#1evotec The R&D Autobahn to Cures Evotec SE, Capital Markets Day, November 19th 2020 SMALL - MOLECULES BIOLOGICS #RESEARCHNEVERSTOPS DRUG DISCOVERY PLATFORM CELL & GENE PRECLINICAL THERAPY COMMERCIAL PRODUCT DISCOVERY#2PAGE 2 evotec Forward-looking statement Information set forth in this presentation contains forward-looking statements, which involve a number of risks and uncertainties. The forward-looking statements contained herein represent the judgement of Evotec as of the date of this presentation. Such forward-looking statements are neither promises nor guarantees, but are subject to a variety of risks and uncertainties, many of which are beyond our control, and which could cause actual results to differ materially from those contemplated in these forward-looking statements. We expressly disclaim any obligation or undertaking to release. publicly any updates or revisions to any such statements to reflect any change in our expectations or any change in events, conditions or circumstances on which any such statement is based.#3evotec Werner Lanthaler CEO Karen Lackey Integrated Drug Discovery PAGE 3 Let's talk about Evotec Capital markets day 2020 Cord Dohrmann CSO Jim Thomas Just - Evotec Biologics G Craig Johnstone COO Enno Spillner CFO#4evotec VROOOOOM#5PAGE 5 evotec Agenda The R&D Autobahn to Cures Our business strategy Data driven precision medicine From patient to patient Drug discovery, development & biologics From machine learning to the factory of the future "...just the beginning"... of the shared economy of drug discovery & development#6PAGE 6 evotec "R&D precision and efficiency is not just a skill, it is an attitude. We want to dramatically expand and accelerate access to better drugs.“ Werner Lanthaler#7PAGE 7 evotec Agenda Our business strategy Co-owned assets Pipeline evolution#8PAGE 8 evotec Manfred Eigen Nobel Prize 1967 ... it is just the beginning Our mission We put drug discovery ideas and leading technologies across all modalities to action. We enable and accelerate the development of precision medicines together with our partners. #RESEARCHNEVERSTOPS#9evotec Development costs per asset Cost per asset increased ~2/3rd since 2010, in US$ bn 1,981 PAGE 9 1,188 2010 2019 2010 Peak sales per drug Average sales more than halved since 2010, in US$ bn 816 Industry dynamics suggest need for disruptive approach in R&D R&D megatrends > 367 2019 +67% -55% R&D budgets continue to grow in US$ bn 160 IRR 2015 Commercial returns decrease 10% - 200 2015 2020 - 2% 2020 Source: Visiongain - Drug Discovery Outsourcing Market Forecast 2015-2025 and Evotec's estimates; -80% +25% "The IRR turn around challenge" needs new business models#10evotec Precision Human genetics supported targets ¹) % success PAGE 10 2x All programmes Genetically supported Biomarker based stratification ²) No biomarker Our focus: More precision, higher efficiency, higher returns Data-driven precision medicine meets operational excellence 3x Selection biomarker + Efficiency Average project evaluation time in months before R&D decision ³) 8 Average project Average cost to safety4) (FGLPD) in % 100 -75% Average project -30% 2 Evotec project 70 Evotec project ¹) Margan, P. et al. Nature Rev Drug Discovery 2018 Mar 17 (3): 167-181 2) Evotec-Bayer report "Excelling Together for the Benefit of Women Suffering from Endometriosis 3) Deloitte Report Unlocking R&D Productivity, Measuring the Return from Pharmaceutical Innovation 2019 || Capitalised cost per launch, in US$ bn4) Target LO Pre- clinical Clinical 2x 4) Evotec internal; Paul S.et al Nature Rev. Drug Discov. 9 203-214 (2010). McKinsey 2014 McKinsey one advantage to product launch (2014). FGLPD= First good laboratory practice dose in safety assessment Launch#11evotec Global pharmaceutical R&D market¹), 2) in US$ bn Cell & Gene Tx Biologics Small molecules PAGE 11 670 Multimodality is reality Small molecules, biologics & other modalities 2017 703 2018 743 2019 +7% ¹) Small molecules forecast from May 2017 and Biologics forecast from Dec 2017 2) Excluding sales not classified by EvaluatePharma Source: EvaluatePharma; Evotec estimates 794 2020e 856 2021e 918 2022e CAGR 2017-22e >50% 8% 5%#12HI aptuit PAGE 12 evotec 1 st J.POD New technologies, more precision, higher speed and efficiency Evotec Sites & number of employees Princeton, Seattle, Branford, Watertown -350 Orth an der Donau -30 Verona -700 - || - TE aptuit O Hamburg (HQ), Goettingen (Manfred Eigen Campus) Cologne, Munich, ~830 Abingdon (Dorothy Crowfoot Hodgkin), Alderley Park -820 Lyon, Toulouse (Campus Curie) ~750#13evotec R&D and IP generation PAGE 13 : evotec VERTEX NOVARTIS BAYER Takeda Lilly Holl novo nordisk ucb gsk AstraZeneca SANOFI : ... Combining best of both worlds Our unique business model Thermo Fisher SCIENTIFIC Wuxi AppTec Catalent. Lonza evotec IQVIA charles river PPD LabCorp evotec Partnership services Partners share with us because of • Unique IP & know-how • Unique platforms Significant efficiency gains#14Revenues in € m 128 evotec 1 PAGE 14 270% Our strategy delivers significant growth and value potential Development from 2015 ... to 2020 (e) 470+ >>1,000% Co-owned programmes Co-owned companies & BRIDGES Top-class employees 20..... 400% 24 1,000 240% 100+ 3,400+ Adjusted Group EBITDA in € m 9...... >1,000% 10 .... Unpartnered R&D projects 100+ 250% 35#15PAGE 15 evotec Agenda Our business strategy Co-owned pipeline & examples Pipeline evolution#16evotec R&D and IP generation PAGE 16 Power of novelty and precision opens path to co-ownership Unique business model Co-owned pipeline World-class partnership services Sources for Co-ownership 1 EVT platforms 2 Indication driven target pipelines 3 BRIDGES, operational ventures, III#17evotec R&D and IP generation PAGE 17 Co-owned pipeline has multiple starting points Unique business model - Sources for "Co-ownership" Co-owned pipeline Focus of today World-class partnership services Sources for Co-ownership 1 2 3 EVT platforms e.g. iPSC, Protein degradation, PanOmics, PanHunter, HAL, High-value IDD Indication-driven target pipelines e.g. P2X3, B1, A2a, . BRIDGES, operational ventures e.g. Lab282, Exscientia, Topas, Breakpoint, ...#181 2 3 PAGE 18 evotec EVT Innovate platforms (Examples) iPSC Boehringer Ingelheim Neurodegeneration UF: $45 m + $ 30 m MS: up to $ 250 m / product % up to double digit 2017 BREAKPOINT THERAPEUTICS Indication driven target portfolios (Examples) Multiple indications P2X3, B1, P2X4 UF: ND Multiple indications MS: ND %: Single digit 2011 novo nordisk 2009 Transactions are the beginning for risk-free value creation Co-owned portfolio¹)- Selected examples (in € m/ US$ m) B BAYER 5 CKD Kidney diseases UF: ND MS: > € 150 m / product % Tiered royalties UF: € 12 m MS: approx. € 580 m % up to double-digit 2018 Multiple indications R&D loan: > € 75 m Europische investitionshank 2012 BRIDGES, operational ventures, spin-offs, grants... (Examples) DNA Repair Spin-off Multiple indications Financing: > € 15 m Financing: € 30 m Protein degradation oncology 2015 UF: ND MS: up to $250 m / product % up to double-digit Solid Tumors UF: US $ 65 m MS: ND % up to double-digit LAB282 2018 ¹) Since 2012 - Status 2020; ND (not disclosed) 2017 2016 BAYER 2 Takeda NephThera PCOS Womans 'Health UF: € 6.5 m MS: > € 330 m % up to double-digit 2019 Multiple UF: ND MS: > € 150 m/product %: up to double-digit JV: CKD Financing: > € 25 m 2017 2018 Infectious Diseases UF: € 60 m SANOFI MS: not applicable %: not applicable BAYER 長 CKD UF: ND MS: > € 300 m % up to double-digit Equitiy & JV Financing: > $ 85 m Exscientia RIVEN HY KNOWLEDGE 2020 2017 2018 Upfronts > € 200 m Potential milestones > € 7 bn VC financing, R&D loans & grants > € 200 m Ø Royalties on more than 110 targets 8% (from 3 -50%)#19evotec • "Free options" through alliance building • Selected risk-shared alliances reduce upfront and research payments in exchange for milestones and royalties