Senzagen Ab · OMXSTO:SENZA
Senzagen Ab — Technical Presentation (February 2020)
59 slides from Senzagen Ab's technical presentation, Feb 25, 2020
Slide 1: All you need to know about GARD™
All you need to know about GARD™ In vitro skin and respiratory sensitization testing based on genomics and machine learning Feburary 25th, 2020
Slide 2: SenzaGen in short
SenzaGen in short Spin-out from Lund University. Research in immunotechnology. assessment of che HQ and Lab in Lund. Sales office in offering: US. Highly multidisciplinary team 3 Founded: 2010. Operational: 2014. ™ PLATFORM Racor High performance Listed on Nasdaq First North, * Broad applicability S
Slide 3: Predictive toxicology: a scientific field in change
Predictive toxicology: a scientific field in change ee A ee >) Predictive Simple workflow In vitro performance > bE ° a a = = o INFORMATIONAL CONTENT
Slide 4: High informational content > Closer to reality
High informational content > Closer to reality AN IN VITRO MODEL, WHERE FEW POINTS ARE MEASURED, GIVES NON-CONCLUSIVE RESULTS. BY MEASURING A LARGE NUMBER OF PARAMETERS, THIS MODEL MIMICS REALITY AND GIVES CON- CLUSIVE RESULTS. Increased number of biomarkers
Slide 5: High informational content > Closer to reality
High informational content > Closer to reality @ sensitizer e@ Non-Sensitizer Z Biomarker 3 Biomarker 1
Slide 6: The GARD™ technology platform - Assay development
The GARD™ technology platform - Assay development Human relevant cells in combination with Genomics and machine learning
Slide 7: The GARD platform - how it works
The GARD platform —- how it works . Prediction model | Genomic biomarker signature | Training data Biological system
Slide 8: The GARD platform - biological system
The GARD platform - biological system | Biological system |
Slide 9: Background - the Adverse Outcome Pathway
Background - the Adverse Outcome Pathway | Dendritic cells Key Event 3 >DC activation >DC maturation >DC migration Keratinocytes Key Event 2 Inflammatory mediators Key Event 1 Hapten formation Covalent binding to proteins Hybrid AOP for skin and respiratory sensitization (Kimber, 2018} Allergic Cont
Slide 10: Background - Dendritic cells (DCs)
Background- Dendritic cells (DCs) Respiratory sensitization T-cells: CD4*Th2 ©) — @ —_- ©) Humoral Immunity naive T-cells Skin sensitization T-cells: CD4*Th1 & CD8* Tc Cellular Immunity
Slide 11: SenzaCell™: a human DC-like cell line
SenzaCell™ : a human DC-like cell line Antigen presentation: MHC I, MHC II Cd1d Potent activators of T-cells: CD80, CD86 T-cell polarization: Contains the functional elements for Th, & Ths polarization
Slide 12: Transcriptomic read-out of the biological response
Transcriptomic read-out of the biological response Assay Development: Hypothesis * There are existing genes and pathways in DCs that are differentially expressed depending on the stimuli * Such genes could be used as predictive tools Sensitizer SenzaCells' Cellular responses U Non-sensitizer Transcr
Slide 13: The GARD platform - Training data
The GARD platform - Training data Training data Biological system
Slide 14: The GARD platform - Training data
The GARD platform - Training data Assay Development: Defining a Training data set to test the hypothesis. Training dataset: A set of well-characterized chemical compounds with known expected outcome. X number of Sensitizers X number of Non-sensitizers Antibiotics Cosmetics & fragrances Ideally, the
Slide 15: The GARD platform - Training data Chemical stimulations
The GARD platform - Training data 0 Chemical stimulations 2] RNA extraction 1 ct VAVATA Prepare chemicals for cellular exposure pee Stimulate cells for 24 h Qny Extract RNA from cells Global transcriptomic profiling i A Measure complete transcriptome; — @affymetrx of cells using microarrays > 29.000
Slide 16: The GARD platform - Genomic biomarker signature
The GARD platform - Genomic biomarker signature Genomic biomarker signature | Training data Biological system
Slide 17: Data-driven biomarker signature identification
Data-driven biomarker signature identification Data analysis identified differentially GARD prediction signature regulated genes in cells stimulated with Gene 1 Sensitizers or Non-sensitizers. Gene? GeneX
