Recommendations for using artificial intelligence in clinical flow cytometry.

Flow cytometry is a key clinical tool in the diagnosis of many hematologic malignancies and traditionally requires close inspection of digital data by hematopathologists with expert domain knowledge. Advances in artificial intelligence (AI) are transferable to flow cytometry and have the potential to improve efficiency and prioritization of cases, reduce errors, and highlight fundamental, previously unrecognized associations with underlying biological processes. As a multidisciplinary group of stakeholders, we review a range of critical considerations for appropriately applying AI to clinical flow cytometry, including use case identification, low and high risk use cases, validation, revalidation, computational considerations, and the present regulatory frameworks surrounding AI in clinical medicine. In particular, we provide practical guidance for the development, implementation, and suggestions for potential regulation of AI-based methods in the clinical flow cytometry laboratory. We expect these recommendations to be a helpful initial framework of reference, which will also require additional updates as the field matures.

flowSim: Near duplicate detection for flow cytometry data.

The analysis of large amounts of data is important for the development of machine learning (ML) models. flowSim is the first algorithm designed to visualize, detect and remove highly redundant information in flow cytometry (FCM) training sets to decrease the computational time for training and increase the performance of ML algorithms by reducing overfitting. flowSim performs near duplicate image detection by combining community detection algorithms with the density analysis of the marker expression values. flowSim clustering compared to consensus manual clustering on a dataset composed of 160 images of bivariate FCM data had a mean Adjusted Rand Index of 0.90, demonstrating its efficiency in identifying similar patterns. flowSim selectively discarded near duplicate files in datasets constructed with known redundancy, and removed 92.6% of FCM images in a dataset of over 500,000 drawn from public repositories.

Single-cell profiling reveals a memory B cell-like subtype of follicular lymphoma with increased transformation risk.

Follicular lymphoma (FL) is an indolent cancer of mature B-cells but with ongoing risk of transformation to more aggressive histology over time. Recurrent mutations associated with transformation have been identified; however, prognostic features that can be discerned at diagnosis could be clinically useful. We present here comprehensive profiling of both tumor and immune compartments in 155 diagnostic FL biopsies at single-cell resolution by mass cytometry. This revealed a diversity of phenotypes but included two recurrent patterns, one which closely resembles germinal center B-cells (GCB) and another which appears more related to memory B-cells (MB). GCB-type tumors are enriched for EZH2, TNFRSF14, and MEF2B mutations, while MB-type tumors contain increased follicular helper T-cells. MB-type and intratumoral phenotypic diversity are independently associated with increased risk of transformation, supporting biological relevance of these features. Notably, a reduced 26-marker panel retains sufficient information to allow phenotypic profiling of future cohorts by conventional flow cytometry.

Automated identification of maximal differential cell populations in flow cytometry data.

We introduce a new cell population score called SpecEnr (specific enrichment) and describe a method that discovers robust and accurate candidate biomarkers from flow cytometry data. Our approach identifies a new class of candidate biomarkers we define as driver cell populations, whose abundance is associated with a sample class (e.g., disease), but not as a result of a change in a related population. We show that the driver cell populations we find are also easily interpretable using a lattice-based visualization tool. Our method is implemented in the R package flowGraph, freely available on GitHub (github.com/aya49/flowGraph) and on BioConductor.

Multi-Omic Data Integration Allows Baseline Immune Signatures to Predict Hepatitis B Vaccine Response in a Small Cohort.

Background: Vaccination remains one of the most effective means of reducing the burden of infectious diseases globally. Improving our understanding of the molecular basis for effective vaccine response is of paramount importance if we are to ensure the success of future vaccine development efforts. Methods: We applied cutting edge multi-omics approaches to extensively characterize temporal molecular responses following vaccination with hepatitis B virus (HBV) vaccine. Data were integrated across cellular, epigenomic, transcriptomic, proteomic, and fecal microbiome profiles, and correlated to final HBV antibody titres. Results: Using both an unsupervised molecular-interaction network integration method (NetworkAnalyst) and a data-driven integration approach (DIABLO), we uncovered baseline molecular patterns and pathways associated with more effective vaccine responses to HBV. Biological associations were unravelled, with signalling pathways such as JAK-STAT and interleukin signalling, Toll-like receptor cascades, interferon signalling, and Th17 cell differentiation emerging as important pre-vaccination modulators of response. Conclusion: This study provides further evidence that baseline cellular and molecular characteristics of an individual's immune system influence vaccine responses, and highlights the utility of integrating information across many parallel molecular datasets.

