BioPharm International - October 2021

BioPharm-October 2021-Regulatory-Sourcebook

Issue link:

Contents of this Issue


Page 22 of 33 October 2021 eBook BioPharm International 23 (ML) is a branch of AI that is used as a technique for designing and training software algorithms to learn and act on data. MOVING AT THE SPEED OF TECHNOLOGY In the life sciences arena, the imple- mentation of AI is ramping up quickly and playing a valuable role in phar- maceutical product development. The article "Scaling Up AI Across the Life Sciences Value Chain" published by Deloitte explains how the technol- ogy is integral in the transformation of biopharma R&D and drug develop- ment, such as the following (3): • An abundance of genotypic and phenotypic data has created a need for techniques that can transform and expedite target identification and validation. AI imaging capa- bilities can detect cell morphology changes that humans cannot see through a microscope. • A p p l y i n g A I t o k n o w l e d g e g r aph s a l low s re s e a rc her s to understand complex relationships between compounds, genes, dis- eases, and proteins. • Organizations are pioneering the use of generative modeling for small-molecule design and pro- tein engineering. • AI-driven digital data f low solu- tions can integrate trial data from multiple-source systems and docu- ments to create standardized digital data elements for transmission to downstream systems. • Facilities are increasingly lever- aging AI to normalize data from different platforms, such as gene expression data, to integrate mul- t iple data points ( genot y pic , imaging, clinical records, and epi- demiological) for patient stratifica- tion and identification. According to data gathered from a survey conducted by Deloitte in 2019, more than 60% of life sciences compa- nies have spent more than $20 million on AI initiatives These investments are already revealing their value in lowering costs and improving opera- tional efficiency. "AI will indeed make it possible to bring all medical knowledge to bear in service of any case. Properly designed AI also has the potential to make our health care system more efficient and less expensive, ease the paper work burden that has more and more doc- tors considering new careers, [and] fill the gaping holes in access to quality care in the world 's poorest places," said Isaac Kohane, head of Harvard Med ic a l S c hool 's D epa r t ment of Biomedical Informatics (4). DATA-DRIVEN TECHNOLOGY The digitization of processes, per- vasive use of mobile technologies, advancements in data gathering and analytics, and inf lux of AI in life sci- ences industries have created new t y pes of a nd more uses for data. Companies have long generated vast amounts of data as a by product of their processes, information manage- ment systems, equipment, personnel, products, communications, and more. This paradigm of data-centricity will continue to evolve as all branches of AI are fueled by data. That said, regulatory agencies such as FDA are getting on board with advancing the use of the technology as well as augmenting their own data management practices. Even before the global pandemic accelerated the drive for modernization, FDA was already engaged in a technolog y upgrade. Acting FDA Director Janet Woodcock asserted that "[FDA] will continue to leverage and maximize every available tool and resource to meet our inspec- tional responsibilities while achieving optimal public health outcomes" (5). Because of the increased emphasis on data, the agency's modernization endeavors include transforming its data enterprise platforms. One example of this practice is the data standards strat- egy created by FDA's Center for Drug Evaluation and Research, which is designed to improve the efficiency and effectiveness of regulatory submission reviews. The organization also plans to engage in a more in-depth review of its approaches to regulatory oversight and employ next-generation assess- ment technologies and improvements. Enhancing support for innovation calls for the ability to gather and analyze larger and more complex data sets. Part of FDA's data initiative is the Data Modernization Action Plan (DMAP). Specifically, this data strat- egy focuses on the stewardship, secu- rity, quality control, analysis, and use of data in developing state-of-the-art prod- ucts and solutions. At a high level, the DMAP will enable the agency to align technology and innovation across mul- tiple industries, enhance its data prac- tices, and promote efficient collaboration across a growing, diverse workforce (6). Understanding the potential of what can be achieved through data is the impetus for FDA's DMAP strategy. Evolving technologies are ushering in next-generation solutions and more precision processes. Data are a highly integral component of these technol- ogies. For example, FDA's ability to track and trace medical and food prod- ucts can lead to more expedient respon- siveness to unplanned events such as pandemics, natural disasters, and global supply chain disruptions. As part of its overall modernization efforts, FDA is also strengthening its understanding of AI and ML. Areas under consideration include the tech- nical and practical application of the advanced technologies, new regulatory questions introduced by AI applica- tions, and the impact of AI solutions across the life cycle of FDA-regulated products. "While the scientific progress of FDA is enabling the entry of inno- vative products into the marketplace to improve the lives of the American public, FDA needs to continue to keep pace with evolving science and technol- Regulatory Sourcebook Data Integrity

Articles in this issue

Links on this page

Archives of this issue

view archives of BioPharm International - October 2021 - BioPharm-October 2021-Regulatory-Sourcebook