Slide 18: From Affymetrix to NanoString
From Affymetrix to NanoString — — nanoString # genes: Assay time: Throughput: Input: Custom - up to 800 genes 22 h, 15 min hands-on 384 samples/day Total RNA From genome-wide arrays to tailor-made biomarker readout - Progress towards routine analysis of skin sensitizing chemicals with GARD @ affymet
Slide 19: GARD assays: number of biomarker signatures
GARD assays: number of biomarker signatures GARD™ skin - 200 genes Skin sensitization testing GARD™ potency - 51 genes Skin sensitization potency testing according to GHS/CLP GARD™ air - 28 genes Respiratory sensitization testing GARD™ skin Medical Device - 200 genes Skin sensitization testing of me
Slide 20: The GARD platform - Prediction model
The GARD platform - Prediction model . Prediction model | Genomic biomarker signature | Training data Biological system
Slide 21: The GARD platform - Prediction model
By observing data from a set of objects (“training data”) with known expected outcomes (“labels”) Training Data from an unlabelled test object is fed into the fixed mathematical model which assigns a label to the new data. The GARD platform - Prediction model Machine learning is acomputer algorithm
Slide 22: The GARD platform - Prediction model
The GARD platform - Prediction model By observing data from a set of objects ("training data”] with known expected outcomes (“labels”) The training data: Gene expression profiles from a set of well- characterized sensitizers and non-sensitizers. © Sensitizer G Non-Sensitizer
Slide 23: The GARD platform - Prediction model
The GARD platform - Prediction model Machine learning is acomputer algorithm designed to learn from previous experience. If Sum <0 = Not sensitizer If sum 20 = Sensitizer £ M genes Class label with known $011 Grr e+ Guy yachine lee"nina sigortnin Training le rns & builds 2 fixed 1 1athematical model
Slide 24: S NS S
S NS ? ? S S S ? S S S S S S S S S S S S NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS
Slide 25: How to GARD™ your product in six steps
Siew to GARD™ your product in six steps SS =
Slide 26: GARD in 6 Steps
© GARD Input Finder PAV ALAA : ey eas aD omarnaliaabans Determine the concentration of the test substance where the cells the test substance tothe cells! react and 90% survive © GARD Main Stimulation eS... extraction RD = q Expose a fresh batch of quality Quality controt ofthe celts controlled cells
Slide 27: GARD: available assays
GARD: available assays GARD™ skin - 200 genes Skin sensitization testing GARD™ potency - 51 genes Skin sensitization potency testing according to GHS/CLP GARD™ air - 28 genes Respiratory sensitization testing GARD™ skin Medical Device - 200 genes Skin sensitization testing of medical devices
Slide 28: GARDskin: assay development
GARDskin: assay development Training Data set 20 Skin sensitizers 20 Non-sensitizers Prediction model Genomic biomarker signature Training data Biological system GARDskin prediction signature (GPS) 200 genes
Slide 29: GARDskin: assay development
GARDskin: assay development The 200 genes in GARDskin cover mechanistically relevant pathways associated 200 with skin sensitization genes @ Bs eratinocytes ®) Dendritic cells T-cells Key Event 1 Key Event 2 Key Event 3 Key Event 4 >Pro-inflammatory >Antigen recognition Covalent binding to signallin
Slide 30: GARDskin: assay development
GARDskin: assay development Full transparency: All assay development steps & the complete list of genes have been published. ‘Gene Symbol Gene Tide ‘Accession D ‘NanoString Probe 1D ABAT “Faminobutyrate aminotransferase 18 7993126 __NM_020686.5:268 ‘ABHDS abhydrolase domain containing 5 51099 807915
Slide 31: GARD: available assays
GARD: available assays GARD™ skin - 200 genes Skin sensitization testing GARD™ potency - 51 genes Skin sensitization potency testing according to GHS/CLP GARD™ air - 28 genes Respiratory sensitization testing GARD™ skin Medical Device - 200 genes Skin sensitization testing of medical devices
Slide 32: GARDpotency: assay development
GARDpotency: assay development Training Data set 22 Class 1A [Strong Sensitizers] 29 Class 1B (weak-sensitizers] Prediction model Genomic biomarker signature | Training data Biological system The GARD platform for potency assessment of skin sensitizing chemicals Kathrin S. Zeller? indy Forrery, Tim
