Exploratory Study on Application of MALDI-TOF-MS to Detect SARS-CoV-2 Infection in Human Saliva

Area of Science:

• Biotechnology
• Infectious Disease Diagnostics
• Computational Biology

Background:

• The COVID-19 pandemic necessitated rapid and accurate diagnostic methods.
• Current gold standard, RT-qPCR, has limitations including invasiveness, cost, and supply chain issues.
• Development of alternative, non-invasive diagnostic tools is crucial for managing outbreaks.

Purpose of the Study:

• To develop and evaluate a PCR-free diagnostic method for SARS-CoV-2 using MALDI-TOF MS and ML on saliva.
• To compare the performance of this novel method against RT-qPCR for COVID-19 diagnosis.
• To assess the diagnostic utility of MALDI-TOF MS and ML profiling of saliva across different time points post-infection.

Main Methods:

• Saliva samples were collected from COVID-19 positive patients and healthy controls at D0, D10, and D30.
• Matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) was used to generate spectral profiles.
• Machine learning models, including Support Vector Machine with Linear Kernel (SVM-LK), were trained and tested for classification.

Main Results:

• Nine spectral peaks significantly differentiated SARS-CoV-2 positive patients from controls at D0.
• The SVM-LK model achieved high accuracy (85.2%) on the training set and 88.9% on D0 independent samples.
• Performance decreased at later time points, with 80.8% accuracy at D30 and 64.3% at D10.

Conclusions:

• MALDI-TOF MS combined with ML offers a promising PCR-free supplementary tool for COVID-19 diagnosis.
• The method demonstrates strong performance for initial diagnosis and later stages (D30).
• Diagnostic accuracy is reduced for convalescent samples (D10), indicating potential limitations for this specific time window.