Evotec Se · XETR:EVT
Evotec — Investor Day (March 2022)
111 slides from Evotec's investor day presentation, Mar 2, 2022
Slide 1: AI/ML Precision Technologies driving Probabilities of Success up
AI/ML Precision Technologies driving Probabilities of Success up (PoS up) Evotec SE, Capital Markets Day, 2nd March 2022
Slide 2: Cautionary statement regarding forward-looking statements
Cautionary statement regarding forward-looking statements Information set forth in this presentation contains forward-looking statements, which involve a number of risks and uncertainties. All statements other than statements of historical fact are forward-looking statements, which are often indicat
Slide 3: Agenda
AGENDA 08.00 – 08.15 am Action Plan 2025 update - “…just the beginning” of the data-driven R&D Autobahn to Cures 08.15 – 09.15 am Precision technologies bring PoS up From molecular databases via iPSCs, to AI/ML AI/ML Precision Technologies driving tools at work Probabilities of Success up (PoS u
Slide 4: Let’s talk about Probabilities of Success
Let’s talk about Probabilities of Success 1st Capital Markets Day 2022 Werner Cord Craig Lanthaler Dohrmann Johnstone CEO CSO COO Uwe Christiane Nele Andag Honisch Schwarz Metabolic Diagnostics & Stem Cell & Diseases Stratification Regenerative Biology Paul Linda Steffen Walker Zuckerman Grimm Toxic
Slide 5: Agenda
Agenda Action Plan 2025 update “…just the beginning” of the data-driven R&D Autobahn to Cures Precision technologies bring PoS up From molecular databases via iPSCs, to AI/ML tools at work Processes bring PoS up From targets, via full suite of AI/ML tools, to manufacturing Roundup & Q&A session PAGE
Slide 6: Bringing the industry closer together
Bringing the industry closer together Our contribution to the industry We discover medicines for difficult to treat diseases in efficient collaborations „The goal of We focus on data driven precision medicine and Evolution is not early disease relevance to improve Probabilities of one single human,
Slide 7: AP 2025 is fully in swing and accelerated by precision medicine
AP 2025 is fully in swing and accelerated by precision medicine Action Plan in numbers 2025 2021 2018 2012 Action Plan 2025 2009 Leadership in data, science, Action Plan 2022 multimodality & access Action Plan 2016 Aspire global Action Plan 2012 Build innovation leadership Restructure for growth see
Slide 8: Our innovation hub is high-tech driven and fully integrated
Our Innovation hub is high-tech driven and fully integrated Capabilities & expertise overview Industry needs Capabilities & Expertise (illustrative) R&D efficiency platforms1) Precision medi- cine platforms Just – Evotec Biologics1) Right modality drug design PAGE 8 1) Also partly accessible as stan
Slide 9: Faster and more learning curves illustrate … “just the beginning”
Faster and more learning curves illustrate … “just the beginning” “Evotec inside” (selected KPI’s 2020/21) 62 142 12 24 > 250 Patent applications High-throughput screens Pre-clinical development INDiGO® programs GMP API batches candidates (PDC) >10 >200 bn >440 bn >10,000 >100 Precision platforms Da
Slide 10: Despite P2X3 set back; strategy for royalty pool is fully intact
Despite P2X3 set back; Strategy for royalty pool is fully intact Steady stream of high-value catalysts Molecule Therapeutic Area/Indication Partner Discovery Pre-clinical Phase I Phase II Phase III EVT201 Insomnia (GABA-A) Selected pipeline events within next 12 – 24 months Not Disclosed Infectious
Slide 11: The iceberg of our product opportunities
The iceberg of our product opportunities In total > 200 proprietary projects with big financial upside Global Neuroscience & Pain Oncology Metabolic Diseases Inflammation & Immunology1) Virology Anti-bacterial Health Ph3 Insomnia ND Ph2 DNP – B1 CC NP OAB Breast cancer Carrick Breast Cancer Endo Cli
Slide 12: Building a massive co-owned clinical pipeline
