NON CONFIDENTIAL OVERVIEWspektronsystems.com/wp-content/uploads/2018/07/2018-Non... · 2018. 7....
Transcript of NON CONFIDENTIAL OVERVIEWspektronsystems.com/wp-content/uploads/2018/07/2018-Non... · 2018. 7....
A Machine Learning / Artificial Intelligence Drug Design Company
NON CONFIDENTIAL OVERVIEW
Spektron Systems is disruptively…
Converting drug discovery into directed drug design and drug engineering.
Curtailing by 75% the conventionally required amounts of time and money to develop medicinal molecules for delivery to the market.
Dramatically reducing the risks of clinical trial and market failure.
The Spektron Advantage: Q-MAP™: AI/ML methods applied to early stage drug design.
2
Drug Discovery is No Longer Viable in Big
Pharma• Average cost of introducing a new drug has
reached $2.6 billion*
• Pre-clinical trial cost alone is $100M+
• Glacially slow R&D: 8-11+ years per drug
• The low-hanging fruit has been plucked
• Investment returns are low and dropping
• Financiers, not scientists, now drive most R&D decisions; too often they back the wrong picks
• Huge, critical patient needs going unmet
*Total capitalized costs per new drug approval (in 2013 dollars) according to ”Journal of Health Economics,” May 2016
The Problem
VALUE OF SPEKTRON’S APPROACH
in vitro & Animal Testing Human
Q-MAP™ models end-to-end in silico prior to the synthesis of any molecules
Spektron’s Q-MAP™ platform reduces the time, effort, and risks involved in developing a viable pre-clinical drug candidate
CUSTOMARY INDUSTRY PRACTICE
Biology Target ID Target Validation
Lead Discovery
Lead Optimization
Preclinical Testing
Phase I Clinical
Phase II Clinical
Phase III Clinical
Q-MAP™ in silico
in vitrovalidation
Preclinical Testing
Chief Operating OfficerTim Dockins is a PhD candidate in Computer Science with an emphasis in Machine Learning. He has industry experience in program management and software development dealing with complex data.
Chief Strategy OfficerRobert Cain is an MBA with background in investment banking with JP Morgan and pharmaceuticals with Sandos. Mr. Cain has experience in strategy, finance and international business development for companies ranging from startups to Fortune 500.
President / Chief Medical OfficerLed by founder David Wolf, MD, recognized expert in taking innovation to operations; pioneer in ultrasound imaging and tissue engineering; professor of biomedical engineering and aeronautics; physician, astronaut; White House advisor.
The Management Team
5
Senior Vice President of Drug Discovery W. Ken Fang, Ph.D. is a veteran in the pharmaceutical industry with over 17 years at Allergan leading early stage drug discovery efforts. With broad experience in developing and testing new chemical entities, Dr. Fang leads our optimization and physical validation strategy.
THE CORE TEAM
Bob D’Agostino, MSc.
DirectorMachine Learning
Max Sharifi, Ph.D.
Principal ComputationalChemist
Ashley Meyer, MSc.
Principal InformaticsScientist
Asawari Kharkar, Ph.D.Sr. Informatics Scientist
Ashley Flory, M.Sc.Sr. Informatics Scientist
Jessica HeidrichInformatics Scientist
Iva Stoyanova-Slavova, Ph.D.FDA – NCTR – ORISE Fellow assigned to CRADA
Shayne Cox-Gad, Ph.D. DABTKey Advisor – FDA-IND Process
EXTENDED TEAM
! ! ! !Executive Management
! ! ! !AI/ML & Platform Development
! ! ! !Informatics / Data Science
! ! !Drug Design / Chemistry
! ! ! ! !Executive Management
! ! ! ! ! !AI/ML & Platform Development
! ! ! ! ! !Informatics / Data Science
! ! !Administrative
! ! ! ! !Drug Program 1
! ! !Administrative
Current Staffing
! Spektron FTE! Shared Resource (Spektron FTE)! Contractor/Consultant
! ! ! !Drug Design / Chemistry
Post-Equity Buildup
6 FTE5 Contractors/Consultants
