AI drug repurposing ♻️ startups: should you date your ex?
"The most fruitful basis of the discovery of a new drug is to start with an old drug" By Nobel Laureate and renowned pharmacologist James Black
Hi 👋 everyone and welcome back to another edition of MetaphysicalCells on AI Drug Repurposing.
Drug repositioning or drug repurposing ♻️ is an approach to accelerate the drug discovery process through the identification of a novel clinical use for an existing drug approved for a different indication.
According to MarketWatch, the global drug repurposing market size is anticipated to reach $ 30.1 billion by 2028, up from $ 24.5 billion in 2021, reflecting a CAGR of 2.9% over the 2022-2028 period.
On October 11, 2023, Cosmos Health Inc. a fully reporting Pharmaceutical Holding Company based in US, that specialises through its subsidiaries in the wholesale of pharmaceutical, nutraceutical and specialised root extracts products throughout Europe, announced that it has entered into a purchase agreement to acquire Cloudscreen®, an AI powered platform developed by Cloudpharm for drug repurposing. The total investment is expected to exceed $5M through 2029.
The drug repurposing platform Cloudscreen has been developed by Cloudpharm 🇬🇷 that is one of the 5 Greek startups entering the final phase of the Start4Health Challenge by the Pfizer Digital Innovation Centre in Thessaloniki, Greece, that focuses on the Digital Health environment. In particular, Cloudscreen (developed under the call of EU RESEARCH-CREATE-INNOVATE project code: T2EDK-03153) integrates both 1D and 3D data types into its AI platform, which thoroughly analyses the druggable proteome and variome, offering accurate predictions for repurposing of existing drugs toward new indications. The platform is accompanied by in vitro validation for both toxicity and effectiveness and is currently being developed in collaboration with the Biomarker Discovery & Translational Research Laboratory at the National Hellenic research Foundation and is set to hit the first milestone IP for a repurposed API by Q3 of 2023.
Cloudpharm is also involved in the following programs:
➡️ The EU-funded ENVIROMED project that aims at reducing the overall environmental impact of pharmaceuticals by reducing the environmental footprint of the manufacturing processes. Cloudpharm will be in charge of developing novel in-silico tools for “green-by-design” pharmaceuticals, along with predicting the ecotoxicity impact of pharmaceuticals using in silico methods.
➡️ A next Generation Food Supplements project, the NextGFS, that aims to implement a new methodological approach which will guide the design of a new generation of nutritional supplements. Specific objectives for the project are: the development of a platform for designing nutritional supplements through computational tools; the pilot production of a new dietary supplement with anti-inflammatory and antioxidant properties, aiming primarily at patients with osteoarthritis (OA) and the implementation of a clinical trial using the new dietary supplement on volunteers with OA. And
➡️ The Next Generation Training on Intelligent Greenhouses, the NEGHTRA project, that is a specialised training EU project that addresses knowledge transfer in precision agriculture based on specific needs and challenges, identified from a comprehensive need analysis. It aims to provide innovative training on smart greenhouse technologies coupled with selected optimum matchings of technologies/cultivations with regions including conditions for economic and environmental sustainability.
Another Greek 🏺 drug repurposing startup is Purposeful focusing on discovering alternative drug effects for rare diseases. Purposeful is using state of the art Modelling and Simulation (M&S) approaches to guide the translation from preclinical to clinical and the clinical development of its candidates in an integrated quantitative framework which has been termed as MID3. Purposeful has so far partnered ➡️ with FRAXA and FRAXA-DVI to identify drug repurposing candidates against Fragile X Syndrome, ➡️ with BIOEMTECH to identify drug repurposing candidates against Glioblastoma and ➡️ with Leo Pharma Ltd to discover and validate drugs to treat Inflammatory and Dermatology indications. Purposeful has raised money from Metavallon VC and PharOS a privately owned pharmaceutical company.
QR Genetics in Israel 🇮🇱 is also provider of personalised drug repurposing services (shelf products, developed drugs and patented drugs) for rare diseases, by using patients' genetic data to confirm disease mutation with molecular and clinical status. QRGenetics developed a powerful AI-supported drug discovery engine that utilises exclusive spatial analytics and proprietary omics datasets, to rapidly produce novel drug and target combinations that are effective for multiple indication.
