AI-Engineered Cancer Drug gets US FDA Approval for Trials

June 7, 2023

This News Covers

The recent developments at Insilico Medicine underline the importance of continuous investment in AI research and development to harness its potential fully.

Insilico Medicine, a company that uses artificial intelligence (AI) to discover new drugs, has made a big announcement. The U.S. Food and Drug Administration (FDA) has given them the green light to start testing a new drug, ISM3091, designed to treat patients with solid tumors. This is a significant milestone as it's the first time their cancer drug has reached this stage of testing.

In simple terms, they will be testing this new drug in both the U.S. and China in what is known as a Phase I clinical trial. These trials are designed to check that the drug is safe, find the best dosage, and see how well it works. This specific drug, ISM3091, is what they call a small molecule inhibitor of USP1 - a sort of a fancy way of saying it's a tiny structure that blocks a certain process in the body, one that's important in the development of cancer.

What's really cool is that they used their own AI technology, named Chemistry42, to design this drug. Imagine a super-smart computer program that can sift through huge amounts of information about biology and chemistry to come up with new ideas for drugs. That's what they used to create ISM3091. Plus, they did it really quickly – they had the drug ready for testing just nine months after they started working on it! This success gives them hope that their AI can help discover more new drugs in the future.

What is ISM3091?

It is a potent and highly selective small molecule inhibitor of USP1 as a target anticancer agent[1]

Essentially, below are the key takeaways from this momentous development:

  1. Insilico Medicine, a comprehensive pharma-technology firm powered by artificial intelligence, whose goal is to expedite the process of drug discovery and development, has secured FDA approval to initiate testing for a new cancer drug, ISM3091. [2]
  2. The Phase I clinical trial will be held in the U.S. and China to test the safety and effectiveness of ISM3091 in treating solid tumors.
  3. The drug was designed using Insilico's AI technology, Chemistry42, highlighting the potential for AI to accelerate and improve the drug discovery process.
  4. ISM3091 was ready for testing just nine months after the project began, illustrating the speed at which AI can aid in drug development.
  5. This is one of several drugs developed using AI by Insilico, showing the broad potential of their technology.

It's thrilling to witness the breakthroughs that are possible at the intersection of AI, chemistry, and biology, and I anticipate many more to come in the near future.

How significant and relevant is this clinical trial to Advanced Cancer Tumor related cases?

Let us understand this one by one in detail.

What is a Phase I clinical trial and what is its purpose?

Phase I clinical trials are the first stage of testing a new drug in humans. In these trials, researchers are primarily trying to figure out if the drug is safe, what's the right dosage to use, and how the body responds to the drug. In short, they're checking if the drug works as intended without causing any harmful side effects.

What does "open-label" and "multicenter" mean?

An "open-label" trial means that both the doctors and the patients know what drug is being given - there's no placebo or 'fake drug' used for comparison. On the other hand, a "multicenter" trial means the testing is happening at more than one place. In this case, the drug testing is happening in both the U.S. and China.

What does 'pharmacokinetics' mean?

Pharmacokinetics is a fancy word scientists use to describe how a drug is absorbed, distributed, metabolized (broken down), and excreted by the body. In simpler terms, it's about what your body does to a drug once it's inside.

What is "dose escalation" and "dose optimization"?

"Dose escalation" is a process where the dose of the drug is gradually increased to find out the highest amount that can be given without causing severe side effects. "Dose optimization" is the process of figuring out the best dosage that's both effective and safe.

What does "IND submission" mean?

"IND" stands for "Investigational New Drug". When a company wants to test a new drug in humans, it must submit an IND application to the FDA in the U.S., or to the equivalent authority in other countries. This application includes data from laboratory and animal testing, showing the drug is safe enough for human testing.

What are they testing the drug ISM3091 for?

They're testing ISM3091 to see if it's safe and effective in treating patients with advanced solid tumors - these are cancers that form in solid parts of the body, like organs or bones. If all goes well, they'll be able to figure out the best dosage for future testing and treatment.

What role has Generative AI played in cancer related drug trial?

