Insilico Medicine is an AI-based drug discovery company and its first drug fully generated by artificial intelligence entered clinical trials with human patients recently. Pulse 2.0 interviewed Insilico Medicine CEO and founder Alex Zhavoronkov, PhD to learn more.
Dr. Zhavoronkov’s Background
Dr. Zhavoronkov said that for as long as he could remember, he wanted to study aging.
“I also realized early in the process that aging is very complex and cannot be solved by just one human scientist. But computers and the Internet could help. I went to Canada (from Latvia) to study computer science and aging research. I received my first two bachelor’s degrees in computer science and in commerce at Queen’s University. I graduated a few years after Elon Musk left. After graduation, I spent a few years in the semiconductor industry working for ATI Technologies where I learned about the applications of graphics processors in AI,” said Dr. Zhavoronkov. “After ATI was acquired, I made enough money to pursue my long-term dream. I pursued a Master’s degree at Johns Hopkins and spent some time at a lab studying long-lived wasps, as back then I thought that research in worms and flies might be translatable to humans. Then I worked for a number of biotechnology companies related to aging research and pursued my Ph.D. in biophysics at Moscow State University, where I studied racemization of amino acids in proteins in the course of aging. In parallel, I started a project analyzing government grants and publications to track biomedical progress and trends. I was very interested in how a dollar invested in a biomedical project today would result in a publication, patent, clinical trial, and finally into the market to extend human longevity. By 2011, when I published my first paper with Dr. Charles Cantor, former director of the Human Genome Project at the Department of Energy (DOE) and founder of Sequenom, we were analyzing approximately $1 trillion dollars in government funding around the world and the data resulting from this funding. It is a massive public data set annotated by many experts. This early work formed the basis of Insilico Medicine’s end-to-end generative AI drug discovery platform called Pharma.AI, which is built around trillions of data points, including omics data, grants, patents, and clinical trials.”
As the founder and CEO of the company, Dr. Zhavoronkov is involved in every aspect of the business.
“I travel the world to talk about the company, our pipeline of assets, generative AI in drug discovery, robotics, aging, and longevity. The company has a co-CEO, Dr. Feng Ren, who has deep pharma and drug discovery experience and is our Chief Scientific Officer. This unique co-CEO structure allows us to expand on all fronts simultaneously. I focus on global expansion and next-generation AI platforms while Dr. Ren drives the company’s drug development efforts,” Dr. Zhavoronkov added.
Favorite Memory
What has been Dr. Zhavoronkov’s favorite memory of working for Insilico Medicine?
“We’ve had a number of exciting, life-changing moments, but one of the most significant was advancing our lead drug to clinical trials. When we announced that the first healthy volunteer had been dosed with our lead AI-discovered and AI-designed drug for the devastating chronic lung disease idiopathic pulmonary fibrosis in Nov. 2021, it was a thrilling moment for me personally, for the company, and for the larger field of AI drug discovery,” Dr. Zhavoronkov reflected. “Ours was the first both AI-discovered and AI-designed drug to reach that stage and the moment represented a total validation of our end-to-end generative AI platform. Now that same drug has just entered Phase II trials with patients – a first for a fully generative AI drug – and we have 31 drugs advancing in our pipeline.”
Challenges Faced
What are some of the challenges Dr. Zhavoronkov faced in building the company and has the current macroeconomic climate affected the company?
“Insilico Medicine’s focus from its inception has been to address aging-related diseases as well as aging itself through the development of dual-purpose therapeutics. Our generative AI platform is trained on both aging and disease. But the industry has been slow to engage in serious discussions about aging. That’s finally beginning to change. I’ve witnessed a surge in interest in the annual Aging Research and Drug Discovery Conference which is now held at the University of Copenhagen (August 28-Sept. 1) and attracts leading scientists, industry leaders, and investors,” Dr. Zhavoronkov acknowledged. “Another challenge in transforming the pharmaceutical industry with artificial intelligence is the silos within traditional drug development. We developed an end-to-end fully customizable drug discovery platform using generative AI called Pharma.AI that allows us to address this in a number of ways. We use this platform – which includes target discovery (PandaOmics), drug design (Chemistry42), and clinical trial prediction (InClinico) to develop our own internal pipeline of assets that we can then license to pharma companies. We also have progressed a lead drug candidate for the chronic lung disease idiopathic pulmonary fibrosis, which will soon be in Phase 2 trials with patients, and another for COVID-19 and related variants that are also in clinical trials. And we have partnerships with leading pharma companies using our platform and expertise to develop their own programs, including Fosun Pharma and Sanofi. In addition, pharma companies can leverage our platform through software agreements.”
Core Products
What are Insilico Medicine’s core products and features? Insilico Medicine has both technology/software assets and a robust pipeline of pharmaceutical assets. These include:
Technology Products
The company’s core technology product is Pharma.AI, which is a generative AI-powered drug discovery and development platform with three components: PandaOmics (for target discovery), Chemistry42 (for molecular synthesis), and InClinico (for clinical trial prediction and optimization).
