Objective: To enhance the quality and efficiency of drug review by leveraging artificial intelligence (AI), facilitate the connection between drug review and AI application in drug research and development (R&D) stages, identify potential risks in drug access procedures, free up reviewers' time from non-technical evaluation tasks, and tilt limited review resources towards clinical urgent innovative technologies and the development of new technical standards, thereby better guiding the innovation of pharmaceutical industry and protecting and promoting public health. Methods: Combining practical work, some application scenarios for AI in drug access procedures were proposed, and the current risks and challenges were analyzed. Some insights and recommendations on regulatory frameworks, data quality, and other aspects were provided. Resultsand Conclusion: While AI has been widely and successfully applied in compound screening, target discovery, and computer-aided drug design, its use in supporting drug access remained at an early exploratory stage. Its inherent complexity, dynamic nature, and heavy dependence on data posed significant challenges to its application in existing regulatory frameworks. Nevertheless, the effective application of AI in drug access procedures holds great promise for improving the quality and efficiency of drug review.

Objective: To explore the application potential and challenges of artificial intelligence (AI) in drug clinical trials, analyze relevant regulatory practices and trends in the United States and the European Union, and provide references for the application of AI in China's drug clinical trial field. Methods: Searched for current regulations, normative documents, and technical guidelines in the United States and the European Union, as well as cases and related literature published on the official websites of regulatory agencies. Discussed the application potential and challenges of AI in drug clinical trials, along with regulatory practices and insights from Europe and the United States. Resultsand Conclusion: AI has gradually begun to be applied in key aspects of drug clinical trials, but it faces issues such as data quality, model performance, ethical risks, and regulatory lag. By summarizing and analyzing relevant regulatory activities in the United States and the European Union, as well as the application of AI in regulation, we conclude that future regulation may focus on ethical compliance and patient safety, establish continuous supervision and accountability systems, and promote standards for trustworthy AI, data governance, and model evaluation. When deploying AI systems in key clinical trials, it is essential to clarify target scenarios, assess decision-making weights and consequences, ensure reproducibility and traceability of results, and conduct training. Collaboration and education play important roles in regulatory practices. Based on these findings, propose recommendations for China, including conducting regulatory science research, strengthening data governance standardization, and promoting the innovation of regulatory tools, to facilitate the standardized and healthy development of AI in China's drug clinical trials.

Object: Driven by continuous improvement of computing power, artificial intelligence (AI) models, with their outstanding data analysis and processing capabilities, are gradually transforming drug development through advanced data analysis, offering new approaches for improving drug quality. The traditional pharmaceutical regulations struggle to address AI's deep integration into manufacturing, failing to meet Good Manufacturing Practice (GMP) requirements. Current implements and potential research directions of AI models in pharmaceutical industry are discussed in order to provide theoretical frame and practical basis for future construction of the regulatory framework that aligns with the characteristics of AI models. Methods: The current applications of AI models in the pharmaceutical quality systems were explored, and the regulatory challenges in process control, risk monitoring and decision-making were analyzed, providing foundational insights for developing AI-specific regulatory frameworks. Resultsand Conclusion: Challenges in data reliability and traceability, model transparency and interpretability, dynamic adaptability and verifiability are discussed. Potential research directions from the aspects of artificial intelligence model validation requirements, large model data management standards, and hierarchical management of artificial intelligence models are explored. We aim to provide theoretical references and practical bases for future research on artificial intelligence model regulatory strategies.

Objective: Amid the global wave of transformation in the pharmaceutical industry, advanced manufacturing technologies (AMTs) such as continuous manufacturing and Process Analytical Technology (PAT) have become the core drivers in developing new drug production models, where a data-driven approach is central to this transformation. However, certain aspects in the digitalization of advanced manufacturing in China still require further strengthening and refinement. We aim to explore a systematic approach for promoting the coordinated development of digitalization and standards in China's biopharmaceutical advanced manufacturing through research on the synergy between standardization and AMTs digitalization, providing insights to support industrial advancement. Methods: Firstly, the challenges in the digitalization of advanced manufacturing in China's pharmaceutical field were analyzed. Secondly, the approaches taken by governments in the United States, Europe, and Japan to strategically guide standardization through national strategies were examined. Finally, the successful experiences of international authoritative standards organizations in the coordinated development of standards in relevant fields, and the implementation pathways of mature universal standards were deeply analyzed and discussed. Resultsand Conclusion: It is proposed that standards serve as the “common language” linking technical rules and industrial ecosystems. The synergistic innovation of standards with advanced technologies represents a key development direction. Establishing a standards system across three dimensions: universal foundational common standards, key technology standards, and industry application standards, and conducting standardization work, are of significant importance for systematically advancing the high-quality development of digitalization and standard synergy of China's biopharmaceutical advanced manufacturing.

Objective: The application of patent big data analysis in the field of drug regulation and decision-making was explored by using patent data as a source of technical information. Methods: Taking transdermal patches as an example, a patent search expression was constructed to obtain relevant patent data. For the structured data part of the patent, descriptive analysis (statistical analysis, trend analysis) and other methods were adopted; for the unstructured data of the patent, text mining was used, and the LDA topic analysis model and TF-IDF word frequency statistics were employed; for the citation network relationship between patents, the main path analysis method and K-means clustering algorithm were adopted. Results: Through patent data analysis, information on the subject categories of transdermal patches, technological development trajectories, and technological evolution trends were obtained. Based on the in-depth understanding of the technological trajectory and development trend, three typical application scenarios were established: first, to provide a global perspective and build a technological panorama; second, to conduct trend prediction and hotspot mining based on the path of technological evolution; third, to refine and form the technological risk points based on the characteristics of specific technologies, to form the regulatory concerns and regulatory considerations. Conclusion: For the first time, we associated the application of patent analysis with drug regulation and decision-making. Through the mining and analysis of massive patent data, it can quickly identify the technical composition, technical evolution path, development trend, and regulatory risks of transdermal patches, providing more abundant information for policy-making and technical development, as well as a more effective and precise basis for decision-making.