1
|
Rodríguez-Belenguer P, Piñana JL, Sánchez-Montañés M, Soria-Olivas E, Martínez-Sober M, Serrano-López AJ. A machine learning approach to identify groups of patients with hematological malignant disorders. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2024; 246:108011. [PMID: 38325024 DOI: 10.1016/j.cmpb.2024.108011] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 01/19/2023] [Revised: 12/06/2023] [Accepted: 01/07/2024] [Indexed: 02/09/2024]
Abstract
BACKGROUND AND OBJECTIVE Vaccination against SARS-CoV-2 in immunocompromised patients with hematologic malignancies (HM) is crucial to reduce the severity of COVID-19. Despite vaccination efforts, over a third of HM patients remain unresponsive, increasing their risk of severe breakthrough infections. This study aims to leverage machine learning's adaptability to COVID-19 dynamics, efficiently selecting patient-specific features to enhance predictions and improve healthcare strategies. Highlighting the complex COVID-hematology connection, the focus is on interpretable machine learning to provide valuable insights to clinicians and biologists. METHODS The study evaluated a dataset with 1166 patients with hematological diseases. The output was the achievement or non-achievement of a serological response after full COVID-19 vaccination. Various machine learning methods were applied, with the best model selected based on metrics such as the Area Under the Curve (AUC), Sensitivity, Specificity, and Matthew Correlation Coefficient (MCC). Individual SHAP values were obtained for the best model, and Principal Component Analysis (PCA) was applied to these values. The patient profiles were then analyzed within identified clusters. RESULTS Support vector machine (SVM) emerged as the best-performing model. PCA applied to SVM-derived SHAP values resulted in four perfectly separated clusters. These clusters are characterized by the proportion of patients that generate antibodies (PPGA). Cluster 1, with the second-highest PPGA (69.91%), included patients with aggressive diseases and factors contributing to increased immunodeficiency. Cluster 2 had the lowest PPGA (33.3%), but the small sample size limited conclusive findings. Cluster 3, representing the majority of the population, exhibited a high rate of antibody generation (84.39%) and a better prognosis compared to cluster 1. Cluster 4, with a PPGA of 66.33%, included patients with B-cell non-Hodgkin's lymphoma on corticosteroid therapy. CONCLUSIONS The methodology successfully identified four separate patient clusters using Machine Learning and Explainable AI (XAI). We then analyzed each cluster based on the percentage of HM patients who generated antibodies after COVID-19 vaccination. The study suggests the methodology's potential applicability to other diseases, highlighting the importance of interpretable ML in healthcare research and decision-making.
Collapse
Affiliation(s)
- Pablo Rodríguez-Belenguer
- Research Programme on Biomedical Informatics (GRIB), Department of Medicine and Life Sciences, Universitat Pompeu Fabra, Hospital del Mar Medical Research Institute, 08003 Barcelona, Spain
| | - José Luis Piñana
- Hematology Department, Hospital Clínico Universitario de Valencia, 46010 Valencia, Spain; Fundación INCLIVA, Instituto de Investigación Sanitaria Hospital Clínico Universitario de Valencia, 46010 Valencia, Spain
| | - Manuel Sánchez-Montañés
- Department of Computer Science, Escuela Politécnica Superior, Universidad Autónoma de Madrid, 28049 Madrid, Spain.
