Diagnostic tools or biomarkers for the heterogenous post-acute COVID-19 syndrome (PASC; also known as long COVID) are needed. Research now published in Cell Host and Microbe shows that the gut microbiome is associated with phenotypic manifestations of PASC and could potentially be used to predict individual symptoms of PASC. Analyzing metagenomic and clinical data form three different cohorts of individuals with PASC and two cohorts of non-PASC control individuals, the authors find that different gut enterotypes associate with specific PASC symptoms. They used a multi-label machine learning model to predict individual symptoms of PACS using data covering hundreds of bacterial species and pathways. Their model achieved an average area under the curve (AUC) of 0.84, an average sensitivity of 86% and an average specificity of 82% for predicting the PACS symptoms in individuals with acute COVID-19 before they developed them. The associations found in this study underscore the role of the gut microbiome in PASC pathogenesis and their diagnostic or therapeutic potential.

Original reference: Cell Host Microbe 32, 651–660.e4 (2024)

需要针对异质性急性后 COVID-19 综合征 (PASC；也称为长期 COVID) 的诊断工具或生物标志物。目前发表在《Cell Host and Microbe》上的研究表明，肠道微生物组与 PASC 的表型表现相关，并且有可能用于预测 PASC 的个体症状。通过分析三个不同的 PASC 个体队列和两个非 PASC 对照个体队列的宏基因组和临床数据，作者发现不同的肠道肠型与特定的 PASC 症状相关。他们使用多标签机器学习模型，利用涵盖数百种细菌种类和途径的数据来预测 PACS 的个体症状。他们的模型在急性 COVID-19 患者出现 PACS 症状之前预测其 PACS 症状的平均曲线下面积 (AUC) 为 0.84，平均灵敏度为 86%，平均特异性为 82%。本研究中发现的关联强调了肠道微生物组在 PASC 发病机制中的作用及其诊断或治疗潜力。

原始参考文献： Cell Host Microbe 32 , 651–660.e4 (2024)