PAGE 19 We optimise long-term value generation Co-owning "blueprint" Alliance building R&D cost Ø € 5-10 m per project Performance-based payments to Evotec (Illustrative) Milestone Payments Ø € 200 m per project Upfront MS 1 and/or research payments € 1-50 m MS 2 MS 3 MS 4 Clinical Start MS 5 MS 6 MS 7 Ø 8-10% royalties Total Royalties & Sales Milestone#20evotec # of fully funded co-owned assets EVT Innovate 118 Bridge projects 36 Others (e.g. oper- ational ventures) 25 PAGE 20 Large portfolio across modalities moving towards market Broad and diversified pipeline of assets Discovery & Pre-clinical Clinical Example Cell & gene therapy >15 0 iPSC derived therapies Biologics > 20 1 Chikungunya virus: HBV Small molecules >65 15 P2X3#21Clinical Pre-clinical evotec Molecule EVT201 BAY-1817080 BAY-1817080 BAY-1817080 CT7001 CT7001 EVT401 BAYXXX BAYXXX BAY2328065 BI 894416 BI 860585 TPM203 DSP-1181 CNTX 6016 EVT894 BAYXXX EVT801 APN411 EXS21546 GLPGXXXX BAYXXXX QRB001 BMSXXXX EVT895 EVTXXXX PAGE 21 Therapeutic Area/Indication Insomnia (GABA-A) Chronic cough (P2X3) Overactive bladder Endometriosis Oncology (CDK7) Oncology (CDK7) > 100 highly attractive co-owned individual assets Partnership portfolio pre-clinical and clinical Immunology & Inflammation (P2X7) Gynaecology Multiple indications Gynaecology Asthma (not disclosed) Oncology (m TORC 1/2) Pemphigus Vulgaris (not disclosed) Obessive-compulsive disorder (5-HT1A) Pain (CB2) Chikungunya (Antibody) Endometriosis (not disclosed) Oncology (VEGFR3) Oncology Immunotherapy Oncology (various programmes) Fibrosis (not disclosed) Nephrology (not disclosed) Metabolic Diabetes (not disclosed) Neurodegeneration (not disclosed) HBV CNS, Metabolic, Pain ... Partner 入 京新药业 ENCARMACEUTICAL Carrick therapeutics Carrick therapeutics 康恩贝集团 CONBA GROUP Boehringer Ingelheim Boehringer Ingelheim Topas Boehringer Ingelheim Exscientia DRIVEN BY KNOWLIGGS XYNOMIC Pharma NIH SANOFI Galapagos Therapeutics SANOFI SANOFI Exscientia DRIVEN BY KNOWLIGGE QRbeta THERAPEUTICS Bristol Myers Squibb APEIRON BIOLOGICS SANOFI >10 further programmes Discovery Pre-clinical Phase I 1) Not disclosed Note: Several projects have fallen back to Evotec, where Evotec does not intend to run further clinical trials unpartnered, e.g. EVT302, EVT101, SGM-1019 Phase II Phase III#22Discovery Molecule Various ND ¹) ND 1) ND 1) INDY inhibitor Various ND¹) ND ¹) ND 1) ND ¹) ND ¹) evotec ND 1) TargetPicV Various Various ND 1) ND 1) Various Various Various ND ¹) ND ¹) ND ¹) Various PAGE 22 Follow-on discovery projects are progressing rapidly Partnership research and discovery portfolio Therapeutic Area/Indication Nephrology Nephrology PCOS Metabolic Oncology Oncology Oncology - Colorectal cancer Oncology - DNA damage response Novel antibiotics Novel antibiotics Anti-bacterial Antiviral Anti-infectives All indications Dermatological diseases Facioscapulohumeral Dystrophy Immunology & Inflammation - Tissue fibrosis Fibrotic disease Immunology & Inflammation Inflammatory Cancer Internal: Oncology, CNS, Metabolic, Pain & Inflammation 1) Not disclosed Partner AstraZeneca Fryze VIFOR PHARMA Bristol Myers Squibb M •O BREAKPOINT indivumed The Mark Foundation" for Cancer Research HELMHOLTZ RESEARCH FOR GRAND CHALLENGES GARDP celmatix THERAPEUTICL FORGE Therapeutics Faplogen 28 facio therapies ucb almirall Inha Pfizer • evotec >5 programmes encon Galápagos LABS91 LAB031 LAB10X Immunitas Aeovian PHARMACEUTICALS >40 further programmes Discovery Pre-clinical Phase I Phase II Phase III#23evotec Pipeline will strongly gain visibility with no clinical costs for us Overview of pipeline and selected upcoming events & internal champions Selected expected upcoming pipeline events in the next 12 - 24 months 1. Phase Ilb with Bayer in RCC (Eliapixant) 2. Phase II with Bayer in Overactive bladder (Eliapixant) 3. Phase II with Bayer in Endometriosis (Eliapixant) 4. Phase II initiation with BI in Oncology / Pain 5. Phase II with Bayer in Gynaecology (B1 antagonist) 6. Phase I initiation in Chikungunya virus 7. Phase I with BMS in CNS 8. Phase I with Exscientia in Oncology (A2a) 9. Phase I with Bayer in Gynaecology (P2X4) 10. Phase I with Sanofi in Immuno-oncology 11. Phase I in HBV Cure PAGE 23 12. Multiple co-owned equity companies will progress in clinic (e.g. Topas, Forge, Carrick, Fibrocor, QRbeta, ...) "Beta cell therapy is the most promising approach to cure diabetes." "IPSCs have game changing potential, they fast forward many key questions for new drugs." Andreas Scheel (Evotec) & Rainer Kuhn (Evotec) "True innovation - an antibody to be used as both: targeted therapy, and prophylactic treatment" Florian von Groote-Bidlingmaier (Evotec) "Potent molecule with sustained activity to achieve HBV functional cure." Antoine Alam (Evotec) "P2X3 is a pipeline in a molecule. E.g. in refractory chronic cough with its selectivity and side effect profile." Adam Davenport (Evotec)#24Fully owned and funded by PAGE 24 evotec VOTEC {6ojosuo available for partnering Retinopathies Хаир!у зиолчо CELL THERAPY Cardiovascular DRUG Diabetes go zaquo Neurodevelopmental Disorders 1 DISCOVERY ipsc. iP % 8 Unparalleled iPSC platform delivers big portfolio of opportunities iPSC platform portfolio licensing ¡PSC patent Neuro-Inflammation Disease Neurodegenerative & safety testing iPSC cells, technology seasuaว! BMS Huntington's Disease Foundation CHDI MORE TO COME THE IPSC LIGHTHOUSE 0 0 i • Unique to select unbiased therapeutic modality for specific disease or target • Perfect starting point for drug discovery and cell therapy - linked to technologies for disease understanding and modelling#25evotec Disease relevance at the start PAGE 25 iPSC-derived therapies have game changing potential Unparalleled fully integrated iPSC-based drug discovery platform Disease-specific drugs Screening 1 Disease in a dish FROM PATIENT TO PATIENT Patient Patient- specific iPSCs Disease-affected cell types, i.e. neurons, ... Cell therapy Drug discovery 00#26evotec 2021 (e) Ill Bristol Myers Squibb Ill Bristol Myers Squibb lli Bristol Myers Squibb™ PAGE 26 iPSC-MS iPSC-RD INK - 10 y8 IT - 10 aß iT - 10 Pre-clinical Clinical development Strong portfolio emerging in CNS, 10, and metabolic diseases Pipeline build-up to turn vision into reality in drug discovery & cell therapy iM ¹) - 10 ¡CM2) - CV iSat³) iPSC-Gaucher QRbeta THERAPEUTICS QRbeta THERAPEUTICS 1) Macrophages 2) Cardiomyocytes 3) Satellite cells muscle 4) Chondrocytes 2022/23 (e) Ill Bristol Myers Squibb Bristol Myers Squibb INK - 10 78 IT - 10 aß iT - 10 iM - IO INKT - 10 iDend - 10 iCM - CV iSat - muscle iNK - fibrosis iChon 4) - OA iPSC-Gaucher QRbeta THERAPEUTICS iPSC-LSD iPSC-ND iPSC-MS iPSC-Psych Ill Bristol Myers Squibb Bristol Myers Squibb Bristol Myers Squibb QRbeta THERAPEUTICS#27evotec PAGE 27 INSULIN oo > 7% of population ¹)2); > 20 US$ bn today - tremendous potential³)4) Paradigm shift for cure Current insulin therapy versus beta cell therapy Old paradigm Insulin injections do not address underlying cause of disease. - • Therapy dominates daily life • Glucose measurements Hypoglycemic episodes Kidney failure • Blindness • Stroke Amputation ¹) Norris et al., Lancet Diabetes Endocrinol 2020; 226-38; Chatterjee et al., Lancet 2017; 389: 2239-51 2) Globaldata list more than 500 companies active in diabetes (count includes affiliates or large pharma companies) 3) Insulin and its analogues, with or without additional technical devices like pumps, closed loop systems, etc. QRbeta THERAPEUTICS New paradigm Beta cell therapy Beta cell implant or infusion • Significant improvement in quality of life • No blood glucose measurements • No daily insulin injections • No hypoglycemic episodes • No diabetic complications Nerve damage, kidney damage, blindness, ... 