Assessment of Automated Flow Cytometry Data Analysis Tools within Cell and Gene Therapy Manufacturing.

Flow cytometry is widely used within the manufacturing of cell and gene therapies to measure and characterise cells. Conventional manual data analysis relies heavily on operator judgement, presenting a major source of variation that can adversely impact the quality and predictive potential of therapies given to patients. Computational tools have the capacity to minimise operator variation and bias in flow cytometry data analysis; however, in many cases, confidence in these technologies has yet to be fully established mirrored by aspects of regulatory concern. Here, we employed synthetic flow cytometry datasets containing controlled population characteristics of separation, and normal/skew distributions to investigate the accuracy and reproducibility of six cell population identification tools, each of which implement different unsupervised clustering algorithms: Flock2, flowMeans, FlowSOM, PhenoGraph, SPADE3 and SWIFT (density-based, k-means, self-organising map, k-nearest neighbour, deterministic k-means, and model-based clustering, respectively). We found that outputs from software analysing the same reference synthetic dataset vary considerably and accuracy deteriorates as the cluster separation index falls below zero. Consequently, as clusters begin to merge, the flowMeans and Flock2 software platforms struggle to identify target clusters more than other platforms. Moreover, the presence of skewed cell populations resulted in poor performance from SWIFT, though FlowSOM, PhenoGraph and SPADE3 were relatively unaffected in comparison. These findings illustrate how novel flow cytometry synthetic datasets can be utilised to validate a range of automated cell identification methods, leading to enhanced confidence in the data quality of automated cell characterisations and enumerations.

Systematic Design, Generation, and Application of Synthetic Datasets for Flow Cytometry.

Application of synthetic datasets in training and validation of analysis tools has led to improvements in many decision-making tasks in a range of domains from computer vision to digital pathology. Synthetic datasets overcome the constraints of real-world datasets, namely difficulties in collection and labeling, expense, time, and privacy concerns. In flow cytometry, real cell-based datasets are limited by properties such as size, number of parameters, distance between cell populations, and distributions and are often focused on a narrow range of disease or cell types. Researchers in some cases have designed these desired properties into synthetic datasets; however, operators have implemented them in inconsistent approaches, and there is a scarcity of publicly available, high-quality synthetic datasets. In this research, we propose a method to systematically design and generate flow cytometry synthetic datasets with highly controlled characteristics. We demonstrate the generation of two-cluster synthetic datasets with specific degrees of separation between cell populations, and of non-normal distributions with increasing levels of skewness and orientations of skew pairs. We apply our synthetic datasets to test the performance of a popular automated cell populations identification software, SPADE3, and define the region where the software performance decreases as the clusters get closer together. Application of the synthetic skewed dataset suggests the software is capable of processing non-normal data. We calculate the classification accuracy of SPADE3 with robustness not achievable with real-world datasets. Our approach aims to advance research toward generation of high-quality synthetic flow cytometry datasets and to increase their awareness among the community. The synthetic datasets can be used in benchmarking studies that critically evaluate cell population identification tools and help illustrate potential digital platform inconsistencies. These datasets have the potential to improve cell characterization workflows that integrate automated analysis in clinical diagnostics and cell therapy manufacturing.

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

Measured variability of product within Cell and Gene Therapy (CGT) manufacturing arises from numerous sources across pre-analytical to post-analytical phases of testing. Operators are a function of the manufacturing process and are an important source of variability as a result of personal differences impacted by numerous factors. This research uses measurement uncertainty in comparison to Coefficient of Variation to quantify variation of participants when they complete Flow Cytometry data analysis through a 5-step gating sequence. Two study stages captured participants applying gates using their own judgement, and then following a diagrammatical protocol, respectively. Measurement uncertainty was quantified for each participant (and analysis phase) by following Guide to the Expression of Uncertainty in Measurement protocols, combining their standard deviations in quadrature from each gating step in the respective protocols. When participants followed a diagrammatical protocol, variation between participants reduced by 57%, increasing confidence in a more uniform reported cell count percentage. Measurement uncertainty provided greater resolution to the analysis processes, identifying that most variability contributed in the Flow Cytometry gating process is from the very first gate, where isolating target cells from dead or dying cells is required. This work has demonstrated the potential for greater usage of measurement uncertainty within CGT manufacturing scenarios, due to the resolution it provides for root cause analysis and continuous improvement.

Current trends in flow cytometry automated data analysis software.

