Potential and limitations of machine meta-learning (ensemble) methods for predicting COVID-19 mortality in a large inhospital Brazilian dataset.

The majority of early prediction scores and methods to predict COVID-19 mortality are bound by methodological flaws and technological limitations (e.g., the use of a single prediction model). Our aim is to provide a thorough comparative study that tackles those methodological issues, considering multiple techniques to build mortality prediction models, including modern machine learning (neural) algorithms and traditional statistical techniques, as well as meta-learning (ensemble) approaches. This study used a dataset from a multicenter cohort of 10,897 adult Brazilian COVID-19 patients, admitted from March/2020 to November/2021, including patients [median age 60 (interquartile range 48-71), 46% women]. We also proposed new original population-based meta-features that have not been devised in the literature. Stacking has shown to achieve the best results reported in the literature for the death prediction task, improving over previous state-of-the-art by more than 46% in Recall for predicting death, with AUROC 0.826 and MacroF1 of 65.4%. The newly proposed meta-features were highly discriminative of death, but fell short in producing large improvements in final prediction performance, demonstrating that we are possibly on the limits of the prediction capabilities that can be achieved with the current set of ML techniques and (meta-)features. Finally, we investigated how the trained models perform on different hospitals, showing that there are indeed large differences in classifier performance between different hospitals, further making the case that errors are produced by factors that cannot be modeled with the current predictors.

Predictors of venous thromboembolism in COVID-19 patients: results of the COVID-19 Brazilian Registry.

Previous studies that assessed risk factors for venous thromboembolism (VTE) in COVID-19 patients have shown inconsistent results. Our aim was to investigate VTE predictors by both logistic regression (LR) and machine learning (ML) approaches, due to their potential complementarity. This cohort study of a large Brazilian COVID-19 Registry included 4120 COVID-19 adult patients from 16 hospitals. Symptomatic VTE was confirmed by objective imaging. LR analysis, tree-based boosting, and bagging were used to investigate the association of variables upon hospital presentation with VTE. Among 4,120 patients (55.5% men, 39.3% critical patients), VTE was confirmed in 6.7%. In multivariate LR analysis, obesity (OR 1.50, 95% CI 1.11-2.02); being an ex-smoker (OR 1.44, 95% CI 1.03-2.01); surgery ≤ 90 days (OR 2.20, 95% CI 1.14-4.23); axillary temperature (OR 1.41, 95% CI 1.22-1.63); D-dimer ≥ 4 times above the upper limit of reference value (OR 2.16, 95% CI 1.26-3.67), lactate (OR 1.10, 95% CI 1.02-1.19), C-reactive protein levels (CRP, OR 1.09, 95% CI 1.01-1.18); and neutrophil count (OR 1.04, 95% CI 1.005-1.075) were independent predictors of VTE. Atrial fibrillation, peripheral oxygen saturation/inspired oxygen fraction (SF) ratio and prophylactic use of anticoagulants were protective. Temperature at admission, SF ratio, neutrophil count, D-dimer, CRP and lactate levels were also identified as predictors by ML methods. By using ML and LR analyses, we showed that D-dimer, axillary temperature, neutrophil count, CRP and lactate levels are risk factors for VTE in COVID-19 patients.

Immune biomarkers to predict SARS-CoV-2 vaccine effectiveness in patients with hematological malignancies.

There is evidence of reduced SARS-CoV-2 vaccine effectiveness in patients with hematological malignancies. We hypothesized that tumor and treatment-related immunosuppression can be depicted in peripheral blood, and that immune profiling prior to vaccination can help predict immunogenicity. We performed a comprehensive immunological characterization of 83 hematological patients before vaccination and measured IgM, IgG, and IgA antibody response to four viral antigens at day +7 after second-dose COVID-19 vaccination using multidimensional and computational flow cytometry. Health care practitioners of similar age were the control group (n = 102). Forty-four out of 59 immune cell types were significantly altered in patients; those with monoclonal gammopathies showed greater immunosuppression than patients with B-cell disorders and Hodgkin lymphoma. Immune dysregulation emerged before treatment, peaked while on-therapy, and did not return to normalcy after stopping treatment. We identified an immunotype that was significantly associated with poor antibody response and uncovered that the frequency of neutrophils, classical monocytes, CD4, and CD8 effector memory CD127low T cells, as well as naive CD21+ and IgM+D+ memory B cells, were independently associated with immunogenicity. Thus, we provide novel immune biomarkers to predict COVID-19 vaccine effectiveness in hematological patients, which are complementary to treatment-related factors and may help tailoring possible vaccine boosters.

Immunological Biomarkers of Fatal COVID-19: A Study of 868 Patients.

Information on the immunopathobiology of coronavirus disease 2019 (COVID-19) is rapidly increasing; however, there remains a need to identify immune features predictive of fatal outcome. This large-scale study characterized immune responses to severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection using multidimensional flow cytometry, with the aim of identifying high-risk immune biomarkers. Holistic and unbiased analyses of 17 immune cell-types were conducted on 1,075 peripheral blood samples obtained from 868 COVID-19 patients and on samples from 24 patients presenting with non-SARS-CoV-2 infections and 36 healthy donors. Immune profiles of COVID-19 patients were significantly different from those of age-matched healthy donors but generally similar to those of patients with non-SARS-CoV-2 infections. Unsupervised clustering analysis revealed three immunotypes during SARS-CoV-2 infection; immunotype 1 (14% of patients) was characterized by significantly lower percentages of all immune cell-types except neutrophils and circulating plasma cells, and was significantly associated with severe disease. Reduced B-cell percentage was most strongly associated with risk of death. On multivariate analysis incorporating age and comorbidities, B-cell and non-classical monocyte percentages were independent prognostic factors for survival in training (n=513) and validation (n=355) cohorts. Therefore, reduced percentages of B-cells and non-classical monocytes are high-risk immune biomarkers for risk-stratification of COVID-19 patients.

Immunologic characterization of COVID-19 patients with hematological cancer.

Not available.