Corrigendum: Clinical Protocol for a Longitudinal Cohort Study Employing Systems Biology to Identify Markers of Vaccine Immunogenicity in Newborn Infants in The Gambia and Papua New Guinea.

[This corrects the article DOI: 10.3389/fped.2020.00197.].

"Age Related Differences in the Biology of Chronic Graft-Versus-Host Disease After Hematopoietic Stem Cell Transplantation".

It is established that pediatric hematopoietic stem cell transplant (HSCT) recipients have a lower rate of chronic graft-versus-host disease (cGvHD) compared to adults. Our group has previously published immune profiles changes associated with cGvHD of clinically well-defined adult and pediatric HSCT cohorts. Since all analyses were performed by the same research group and analyzed using identical methodology, we first compared our previous immune profile analyses between adults and children. We then performed additional analyses comparing the T cell populations across age groups, and a sub-analysis of the impact of the estimated pubertal status at time of HSCT in our pediatric cohort. In all analyses, we corrected for clinical covariates including total body irradiation and time of onset of cGvHD. Three consistent findings were seen in both children and adults, including elevations of ST2 and naive helper T (Th) cells and depression of NKreg cells. However, significant differences exist between children and adults in certain cytokines, B cell, and Treg populations. In children, we saw a broad suppression of newly formed B (NF-B) cells, whereas adults exhibited an increase in T1-CD21lo B cells and a decrease in T1-CD24hiCD38hi B cells. Prepubertal children had elevations of aminopeptidase N (sCD13) and ICAM-1. Treg abnormalities in children appeared to be primarily in memory Treg cells, whereas in adults the abnormalities were in naïve Treg cells. In adults, the loss of PD1 expression in naïve Treg and naïve Th cells was associated with cGvHD. We discuss the possible mechanisms for these age-related differences, and how they might theoretically impact on different therapeutic approaches to cGvHD between children and adults.

Systems Biology Methods Applied to Blood and Tissue for a Comprehensive Analysis of Immune Response to Hepatitis B Vaccine in Adults.

Conventional vaccine design has been based on trial-and-error approaches, which have been generally successful. However, there have been some major failures in vaccine development and we still do not have highly effective licensed vaccines for tuberculosis, HIV, respiratory syncytial virus, and other major infections of global significance. Approaches at rational vaccine design have been limited by our understanding of the immune response to vaccination at the molecular level. Tools now exist to undertake in-depth analysis using systems biology approaches, but to be fully realized, studies are required in humans with intensive blood and tissue sampling. Methods that support this intensive sampling need to be developed and validated as feasible. To this end, we describe here a detailed approach that was applied in a study of 15 healthy adults, who were immunized with hepatitis B vaccine. Sampling included ~350 mL of blood, 12 microbiome samples, and lymph node fine needle aspirates obtained over a ~7-month period, enabling comprehensive analysis of the immune response at the molecular level, including single cell and tissue sample analysis. Samples were collected for analysis of immune phenotyping, whole blood and single cell gene expression, proteomics, lipidomics, epigenetics, whole blood response to key immune stimuli, cytokine responses, in vitro T cell responses, antibody repertoire analysis and the microbiome. Data integration was undertaken using different approaches-NetworkAnalyst and DIABLO. Our results demonstrate that such intensive sampling studies are feasible in healthy adults, and data integration tools exist to analyze the vast amount of data generated from a multi-omics systems biology approach. This will provide the basis for a better understanding of vaccine-induced immunity and accelerate future rational vaccine design.

Clinical Protocol for a Longitudinal Cohort Study Employing Systems Biology to Identify Markers of Vaccine Immunogenicity in Newborn Infants in The Gambia and Papua New Guinea.