Automated flow cytometry (FC) data analysis tools for cell population identification and characterization are increasingly being used in academic, biotechnology, pharmaceutical, and clinical laboratories. The development of these computational methods is designed to overcome reproducibility and process bottleneck issues in manual gating, however, the take-up of these tools remains (anecdotally) low. Here, we performed a comprehensive literature survey of state-of-the-art computational tools typically published by research, clinical, and biomanufacturing laboratories for automated FC data analysis and identified popular tools based on literature citation counts. Dimensionality reduction methods ranked highly, such as generic t-distributed stochastic neighbor embedding (t-SNE) and its initial Matlab-based implementation for cytometry data viSNE. Software with graphical user interfaces also ranked highly, including PhenoGraph, SPADE1, FlowSOM, and Citrus, with unsupervised learning methods outnumbering supervised learning methods, and algorithm type popularity spread across K-Means, hierarchical, density-based, model-based, and other classes of clustering algorithms. Additionally, to illustrate the actual use typically within clinical spaces alongside frequent citations, a survey issued by UK NEQAS Leucocyte Immunophenotyping to identify software usage trends among clinical laboratories was completed. The survey revealed 53% of laboratories have not yet taken up automated cell population identification methods, though among those that have, Infinicyt software is the most frequently identified. Survey respondents considered data output quality to be the most important factor when using automated FC data analysis software, followed by software speed and level of technical support. This review found differences in software usage between biomedical institutions, with tools for discovery, data exploration, and visualization more popular in academia, whereas automated tools for specialized targeted analysis that apply supervised learning methods were more used in clinical settings.

Quantifying Operator Subjectivity within Flow Cytometry Data Analysis as a Source of Measurement Uncertainty and the Impact of Experience on Results.

Flow cytometry is a complex measurement characterization technique, utilized within the manufacture, measurement, and release of cell and gene therapy products for rapid, high-content, and multiplexed discriminatory cell analysis. A number of factors influence the variability in the measurement reported including, but not limited to, biological variation, reagent variation, laser and optical configurations, and data analysis methods. This research focused on understanding the contribution of manual operator variability within the data analysis phase. Thirty-eight participants completed a questionnaire, providing information about experience and motivational factors, before completing a simple gating study. The results were analyzed using gauge repeatability and reproducibility techniques to quantify participant uncertainty. The various stages of the gating sequence were combined through summation in quadrature and expanded to give each participant a representative uncertainty value. Of the participants surveyed, 85% preferred manual gating to automated data analysis, with the primary reasons being legacy ("it's always been done that way") and accuracy, not in the metrological sense but in the clear definition of the correct target population. The median expanded uncertainty was calculated as 3.6% for the population studied, with no significant difference among more or less experienced users. Operator subjectivity can be quantified to include within measurement uncertainty budgets, required for various standards and qualifications. An emphasis on biomanufacturing measurement terminology is needed to help understand future and potential solutions, possibly looking at translational clinical models to engage and enhance better training and protocols within industrial and research settings.

Tumour cell invasiveness and response to chemotherapeutics in adipocyte invested 3D engineered anisotropic collagen scaffolds.

Breast cancers are highly heterogeneous and their metastatic potential and response to therapeutic drugs is difficult to predict. A tool that could accurately gauge tumour invasiveness and drug response would provide a valuable addition to the oncologist's arsenal. We have developed a 3-dimensional (3D) culture model that recapitulates the stromal environment of breast cancers by generating anisotropic (directional) collagen scaffolds seeded with adipocytes and culturing tumour fragments therein. Analysis of tumour cell invasion in the presence of various therapeutic drugs, by immunofluorescence microscopy coupled with an optical clearing technique, demonstrated the utility of this approach in determining both the rate and capacity of tumour cells to migrate through the stroma while shedding light also on the mode of migration. Furthermore, the response of different murine mammary tumour types to chemotherapeutic drugs could be readily quantified.

Rapid comparative immunophenotyping of human mesenchymal stromal cells by a modified fluorescent cell barcoding flow cytometric assay.