In the context of Insilico Medicine's development and FDA approval, Generative AI has played a pivotal role in several aspects:

  1. Speeding up Drug Discovery: Insilico's proprietary AI platform, Chemistry42, used Generative AI to design a novel cancer drug - ISM3091. This process, which would traditionally take several years, was completed in just nine months. The AI sifted through large amounts of biological and chemical information to generate a new drug candidate that could effectively treat solid tumors.
  2. Enhanced Drug Design: Generative AI enabled the development of ISM3091, a small molecule inhibitor of USP1. In simpler terms, it's a tiny structure that disrupts a particular process in the body, one crucial for cancer development. This wouldn't have been possible (or would have taken a much longer time) without the aid of AI.
  3. Clinical Trial Planning: Generative AI also likely assisted in planning the Phase I clinical trials. It would have helped predict the optimal dosage levels, potential side effects, and the patient response, which is critical in the early stages of human testing.
  4. Regulatory Approval: Insilico's successful IND submission to the FDA (and NMPA) for ISM3091 was built on preclinical data generated with the aid of AI. The ability of AI to predict the drug's safety and efficacy based on lab and animal studies likely contributed significantly to gaining regulatory approval for human trials.

In essence, Generative AI not only expedited the drug design process but also assisted in preclinical testing, regulatory submissions, and in planning the upcoming clinical trials. It's an example of how AI is revolutionizing drug discovery and development, particularly in oncology.

What are the top 5 features and impacts of this usage of Generative AI in Biotech

Speed of Drug Discovery: It is noteworthy that Insilico nominated the preclinical candidate targeting USP1 in just 9 months, highlighting the capacity of AI to expedite the drug discovery process. Traditional methods usually take years to reach the same stage, demonstrating a significant improvement in efficiency brought by AI technologies.

Innovative Drug Design: The use of Insilico's proprietary AI platform, Chemistry42, to design and optimize the structure of ISM3091 is remarkable. It shows the potential of AI to not only expedite the drug discovery process but also contribute to the innovative and efficient design of molecules.

Efficacy and Tolerance: Preliminary results indicating the potent efficacy of ISM3091 against multiple tumor cell lines and its well-tolerated nature in different species reinforce the role of AI in producing viable drug candidates. The drug's potential to provide a novel solution for patients with solid tumors who do not respond to or develop resistance to PARPi is another promising feature.

Versatility of AI in Drug Discovery: Insilico's success isn't limited to one drug. They've nominated 12 preclinical candidate compounds since 2021 and have three leading programs in the clinical stage. This indicates the versatility and breadth of application that AI-powered drug discovery platforms can offer.

AI-Driven Synthetic Lethal Therapy: ISM3091 represents a new generation of synthetic lethal therapy. This therapy exploits the genetic weaknesses in cancer cells, thus potentially paving the way for innovative treatments for cancer patients.

An overview of Generative AI

Generative Artificial Intelligence (AI) is a cutting-edge technology that has emerged as a transformative force in various industries, including biotechnology. It focuses on the creation, synthesis, and generation of new content using advanced deep learning techniques and neural networks. One of the pivotal milestones in the development of generative AI is the introduction of Generative Adversarial Networks (GANs) by computer scientist Ian Goodfellow in 2014. GANs consist of two neural networks, a generator and a discriminator, which collaborate to generate realistic data and distinguish between authentic and generated content.

The primary purpose of generative AI lies in its capacity to produce novel and creative outputs. In the biotechnology industry, this technology has profound implications, ranging from accelerating drug discovery and optimizing bioprocess engineering to advancing biomedical research. By leveraging generative AI, scientists and researchers can design new compounds, predict drug-target interactions, optimize bioprocesses, and gain insights from complex biomedical data.

Transforming Biotechnology: GPT-4 and Generative AI's Expanding Role in Advancements

Biotechnology, coupled with the advancements in generative AI, has opened up new possibilities and accelerated progress in various areas. The introduction of GPT-4 (Generative Pre-trained Transformer 4), the next iteration of AI language models, further enhances the capabilities of generative AI. GPT-4 exhibits improved natural language understanding, context comprehension, and creative generation, enabling more sophisticated applications in the biotechnology industry.

