Artificial Intelligence (AI) and machine Learning (ML) have been on the rise in the medical field over the past decade, with many studies and papers discussing their potential to diagnose, manage, and treat a wide variety of medical conditions. These rapid advances in AI technology present particularly promising opportunities in clinical research for developing new and effective treatments for various diseases, a process normally fraught with failure. To illustrate, between 2000 and 2015, nearly 86% of all drug candidates developed were unable to reach their objective endpoints.
The drug discovery process is often long, complex, and expensive for the sponsor and contract research organization (CRO) researchers. Therefore, AI-based drug development has the potential to significantly reduce the time and cost involved in drug discovery, increase the success rate of drug development, and lead to the discovery of new treatments.
In this article, we will explore the role of AI technology in drug discovery and what, if any, regulatory approvals have resulted from AI-based drug design to date.
What Are AI-Designed Drugs?
AI-designed drugs are a class of medications that are developed using AI technology at any point during the drug discovery process. These types of drugs work by leveraging the power of ML algorithms to identify potential drug targets and design molecules that can interact with those targets. What makes this possible is the ability of AI algorithms to analyze large amounts of data, including genetic and protein data, to identify potential drug targets.
One of the main advantages of AI-designed drugs is that they can significantly reduce the time and cost invested in drug discovery and development by sponsors and CROs. Approving a new drug developed using traditional drug discovery and development methods can take up to 10-15 years and cost billions of dollars.
AI technology could reduce this timeline by identifying potential drug targets and designing molecules for these targets in a fraction of the time it would take using traditional screening and testing methods. Additionally, by utilizing AI technology to predict the efficacy and safety of drug candidates before human testing, researchers could see an increase in the success rate of drug discovery, particularly for orphan or rare diseases with no existing treatment options.
The Role of AI Technology in Drug Discovery
The process of drug discovery involves identifying a target, designing a molecule that can interact with the target, optimizing the molecule to make it more effective, and assessing the molecule for safety and efficacy. Integrating AI technology into the pharmaceutical drug development pipeline can help expedite each of these steps. In the early stages of drug discovery, AI technology can be used to analyze large datasets, including genetic and protein data, to identify potential drug targets. ML algorithms often identify patterns in the data that humans may miss, which could lead to the discovery of new targets that were previously unknown, or that were not considered to be viable targets.
Once a target has been identified, AL/ML algorithms can be used to design molecules that interact with the target; this process typically involves generating a large number of potential molecules and using these algorithms to select the best candidates. Using technology, researchers can optimize the best candidates by predicting their efficacy, toxicity, pharmacokinetic properties, as well as safety in vivo prior to the start of clinical testing.
Current State of AI-Based Drug Development
Today, there are numerous applications in existence leveraging the power of AI to streamline the drug development process. For example, within the drug design stage, generative models can learn the underlying patterns in chemical datasets to generate new molecular structures with specific properties.
Alternatively, DeepMind’s AlphaFold, Meta’s ESM (Evolutionary Scale Modelling) Metagenomic Atlas, RoseTTAFold, and other similar technology have become popular in the world of 3D structure prediction for drug discovery. Read more about the impact of ESM and AlphaFold on the pharmaceutical industry by visiting Vial CRO’s blog.
In 2018, AlphaFold represented a significant breakthrough in AI-power drug discovery, generating sustained excitement across the industry about its potential to accelerate this phase of the pipeline. However, as of August 2023, no AI-derived clinical-stage programs have come from a drug candidate discovered by AlphaFold specifically, nor have any AI-designed drugs received approval from the United States Food and Drug Administration (FDA) to reach the market.
AI-Designed Drugs FDA-Approved for Clinical Trials
While AI-created drugs have not gained FDA approval to enter the market, encouraging developments have materialized within the clinical trial landscape. As of March 2022, Boston Consulting Group reported that biotechnology firms using AI as a primary strategy had over 150 small-molecule drugs in the discovery phase, with over 15 already advancing through clinical trials. Below, we list two prominent drugs that were FDA-approved for clinical trials.
