Introduction: First AI designed drug in human trials.
First AI designed drug in human trials, we are on the cusp of AI breaking into the field of drug design to improve healthcare. We have been hearing about the potential of Artificial Intelligence (AI) in improving traditional drug discovery and development for several years. Clinical trials have begun over the past two years.
Exscientia made headlines last April when it announced that it had begun a Phase 1 clinical trial for a drug it developed using AI to treat an established protein target. In Utah, Recursion Pharmaceuticals uses artificial intelligence to find new uses for drugs owned by other companies.
The company has announced the next step: the first Phase 1 clinical trial of a drug developed from scratch using artificial intelligence. AI can be applied to biology to discover targets, and to chemistry to design drugs using its end-to-end platform. The company believes the drug may have anti-aging properties as well as its immediate therapeutic effect.
Last week, AI-backed drug discovery company Insilico Medicine announced that it had dosed the first healthy volunteer in a microdose trial of ISM 001-005.
A small-molecule inhibitor of a biological target was discovered by Pharma AI and designed with the help of AI. The trial is taking place in Australia.
The AI-designed drug is being developed to treat IPF, or idiopathic pulmonary fibrosis, which is a chronic lung disease. IPF leads to progressive and irreversible lung function decline, and it affects 20 out of every 100,000 people worldwide.
Freng Ren, Chief Scientific Officer at Insilico, said in a press release that this drug discovery marks a significant step forward for the AI-drug discovery field. Since the said candidate is the first-ever molecule discovered by AI based on a target discovered by AI, it is a first.
In this new drug, the disease being targeted is idiopathic pulmonary fibrosis (IPF). The cause of this condition is unknown (hence “idiopathic”), and it stiffens the lungs of older people, eventually killing them. It affects approximately 5 million people annually.
According to Ren, the team has utilized an end-to-end AI-powered drug discovery platform along with novel molecules that have drug-like properties to discover biological targets. A machine learning model called Generative Tensorial Reinforcement Learning, or GENTRL, was used to develop the drug. It cost $2.6 million and took less than 18 months for the process to reach the preclinical stage.
This drug is the first of its kind ever to enter the clinic, and we hope that it will be the first of many.
The ISM 001-005 drug has shown promising results in several preclinical studies, according to the company. The studies include pharmacokinetic, safety, and antibacterial testing. According to Insilico, the compound has also shown significant improvement during myofibroblast activation, which results in fibrosis. Potentially, the drug could be used for a wide range of fibrotic indications.
When a drug is tested on humans for the first time at a dosage level required to produce therapeutic effects, this is called a Phase 1 trial. Safety is the primary concern, not clinical efficacy.
Then, drugs must undergo preclinical trials on animals and Phase 0 trials at subtherapeutic doses. Phase 2 and Phase 3 trials test a drug’s effectiveness, first with a sample of a few hundred people, then with a much larger sample of thousands of people.
Insilico’s Phase 1 trial will involve 80 volunteers, half of whom will receive increasing dosages. The goal of the trial is to determine how the drug molecule impacts the human body, a process known as pharmacokinetics.
The breakthrough isn’t just important for people with IPF. It adds to the growing body of evidence showing that AI is a valuable tool for drug development. Human biology is very complex, making drug development extremely difficult. Scientists noticed that pharmaceutical research and development costs doubled every decade or so as far back as the 1980s. It was later called Eroom’s Law, since it was the reverse of Moore’s Law, which reflected exponential improvements in computing.
The year 2022 marks the tenth anniversary of the Big Bang in AI, which happened when Geoff Hinton finally got a machine learning algorithm called backpropagation to work and achieved a famous breakthrough in image recognition. The explosion of interest in AI this decade has been largely fueled by machine learning, especially deep learning, which has enabled the creation of interactive maps, near-omniscient search engines, and machines that can write, draw, and create music.
Alex Zhavoronkov, founder of Insilico, had previously worked in a company producing GPUs, a type of chip that enabled deep learning, and he began exploring the potential for machine learning in the field of drug development at the time of the Big Bang.
The development process
A three-part, AI-powered process led to ISM001-055, a drug now in Phase 1 trials. PandaOmics is the first part, in which Insilico used natural language processing AI to trawl through massive medical data sets, looking for proteins which could be causing fibrosis. This protein is referred to as a target protein.
The disease of fibrosis is closely linked with aging, and Insilico’s deep neural networks were trained to identify potential targets based on millions of data files, including patents, research publications, grants, and databases of clinical trials.
Insilico used Chemistry42, a type of artificial intelligence (AI) called a Generative Adversarial Network (GAN), to identify a drug molecule that could combat the disease. A GAN pits two AIs against each other: one generates suggestions, while the other critiques them.
In the third step, the company used artificial intelligence to monitor the effects of the drug on eight volunteers in Australia. The whole three-part process took 30 months and £2.6m, which is a small fraction of the traditional drug development process, which costs billions of dollars, takes a decade or more, and suffers from a 90% failure rate.
Along with its work with fibrosis, Insilico researches innovative drugs for cancer, immunity, CNS disease, and age-related diseases. Oncology is one of the fields where it collaborates with Fosun Pharmaceuticals, a major Chinese drug development and manufacturing company.
Conclusion: First AI designed drug.
Due to efficiencies gained from utilizing AI, Moderna and BioNTech developed the two leading mRNA (messenger RNA) vaccines against Covid-19 much more quickly. As the industry leader in drug development, Insilico strives to achieve even more impressive results.
