The gut microbiome associates with phenotypic manifestations of post-acute COVID-19 syndrome, 2024, Su et al.

[SNT Gatchaman]

The gut microbiome associates with phenotypic manifestations of post-acute COVID-19 syndrome
Qi Su; Raphaela I. Lau; Qin Liu; Moses K.T. Li; Joyce Wing Yan Mak; Wenqi Lu; Ivan S.F. Lau; Louis H.S. Lau; Giann T.Y. Yeung; Chun Pan Cheung; Whitney Tang; Chengyu Liu; Jessica Y.L. Ching; Pui Kuan Cheong; Francis K.L. Chan; Siew C. Ng

SUMMARY
The mechanisms underlying the many phenotypic manifestations of post-acute COVID-19 syndrome (PACS) are poorly understood. Herein, we characterized the gut microbiome in heterogeneous cohorts of subjects with PACS and developed a multi-label machine learning model for using the microbiome to predict specific symptoms. Our processed data covered 585 bacterial species and 500 microbial pathways, explaining 12.7% of the inter-individual variability in PACS. Three gut-microbiome-based enterotypes were identified in subjects with PACS and associated with different phenotypic manifestations.

The trained model showed an accuracy of 0.89 in predicting individual symptoms of PACS in the test set and maintained a sensitivity of 86% and a specificity of 82% in predicting upcoming symptoms in an independent longitudinal cohort of subjects before they developed PACS. This study demonstrates that the gut microbiome is associated with phenotypic manifestations of PACS, which has potential clinical utility for the prediction and diagnosis of PACS.

HIGHLIGHTS

• The gut microbiome in subjects with post-acute COVID-19 syndrome (PACS) was analyzed

• Gut microbiome composition associates with the heterogeneity of PACS

• Gut enterotypes associate with distinct phenotypic manifestations of PACS

• Microbiome-based multi-label machine learning model accurately predicts PACS symptoms

Link | PDF (Cell Host & Microbe)

 

[SNT Gatchaman]

To validate the above findings in other independent cohorts, we trained a microbiome-based multi-class model based on a randomly selected training set (70%, n = 495) from the discovery cohort (n = 707) [...] which achieved an average area under the curve (AUC) of 0.96 (sensitivity 94.7% and specificity 98.3%) in the withheld test set (30%, n = 212)

We then performed the enterotype prediction using the trained model in the cohort of post-COVID subjects without PACS (n = 201) and the cohort of healthy controls without COVID-19 exposure (n = 653). We found that both E1 and E2 were significantly enriched in the PACS patients (p < 0.001), suggesting that these two enterotypes are associated with a higher risk of PACS.

In another longitudinal cohort, the results verified the associations between the prevalence of fatigue, memory loss, insomnia, and shortness of breath and the three enterotypes. These observations confirmed the robustness of our identified associations between gut-microbiome-based enterotypes and PACS symptoms.