15 FTE14 Contractors/Consultants
The Business Model
Multiple Paths to Revenue & Profit• Sell clinical trial-ready investigational new drugs
(IND) to other pharmaceutical companies• Collect upfront payments, progress payments &
ongoing royalties
• Provide toxicity screening services
• Develop drugs initiated by JV partners in exchange for ownership share
• Fix “broken” or withdrawn drugs
Q-MAP technology/trade secrets remain in-house and proprietary to Spektron Systems
8
Target Customers
New drug molecules $
Upfront &Ongoing
Payments
SPEKTRON MOLECULE DEAL STRUCTURE
$10MEarly stageBefore preclinical testing
1) Licensed at 2) Initial Payments 3) Milestone Payments
$20M $370M
PreclinicalInvestigational New Drug
$30M $370M
Phase 1aHuman Safety $100M $500M
Early Stage Preclinical Testing Phase I Clinical Phase II Clinical Phase III Clinical
COMPARABLE DEALS
Early Stage Preclinical Testing Phase I Clinical Phase II Clinical Phase III Clinical
$14.2MBicycle Therapeutics & Bioverativ $410M
$300MIFM Therapeutics & Bristol Meyers $1B
$150MNektar Therapeutics &Eli Lilly $1B
2017 – Hemophilia / Sickle Cell Anemia
2017 – Oncology (STING)
2017 – Autoimmune (Tregs)
Psychiatric
Schizophrenia
Depression
Anxiolytics
Drug PipelinesCardiovascular
Bipolar Disorder
Antiarrythmic
Hypercholestoreolemia
Hypertension
Analgesics
Platform Development used Schizophrenia as a target disease state with lower safety and tolerability issues.
Data in other Psychiatric areas was available and overlapping suggesting the possibility of repurposing or separate programs.
Other drug classes are being explored.
Hypoglycemics
Antiemetics
Neurology
Respiratory
OPERATING STRUCTURE
Subsidiaries hold compound and target data, IP, know how, IND, etc.These can be sold separately
Investors
Funding/services flow from Investors to Spektron
Distributions and capital gains flow to Investors
Spektron Systems, Inc.
Compound IP, know how, data, etc. flow to Sub (funded by Parent)
Sub 1 LLC Sub 2 LLC
Compound gains/losses flow to Parent
Sub 3 LLC
COMPETITION
Q-MAP™ models end-to-end in silico prior to the synthesis of any molecules
MOA asserted through
existing human clinical
endpoint data
Some companies focus on aselect portion of the traditionaldrug discovery process
Some companies focus on certain drug classes outside of Spektron’s arena
Biology Target IDTarget
Validation
Lead
Discovery
Lead
Optimization
Preclinical
Testing
Phase I
Clinical
Phase II
Clinical
Phase III
Clinical
Q-MAP™
in silico
in vitro
validation
Preclinical
Testing
IP STRATEGY
• Protect core platform technology with:• Trade secrets• Provisional patents• PCT filings
• Protect drug molecules with full patent protection on molecules and Markush structures
• Protect other important IP through trademarks and copyrights
Introducing Q-MAP™The Quantum-Molecular Activity Predictor
An extension of research from the FDA’s National Center for Toxicological Research through licensing and a Collaborative Research and Development Agreement
Combines modern methods of molecular fingerprinting to reveal key characteristics involved in biological activity.
A key component in Spektron’s platform for new drug discovery
PUBLIC HEALTH SERVICE
COOPERATIVE RESEARCH AND DEVELOPMENT AGREEMENT
PHS CRADA Agreement Ref. No. 2017-0027-CRD A1 MAY 15, 2018 Page 18 of 25 Confidential Revised August 1, 2012
SUMMARY PAGE
EITHER PARTY MAY, WITHOUT FURTHER CONSULTATION OR PERMISSION, RELEASE THIS SUMMARY PAGE TO THE PUBLIC.
TITLE OF CRADA: Advancing SDAR modeling for predictions of drug biological activity
FDA Component: National Center for Toxicological Research
FDA Principal Investigator: William Mattes, Ph.D.
Collaborator: Spektron Systems, LLC.
