To face the COVID-19 pandemic caused by SARS-CoV-2 infection that continues spreading worldwide andlead more than 4 million of deaths, many efforts are still being employed to better understand themolecular pathology of the different levels of COVID-19 severity and develop biomarker strategies enablingits early diagnosis. Knowing that the plasma secretome reflects the functional state of cells underpathological evolution, we used a label free mass spectrometry method to define a protein plasmasecretome signature of COVID-19 severe forms leading to intensive care unit (ICU) hospitalization andcompare the secretome signature to severe pneumonia not from COVID patients also leading to ICUhospitalization. This intrigued us to reveal specific SARS-CoV-2 infection effects on plasma content. For thispurpose, the proteome plasma profile of 33 patients hospitalized from severe pneumonia after a SARS-CoV-2 infection, 126 patients hospitalized from severe pneumonia without prior to SARS-CoV-2 infection and 20control patients were analyzed. The proteome profile was obtained after an enrichment of the lowabundant proteins (Proteominer® technology), a trypsin/Lys C digestion and a mass spectrometry analysisby using a NanoAcquity C18 and SYNAPT G2Si mass spectrometer system (Waters) operating in a highdefinition LC-MSE mode. Based on the quantification of 302 proteins (FDR 1%), we identified 57 significantlyderegulated proteins (DEP) in patients hospitalized after a SARS-CoV-2 infection. Several of those DEP areinvolved in complement activation, platelet degranulation that are known to contribute to the pathogenesisand other biological processes that are less known to contribute to the pathogenesis, such as lipoproteinsynthesis. Our approach contributes i) to identify several promising biomarkers for severe forms of SARS-CoV-2 and also ii) to distinguish the molecular alterations induced by SARS-CoV-2 infection.