Slide 33: GARDpotency: subclassification of skin sensitizers
GARDpotency: subclassification of skin sensitizers according to GHS/CLP GARDskin ~—- GARDpotency a 1A Sensitizer Test ; Non- Sensitizer
Slide 34: Validation study: GARDskin & GARDpotency
Validation study: GARDskin & GARDpotency _ Tech transfer ; | & Training : Validation Phase |: Validation Phase II | N=28 N= 28 Lead Laboratory: ! ' (SE) ' Hy blinded compounds blinded compounds N= 28 Ne ae Eurofins (EU blinded compounds ' \ ' | f ' i GARDskin | GARDpotency
Slide 35: Validation results: GARDskin
Validation results: GARDskin GARD Decision Value Sensitizers SenzaGen Burleson Eurofins Concordance: Within Laboratory Reproducibility: 82% 83% 89% Between Laboratory Reproducibility: | 92% Non-sensitizers WExperiment 1 MeExperiment 2 Experiment 3 Burleson Research Technologies Accuracy Sensitivity
Slide 36: Experiment 1 Experiment 2 Experiment 3
Experiment 1 Experiment 2 Experiment 3
Slide 37: GARDskin & GARDpotency: REACH registration
GARDskin & GARDpotency: REACH registration “The REACH Regulation allows the use of non-adopted in vitro methods in case they meet the EURL ECVAM criteria for entering pre-validation. For the GARD assay this Included in OECD criteria is met, as it is currently being validated.” Users Accepted by ECHA
Slide 38: GARD: available assays
GARD: available assays GARD™ skin - 200 genes Skin sensitization testing GARD™ potency - 51 genes Skin sensitization potency testing according to GHS/CLP GARD™ air - 28 genes Respiratory sensitization testing GARD™ skin Medical Device - 200 genes Skin sensitization testing of medical devices
Slide 39: Horizon 2020 funding acknowledgement
This project has recieved funding from the European Union’s Horizon 2020 Research and Innovation Programme under grant agreement No 756014.
Slide 40: GARDair: assay development
GARDair: assay development Training Data set GARDair prediction signature 10 respiratory sensitizers 20 Non-respiratory sensitizers (incl. skin sensitizers) } Prediction model { Genomic biomarker signature { Biological system wy. lane Forreryd A, Johansson H, Albrekt A-S, @ PLOS | Borrebaeck CAK, Li
Slide 41: GARDair: coverage of mechanistically relevant pathways
GARDair: coverage of mechanistically relevant pathways Keratinocytes Dendritic cells T - cells Key Event 1 Key Event 2 Kev Event 3 Key Event 4 Hapten formation Inflammatory >DC activation Activation/ > covalent binding mediators >DC maturation proliferation of to proteins >DC migration T-cells Adver
Slide 42: Accuracy Sensitivity Specificity
Non-respiratory sensitizers Respiratory sensitizers Skin sensitizers Accuracy 90% Sensitivity 100% Specificity 88%
Slide 43: GARD: available assays
GARD: available assays GARD™ skin - 200 genes Skin sensitization testing GARD™ potency - 51 genes Skin sensitization potency testing according to GHS/CLP GARD™ air - 28 genes Respiratory sensitization testing GARD™ skin Medical Device - 200 genes Skin sensitization testing of medical devices
Slide 44: GARDskin Medical Device: assay development
GARDskin Medical Device: assay development Requirements * Polar and non-polar extraction vehicles according to ISO 10993-12:2012 * Sensitive enough to detect low-level of skin sensitizers in extracts Development «Find oil that works with the assay * Demonstrate sensitivity of the assay, by perform a
Slide 45: GARDskin Medical Device: in house validation
GARDskin Medical Device: in house validation Extraction vehicles * Saline + Super Refined Olive Oil * Sesame oil, Ph Eur Material + Silicone and TPU spiked with five known skin sensitizers * Tubes (Silicone, TPU and PVC) Controls * Negative control, vehicle control * Positive control, vehicle spiked
Slide 46: GARDskin Medical Device: in house validation
GARDskin Medical Device: in house validation Summary of the GARDskin Medical Device results from the materials used in this study compared with LLNA [as listed in the CE STTF database} and Human potency classification (HP) for the chemicals (Basketter et al. 2014) Test material Chemical The spiked m
Slide 47: Why use GARD™?