Collaborator Principal Investigator(s): Tim Dockins
TERM OF CRADA: Three (3) years from the Effective Date
ABSTRACT OF THE RESEARCH PLAN:
Spectral Data-Activity Relationships (SDARs) is an FDA-patented molecular modeling technology that can generate accurate chemical and biological predictive data. The technology has been demonstrated to discern important chemical features for a variety of molecules as related to biological effects and toxicity. The SDAR method utilizes NMR chemical shifts related to, but independent of, specific chemical structure. SDAR molecular fingerprints are used in conjunction with their associated biological endpoints and modeled using various pattern recognition methods to create predictive models. These mathematical associations between the fingerprint characteristics produce predictions for unknowns. Although SDAR predictive models have been demonstrated for several relevant toxicities, the limitations of the method relative to small (<70 molecules) and large (>500 molecules) datasets have not been determined. The collaborators will seek to 1) produce a robust, user-friendly SDAR modeling software; 2) identify and explore limitations of the method; and 3) automate/replace expert user functions. Collaboratively created in silico models will be available to Collaborator and FDA as shared work products.
COLLABORATION
Spektron Systems & FDA Cooperative Research and Development Agreement (CRADA)
Spektron Systems maintains a CollaborativeResearch And Development Agreement with theFDA – National Center for Toxicological Research
This collaboration provides bi-directional accessto data, methodology, and expertise enabling akey technology transfer into commercialization
Proprietary & Confidential
Molecular Design Generator
AI/ML Modeling
Expertise Expertise
Molecular Design Evaluator
AI/ML Modeling
Public and PrivateTraining Data
producesQ-MAP™ Qualified
Lead Molecules
Q-MAP™ - AI-assisted Drug Design
TECHNOLOGY VIABILITY DEMONSTRATIONS
Benchmarking
Toxicity
Models for all
Major Organ
Systems
Corroboration
with Published
Research
Engineering
Novel Molecules
PREDICTION DASHBOARDS000001 S000002 S000003 S000004 S000005 S000006 S000007
Tar
gets
Efficacy OK UNDF OK OK FAIL OK OK
Relative Ranking OK UNDF OK OK FAIL OK OK
Pharmacological Class OK OK OK OK FAIL OK OK
Animal Models OK FAIL OK UNDF FAIL UNDF OK
Ant
itar
gets
Blood Dyscrasias OK OK OK OK OK OK OK
Cardiotoxicity OK OK OK FAIL OK UNDF OK
Hepatotoxicity OK OK UNDF OK OK OK UNDF
Nephrotoxicity OK OK OK OK OK OK OK
Neurotoxicity OK UNDF OK UNDF OK UNDF UNDF
Phospholipidosis UNDF OK OK FAIL OK OK OK
Mutagenicity FAIL OK OK OK OK OK OK
Carcinogenicity OK OK OK OK OK OK OK
Pro
file
Phase 1 P450 Inhibition NON INHIBIT NON NON NON NON NON
Receptor Binding TARGET TARGET TARGET OFF-TGT OFF-TGT TARGET TARGET
Receptor Action TARGET TARGET TARGET TARGET OFF-TRGT TARGET TARGET
Q-MAP’s virtual screening toolbox provides a range of predictive models employed to predict biological and in-vitro activity.
Predictions are combined into target, off-target, and profile to create a composite score.
19
Tox21 Data ChallengeThe 2014 Tox21 data challenge is designed to help scientists understand the potential of the chemicals and compounds being tested through the Toxicology in the 21st Century initiative to disrupt biological pathways in ways that may result in toxic effects.
The goal of the challenge is to "crowdsource" data analysis by independent researchers to reveal how well they can predict compounds' interference in biochemical pathways using only chemical structure data. The computational models produced from the Challenge could become decision-making tools for government agencies in determining which environmental chemicals and drugs are of the greatest potential concern to human health.
Spektron operates at the top of the field of challengers for a balanced metric and in the crucial metrics for screening for toxicity.
Specificity
hERG InhibitionhERG blocking has been one of the leading causes for withdrawal from the market of drugs approved by the FDA and is implicated as leading to cardiac arrhythmia.
Spektron has developed in silico models for predicting
drug-induced hERG Inhibition achieving high accuracy for identifying
active molecules.
Stoyanova-Slavova, Iva B., Svetoslav H. Slavov, Dan A. Buzatu, Richard D. Beger, and Jon G. Wilkes. 2017. “3D-SDAR Modeling of HERG Potassium Channel Affinity: A Case Study in Model Design and Toxicophore Identification.” Journal of Molecular Graphics and Modelling 72. Elsevier Inc.: 246–55
Q-MAP™ corroborates toxicophore identification using
the same methodology as developed at the FDA – National Center for Toxicological Research
Psychiatric Schizophrenia
Antipsychotics
Novel Molecules Q-MAP Qualified Leads Q-MAP Optimized Leads Validated Leads Pre-Clinical
> 15,000Novel Molecules
generated by Q-MAP™
~100Leads qualified by
Virtual Screening and Multi-Parameter
Optimization
~60Leads optimized for
patents and synthesis
-Small scale synthesis
and validation for efficacy and pre-IND
readiness
-In pre-clinical testing
targeting FDA Investigational New
Drug Process
Q-MAP™ has generated over 15,000 novel molecules targeting improved atypical antipsychotics in a first iteration.
