Mass Cytometry and Topological Data Analysis Reveal Immune Parameters Associated with Complications after Allogeneic Stem Cell Transplantation

Lakshmikanth, Tadepally; Olin, Axel; Chen, Yang; Mikes, Jaromír; Fredlund, Erik; Remberger, Mats; Omazic, Brigitta; Brodin, Petter

doi:10.1016/j.celrep.2017.08.021

Cited by 60 publications

(44 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A study using the same analysis method as the present study, but with patients undergoing allogeneic HSCT for acute leukemia and/or myelodysplastic syndrome noted a transient burst, decay, or slow increase or decrease in several of the proteins at 3 months after the procedure [36]. The authors suggested that the overrepresentation of proteins involved in cell proliferation, IL-2 signaling, and allograft rejection among the "3-month burst" proteins implied that these processes are particularly active during this phase of immune regeneration.…”

Section: Discussionmentioning

confidence: 55%

Profound but Transient Changes in the Inflammatory Milieu of the Blood During Autologous Hematopoietic Stem Cell Transplantation

Wiberg

Olsson‐Strömberg

Herman

et al. 2020

Biology of Blood and Marrow Transplantation

View full text Add to dashboard Cite

A B S T R A C TLittle is known about the inflammatory milieu in the blood during autologous hematopoietic stem cell transplantation (AHSCT) and how it is affected by the stem cell mobilization, collection, and reinfusion and conditioning regimen. In this study, we analyzed 92 proteins connected to inflammation at 10 time points during and after AHSCT in 16 patients with multiple sclerosis (MS). Serum from 29 patients with newly diagnosed MS and 15 healthy controls were included for comparative analysis. There were no significant differences in inflammatory serum protein levels between patients with newly diagnosed MS and healthy controls, but 29 out of 73 detectable proteins were significantly altered between at least 2 adjacent sampling time points during AHSCT. The predominant changes occurred after the conditioning regimen had been administered, whereas stem cell mobilization, collection, and reinfusion appeared to have less impact. Two distinct response patterns could be discerned, likely representing loss of basal cytokine production and homeostasis. The analyzed serum proteins gradually returned to baseline levels after treatment, with no remaining differences at 3 months after AHSCT. We conclude that treatment with AHSCT has a major but transient impact on the inflammatory milieu of peripheral blood.

show abstract

Section: Discussionmentioning

confidence: 55%

Profound but Transient Changes in the Inflammatory Milieu of the Blood During Autologous Hematopoietic Stem Cell Transplantation

Wiberg

Olsson‐Strömberg

Herman

et al. 2020

Biology of Blood and Marrow Transplantation

View full text Add to dashboard Cite

show abstract

“…Most notably, none of the 15 publications explicitly stated that the provided data underwent randomization. To the best of our knowledge, the only non‐randomized public data set has been announced in , during the preparations of this article.…”

Section: Resultsmentioning

confidence: 99%

Challenges in the Multivariate Analysis of Mass Cytometry Data: The Effect of Randomization