| | - Emilio Soria-Olivas
- IDAL, Intelligent Data Analysis Laboratory, ETSE, Universitat de València, 46100 Valencia, Spain
| | - Marcelino Martínez-Sober
- IDAL, Intelligent Data Analysis Laboratory, ETSE, Universitat de València, 46100 Valencia, Spain
| | - Antonio J Serrano-López
- IDAL, Intelligent Data Analysis Laboratory, ETSE, Universitat de València, 46100 Valencia, Spain
| |
Collapse
|
2
|
Merino-Vico A, Frazzei G, van Hamburg JP, Tas SW. Targeting B cells and plasma cells in autoimmune diseases: From established treatments to novel therapeutic approaches. Eur J Immunol 2023; 53:e2149675. [PMID: 36314264 PMCID: PMC10099814 DOI: 10.1002/eji.202149675] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/28/2022] [Revised: 09/27/2022] [Accepted: 10/27/2022] [Indexed: 02/02/2023]
Abstract
Autoimmune diseases are characterized by the recognition of self-antigens by the immune system, which leads to inflammation and tissue damage. B cells are directly and indirectly involved in the pathophysiology of autoimmunity, both via antigen-presentation to T cells and production of proinflammatory cytokines and/or autoantibodies. Consequently, B lineage cells have been identified as therapeutic targets in autoimmune diseases. B cell depleting strategies have proven beneficial in the treatment of rheumatoid arthritis (RA), systemic lupus erythematous (SLE), ANCA-associated vasculitis (AAV), multiple sclerosis (MS), and a wide range of other immune-mediated inflammatory diseases (IMIDs). However, not all patients respond to treatment or may not reach (drug-free) remission. Moreover, B cell depleting therapies do not always target all B cell subsets, such as short-lived and long-lived plasma cells. These cells play an active role in autoimmunity and in certain diseases their depletion would be beneficial to achieve disease remission. In the current review article, we provide an overview of novel strategies to target B lineage cells in autoimmune diseases, with the focus on rheumatic diseases. Both advanced therapies that have recently become available and more experimental treatments that may reach the clinic in the near future are discussed.
Collapse
Affiliation(s)
- Ana Merino-Vico
- Department of Experimental Immunology, Amsterdam University Medical Centers, University of Amsterdam, Amsterdam, Netherlands.,Department of Clinical Immunology and Rheumatology, Amsterdam Rheumatology and Immunology Center, Amsterdam University Medical Centers, University of Amsterdam, Netherlands
| | - Giulia Frazzei
- Department of Experimental Immunology, Amsterdam University Medical Centers, University of Amsterdam, Amsterdam, Netherlands.,Department of Clinical Immunology and Rheumatology, Amsterdam Rheumatology and Immunology Center, Amsterdam University Medical Centers, University of Amsterdam, Netherlands
| | - Jan Piet van Hamburg
- Department of Experimental Immunology, Amsterdam University Medical Centers, University of Amsterdam, Amsterdam, Netherlands.,Department of Clinical Immunology and Rheumatology, Amsterdam Rheumatology and Immunology Center, Amsterdam University Medical Centers, University of Amsterdam, Netherlands
| | - Sander W Tas
- Department of Experimental Immunology, Amsterdam University Medical Centers, University of Amsterdam, Amsterdam, Netherlands.,Department of Clinical Immunology and Rheumatology, Amsterdam Rheumatology and Immunology Center, Amsterdam University Medical Centers, University of Amsterdam, Netherlands
| |
Collapse
|
3
|
Agarbati S, Benfaremo D, Viola N, Paolini C, Svegliati Baroni S, Funaro A, Moroncini G, Malavasi F, Gabrielli A. Increased expression of the ectoenzyme CD38 in peripheral blood plasmablasts and plasma cells of patients with systemic sclerosis. Front Immunol 2022; 13:1072462. [PMID: 36618427 PMCID: PMC9811259 DOI: 10.3389/fimmu.2022.1072462] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/17/2022] [Accepted: 12/08/2022] [Indexed: 12/24/2022] Open
Abstract
Objective CD38 is a type II glycoprotein highly expressed on plasmablasts and on short- and long-lived plasma cells, but weakly expressed by lymphoid, myeloid, and non-hematopoietic cells. CD38 is a target for therapies aimed at depleting antibody-producing plasma cells. Systemic sclerosis (SSc) is an immune-mediated disease with a well-documented pathogenic role of B cells. We therefore analyzed CD38 expression in different subsets of peripheral blood mononuclear cells (PBMCs) from a cohort of SSc patients. Methods Cell surface expression of CD38 was evaluated on PBMCs from SSc patients using eight-color flow cytometry analysis performed with a FacsCanto II (BD). Healthy individuals were used as controls (HC). Results Forty-six SSc patients (mean age 56, range 23-79 years; 38 females and 8 males), and thirty-two age- and sex-matched HC were studied. Twenty-eight patients had the limited cutaneous form and eighteen the diffuse cutaneous form of SSc. The mean disease duration was 7 years. Fourteen patients were on immunosuppressive therapy (14 MMF, 5 RTX). The total percentages of T, B and NK cells were not different between SSc and HC. Compared to HC, SSc patients had higher levels of CD3+CD38+ T cells (p<0.05), higher percentage (p<0.001) of CD3+CD4+CD25+FOXP3+ regulatory T cells, lower percentage (p<0.05) of CD3+CD56+ NK T cells. Moreover, SSc patients had higher levels of CD24highCD19+CD38high regulatory B cells than HC (p<0.01), while the amount of CD24+CD19+CD38+CD27+ memory B cells was lower (p<0.001). Finally, the percentages of circulating CD38highCD27+ plasmablasts and CD138+CD38high plasma cells were both higher in the SSc group than in HC (p<0.001). We did not observe any correlations between these immunophenotypes and disease subsets or duration, and ongoing immunosuppressive treatment. Conclusions The increased expression of CD38 in peripheral blood plasmablasts and plasma cells of SSc patients may suggest this ectoenzyme as a candidate therapeutic target, under the hypothesis that depletion of these cells may beneficially downregulate the chronic immune response in SSc patients. Validation of this data in multicenter cohorts shall be obtained prior to clinical trials with existing anti-CD38 drugs.