“PandaOmics includes more than 20 AI and bioinformatics models. It uses insilico pathway activation network decomposition analysis and generative AI approaches to identify novel targets and biomarkers. PandaOmics includes a knowledge graph for target-disease associations, which is generated by transformer-based natural language processing models applied to biomedical information published in the scientific literature for the last 30 years. Chemistry42 is a software platform launched in 2020 that integrates generative AI algorithms with computational and medicinal chemistry methods to generate novel molecules with drug-like properties. Chemistry42 uses over 40 generative models, including generative autoencoders and generative adversarial networks. InClinico is a generative AI-powered software platform utilizing a data-driven, multimodal forecast for predicting the outcome of Phase II clinical trials launched in November 2022. The platform’s prediction engines utilize data from small molecule chemical properties, transcriptomics, text data, and clinical trial protocols. InClinico’s predictive accuracy has been tested through retrospective, quasi-prospective, and prospective validation studies,” Dr. Zhavoronkov explained. “In addition to this end-to-end platform, Insilico also opened an AI-powered robotics laboratory in Suzhou’s BioBAY Industrial Park. The automated laboratory performs target discovery, compound screening, precision medicine generation, and translational research.”
Pharma Pipeline Assets
Insilico Medicine has 31 programs across 29 targets in its pipeline, including fibrosis, oncology, immunology, and central nervous system disorders.
“The lead program, INS018_055, is a small molecule inhibitor against a novel target known as Target X, with the drug candidate currently being evaluated in Phase II clinical trials with patients as a potential treatment for idiopathic pulmonary fibrosis. Insilico has many potentially first-in-class and best-in-class pipeline assets available for licensing and recently presented four anti-cancer AI-designed molecules at the American Association for Cancer Research annual meeting in March. These assets include a QPCTL inhibitor designed as novel cancer therapy for immunologically cold tumors; a USP1 inhibitor for BRCA-mutant cancer which received IND approval for clinical trials; a 3CL inhibitor for COVID-19 and related variants currently in clinical trials; a MAT2A inhibitor for MTAP-deficient cancer; a Ph.D. inhibitor for inflammatory bowel disease and anemic chronic kidney disease; a KAT6A inhibitor for ER+/HER2- breast cancer; and a DGKA inhibitor being evaluated as an immunotherapy treatment for solid tumors,” Dr. Zhavoronkov continued.
Evolution Of Insilico Medicine’s Technology
How has Insilico Medicine’s technology evolved since launching? “We are constantly evolving our technology. In January 2023, we announced the opening of our 6th generation fully autonomous AI-powered robotics lab capable of performing target discovery, compound screening, precision medicine development, and translational research. We incorporated AlphaFold 2 into our platform in a highly publicized study with the University of Toronto’s Accelerative Consortium in order to demonstrate that the AlphaFold protein database has practical utility for drug discovery – we ultimately used the AlphaFold predicted structure to find a small molecular inhibitor for liver cancer in under 30 days. And this past March we incorporated large language models into our platform – called ChatPandaGPT – to allow for more seamless discovery and navigation of our knowledge graphs,” Dr. Zhavoronkov pointed out.
Significant Milestones
What have been some of Insilico Medicine’s most significant milestones? Dr. Zhavoronkov cited the following:
— First publication in generative chemistry: In Dec. 2016, the company published in Oncotarget, “The cornucopia of meaningful leads: Applying deep adversarial autoencoders for new molecule development in oncology. This publication coincided closely with the publication of a similar idea by Alan Aspuru-Guzik’s team in their ArXiv paper “Automatic chemical design using a data-driven continuous representation of molecules.” During this period, the company began to collaborate and build a global community around generative chemistry.
— Building a proprietary generative AI system for drug discovery: In 2018, the company progressed towards building and validating a powerful deep generative model, generative tensorial reinforcement learning called GENTRL – an AI system for drug discovery that dramatically accelerates the process of lead discovery from years to days. The company made the code publicly available on GitHub to encourage a broader community of scientists to continue building on this work.
— Launching their end-to-end Pharma.AI platform: By 2020, the company had released its own end-to-end AI platform with three key components: PandaOmics, target discovery, and multi-omics data analysis engine; Chemistry42, a de novo molecular design engine; and Inclinico, a clinical trial outcomes prediction engine InClinico.
— Signing their largest partnership deal: In Nov. 2021, Insiico signed its largest partnership to date, a strategic research collaboration to advance drug development targets for up to 6 candidates, worth up to $1.2 billion.
— Signing another major pharma partnership deal: In January 2022, Insilico Medicine announced a multi-target collaboration with Fosun Pharma, along with co-development of Insilico’s QPCTL program, including an upfront payment of $13 million, equity investment, and other development- and commercial-based milestones.