Always in Israel ✡️ we have 9xchange, a biopharma marketplace created for innovators to match, buy and sell drug assets, that has joined forces with BenevolentAI, a leading AI-enabled drug discovery and development company, to leverage BenevolentAI's technology to support decision-making related to indication expansion and drug repurposing for assets within the 9xchange platform.
Specifically, the 9xchange marketplace is a curated, members-only biopharma marketplace that matches asset buyers and sellers and removes friction by anonymising, automating and enhancing many steps required to get a deal done. The power of BenevolentAI’s AI technologies will help 9xchange marketplace members (including biotech and pharma partners, VCs, investors, and academic institutions) to investigate drug repurposing hypotheses on potential new indications for a given asset and ultimately transact based on these findings.
Another company in Israel combining over 1 trillion data points across the clinical and pharmacological domains to optimise clinical development is QuantHealth, with a unique AI-based clinical trial platform (based on a vast dataset of 350M patients and over 700K therapeutics) that allows pharmas to simulate their upcoming clinical trials in a cloud environment, and test thousands of protocol variations to discover the trial design that will maximise the likelihood of success. In fact, their in-silico platform can be used to support trial planning, as well as indication selection, drug repurposing and in-licensing asset evaluation.
QuantHealth has a Large Healthcare Model (LHM) that learns deep patient-drug interactions that drive high-resolution outcome models with 86% endpoint accuracy, and continually adapts as new research comes forward, and can be further enhanced with internal trial data and proprietary research. QuantHealth emerged from stealth 🥷 last year, along with a $2.6M seed, and has raised two weeks ago $15M in Series A funding.
Nebion AG in Switzerland has been acquired by Immunai (with an ambitious goals of mapping the entire immune system) but its main product GENEVESTIGATOR that curates public and private transcriptomic data continues to be developed and commercialised. GENEVESTIGATOR, that has been over 10+ years in development, has the following key applications for curated gene expression data:
Effective discovery through compendium-wide analysis
Unbiased data-driven decision making
Straightforward prioritisation and validation of targets and biomarkers
Innovative drug repurposing and repositioning
Convenient identification of relevant experimental model systems
Reliable prediction of side effects, and
Robust ML through clean data.
Oncocross in South Korea 🇰🇷 utilises AI to analyse gene expression patterns (transcriptome level of cells, human biopsy and blood samples), allowing researchers to develop biomarkers, repurpose drugs and find novel drugs. The company is repurposing drugs by mining diseases against given drugs and mining drugs against diseases to treat rare, intractable diseases and diseases without treatments. By using their technology (RAPTOR AI, ONCO-RAPTOR AI and ONCOfind AI), they have identified OC-501 and OC-504, both marketed drugs, as a possible drug combination for Sarcopenia. ➡️ In March 2013, they announced the completion of the phase 1 global clinical trial of OC514 (a drug-drug combination containing 2 active pharmaceutical ingredients for cancer cachexia), a treatment targeting sarcopenia and other rare muscular diseases. Oncocross has raised about $19M so far.
Healx in UK 🇬🇧 🎡 uses AI to match existing drugs with rare diseases. Their AI platform uses NL processing to extract disease knowledge from published sources and to complement biomedical databases and proprietary curated data. ➡️ In 2021, Healx begun a clinical trial on a repositioned drug (a non-steroidal anti-inflammatory drug) for Fragile X Syndrome. ➡️ On February 08, 2022, Ovid Therapeutics and Healx announced that Healx has secured from Ovid an exclusive option to license rights to develop and commercialise gaboxadol, as part of a potential combination therapy (with HLX-0201, that has been identified by the company’s proprietary AI platform, Healnet) for Fragile X syndrome, as well as a treatment for other indications. ➡️ On September 07, 2023, the FDA has awarded orphan drug designation to Healx’s HLX-1502 for the treatment of neurofibromatosis type 1 (NF1), a genetic condition that causes tumours to grow along the nerves. Healx has raised so far $66M.