Building a massive co-owned clinical pipeline EVT Innovate pipeline evolution 2015-2025 (e) 20153) 20221,3) 2025 (e)2,3) # of projects # of projects # of projects >170 >130 2 0 3 >100 18 12 >100 49 34 9 2 0 0 4 Platf. & Pre- Phase Phase Phase Appro- Total Platf. & Pre- Phase Phase Phase Appro- Total
Slide 13: Building co-owned product upside with limited financial risk
Building co-owned product upside with limited financial risk Co-ownership business model 2010-2025 (e) in products / in € m >170 >140 >100 >50 >5 First royalties 2010 2012 2014 2016 2018 2020 2022(e) 2024(e) 2025 (e) … Co-owned products R&D investments1) EVT Innovate EBITDA PAGE 13 1) without EVT Eq
Slide 14: Clear strategy in place
Clear strategy in place Growth and investment strategy overview – Action Plan 2025 Targeted revenue development Revenue Composition of (in € m) composition revenue mix expected >1,000 2020 ~20% to change over time while ALL fields EVT Execute ~10% continue to grow CAGR Just – Evotec ~70% Shiftin
Slide 15: Quote: Werner Lanthaler on PoS and IRR
“To improve IRR we have to integrate tech- driven decisions in discovery. Getting PoS up is Evotec’s key contribution to the sharing economy in R&D.“ Werner Lanthaler PAGE 15
Slide 16: Integrating and accelerating better data for more precise medicine
Integrating and accelerating better data for more precise medicine What does ‘data-driven’ mean in practice? Multiple sources to drive innovation AI-enabled Data small R&D integration molecules efficiency EVOiR&D RNA Linking platforms Single analytics genetic, Precision cell tech- nologies Structure
Slide 17: Still a highly fragmented industry
Still a highly fragmented industry Selected AI/ML companies (Industry landscape – Illustrative & highly simplified) Many players at Data integration work to initiate a new “data driven” Patient industry paradigm outcomes Academic data Data Early Discovery Approval and post-approval aggregation PAGE
Slide 18: PoS up is key leverage for better IRR
PoS up is key leverage for better IRR Current challenges in R&D Key challenges Development costs per asset increase Cost per asset doubled since 2010, in US$ m R&D model is Challenging returns due to ~2,500 inefficient “too late and “expensive failure” 1,188 +100% “One drug fits all” 90% of drugs ar
Slide 19: Significant improvements of PoS are possible
Significant improvements of PoS are possible How we can improve PoS in early R&D (Examples/simplified) Todays use of AI/ML Integrated drug design Data integration in % Target identification ~10% Efficacy & safety predictions System biology to better understand diseases e.g. associating existing targ
Slide 20: Agenda
Agenda Action Plan 2025 update “…just the beginning” of the data-driven R&D Autobahn to Cures Precision technologies bring PoS up From molecular databases via iPSCs, to AI/ML tools at work Processes bring PoS up From targets, via full suite of AI/ML tools, to manufacturing Roundup & Q&A session PAGE
Slide 21: Quote: Cord Dohrmann on AI/ML data analysis
“AI/ML driven data analysis based on high quality data is the key to improving probabilities of success.“ Cord Dohrmann PAGE 21
Slide 22: Increasing the probability of success is the key challenge
Increasing the probability of success is the key challenge Attrition rates have not improved Selected KPIs False discovery rate in the pre- Clinical attrition up to market Post-market safety events of FDA- clinic1) launch2) approved drugs3) ~95 % > 92 % >32 % Development cost per NME launch4) >$ 3 b
Slide 23: Probability of success on the decline in most indications
Probability of success on the decline in most indications PoS from Phase II to Phase III transition is below 30% Probability of success Overall clinical success rates in % success rate = number that advanced to next phase/total number advanced & suspended 26.1 2016 2021 23.9 90.6% 19.1 17.1 15.3 15.
