AI (Artificial Intelligence) pharmaceuticals refer to the use of artificial intelligence technologies such as big data, cloud computing, and deep learning to assist in drug discovery, drug management, efficacy analysis, and other processes, with the aim of shortening trial cycles, saving costs, and improving trial success rates.

Since the first proposal of AI technology in 1956, as of 2022, according to Deep Pharma Intelligence, nearly 800 AI pharmaceutical companies have been established worldwide, with a total investment of tens of billions of dollars. Today, this trend has not yet faded.

In August alone this year, 15 AI pharmaceutical companies announced the completion of a new round of financing. Recently, multinational pharmaceutical giant Novo Nordisk announced a cooperation agreement with AI pharmaceutical company Valo Health, with a total amount of more than 2.7 billion U.S. dollars. With such momentum, can AI pharmaceuticals take off?

MNC ramps up AI pharmaceutical layout

On September 26, Novo Nordisk reached a cooperation agreement with Valo Health, and Novo Nordisk will use Valo's large human data set and AI-driven calculations to discover and develop innovative therapies for heart metabolic diseases, with a total cooperation amount of 2.7 billion U.S. dollars.

Specifically, the cooperation between Novo Nordisk and Valo will use the key joint capabilities of human data and genetics to develop innovative therapies for heart metabolic diseases. Valo can identify and validate preclinical new drug targets, develop candidate drugs targeting these targets, and help predict the safety and efficacy of compounds. Novo Nordisk and Valo also plan to use the latter's Opal computing platform to discover and develop three preclinical drug discovery projects for cardiovascular diseases.

The Opal computing platform is a comprehensive, end-to-end drug discovery and development tool that relies on high-quality, proprietary, and differentiated data resources and specialized artificial intelligence technologies to redefine the drug development process from start to finish.

In fact, many multinational pharmaceutical companies have made AI pharmaceuticals an important part of their strategic layout. AstraZeneca has been cooperating with AI and data-driven companies since 2017. In 2019, AstraZeneca reorganized its structure and created a data science and AI department, which is a team composed of data scientists, bioinformaticians, machine learning experts, etc., aimed at enhancing basic research, drug development, and other related research through data and artificial intelligence algorithms. With the assistance of AI pharmaceuticals, AstraZeneca has developed new drugs and new targets, and improved the success rate of drug development.

In September this year, Alexion, a subsidiary of AstraZeneca, reached a cooperation agreement with Verge Genomics. Verge Genomics is a biotechnology company that uses artificial intelligence to develop therapies for neurological diseases. The two jointly developed new drug targets for neurodegenerative diseases and neuromuscular diseases.

In addition, AstraZeneca is also actively deploying AI diagnosis and disease management to enhance its capabilities in precision medicine and chronic disease management. These measures have made AstraZeneca a leader in the field of AI pharmaceuticals.

Similarly, in 2017, GlaxoSmithKline (GSK) reached a cooperation agreement with Insilico Medicine to jointly develop new biological targets and molecules. The following year, GSK established a powerful AI/ML (machine learning) team of 120 people. In 2020, GSK established a dedicated research center to discover new drugs for cancer and other diseases using AI.

Also in 2020, GSK, NVIDIA, AstraZeneca, and the UK National Health Service (NHS) established a new partnership to build the UK's most powerful supercomputer and use it for AI research in healthcare.

In April 2022, GSK reached a cooperation agreement with PathAI to advance clinical trials and drug development for oncology and NASH (non-alcoholic fatty liver disease). PathAI mainly develops AI-driven digital pathology tools to analyze tissue samples and identify specific biomarkers.

In February 2023, GSK reached a cooperation agreement with Cytel, a US biostatistics CRO company. The two will combine algorithms with cloud computing capabilities to provide the ability to quantify scientific and business trade-offs through visualization.

GSK's layout in the field of AI pharmaceuticals emphasizes its leading position in the healthcare industry. Through innovative technologies and strategic partnerships, it will provide forward-looking solutions for future drug development and medical treatment.

In addition to AstraZeneca and GSK, multinational giants such as Roche, Pfizer, Takeda, Novartis, BMS, Eli Lilly, and Merck are also deploying AI pharmaceuticals.

Can AI and pharmaceuticals collide to spark "fireworks"? 

China is also actively exploring the application of AI in drug discovery. China plans to become the world leader in artificial intelligence by 2025. The National Development and Reform Commission has identified artificial intelligence in the pharmaceutical industry as a key area in the "Made in China 2025" plan.

Currently, there are nearly 80 start-up AI pharmaceutical companies in China, among which Jing Tai Technology and Insilico Medicine have completed D-round financing and become the leaders in the domestic AI pharmaceutical field.

On September 12 this year, Insilico Medicine also achieved the first AI drug "going global". Insilico Medicine and Exelixis announced that they have signed an exclusive license agreement, and Exelixis will obtain global development and commercialization rights for ISM3091 and other compounds targeting USP1 (synthetic lethal target deubiquitinase) from Insilico Medicine; Insilico Medicine will receive a prepayment of 80 million U.S. dollars, as well as milestone payments based on subsequent development, commercialization, and sales.

ISM3091 is a small molecule inhibitor targeting USP1 for BRCA-mutated tumors. It is the third clinical-stage project discovered under the company's AI-generated platform Chemistry42. Using AI research and development, the drug development cycle was greatly shortened. ISM3091 took only a little over a year (from April 2022 to July 2023) from nomination to the clinical stage.

Preclinical data shows that ISM3091 has strong anti-tumor activity, good tolerability, and ideal pharmacokinetic characteristics. The excellent data contributed to the first AI drug "going global" by Insilico Medicine.

The above example shows that AI pharmaceuticals have great development prospects. Compared with the traditional drug discovery process, AI can shorten the drug development cycle in the preclinical stage. According to Exscientia's data, AI pharmaceuticals can save an average of 40-60% of drug synthesis time, reduce R&D costs, and increase success rates by 12-14%. AI pharmaceutical technology can currently save about $55 billion annually for compound screening and clinical trials worldwide.

In particular, AI pharmaceuticals can be effectively used in different stages, such as drug design, chemical synthesis, drug screening, and drug reuse, thereby promoting the widespread transition to data-centric drug discovery.

Although AI pharmaceuticals have many functions, they are still in the early stages of research and development, and there are still many problems to be solved. For example, AI pharmaceuticals face regulatory and ethical issues and require the establishment of regulatory constraints and standards; AI drug development requires multidisciplinary collaboration, and there is a lack of comprehensive talent; AI pharmaceuticals cannot achieve high-quality data sharing, etc. In summary, AI pharmaceuticals still have a long way to go from technology to market.

However, in the future, AI technology will undoubtedly profoundly change the drug development model.

About  the Author:    

Yefenghong

Ye  Fenghong, a medical editor specializing in oncology at a healthcare internet  company, has conducted in-depth research on the pathogenesis and clinical  treatment of lung cancer and breast cancer. She has previously been involved  in the design and synthesis of anti-tumor drugs and has some experience in  computer-aided drug design. She is currently devoted to introducing  cutting-edge cancer treatment drugs to a wide range of readers, aiming to  help more people avoid cancer pain and embrace good health.