Unsupervised machine learning reveals key immune cell subsets in COVID-19, rhinovirus infection, and cancer therapy

Development of a 27-color panel for the detection of measurable residual disease in patients diagnosed with acute myeloid leukemia

Acute myeloid leukemia (AML) measurable residual disease (MRD) evaluated by multiparametric flow cytometry (MFC) is a surrogate for progression-free and overall survival in clinical trials and patient management. Due to the limited number of detection channels available in conventional flow cytometers, panels used for assessing AML MRD are typically split into multiple tubes. This cripples the simultaneous and correlated assessment of all myeloblast measurements. In response, we prototyped a single-tube 27-color MFC assay for the evaluation of AML MRD, incorporating all recommended markers. Marrow aspirates from 22 patients were processed for analysis using full spectrum flow cytometry (FSFC). The signal resolution of each marker was compared between samples stained with single antibody vs. the fully stained panel. The analytical accuracy for quantifying hematopoietic cells between our established 8-color assay and the new 27-color method were compared. Variations within an operator and between separate operators were assessed to evaluate the assays reproducibility. The limited of blank (LOB), limit of detection (LOD), and lower limit of quantification (LLOQ) of the 27-color method were empirically determined using limiting dilution experiments. The stability of antibody cocktails over a period of 120 h was also studied using cryopreserved marrow cells. The stain indices for all antibodies were lower in the fully stained panel compared to cells stained with one antibody but clear separations between negative and positive signals were achieved for all antibodies. Our results demonstrated a high concordance between the established 8-color method and the new 27-color assay for enumerating myeloblasts and MRD interpretation within and between operators. The data further showed that the single-tube 27-color assay easily achieved the minimum required detection sensitivity of 0.1%. When antibodies were combined, however, expression intensity of some antigens deteriorated significantly when stored. Our single-tube 27-color panel is a suitable, high sensitivity flow cytometric approach that can be used for AML MRD testing, which improves the correlation of aberrant antigens and detection of asynchronous differentiation patterns. Based on the stability study, we recommend the full panel be made prior to staining.

Single-cell profiling of the antigen-specific response to BNT162b2 SARS-CoV-2 RNA vaccine

RNA-based vaccines against SARS-CoV-2 have proven critical to limiting COVID-19 disease severity and spread. Cellular mechanisms driving antigen-specific responses to these vaccines, however, remain uncertain. Here we identify and characterize antigen-specific cells and antibody responses to the RNA vaccine BNT162b2 using multiple single-cell technologies for in depth analysis of longitudinal samples from a cohort of healthy participants. Mass cytometry and unbiased machine learning pinpoint an expanding, population of antigen-specific memory CD4+ and CD8+ T cells with characteristics of follicular or peripheral helper cells. B cell receptor sequencing suggest progression from IgM, with apparent cross-reactivity to endemic coronaviruses, to SARS-CoV-2-specific IgA and IgG memory B cells and plasmablasts. Responding lymphocyte populations correlate with eventual SARS-CoV-2 IgG, and a participant lacking these cell populations failed to sustain SARS-CoV-2-specific antibodies and experienced breakthrough infection. These integrated proteomic and genomic platforms identify an antigen-specific cellular basis of RNA vaccine-based immunity.

Subphenotyping of COVID-19 patients at pre-admission towards anticipated severity stratification: an analysis of 778 692 Mexican patients through an age-sex unbiased meta-clustering technique

Background

The COVID-19 pandemic has led to an unprecedented global health care challenge for both medical institutions and researchers. Recognizing different COVID-19 subphenotypes—the division of populations of patients into more meaningful subgroups driven by clinical features—and their severity characterization may assist clinicians during the clinical course, the vaccination process, research efforts, the surveillance system, and the allocation of limited resources.

Objective

We aimed to discover age-sex unbiased COVID-19 patient subphenotypes based on easily available phenotypical data before admission, such as pre-existing comorbidities, lifestyle habits, and demographic features, to study the potential early severity stratification capabilities of the discovered subgroups through characterizing their severity patterns, including prognostic, intensive care unit (ICU), and morbimortality outcomes.

Methods

We used the Mexican Government COVID-19 open data, including 778,692 SARS-CoV-2 population-based patient-level data as of September 2020. We applied a meta-clustering technique that consists of a 2-stage clustering approach combining dimensionality reduction (ie, principal components analysis and multiple correspondence analysis) and hierarchical clustering using the Ward minimum variance method with Euclidean squared distance.

