Introduction

Cytometry is a platform for analysing cellular heterogeneity in clinical studies. Flow cytometry samples often have to be run real-time, making careful planning ahead of the study crucial: The staining protocol has to be established, sample stability ensured and data acquisition and analysis standardized across multiple laboratories. Furthermore, samples are discarded post-analysis, making sample reanalysis impossible.

Chipcytometry is an image-based cytometric system that has been designed as an alternative platform to overcome those limitations of flow cytometry. It combines high-plex quantitative phenotyping, cell imaging, sample preservation and long-term biomarker integrity.

A platform cross-comparison between Zellkraftwerk fully-automated Chipcytometry instrument CYTOBOT^TM and BD FACSCanto II was performed on samples at time point 0. Backup chips were stored for 6 months and analysed on the CYTOBOT^TM to assess sample and biomarker stability. In addition, intra-assay and inter-assay precision were measured in a third experiment.
Monoclonal antibodies against the proteins were used for the assay. The biomarkers were chosen to represent a typical marker panel ranging from common lineage biomarkers to immune check point proteins. Sequential immune phenotyping was performed according to the Chipcytometry principle. Data analysis was performed either in FlowJo (BD FACSCanto II data) or using Zellkraftwerk ZellExplorer software and the statistical software language R (Chipcytometry data).

Crosscomparison Flow cytometry vs. Chipcytometry
Cell frequencies as measured in Flow and in Chipcytometry at time point 0 (fresh on day of preparation for Flow cytometry, 1-2 weeks after preparation for Chipcytometry) for three different donors are shown. For the majority of the analysed cell populations, the results obtained by both technologies are comparable, the CVs ranging from 0.1% to 23.6%. For the ICOS-positive cells, there is a significant discrepancy, with Chipcytometry yielding consistently higher results than Flow cytometry. Investigation of this issue revealed that different clones were used on CYTOBOT and FACSCanto (C398.4A and ISA-3, respectively). When the clones were compared on a single system, ISA-3 stained less cells than C398.4A.

Precision assessment in Chipcytometry
For the intra-assay precision assessment, a single sample was split across multiple chips to measure the level of reproducibility. In contrast, the inter-assay precision was defined as the combined level of variability from different sample preparations (PBMC isolation tubes) and intra-assay variation. Both precision measures were assessed for three donors. The mean intra-assay and inter-assay coefficient of variation (CV) was 13.4% and 18.3%, respectively (Figure 3). Both CVs were inversely correlated with the size of the cell population (0.02-65%). With this performance, this assay is within the recommended precision for cytometry testing during drug development.

After 6 months storage, replicate chips were analysed with the same marker panel as the chips at time point 0. The mean % deviation between 0 and 6 months was 11.2%. All but 2 cell populations for the three donors were within the 25-30% accepted deviation level as has been recommended for cell-based fluorescence assays. Again, the CV was inversely correlated with the size of the cell population.