Why use GARD™? Features and benefits
Slide 48: Compliance
Compliance Regulatory and quality standards OECD * GARDskin and GARDpotency included in OECD Test Guideline Program (TGP no. 4.106). REACH and CLP * Accepted by ECHA for REACH dossiers. ¢ REACH registration of chemicals * CLP 1A &1B potency classification GLP * GLP audit and approval expected in spr
Slide 49: Broad applicability
Broad applicability COSMETIC INGREDIENTS ACTIVE PHARMACEUTICAL & FORMULATIONS INGREDIENTS Essential oils Enzyme inhibitors Natural extracts NSAIDs Oil-based emollients Antiretrovirals, UV protection agents Antibiotics Hair dyes HOUSEHOLD AND PERSONAL CARE Over 400 items successfully INGREDIENTS & FO
Slide 50: Expertise in “difficult-to-test samples”
Expertise in “difficult-to-test samples” Cell membrane damaging substances e.g. surfactants Pre-/pro- haptens Low water-soluble substances Natural extracts Complex mixtures E-liquids Finished products
Slide 51: Number of substances
Active Pharmaceutical Ingredients 37 • Disinfectants Dyes • Flavoring agents/Frangrances • Intermediates • Number of substances 23 Organic solvents • 20 Paint and coating additives 17 17 Pesticides Preservatives • 13 10 Reactive agents • 8 Surfactants • 4 5 UV protection agents 3 3 2 Others Types of
Slide 52: GARDskin: pre-/pro-haptens, low water-soluble substances
GARDskin: pre-/pro-haptens, low water-soluble substances Compound Pre/pro High DPRA ARE-NRF2 _h-CLAT GARD hapten logP_(TG442C)_ (1G 4420) __(TG442E)_ {TGP 4.106) Challenge 2-Aminophenol 2-nitro-1,4-Phenylendiamine + Pre-/pro-haptens need to be activated * Pro-haptens more difficult to identify using
Slide 53: GARDskin: how about complex mixtures?
GARDskin: how about complex mixtures? Natural hair dyes Oil-based emollients E-liquids Tea extracts in oil Bee formulations Painting oils Detergents UVCBs Examples of successfully tested complex mixtures Challenge - Why difficult to test? * Complexity and uncertainty of the compositions * Solubility
Slide 54: Vehicle Test substance
Vehicle Test Vehicle Test Vehicle Test substance substance substance
Slide 55: Versatility
Versatility One sample several readouts Skin sensitizer? GARDskin Respiratory sensitizer? GARDair Test sample — Sensitizer GARDpotency Strong or weak? GHS/CLP 1A or 1B
Slide 56: Efficiency
Efficiency Time and cost * Quick and reliable results. * Short test time: 2 weeks. * Less expensive than animal testing. * “Pre-test” available for difficult-to-test samples Sample requirements * Only 0.5 g (solids) or 1 ml (liquids). * Potential to be adapted to significantly lower amounts if neede
Slide 57: Study timeline overview
Study timeline overview 4 Sample shipment Study plan Agreement Feasibility check (optional) Test time: 2 weeks Lab work Input finder Final report
Slide 58: Where to get my sample tested?
Where to get my sample tested? SenzaGen and its partners over the world Sweden SenzaGen AB License Labs Germany Eurofins BPT US Burleson Research Technologies MB Research Laboratories Distributors - Europe France Eurosafe GenEvolutioN PkKDerm Italy VitroScreen The Netherlands Charles River Laborator
Slide 59: Contact us
Contact us Andy Forreryd, PhD Key Account Manager & Scientific Liaison andy.forreryd(@senzagen.com +46 734331777 Home Services Science About «Partners Investors News and Events Joshua Schmidt, PhD Business Development Director Americas Joshua.schmidt(@senzagen.com +1 651.440.5691 contact a