The Importance of AI-Based Drug Research to Develop Effective Therapies
Many of us interact with artificial intelligence (AI) in our daily lives, from searching online and receiving personalized shopping recommendations to booking travel, making reservations and asking questions to digital assistants in our homes. The use of AI has grown significantly due to its ability to perform complex tasks quickly and solve problems without human intervention.
AI’s ability to quickly analyze and synthesize huge volumes of information is also being embraced for many healthcare initiatives, including use in pharmaceutical drug discovery and development.
The traditional process of drug discovery and development of a single new drug can take more than a decade and cost billions of dollars. Most drug candidates end up failing to show efficacy in clinical trials and big pharma often abandons investments in early development of drug candidates even before the clinical trial stage due to cost. As we learn more about human diseases and the complexity of human biology, it’s clear that we need more sophisticated tools.
Benefits of AI in drug development
Using AI in drug development has many benefits, including the potential to accelerate the pace of discovery while reducing the need for extensive lab work and shortening the clinical trial phase.
AI is also being used to research and find existing drugs that might be effective for newly discovered diseases. During the rapidly escalating and deadly Covid-19 pandemic, researchers applied AI to assess existing drugs that could be repurposed to treat Covid-19, alone or in combination with other drugs, without causing serious complications.
Despite recent interest, the use of computer and data science has not evolved as rapidly in basic science research and clinical trials as in other sectors; however, evidence of its benefits is mounting.
Here are some of the potential benefits of using AI in drug development:
- Target identification: AI can be used to filter large amounts of structured and unstructured data to improve understanding of a disease and more accurately identify relevant drug targets.
- Generation of lead compounds: AI can help filter trillions of molecules to identify the most promising; or predict protein structures and interactions to increase the likelihood that compounds will be effective.
- Predict efficacy and safety: AI can be used to predict absorption, distribution, metabolism, and excretion (ADME) properties in silico or to assess whether a drug compound that works in animal models would work in humans and also predict safety issues.
- Patient selection: AI can help determine which populations would most likely respond to a therapy to develop the right inclusion and exclusion criteria and biomarkers.
- Preventing patient dropout in clinical trials: AI can be used to develop “patient coaches”, i.e. personalized digital messages to support patients during the trial to encourage their continued participation and to ask for feedback. The AI can also give personalized answers to participants’ questions.
- Find ideal sites to run clinical trials: Since some sites have difficulty recruiting or enrolling patients in trials, AI can be used to understand these demographic variables and see what trials competitors are running on these same Site (s.
Use of AI in clinical trials
Importantly, AI, in combination with computational biology and modeling, can be used to form a synthetic control arm of a study to compare investigational drugs to standard of care, to other drugs or combinations of drugs. This is a relatively new and growing field known as “in silico” clinical trials, where disease-specific computer models train virtual cohorts to test safety and/or efficacy new drugs.
Using in silico trials, we can integrate different disease, age and gender parameters, for example, one at a time and in different combinations, to test the potential efficacy of a drug among thousands of virtual patients . We have the potential to examine thousands of variations. We can test multiple compounds, which cannot be done in traditional clinical trials. Therefore, we can achieve statistically significant results with fewer failed trials.
AI can also mitigate the need for animal testing at some levels. In silico, translational medicine modeling gives drug developers a better idea of how a drug will react in the human body rather than seeing how animals react and hoping the same effect will translate to humans. . Nearly 90% of promising drugs in animal research are not safe or effective in humans.
The ability of AI to reduce the cost of drug development
Finally, there is the cost factor. We know the expense involved in developing new and better therapies. By being able to design drugs faster with better efficacy and fewer side effects, pharmaceutical manufacturers can develop and eventually sell drugs more affordably.
I think in five to ten years of increased use of AI in drug development and testing, we will see a completely changed economic structure in pharmaceutical development. This is particularly important for narrowly targeted therapies and therapies for orphan diseases – those that affect smaller numbers of patients, where the investment in drug development may not be balanced with the rate of return.
The increasing use of AI will also create opportunities for new small companies to succeed in the field of drug development.
The refinement and use of AI in drug development will be essential in order to pursue a patient-centered approach to care, improving and extending the lives of people – even those with rare but serious disorders for which it is not possible. There is no significant financial incentive to devote the time and resources needed to develop effective new therapies in the traditional way.
Photo: Metamorworks, Getty Images