4) Global data, EU5, USA and Japan, patients 20 - 64 year old; Accumulated sales of top 10 Diabetes products in 2018#28Blood glucose [mg/dl] evotec Random-fed blood glucose in diabetic mice implanted with GMP iPSC-derived beta cells ¹) 600- 500 400- 300- 200 100 PAGE 28 STZ 234567 10 14 Durable normalisation of blood glucose levels iPSC-derived islet-like clusters in diabetic animal PoC study 21 H 28 35 42 49 56 63 70 77 84 Days 92 99 Animals with control implant Animals with 1.5 m iBeta cells 107 114 122 129 136 142146 Late pre-clinical development; Phase I expected in 2021/22 QRbeta THERAPEUTICS ¹) Cells implanted in Theracyte non-oxygenated encapsulation devices subcutaneously in STZ-diabetic NOD-SCID mice 2) In this model, 4000 human donor islet IEQ are needed to establish normoglycemia (kidney capsule). Assuming a B-cell content of 35%, corresponding to 1.4 M B-cells (1000 total cells per IEQ, see https://pubmed.ncbi.nlm.nih.gov/24835624/) iPSC islet-like clusters deliver long-lasting normoglycemia at human glucose setpoint²) Significantly increased. resistance to hypoxia and post-implantation stress relative to primary human islets ³) . GMP capabilities with unique know-how • Direct efficacy comparison. to standard treatments not feasible 3) Evotec results, in agreement with Lee et al., 2009; https://pubmed.ncbi.nlm.nih.gov/19352116/, Shapiro et al.#29evotec > 1.3 bn people in endemic areas Mosquito-borne infection, most prevalent in tropical and subtropical regions with recent cases in Europe • Often misdiagnosed due to unspecific symptoms • Illness transitions to chronic arthritis like condition associated with high cost of disease management ● Infection with significant public health burden EVT894 - Chikungunya virus (CHIKV) • No effective therapies and approved vaccines; no rapid point of care test for diagnosis Chikungunya on FDA priority review voucher list PAGE 29 WHO designated neglected tropical disease Chikungunya in the world Chikungunya virus disease cases 1 100 10,000 Affected territories NIH World Health Organization ecoc Date of production: 15/07/2020#30evotec PAGE 30 Pre-clinical development Strong neutralising activity in mice and non human primates • Data suggest long half-life CHIKV Genomes/ml of Plasma 109 108 107 106 105 104 103 10² 0 T 1 Strong data in various in vivo models lead to Phase I EVT894: FIH study at Duke University initiated November 2020¹) T Control - 12.5 mg/kg SAR440894 - 0.5 mg/kg SAR440894 - 2.5 mg/kg SAR440894 - 12.5 mg/kg 2 3 4 5 Days Post Infection 1 Phase I FIH study Single ascending IV doses of EVT894 (0.3, 1, 3, 10, 20 mg/kg) • 8 subjects (6 active, 2 placebo) per cohort Projected duration 14 months, started Nov 2020 Very good synergy with Just - Evotec Biologics Reference case for "Pandemic preparedness" Phase I initiated NIH 1) www.clinicaltrials.gov: NCT04441905; Sponsor: NIH Site: Duke Clinical Research Center#31evotec A novel biologic to cure HBV EVT895: HBV infections are a major global health burden >900,000 deaths in 2015 • HBV kills as many people world-wide as HIV Current therapeutics have a low cure rate • Some being poorly tolerated Though a safe and effective vaccine is available, it is not used in all countries • It will be decades before impact is seen on global disease burden PAGE 31 HBV cure is WHO target for 2030 agenda for sustainable development Viral Hepatitis B in the world 7m AMERICAS 257 m GLOBAL 15 m EUROPE 60 m AFRICA SANOFI 21 m EASTERN MEDITERRANEAN 39 m SOUT-EAST ASIA World Health Organization 115 m WESTERN PACIFIC#32evotec PAGE 32 Interferon pathway stimulation compared to Pegasys 3,000 2,500 2,000 1,500 1,000 0,500 0,000 0,3 1 Parental Ab Abk77 3 HAH HA ng/ml T Stimulate interferon pathway and agonise CD40 EVT895: A potent biologics antiviral - Bifunctional molecule 1 10 30 100 300 1000 CD40 agonism compared to CD40 agonistic antibody O.D 2,500 2,000 1,500 1,000 0,500 0,000 1000'0 100'0 0,01 Abk 77 Pegasys ng/ml 100 Hbe release: Comparison with Pegasys (donor 2) Late pre-clinical development; Phase I in 2021 (e) Hbe release ( 160,0 140,0 120,0 100,0 80,0 60,0 40,0 20,0 0,0 E MOI 500 0,0000001 HA 0,00001 SANOFI 0.0001 Concentration (nM) 0.10 Abk77 -Pegasys#33BAYER evotec BAYER Expertise in inflammation / Women's health; market access PAGE 33 Sex hormones Key target in nerve fiber hypersensitization P2X3 antagonist – Eliapixant (BAY1817080) Endometriotic lesions Proliferating endometriotic cells Innate immune system • Innate immune .. system activation DAMPS, Iron (OS) Menstruating tissue Cell death Direct sensory nerve stimulation PGE2, Acidosis (H+), Iron (OS) NF-KB NF-KB. ● Macrophages Mast cells P2X3 Nervous system response SP, CGRP Sensory nerve stimulation NGF, IL-1B, TNF-a, PGE2 P2X3 Peripheral nervous system Primary sensory neurons Nociceptive signal Sensory nerve endings Dorsal root ganglion Derived from the strategic multi-target research alliance of Bayer and Evotec on endometriosis Infographic: Modified from Laux-Biehlmann et al, Trends Pharmacol Sci. 2015 May;36(5):270-6. * RCC: Refractory Chronic Cough Spinal cord evotec Expertise in pain research, in vivo; ion channels Multi-target notential RCC* Overactive Bladder Others e.g. neuropathic Endometriosis BAYER BAYER pain, ...#34evotec PAGE 34 سمجھے - 15 million RCC patients in US and EU High unmet medical need, no effective treatment P2X3 antagonist - Eliapixant (BAY1817080) Refractory Chronic Cough (RCC) ● Disease • RCC persists > 8 weeks; present despite guideline-based treatment Symptoms include dry irritable sensation in the throat. Symptoms not limited to coughing, may include globus, dyspnea, and dysphonia¹) • Cough refractory to treatment and/or unexplained Standard of Care • No effective treatment approved BAYER ¹) Song WJ, Chang YS, Faruqi S, et al. (2016). Defining Chronic Cough: A Systematic Review of the Epidemiological Literature. Allergy Asthma Immunol Res. 2016 Mar. 8 (2):146-55. BAYER#35evotec Phase II data support best-in-class potential P2X3 antagonist - Eliapixant (BAY1817080) Refractory Chronic Cough (RCC) Safety • Low rates of AEs including taste- related AEs PAGE 35 . All taste-related AEs mild and resolved after cessation of therapy Efficacy • Dose-dependent reduction in cough frequency over 24 hours (plateau for 200 mg and 750 mg) Daytime (awake) cough frequency similar to 24-hour frequency, showing a reduction of 36% versus baseline with 750 mg dose. • Dose-dependent improvements in cough severity and LCQ Cough frequency¹) Figure 2. Mean relative change in hourly cough frequency assessed over 24-hour periods versus placebo. ¹) Dose-dependent improvements in patients also seen in the Leicester Cough Questionnaire (LCQ) Sources: https://conference.thoracic.org/ Poster presented on the ATS 2020 International Conference Virtual Platform Mean relative change in cough frequency versus placebo (%) 35 25 15 4.0 4.0 5 -5 -15 -25 -35 -45 10% 10 mg Status: Phase Ilb initiated October 2020 - expected completion Q4/2021 -15% -23% p=0.004 200 mg BAY 1817080 dose 50 mg -25% p=0.002 750 mg Vertical lines show 90% credible limits. Duration of treatment was 1 week with each dose of BAY 1817080. BAYER BAYER#36evotec ~8-10% of women in reproductive age PAGE 36 Two more large indications already on their way P2X3 | Eliapixant (BAY1817080) – Endometriosis // Overactive bladder (OAB) Endometriosis - No proper diagnosis • Estrogen-dependent, chronic inflammatory disease caused by endometrial tissue outside the uterus • Øage at first diagnosis 28 years Symptoms include dyspareunia, cyclic menstrual pain, chronic pelvic pain, subfertility Current treatment options . No safe & efficacious long-term treatment available • No non-hormonal treatments available Status: Phase II expected to be initiated shortly Sources: Ahn S et al. (2015). Pathophysiology and Immune Dysfunction in Endometriosis. Biomed Res Int. 2015;2015:795976 Alimi Y et al. (2018). The Clinical Anatomy of Endometriosis. A Review. Cureus. 