Collapse
Affiliation(s)
- S. Agarbati
- Department of Clinical and Molecular Sciences, Marche Polytechnic University, Ancona, Italy
| | - D. Benfaremo
- Department of Clinical and Molecular Sciences, Marche Polytechnic University, Ancona, Italy,Clinica Medica, Department of Internal Medicine, Azienda Ospedaliero Universitaria delle Marche, Ancona, Italy
| | - N. Viola
- Immunologia Clinica, Department of Internal Medicine, Azienda Ospedaliero Universitaria delle Marche, Ancona, Italy
| | - C. Paolini
- Department of Clinical and Molecular Sciences, Marche Polytechnic University, Ancona, Italy
| | - S. Svegliati Baroni
- Department of Clinical and Molecular Sciences, Marche Polytechnic University, Ancona, Italy
| | - A. Funaro
- Department of Medical Sciences, University of Turin, Torino, Italy
| | - G. Moroncini
- Department of Clinical and Molecular Sciences, Marche Polytechnic University, Ancona, Italy,Clinica Medica, Department of Internal Medicine, Azienda Ospedaliero Universitaria delle Marche, Ancona, Italy,*Correspondence: G. Moroncini,
| | - F. Malavasi
- Department of Medical Sciences, University of Turin, Torino, Italy,Fondazione Ricerca Molinette, Torino, Italy
| | - A. Gabrielli
- Department of Clinical and Molecular Sciences, Marche Polytechnic University, Ancona, Italy
| |
Collapse
|
4
|
Werner A, Schäfer S, Gleußner N, Nimmerjahn F, Winkler TH. Determining immunoglobulin-specific B cell receptor repertoire of murine splenocytes by next-generation sequencing. STAR Protoc 2022; 3:101277. [PMID: 35434659 PMCID: PMC9010798 DOI: 10.1016/j.xpro.2022.101277] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/19/2022] Open
Affiliation(s)
- Anja Werner
- Division of Genetics, Department of Biology, Friedrich-Alexander-University Erlangen-Nürnberg (FAU), Erwin-Rommelstr. 3, 91058 Erlangen, Germany
- Corresponding author
| | - Simon Schäfer
- Division of Genetics, Department of Biology, Friedrich-Alexander-University Erlangen-Nürnberg (FAU), Glückstrasse 6, 91054 Erlangen, Germany
- Corresponding author
| | - Nina Gleußner
- Division of Genetics, Department of Biology, Friedrich-Alexander-University Erlangen-Nürnberg (FAU), Glückstrasse 6, 91054 Erlangen, Germany
| | - Falk Nimmerjahn
- Division of Genetics, Department of Biology, Friedrich-Alexander-University Erlangen-Nürnberg (FAU), Erwin-Rommelstr. 3, 91058 Erlangen, Germany
| | - Thomas H. Winkler
- Division of Genetics, Department of Biology, Friedrich-Alexander-University Erlangen-Nürnberg (FAU), Glückstrasse 6, 91054 Erlangen, Germany
- Corresponding author
| |
Collapse
|