Slide 24: Unique combination of precision platforms
Unique combination of precision platforms Precision medicine is driving probability of success Molecular patient Patient (iPSC) – Patient stratification databases derived disease models and biomarkers Re-defining health and Modelling disease profiles Precision diagnostics disease via molecular u
Slide 25: Biological insights enable precision medicine
Biological insights enable precision medicine Transcriptomics and proteomics provide more biological insight than genomics Robustness Scalability Cost efficiency Biological insight Genomics Transcriptomics Proteomics Metabolomics Reproducibility Throughput Cost efficiency Molecular insights in Day
Slide 26: EVOpanOmics – High performance transcriptome and proteome analysis
EVOpanOmics – High performance transcriptome and proteome analysis Efficient data generation combined with superior data analysis increases PoS Predictive drug discovery – Evolving paradigm Genomics Standard process Seamless and cost effective platforms and processes Patient from pre-clinic to c
Slide 27: Connecting complex clinical and pre-clinical data increases PoS
Connecting complex clinical and pre-clinical data increases PoS EVOpanHunter – easy access to complex data analysis Omics data Meta data Transcriptomics Gene/protein information Genomics Experimental parameters Proteomics Metabolomics Chemical data Structures Compound information Cli
Slide 28: Demand for PanHunter is accelerating exponentially
Demand for PanHunter is accelerating exponentially EVOpanHunter – “…just the beginning” Cumulative active users 500 per month 450 400 350 300 250 200 150 100 50 0 2014 2015 2016 2017 2018 2019 2020 2021 PAGE 28
Slide 29: We integrate AI/ML throughout the value chain
We integrate AI/ML throughout the value chain AI/ML driven drug discovery Target ID/ Hit Lead Pre-clinical Phase I Phase II Phase III Market validation identification optimisation tox testing Multi-omics data analytics and compound profiling along the full value chain Generative design1) Safety pred
Slide 30: Predicting drug profiles will improve POS
Predicting drug profiles will improve POS Comprehensive Omcis data enables prediction of efficacy and safety Reversal of disease signatures Determination of MoA Safety prediction PAGE 30
Slide 31: From patient data to discovery & stratification
From patient data to discovery & stratification PAGE 31
Slide 32: Quote: E.MPD and precision medicine
“E.MPD re-defines patient popu- lations according to disease mech- anisms rather than symptoms. This paves the way for true precision medicine“ Uwe Andag / Christiane Honisch PAGE 32
Slide 33: Deep understanding of biology for precision medicine
Deep understanding of biology for precision medicine The Evotec Molecular Patient Database (E.MPD) Tuber- Liver culosis disease Multiple rare Infectious diseases diseases Autoimmune CNS disease Kidney diseases Fibrosis Lung & Multiple cancers Inflammation Womens‘ … Health PAGE 33 EVOpanOmics: Genomi
Slide 34: Constantly fast growing unique human data source
Constantly fast growing unique human data source The Evotec Molecular Patient Database (E.MPD) in numbers # of patients # of samples # of data points generated so far 15,000+ 200,000+ 200 bn+ # of searchable data categories # of clinically described diseases / # of targets from E.MPD in target per p
Slide 35: Globally leading in kidney diseases
Globally leading in kidney diseases Overview/Example: >12,000 participants; >3,000 kidney biopsies 2018 2019 2020 2021 2022 CKD & NephSyn CKD (SKS) Healthy controls CKD NephSyn ~ 4,000 patients ~ 2,000 patients 1,000 patients ~ 5,000 patients ~ 300 patients ~ 900 kidney ~ 250
Slide 36: Re-defining health and disease
Re-defining health and disease Starting point for drug discovery, patient stratification & biomarker Modality agnostic 1st in class >200,000 samples Billions of data points EVOpanOmics & EVOpanHunter analysis treatments Patient cohorts Clinical data Renal diseases Co-morbidities, Histology, Accurate
Slide 37: Significant advantages over public domain data sets
Significant advantages over public domain data sets Standard technology, data QC and rich data annotations to improve outcomes Public Domain Cohort planning/design Physician engagement Prospective and longitudinal studies Clinical data Sample ID linked to source data (e.g. hospital) Sample tracking
Slide 38: High quality data sources for increased probability of success
