Metabolomics Markers of COVID-19 Are Dependent on Collection Wave

Abstract

The effect of COVID-19 infection on the human metabolome has been widely reported, but to date all such studies have focused on a single wave of infection. COVID-19 has generated numerous waves of disease with different clinical presentations, and therefore it is pertinent to explore whether metabolic disturbance changes accordingly, to gain a better understanding of its impact on host metabolism and enable better treatments. This work used a targeted metabolomics platform (Biocrates Life Sciences) to analyze the serum of 164 hospitalized patients, 123 with confirmed positive COVID-19 RT-PCR tests and 41 providing negative tests, across two waves of infection. Seven COVID-19-positive patients also provided longitudinal samples 2-7 months after infection. Changes to metabolites and lipids between positive and negative patients were found to be dependent on collection wave. A machine learning model identified six metabolites that were robust in diagnosing positive patients across both waves of infection: TG (22:1_32:5), TG (18:0_36:3), glutamic acid (Glu), glycolithocholic acid (GLCA), aspartic acid (Asp) and methionine sulfoxide (Met-SO), with an accuracy of 91%. Although some metabolites (TG (18:0_36:3) and Asp) returned to normal after infection, glutamic acid was still dysregulated in the longitudinal samples. This work demonstrates, for the first time, that metabolic dysregulation has partially changed over the course of the pandemic, reflecting changes in variants, clinical presentation and treatment regimes. It also shows that some metabolic changes are robust across waves, and these can differentiate COVID-19-positive individuals from controls in a hospital setting. This research also supports the hypothesis that some metabolic pathways are disrupted several months after COVID-19 infection.

Keywords: COVID-19; LC-MS; machine learning; targeted metabolomics.

Figure 1

Orthogonal partial least squares discriminant analysis (OPLS-DA) of (A) all of the data (n = 164) color coded by RT-PCR test result comparing positive patients (n = 193) and negative patients (n = 41) and (B) only taking into account positive patients who were sampled within 14 days of the positive RT-PCR test (n = 97), and removing patients who were recovering from COVID-19 leaving only negative patients with no known history of COVID-19 (n = 37).

Figure 2

Left: Orthogonal partial least squares discriminant analysis (OPLS-DA) of (A) patients with positive RT-PCR tests in Wave 1 (n = 32) sampled from May 2020 to July 2020 compared to negative controls (n = 37) color coded by RT-PCR test result, (B) patients with positive RT-PCR tests in Wave 2 (n = 91) sampled from September 2020 to June 2021 compared to negative controls (n = 37) color coded by RT-PCR test result, and (C) patients with positive RT-PCR tests in Wave 1 (n = 32) and Wave 2 (n = 91) color coded by wave. Right: The top 10 VIP scores for each model, showing which metabolites are up and down regulated.

Figure 3

(Top) Venn diagram for the features identified by machine learning to be predictive for COVID-19 diagnosis in Wave 1, Wave 2 and across waves (where DG: diglycerides and TG: triglycerides); and (Bottom) Machine learning model prediction results in test set of Wave 1, Wave 2 and across waves, the 95% confidence intervals are enclosed in the parentheses.

Figure 4

Box plots (outliers shown as symbols outside of whiskers) to compare positive and negative patients for the 6 metabolites shown to differentiate across both waves (**—p < 0.01, ****—p < 0.001).

Figure 5

Box plots (outliers shown as symbols outside of whiskers) to show the 3 metabolites shown to differentiate across both waves for the 7 patients with longitudinal samples where (A) First sample taken in hospital, (B) average of samples over hospital stay, (C) longitudinal samples, (D) all negative patients and (E) Healthy controls (*—p < 0.05, **—p < 0.01, ns—not significant).