Advancements in Medicine: Generative AI, including GPT-4, is being utilized in drug discovery and development. It aids in virtual screening of large compound libraries, predicting drug-target interactions, and optimizing molecular designs for improved efficacy and safety. This accelerates the process of identifying potential drug candidates and expedites the development of life-saving therapies.

Improved Agricultural Practices: Generative AI contributes to biotechnology's impact on agriculture by assisting in the development of genetically modified crops. It aids in optimizing genetic traits, improving crop yields, and enhancing resistance to pests, diseases, and environmental stressors. GPT-4's capabilities enable more precise genetic engineering and data-driven decision-making in crop breeding and biotechnology-based agriculture.

Diagnostic and Therapeutic Innovations: Generative AI, including GPT-4, is instrumental in the field of diagnostics and therapeutics. It facilitates the analysis of large-scale biomedical data, such as genomics, proteomics, and medical imaging, leading to more accurate disease diagnosis and personalized treatment plans. GPT-4's advanced language understanding enables better interpretation of medical records, aiding in clinical decision support and patient stratification.

The integration of generative AI and GPT-4 in biotechnology applications empowers researchers and healthcare professionals to harness the power of data, accelerate discoveries, and deliver more effective solutions. By leveraging generative AI's creative generation capabilities and GPT-4's enhanced language understanding, the biotechnology industry can drive innovation, improve patient outcomes, and contribute to the advancement of medical science and agricultural practices.

The implications of generative AI in the biotechnology industry are poised to revolutionize various subdomains, accelerating advancements and unlocking new possibilities. From drug discovery and development to agricultural biotechnology and diagnostic innovations, generative AI, including the latest GPT-4 model, holds immense potential. In the realm of biotechnology, major subdomains such as pharmaceutical biotechnology, agricultural biotechnology, industrial biotechnology, and medical biotechnology will experience transformative changes. By leveraging generative AI's capabilities, researchers and professionals in these subdomains can enhance precision medicine, optimize crop breeding, improve manufacturing processes, and drive breakthroughs in therapeutic development. The integration of generative AI in biotechnology heralds a new era of innovation, propelling the industry forward to address global challenges and improve human well-being.

The biotechnology market is a rapidly growing sector that is poised to be significantly impacted by the implications of generative AI. According to market reports, the global biotechnology market size was valued at over $450 billion in 2020 and is projected to reach new heights in the coming years. With the integration of generative AI technologies, the market's potential for growth and innovation expands even further. The application of generative AI in drug discovery, agricultural biotechnology, precision medicine, and other subdomains has the potential to revolutionize processes, accelerate advancements, and drive market expansion. The transformative power of generative AI positions the biotechnology market to witness significant growth and drive positive change across industries worldwide.

Who is likely to play a major role in the near future in Biotech?

  1. Precision Medicine Software Market:
    1. Tempus: Data analytics platform for personalized cancer care.
    2. Syapse: Precision medicine platform for healthcare providers and researchers.
    3. Foundation Medicine: Molecular information platform for precision oncology.
    4. Flatiron Health: Oncology-focused electronic health record and data analytics platform.
    5. Sophia Genetics: AI-powered platform for genomic data analysis and clinical decision support.
  2. Aptamers Market:
    1. Aptamer Sciences: Custom aptamer synthesis and selection services.
    2. NeoVentures Biotechnology: Aptamer-based products for diagnostics and therapeutics.
    3. Base Pair Biotechnologies: Custom aptamer development and conjugation services.
    4. AM Biotechnologies: Aptamer discovery and development services.
    5. Aptagen: Aptamer products and screening services for various applications.
  3. Label-free Detection Market:
    1. Biacore (GE Healthcare): Label-free interaction analysis systems and consumables.
    2. Attana: Label-free cell-based and biochemical interaction analysis systems.
    3. Corning Incorporated: Microplate-based label-free detection instruments and assay kits.
    4. Horiba Scientific: Label-free analysis instruments based on surface plasmon resonance (SPR) and other techniques.
    5. Nanotemper Technologies: Microscale thermophoresis instruments for label-free biomolecular interaction analysis.
  4. Bioprocess Validation Market:
    1. Merck KGaA: Bioprocess validation services, media, and equipment.
    2. Charles River Laboratories: Bioprocess validation and testing services for pharmaceutical and biotech industries.
    3. Eurofins Scientific: Validation services for bioprocessing, including cleaning, sterilization, and environmental monitoring.
    4. Sartorius AG: Bioprocess validation tools, sensors, and analytics for biopharmaceutical manufacturing.
    5. Pall Corporation: Bioprocess validation services, filtration, and single-use technologies.