High quality data sources for increased probability of success Superior signal-to-noise ratio on Evotec data compared to public domain data sets Cohort information OMICS / and medical records sequencing data Kidney cohorts Public domain data Public domain data No data QC info Several platforms
Slide 39: Molecular profiling leads to deep understanding of diseases
Molecular profiling leads to deep understanding of diseases Multi-omics analysis on patient samples SNP analysis (DNA) Transcriptomics (mRNA) Proteomics (Proteins) Clinical parameter Polygenic risk score RNA sequencing to identify key Proteomics analysis to Integration of clin- based on DNA analysis
Slide 40: From bedside to bench and back to bedside
From bedside to bench and back to bedside Example: Human target ID increases PoS Patient serum proteomics analysis Rodent kidney Negative correlation of target transcriptomics abundance to kidney function Healthy vs diseased rats: Counts per million (CPM) 1 2 eGFR (kidney function) (eGFR) in CKD pat
Slide 41: E.MPD core component of alliances in CKD
E.MPD core component of alliances in CKD CKD strategic drug discovery deals – “…just the beginning” 2016/17 2018/19 2020 2021 2022 Upfront Vifor funded: € 25 m Upfront Upfront Upfront Research funding 50% on all projects Research funding Research funding Research funding Milest
Slide 42: Accelerating changes in the healthcare paradigm
Accelerating changes in the healthcare paradigm Omics based patient stratification, disease prevention and early intervention Established Doctor exams diagnostics Patient sees doctor Follow-up diagnostic Disease specific patient and makes Too late Interventions only with symptoms tests (blood, urine
Slide 43: World leading in blood transcriptomics
World leading in blood transcriptomics Deep insights into kidney disease E.MPD NURTuRE and GCKD cohorts Blood transcriptomics at Evotec, # of blood samples for blood transcriptomics are unique in kidney disease: 6,450 by far highest sample numbers 4,890 extensive clinical data most comprehen
Slide 44: OMICS-based patient stratification
OMICS-based patient stratification A new way to define patient populations Patients Blood EVOpanOmics & EVOpanHunter A map of health and disease Risk for disease and disease progression – Personalised disease monitoring PAGE 44 PoS: Probability of Success
Slide 45: EVOgnostic learning models (AI/ML) in kidney disease
EVOgnostic learning models (AI/ML) in kidney disease Patient stratification and personalized disease monitoring AI/ ML used to assign disease >2,000 CKD patients plotted NURTuRE – patient cohort CKD risk score to each patient according to CKD risk score CKD risk score CKD risk groups Todays diagnost
Slide 46: Disease insights down to the single cell level resolution
Disease insights down to the single cell level resolution EVOgnostic AI/ML tools in kidney disease >2,000 CKD patients plotted according to CKD score Risk Groups PAGE 46
Slide 47: Genetic predisposition adds an additional layer for prevention
Genetic predisposition adds an additional layer for prevention Predisposition for disease/phenotype based on genetics Genomics Genetic predisposition determined for >5,000 CKD patients Patients with genetic predis- Proportion of patients position show higher risk of developing disease No High Risk L
Slide 48: E.MPD expansion will accelerate multiple therapeutic areas
E.MPD expansion will accelerate multiple therapeutic areas Selected key events to watch out for Access to selected kidney disease etiologies Prospective longitudinal studies Pre-disease / early stage disease cohorts Kidney diseases Access to NAFLD, NASH & liver fibrosis patient cohorts Add
Slide 49: iPSC drug screening to end erratic target selection
iPSC drug screening to end erratic target selection PAGE 49
Slide 50: Quote: iPSC-based drug screening
“iPSC-based drug screening enables identification of novel, highly disease-relevant targets.“ Nele Schwarz / Cord Dohrmann PAGE 50
Slide 51: iPSC drug discovery platform in numbers
iPSC drug discovery platform in numbers Selected KPIs # of patient derived iPS cell lines # of iPSC-derived cell types # of iPSC-derived cells per year 300+ 15+ 500 bn+ # of drug discovery programs # of cpds screened per year First molecule in the clinic 20+ 1.0 m+ EVT8683 PAGE 51
Slide 52: iPSC technology shifts drug discovery paradigm
iPSC technology shifts drug discovery paradigm Focus on disease relevance Hit Lead Target ID/ Pre-clinical identi- optimi- Phase I Phase II Phase III Approval Market validation development fication sation Con- Limited disease relevance High attrition rate Limited disease ventional relevance immortal
Slide 53: >15 cell types, co-cultures and organoids established