Pharnext in France is an AI-screening and repositioning company of known drugs in unrelated indications at new and lower doses, in order to identify synergistic combinations of repositioned drugs for diseases with high unmet medical needs. These new therapeutic entities are called PLEODRUG™. At the beginning on this year, Pharnext walked away from its work developing a treatment for Alzheimer’s with PXT864, and decided to focus on a PXT3003 phase 3 trial to treat Charcot-Marie-Tooth disease type 1A. Pharnext has raised so far $157.6M.
Standigm in South Korea 🇰🇷 has a proprietary AI platform to generate commercially valuable drug pipelines. Pursuing full-stack, AI-driven industrialising drug discovery. Standigm has achieved the automation of molecular design workflow through DarkMolFactory, and the automation effort has been expanding to the whole drug discovery process on the basis of Standigm AI platforms, including: ➡️ Standigm ASK for target discovery, ➡️ Standigm BEST for lead design and ➡️ Standigm Insight for drug repurposing.
On August 10, 2013, Standigm and Nashville Biosciences, a wholly-owned subsidiary of Vanderbilt University Medical Centre, announced a collaboration on early drug discovery research in order to use Nashville Bioscience's extensive, de-identified genomic and clinical data to build new AI models for drug discovery. Standigm has raised so far $71.2M.
Computational drug repurposing based on electronic health records: a scoping review
Electronic health records (EHRs) datasets can facilitate the generation and validation of drug repurposing.
This review summarised the contribution of EHR used in drug repurposing as well as revealed that the utilisation is hindered by the validation, accessibility, and understanding of EHRs.
Conclusion: EHRs are an invaluable source of large-scale clinical data capable of simulating drug repurposing strategies in an uncontrolled, real-world environment as opposed to the controlled environment of clinical trials, which is one of the biggest benefits of drug repurposing.
However, the biggest issue of EHR-based methods is the inaccessibility of any data involved. Thus only primary academic medical centers can fully leverage any advantages of utilising EHR-based methods.
Lastly, EHR provides large longitudinal medical records in clinical settings, which improves the applicability/reliability of the predictive models. However, temporal information is seldom incorporated into the experiment. The data model that incorporates temporal information will reduce the bias in the validation. Another issue of EHR-based methods is that they cannot reveal causality. So more mechanisms need to be studied to better understand the toxicity and tolerability of drugs in humans. Then the leverage of these genetic information can provide another potential to improve the performance and interpretability of drug repurposing. For example, both the eMERGE network and the All of US Initiative can link EHR to genetic information for multisite studies and increase the number of features (e.g., phenotypes and genotypes) to further promote the development of drug repurposing.
Always in South Korea 🇰🇷 Syntekabio is a global partner for repurposing drugs, using DeepMatcher, an AI drug development platform that combines new drug development technology and big data technology (such as biomarkers for individual cancer drug screening and disease susceptibility by immunotyping, pharmacogenomic typing, and predicting multi-omics information-based drug adverse effects, etc).
➡️ On July 31, 2023, Syntekabio has signed a MOU with Metaclipse Therapeutics, a USA based biotechnology company developing immunotherapies for cancer and vaccines for infectious disease, for joint research on development of a personalised immunotherapy for cancer. ➡️ Again on August 31, 2023, Syntekabio announced that it has signed another MOU this time with POLARISqb, a drug discovery company that utilises the optimisation power of quantum computing, in order to combine the companies’ vanguard drug research platforms. Syntekabio has raised a total funding of $11.1M over 1 round.
Lantern Pharma in US 🇺🇸 🏉 is a uniquely positioned life science company pushing innovations in pharmaceutical biotechnology and drug-companion diagnostic co-development through AI, ML and genomics. By utilising AI to analyse genetic signals and molecular markers for patient response to drugs (small molecules), they are allowing researchers to find clinical uses for validated cancer treatments whose development has been discontinued.
➡️ On September 26, 2023, Lantern has dosed its first patient in a phase I trial of its drug LP-184 to treat advanced solid tumours. A total of 35 patients will receive an LP-184 infusion on a 21-day cycle, for at least two cycles. ➡️ On October 3, 2023, Lantern announced that the in vivo data highlighting the enhanced efficacy of its drug candidate LP-184 in glioblastoma were published in Clinical Cancer Research. LP-184 is a small molecule with favourable CNS penetration, that utilises the mechanism of action known as synthetic lethality, to exploit common vulnerabilities in solid tumour and CNS cancers with DNA damage repair (DDR) deficiencies. In addition, Lantern’s AI platform, RADR, has highlighted overlapping gene dependency profiles between glioblastoma tumourigenesis and sensitivity to LP-184, such as EGFR activation pathways. Lantern Pharma went public in 2020, and had post-IPO fundraising, all totalling in $95 million.
