by
Shannon Eaker, PhD, Chief Technology Officer at Xcell Biosciences *
Dominic Clarke, PhD, Consultant & Chief Strategy Officer at Orange County Bio *
*Representing the International Society for Cell & Gene Therapy (ISCT) Process Development and Manufacturing (PDM) Committee.
The Cell and Gene Therapy (CGT) field has witnessed remarkable progress in recent years, opening up new possibilities for precision medicine and personalized treatments. In this era of rapid advancements, the integration of artificial intelligence (AI), machine learning (ML), Industry 4.0, and Digitalization has emerged as a driving force behind innovation in the field. These technologies hold the potential to enhance manufacturing processes, improve patient access, and optimize therapeutic outcomes. The annual International Society for Cell & Gene Therapy (ISCT) meeting in Paris this year provided a platform for experts to share insights, research, and discussions on the application of AI, ML, Industry 4.0, and Digitalization in CGT. This article aims to explore the key highlights and discussions from the ISCT meeting, shedding light on the recent advancements and their implications for the field.
At the ISCT meeting, numerous talks, roundtable discussions, and poster presentations highlighted the significance of AI and digitalization across various domains within CGT (Ref 1). These technologies were discussed in the context of potency and killing assays, visualization/microscopy, cellular trafficking, chimeric-antigen receptor (CAR) target identification, induced-pluripotent stem cell (iPSC) and mesenchymal stem/stromal cell (MSC) bioreactors, long-term patient safety monitoring, clinical trial design, adeno-associated virus (AAV) seroprevalence screening, and media/sera screening for cell culture. The widespread mention of AI and digital tools in these discussions demonstrates their growing importance in advancing research and manufacturing processes within CGT.
During a roundtable discussion led by Shannon Eaker, Chief Technology Officer at Xcell Biosciences, the slow adoption rate of automation and digitalization within CGT manufacturing was addressed. Jeet Sarkar, Vice President of Information Technology at the Center for Breakthrough Medicines, emphasized that digitalization should be viewed as a transformative process for businesses, and proposed workshops involving internal stakeholders from academic institutions, pharmaceutical companies, and industry partners. The aim of these workshops would be to discuss the capture and utilization of data sets that can feed into AI systems. The education of stakeholders (Manufacturing, PD, Supply Chain, QA/QC, key external vendors, etc.) was emphasized as a critical factor in driving adoption. Edwin Stone, Chief Executive Office at Cellular Origins, highlighted that automation and digitalization have become commonplace in many other industries and urged the CGT field to learn from their evolution. The discussion raised questions about who should take the lead in driving this transformation - therapy companies, equipment manufacturers, or other stakeholders. “Tools have taken a huge leap, and the field must understand how to use the tools, which ones and at what time.” Stone stated. It was emphasized that CGT must understand how to effectively use the tools available, identifying the right tools and the appropriate timing for their implementation. Standardization of data set formats for both processes and equipment/unit operations was proposed as a possible regulatory step to facilitate automation and digitalization. Kelly Ganjei, President and CEO at AmplifyBio, spoke to the importance of interfaces between unit operations or, at the very least, finding a way to connect data from each operation. Device integration platforms such as OPC-UA and SiLA 2 should be evaluated by instrumentation, technology and software providers. Validation and data integrity were also identified as critical factors that need to be addressed from research and development stages to process qualification and product release.
The reasons for including automation and digitalization in CGT manufacturing were explored, with cost reduction, time and effort savings in implementation and validation, data security, and the fear of locking down processes that could be further optimized identified as key drivers. The importance of considering future scalability of therapies and prioritizing patient needs over manual processes that may currently work was also emphasized. Starting with simpler aspects such as supply chain operations and weighing/dispensing in manufacturing was suggested as a potential approach to implementing automation and digitalization. The need to "fail forward" when establishing data platforms was discussed, emphasizing the importance of risk-based analysis of current processes to determine whether failures should lead to disruption or transformation and evolution.
In a roundtable discussion led by Rodney Reitze, co-Founder and Chief Executive Officer at iVexSol, the concept of Industry 4.0 and its implications for cell therapies were explored. The transition from Industry 1.0 (mechanization and steam power), 2.0 (mass production and assembly line), and 3.0 (automation and computers) to Industry 4.0 (smart automation, big data, cyber and autonomous systems) was discussed, with specific reference to CGT as the "Factory of the Future." The participants highlighted the need to define process problems versus data problems during internal stakeholder meetings to effectively address bottlenecks in patient access.
Dominic Clarke, Chief Strategy Officer at Orange County Bio, led a session discussion on implementing the first mile of big data in CGT manufacturing. He was joined by Mark Lowdell, Chief Scientific Officer at INmuneBio and Co-Founder of Autolomous, Najib Rehman, Digital Automation Lead at ATMPs and Kevin Gordon, Chief Digital Officer at Ori Biotech. Each of the speakers discussed the need for the industry to create an integrated and harmonized data ecosystem. Big data is a coined term that encompasses the large amount of information that the industry is generating from discovery, development, manufacturing and culminating with the patient. Harnessing the data early will lead to smarter, more efficient manufacturing but also aid in standardization.
The key question for the panel and the audience centered around our ability as an industry to not only capture the data but to enable access to the data. The potential need for patient lifetime data, both before and after treatment, was discussed, although the challenges of incorporating this level of data sharing were acknowledged. The importance of utilizing data to support patient access was emphasized, with examples highlighting the impact of data access and emphasizing the need to fail fast but importantly to fail forward. The idea of data capture combined with increasing data accessibility could lead to decoding cell and gene therapies and global patient availability.
As an example of AI-driven funding initiatives, Laura Herbst (Fraunhofer Institute for Production Technology) presented AIDPATH, an AI-powered Decentralized Production system for Advanced Therapies in the Hospital setting. “Partners from industry and research are building an automated and intelligent facility over a period of four years that is capable of producing targeted and patient-specific cell therapy directly at the point of treatment.” Weltin states. This project, funded by the European Union's Horizon 2020 research and innovation program, aims to build an automated and intelligent facility capable of producing targeted and patient-specific cell therapies directly at the point of treatment. The initial program focuses on CART therapy.
The discussion highlighted the challenges related to hardware and software, including the need for standardized communication protocols and improved device interoperability. “Closed, semi-automated systems have been developed to address these issues, such as the Miltenyi Prodigy and the Lonza Cocoon. These devices follow a “one-device-per patient approach” to minimize the risk of cross-contamination. Unfortunately, this manufacturing approach is unsuited for large-scale deployment, limiting the reduction of manufacturing costs and widespread application of CAR-T cell therapy” (Ref 2). The importance of patient data, including various parameters, and the concept of the digital twin (or digital ghost) were also addressed.
In Japan, the automated cell platform CellQualiaTM is being developed by Sinfonia and Foundation for Biomedical Research and Innovation at Kobe (FBRI). FBRI was established in March 2000 through Governmental funding from Kobe City and Hyogo Prefecture. This system is promoted as an “Intelligent Cell Processing System” for adherent iPSC and MSC cultures, providing microscopy and real-time sensors, environmental monitoring and media handling to provide automation in cell expansion.
The true test will be placing such a system in a Hospital environment, where logistics processes and data management/security are critical and cumbersome to navigate and advance. “The making of AI models enables an intensive insight into biological processes and informed decision-making. However, they also require various patient data (e.g., age, gender, previous illnesses) and the manufacturing process (e.g., process and cell parameters, device information) (Ref 2). The same team is also developing AutoCRAT, an extension of the EU funded AUTOSTEM project, a platform expanding MSC’s on microcarriers (iPSC derived MSCs for Osteoarthritis). AI has not been incorporated yet, but “The user can also incorporate decision trees that execute different process branches based on measurement data or user input to automate adaptive processes”.
Similar European Initiatives are helping to spark innovation in the space. Autolomous and Cell and Gene Therapy Catapult were recently awarded £1.2m grant by UK Research and Innovation to “digitise Process Analytical Technologies and accelerate industrial scale cell and gene therapy manufacturing” (Ref 3).
In contrast, the same month as the ISCT conference, the US President Joe Biden appointed Vice President Kamala Harris to be the "AI Czar", and the US FDA is currently requesting information and comments related to AI and drug manufacturing (including CGT, Ref 4 and 5). In July of this year, the FDA published a Draft Guidance document for Manufacturing Changes in CGT (Manufacturing Changes and Comparability for Human Cellular and Gene Therapy Products)(Ref 6). Although it does not mention manufacturing Automation, Digitalization or other related terms related to overall manufacturing process advancements, it provides a cautious and conservative approach when comparing manual versus automation within specific unit operations with respect to comparability. Once would suspect this same theme when transitioning to AI and process automation. The FDA is also currently struggling with common drug shortages, specifically with oncology drugs such as such as cisplatin and carboplatin, which some say such digitalization and AI strategies could help improve (Ref 7).
Investment in the space is also on the rise, as CGT and AI are both viewed as individual “Opportunity Areas” (Ref 8). It’s up to us in the CGT field to combine the two into a powerful machine.
The recent ISCT meeting provided valuable insights into the integration of AI, machine learning, Industry 4.0, and digitalization in the CGT field. Discussions and presentations emphasized the potential benefits of these technologies in improving precision medicine, enhancing manufacturing processes, and facilitating patient access. Challenges such as standardization, data integrity, scalability, and stakeholder education were discussed, urging collaboration and the development of strategies that prioritize patient care and safety. What was also noted is that digital terms which are new to most in CGT (e.g. AI, Digitalization, Digitization, Machine Learning, Industry 4.0, Data Set, Predictive Analytics, Digital Twin, IOT/Internet of Things, Data Lake, and Big Data) were often misused or mischaracterized in the context with which they were meant withing talks, posters and roundtable discussions.
Cell and gene therapies are anticipated to become one of the largest segments across oncology trailing only antibodies and small molecules (Ref 9). Notably, AI coupled with digitalization can facilitate the development of these therapies in almost every step, including target identification, payload design optimization, translational and clinical development, and complete digitization. The advancements in AI, machine learning, Industry 4.0, and digitalization hold great promise for the future of CGT, paving the way for more effective and personalized treatments. ISCT’s Process Development and Manufacturing (PDM) Committee is developing a Guidance Document based on these discussions to draft industry definitions, challenges and opportunities in order to advance and incorporate these tools into the CGT field.
References:
1. https://www.isctglobal.org/isct2023/home
2 https://www.frontiersin.org/articles/10.3389/fmed.2022.913287/full
3. https://ct.catapult.org.uk/news/autolomous-and-the-cell-and-gene-therapy-catapult-awarded-1-2m-grant-by-uk-research-and-innovation
4. https://pink.pharmaintelligence.informa.com/PS148407/Cell-Therapy-Manufacturing-Hurdles-Could-Be-Solved-Through-AI-Great-Advancements-In-Next-10-Years
5. https://www.federalregister.gov/documents/2023/03/01/2023-04206/discussion-paper-artificial-intelligence-in-drug-manufacturing-notice-request-for-information-and
6. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/manufacturing-changes-and-comparability-human-cellular-and-gene-therapy-products
7. https://www.fiercepharma.com/manufacturing/failure-market-results-persisting-shortages-fda-oncology-director-says-report
8. https://www.genengnews.com/gen-edge/ai-cell-gene-therapy-precision-medicine-drive-bullish-biotech-investment/
9. https://www.isctglobal.org/telegrafthub/blogs/lauren-reville/2023/08/08/the-ai-revolution-in-cell-and-gene-therapy-overcom
#IndustryMemberNews
#CommunityFeature