>15 cell types, co-cultures and organoids established Development of iPSC drug discovery platform What is next? 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 T-cells Alpha-actinin Islet1/βIII tubulin DDR2 GFAP βIII tubulin/TH Iba1/PU1 MBP/OLIG2 ZO1/Aquaporin Rhodopsin Skeletal muscle Hepat
Slide 54: Unbiased identification of relevant drug candidates
Unbiased identification of relevant drug candidates Screening to revert molecular patient profiles to the healthy state Transcriptome profiles induced by individual com- pounds in patient-derived cellular disease model Healthy Patient-derived in vitro Identifies most suitable disease model chemi
Slide 55: iPSC already highly productive, …but just the beginning
iPSC already highly productive, …”but just the beginning” Continuous expansion of disease models and programmes & iPSC neurodegeneration Oct 2017 Oct 2018 Jun 2019 Nov 2019 Oct 2020 Oct 2021 Jan 2022 US$ 5 m – US$ 6 m – US$ 9 m – US$ 30 m – US$ 6 m – US$ 9 m – US$ 15 m – Development of novel Scree
Slide 56: iPSC platform shifts drug discovery & cell therapy paradigms
iPSC platform shifts drug discovery & cell therapy paradigms Holistic approach to identification of novel therapeutic options Neurodegeneration, Diabetes Neuroinflammation & Neurodevelopmental Diseases Beta cells Cortical neurons, Microglia, Astrocytes, Oligodendrocytes, Cortical neurons Immuno-onco
Slide 57: AMD – a global pandemic affecting ~200 million people
AMD – a global pandemic affecting ~200 million people High disease prevalence coupled with huge unmet medical need People effected to rise to 288 million by 2040 Age-related macular degeneration in the world Major cause of blindness in developed world Current treatment options severely limited
Slide 58: AMD – a complex, challenging eye disease
A complex, challenging eye disease Age-related macular degeneration (AMD) Pathogenesis Progressive disease with loss of vision in center of visual field (macula) Family Advancing Obesity history age Challenges No predictive in vitro models and in vivo models Strong genetic heterogeneity AM
Slide 59: Taking human iPSC-based drug discovery to a new level
Taking human iPSC-based drug discovery to a new level Biologically relevant translational iPSC models combined with innovative platforms Phenotypic screening rule of 31) System – HUMAN High relevance to human disease at screening stage Replacement of rodent cell & in vivo models Stimulus – R
Slide 60: Predictive cell models to target retinal cell function & cure AMD
Predictive cell models to target retinal cell function & cure AMD Better strategy to develop therapies Unbiased phenotypic screening to identify new targets driving processes towards iPSC-derived retinal cells retina protection Assay system with high translatability to clinical setting Phenotypi
Slide 61: Leading proprietary platforms drive partnership
Leading proprietary platforms drive partnership Another step towards precision medicine iPSC platform in combination with EVOpanOmics Identify and validate new approaches for ophthalmology Upfront, FTE-based research payment, milestones and layered royalties PAGE 61
Slide 62: AI/ML specific application: DILI prediction
AI /ML specific application Drug induced liver injury (DILI) prediction PAGE 62
Slide 63: Quote: Paul Walker on safety assessment
“Advanced human relevant cell models combined with omic-technologies is transforming safety assessment.“ Paul Walker PAGE 63
Slide 64: AI/ML application: Drug induced liver injury (DILI) prediction
AI/ML application: Drug induced liver injury (DILI) prediction Transcriptomics safety database (EVOpanOmics) & AI (EVOpanHunter) Building AI/ML knowledge in Organ safety Transcriptomic database in e.g. DILI, cardiotoxicity many safety areas expansion and teratogenicity, utilizing iPSC capabilities I
Slide 65: Toxicity – leading source of drug attrition
Toxicity – leading source of drug attrition Example: AI/ML application – Drug Induced Liver Injury (DILI) prediction Drug Induced Liver Injury (DILI) Transcriptomics and AI/ML to proactively de-risk tox liabilities in integrated drug development 1% 3% 3% 2% Hepatic only 2% 21% Cardiovascular 50% 4%
Slide 66: Largest DILI transcriptomic database in the world
Largest DILI transcriptomic database in the world ….and growing Safety database creation AI/ML DILI database DILI Prediction DILI risk Partners compounds LTKB Data base Improving sensitivity 1,372 FDA DILI and specificity of assigned compounds Mechanistic safety prediction information 1,603 Marketed
