Evaluation of methylation changes in blood cells of COVID-19 patients as a biomarker of severity of the infection.

Background: A number of studies have shown that methylomes of blood cells of COVID-19 patients display infection-related methylation changes. Those methylation abnormalities were associated with clinical outcomes of the infection and, thus, can potentially be utilized as biomarkers in clinical COVID-19 management. However, a number of parameters need to be assessed, before a clinical use of a biomarker candidate is proposed, with the most important one being reproducible detection in independent target populations.

Methods: We benchmarked data reported in publications investigating genome-wide methylation changes occurring in blood cells during SARS-CoV-2 virus infection to assess potential applicability of these changes as biomarkers in clinical management of COVID-19 patients.

Results: We identified nine studies, in which infection-related genome-wide methylation changes in blood cells of COVID-19 patients were reported and associations of those changes with outcomes of the infection was assessed. Our benchmarking analysis showed that each of those studies included patients with different clinical characteristics and each study utilized different symptoms assessment criteria. Most importantly, no uniform data analysis methodologies to identify virus-related methylation changes were used in the analyzed studies. Consequently, analyzed studies, except for one, which was based on uniform data analysis workflow applied across independent patient cohorts, reported sufficiently uniform results that would allow to propose the use of the identified infection-related methylation changes as biomarkers for clinical management of COVID-19 patients.

Conclusions: Due to lack of coherence between studies reporting SARS-CoV-2 associated methylation changes in blood, the potential use of already reported methylation changes as biomarkers for clinical management of COVID-19 patients needs to be further assessed in rigorously controlled studies. At the same time unified data analysis framework appears to be the most critical in development of biomarkers associated with the severity of SARS-CoV-2 infection.