#2349 Increased baseline circulating ACE2 activity and Gas6 levels predict COVID-19 severity

Abstract Background and Aims The angiotensin converting enzyme 2 (ACE2) is the main cell entry receptor for SARS-CoV-2. SARS-CoV-2 participates in ACE2 shedding from cell membrane, increasing soluble ACE2. On the other hand, the receptor AXL also cooperates in the entry of SARS-CoV-2. The present study aims to evaluate if the initial serum levels of circulating ACE2 activity, sAXL and its ligand, the vitamin-K dependent protein Gas6, are related with the disease prognosis. Method Serum samples were collected from 2123 COVID-19 diagnosed patients, at two referent centers from Spain, on their first emergency between 2020 and 2021. Age, sex, previous comorbidities and clinical and analytical variables were recorded. Patients were categorized depending on their disease progression during hospitalization considering severe patients those who required intensive care unit (ICU) admission or died because of COVID-19. ACE2 activity was measured by a fluorescence enzymatic activity assay. Serum Gas6 and sAXL levels were measured by ELISA. Data was analyzed using RStudio. Results Patients with severe COVID-19 were older and with a major number of comorbidities. We used the cohort from one hospital (n = 1655)to perform multivariate cox regression model adjusted by age, sex, previous conditions and different analytical parameters (IL6, PCR, DD, lymphocytes count), and we found that elevated circulating ACE2 activity and higher levels of Gas6 were associated with a worse prognosis (HR: 1.23, 95% CI:1.02–1.48, p-value = 0.027; HR: 1.03, 95% CI: 1.01–1.04, p-value>0.001, respectively). In addition, we followed-up the patients for 30 days during hospitalization and carried out a Kaplan Meier Survival test that revealed that higher circulating ACE2 activity and levels of Gas6 were associated to worse survival (Log-rank = 0.0014 and Long-rank <0.001, respectively) (Fig. 1A). On the contrary, sAXL levels did not associate to a worse outcome of the disease. Finally, to analyze the predictive power of our model, we performed a time dependent ROC curve analysis obtaining an area under the curve (AUC) of 0.80 (Fig. 1B). To validate the results, the same study was also performed in the external cohort from the second hospital (n = 468) where the AUC was 0.79 providing a sensitivity of 84% and a specificity 66%. Conclusion Our results indicate that elevated circulating ACE2 activity and higher levels of Gas6 at admission are risk factors for severe COVID-19 disease. In addition, according to our multivariate cox model circulating ACE2 activity and higher levels of Gas6 at admission could be useful tool to predict COVID-19 severity.