Using artificial intelligence to accelerate cell and gene therapies
"We are not made of drugs, we are made of cells". By Cade Hildreth,the Founder of BioInformant.com, the world's largest publisher of stem cell industry news
Cell and Gene Therapy
According to an article published last year by McKinsey (“How AI can accelerate R&D for cell and gene therapies”), within oncology the cell therapy 🧫 modality is expected to become the third-largest segment across all therapies (behind antibodies and small molecules) by 2030, with 35% CAGR in sales over 2021–30 (and 437 studies so far in phase I-III versus 1262 studies for small molecules and 747 studies for antibodies).
On the other hand, gene- and RNA-based therapies 🧬 are unlikely to play a major role in this short-medium term, even though there are currently more than one hundred such therapies in phase I–III studies.
Moreover, always according to the same article, all cell and gene therapies still remain challenging 🤷♂️
for the complexity and heterogeneity of the solution space,
for manufacturing 🔩 and supply chain challenges (especially for personalised therapies), and
for the difficulty of appropriately matching therapies to the suitable patients.
On top of that, AI applications are only starting to explore how to apply their potential to cell and gene therapies.
Very briefly, the cell therapy modality includes
stem cell- and non–stem cell-based therapies,
unicellular and multicellular therapies,
therapies with different immunophenotypic profiles. For example, immunotherapy is a form of treatment that uses the cells of our immune system to eliminate cancer cells, like the
Tumor-Infiltrating Lymphocyte (TIL) Therapy,
Engineered T Cell Receptor (TCR) Therapy,
Chimeric Antigen Receptor (CAR) T Cell Therapy, and the
Natural Killer (NK) Cell Therapy;
and cell therapies with different isolation techniques, mechanisms of action and regulatory levels.
When it comes to the gene therapy modality (with both viral and non viral vectors) we have
DNA based therapies such as
CRISPR gene editing, that enables a correction of a particular genetic defect,
Prosthetic gene therapy, and
RNA based based therapies (mRNA, antisense oligonucleotides, shRNA, siRNA, miRNA).
Let's see now the latest news 🗞️ on cell and gene therapy startups using AI/ML tools.
Cell therapy startups using AI
👉 Twist Bioscience founded in 2013 by Emily Leproust, Bill Banyai, and Bill Peck, is developing a proprietary semiconductor-based synthetic DNA manufacturing process featuring a high-throughput silicon platform that allows to miniaturise the chemistry necessary for DNA synthesis. In 2021, Twist Bioscience incorporated the MOLCURE AI technology to augment therapeutic antibody discovery. MOLCURE's patented and validated AI platform is based on evolutionary molecular engineering, next-generation sequencing (NGS), and laboratory automation technologies.
Last month, Twist Bioscience announced the launch of Twist TCR libraries and Twist CAR libraries. These new libraries provide researchers with the ability to rapidly create custom libraries that can enable the identification and development of next generation cell therapies. Twist Bioscience has raised a total of $503.1M in funding over 9 rounds.
“Twist CAR and TCR Libraries allow researchers and drug developers to fine tune these receptors to address the challenges of cell therapies, including off-target toxicity, enabling rapid advancement toward the clinic." said Emily M. Leproust
👉 Outpace Bio founded in 2020 is pushing the boundaries of biology to create tailored cell therapy solutions through protein design and custom programming. In particular, by using Rosetta (an advanced protein design software) to custom design new biological functions combined with their cell engineering expertise they try to solve holy grail problems in cell and gene therapy.
On May 18, 2023, Outpace Bio presented advancements to its OutSmart platform during a poster presentation at the American Society of Gene & Cell Therapy (ASGCT) 26th Annual meeting held from May 16-20, 2023 in Los Angeles. The OutSmart platform provides a way to regulate cytokine signalling in the tumour microenvironment (that is a major barrier to efficacy for cell therapies in solid tumours) by coordinating a productive immune response directly in the tumour microenvironment (TME) from both the engineered cell therapy as well as endogenous immune cells. Outpace Bio has raised a total funding of $30M over 1 round.