Results

In the independent age-sex clustering analyses, 56 clusters supported 11 clinically distinguishable meta-clusters (MCs). MCs 1-3 showed high recovery rates (90.27%-95.22%), including healthy patients of all ages, children with comorbidities and priority in receiving medical resources (ie, higher rates of hospitalization, intubation, and ICU admission) compared with other adult subgroups that have similar conditions, and young obese smokers. MCs 4-5 showed moderate recovery rates (81.30%-82.81%), including patients with hypertension or diabetes of all ages and obese patients with pneumonia, hypertension, and diabetes. MCs 6-11 showed low recovery rates (53.96%-66.94%), including immunosuppressed patients with high comorbidity rates, patients with chronic kidney disease with a poor survival length and probability of recovery, older smokers with chronic obstructive pulmonary disease, older adults with severe diabetes and hypertension, and the oldest obese smokers with chronic obstructive pulmonary disease and mild cardiovascular disease. Group outcomes conformed to the recent literature on dedicated age-sex groups. Mexican states and several types of clinical institutions showed relevant heterogeneity regarding severity, potentially linked to socioeconomic or health inequalities.

Conclusions

The proposed 2-stage cluster analysis methodology produced a discriminative characterization of the sample and explainability over age and sex. These results can potentially help in understanding the clinical patient and their stratification for automated early triage before further tests and laboratory results are available and even in locations where additional tests are not available or to help decide resource allocation among vulnerable subgroups such as to prioritize vaccination or treatments.

Single-Cell Profiling of the Antigen-Specific Response to BNT162b2 SARS-CoV-2 RNA Vaccine

RNA-based vaccines against SARS-CoV-2 are critical to limiting COVID-19 severity and spread. Cellular mechanisms driving antigen-specific responses to these vaccines, however, remain uncertain. We used single-cell technologies to identify and characterized antigen-specific cells and antibody responses to the RNA vaccine BNT162b2 in longitudinal samples from a cohort of healthy donors. Mass cytometry and machine learning pinpointed a novel expanding, population of antigen-specific non-canonical memory CD4 ⁺ and CD8 ⁺ T cells. B cell sequencing suggested progression from IgM, with apparent cross-reactivity to endemic coronaviruses, to SARS-CoV-2-specific IgA and IgG memory B cells and plasmablasts. Responding lymphocyte populations correlated with eventual SARS-CoV-2 IgG and a donor lacking these cell populations failed to sustain SARS-CoV-2-specific antibodies and experienced breakthrough infection. These integrated proteomic and genomic platforms reveal an antigen-specific cellular basis of RNA vaccine-based immunity.

ONE SENTENCE SUMMARY

Single-cell profiling reveals the cellular basis of the antigen-specific response to the BNT162b2 SARS-CoV-2 RNA vaccine.

Full Spectrum Flow Cytometry as a Powerful Technology for Cancer Immunotherapy Research

The Nobel Prize-deserving concept of blocking inhibitory pathways in T cells, to unleash their anti-tumoral capacity, became one of the pillars of cancer treatment in the last decade and has resulted in durable clinical responses for multiple cancer types. Currently, two of the most important goals in cancer immunotherapy are to understand the mechanisms resulting in failure to checkpoint blockade and to identify predictive immunological biomarkers that correlate to treatment response, disease progression or adverse effects. The identification and validation of biomarkers for routine clinical use is not only critical to monitor disease or treatment progression, but also to personalize and develop new therapies. To achieve these goals, powerful research tools are needed. Flow cytometry stands as one of the most successful single-cell analytical tools used to characterize immune cell phenotypes to monitor solid tumors, hematological malignancies, minimal residual disease or metastatic progression. This technology has been fundamental in diagnosis, treatment and translational research in cancer clinical trials. Most recently, the need to evaluate simultaneously more features in each cell has pushed the field to implement more powerful adaptations beyond conventional flow cytometry, including Full Spectrum Flow Cytometry (FSFC). FSFC captures the full emission spectrum of fluorescent molecules using arrays of highly sensitive light detectors, and to date has enabled characterization of 40 parameters in a single sample. We will summarize the contributions of this technology to the advancement of research in immunotherapy studies and discuss best practices to obtain reliable, robust and reproducible FSFC results.