— Orphan Drug Designation for their lead candidate: In Feb. 2023, the company announced that the FDA had granted Orphan Drug Designation for their lead AI-discovered and AI-designed drug for IPF, moving us one step closer to Phase 2 trials with patients.
— Opened the largest AI and Quantum Computing R&D Center in the Middle East: Insilico opened its Abu Dhabi-based R&D center in Feb. 2023, and is rapidly expanding its presence and partnerships in the region.
Customer Success Stories
Can you share any specific customer success stories?
Deerfield Management was one of the investors behind Insilico’s $225 million Series C round in June 2021 and Insilico Medicine has its New York City office in Deerfield’s CURE building – a life sciences hub on Park Ave. South. Soon after, the Deerfield Discovery and Development team partnered with Insilico to utilize its Pharma.AI platform for Deerfield’s incubator companies and the company talked about the value that Insilico’s platform brings to the drug discovery process.
Andy Stamford, Ph.D., VP of Chemistry at Deerfield Discovery & Development, observed that they are extremely impressed by the expert team of scientists at Insilico and the sophistication of the technology and chemistry. And he believes the platform has the potential to broadly impact drug discovery and break ground on novel targets across multiple disease indications.
Funding
Insilico Medicine has raised over $400 million since its founding in 2014. Last year, the company raised Series D financing of $95 million from new and existing investors including Prosperity7 (Aramco’s growth venture fund), a large US-based asset management firm, B Capital Group, Warburg Pincus, BHR Partners, Qiming Ventures, Deerfield, Pavilion Capital, BOLD Capital, along with Alex Zhavoronkov and other investors. In June 2021, the company raised Series C financing of $255 million. Additional investors include Eight Roads, Lilly Asia Ventures, Sinovation Ventures, CPE, OrbiMed, Mirae Asset Capital, Maison Capital, Lake Bleu Capital, President International Development Corporation, Sequoia Capital China and Sage Partners.
Total Addressable Market
What total addressable market (TAM) size is Insilico Medicine pursuing? “Our market is pharmaceuticals, with over 30 programs in development – some of these markets are in the billions,” Dr. Zhavoronkov assessed.
Differentiation From The Competition
What differentiates Insilico Medicine from its competition? Dr. Zhavoronkov replied:
“Insilico Medicine is unique from other companies in the AI drug discovery space in a number of key ways, beginning with our end-to-end discovery AI capabilities across target discovery, drug candidate generation, and clinical trials. While 80% of AIDD companies are focused on chemistry, we can discover and identify novel targets through our target identification platform, PandaOmics, and then design compounds using our small molecule generation platform, Chemistry42, which is how the potentially first-in-class drug candidate INS018_055 was discovered,”
“The second point is that we need to ensure the effectiveness of our AI platform, something that Insilico has been working on. Validation is expensive and time-consuming. At Insilico we use both external and internal programs to validate our AI-enabled drug discovery capabilities. For external programs, we work with big pharmaceutical companies, providing them with services and software to perform validation and receive feedback. For internal programs, we have nominated 11 preclinical candidate compounds since 2021, which further validates Insilico’s AI drug discovery capabilities in the early stages.”
“The third point is that Insilico is a company co-driven by AI and drug discovery, with two co-CEOs, one focused on AI development and the other on drug discovery. Insilico is doing well in terms of combining the data-driven nature of artificial intelligence with the experience and curiosity of human intelligence.”
“Also, we have invested deeply into AI and have accumulated a lot of data. We followed $2 trillion worth of research data and invested significant time and resources in making the data machine readable so that it can be used in our AI platform. Our AI engines are quite unique. For example, we have 42 generative models in our Chemistry42 platform, which can generate compounds with desired properties without the use of large molecular libraries.”
“In addition, we have an open-source model and license our technology platforms to pharma companies and academics. The benefit of our unique business model and open platform is the feedback we receive which continually improves our technology. We are in a constant feedback loop with partners who license our software.”
Future Company Goals
What are some of Insilico Medicine’s future company goals? “We will continue to further develop our platform, including robotics capabilities, partnering with pharma companies, and advancing the 30+ assets in our pipeline. Our lead drug for IPF has entered Phase II trials, we have a COVID-19 drug just accepted into clinical trials and a cancer drug for the treatment of solid tumors approved for clinical trials, and we are eager to see more of these new therapies advance into the clinic,” Dr. Zhavoronkov concluded. “While our primary focus is on developing new therapeutics for diseases where there is a high unmet need, we are also able to use our platform to develop sustainable chemicals and have been exploring this avenue through partnerships with Syngenta and Aramco. In much the same way we are able to develop and design novel drugs optimized to treat disease, the platform can also be used to optimize the design of chemicals with properties that are less harmful to the environment.”