Flow cytometry immunophenotyping is a sensitive technique allowing rapid characterization of single cells within heterogeneous populations, but it includes several subjective steps during sample analysis that impact the development of standardized methodology. Proposed strategies to overcome these limitations include fluorescent cell barcoding (FCB), which facilitates multiplexed sample evaluation with increased data reproducibility whilst reducing labeling variation, materials, and time. To date, the FCB assay has found utility for analyzing the phosphorylation status of intracellular targets but has not been intensively employed for cellular immunophenotypic analyses using cell surface markers. In this study we developed a modified FCB assay for multiplexed analysis of human mesenchymal stromal cells (hMSCs) to evaluate the quality of these cells during bioprocessing. A panel of fluorochrome-conjugated antibodies was used to target 15 ubiquitously expressed or stage-specific markers together with a fixable viability dye eFluor 506 acting as the cell barcoding agent. Critical technical considerations and validation steps were presented in the context of monitoring hMSC status, defined by generic, and specific surface markers for cell identity and quality. It was found that at discrete passages, inter-analyst expression patterns between hMSCs cultures were similar, but in contrast, diverse marker expression was evident between passages. A side-by-side analysis of barcoded and non-barcoded cells demonstrated the potential of this technique for the rapid phenotypic characterization of cells exposed to different bioprocessing conditions. Additionally, the method incorporates fewer subjective factors; including sample preparation and instrument day-to-day variations and is customizable across a diversity of cell types. © 2017 International Society for Advancement of Cytometry.

Development of three-dimensional collagen scaffolds with controlled architecture for cell migration studies using breast cancer cell lines.

Cancer is characterized by cell heterogeneity and the development of 3D in vitro assays that can distinguish more invasive or migratory phenotypes could enhance diagnosis or drug discovery. 3D collagen scaffolds have been used to develop analogues of complex tissues in vitro and are suited to routine biochemical and immunological assays. We sought to increase 3D model tractability and modulate the migration rate of seeded cells using an ice-templating technique to create either directional/anisotropic or non-directional/isotropic porous architectures within cross-linked collagen scaffolds. Anisotropic scaffolds supported the enhanced migration of an invasive breast cancer cell line MDA-MB-231 with an altered spatial distribution of proliferative cells in contrast to invasive MDA-MB-468 and non-invasive MCF-7 cells lines. In addition, MDA-MB-468 showed increased migration upon epithelial-to-mesenchymal transition (EMT) in anisotropic scaffolds. The provision of controlled architecture in this system may act both to increase assay robustness and as a tuneable parameter to capture detection of a migrated population within a set time, with consequences for primary tumour migration analysis. The separation of invasive clones from a cancer biomass with in vitro platforms could enhance drug development and diagnosis testing by contributing assay metrics including migration rate, as well as modelling cell-cell and cell-matrix interaction in a system compatible with routine histopathological testing.

Engineering mammary gland in vitro models for cancer diagnostics and therapy.

Breast cancer is a complex disease with many distinct subtypes being recognized on the basis of histological features and molecular signatures. It is difficult to predict how cancers will respond to therapy, which results in many women receiving unnecessary or inappropriate treatment. Advances in materials science and tissue engineering are leading the development of complex in vitro 3D breast tissue models that will increase our understanding of normal development and tumorigenic mechanisms. Ultimately, platforms that support primary tissue culture could readily be adapted to form high-throughput drug screening tools for personalized medicine. This review will summarize the control of mammary gland phenotype within in vitro 3D environments, in the context of a detailed analysis of mammary gland development and stem and progenitor cell controlled tumorigenesis.

A 3-D in vitro co-culture model of mammary gland involution.

Involution is a process whereby the mammary gland undergoes extensive tissue remodelling involving exquisitely coordinated cell death, extracellular matrix degradation and adipose tissue regeneration following the weaning of offspring. These processes are mediated in part through Jak/Stat signalling pathways, which can be deregulated in breast cancer. Synthetic in vitro analogues of the breast could become important tools for studying tumorigenic processes, or as personalized drug discovery platforms and predictors of therapeutic response. Ideally, such models should support 3D neo-tissue formation, so as to recapitulate physiological organ function, and be compatible with high-throughput screening methodologies. We have combined cell lines of epithelial, stromal and immunological origin within engineered porous collagen/hyaluronic acid matrices, demonstrating 3D-specific molecular signatures. Furthermore seeded cells form mammary-like branched tissues, with lobuloalveolar structures that undergo inducible involution phenotypes reminiscent of the native gland under hormonal/cytokine regulation. We confirm that autophagy is mediated within differentiated mammary epithelial cells in a Stat-dependent manner at early time points following the removal of a prolactin stimulus (H/WD). In addition, epithelial cells express markers of an M2 macrophage lineage under H/WD, a process that is attenuated with the introduction of the monocyte/macrophage cell line RAW 264.7. Thus, such 3D models are suitable platforms for studying cell-cell interactions and cell death mechanisms in relation to cancer.

