313. Host Protein Biomarkers Predicting Severity of Lung Damage due to COVID-19. (4th December 2021)

• Record Type: Journal Article
• Authors: Kerschbaumer, Isabell; Fazzari, Melissa; Burstein, Shana R; Furey, Aisha; Fox, Amy S; Cole, Adam M; Gendlina, Inessa; Gendlina, Inessa; Achkar, Jacqueline M
• Is Part Of: Open forum infectious diseases. Volume 8(2021)Supplement 1
• Page Start: S262
• Page End: S262
• Publication Date: 2021-12-04
• Subjects: Communicable diseases -- Periodicals; Medical microbiology -- Periodicals; Infection -- Periodicals; 616.9
• DOI: 10.1093/ofid/ofab466.515
• Languages: English
• ISSNs: 2328-8957

Abstract

Abstract: Background: Biomarkers to predict the severity of lung damage due to COVID-19 are urgently needed to inform management and treatment decisions. Our objective was to investigate the predictive value of host proteins for worsening respiratory failure in one of the by COVID-19 most affected and diverse patient populations in the US. Methods: We performed a prospective single-center cross-sectional study of 34 adult patients admitted to Montefiore Medical Center in the Bronx, New York, for respiratory symptoms due to PCR-confirmed COVID-19. Exclusion criteria were age < 21, history of prior SARS-CoV-2 infection, and/or underlying severe chronic lung diseases requiring home O2 and/or high dose steroids. We stratified and compared patients by whether they developed worsening respiratory failure, necessitating transfer to the intensive care unit (ICU) during their hospital stay. Using a custom Luminex Assay, we measured hospital admission serum concentrations of 8 host proteins, representing respiratory-associated epithelial (RAGE, SP-D, CC16), endothelial (Ang-2, vWF), and immune pathways (S100A12, ICAM-1, VCAM-1). Results: Except for race and WHO COVID-19 scores, demographics, co-morbidities, symptoms, and symptom duration were not statistically significantly different between patients requiring transfer to the ICU (n=15) and non-ICU patients (n=19). Higher log-transformed levels for 5/8 proteins (S100A12, ICAM-1, Ang-2, RAGE, SP-D) showed significant or marginally significant increased cause-specific hazard for ICU transfer (n=15). Estimated cumulative incidence functions further showed a significantly or near significantly increased risk for ICU transfer for patients with above the median values of S100A12 or ICAM-1 (p=0.013), Ang-2 (p=0.056) or RAGE (p=0.077), respectively (Figure 1). Host proteins predicting need for ICU transfer did not correlate strongly with other clinical laboratory markers for COVID-19 severity (CRP, LDH, D-Dimer, Fibrinogen, Ferritin). Figure 1. Patients with above median levels of host protein markers S100A12, ICAM-1, Ang-2, and RAGE have a significantly or near significantly increased risk for severe respiratory failure requiring transfer to the ICU. Comparison of estimated cumulative incidence at 7 days post admission for host protein markers above and below median levels for (A) S10012 (median 96, 675 pg/ml); (B) ICAM-1 (median (1, 192, 277 pg/ml); (C) Ang-2 (median 3463 pg/ml); (D) RAGE (median 6356 pg/ml); and (E) SP-D (median 11, 832 pg/ml). Conclusion: These results suggest that host proteins have additional predictive value for the severity of COVID-19-associated lung damage at time of presentation to the hospital. Disclosures: Inessa Gendlina, Nothing to disclose