👉 Modulus Therapeutics founded in 2020, envisions a future where the design of cell therapies is guided by ML via their Convergent Design™ platform, that combines experimental and computational advances for cell optimisation. In particular, their design process includes four steps
Molecular Engineering to assemble modules in immune cells from libraries of proprietary CARs, switch receptors (a chimeric molecule that combines a ligand binding domain with an alternative signalling domain) and responsive DNA elements (short sequences of DNA within a gene promoter or enhancer region that are able to bind specific transcription factors and regulate transcription of genes),
Rapid Prototyping to quickly produce and test thousands of module designs in a single experiment,
Tunable Assays to replicate mechanisms of immune cell dysfunction and rescue with high fidelity, and
Machine Learning to efficiently search component combinations to meet design goals.
The result is to develop next-generation therapies that outperform industry benchmarks. Modulus is supported by Madrona Venture Group, Allen Institute for AI and KdT Ventures and has raised a total of $3.5M.
👉 Serotiny founded in 2014 invents new therapeutic genes in high-throughput, and designs its experiments for analysis enabling the use of ML. Serotiny is a pre-clinical discovery company using high-throughput engineering of therapeutic Multi-Domain Proteins (tMDP) such as CARs, CAR alternatives, accessory proteins and next generation gene editors. Serotiny has raised a total funding of $3M over 1 round.
So far, Seronity had collaborations with:
Twist Biopharma, to discover novel Chimeric Antigen Receptors for CAR T-Cell therapies by combining Twist’s discovery platform with Serotiny’s core software engineering and high throughput assay platform and ultimately address difficult-to-treat cancers,
Tessera, to engineer programmable Gene Writer proteins that can allow the substitution, insertion or deletion of DNA sequences, as well as the writing of entire genes into the genome, and with
Janssen, to optimise CAR designs for cellular therapy, in which a patient's T cells are engineered with chimeric antigen receptors/CARs and used as a potent new class of therapeutics for cancer.
👉 Cellino Biotech founded in 2017 by Nabiha Saklayen (@nabsicle) is using AI and ML to automate stem cell production, for stem cell-derived regenerative medicines for Parkinson’s 🧠, diabetes, heart 🫀 disease and many more. Cellino’s platform combines label-free imaging and high-speed laser editing with ML to automate cell reprogramming, expansion and differentiation in a closed cassette format, enabling thousands of patient samples to be processed in parallel in a single facility, combining biology, laser physics, gene editing tools and ML.
On May 17, 2023, Cellino has been named the winner of the “Scalable Solutions” category of the Advanced Research Projects Agency for Health’s inaugural Dash to Accelerate Health Outcomes. Cellino was also named overall first runner-up out of 64 teams after six rounds of voting.
CEO & Co-Founder Nabiha Saklayen defined Cellino as a closed loop ➰ cell therapy manufacturing company and has raised so far $96M. In particular, last year they completed a Series A financing of $80 million led by the impact investment arm of Bayer AG, Leaps by Bayer, 8VC and Humboldt Fund.
👉 Rubik Therapeutics founded in 2020, is also leveraging its proprietary computational platform to identify novel tumour antigens and map immune cell signalling pathways, to develop cell therapies that target highly selective, unique solid tumour targets.
LG Chem Life Sciences, a business division within LG Chem engaged in development, manufacturing and global commercialisation of pharmaceutical products, has selected Rubik Therapeutics as the winner of their 2022 LabCentral Golden Ticket Program. LabCentral is a first-of-its-kind shared laboratory space designed as a launchpad for high-potential life-sciences and biotech startups located in the heart of Kendall Square in Cambridge, MA. Rubik Therapeutics is backed by Illumina Accelerator in the U.S.
👉 Exogene (founded in 2019, Oxford, UK 💂♂️), combines AI, high-throughput TCR-cell screening and deep sequencing to mine the immune system for rare, natural TCRs against novel targets, that uses to develop its own pipeline of TCR-based cell therapies. In 2022, Exogene secured $2 million to advance its AI platform for solid-tumour cancer cell therapies.