Delta 4 in Austria 🇦🇹 uses AI to conduct in-silico screening prior to experimental screening. Delta 4 leverages on a proprietary computational analytics platform, combined with straightforward biomedical testing and clinical validation of candidate drugs. Its first therapeutic area is Focal Segmental Glomerulosclerosis (FSGS), a devastating renal disease associated with nephrotic syndrome that carries a high risk of kidney disease progression with high and urgent medical need. ➡️ On June 19, 2023, Delta 4 has announced that has identified clopidogrel, an anti-platelet drug, as a promising therapeutic candidate for FSGS. Clopidogrel demonstrated noteworthy potential in improving key FSGS outcome parameters and reducing histopathological damage, as evidenced in the Adriamycin FSGS mouse model.
BioXcel Therapeutics in US 🇺🇸 🏉 was spun out of BioXcel Corporation in March 2017 to leverage AI and ML for drug re-innovation, and now utilises AI to find applications for existing approved drugs or clinically validated candidates in immuno-oncology, neuroscience and rare diseases. BioXcel’s leading product is IGALMI™, a sublingual film used to treat agitation associated with schizophrenia or bipolar disorder. It was approved in April 2022 and is currently on the market. ➡️ On October 10, 2023, BioXcel Therapeutics announced positive overall survival data from its phase 2 trial of BXCL701, an investigational oral innate immune activator in combination with KEYTRUDA (pembrolizumab) in patients with small cell neuroendocrine prostate cancer. BioXcel Therapeutics has raised so far $140M.
Acurastem in US 🥞 utilises AI to analyse data from sources including patient stem cells. AcuraStem’s iNeuroRx® technology uses living motor neurons (muscle-controlling nerve cells from many amyotrophic lateral sclerosis or ALS patients) and advanced informatics as a tool for the identification and development of novel treatments. So far, they have identified 37 compounds that improve nerve cell survival and might help treat Charcot-Marie-Tooth disease type 2A (a group of inherited disorders that cause nerve damage). ➡️ On September 26, 2023, Takeda inked a development and commercialisation agreement with AcuraStem to target ALS and other neurodegenerative diseases, of approximately $580M. In return, Takeda will gain access to AcuraStem’s PIKFYVE program, which includes AS-202, an investigational antisense oligonucleotide that is an emerging therapeutic target in ALS. AcuraStem has raised a total of $5M in funding over 3 rounds.
Until next time 🍂,
Reference
Top drug repurposing companies
PS
➡️ A Startup for Accessible AI for every Drug Hunter
Deep Mirror in UK is enabling every Life Sciences laboratory with intuitive and secure scientific AI powered software to accelerate discovery, unlock creativity and elevate health for all. They are offering AI tools for Bioprocess Optimisation and Image Analysis. For example, KymoButler (a cloud-app for the analysis of particle traces in kymographs/space–time plots) helps hundreds of researchers across the world track and analyse molecule dynamics in brain cell imaging (neurons).
Deep Mirror was founded by a team of researchers at the University of Cambridge in 2019 and is supported by Santader, NVIDIA, AWS Activate, Microsoft for Startups Founders Hub, MaRS, Royal academy of engineering and Innovate UK. Co-founder and Product Lead is Dr. Andrea Dimitracopoulos.
➡️ Building Generative AI Applications for Enterprise Demands
An NVIDIA Webinar will be presented by Amanda Saunders and Donald Chouinard on Thursday, October 26, 2023 (Time: 9:00-10:00 a.m. PT | 6:00-7:00 p.m. CET) and attendees will learn (Register):
How to build generative AI applications, from data acquisition to model deployment, including
What AI teams need to know about data acquisition
What data curation does and why it’s necessary
Key concepts behind model training
Sising considerations for model training
Varying levels of model customisation, top use cases for each customisation technique, and how a business should evaluate and determine its solution.
Love the newsletter’s name!!!
Love your newsletter’s name!!!