2018 Sep; 10(9): e3361. Wee-Stekly et al. (2015). Endometriosis: A review of the diagnosis and pain management. Gynecology and Minimally Invasive Therapy G -12% of adults worldwide OAB- Growing topic with ageing population 1)2)3) • First line: behavioral training . Second line: medications e.g. anticholinergics • Third line: e.g. onabotulinumtoxin • Urinary urgency, with or without urinary incontinence. Usually with urinary frequency and nocturia4),5) Standard of Care Status: Phase II initiated September 2020 BAYER ¹) MedScape: https://emedicine.medscape.com/article/459340-guidelines (retrieved March 2020). 2) MedScape: https://emedicine.medscape.com/article/459340-overview#a1 (retrieved March 2020). 3) Mayoclinic: https://www.mayoclinic.org/diseases-conditions/overactive-bladder/symptoms-causes/syc-20355715 (retrieved March 2020). 4) Milsom I, Abrams P, Cardozo L, Roberts RG, Thüroff J, Wein AJ. How widespread are the symptoms of an overactive bladder and how are they managed 5) Irwin DE, Milsom I, Hunskaar S, et al. Population-based survey of urinary incontinence BAYER#37PAGE 37 evotec Agenda Our business strategy Co-owned pipeline & examples Pipeline evolution#382015 # of projects 34 evotec PAGE 38 4 1 2 0 Building a massive co-owned clinical pipeline EVT Innovate pipeline evolution 2015-2025 (e) 0 49 Platf. & Pre- Phase Phase Phase Appro- Total Disc. clinical | __|| ||| ved 2020¹) # of projects 83 16 7 15 5 0 ¹) Does not include projects that were completely stopped, e.g. Diap277, EVT302 2) Not risk adjusted 3) Does not include EVT equity investments 112 Platf. & Pre- Phase Phase Phase Appro- Total Disc. clinical | || ||| ved 2025 (e)²) # of projects 90 > 20 >10 >20 >10 1-5 >150 Platf. & Pre- Phase Phase Phase Appro- Total Disc. clinical | ||| ved ||#39in products / in € m evotec 11 "...JUST THE BEGINNING" >5 2010 Co-owned products PAGE 39 Building co-owned product upside with limited financial risk Co-ownership business model 2010-2025 (e) 2012 2014 R&D investments ¹) >50 2016 2018 EVT Innovate EBITDA 1) without EVT Equity >100 wwwwww 100 wwwww 2020 wwwwwwwwwwwwww wwwwwww 2022(e) > 150 First royalties 2024(e) MUW MAN www. UUN 2 000 2030(e)#40PAGE 40 evotec Agenda The R&D Autobahn to Cures Our business strategy Data driven precision medicine From patient to patient Drug discovery, development & biologics From machine learning to the factory of the future "...just the beginning" ... of the shared economy of drug discovery & development#41PAGE 41 evotec Agenda Precision medicine requires a multi-omics approach Evotec's precision medicines platforms: Patient data bases - PanOmics - PanHunter Molecular patient databases the foundation of precision medicine An emerging paradigm shift in drug discovery#42evotec From patient to patient Integration of big "omics" data into drug discovery is driving precision medicine#43PAGE 43 evotec „We are living in a golden age of disruptive technologies. Applying these to drug discovery is highly exciting and rewarding." Cord Dohrmann#44evotec Genomics Transcriptomics Proteomics Metabolomics PAGE 44 MO VAN e Precision medicine requires a multi-omics approach Molecular understanding of disease mechanisms enables precise interventions Robustness Reproducibility • Day to day • Month to month • Year to year Scalability Throughput • High • Medium • Low Cost efficiency Cost efficiency • High • Medium • Low Biological insight Molecular insights in • Cause of disease • Manifestation of disease • Organs, tissues, cells#45Normalized database growth evotec Omics data entered exponential growth phase PAGE 45 140 120 100 80 60 50% 40 20 Genomics ¹) Transcriptomics ²) Proteomics 3) Pan Omics Trendline The acceleration of multi-omics data generation Lower costs and AI/ML are key drivers of a coming Omics Tsunami 0 2009 2010 2011 2012 2013 2014 2015 2016 2017 R² = 0,9995 2018 2019 2020 Sequencing costs dropped dramatically4) per Mb4) [in USD] 1000 Normalized number (2009= 0; 2019 = 100) 10 0,1 0,001 2001 2002 2003 2004 2005 ML + drug Al + drug 2006 2007 2012 2008 2009 2010 2011 2012 2013 2014 2013 2014 2015 2015 AI/ML entering into exponential growth phase 5) 100 80 60 40 20 0 2009 2010 2011 2016 2017 Moore's law 2018 2016 2017 2018 1) NCBI GenBank and WGS Statistics (https://www.ncbi.nlm.nih.gov/genbank/statistics/); 2) NCBI - Sequence Read Archive (SRA; https://www.ncbi.nlm.nih.gov/sra/docs/sragrowth/); 3) Perez-Riverol et al., The PRIDE database and related tools and resources in 2019 (doi.org/10.1093/nar/gky1106); 4) Wetterstrand KA. DNA Sequencing Costs: Data from the NHGRI Genome Sequencing Program (genome.gov/sequencingcostsdata); 5) Web of Science (http://www.webofknowledge.com) 2019 2020 2019#46evotec Molecular patient databases • Re-defining health and disease ● Defining molecular disease profiles Precision medicine is our focus Patient databases & disease models combined with PanOmics & PanHunter PAGE 46 Patient (IPSC) - derived disease models • Focus on disease relevance throughout the process • Screening / H2L / LO ... Genomics - Transcriptomics - Proteomics - Metabolomics Industrialised data generation Data science - Machine learning / Artificial intelligence - Bioinformatics AI/ML driven data analytics Molecular profiles turned biomarkers . More precise measure of efficacy and safety • Differentiation from SOC PanOmics Data generation PanHunter Data analytics -I#47PAGE 47 evotec Agenda Precision medicine requires a multi-omics approach Evotec's precision medicines platforms: Patient data bases - PanOmics - PanHunter Molecular patient databases the foundation of precision medicine An emerging paradigm shift in drug discovery#48evotec Genomics Transcriptomics Proteomics Metabolomics PAGE 48 MO VAN u PanOmics & PanHunter accelerate precision medicine Efficient data generation combined with superior data analysis PanOmics Data generation Commodity processes Proprietary RNA-Seq processes deliver unprecedented throughput and depth Proprietary proteomics processes deliver unprecedented coverage and sensitivity Commodity processes PanHunter Data analytics Proprietary multi-omics data analysis platform • Integrates bioinformatics and data science for proprietary and public domain data . Incorporates purpose-built machine. learning and artificial intelligence#49Bulk RNA-Seq evotec ● PAGE 49 -0- High-throughput transcriptomics is game changing Transcriptomics # Transcriptomics High-throughput RNA-Seq D -0- CP -0- الا -0- High-throughput RNA-Seq enables Building of molecular database to re-define health and disease • Unbiased drug screening / profiling at screening, hit to lead and lead optimization Transparent animal models with unbiased universal read-out Single cell RNA-Seq & Spatial RNA-Seq#50evotec Set of 384 well plates PAGE 50 ScreenSeq™ is scalable to >100,000 samples Evotec's automated 384-well transcriptomics platform Cell lysis and well- specific barcoding Precise gene expression quanti- fication at unsurpassed depth • Detection limit at @ 15,000 genes per sample 1 2 X A Z Z !!! 