment to score 53. Available now: CHOPPI CHO Protein Predicted Immunogenicity CHOPPI http://bit.ly/11fZqfJ 54. Formulation (VLP; aggregates) Danger Signal Route: Subcutaneous delivery? Dose (high/low, persistent, intermittent) T cell epitope content Differing T cell epitope content = HCP 55 In Closing Factors affecting Immunogenicity 55. While CHO are the most commonly used cell lines for mammalian cell protein expression, Company-specific cell lines may vary. Furthermore, we cant anticipate Genetic engineering Batch-to-batch variation Expression (based on above) Which protein will hitchhike CHO Cell lines may differ 56. Genomics ExpressomeInformatics In the future Obtain proteins through MS/MS HPLC and Sequence, ID epitopes 57. Thank you! And . . . CHOPPI: http://bit.ly/11fZqfJ or contact me. Translational Immunology Research and Accelerated [Vaccine] Development Institute for Immunology and Informatics University of Rhode Island Dartmouth College EpiVax, Inc. SL cytokine 58. Institute for Immunology and Informatics (iCubed) D. Spero icubed overview 2011 www.immunome.org URI Alumni Board 2012 59. New Concept: Tregitopes induce tolerance to protein Therapeutics (Friday April 20th Session) Epitope may induce different types of Response 60. CHO Adjustment for Immunogenicity? + + Conserved epitope Neo-Epitope Neo-Epitope 61. Immune Response = Sum of Epitopes Sum includes + (T effectors) and (Tregs) scores ISPRI approach to analyzing mAbs ... T cell response depends on: T cell epitope content x HLA Treg Epitope content x HLA Protein Immunogenicity can be Ranked Treg epitope Protein Therapeutic 1 + 1 - Treg = Response epitopeepitope 62. T reg Stimulus IL 10, TNF alpha Additional Treg Epitope Modify Effector T cell response: Reduce T effector Stimulus Current Hypothesis: More Tregitopes Lower Immunogenicity De Groot A.S. and D. Scott. Immunogenicity of Protein Therapeutics. Trends in Immunology. Invited Review. Trends Immunol. 2007 Nov;28(11):482-90. 6363Confidential and Copyrighted EpiVax 63. 64 EpiVax: Immunogenicity scale is 64. Correlation of antibody immunogenicity with Tregitope adjusted EPX Scores 65. Correlation of EpiMatrix Scores and Immunogenicity in Human studies 40% 37% 21.97 FPX 1 0% 9.3% -111.25 FPX 5 NA0.5%12% Neutralizing Antibodies 5.6%7.8%53%Binding Antibodies -1.761.6234.37EpiMatrix score FPX 4FPX 3FPX 2Protein Na: not analyzed Negative score indicates presence of Treg epitope 66. - 80 - - 70 - - 60 - - 50 - - 40 - - 30 - - 20 - - 10 - - 00 - - -10 - - -20 - - -30 - - -40 - - -50 - - -60 - - -70 - - -80 - Thrombopoietin Human EPO Immunogenic Antibodies* Tetanus Toxin Influenza-HA Albumin IgG FC Region EBV-BKRF3 Fibrinogen-Alpha Non-immunogenic Antibodies Follitropin-Beta Ab K (-38.23) Ab E (-16.03) Ab N (-53.88) Ab P (-70.14) Ab B (-00.32) Ab A (13.82) Ab D (-08.87) Ab F (-22.13) Ab I (-25.77) Ab O (-54.26) Ab L (-48.49) Ab C (-02.03) Ab M (-52.25) Ab H (-24.99) Ab J (-28.94) Ab G (-24.33) *Tregitope adjusted Application - Germline Abs*