Which top markets will Generative AI impact

Generative AI has the potential to significantly impact the Precision Medicine Software, Aptamers, Label-free Detection, and Bioprocess Validation markets in several ways:

  1. Precision Medicine Software: Generative AI can enhance precision medicine software by enabling more accurate analysis and interpretation of genomic data. It can assist in identifying patterns, biomarkers, and potential therapeutic targets, leading to improved patient stratification and personalized treatment plans. Generative AI algorithms can also aid in predicting drug responses and adverse events, facilitating more informed clinical decision-making.
  2. Aptamers: Generative AI can play a crucial role in aptamer development by assisting in the design and optimization of novel aptamer sequences with improved specificity and binding affinity. By leveraging generative AI algorithms, researchers can generate a diverse range of aptamer candidates and identify those with desirable characteristics for various applications, including diagnostics, therapeutics, and biosensors.
  3. Label-free Detection: Generative AI can enhance label-free detection technologies by improving data analysis and interpretation. It can aid in extracting meaningful insights from complex label-free interaction data, enabling more accurate quantification and characterization of molecular interactions. Generative AI algorithms can also contribute to the development of advanced data analysis software and algorithms, empowering researchers to extract valuable information and identify important trends from label-free detection experiments.
  4. Bioprocess Validation: Generative AI can streamline and optimize bioprocess validation by improving data analysis, modeling, and decision-making. It can help in identifying critical process parameters, optimizing process conditions, and predicting potential risks or deviations. Generative AI algorithms can also assist in developing real-time monitoring and control systems for bioprocesses, ensuring robust and compliant manufacturing processes.

Overall, generative AI can revolutionize these markets by enhancing the capabilities of existing technologies, accelerating innovation, and improving outcomes. By leveraging generative AI's ability to generate novel solutions and insights, these industries can benefit from improved efficiency, accuracy, and productivity in their respective applications.

How will the Technology Roadmap for Generative AI be in coming years?

A. Short Term Technology Roadmap (2023-2025): In the short term, the focus of generative AI technology will be on:

  1. Technological improvements in language modeling to enhance natural language understanding and generation.
  2. Advancements in Generative Adversarial Networks (GANs) for more realistic and diverse data generation.

B. Mid-term Technology Roadmap (2026-2028): In the mid-term, the generative AI market will see:

  1. Advances in meta-learning and few-shot learning techniques, enabling faster adaptation to new tasks and domains.
  2. Development of generative AI systems capable of learning from multi-modal data, combining different data types such as text, images, and audio.

C. Long Term Technology Roadmap (2029-2030): In the long term, the generative AI market will witness:

  1. The emergence of fully autonomous generative AI systems, capable of self-learning and decision-making.
  2. Increased utilization of generative AI in scientific research, enabling researchers to generate novel hypotheses and explore complex datasets more efficiently.


1. Insilico Medicine (“Insilico”)

2. Insilico Medicine


Sustainable Digital Transformation & Industry 4.0

Sustainable Digital Transformation & Industry 4.0

Sanjay Kaul, President-Asia Pacific & Japan, Cisco, and host Aashish Mehra, Chief Research Officer, MarketsandMarkets, in conversation on unraveling 'Sustainable Digital Transformation and Industry 4.0'

11 July 2023|S2E12|Listen Now

Future of Utilities with Thomas Birr from E.ON

Generative AI

Prasad Joshi, Senior Vice President-Emerging Technology Solutions, Infosys, and host, Vinod Chikkareddy, CCO, MarketsandMarkets, in exploring the recent advances in AI and the generative AI space.

7 Nov 2023|S2E13|Listen Now

Download Whitepaper


Subscribe Email Protection Status