3 Library preparation & sequencing BBB BILD Protocols work for wide range of samples • Primary human cells, tissues, cell lines, 3D microtissues. Bioinformatics analysis 20 10 -20 -30 40 -30 -25 -20 -15 -10 56 1-SNE Component 1 10 15 20 Automated high-throughput process • For high-throughput compound screening 25#51evotec Format Input material Gene-targeted option Throughput Data analysis Data quality Cost efficiency PAGE 51 ScreenSeq TM platform is industry leading Benchmarking against the leading high-throughput transcriptomics platforms Competitor 1 96 well O Competitor 2 384 well O O Competitor 3 384 well O Competitor 4 96 well ScreenSeq 384 well TM#52● Mass spectrometry at industrial scale High-end mass spectrometers embedded in proprietary work flows Throughput: >100,000 samples per year ● evotec First partner: Biological samples • In vivo (patients) • In vitro (compounds) PAGE 52 TM Screen Pep ™M - Deep proteomics at industrial scale Automated sample preparation Proteomics with unprecedented performance World-leading proteomics technology and performance Exceptional proteome coverage: Up to 10,000 proteins Highest reproducibility ● Bristol Myers Squibb Deep proteome single-shot MS analysis Driven by proprietary processes and workflows Fully automated sample preparation processes Highly optimised, single-shot mass spectrometry • Dedicated bioinformatics pipeline and IT infrastructure ● Dedicated bioinformatics pipeline Activity profiles Targets Biomarker#53HT-Proteomics Integrated proteomics evotec PAGE 53 Throughput Coverage Accuracy PTMs Target Deconvolution Data analysis / Machine learning Evotec's Screen Pep™ platform is industry leading High-Throughput Proteomics and Integrated Proteomics Competitor 1 Competitor 2 Competitor 3 O Competitor 4 Screen Pep TM#54PAGE 54 evotec Agenda Precision medicine requires a multi-omics approach Evotec's precision medicines platforms: Patient data bases - PanOmics - PanHunter Molecular patient databases the foundation of precision medicine An emerging paradigm shift in drug discovery#55PAGE 55 evotec Lung & Multiple cancers Womens' Health The foundation of precision medicine Molecular patient data bases are re-defining health and disease Kidney diseases CNS Multiple rare diseases Tuber- culosis Go Liver disease PanOmics Data generation INFECTIONS PATIENT DATABASES WITH CLINICAL AND MOLECULAR PROFILES PanHunter Data analytics Infectious diseases Autoimmune disease Fibrosis Inflammation PanOmics: Genomics, Transcriptomics, Proteomics and Metabolomics PanHunter: Bioinformatics, AI/ML#56evotec Kidney diseases are very diverse and not well defined More precise definitions and disease models are needed Broad spectrum of kidney diseases • Major category is chronic kidney diseases of unknown etiology Category "Glomerular" includes - Idiopathic membrane nephropathy - Focal segmental glomerulare sclerosis -Alport syndrome - Fabry disease PAGE 56 . Even is the cause is known, treatment is. not always clear Source: ttps://www.nurturebiobank.org/dynamic-data-ckd/ Reflux nephropathy SLE DKD_type 1 Toxic AKI 2% 1% 1% Obstructive nephropathy Circulatory Ischaemic nephropathy hypertensive CKD Other TI_Nephritis 4% 4% 5% 4% Vasculitis 6% 3% IgAN 3% 2% 6% 2% 8% Glomerular 9% Cystic 30% 9% CKD of unknown etiology DKD_type 2#57Cohort CKD NS CKD Healthy donors Healthy donors CKD evotec NS PAGE 57 NURTORE National Unified Renal Translational Research Enterprise NURTO RE National Unified Renal Translational Research Enterprise Salford Royal NHS NHS Foundation Trust QUOD Quality in Organ Donation Not disclosed Not disclosed Not disclosed Worldwide largest PanOmics approach to CKD > 10.000 patients in Chronic Kidney Disease and growing Patients 3000 800 2500 1000 200 3000 100-200 Biopsies 450 450 200 1000 100+ 500 + tbd Other Samples Blood, serum, urine Blood, serum, urine DNA, serum n/a Biopsies, glomeruli, blood, serum, urine Blood Blood, serum, urine Comment Baseline recruitment completed; 1 year follow up ongoing Recruitment 65% completed Cohort completed Kidney & donor-matched liver and heart tissues Scalable; 100 HD samples in Q3 2020 Blood: baseline & 6-years follow up option for further follow up samples Remission vs relapse https://www.nurturebiobank.org/; NURTURE: National Unified Renal Translational Research Enterprise; https://www.hra.nhs.uk/planning-and-improving-research/application-summaries/research-summaries/salford-kidney-study/; SKS: Salford Kidney Study (CRISIS); HD: Healthy Donor; NS: Nephrotic Syndrome; QUOD: https://quod.org.uk/ EVT Data Exclusivity 7 years 7 years 5 years 5-7 years 5 years TBD TBD#58evotec PAGE 58 Genomics SNP analysis • Stratification of patients according to genetic background / ethnicity 20- 0- -20- ■ South-East Asian Caucasian (British) 20 PanOmics strategy is delivering on multiple fronts Genomics, transcriptomics, proteomics African Pcal East Asian Transcriptomics - blood . Molecular patient diagnostics / patient stratification t-SNE Component 2 15- 10- 5- 0 -10- -15- -20 Genomics, transcriptomics, proteomics -15 -10 Disease X CKD predicted Disease X CKD unknown etiology O Female + Male 0 5 t-SNE Component 1 15 20 Proteomics - blood • Correlation of target expression with kidney function. $460 ligand concentration vs eGFR 80- CKD eGFR 60- 20- 6 S460 ligand#59evotec BAYER PAGE 59 BAYER Start: 2016 • Strong pipeline • Financials - UF payment: ND - Research funding - MS of > € 300 m - Tiered royalties Molecular patient databases translate to high value partnerships Partnerships deliver significant cash flow and upside evotec V VIFOR PHARMA Start: 2019 • Multiple projects • Financials evotec Neph Thera Funding of € 25 m • Evotec owns 50% of Neph Thera novo nordisk evotec Start: 2020 • Pipeline building initiated • Financials - UF payment: ND - Research funding - MS of > € 150 m / per product - Tiered royalties#60PAGE 60 evotec Agenda Precision medicine requires a multi-omics approach Evotec's precision medicines platforms: Patient data bases - PanOmics - PanHunter Molecular patient databases the foundation of precision medicine An emerging paradigm shift in drug discovery#61evotec Disease signatures CHE CHO PAGE 61 Reverting molecular disease phenotypes towards healthy state Reversal of molecular disease phenotypes ensures disease relevance diseased healthy Identifying molecular disease signatures ● • Signatures capture more complete picture disease Hit selection OH T • HTS using molecular disease signatures • Hit selection according most disease relevant profile Effective drugs normalise disease signatures .OH zg IT Reverting molecular disease signatures Ensures disease relevance in key cell types#62evotec PanOmics Data generation • Patient-derived in vitro disease model • High-throughput screen • Transcriptome analysis in 384 well format PAGE 62 Unbiased identification of disease relevant drug candidates Screening to revert molecular patient profiles to the healthy state Transcriptome profiles induced by individual compounds in patient-derived cellular disease model UMAP Component 2 3- 2 1 -2- -3- Diseased -5 -1 0 UMAP Component 1 Healthy Unwanted / toxic effects 2 PanHunter Data analytics Identifies most suitable chemical hits • Focus on reversal of molecular disease phenotype • Weed out unwanted mechanisms#63evotec 90% of all drugs fail in late stages of clinical development Drug induced liver injury (DILI) is a major contributor for drug failure • The liver is the most frequent site of adverse drug reactions - 18% of marketed drug withdrawals are due to DILI alone PAGE 63 • Animal models predict only approximately 50% of the human DILI events - More predictive models are urgently needed Primary human liver cultures combined with transcriptomics and AI/ML supported analysis - Opportunity to transform DILI prediction 90₂ % of drugs fail in late clinical development 18% drug withdrawals from the market caused by DILI US$ 2.6 and 15 years to develop a drug billion nly 50% DILI picked up in animal studies#64evotec PAGE 64 Superior DILI prediction based on PanOmics & PanHunter Gold standard High-content imaging vs. Transcriptomics (PanOmics) & AI (PanHunter) Current gold standard HCI based DILI platform ¹) • Primary human hepatocytes • Seven (7) read-outs - High-content imaging Accuracy of DILI prediction: 70% Evotec's new DILI prediction platform¹) • Primary human hepatocytes • One (1) read-out - Transcriptomics PanOmics Data generation & PanHunter Data analytics Accuracy of DILI prediction: 82% & Insights into mechanism of tox Prediction are based on 2D Primary Human Hepatocyte assay with 128 reference compounds tested (largest reference compound data base reported)#65evotec Focus on ... Focus on ... PAGE 65 Drug screening Target Molecular disease profiles are driving a paradigm shift Disease relevance is paramount: 54% of phase 3 trials fail due to inadequate efficacy¹ Disease relevance Drug optimization Efficacy IND enabling Current paradigm: Target driven / Low probability of success / 12 to 15 years Drug likeness Safety Confirmatory safety Clinical development Disease relevance Confirmatory disease relevance Future paradigm: Molecular profile driven / Higher probability of success / 8 to 12 years 1) Fogel DB. Factors associated with clinical trials that fail and opportunities for improving the likelihood of success: A review. Contemp Clin Trials Commun. 