Gene therapy and adeno-associated virus (AAV) startups using AI
👉 Dyno Therapeutics founded in 2018, is transforming the gene therapy landscape with AI-powered AAV vectors. In 2021, Dyno Therapeutics Closed a $100M 💰 Series A Financing Led by Andreessen Horowitz (for a total of $109M over 2 rounds). Dyno Therapeutics, that is a TechBio company, is using AI to engineer AAV capsids that can expand the potential of genetic medicine.
Just a month ago Dyno announced the launch of its Dyno bCap 1™ capsid product, a breakthrough central nervous system (CNS)-targeted AAV gene delivery vector with best-in-class potential. The Dyno bCap 1 vector provides dramatically improved CNS delivery (since delivering genes to the brain has been an elusive goal due to the blood-brain barrier) and liver detargeting (when compared to leading natural capsids) and stronger all-around characteristics relative to other engineered CNS-IV capsids.
Moreover, in 2021 Astellas Pharma signed an option and licence agreement with Dyno Therapeutics to develop AAV gene therapy vectors for skeletal 💪 and cardiac 🫀 muscle, in order to utilise its CapsidMap platform (which uses AI to rapidly optimise AAV capsids) for delivery across multiple organs, while Astellas will conduct preclinical, clinical and marketing activities and also produce gene therapy product candidates using the capsids.
👉 Always in relation to AI and AAV gene therapy, on November 8, 2022 Cyagen and the clinical-stage Neurophth Therapeutics have entered a strategic partnership to jointly develop next-generation AAV gene therapy vectors (the adeno-associated virus/AAV vectors are the leading platform for gene delivery) for specific kinds of genetic ophthalmic ailments. According to this deal, Cyagen will leverage its AI-driven high-throughput platform for discovering new AAV vectors with enhanced tissue targeting expertise, tissue specificity and productivity.
In particular, Cyagen’s platform by using AI and single-cell RNA-sequencing technologies aims to overcome the current limitations of gene therapy research and development by rapidly detecting next-generation AAV capsids. Specifically, the “AI Delivery vehicle design” process in order to increase AAV capsids tissue specificity, load capacity and stability:
starts with millions of mutated DNA-encoded capsid designs, computing structures and features, as well as from resulting mRNA and protein,
performs high-throughput measurement of the experimental capsid properties, and
links back the measurements to the original design space to improve the capsid designs.
Under this collaboration, Cyagen is entitled to get research phase and clinical phase milestone payments and sales-based royalty payments totalling over $140M 💰. Neurophth (founded in 2016) is China's ⛩️ leading in-vivo gene therapy company for ophthalmic 👀 diseases.
👉 Another Gene Therapy company with viral AAV modality utilising AI is Patch Biosciences. Nicholas Bogard of Patch Biosciences, a former UW postdoctoral research associate, was a co-author of a paper published last year “Multiplexed direct detection of barcoded protein reporters on a nanopore array” , explaining how a group of researchers developed the “Nanopore-Tal” technology that allows “Cells To Chat 🗨️ with Computers”, thanks to a new class of reporter proteins that can be directly read by a commercially available nanopore sensing device.
Patch Biosciences, founded in 2020 has a team at the interface of synthetic biology and ML, and is hiring right now.
mRNA-based therapy startups using AI
👉 Anima Biotech founded in 2005 is advancing mRNA Lightning, that is a novel platform for the discovery of selective small molecule mRNA drugs and their mechanisms of action MoA 🔩⛓️🔗🗜️🔐 (for more about MoA), combining high scale phenotypic screening and AI.
So far, Anima Biotech had strategic collaborations with Lilly and Takeda, while this year on January 10, 2023, AbbVie paid $42 million upfront to work with the Anima Biotech on small molecule mRNA drugs against three “undruggable” targets in oncology and immunology.
Anima Biotech, that is considered a leader in the discovery of small molecule mRNA drugs, launched this year a groundbreaking webinar series in partnership with leading life science media outlets, to provide up-to-date, high-quality information on mRNA biology: Exposing hidden targets within the mRNA regulation space.
