Quantification of biological variation in blood‐based therapy – a summary of a meta‐analysis to inform manufacturing in the clinic

Abstract

Background and ObjectivesBiological raw materials, the basis for cellular therapies such as stem cells, have a significantly greater degree of complexity than their traditional pharmaceutical counterparts. This can be attributed to the inherent variation of its source – human beings. Currently, cell therapies are made in small, ad hoc batches, but larger scale production is a prerequisite to meeting future demand and will require a quality‐by‐design approach to manufacturing that will be designed around, or be robust to this variation. Quantification of variation will require understanding of the current baseline and stratification of its sources.Materials and MethodsHaematopoietic stem cell therapy was chosen as a case study to explore this variation, and a PRISMA‐guided (Preferred Reporting Items for Systematic Reviews and Meta‐Analyses) systematic meta‐analysis was carried out for a number of predetermined cell measurements.ResultsFrom this data set, it appears that the extent of variation in therapeutic dose (in terms of transplanted total nucleated cells and CD34+ cells per kilogram) for HSCT is between one and four orders of magnitude of the median.ConclusionsThis is tolerated under the practice of medicine but would be unmanageable from a biomanufacturing perspective and raises concerns about comparable levels of efficacy and treatment. A number of sources that will contribute towards this variation are also reported, as is the direction of travel for 4 greater clarity of the scale of this challenge.

Understanding the contribution of operator measurement variability within Flow Cytometry data analysis for Quality Control of Cell and Gene Therapy manufacturing

“…However, when applying similar control limits or error estimators to patient data or a cell therapy product, 'outliers' will be common within the probability distribution of the data (parametric or non-parametric) and cannot be excluded because they represent a specific patient or patient data upon which decisions must be based. Previous work has shown that patient data for cell counts for cell therapy starting materials can vary by up to 4 orders of magnitude with positively skewed distributions, which potentially creates significant difficulty when manufacturing a product that has to cope with this incoming variation [20].…”

The Early Career Researcher’s Toolkit: Translating Tissue Engineering, Regenerative Medicine and Cell Therapy Products

“…It is also worth noting that whilst consistency is paramount to any TERM manufacturing process, achieving this can be extremely difficult, particularly for inherently variable bioprocesses such as autologous processes where donor material will differ from patient to patient [53], as opposed to allogeneic processes where there is a degree of control with respect to biological starting material. Whilst it is unfeasible to eliminate variation entirely, understanding the sources of variation and identifying strategies to minimise these is critical and may result in a situation where differences in the product do not affect patient outcome [53].…”

“…Whilst it is unfeasible to eliminate variation entirely, understanding the sources of variation and identifying strategies to minimise these is critical and may result in a situation where differences in the product do not affect patient outcome [53].…”

Assessment of Protocol Impact on Subjectivity Uncertainty When Analyzing Peripheral Blood Mononuclear Cell Flow Cytometry Data Files

“…Variability within Cell and Gene Therapy (CGT) manufacturing arises from numerous sources and it has been well documented that starting material [1,2], inter-site comparability [3] and final product comparability [4] to name a few examples cause variability. However, operators are a key function of the manufacturing processes and cause inherent variation due to numerous factors, such as experience with a specific instrument, assay or specific cell lines [5].…”