2018;11:156-164. Published 2018 Aug 7. doi:10.1016/j.conctc. 2018.08.001. Market One drugs fits all Right drug, right patient, right dose#66evotec PAGE 66 Looking at the whole picture with unbiased molecular profiles Too much target focus is limiting Target driven drug discovery BE Molecular profile driven drug discovery#67evotec Quantum leap in Drug Discovery, Development & Biologics Operational excellence, from Machine Learning to the Factory of the future in all modalities QUANTUM LEAP PF! BRAGGER. AA 1#68PAGE 68 evotec Agenda The R&D Autobahn to Cures Our business strategy Data driven precision medicine From patient to patient. Drug discovery, development & biologics From machine learning to the factory of the future "...just the beginning"... of the shared economy of drug discovery & development#69PAGE 69 evotec "More efficient and effective drug discovery and development is a global necessity. Applying machine learning is the natural evolution beyond operational excellence" Craig Johnstone#70PAGE 70 evotec Agenda Next generation drug discovery & development Al & ML in small molecules Biologics#71evotec Opening of R&D Autobahn 2015-2018 • M&A to enhance capabilities and capacity Talent acquisition PAGE 71 R&D Autobahn creates quantum leap for partners and patients. Creating the future with long and consistent vision • Cycle time, process excellence and quality enhancements *** Current & near-future state 2018-2023 • Combination of multi-modality expertise, experience, technologies, slick processes ● Application of AI/ML to high-impact problems Integration in benchmark-busting performance and unique discovery launch-pad Dal = Medicines of the Future 2023-2030 • Integration and exploitation of data surface on R&D Autobahn for even better holistic prediction • Massive reduction in costs and time in inventive, iterative discovery phases • Quantum leap to novel medicines YYY#72evotec SMALL MOLECULES BIOLOGICS --- PAGE 72 DRUG DISCOVERY PLATFORM CELL & GENE THERAPY PRECLINICAL ANTISENSE THERAPY COMMERCIAL PRODUCT DISCOVERY Growth driven by multi-modality, integration and data surface Key growth drivers for high-impact and high-value business Capabilities and expertise creates multi-modality R&D Autobahn for growth • Biologics technology disruption "Small molecules" extension to difficult targets • Gene therapy; iPSCs and scalable cell therapy Integration drives differentiation and high value Knowledge, experience and know-how creates success loop in discovery and development (> 90% return rate of partners) • Integrated working creates quality, speed, performance and inventive steps Combination of experimental data and AI/ML surface is cutting edge • Creating and exploiting data in optimised infrastructure holds huge potential POC examples: HAL, leading with AI/ML in molecular design and predictive ADMET • Laying "data surface" onto R&D Autobahn further drives competitive advantage#73evotec More efficient to high value value inflection points Key advantages to consistently deliver outstanding performance Integration across value chain • Problem-solving and inventive step creation through Integrated drug discovery & drug development • Smooth and efficient transitions within end to end process Flexible R&D Autobahn access • Capital elasticity driven resourcing High speed execution on multimodality platforms PAGE 73 Top-class scientific leaders, teams & Demonstrable know-how Overseeing, driving and piloting projects and portfolios across therapeutic areas (Disease models, design, breakthrough biology, formulation, ...) Benchmark-busting performance ¹) including attrition Cost to IND¹), in m USD 75 50 25 3 ←-30% Evotec 4 Faster & more efficient to IND inflection 30% reductions in time, 50% reductions in cost 5 Industry benchmark -50% 6 Years#74evotec Industry Evotec PAGE 74 Targets Screen Search Select M Hits D Disease area knowledge Significantly faster and more efficient on R&D continuum Key performance Indicators Rapid Inventive process Leads A INDIGO Cutting edge technologies PDC 18 months faster to IND IND 4yr US$ 30-40 m ¹) ¹) including attrition; DMTA = Design make test analyse I I I I IND 5.5yr US$ 75 m ¹) • Benchmarks for speed not improving in last 10 years • Attrition getting even worse - more complex targets/biology? • Costs rising • Integrated processes create speed and early prioritization in cascades Expertise solves problems, creates inventive step and solutions • Costs and time go down "Innovation efficiency"#75evotec Padlock THERAPEUTICS IDD in Autoimmunity Initiated 2014 Acquired by BMS¹) TESARO IDD in Oncology Initiated 2016 PAGE 75 Acquired by GSK 5) FIBROCOR IDD in Fibrosis Initiated 2017 Partnered with Galapagos 20192) disarm THERAPEUTICS IDD in CNS Initiated in 2019 Acquired by Lilly6) R&D Autobahn creates much better exit points for our partners Selected examples of impact and value inflection Aeovian PHARMACEUTICALS IDD & INDIGO in Rare & Age Related Initiated 2017 PDC Milestone 20183) enterprise THERAPEUTICS IDD in Respiratory Initiated in 2018 Project acquired by Roche7) FORGE Therapeutics IDD & DEV in Infectious Dis. Initiated 2016 Partnership with Roche 20204) Multiple others in stealth mode... Pain, Oncology, ID, Metabolic, etc. ¹) https://lifescivc.com/2016/03/bms-secures-keys-padlock/ 2) https://www.evotec.com/en/innovate/spin-offs-and-other-holdings/fibrocor-therapeutics 3) 4) https://forgetherapeutics.com/forge-enters-into-collaboration-with-roche-to-develop-novel-sntibiotic-to-treat-lung-infections/ "With Padlock, we decided to do something different - and signed up for a single, large collaboration with Evotec, where they would cover all of our research. They were essentially our entire discovery execution team. It's obviously worked well ... It also simplified the operating model enormously. "We are extremely glad about the fast progress of our programme which has been made possible by Evotec's expertise and technical excellence. The successful identification and de-risking of our lead clinical candidate was instrumental in the closing of our Series A financing. Stelios T. Tzannis, PhD, President & CEO of Aeovian https://www.evotec.com/en/invest/news--announcements/press-releases/p/evotec-reaches-milestone-in-integrated-drug-discovery-and-development-partnership-with-aeovian-5851 Bruce Booth, Partner Atlas Ventures 5) https://www.gsk.com/en-gb/media/press-releases/gsk-completes-acquisition-of-tesaro-an-oncology-focused-biopharmaceutical-company/ 6) https://www.bloomberg.com/press-releases/2020-10-15/lilly-announces-agreement-to-acquire-disarm-therapeutics 7) https://enterprisetherapeutics.com/enterprise-therapeutics-first-in-class-tmem16a-potentiator-program-for-treatment-of-cystic-fibrosis-and-other- respiratory-diseases-acquired-by-roche/#76PAGE 76 evotec Integrating it all for higher productivity SMALL MOLECULES BIOLOGICS & ANTIBODIES CELL & GENE THERAPY 1 1 R&D-AUTOBAHN TO CURES 1 10 ANTISENSE O O O LO ..