Next Step: revalidate Q-MAP™ leads through Version 2 of the platform and prepare for small-scale synthesis.
FINANCING OVERVIEWSpektron is
presently seeking Series A equity
financing with follow-on Rounds totaling
$40M
Concurrently, Spektron is filling
$2.2M in Convertible Debt via Friends & Family and Angels for a total of $5M
CDNs
To date, all CDN financing has come
from investors outside of Arkansas
Proprietary & Confidential
$2.8MConvertible Debt
$2.2MConvertible Debt
$10MSeries A Equity
2016 Mid 2018 Late 2018
Terms:1.5x on Equity36mo MaturitySEC Form D Filing
Terms:1.5x on Equity36mo MaturitySEC Form D Filing
Terms:Institutional and Venture Capital
EQUITY RAISE HIGHLIGHTS
$10M+ $30M $370M
Current Funding Round Upfront Payment Milestone Payments
Spektron Systems is seeking 24mos financing to carry its first drug target through preclinical trials to submit to the FDA as an IND
Estimated proceeds from first program
Used to expand Spektron’s platform and complete the first program milestones
Proprietary & Confidential
The FinancialsInvestment Thesis
• Deploy $40 million to complete tech platform and develop first 3 drugs
• License each drug to Big Pharma companies to manage FDA clinical trials
• Collect payments as drugs move toward and beyond commercial launch: • Up-front payments: est. $30M per drug
• Milestone payments: est. $200M to $1B+ per drug
• Ongoing royalties: est. 5% - 20% of sales revenue
• Aim for USD multi-billion sale or IPO after 2nd
successful licensing agreement
25
Pro Forma Financial Projections$37m invested in 3-drugs, 2nd & 3rd drugs succeed
2018 2019 2020 2021 2022 2023
$10M $14M $13.2M
Drug 1Sale
Drug 2Sale
Drug 3Sale
Anticipated Financing & Liquidity Events
Use of Funds
Liquidity
Proprietary & Confidential
FUNDING / VALUATION COMPARABLE
Total FundingLatest Round
ValuationSeed Round Average
Debt Financing
AverageSeries A Average Series B Average Series C Average
twoXAR $14.3M $66.67M
Spektron is on target with seed round and is expected to close a series A in the same range as
these comparable companies..
Recursion Pharmaceuticals $84.3M $400M
Numerate $17.4M Unknown
NuMedii $5.5M $23.33M
Atomwise $51.6M $300M
Verge Genomics $4M $26.67M
Average $29M $163M $2M $2.5M $10M $33M $8M
Spektron’s Valuation Milestones
Year 1 Year 2 Year 3 Year 4 Year 5
~ $100 M
Q-MAP fully operational
1st Q-MAP Optimized Lead
Selected
~ $150 M
1st lead molecule validated
2nd Q-MAP Optimized Lead
Selected
~ $300 M
1st pre-clinical candidate
2nd lead molecule validated
3rd Q-MAP Optimized Lead
Selected
~ $500 M
1st Investigational New Drug
2nd pre-clinical candidate
3rd lead molecule validated
~ $1.2 B
1st drug passes Phase 1 CT
2nd Investigational New Drug
3rd pre-clinical candidate
Investor Exit Strategies
Shareholder Buyout options
with proceeds from each Drug Program
(any 2 out of 3 = Grand Slam)
Mergers / Acquisitions / IPO
Positive Business Environment
Strong Market for medicinal molecules
Strong merger/acquisition and liquidity options
Big Pharma Exiting Discovery
2025202420232022
@SpektronSystems
www. linkedin.com/company/spektron-systems-llc.
www.spektronsystems.com
Contacts:
Spektron Systems400 W. Capitol Ave., 17th fl.Little Rock, AR 72201
Tim [email protected]+1.469.337.0788
David [email protected]+1.832.372.3834
Robert [email protected]+1.310.663.8811