Rapid classification and prediction of COVID-19 severity by MALDI-TOF mass spectrometry analysis of serum peptidome

Classification and early detection of severe COVID-19 patients is urgently required to establish an effective treatment. Here, we tested the utility of matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) to classify and predict the severity of COVID-19 in a clinical setting. We used this technology to analyse the mass spectra profiles of the sera from 80 COVID-19 patients, clinically classified as mild (33), severe (26) and critical (21), and 20 healthy controls. We found a clear variability of the serum peptidome profile depending on COVID-19 severity. Seventy-eight peaks were significantly different and 12 at least four fold more intense in the set of critical patients than in the mild ones. Analysis of the resulting matrix of peak intensities by machine learning approaches classified severe (severe and critical) and non-severe (mild) patients with a 90% of accuracy. Furthermore, machine learning predicted correctly the favourable outcome of the severe patients in 85% of the cases and the unfavourable in 38% of the cases. Finally, liquid chromatography mass spectrometry analysis of sera identified five proteins that were significantly upregulated in the critical patients. They included serum amyloid proteins A1 and A2, which probably yielded the most intense peaks with m/z 11,530 and 11,686 detected by MALDI-TOF MS. In summary, we demonstrated the potential of the MALDI-TOF MS as a bench to bedside technology to aid clinicians in their decisions to classify COVID-19 patients and predict their evolution.