Flow cytometry applied to cell analysis and therapies – why is there variation in the metrology?
Cell analysis and production of cell therapies, whether by single clinical or multiple biomanufacturing sites, introduces challenges to be solved to achieve the delivery of safe, viable diagnosis and therapeutics. From a biomanufacturing viewpoint, these challenges include the translation of the process from T-flask to bioreactor, inline measurement and assessment of the cell journey, concerns around infections and asepsis, correct cell differentiation, supply logistics, business considerations, and the quality and variation of starting materials. This latter component was investigated and considered in the context of Haematopoietic Stem Cells Therapy (HSCT) with orders of magnitude variation found within the literature. A single site clinical centre HSCT exemplar provided opportunity to further investigate the associated HSCT cell manipulation processes. It was observed that the Flow Cytometry processes using operator defined manual gating techniques was potentially prone to variation in data analysis as a function of changing operators. Bespoke studies were subsequently commissioned with 30+ operators being asked to manually gate and process increasingly complex cell data. The results demonstrated that as complexity increased then cell count data variation increased, but, if systematic protocols were put into place then this human operator variation could be mitigated. Whilst manual gating is (in many cases) the preferred method for flow cytometry data processing, it was clear that a proliferation of unsupervised automated software platforms were being made increasingly available and competing for the data processing task. The aim of many of these software solutions is to reduce operator variation and increase the repeatability and accuracy of cell enumeration. Yet these software solutions involve a range of different mathematical paradigms that in themselves offer the potential for introducing different final cell count values and hence variation into the clinical and biomanufacturing setting. The research completed here has been to develop agnostic synthetic two and three cluster data sets that allows bespoke testing of a range of softawe platforms. This offers opportunity for unambiguous statements of accuracy and repeatability for each software solution, thus leading to a comparability analysis. This presentation considers how variation in the metrology of flow cytometry (specifically the data analysis component) can potentially impact quality of diagnosis within the clinical setting, and also affect the decision making associated with cell therapy critical quality attributes from a biomanufacturing perspective.
Funding
EPSRC/MRC Doctoral Training Centre for Regenerative Medicine at Loughborough University (EP/L105072/1)
EPSRC Centre for Innovative Manufacturing in Regenerative Medicine
Engineering and Physical Sciences Research Council
Find out more...Dana Farber Cancer Institute (Boston, USA)
UK Biobank – Project 4047
GlaxoSmithKline
LGC
History
School
- Mechanical, Electrical and Manufacturing Engineering
Source
Midlands Innovation Flow Cytometry Meeting 2022Version
- AM (Accepted Manuscript)
Rights holder
© The AuthorAcceptance date
2022-03-07Publication date
2022-04-25Copyright date
2022Publisher version
Language
- en