Slide 67: Turning transcriptomic data into DILI prediction using AI
Turning transcriptomic data into DILI prediction using AI The largest DILI transcriptomic database at work High Low risk High risk risk Cmax DILI probability Low threshold Cmax risk PAGE 67 This representation of the workflow is simplified
Slide 68: EVOpanOmics & EVOpanHunter deliver superior DILI prediction
EVOpanOmics & EVOpanHunter deliver superior DILI prediction Gold Standard High Content Imaging vs. EVOpanOmics & EVOpanHunter Current gold standard: Prediction Image based DILI platform1) accuracy Primary human hepatocytes High content imaging – 2D PHH 70% Seven read-outs 16% Evotec’s new DILI P
Slide 69: Future of safety prediction is Omics-driven
Future of safety prediction is Omics-driven Robust human relevant cellular models combined with HT-Omics Safety database Safety liability prediction Relevant cellular models DILI risk Organotypic models & HT transcriptomics Toxicity type Hepatic Cardiovascular Mechanistic Haematological information
Slide 70: Short Break – Q&A
@ evotec Short Break — Q&A PAGE 70
Slide 71: Agenda
Agenda Action Plan 2025 update “…just the beginning” of the data-driven R&D Autobahn to Cures Precision technologies bring PoS up From molecular databases via iPSCs, to AI/ML tools at work Processes bring PoS up From targets, via full suite of AI/ML tools, to manufacturing Roundup & Q&A session PAGE
Slide 72: Matching technology with modality
Matching technology with modality PAGE 72
Slide 73: Quote: Steffen Grimm on RNA targeting
“Drug targeting is my bread and butter, targeting RNA is the luscious topping.“ Steffen Grimm PAGE 73
Slide 74: Proprietary platform drives high value partnerships
Proprietary platform drives high value partnerships Another step towards precision medicines Extensive RNA targeting platform to optimally identify RNA sequences to target small molecule ligands Research funding, pre-clinical, clinical, commercial milestones of US$ 160 m per programme and high r
Slide 75: One of the most comprehensive platforms in the industry
One of the most comprehensive platforms in the industry Key indicators # of active targets # of scientists in two largest # integrated technologies in RNA collaborations in next 3 years targeting platform 15+ 80+ >20 # of completed RNA target # people involved on Relevant players in the field specif
Slide 76: Almost limitless future of RNA targets
Almost limitless future of RNA targets Expanding the drug target space Looking beyond the undruggable proteins (mRNAs) as the target species Targeted 0% 20% 40% 60% 80% 100% Functionally characterised lncRNA genome Genome <10 >100 Transcribed untranslated RNAs Translated “undruggable” proteins Poten
Slide 77: Applying multiple modalities with new target biology
Applying multiple modalities with new target biology Strengthening offering for drug discovery Antisense (ASO) and Interfering RNA (siRNA) Small molecule Oligonucleotide therapeutics – sequence targets Structural targets Small Antisense molecule Drug Traditional Traditional drug drug Up or Down regu
Slide 78: Recognizing RNA as druggable molecule
Recognizing RNA as druggable molecule Examples of drug like molecules specifically targeting RNA Small molecules substances can be developed into potent and therapeutically active RNA-drugs Example: Ribocil, which binds to a structural “pocket” in their target RNA with high affinity and selectiv
Slide 79: Our platform makes RNA targeting possible
Our platform makes RNA targeting possible Thinking in tool boxes of technologies Structural model tool box Biological Validation Tool Box RNA-Seq Ribo-Seq RMSD (Ä) Construct Race Research Ab initio Consensus CLIP ALGORITHM design & informatics prediction validation # nucleotides In vitro CMV Lucifer
Slide 80: Bringing technology and experience together
Bringing technology and experience together RNA biology core strengths Structure biology RNA COM- SAXS DMS-Seq Cryo EM SHAPE X-ray Mass spec Comp chem structures RADES Patient samples Bioinformatics RNA LC-MS Nano-Seq RBSeq modifications AI/ML AI/ML RNA RNA structures Plasmid PCR IVT Capping RNA mod
Slide 81: Our RNA platform is industry leading
Our RNA platform is industry leading Benchmarking against the leading platforms Competitor 1 Competitor 2 Competitor 3 RNA production RNA structure analysis Screening throughput RNA CMPD library In vitro / in vivo cascade IND capacity PAGE 81 Illustrative assessment of capabilities
Slide 82: Our RNA platform in action
