29-mRNA host response signatures for classification of bacterial infection, viral infection and disease progression in COVID-19 pneumonia: a post hoc analysis of the SAVE-MORE randomized clinical trial

Résumé fourni par la source

Abstract Background Biomarkers based on host response signatures are currently under development for the critically ill. We applied a 29-mRNA classifier for the diagnosis and prognosis of suspected acute infection and sepsis (TriVerity™, Inflammatix Inc.) in patients hospitalized with COVID-19. Methods We applied three scores from locked classifiers (IMX-BVN-4 and IMX-SEV-4) from the 29-mRNA TriVerity™ blood test in participants of the SAVE-MORE randomized clinical trial (ClinicalTrials.gov NCT04680949) at baseline and days 4 and 7 of treatment, to classify bacterial infection, viral infection and decompensation. Participants were adults hospitalized with confirmed COVID-19 pneumonia and plasma soluble urokinase plasminogen activator receptor (suPAR) levels of ≥ 6 ng/ml, randomized to placebo or anakinra treatment. Results A total of 471 patients were studied. At baseline nearly 90% had a Very Low or Low IMX-BVN-4 Bacterial Score and Moderate, High or Very High IMX-BVN-4 Viral Score. Anakinra treatment had an effect on the expression of genes indicating IMX-SEV-4 High or Very High scores after a 7 day treatment compared to baseline (12.9% of anakinra-treated patients continued being classified as high severity vs 20.4% of placebo-treated patients, p 0.046). Conclusions The classifiers were well tested in COVID-19 pneumonia and may become a useful tool for hospitalized patients.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.