Background: Infection contributes to significant morbidity and mortality particularly in the very young and in low- and middle-income countries. While vaccines are a highly cost-effective tool against infectious disease little is known regarding the cellular and molecular pathways by which vaccines induce protection at an early age. Immunity is distinct in early life and greater precision is required in our understanding of mechanisms of early life protection to inform development of new pediatric vaccines. Methods and Analysis: We will apply transcriptomic, proteomic, metabolomic, multiplex cytokine/chemokine, adenosine deaminase, and flow cytometry immune cell phenotyping to delineate early cellular and molecular signatures that correspond to vaccine immunogenicity. This approach will be applied to a neonatal cohort in The Gambia (N ~ 720) receiving at birth: (1) Hepatitis B (HepB) vaccine alone, (2) Bacille Calmette Guerin (BCG) vaccine alone, or (3) HepB and BCG vaccines, (4) HepB and BCG vaccines delayed till day 10 at the latest. Each study participant will have a baseline peripheral blood sample drawn at DOL0 and a second blood sample at DOL1,-3, or-7 as well as late timepoints to assess HepB vaccine immunogenicity. Blood will be fractionated via a "small sample big data" standard operating procedure that enables multiple downstream systems biology assays. We will apply both univariate and multivariate frameworks and multi-OMIC data integration to identify features associated with anti-Hepatitis B (anti-HB) titer, an established correlate of protection. Cord blood sample collection from a subset of participants will enable human in vitro modeling to test mechanistic hypotheses identified in silico regarding vaccine action. Maternal anti-HB titer and the infant microbiome will also be correlated with our findings which will be validated in a smaller cohort in Papua New Guinea (N ~ 80). Ethics and Dissemination: The study has been approved by The Gambia Government/MRCG Joint Ethics Committee and The Boston Children's Hospital Institutional Review Board. Ethics review is ongoing with the Papua New Guinea Medical Research Advisory Committee. All de-identified data will be uploaded to public repositories following submission of study output for publication. Feedback meetings will be organized to disseminate output to the study communities. Clinical Trial Registration: Clinicaltrials.gov Registration Number: NCT03246230.

Single Cell Phenotypic Profiling of 27 DLBCL Cases Reveals Marked Intertumoral and Intratumoral Heterogeneity.

Diffuse large B-cell lymphoma (DLBCL) is the most common histologic subtype of non-Hodgkin lymphoma and is notorious for its clinical heterogeneity. Patient outcomes can be predicted by cell-of-origin (COO) classification, demonstrating that the underlying transcriptional signature of malignant B-cells informs biological behavior in the context of standard combination chemotherapy regimens. In the current study, we used mass cytometry (CyTOF) to examine tumor phenotypes at the protein level with single cell resolution in a collection of 27 diagnostic DLBCL biopsy specimens from treatment naïve patients. We found that malignant B-cells from each patient occupied unique regions in 37-dimensional phenotypic space with no apparent clustering of samples into discrete subtypes. Interestingly, variable MHC class II expression was found to be the greatest contributor to phenotypic diversity. Within individual tumors, a subset of cases showed multiple phenotypic subpopulations, and in one case, we were able to demonstrate direct correspondence between protein-level phenotypic subsets and DNA mutation-defined subclones. In summary, CyTOF analysis can resolve both intertumoral and intratumoral heterogeneity among primary samples and reveals that each case of DLBCL is unique and may be comprised of multiple, genetically distinct subclones. © 2019 International Society for Advancement of Cytometry.

Guidelines for the use of flow cytometry and cell sorting in immunological studies (second edition).

These guidelines are a consensus work of a considerable number of members of the immunology and flow cytometry community. They provide the theory and key practical aspects of flow cytometry enabling immunologists to avoid the common errors that often undermine immunological data. Notably, there are comprehensive sections of all major immune cell types with helpful Tables detailing phenotypes in murine and human cells. The latest flow cytometry techniques and applications are also described, featuring examples of the data that can be generated and, importantly, how the data can be analysed. Furthermore, there are sections detailing tips, tricks and pitfalls to avoid, all written and peer-reviewed by leading experts in the field, making this an essential research companion.

Improving the Quality and Reproducibility of Flow Cytometry in the Lung. An Official American Thoracic Society Workshop Report.

Defining responses of the structural and immune cells in biologic systems is critically important to understanding disease states and responses to injury. This requires accurate and sensitive methods to define cell types in organ systems. The principal method to delineate the cell populations involved in these processes is flow cytometry. Although researchers increasingly use flow cytometry, technical challenges can affect its accuracy and reproducibility, thus significantly limiting scientific advancements. This challenge is particularly critical to lung immunology, as the lung is readily accessible and therefore used in preclinical and clinical studies to define potential therapeutics. Given the importance of flow cytometry in pulmonary research, the American Thoracic Society convened a working group to highlight issues and technical challenges to the performance of high-quality pulmonary flow cytometry, with a goal of improving its quality and reproducibility.

