Estudo do metaboloma global de amostras de urina analisadas por espectrometria de massas para investigação de biomarcadores associados à COVID-19

Abstract

The COVID-19 pandemic, caused by the SARS-CoV-2 virus, represented one of the greatest challenges to global public health, requiring effective diagnostic strategies to control transmission. Although the RT-PCR test is the gold standard, it involves invasive sampling, which leaves room for the search for alternative matrices. Urine stands out as a promising biological sample due to its non-invasive collection, ease of handling, and representative composition of an individual's health status. In this context, this study applied the global metabolomics approach to investigate metabolites associated with COVID-19 in samples using high-performance liquid chromatography coupled with mass spectrometry (HPLC-MS). One hundred samples from volunteers were used, divided into a Test Group (38 positive for COVID-19, confirmed by RT-PCR) and a Control Group (62 negative). The samples underwent viral inactivation and protein precipitation prior to analysis in both positive and negative ionization modes. Data processing was performed on the XCMS Online and MetaboAnalyst 6.0 platforms. After filtering signals from analytical blanks and molecular features with low reproducibility (RSD > 30%), the data were normalized by creatinine and median, logarithmically transformed, and Pareto scaled. PCA and PLS-DA models were applied. PCA analysis showed considerable overlap between groups, indicating the absence of natural separation in the data. PLS-DA exhibited signs of overfitting, suggesting limited predictive performance. Fold Change analysis (FC > 2) identified m/z and retention time pairs with relevant intensity variations between groups; however, when combined with statistical significance criteria (p-value < 0.05), no significant differences were observed, indicating limited statistical discrimination under the tested protocols. Cluster analyses reinforced the absence of clear separation, suggesting that the similarity between samples may be the result of individual factors that were beyond the scope of this work. It is concluded that, although urine presents a viable matrix, metabolic complexity and individual variability pose challenges for the definition of metabolites suitable for diagnostic purposes, highlighting the need for future studies that incorporate additional clinical variables to achieve a more comprehensive understanding of the metabolic response to COVID-19.