Longitudinal immunological analyses reveal inflammatory misfiring in severe COVID-19 patients
bioinformatics clinical immunology/immunity inflammation therapeutics
Authors:Lucas, C.; Wong, P.; Klein, J.; Castro, T.B.R. et al.
Journal/ Pre-Print:medRxiv (preprint)
Tags: Bioinformatics, Clinical, Immunology/Immunity, Inflammation, Statistics, Therapeutics
SARS-CoV-2 triggers a maladaptive immune response, characterized by increased type 1, 2 and 3 responses.
Cytokine profile (in PBMCs) measured up to 12 days from symptoms onset predicts disease trajectory.
Unsupervised clustering analysis revealed the presence of 4 distinct immune signatures that correlated with three distinct disease trajectories in COVID-19 patients.
This study provides a comprehensive longitudinal analysis of immune responses in 113 COVID-19 patients, and includes both cellular and cytokine measurements. Efforts focused on characterizing the cytokine storm extensively reported to happen upon SARS-CoV-2 infection. By doing that, authors were able to identify a core COVID-19 signature of cytokine expression that is shared between patients that develop moderate or severe disease up to 10 days post onset of symptoms. Further investigation and unsupervised clustering analysis allowed the determination of 4 different immune signatures, which were used to cluster patients in 3 different groups that can be correlated with disease outcome. The trajectory analyses done in this study offer relevant information for better targeted treatment of COVID-19 patients, specially to what concerns therapeutic strategies to block anti-inflammatory cytokines and promote tissue damage repair.
Impact for SARS-CoV2/COVID19 research efforts
Understand the immune response to SARS-CoV2/COVID19
Longitudinal analysis allow a better understanding of the immune responses to SARS-CoV-2
Clinical symptoms and pathogenesis of SARS-Cov2/COVID19
Study provides some interesting findings: viral load is not significantly different between moderate and severe disease outcomes, but correlates with pro-inflammatory cytokines; cytokine expression profile in PBMCs up to 12 days after symptoms onset can predict disease trajectory; moderate disease is associated with a cluster of cytokines related to tissue repair and growth factors for lymphocytes and stromal cells while disease severity and coagulopathy is associated with a mix pro-inflammatory signature (type 1, 2 and 3).
Treat of SARS-CoV2/COVID19 positive individuals
The study contributes indirectly to the development of more efficient therapies because: 1. It provides a strategy for early identification of patients that can potentially develop severe disease; 2. Suggests that therapies focusing on anti-inflammatory cytokines must be combined to target type 1, type 2 and type 3 responses to potentially achieve better outcomes.
In silico study / bioinformatics study
Patient Case study
Strengths and limitations of the paper
Novelty: This is a robust longitudinal study, done with 113 patients (and 107 health controls), which were classified into moderate or severe depending on disease outcome. This allowed the identification of immune response signatures early after onset of symptoms that can predict disease severity. Furthermore, the study is using an unbiased approach analysing a variety of cytokines and growth factors which allows to highlight that severity is associated with increased type 2 responses (IL-5, IL-13), eosinophils and IgE production.
Standing in the field:The study corroborates several previous observations: 1. T cells numbers drop in severe patients, which is accompanied by an increase in monocytes; 2. Severity correlates with maladaptive immune responses, majorly due to excessive release of pro-inflammatory cytokines (cytokine storm).
Viral model used:blood (plasma and PBMCs) of patients infected with SARS-CoV-2
Translatability:the study profiles immune signatures that can be used to predict disease outcome, which has relevance to clinics, but also provides a comprehensive overview of the immune responses against SARS-CoV-2, offering an important resource for hypothesis generation and potential cytokines targets for therapies.
Main limitations: Therapies received by patients might influence immune responses beyond IL-6 and T cell profiling (which was already done by authors). Authors analysed PBMC, therefore their conclusions on eosinophils and neutrophils remain to be confirmed. The unsupervised analysis failed to identify a specific immune signature that can predict mortality as opposed to disease severity. It would have been interesting to analyse other clinical parameters to characterise further the difference between the patient groups identified by their unsupervised analysis.