Data-Driven Flow Cytometry Analysis.

The emergence of flow and mass cytometry technologies capable of generating 40-dimensional data has spurred research into automated methodologies that address bottlenecks across the entire analysis process from quality checking, data transformation, and cell population identification, to biomarker identification and visualizations. We review these approaches in the context of the stepwise progression through the different steps, including normalization, automated gating, outlier detection, and graphical presentation of results.

Flow cytometry data analysis: Recent tools and algorithms.

Flow cytometry (FCM) allows scientists to rapidly quantify up to 50 parameters for millions of cells per sample. The bottleneck in the application of the technology is data analysis, and the high number of parameters measured by the current generation of instruments requires the use of advanced computational algorithms to make full use of their capabilities. This review summarizes the main steps of FCM data analysis, focusing on the use of the most recent bioinformatic tools developed for an R-based programming environment. In particular, for each stage of the data analysis, libraries and packages currently available are listed, and a brief description of their functioning is included.

Dynamic molecular changes during the first week of human life follow a robust developmental trajectory.

Systems biology can unravel complex biology but has not been extensively applied to human newborns, a group highly vulnerable to a wide range of diseases. We optimized methods to extract transcriptomic, proteomic, metabolomic, cytokine/chemokine, and single cell immune phenotyping data from <1 ml of blood, a volume readily obtained from newborns. Indexing to baseline and applying innovative integrative computational methods reveals dramatic changes along a remarkably stable developmental trajectory over the first week of life. This is most evident in changes of interferon and complement pathways, as well as neutrophil-associated signaling. Validated across two independent cohorts of newborns from West Africa and Australasia, a robust and common trajectory emerges, suggesting a purposeful rather than random developmental path. Systems biology and innovative data integration can provide fresh insights into the molecular ontogeny of the first week of life, a dynamic developmental phase that is key for health and disease.

Implementation and Validation of an Automated Flow Cytometry Analysis Pipeline for Human Immune Profiling.

Automated reagent preparation, sample processing, and data acquisition have increased the rate at which flow cytometry data can be generated. Furthermore, advances in technology and flow cytometry instrumentation continually increase the complexity and dimensionality of this data. Together, this leads to increased pressure on manual data analysis, which has inherent limitations including subjectivity of the analyst and the length of time needed for data processing. These issues can create bottlenecks in the data processing workflow and potentially compromise data quality. To address these issues, as well as the challenges associated with manual gating in a high-volume human immune profiling laboratory, we sought to implement an automated analysis pipeline. In this report, we discuss considerations for selecting an automated analysis method, the process of implementing an automated pipeline, and detail our successful incorporation of an automated gating strategy with flowDensity into our analysis workflow. This validated pipeline augments our laboratory's ability to provide rapid high-throughput immune profiling for patients participating in cancer immunotherapy clinical trials. © International Society for Advancement of Cytometry.

A standardized immune phenotyping and automated data analysis platform for multicenter biomarker studies.

The analysis and validation of flow cytometry-based biomarkers in clinical studies are limited by the lack of standardized protocols that are reproducible across multiple centers and suitable for use with either unfractionated blood or cryopreserved PBMCs. Here we report the development of a platform that standardizes a set of flow cytometry panels across multiple centers, with high reproducibility in blood or PBMCs from either healthy subjects or patients 100 days after hematopoietic stem cell transplantation. Inter-center comparisons of replicate samples showed low variation, with interindividual variation exceeding inter-center variation for most populations (coefficients of variability <20% and interclass correlation coefficients >0.75). Exceptions included low-abundance populations defined by markers with indistinct expression boundaries (e.g., plasmablasts, monocyte subsets) or populations defined by markers sensitive to cryopreservation, such as CD62L and CD45RA. Automated gating pipelines were developed and validated on an independent data set, revealing high Spearman's correlations (rs >0.9) with manual analyses. This workflow, which includes pre-formatted antibody cocktails, standardized protocols for acquisition, and validated automated analysis pipelines, can be readily implemented in multicenter clinical trials. This approach facilitates the collection of robust immune phenotyping data and comparison of data from independent studies.