This was the main topic of discussion at the 11th Annual Global Summit on Regulatory Science, where regulators from Brazil, Canada, India, Italy, Japan, Germany, Switzerland, Singapore, the UK and the US RWD to agency operations and regulatory mechanisms. The conference will be held virtually in October 2021 and sponsored by the Global Coalition for Regulatory Science Research (GCRSR).
The proceedings were recently summarized in a journal by Shraddha Thakkar, Ph.D., MSc, MSc, from the Center for Drug Evaluation and Research (CDER) at the U.S. Food and Drug Administration (FDA), and colleagues from regulatory agencies in the above countries. Regulatory toxicology and pharmacology.
In a series of discussions, workshops, and presentations, regulators will discuss how AI and RWD can be applied to food and drug safety assessment, whether regulatory science is ready for the arrival of AI, and how data science tools can be applied to regulatory applications. We discussed how to make it better fit. The future of regulatory science research.
According to the authors, “Continued advances in AI and RWD will help us in two key aspects: improving agency operations and preparing regulatory mechanisms for reviewing and approving products using these innovations. “This is especially important for drug development, which typically takes many years and is very costly. AI and RWD will help improve drug safety and review. are demonstrating.”
Noting that the regulator sees AI and RWD as having potential for food safety, pattern recognition, and foodborne outbreaks, it “mainly uses imaging, spectroscopic data, genomic data, chemical It relies on specific manual analysis of composition and contaminants.” said the author. AI and machine learning (ML) have the potential to reduce review time and human variability in manual processes. In many ways, AI and RWD already exist, and agencies such as the FDA and the Canadian Food Inspection Agency are incorporating AI and RWD methodologies into their existing programs. AI and RWD also serve as extensions of existing information aids. For example, Swissmedic is considering developing automated pharmacovigilance signal detection using severe side effects on admission as his RWD. Another example is the crowdsourcing used by the National Institute of Health Sciences of Japan to develop a quantitative structure-activity relationship model for Ames mutagenicity prediction.
In two debates, presenters argued that regulators may or may not be ready for AI and RWD advances in the realm of scientific knowledge and assessment practice. . One presenter argued that “AI is playing an ever-growing role” in drug discovery and development, and some regulatory agencies, such as the FDA, are preparing for innovative scientific and technological approaches ( ISTAND) initiative to develop programs such as: Other considerations discussed were the role of AI in clinical applications and how comfortable patients are using AI-enabled applications in different situations.
“Regulatory science could play an important role in developing regulatory structures and frameworks for the evaluation of AI applications, including promoting the credibility and reliability of these technologies.” The author writes
Another opportunity for AI, Thakkar and colleagues noted, was a workshop where regulators detailed data analysis tools. AI has the potential to automate the process of manually reading texts related to food and drug safety and efficacy. “The majority of the data used for regulatory decision-making is presented in text documents, and he said AI could be important to expedite the review process,” they wrote. “Regulators around the world have not only reviewed a vast amount of submitted applications, papers and/or literature data, but they have also created a large amount of documentation during the product review process. are typically unstructured texts and often do not follow the use of standard vocabularies.”
Lacking data standardization and fragmentation, Thakkar and colleagues explain that leveraging AI to interpret datasets is a “substantial regulatory challenge.” “The biggest challenge facing the research community is the current fragmentation of data in many repositories with multiple formats and definitions,” they said. “Another challenge is that sometimes data codes are not uniform. Each data source has a coding system and different methods of assigning codes to medicines are adopted without national or international standardization. ”
The future of regulatory science research related to AI and RWD is one in which AI augments the work of human clinicians, but does not replace them. “One of the most important benefits of AI/ML is its ability to learn from real-world usage and improve performance,” the authors say. But because AI is an emerging technology, they say, “it needs to be constantly evaluated to actively promote the use of these new tools in a regulatory environment.”
Regular Toxicol Pharmacol