A multifunctional 3D co-culture system for studies of mammary tissue morphogenesis and stem cell biology.

Studies on the stem cell niche and the efficacy of cancer therapeutics require complex multicellular structures and interactions between different cell types and extracellular matrix (ECM) in three dimensional (3D) space. We have engineered a 3D in vitro model of mammary gland that encompasses a defined, porous collagen/hyaluronic acid (HA) scaffold forming a physiologically relevant foundation for epithelial and adipocyte co-culture. Polarized ductal and acinar structures form within this scaffold recapitulating normal tissue morphology in the absence of reconstituted basement membrane (rBM) hydrogel. Furthermore, organoid developmental outcome can be controlled by the ratio of collagen to HA, with a higher HA concentration favouring acinar morphological development. Importantly, this culture system recapitulates the stem cell niche as primary mammary stem cells form complex organoids, emphasising the utility of this approach for developmental and tumorigenic studies using genetically altered animals or human biopsy material, and for screening cancer therapeutics for personalised medicine.

Stem cell mechanobiology.

Stem cells are undifferentiated cells that are capable of proliferation, self-maintenance and differentiation towards specific cell phenotypes. These processes are controlled by a variety of cues including physicochemical factors associated with the specific mechanical environment in which the cells reside. The control of stem cell biology through mechanical factors remains poorly understood and is the focus of the developing field of mechanobiology. This review provides an insight into the current knowledge of the role of mechanical forces in the induction of differentiation of stem cells. While the details associated with individual studies are complex and typically associated with the stem cell type studied and model system adopted, certain key themes emerge. First, the differentiation process affects the mechanical properties of the cells and of specific subcellular components. Secondly, that stem cells are able to detect and respond to alterations in the stiffness of their surrounding microenvironment via induction of lineage-specific differentiation. Finally, the application of external mechanical forces to stem cells, transduced through a variety of mechanisms, can initiate and drive differentiation processes. The coalescence of these three key concepts permit the introduction of a new theory for the maintenance of stem cells and alternatively their differentiation via the concept of a stem cell 'mechano-niche', defined as a specific combination of cell mechanical properties, extracellular matrix stiffness and external mechanical cues conducive to the maintenance of the stem cell population.

Three-dimensional culture models of mammary gland.

The mammary gland is a complex tissue comprised of a branching network of ducts embedded within an adipocyte-rich stroma. The ductal epithelium is a bi-layer of luminal and myoepithelial cells, the latter being in contact with a basement membrane. During pregnancy, tertiary branching occurs and lobuloalveolar structures, which produce milk during lactation, form in response to hormonal and cytokine signals. Postlactational regression is characterized by extensive cell death and tissue remodeling. These complex developmental events have been difficult to mimic in cell culture although many useful culture models exist. Recently, considerable advances in three-dimensional modelling of the mammary gland have been made with the use of collagen and other biomaterials for the study of branching morphogenesis and tumorigenesis, techniques which may enable rapid advances in our understanding of both basic biology and the study of cancer therapeutics.

Dynamic compressive strain influences chondrogenic gene expression in human mesenchymal stem cells.

This study tests the hypothesis that dynamic compressive strain selectively enhances chondrogenic differentiation by human mesenchymal stem cells (MSCs). Primary MSCs were isolated and expended in monolayer culture. The cells were seeded in alginate constructs or in pellet culture. The time course of chondrogenic differentiation was assessed by real-time QPCR of mRNA expression analysis for cartilage specific markers. Collagen types II and X mRNA, not present in undifferentiated MSCs, were detectable by 2-4 days of chondrogenic induction and continued to rise significantly throughout the culture period of 10 days (p < 0.001). Basal levels of gene expression for Sox-9 and aggrecan were evident in undifferentiated MSCs, although chondrogenic induction for a period of 8 days resulted in an increased trend in the gene expression levels. The alginate system was also used in mechanical conditioning studies. Dynamic compression was applied, in an intermittent regimen, at a strain amplitude of 15% and frequency of 1 Hz in the presence and absence of 10 ng/ml TGFbeta3, for a period of 8 days. Results indicated significant changes in the levels of mRNA expression for the chondrogenic markers. For example, by day 8, the application of the strain regimen alone caused an up-regulation in all the chondrogenic markers compared to the control samples (no TGFbeta, no compression). However, the combined effects of strain and TGFbeta on these markers were more complex than purely additive.