#77PAGE 77 evotec Agenda Next generation drug discovery & development Al & ML in small molecules Biologics#78PAGE 78 evotec "My passion is putting our inventions in patients" Karen Lackey#79evotec Current capabilities: Strengths in full value chain New molecules generated PAGE 79 1 Deep learning 3 Score 2 Multi-objective optimisation Policy gradient reinford to deliver optimal solution Small molecule computational drug discovery & development Overview Growing in deep learning & knowledge building Future state in applications of knowledge built 101101011101010010 10011 1001 110011010111 1 101 01110 10011 110110011 11010101001101 1001101011101 1001 101#80Less data available: HTS Ongoing investment in screening collection 74 66 evotec 49 41 33 25 16 8 0 PAGE 80 Row Count 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 QEDw mo3 (16 bins) Evotec Screening Collection Known Drugs 49000 44300 39400 34500 29500 24600 -19700 15000 -10000 5000 0 Balancing speed, cost & probability of technical success Early Hit ID: Extensive capabilities in small molecules Highest quality chemical start points • Best value Desirable feature QED 3 D shape Lip E LE N=N ara Molecular term Composite property clog D MW Value 0.91 chiral 1.1 314 GPCR homology model directed screen Protein//DNA endonuclease SBDD/LBDD screens Protein//Protein inter-action SBDD/LBDD screens ¹) Wu etaCBB 201 More data available: Virtual Screening Skp2 Chan of Cill 2013 Field Pharmacophore guided LBDD Protein//RNA transferase SBDD/LBDD screens Pharmacophore guided LBDD excluded volumes F8: Aro (aternativel F1: Cat&Don Få: Ara P3: Hyd F7: Aco FG: Don F5: Acc F4: Aro F2: Aro#81evotec 1. Bayesian optimisation Exploration PAGE 81 What molecule provides maximum information to the model? 2. Generative design Exploitation New molecules generated Deep Learning 2 Score Multi-objective optimization Policy gradient reinforces to deliver optimal solution Optimising features with Evotec's molecular design apps Fit-for-Purpose application of tools to drive project success Project MPO Score 5.5 5 45 3.5 0 1 2 20 3 40 6 months (intermediate goal met) Early lead Project MPO score vs Sequential ID 60 A pre-clinical drug candidate in 12 months and < 150 compounds 80 Sequential Compound ID 100 4 pre-candidates identified for downstream profiling 120 12 months SEN CRECIO 140 3. Quantum mechanics | Ki 460 nM Y. Global Model FMO-based SBDD 4. Machine learning DMPK Project Models EVO PPS HUMAN (RFR) EVO PPB RAT (RFR) EVO LOGO (RR) EVO MICS HUMAN (RFC) EVO MICS RAT (RFC) EVO MICS MOUSE (RFC) EVO PPB MOUSE (RFR) EVO SOLUBILITY (RFC) EVO HERG IRFRI EVO HERG CLASS (RFC) EVO CACO2 (FR) Your Models Shared Models HERG 112 Ki 4.1 nM 780 371 Model Performance-fold strated cross vadation)#82evotec Human PK and dose prediction • Multispecies prediction Improve quality and reduce costs to accelerate to INDs Development readiness: discovery to development continuum Target tissue concentration Continuum of predictions to optimise human PK during discovery process Predicting solid state • Design for solubility, polymorphism screening & crystallisation processes • Batch physical purity & crystal structure determination PAGE 82 Vein Portal Pancreas Gut t (faka) Spleen R Venous Blood Muscle Kidney R K Telt QTA CL Arterial Blood In Silico Relative Solubility Solvent Screening of Structure 1 @25°C Nared to the best solvent Predictive sciences drive faster and higher quality IND R Luno R Building development into Dx chemistry • >13,808 development transformations in a Dx reactions database Right First Time approach = no reengi- neering of process route for development Drug-induced liver injury Library of toxicology profiles • Integrated Al and machine learning to enhance predictive power Unrivalled mechanistic insight ● Action Ectrophilic reaction of 126 Aside formation R AI Alle An 11 SE X-OBLOm ->> GOPA Accuracy 78-82%#83evotec Biomarker identification • Big Data analysis platform • In-house quality data sets . Data curation. PAGE 83 Biomarkers link all discovery and development work to patients Integration of AI approaches to increase success in translation of pre-clinical discovery GCP Good Clinical Practice Biomarker validation & optimisation t SNE Component 2 15 10 5 -10- -15- -20 -15 -10 Etiology X CKD predicted etiology X CKD unknown etiology Female Male -5 0 5 t SNE Component 1 • Hypothesis testing and cross- validation on new cohorts • Multi-variate signatures 10 Translation of biomarkers & companion diagnostics € Integration of clinical results . Retroactive refinement of predictivity and safety#84evotec PAGE 84 Data collection Unstructured data Commercial data Public data New and pro- prietary data Training algorithms with high quality well-curated data Medium term objective: Dramatically improved designs through prediction Data curation and prediction Machine learning models. Deep learning обек Connected prediction, end-to-end Experiment Evotec data surface SMALL MOLECULES BIOLOGICS & ANTIBODIES R&D-AUTOBAHN TO CURES CELL & GENE THERAPY 0 0 1 1 0 100 1 Comprehensive cartography Connect molecular disease and molecular interventions Knowledge graph#85evotec Biomarker identification PAGE 85 e Future state: Quantum leap in exploitation of knowledge in all domains to invent and produce medicines of the future Schematic representation of future state medicines discovery and development Redefined diseases at molecular level 1101011101010010 10011 1001 110011010111 101 01110 10011 110110011 11010101001101 1001101011101 1001 1100 101 Optimal phenotypic, drug-like and developable properties at point of molecular invention Uncovering disease-disease relationships through the incomplete interactome. DOI: 10.1126/science.1257601 Translation of biomarkers & companion diagnostics R&D-AUTOBAHN TO CURES SMALL MOLECULES BIOLOGICS & ANTIBODIES CELL & GENE THERAPY 01 O ANTISENSE From invention to patients on digital, frictionless surface on the multi- modality Autobahn#86PAGE 86 evotec Agenda Next generation drug discovery & development Al & ML in small molecules Biologics#87PAGE 87 evotec C "We're using our deep understanding of data science to deliver critical industry solutions and drive global access to important biotherapeutics" Jim Thomas#88500LSUB evotec Perfusion PAGE 88 другор J.HAL Surge Vessel Continuous ProA in vivo Continuous Low pH VI Mixer Mixer Surge Vessel Cont. Depth Filtration A Al generated and in vivo discovery Surge Vessel Common data platform coupled to powerful data science Integrating molecular, process and manufacturing design delivers excellence Cont. Polishing Chrom Surge Vessel Continuous ILC/ILDF Pool Vessel MOLECULE DISCOVERY J.DISCOVERY™ T( MANUFACTURING DESIGN J.PODⓇ J.DESIGN MOLECULE DESIGN J.MD™ PROCESS & PRODUCT DESIGN JP3Ⓡ Media Cells Continuous Harvest (Cell-Free) Cell Bleed Abacus optimisation to fit PD Robotic High- throughput PD End to end continuous processing (E2E) Machine learning (ML) and Artificial intelligence (AI) are maturing our integrated biologics platform (J.DESIGN) Dynamic predictive process control#89Amount of data available evotec PAGE 89 (100,000s) MOLECULE DISCOVERY J.DISCOVERY TM Sequence to structure model HTLYDTLY- RRGPTTLFGV. Data are captured and archived using common data platform, ML tools accelerate learning Platform overview (10,000s) MOLECULE DESIGN J.MD" TM Structure to characterisation model (100s) PROCESS & PRODUCT DESIGN JP3Ⓡ Characterisation to PD model a 300 250 200 150 100 50 hacon eve 5.5 6.0 ******** pH 7.0 7.5 Common data set (J.DESIGN) Intense learning is focused on the most abundant, least expensive data - DNA sequence (10s) MANUFACTURING DESIGN J.POD PD to manufacturing model GMP#90evotec Example MOLECULE DISCOVERY J.DISCOVERY TM • Discriminator neural network is lightly trained on human faces PAGE 90 Generative Adversarial Networks (GANS) to create faces in silico Go to www.thispersondoesnotexist.com to find out more • Generator creates images that sometimes fools the Discriminator, and learns from this experience • Discriminator is trained with more real human faces, forcing the Generator to improve . Eventually Generator can fool a human Training faces Noise MOLECULE DESIGN J.MD Generator PROCESS & PRODUCT DESIGN JP3 Discriminator Generated faces Real/ Synthetic MANUFACTURING DESIGN J.PODⓇ AB CD#91evotec MOLECULE DISCOVERY J.DISCOVERY PAGE 91 TM GAN technology to create human-like antibodies 100,000s of natural