et al. 2019

View full text Add to dashboard Cite

Cytometry by time-of-flight (CyTOF) has emerged as a high-throughput single cell technology able to provide large samples of protein readouts. Already, there exists a large pool of advanced high-dimensional analysis algorithms that explore the observed heterogeneous distributions making intriguing biological inferences. A fact largely overlooked by these methods, however, is the effect of the established data preprocessing pipeline to the distributions of the measured quantities. In this article, we focus on randomization, a transformation used for improving data visualization, which can negatively affect multivariate data analysis methods such as dimensionality reduction, clustering, and network reconstruction algorithms. Our results indicate that randomization should be used only for visualization purposes, but not in conjunction with high-dimensional analytical tools.Single cell cytometry allows the detection of cell components in a high-throughput fashion. One of its latest versions is Cytometry by Time Of Flight (CyTOF) (1). The advantage of CyTOF compared to traditional flow cytometry is that high atomic weight metal reporters typically not found in a biological sample are employed for cell tagging, allowing the quantification of more than 40 cell parameters simultaneously. Such large number of parameters enables this technology to provide multivariate data sets with emerging properties that are well suited to advanced computational analysis (2,3). For example, unsupervised learning techniques like clustering and dimensionality reduction are typically used for cell phenotyping (4,5). In combination with statistical tests or supervised learning approaches, these methods are also employed for associating phenotypes or clinical outcomes to relevant cell subsets or protein markers (6,7). Clustering and dimensionality reduction are also commonly employed to visualize patterns in the data, marker relationships in the high-dimensional space or the phenotypic progression trajectories of cell subsets (8). Very recently, network-based methods have also been applied on CyTOF data, for automatic cell population identification (9) and the prediction of protein signaling networks using automated causal discovery algorithms (10).Despite the accelerated development of CyTOF-dedicated analysis methods, there is still no well-established data preprocessing consensus. This is mainly because the preceding standardization of experimental procedures is still in its infancy (11,12). In general, there are at least three distinct sources of technical variation in CyTOF. The first is the drop in the instrument sensitivity and the change in oxidation rate over long sample running times that causes signal fluctuations (13). Second,

show abstract

“…Other approaches dedicated to identifying differential immune cell frequencies in groups of samples can be applied, including approaches relying on statistical comparisons of cluster frequencies (Bruggner et al, 2014;Spitzer et al, 2017), convolutional neutral networks (Arvaniti and Claassen, 2017), empirical Bayes moderated tests (Weber et al, 2018) or hyperspheres (Lun et al, 2017). Since these approaches typically require advanced computational skills, we additionally demonstrate compatibility of this experimental workflow with a fully-automated, commercial analysis platform (astrolabediagnostics.com) to perform a systems-level analysis of immune cell reconstitution following BMT (Lakshmikanth et al, 2017;Stern et al, 2018) and to identify factors associated with the development of acute GvHD (Stikvoort et al, 2017). Albeit preliminary due to the limited sample number and potential confounding factors, we demonstrated the utility of our framework to identify such diseaseassociated cellular immune signatures in a clinical cohort.…”

Section: Discussionmentioning

confidence: 99%

Comprehensive Immune Monitoring of Clinical Trials to Advance Human Immunotherapy

Hartmann

Babdor

Gherardini

et al. 2018

Preprint

View full text Add to dashboard Cite

The success of immunotherapy has led to a myriad of new clinical trials. Connected to these trials are efforts to discover biomarkers providing mechanistic insight and predictive signatures for personalization. Still, the plethora of immune monitoring technologies can face investigator bias, missing unanticipated cellular responses in limited clinical material. We here present a mass cytometry workflow for standardized, systems-level biomarker discovery in immunotherapy trials. To broadly enumerate human immune cell identity and activity, we established and extensively assessed a reference panel of 33 antibodies to cover major cell subsets, simultaneously quantifying activation and immune checkpoint molecules in a single assay. The resulting assay enumerated ³ 98% of peripheral immune cells with ³ 4 positively identifying antigens. Robustness and reproducibility were demonstrated on multiple samples types, across research centers and by orthogonal measurements. Using automated analysis, we monitored complex immune dynamics, identifying signatures in bone-marrow transplantation associated graft-versus-host disease. This validated and available workflow ensures comprehensive immunophenotypic analysis, data comparability and will accelerate biomarker discovery in immunomodulatory therapeutics.

show abstract

Mass Cytometry and Topological Data Analysis Reveal Immune Parameters Associated with Complications after Allogeneic Stem Cell Transplantation

Cited by 60 publications

References 36 publications

Profound but Transient Changes in the Inflammatory Milieu of the Blood During Autologous Hematopoietic Stem Cell Transplantation

Profound but Transient Changes in the Inflammatory Milieu of the Blood During Autologous Hematopoietic Stem Cell Transplantation

Challenges in the Multivariate Analysis of Mass Cytometry Data: The Effect of Randomization

Comprehensive Immune Monitoring of Clinical Trials to Advance Human Immunotherapy

Contact Info

Product

Resources

About