👉 StrandTx, co-founded by world-leading mRNA researchers from the MIT Synthetic Biology Centre (Jake Becraft, Ph.D., Tasuku Kitada, Ph.D., Prof. Darrell J. Irvine, Ph.D., Prof. Ron Weiss, Ph.D.) and founded in 2017, is an emerging biopharmaceutical company applying synthetic biology to RNA therapeutics and developing the first platform for the creation of programmable, long-acting mRNA drugs capable of delivering precise, multi-functional, potentially curative treatments with a single dose.
They are merging synthetic biology, ML and mRNA therapeutics to program mRNA with more precision and control in cell behaviour, in order to enable precise control of the location, timing, intensity and duration of therapeutic protein expression for improved efficacy and lower toxicity. StrandTx is backed by Anri operates, Alexandria Venture Investments and Playground and has raised a total of $104.2M.
Gene & Cell therapy startups using AI
👉 In 2022, Form Bio was spun out of Colossal Biosciences and entered the market with an oversubscribed $30M Series A funding round led by JAZZ Venture Partners with participation from Thomas Tull, Colossal lead investor. Form Bio introduced FORMsightAI A breakthrough AI-powered solution to predict and optimise cell and gene therapy manufacturing.
FORMsight (Simulate and Optimise) is solving the replication and packaging problem for cell and gene therapy manufacturing (that can cost millions in manufacturing costs and months or years in time-to-market), by predicting non-empty/empty capsid ratios, capsid content proportions by full, truncated and chimeric components, truncation propensities and by breaking down truncation contribution proportions by sequence region. Finally, with FORMsight Optimise you can
generate the ideal construct for maximised predicted therapeutic drug products produced while minimising key inefficiencies, and develop a new construct design blueprint.
The Potential Economic Impact of FORMsight translates to reducing cost per dose by more than 45%, to increasing manufacturing dose yields by 88% and to treating a patient population of 50,000 in 5 Years instead of 10.
On May 17, 2023, Form Bio announced that it has been selected to help organisations adopt Google Cloud's Multiomics Suite, accelerating the research of life science companies and academic institutions. In addition, Form Bio has been designated by Google as a Google Cloud Premier partner. And last year, in 2022, Form Bio announced the integration of the NVIDIA Clara Parabricks suite of genomic analysis tools with Form Bio's genomic analysis workflows. You can download here the Whitepaper “Machine Learning for Cell and Gene Therapy”, by Joseph Nipko, PhD, Head of Computational Research at Form Bio. Form Bio has raised $30M.
👉 On September 13, 2022, Whitelab Genomics Raised $10M 💰 in Series A Funding from Debiopharm Innovation Fund, the strategic investment arm of Swiss 🧀🫕 biopharmaceutical company Debiopharm, and Omnes Capital. At the interface between gene and cell therapy companies the French 🥐🥖 AI biotech company WhiteLab Genomics provides predictive software simulations to facilitate 🧞 the design of gene and cell therapies.
The company (founded in 2019 and a YC graduate) has a proprietary platform that leverages graph knowledge technology and ML to help scientists discover and design new gene, RNA and cell therapies, for:
target 🎯 discovery (target expression, profiling, epigenetic signatures identification, protein structure and interaction simulations, cellular biomarker screening),
payload design 👨💻 (sequence design, promoter/enhancer choice, chimeric construction, transgene expression modulation) and
vector design (viral capsid & envelope design, transduction optimisation, biodistribution prediction, immunoreactivity modulation).
Finally, a team of researchers by using the AI system AlphaFold managed to modify a syringe-like protein naturally found in Photorhabdus asymbiotica (a species of bacteria that primarily infects insects) to inject proteins, including cancer-killing drugs and gene therapies, directly into cells (DeepMind's AI used to develop tiny 'syringe' for injecting gene therapy and tumor-killing drugs). However, the modified syringe-like protein has not yet been tested in humans, only in lab dishes and live mice.
Until next time,