AI revolutionizing drug testing: Greater accuracy and Biological relevance

Why in the NEWS?

• The US Food and Drug Administration (FDA) has proposed draft guidelines on AI use for the safety and effectiveness of drugs.

Key Points:

• Traditionally, the pharmaceutical industry relies on the response of animal models to assess whether a drug candidate can proceed to human clinical trials,
  • But there is now a growing need to improve the quality of data.
• Over the past decade, submissions by drug manufacturers that include an AI and machine-learning component have grown rapidly.
• In 2021, the FDA recorded 132 submissions that included an AI and/or machine-learning component, a 10-fold increase over the previous year.

What will you read next in this topic?

1. Use of AI in drug development: Benefits and challenges
2. Approval of the use of AI in India
3. Guidelines for Regulators and the Pharmaceutical Industry

Use of AI in drug development: Benefits and challenges

Traditional process of drug development

• The traditional drug development process takes about 10 years and costs more than $1 billion.
• It has a success rate of only 14%.
• Due to the reliance on animal models, the response of drugs to the human body cannot be measured accurately.
• Animals such as mice respond differently to drugs than humans, which leads to adjustments in results.

Impact of AI in drug development

• Emerging technologies such as AI and machine learning can improve the speed and effectiveness of drug development.
• AI helps in the selection of compounds, allowing potential candidates to be identified quickly.
• This can be helpful in finding safe and effective compounds for human clinical trials.
• AI can predict drug response for children and other vulnerable populations that are difficult to test ethically.

Benefits of using AI in drug development

• With the help of AI, unexpected side effects of drugs can be identified in advance, which helps to avoid potential risks.
• An AI-based "safety toolbox" can predict the undesirable effects of chemical compounds. This is helpful in identifying the side effects of the drug on different organs of the body.
• Computer-based toxicological models help predict the effects of the drug on the whole body or specific organs, thereby ensuring drug safety.
• AI can be used to assess the effectiveness of drugs, ensuring that drugs act correctly on the human body.
• The use of AI helps to determine the appropriate dosage of the drug, which is more effective and safe based on the response of the individual body.
• AI uses machine learning algorithms to help identify new and effective drugs, thereby accelerating the drug development process.
• AI can also be used to monitor the safety of drugs after marketing. AI models can provide timely information about adverse effects, allowing drug manufacturers to take timely action.
• With the help of AI, drugs can be created as custom medicines, individually suited to different patients according to their genetic structure and medical history.
• AI is used in preclinical trials, allowing the results of animal tests to be more accurately linked to the response of the human body. It also reduces the reliance on animal tests.
• AI can make the drug development process faster and cheaper, as AI models can quickly identify potential drugs by analysing large data sets, saving time and resources.
• AI is able to recognize and understand complex patterns, allowing the effects of drugs to be measured very accurately. AI can advance drug development by identifying new patterns from old or existing data.
• AI can analyse multi-dimensional data such as genome data, laboratory results, and a patient’s medical history to better predict individual drug effects.

Challenges of AI

• The reliability of AI models depends on the quality of the data they are trained on.
• This problem is evident in the saying “garbage in, garbage out” – if the data is biased or less representative, it has a negative impact on the results.
• Lack of transparency is also a major challenge, as the internal workings of most AI models and the data used to train them are not opened to independent scrutiny.

FDA’s Draft Guidelines

• The FDA has released draft guidelines for the use of AI, which provide a framework to assess the reliability of AI models.
• These guidelines ensure that risks posed by AI are identified and that incorrect predictions do not lead to life-threatening consequences.
• It also states that AI models require continuous monitoring and maintenance throughout their lifecycle.

Potential uses of AI in drug development

• The use of AI is most important in the preclinical testing phase of a drug, where it can ensure that a compound is safe enough for human trials.
• This can free drug manufacturers from relying on animal trials and ensure its effective use.

Approval of the use of AI in India

New Drugs and Clinical Trials (Amendment) Rules 2023

• India passed the New Drugs and Clinical Trials (Amendment) Rules 2023, which allow advanced computational models to be used to assess the safety and efficacy of drugs. 
• This move is significant towards promoting technological improvements in the drug development process.

Reducing dependence on animal tests

• This amendment will increase the use of AI and computational models in drug development, which can reduce the dependence on traditional animal tests. 
• This will not only make the drug testing process faster and accurate, but it will also reduce the suffering of animals.

Promoting the use of AI

• This move to promote the use of AI in India will help make the pharmaceutical industry more modern and accurate. 
• AI will be used to predict the efficacy, safety and potential side effects of drugs, ensuring the safety of patients.

Smart and efficient drug development

• These regulations support the use of AI technology, which can make the drug development process in India smarter, efficient and time-saving. 
• AI-based models can analyze complex data, allowing new drugs to be identified and tested for efficacy quickly and accurately.

Matching global standards

• India’s move matches global standards, such as the European Medicines Agency and the FDA have also accepted the use of AI in the drug development process. 
• This will give India an opportunity to compete on the global stage of drug development.

Innovation in research

• The use of AI will help researchers understand the complexities of drug effects, genomic data and ethical trials in new ways, opening up new areas of drug development.

Guidelines for Regulators and the Pharmaceutical Industry

FDA Guidelines

• The FDA (Food and Drug Administration) guidelines are important to ensure transparency and quality of the use of AI in drug development. 
• These guidelines ensure that AI models are used under standard procedures and that the results are reliable.

Objectives and Benefits

• The main objective of these guidelines is to improve policy, manufacturer expectations, and consumer safety. 
• This is to ensure that the AI ​​models used by drug manufacturers are safe and effective, and do not pose any risk to patients.

Risk Assessment as Required

• The guidelines also point out the importance of assessing the risk posed by AI models. 
• If an AI model underestimates the risk for a drug, it can be life-threatening due to an incorrect prediction. 
• Therefore, risk assessment is an important step in these guidelines.

Transparency and Monitoring

• The FDA guidelines have a special focus on the transparency of AI models. 
• These models are not open to independent scrutiny, so verifying their functioning and training data can be challenging. 
• Because of this, the need for continuous monitoring and maintenance throughout the lifecycle of an AI model has been recognised.

European Medicines Agency and ICH Guidelines

• The European Medicines Agency and ICH (International Council for Harmonisation) have also issued similar guidelines on the use of AI. 
• This ensures that there are uniform standards for the use of AI in the pharmaceutical industry globally, and that there are similarities in the drug development process across different countries.

Need for system improvements

• Improving the drug development process is possible using AI, but it is imperative to follow proper procedures under the guidelines. 
• This can help policymakers maximise the benefits of AI and avoid potential risks.