During the June 27 Drug Information Association Global Annual Meeting, Hao Zhu, Director of the Division of Pharmacometrics in the US Food and Drug Administration’s (FDA) Center for Drug Evaluation and Research (CDER) reported the fact that there were 170 new drugs and biologic submissions that incorporated artificial intelligence (AI) and machine learning (ML) elements in 2022. The previous year there were 132 applications that included AI and ML elements. It was only five years ago that there were none.
This increase in applications comes at a time when researchers are realizing that AI can play a substantial role in the drug development process. Pharmaceutical companies are using the tool for nearly every step in the discovery process including constructing small molecules, determining appropriate dosages, discovering active components, reclaiming drugs, predicting toxicity and bioactivity properties and identifying the mechanism of action (MOA) of the drug.
Drug discovery today – the challenges
The introduction of AI couldn’t have come at a better time as the drug industry, like other industries, faces high inflation, talent shortages, pressures on consumer spending and rising capital costs. In addition to these economic challenges, the industry is being pushed to consider reducing the duration of the time-to-patient process.
Companies are also being called to bring down development costs, which can easily skyrocket into the billions of dollars range due to the high failure rates of new drug candidates. One 2022 study points to the fact that more than 90 percent of all clinical drug development – which takes on average ten to 15 years to begin with – fails.
Finally, there are numerous regulatory compliance requirements that are often seemingly nearly impossible for companies to keep up with, Seema Sayani, Ph.D., Senior Director of Life Sciences at Cognizant explains. “There has been an exceptionally high pace of change in the global regulatory landscape in the past five years, which translates into approximately one significant change per day,” she says. “The vast array of variables presents an immensely intricate business challenge, warranting the use of sophisticated AI solutions.”
Bringing AI into the equation
Forward-thinking companies are integrating AI across all phases of drug development, such as drug discovery, which includes foreseeing protein-protein interconnections, determining protein conformation and distortions, conducting structure and ligand-based virtual screening, employing quantitative structure-activity relationship (QSAR) modeling. It has been established that employing AI in the drug discovery process can bring about significant transformation. This includes precise prediction of pertinent physiochemical processes and ADMET (Absorption, Distribution, Metabolism, Excretion and Toxicity) properties that can minimize drug attrition rates by detecting and addressing any toxicity and safety concerns that may arise.
In the past, integrating AI into the development process was very challenging. The tools were complicated and required advanced knowledge of coding. Today, however, there are AI tools available that are automated and easy to use, making AI accessible to everyone involved in the drug discovery process including medicinal chemists and pharmaceutical scientists who may not be well-versed in the technology. The tools feature ready-to-use formats such as pre-trained, pre-configured models, frameworks and drag-and-drop AI pipeline-building platforms.
This is significant since the benefits of AI and ML in drug discovery are transformative. For instance, AI has shortened the time to review drug outcomes. One study in the March 2022 issue of Drug Discovery Today, for instance, profiles Pfizer’s success using AI. The study found that the pharmaceutical company, using AI and other industry-leading techniques and strategies, achieved an end-to-end clinical success rate of 21 percent – well above the 11 percent average success rate of its peers.
Sayani states that, although AI is still relatively nascent, integrating the technology into existing drug discovery initiatives will rapidly enhance the pipeline and unlock new opportunities in the market for those organizations that proactively leverage it.
“Keeping pace with the leading industry trend of democratizing AI and leveraging an AI-driven integrated discovery and analysis platform can effectively optimize the drug development process and facilitate decisions resulting in improved outcomes,” she says.
The types of AI-driven platforms that are making this happen can integrate various public, commercial and custom catalogs for virtual screening. This is because of built-in technology for analysis and visualization and includes advanced features like generative AI-based models for expedited generation of novel molecules. The most important benefit is the accessibility of data across all stakeholders for seamless collaboration, Sayani says.
“If you use an AI-driven drug discovery platform that stores all your previous work when you start researching something, it says, ‘Hey, this has been tried before, so you don't have to try it again,’ and lets you move on to the next step sooner,” she says. Sayani likens an AI-driven drug discovery platform to a “prime mover,” (source of propulsion) that links and unleashes the combined power of the drug discovery ecosystem, providing insights, predictions and solutions that were previously challenging to obtain.
“When pharmaceutical companies can extend the power of AI usage across labor-intensive drug development processes, such as clinical trials, then they can truly drive efficiencies across the drug development lifecycle,” Sayani says.
Taking drug discovery into the future
AI is quickly transforming the drug discovery process. With its ability to process vast amounts of data and identify patterns, it can significantly reduce the time and cost involved in developing new drugs. Moreover, the accuracy and efficiency of AI-powered drug discovery can lead to breakthrough treatments that may not have been possible through traditional methods. While there are challenges -- such as ethical considerations, data privacy issues and a lack of standardization -- the potential benefits of AI in drug discovery are too compelling to overlook. As we move forward in this exciting era of medical research, we can expect more AI-powered drugs to make their way to the market.
To learn about Cognizant’s Life Sciences practice, please visit cognizant.com/lifesciences.