Our RNA platform in action Affinity, structure , therapeutic effect Validation and hit profiling Influence on RNA conformations In cellular target engagement of selective with selective binders by selective binders binders and effect on pathway Stability changes of RNA Overlay unbound and bound RNA
Slide 83: Partnerships/Alliances
Partnerships/Alliances Ongoing and upcoming projects currently planned/running 1 Takeda >10 targets ImmunOncology, rare disease SM First targets entering Hit ID 2 Not disclosed >5 targets Oncology, CNS SM Cellular active cmpds showing SAR 3 Not disclosed 2 targets Oncology SM RnD Collaborations
Slide 84: Wide open space to be conquered
Wide open space to be conquered Next actions to watch Maturation of RNA Enabling launch of additional target classes and advance platform product pipeline Develop proprietary chimeric degrader molecules RNA degraders (“RiboTac”) Fuse small molecules and Introduce switchable RNA translation for mRNA
Slide 85: Full suite of AI and advanced computational tools
Full suite of AI and advanced computational tools PAGE 85
Slide 86: Quote: Craig Johnstone on AI/ML and domain knowledge
“The intersection of quality data generation, AI/ML exploitation and deep domain know- ledge is our sweet-spot of highest performance.“ Craig Johnstone PAGE 86
Slide 87: Generative and predictive tools in daily use
Generative and predictive tools in daily use AI/ML applications at work – E.INVENT-AI # of Downloads # of predicted datasets Average time saved 20,000 in 2 months 7,392 >35% # of PubMed articles mined using # of E.INVENT-AI users # patent applications in last 24 Natural Language Processing months 29
Slide 88: Cost-effective and rapid progression to clinic with higher PoS
Cost-effective and rapid progression to clinic with higher PoS Essential components for high-performance discovery and development Expert drug hunting, problem- solving & decisions Cutting-edge AI/ML technologies Disease biology with translational focus PAGE 88
Slide 89: E.INVENT-AI and drug design infrastructure is world leading
E.INVENT-AI and drug design infrastructure is world leading Benchmarking against leading SMol technology companies Competitor 1 Competitor 2 Competitor 3 Generative AI tools Proprietary prediction & selection methods Infrastructure for high volume data generation Protein structure and SMol 4D confor
Slide 90: E.INVENT-AI: Full suite of AI and advanced computational tools
E.INVENT-AI: Full suite of AI and advanced computational tools Comprehensive AI/ML toolkit for design AI-driven generative tools Predictive models Molecular autoencoders, SLERP, Streamlined ML models – global and local virtual expansion models for virtual selection / filtration Transforms Bayesian O
Slide 91: Data and experience across entire value stream
Data and experience across entire value stream Rapid learning iterations through the drug discovery process Sourcing Target ID/ Hit-Identi- Lead Pre-clinical Phase I Phase II Phase III Approval Market novel ideas Validation fication Optimisation Data Global Local Global Local pools Models Models Mod
Slide 92: Rapid identification of problem and generation of solutions
Rapid identification of problem and generation of solutions Example: Combining SM generative design & drug hunting expertise to target nerve exposure “Do we have a candidate?” “Can we find way forward?” Solution: Bayesian Optimisation directed exploration to Series 1b; combining desirable potency, d
Slide 93: Efficient invention of quality candidate in compounds
Efficient invention of quality candidate in 350 compounds Example: Project progression as a function of iterative design Target X Inhibitors Progression and Outcome 4 5 1. First data package. High eDtM and safety issue 3 1 2 2. Bayesian Optimisation and E.INVENT-AI directed exploration to Series 1b:
Slide 94: Just-Evotec Biologics: From J.HAL to J.POD
Just-Evotec Biologics From J.HAL to J.POD Successful Execution
Slide 95: Quote: Linda Zuckerman on biotherapeutics access
“We are successfully executing on our mission to design and apply innovative technologies to dramatically expand global access to bio- therapeutics. AI/ML/NLP driven technologies provide the foundation.“ Linda Zuckerman PAGE 95
Slide 96: Successful execution in numbers
Successful execution in numbers Selected key indicators J.PLANT’s 3-Year Success rate Longest Perfusion Culture J.HAL library projected diversity in days (2022) 100% >24 60 bn Fabs Kg of mAbs against an infectious # of consecutive successful runs Highest productivity in a 500L disease target from 2
Slide 97: Enabling global access to modern biologics