1. DSP-1181
Despite the nascent stage this type of drug discovery and development is in, the first drug designed with the help of AI technology to enter clinical trials was DSP-1181, announced in January 2020. DSP-1181 is a drug created by British start-up Exscientia and Japanese pharmaceutical firm Sumitomo Dainippon Pharma that is used to treat obsessive compulsive disorder. The compound was discovered by algorithm-based candidate screening and managed to reach clinical testing in 12 months, as opposed to the typical 5 years seen with traditional methods. However, the drug did not progress past Phase I. In July 2022, it was discontinued as it did not achieve the evaluation criteria during its Phase I study.
2. INS018-055
Insilico Medicine, a biotech company headquartered in Hong Kong, has created the world’s first AI-designed anti-fibrotic small molecule inhibitor drug to be tested in human patients. Distinguishing itself from other AI-driven medications currently undergoing testing, INS018-055 was both discovered and designed using AI. INS018-055 has already completed initial testing and early phases (NCT05154240).
In June 2023, Phase II, double-blind, randomized clinical trials began. As of August 2023, the trials are currently being conducted in both the U.S. and China, to investigate the safety, tolerability, pharmacokinetics, and efficacy of INS018_055 administered orally in subjects with idiopathic pulmonary fibrosis (IPF).
At the time of writing this article, there have been no FDA-approved AI-created drugs, despite promising advancements witnessed in the clinical trial landscape. Although notable developments have been observed with INS018-055, only time will tell if it will reach the market.
Challenges and Opportunities for AI-Designed Drugs
Despite its promise, the implementation of AI/ML into the drug discovery pipeline faces many obstacles. One possible reason why AI-designed drugs have yet to reach FDA approval could be the lack of sufficient data. Developing a new drug requires extensive testing and clinical trials to ensure its safety and efficacy; while AI algorithms can help identify potential drug targets and design molecules that can interact with those targets, they cannot replace the need for clinical trials.
As the FDA requires a significant amount of data from clinical trials before it will approve a drug, many AI-designed drugs are still in the early stages of development or clinical testing. It will take time to collect and analyze the data needed to demonstrate the safety and efficacy of these drugs.
Additionally, there may be issues with transparency surrounding these software programs. The FDA requires a thorough understanding of the mechanism of action of a drug before it can be approved, but depending on the complexity of the AI algorithm used, this may not be clearly understood yet. Without a definite picture of how a drug works, predicting its safety and efficacy in humans becomes more challenging.
Lastly, stringent regulatory requirements posed by the FDA could also be a possible obstacle slowing AI-designed drugs from reaching the market. For example, drugs discovered by AI technology may require new manufacturing processes or quality control measures that need to be validated prior to approval.
Despite these challenges, we can expect to see more AI-powered drugs moving towards approval by regulatory agencies like the FDA in the future as a result of advances in digital technologies. Possible opportunities for sponsors, CROs, and other researchers to integrate AI into the drug discovery process include target identification, molecule design, and optimization.
Past preclinical testing, AI technology can assist by predicting a candidate’s safety and efficacy before clinical trials, as well as supporting precision medicine by analyzing patient data for specific biomarkers to determine whether a patient will respond to a drug or not.
Vial CRO: An Emerging Leader in AI-based Drug Design
Vial is a tech-first CRO delivering faster, better, and more affordable clinical trial results for biotech sponsors. Battery Bio is Vial CRO’s newest drug discovery arm, a modern approach to drug discovery to transforming the pharmaceutical industry. Our platform combines AI-designed drugs, fully automated labs, and hyper-scaled trials to create a more targeted and efficient approach to drug development that can turn AI designs into reality in days.
We fundamentally believe that a dramatic reduction in the cost of trials combined with a high-velocity clinical strategy will transform medicine and enable scientists to bring curative therapies to patients at a dramatically faster rate. By leveraging Vial’s 90%+ lower cost structure, we are able to take more radically more shots on goal, redefining drug discovery with a systems engineering approach.
To learn more about what Battery Bio or Vial’s CRO can do for you, visit our website or contact a Vial representative today!
