A model including CD15, ACE2 and age efficiently predicts COVID-19 severity

The COVID-19 pandemic presents a spectrum of clinical outcomes ranging from respiratory conditions to cardiovascular complications that challenge management and resource allocation. Identification of early predictive biomarkers that are easy to detect is a priority to optimize medical care and resources. ACE2 and TMPRSS2 have received special attention due to their role in viral infectivity but also due to their physiological anti-inflammatory activities. CD15 and CD45 are key proteins in the immune response also associated with SARS-CoV-2 response. This study focused on analyzing the expression of ACE2, TMPRSS2, CD15, and CD45 in a cohort of 216 patients (111 mild and 105 severe disease) to ascertain their potential as biomarkers for predicting disease severity. We aimed to assess the correlation between these markers and the severity of symptoms, utilizing qPCR and flow cytometry. We used mixed-effects linear regression models and Receiver Operating Characteristic (ROC) curves to test the performance of the biomarkers in the prediction of the severity of the disease. Significant lower surface expression of CD15 and ACE2 was observed in severe cases in addition to a strong association between aging and the severity of the disease. By integrating these findings, we developed a predictive model achieving 92.9% specificity and 79.3% sensitivity (AUC = 0.91; 95% CI: 0.87-0.96). The study concludes that our combined biomarker model could significantly enhance the management of COVID-19 by enabling early identification of patients at risk for severe outcomes, thus improving treatment strategies and resource distribution.