Enabling global access to modern biologics Example: Efficient and flexible biologics manufacturing (EVOaccess) Partners Large and diverse library to generate antibodies Reviews and im- proves native anti- body sequences to enhance manu- facturabiliy J.DESIGN Disruptive, intensified production proces
Slide 98: A quality solution at every stage of biotherapeutic development
A quality solution at every stage of biotherapeutic development From sequence discovery to vial AbacusTM Partner Cell Line In sillico Design Tool Ig/VHH Discovery HTS Hybridoma Generator Discriminat or RANDŸ GAN ML Stability Prediction & Rank based Order by NLP Phage Display Modular, automated GMP H
Slide 99: DNA to IND to Commercialization
DNA to IND to Commercialization Benchmarking against the leading platforms Just-Evotec Competitor 1 Competitor 2 Competitor 3 Biologics Antibody discovery with CMC readiness Integrated Discovery, Pre-clinical, CMC, Clinical Low-Cost Commercial Manufacturing Mature Continuous Platform Complex Molecul
Slide 100: Abacus & RANDŸ lead to the best candidate mAbs
Abacus & RANDŸ lead to the best candidate mAbs Example: From sequence discovery to vial <1 month for characteriz. <5 months to cGMP start B-Cell Sequence Abacus & RANDŸ NLP developability Best 2 mAbs into JP3 cGMP manufacturing Panel assessment and optimization Process & Product Design EVQLQE … VEIK
Slide 101: Rapid progress to cGMP start
Rapid progress to cGMP start Example: AI/MD Abacus and RANDŸ enabled rapid cGMP start <5 months to cGMP start Cells PD Scale-down Robotics Systems 2 RCBs Cloning Process & Pro- cGMP Start Best 2 mAbs into JP3 of pools both mAbs duct Design Process & Product Design Optimized HTS Screening Intensified
Slide 102: Continuous harvest outperforms fed batch by 10x
Continuous harvest outperforms fed batch by 10x Example: Perfusion and continuous manufacturing compared to traditional fed batch Viable Cell Density (10e6 cells/mL) Comparison of 1 x 500L Run, in kg DS per 500L Bioreactor 120 10 100 8 80 6 60 4 40 2 20 0 0 Fed batch 12 day 15 day 0 1 2 3 4 5 6 7 8
Slide 103: Fully E2E continuous process outperforms fed batch 20x
Fully E2E continuous process outperforms fed batch 20x Example: More intensification, higher productivity, lower COGMs Fully E2E continuous process for late-stage products (> 25-day production) Kg DS per Bioreactor 40 1,000L 35 30 25 20 15 500L 10 5 0 Fed batch Hybrid 15d E2E 25d PAGE 103
Slide 104: Global manufacturing network is “…just at the beginning”
Global manufacturing network is “…just at the beginning” J.POD® – Current status and timings J.PLANT J.POD-2 Seattle, WA Toulouse, France 500L SUB 500L & 1,000L SUB Ph1 – 2 Clinical Ph1 – Commercial Over 34 runs Groundbreaking 2022 100% success CQV 2024 3 years J.POD-1 Redmond, WA
Slide 105: Robust growth fueled by successful execution
Robust growth fueled by successful execution Forecast revenue for J.POD in Redmond, WA (USA) Outlook of revenue for J.POD in Redmond, WA (USA) J.POD-1 Redmond, WA, (USA) €250M expansion and build-out towards optimal efficiency 2024 full commercial readiness anticipated J.POD-2 Toulouse, Fran
Slide 106: Agenda
Agenda Action Plan 2025 update “…just the beginning” of the data-driven R&D Autobahn to Cures Precision technologies bring PoS up From molecular databases via iPSCs, to AI/ML tools at work Processes bring PoS up From targets, via AI/ML tools, to manufacturing Roundup & Q&A session PAGE 106
Slide 107: Quote: Matthias Evers
“I have spent my career as a scientist and con- sultant in life sciences on the confluence of bio- logy, chemistry, and technology more broadly, i.e., data, A.I./ML as well as automation. Evotec’s platform is uniquely positioned to capitalize on this opportunity as we enter a new era of science and
Slide 108: Matthias Evers new Chief Business Officer
Matthias Evers new Chief Business Officer Portrait PhD molecular biochemist and bioinformatician 20 years of experience as Senior Partner of McKinsey & Company helping R&D organizations globally to excel at innovation through large-scale performance transformations, innovation programs, and nove
Slide 109: Setting the pace to accelerate growth along Action Plan 2025
Setting the pace to accelerate growth along Action Plan 2025 Selected key events to wach 2022 R&D Undisrupted growth in line with AP 2025; Continued double digit growth of efficiency base business (EVOiR&D) platforms Significant capacity and value chain expansion for all modalities and sites N
Slide 110: Many thanks for your participation!
Many thanks for your participation! Evotec SE, Capital Markets Day, 2nd March 2022
Slide 111: Questions and Answers