Review 4: "Serum Sphingosine-1-Phosphate as Novel Prognostic and Predictive Biomarker for COVID-19 Severity and Morbidity and Its Implications in Clinical Management"

RR:C19 Evidence Scale rating by reviewer:

• Potentially informative. The main claims made are not strongly justified by the methods and data, but may yield some insight. The results and conclusions of the study may resemble those from the hypothetical ideal study, but there is substantial room for doubt. Decision-makers should consider this evidence only with a thorough understanding of its weaknesses, alongside other evidence and theory. Decision-makers should not consider this actionable, unless the weaknesses are clearly understood and there is other theory and evidence to further support it.

***************************************

Review:

The manuscript by Marfia et al. addresses the role of S1P as a potential biomarker for severity of Covid-19-infection in patients either being admitted at the ICU or remaining clinically stable. Despite the well-written manuscript and the sound investigation regarding the role of S1P as a biomarker of Covid-19-infection I have some major concerns regarding methodology of S1P-assessment, interpretation and discussion of the data.

1. S1P was assessed by an ELISA which is not the best established methodology of exact measurement of S1P-levels. In most studies so far the mass-spectrometric assessment of S1P-levels is propagated. Yet, the assessment of S1P by an ELISA may constitute an innovative method for the higher practicability of the assessment of S1P in clinical practice. In my opinion methodology of S1P-assessment should be discussed by the authors and the limitations of the ELISA should be stated clearly.

2. Since S1P associates with RBC- and PLT-levels as correctly stated by the authors it seems logical that throughout an acute severe infection, as this is the case for Covid-19, a transient infection-associated anemia develops and this associates with low RBC- and also with low S1P-levels. Interestingly, S1P is an independent predictor of the probability of ICU-admission as shown in the multivariate analysis in Table 6. However, the sample size of 22 patients with admission to the ICU-department limits the significance of the conclusions drawn. Furthermore, no data are available regarding therapeutic medications and supportive therapy applied to the patient sample during serum asservation. S1P-levels could be severely confounded by different treatment of the patient sample.

3. S1P reduction probably resembles an epiphenomenon of the severe inflammation seen in these patients. The authors discuss a mechanistic approach involving the pathophysiology of S1P as an etiologic factor affecting the clinical course of Covid-19 infection. From my point of view, it seems quite preliminary to draw such conclusions from the given data in the respective study.

In summary, despite my major concerns regarding the methodology of the study and the patient sample I rate the respective manuscript as potentially informative since it is so far the first study demonstrating the significant decrease of S1P levels in patients with severe Covid-19 infection. Yet, the current data are certainly not sufficient in order to enforce the statement that S1P could be a predictive biomarker for COVID-19 severity and prognosis and further studies are needed in the same direction.