Summary COVID-19 is highly variable in its clinical presentation, ranging from asymptomatic infection to severe organ damage and death. We characterized the time-dependent progression of the disease in 139 COVID-19 inpatients by measuring 86 accredited diagnostic parameters, such as blood cell counts and enzyme activities, as well as untargeted plasma proteomes at 687 sampling points. We report an initial spike in a systemic inflammatory response, which is gradually alleviated and followed by a protein signature indicative of tissue repair, metabolic reconstitution, and immunomodulation. We identify prognostic marker signatures for devising risk-adapted treatment strategies and use machine learning to classify therapeutic needs. We show that the machine learning models based on the proteome are transferable to an independent cohort. Our study presents a map linking routinely used clinical diagnostic parameters to plasma proteomes and their dynamics in an infectious disease. Graphical abstract Highlights • Plasma proteomes combined with clinical parameters characterize COVID-19 progression • Machine learning models allow highly precise prediction of the disease phenotype • The early molecular host response is predictive of COVID-19 progression • The molecular response to COVID-19 is age specific Demichev, Tober-Lau et al., present a time-resolved molecular map of the COVID-19, measuring plasma proteomes of patients with COVID-19 along with an extensive panel of clinical diagnostic parameters at 687-time points. They describe the specificity and dynamics, as well as the predictive and prognostic power of the molecular signatures in COVID-19.
【저자키워드】 COVID-19, proteomics, Biomarkers, machine learning, physiological parameters, patient trajectories, clinical disease progression, longitudinal profiling, disease prognosis, 【초록키워드】 machine learning, diagnostic, immunomodulation, host response, Infectious disease, progression, Protein, specificity, Cohort, therapeutic, asymptomatic infection, death, Prognostic marker, phenotype, plasma, age, proteome, molecular, prognostic, parameters, tissue repair, Clinical parameters, treatment strategy, blood cell count, machine learning models, followed by, blood cell, Predictive, Inpatient, COVID-19 progression, Treatment strategies, Abstract, systemic inflammatory response, blood cell counts, organ damage, enzyme activities, machine, disease phenotype, clinical parameter, parameter, independent, initial, identify, the disease, characterized, machine learning model, patients with COVID-19, 【제목키워드】 prognostic, proteomic,