human antibody sequences in the public domain serve as the training set Example • Discriminator neural network is lightly trained on normal human antibodies MOLECULE DESIGN J.MD OAS set- 400,000 per chain Human Antibody Sequences (Real) Training Sequences EVQLQE... VEIKR QVQLQQ ... VEIKR QVQLVE... VEIKR Noise- PROCESS & PRODUCT DESIGN JP3 Generator Discriminator • Generator creates antibody structures that sometimes fool the Discriminator, and learns from this experience • Discriminator is trained with more real human antibodies, forcing the Generator to improve • Eventually Generator produces a diverse library of antibodies indistinguishable for human antibodies We can use GAN technology to create human-like antibodies - indistinguishable from normal human antibodies 43 QLQLVE ... VEIKR Generated molecules (Synthetic) MANUFACTURING DESIGN J.PODⓇ Real or Synthetic Human-like antibody#92evotec MOLECULE DISCOVERY J.DISCOVERY TM PAGE 92 Billions of human-like antibodies created to screen for activity Transfer learning can bias libraries toward antibodies with superior qualities Example • Discriminator neural network is lightly trained on normal human antibodies • Generator creates antibody structures that sometimes fool the Discriminator, and learns from this experience • Discriminator is trained with more real human antibodies, forcing the Generator to improve • Eventually Generator produces a diverse library of antibodies indistinguishable for human antibodies MOLECULE DESIGN J.MD OAS set- 400,000 per chain Human Antibody Sequences Noise- (Real) Training Sequences EVQLQE... VEIKR QVQLQQ ... VEIKR QVQLVE... VEIKR Generator PROCESS & PRODUCT DESIGN JP3 Discriminator 4* QLQLVE ... VEIKR Generated molecules (Synthetic) Real or Synthetic MANUFACTURING DESIGN J.PODⓇ Screen Diverse library of human-like antibodies (> 109) Libraries containing billions of human-like antibodies are being created to screen for therapeutic activity Target antigen#93PAGE 93 evotec MOLECULE DISCOVERY J.DISCOVERY Partner or client antibodies from animals or people are improved for manufacturing and formulation Abacus - an in silico computational toolset of ML algorithms Abacus Antibody Fv with hot spots displayed IGHV3-30-01 o IGMV3-30-3-01 01GHV3-33-01 MOLECULE DESIGN J.MD™ 67 TM KV67 ASFLYSGVPSRFSGS KV:81 KV:56 YSASFLYSGVPSRF KV78 2.18 KV:52 PKLLIYSASFLYSGV KV:74 2.25 KV51 APKLLIYSASFLYSG KV 73 2.07 KV:50 LIYSASFLYS KV:72 1.84 Hot spot tables Clading Germline Switching 1.88 1.95 Sequence Predicted binding to HLA DRB!* start peptide end 0101 0301 0401 0701 0801 1101 1301 1501 KV.57 YSASFLYSGVPSRFS KV:79 KV-58 SASFLYSGVPSRFSG KV:80 2.07 3.48 1.90 1.69 3.33 1.77 1.83 Variable Region Hotspots gut Export to Exca Export SV for HotSpot Highlighting in MOE Covariance Ste Coverance Site Covariance Site Non-Standard N-La Cycosylation Site Potential Isomerization Site (CDR) Potental Deamidation Site (CDR) alignments 1.74 CONSENSUS Structure-o ipilimumab Odenosumab ML Tier 2 T2 based gantenerumab Ter 2 ASN a 3.56 4.05 Predictor 4.04 PROCESS & PRODUCT DESIGN JP3 Immunogenicity Chain Kappe G HC GMC G HC DG HC lgG: HC mmomm teprotumumab E robatumumab Kapp Variable Heavy Variable Heavy Varable Heavy Variable IVLTQ SP Germline ipilimumab_LC Analysis Framework 3 103 Framework HS Framework 1 M3 CDR 2 CDR 3 KV FR1 LS ASN # ASN Location TTTTT ADDLX Score hu IGKV3-20 100.0% ohu IGKV3D-20 97.1% hu IGKV3-11 95.7% hu IGKV3D-11 94.3 % hu IGKV3D-7 92.9% HV:59.65 198:72.73 H112 113 Molecular optimisation builds in quality SP Qmax N E24 ton $3 veslation max NT >KV1:39 GERAT S Molecules optimised for Expression in cells ● MANUFACTURING DESIGN J.PODⓇ • Purification . Formulation ● Long-term stability#94evotec mAb 1 PAGE 94 MOLECULE DISCOVERY J.DISCOVERY™ mAb1 variant aggregate reduction and yield improved by 60% by resistance to low pH (3.3) aggregation Dramatic improvement in low pH aggregation achieved through optimised molecular design with Abacus Sequence optimisation improves manufacturability and yield 40.0 mAb 1 MOLECULE DESIGN J.MD" 3.6 mAb1 optimised. Kerwin et al. 2019 J Pharm Sci (In Press) PROCESS & PRODUCT DESIGN JP3 mAb2 3.4 mAb2 optimised. MANUFACTURING DESIGN J.PODⓇ m Ab2 mAb2 variant aggregate reduction and yield improved. by 700% by resistance to low pH (3.3) aggregation#95● ● evotec • Powerful expression vectors ● Optimised cell hosts • Custom media tuned for productivity High density perfused culture conditions. • Connected downstream processing • High resolution analytical methods Highly stable formulation conditions MOLECULE DISCOVERY J.DISCOVERY PAGE 95 Propriety reagents and methods, coupled to robotics and ML can rapidly move client or partner molecules into the clinic Highly efficient process & product design delivers high quality, low cost therapeutics • Current process yields are generally 2-4 grams per reactor/L per day MOLECULE DESIGN J.MD Scale-down system Continuous Harvest (Cell-Free) Media Cells Cell Bleed 500LSUB Perfusion Surge Vessel Continuous ProA PROCESS & PRODUCT DESIGN JP3Ⓡ Robotics Continuous Low pH VI Mixer Mixer Surge Vessel Cont. Depth Filtration Surge Vessel Cells Cont. Polishing Chrom Surge Vessel MANUFACTURING DESIGN J.PODⓇ DNA Continuous ILC /ILDF HAND Pool Vessel Full scale manu- facturing#96evotec Intensified processing 500LSUB PAGE 96 MOLECULE DISCOVERY J.DISCOVERY™ Perfusion Surge Vessel Production processes are small and fit into modular clean rooms that can be reconfigured for flexibility J.POD® facility design reduces scale-up risk by scaling out, not up Continuous ProA 2010 au Continuous Low pH VI MOLECULE DESIGN J.MD Mixer Mixer Surge Vessel Cont. Depth Filtration A Surge Vessel PROCESS & PRODUCT DESIGN JP3 Cont. Polishing Chrom Surge Vessel Continuous ILC / ILDF Pool Vessel Production from a few kilograms to metric tons in the same facility MANUFACTURING DESIGN J.POD#97evotec PAGE 97 MOLECULE DISCOVERY J.DISCOVERY™ J.POD facilities are ready for precision medicine while delivering capacity for high demand biologics for a variety of partners The future is smaller, modular, flexible and highly automated Conventional manufacturing plant MOLECULE DESIGN J.MD VS PROCESS & PRODUCT DESIGN JP3 J.POD® manufacturing network Complexity managed at the process and not the plant level MANUFACTURING DESIGN J.POD#98PAGE 98 evotec Evotec is creating a multi-modality digital Autobahn for delivering critical industry solutions to partners and clients Using the power of data science to deliver enhanced speed, lower cost and predictive efficacy SMALL MOLECULES BIOLOGICS & ANTIBODIES CELL & GENE THERAPY 1 1 R&D-AUTOBAHN TO CURES 1 10 ANTISENSE O O O LO ..#99PAGE 99 evotec Agenda The R&D Autobahn to Cures Our business strategy Data driven precision medicine From patient to patient. Drug discovery, development & biologics From machine learning to the factory of the future "...just the beginning” ... of the shared economy of drug discovery & development#100PAGE 100 evotec The shared economy in discovery & development Summary Precision Medicine is paramount • Disease relevance from the beginning will redefine "drug hunting" process • Novel targets will only be progressed if disease relevance is visible in early stages of discovery, or latest early clinical evaluation ML & Al will increase R&D IRR • Unbiased application of right tools and modalities to novel biology will make drug discovery much more data driven and cost effectivel • Access to all patients has to be core consideration from start Creating co-owned pipeline is unique strategy that holds massive value • Reducing cost of capital via efficient service and sharing partnering processes is helping all parties, and most importantly patients#101evotec VROOOOOM#102evotec QUESTIONS AND ANSWERS 3 #RESEARCHNEVERSTOPS#103evotec Many thanks for your participation! WE'D LOVE TO SEE YOU FACE-TO-FACE. Evotec SE, Capital Markets Day, November 19th 2020 #RESEARCHNEVERSTOPS