Introduction: First AI designed drug in human trials.
First AI designed drug in human trials, we are on the cusp of AI breaking into the field of drug design to improve healthcare. We have been hearing about the potential of Artificial Intelligence (AI) in improving traditional drug discovery and development for several years. Clinical trials have begun over the past two years.
Exscientia made headlines last April when it announced that it had begun a Phase 1 clinical trial for a drug it developed using AI to treat an established protein target. In Utah, Recursion Pharmaceuticals uses artificial intelligence to find new uses for drugs owned by other companies.
The company has announced the next step: the first Phase 1 clinical trial of a drug developed from scratch using artificial intelligence. AI can be applied to biology to discover targets, and to chemistry to design drugs using its end-to-end platform. The company believes the drug may have anti-aging properties as well as its immediate therapeutic effect.
Last week, AI-backed drug discovery company Insilico Medicine announced that it had dosed the first healthy volunteer in a microdose trial of ISM 001-005.
A small-molecule inhibitor of a biological target was discovered by Pharma AI and designed with the help of AI. The trial is taking place in Australia.
The AI-designed drug is being developed to treat IPF, or idiopathic pulmonary fibrosis, which is a chronic lung disease. IPF leads to progressive and irreversible lung function decline, and it affects 20 out of every 100,000 people worldwide.
Freng Ren, Chief Scientific Officer at Insilico, said in a press release that this drug discovery marks a significant step forward for the AI-drug discovery field. Since the said candidate is the first-ever molecule discovered by AI based on a target discovered by AI, it is a first.
In this new drug, the disease being targeted is idiopathic pulmonary fibrosis (IPF). The cause of this condition is unknown (hence “idiopathic”), and it stiffens the lungs of older people, eventually killing them. It affects approximately 5 million people annually.
According to Ren, the team has utilized an end-to-end AI-powered drug discovery platform along with novel molecules that have drug-like properties to discover biological targets. A machine learning model called Generative Tensorial Reinforcement Learning, or GENTRL, was used to develop the drug. It cost $2.6 million and took less than 18 months for the process to reach the preclinical stage.
This drug is the first of its kind ever to enter the clinic, and we hope that it will be the first of many.
The ISM 001-005 drug has shown promising results in several preclinical studies, according to the company. The studies include pharmacokinetic, safety, and antibacterial testing. According to Insilico, the compound has also shown significant improvement during myofibroblast activation, which results in fibrosis. Potentially, the drug could be used for a wide range of fibrotic indications.
When a drug is tested on humans for the first time at a dosage level required to produce therapeutic effects, this is called a Phase 1 trial. Safety is the primary concern, not clinical efficacy.
Then, drugs must undergo preclinical trials on animals and Phase 0 trials at subtherapeutic doses. Phase 2 and Phase 3 trials test a drug’s effectiveness, first with a sample of a few hundred people, then with a much larger sample of thousands of people.
Insilico’s Phase 1 trial will involve 80 volunteers, half of whom will receive increasing dosages. The goal of the trial is to determine how the drug molecule impacts the human body, a process known as pharmacokinetics.
The breakthrough isn’t just important for people with IPF. It adds to the growing body of evidence showing that AI is a valuable tool for drug development. Human biology is very complex, making drug development extremely difficult. Scientists noticed that pharmaceutical research and development costs doubled every decade or so as far back as the 1980s. It was later called Eroom’s Law, since it was the reverse of Moore’s Law, which reflected exponential improvements in computing.
The year 2022 marks the tenth anniversary of the Big Bang in AI, which happened when Geoff Hinton finally got a machine learning algorithm called backpropagation to work and achieved a famous breakthrough in image recognition. The explosion of interest in AI this decade has been largely fueled by machine learning, especially deep learning, which has enabled the creation of interactive maps, near-omniscient search engines, and machines that can write, draw, and create music.
Alex Zhavoronkov, founder of Insilico, had previously worked in a company producing GPUs, a type of chip that enabled deep learning, and he began exploring the potential for machine learning in the field of drug development at the time of the Big Bang.
The development process
A three-part, AI-powered process led to ISM001-055, a drug now in Phase 1 trials. PandaOmics is the first part, in which Insilico used natural language processing AI to trawl through massive medical data sets, looking for proteins which could be causing fibrosis. This protein is referred to as a target protein.
The disease of fibrosis is closely linked with aging, and Insilico’s deep neural networks were trained to identify potential targets based on millions of data files, including patents, research publications, grants, and databases of clinical trials.
Insilico used Chemistry42, a type of artificial intelligence (AI) called a Generative Adversarial Network (GAN), to identify a drug molecule that could combat the disease. A GAN pits two AIs against each other: one generates suggestions, while the other critiques them.
In the third step, the company used artificial intelligence to monitor the effects of the drug on eight volunteers in Australia. The whole three-part process took 30 months and £2.6m, which is a small fraction of the traditional drug development process, which costs billions of dollars, takes a decade or more, and suffers from a 90% failure rate.
Along with its work with fibrosis, Insilico researches innovative drugs for cancer, immunity, CNS disease, and age-related diseases. Oncology is one of the fields where it collaborates with Fosun Pharmaceuticals, a major Chinese drug development and manufacturing company.
Conclusion: First AI designed drug.
Due to efficiencies gained from utilizing AI, Moderna and BioNTech developed the two leading mRNA (messenger RNA) vaccines against Covid-19 much more quickly. As the industry leader in drug development, Insilico strives to achieve even more impressive results.
Share this: