Understanding and innovating in cell therapy quality control: an interview with Stephen Sullivan, GAiT
Stephen Sullivan (GAiT) reflects on the evolution of quality control in the field of cell therapy, and discusses what the future may hold and what obstacles still remain.
How have quality control needs changed since you began working in this field?
Derivation of human induced pluripotent stem cells (iPSCs) was first described by Prof Yamanaka and members of his laboratory in 2007 and, given these reprogrammed cells capacity to self-renewal or produce differentiated progeny, their clinical potential for cell replacement therapy was manifest since that discovery. However, it is only recently institutions around the world have begun to entertain the practical challenges to scalable manufacturing of appropriately ‘clinical-grade’ iPSC lines for allogeneic or ‘off the shelf’ cell replacement strategies.
Before appropriate quality control (QC) can be implemented, what quality is, in the context of iPSC lines to be used as starting material for therapeutic development, needs to be considered and agreed. Conversations about iPSC quality between cell therapy developers and manufacturers began and to date these are strongly informed by what has been learned by manufacture of other cell types and tissues.
The Global Alliance for iPSC Therapies (GAiT) hosted a series of Quality Workshops in 2017 and 2018 where its members were surveyed on how their members understood critical quality attributes in the context of clinical grade iPSC lines as starting material for allogeneic therapeutic development, how they tested for such quality and who tested for it, and, while extensive diversity was found in the replies, a list of critical quality attributes for clinical-grade iPSC was agreed.
Read GAiT’s white paper on QC guidelines for clinical-grade human induced pluripotent stem cell lines >>
Assessing purity (absence of adventitious agents and cells of different origin), identity (authentic genotype and phenotype as originally described), performance (functional response or expression), stability both in vitro (degree of phenotypic and genotypic variation that occurs on passage or use in different culture environments) and in vivo (tumorigenicity and oncogenicity) would be required.
As several of selected quality tests deemed mandatory presently lack pharmacopeial guidance, GAiT subsequently has organized a ‘Quality Round’ where as part of a Quality Assessment Exercise standards were circulated to members to qualify their own testing. 18 institutions, across 11 countries are participating with a second quality round planned for next year.
Technologies have evolved and continue to evolve for both processing and analysis. In terms of QC, the progression of technologies and guidance to assess the genetic stability of the bank continues to evolve. The focus is increasingly moving to the understanding of genetic mutations/inconsistencies. This will be key going forward.
How can QC protocols be adapted to meet the different needs of different applications?
Quality being considered from the beginning and being purposefully built into the process is desirable. In terms of applications for different indications, development of QC standards for donor selection, consent, and master cell bank testing, will facilitate this. Further acceptance by regulators in different territories of acceptable data is also required for the transferability of banks for different indications. The use of platform control strategies can also be deployed, by adapting for different protocols once proven for a given product type.
With the important role that quality control plays in the manufacturing process, how can developers simplify this process?
Embedding QC into the process is essential for long, complex differentiation and expansion processes. Decisions need to be made as early as possible to ensure product consistency, potency and safety. Manufacturers should build QC checks into the process; utilizing offline or atline technologies.
Moving forward, efforts should be made to embed inline soft sensors into the process where possible: for example, pluripotency testing can be at least a partial functional readout for identity and stability. To simplify the process, manufacturers should resist measuring an attribute just because it is easily measured; rather, they should focus on the attributes which are good predictors of patient risk.
What is the biggest obstacle remaining in the field?
The understanding of the importance of mutations in iPSC lines within the GAiT haplobank network, and without, will be needed. Preventing such mutations and minimizing their potential impact on the final product are key obstacles for the field.
While this and other significant technical and reimbursement hurdles persist, the biggest obstacle is a human one: the forging and adaptation of common definitions, standards and practices. A lack of alignment around these has caused waste and confusion in the mesenchymal stem cell field, for example, and it is important for a balanced funding policy to stage clinical trials. However, to balance this, the establishment of definitions and standards and developing enabling technologies should also be funded, which could well have more clinical impact in the long run.
Information mobilization is a key activity of GAiT, bringing different stakeholders into dialogue around common challenges and building a roadmap for how iPSCs therapeutics can be developed efficiently.
We talk about ‘silo breaching’ within GAiT. While personnel in different functions can have a very deep level of knowledge within their own professional silo, they might not have a holistic view of what is involved in manufacturing clinical-grade iPSCs or generating a therapeutic therefrom. Much effort and resources can then be wasted if donor material is improperly sourced or inadequately consented, for example, and much that is required cannot be intuitively deduced without specialized knowledge.
GAiT attempts to broaden its members appreciation of what is required for the manufacture of clinical-grade iPSC lines through free virtual workshops through its website. The most recent ones have focused on donor recruitment and donor consent.
What’s the best/most effective technology innovation you’ve seen in the last 5 years?
Maintaining viability of cells iPSCs during routine culture was a major challenge in the early days. More optimized culture systems have been developed which are now in common use in basic research. However, volumes and costs grow significantly when considering large scale GMP manufacturing of iPSC cells where billions of clinical-grade cells are needed for large transplantation programs. The current diversity of culture conditions iPSCs are grown in make it difficult to determine whether variability originates from biological, process or other sources.
We also need more stringent markers that are not only identifiers of cell identity and functionality but are also good indicators of patient risk. Even though we refer to ‘pluripotent stem cell markers’, none have yet been identified. This misnomer commonly refers to the co-expression of self-renewal and undifferentiated cell markers. There are emerging technologies that could help address this need, however.
One of the most promising is single cell transcriptomics. This approach can provide a highly detailed understanding of gene expression and identify the genes and pathways that define differences in pluripotent cell states. In turn this information can then be used to define panels of markers for cell monitoring and control during GMP manufacture.
While the continuing optimization of the bulk cell culture is significant, the means by which cell culture quality is currently assessed is highly subjective. Quantitative software being used to analyze live cell imaging data is also improving.
How do you see quality control processes, protocols and products evolving in the next 10 years?
As nuclear reprogramming mechanisms are better elucidated, we will be better able to characterize iPSCs at a molecular level, removing some of the starting material variability and consequently also allowing better markers to be identified. It will be important to build QC into the manufacturing process as much as possible.
Standardization of how manufacturers test for tumorigenicity will be key. HESI CT-TRACS is doing a lot to build consensus in this area. GAiT is co-hosting a series of roundtables around quality and manufacturing challenges for human pluripotent stem cells on 30 June 2019 at the University of Southern California (CA, USA).
However, some of the complex variances cannot be controlled due to the intrinsically complex nature of cells and there is a lack of established standard references to aid the construction of robust functional (orthologonal) assays.