Evaluation of Potential Prognostic Biomarkers for Identifying Adults Hospitalised with COVID-19 at Increased Risk of Adverse Clinical Outcomes.

INTRODUCTION: Elevated levels of circulating cytokines and immune cells are often observed in patients with coronavirus disease 2019 (COVID-19), some of which have been associated with disease severity and clinical outcomes across various medical interventions. We evaluated 50 biomarkers and immune cell types in blood from patients in a randomised controlled trial and hypothesised that baseline inflammatory cytokines and haematologic markers would be associated with COVID-19 severity and outcomes, and a composite biomarker score derived through statistical and machine learning capturing complementary immune pathways could provide stronger prognostic value than individual biomarkers. METHODS: COMMUNITY was a phase 3, randomised, adaptive-platform study in adult patients hospitalised with COVID-19 with clinical severity grade 2-5 (8-point ordinal scale), enrolled November 24, 2020, to June 4, 2021. An exploratory biomarker analysis used clinical laboratory assessments and serum cytokine data collected from baseline up to day 29 (or discharge) from patients randomised to placebo or apremilast + standard of care (baseline only). Healthy volunteers and ambulatory patients with COVID-19 (not part of the study) were evaluated for comparison. RESULTS: Interleukin (IL)-6, IL-8, IL-10, and tumour necrosis factor alpha (TNFα) were higher at baseline in 378 hospitalised patients than healthy volunteers or ambulatory patients. These cytokines and several haematological markers (including neutrophil count and percentage) showed positive associations with clinical severity and were inversely associated with clinical outcomes. A composite score was derived from baseline IL-6 levels and percentage of neutrophils and showed a strong association with patient overall survival. With a tertile cutoff at which composite scores ≤ 72.1 indicate high-risk patients, the estimated hazard ratio of mortality from COVID-19 infection was 14.3 (95% confidence interval 4.1, 49.3; P < 0.0001). CONCLUSION: This novel composite score was strongly associated with overall survival in adults hospitalised with COVID-19 and showed improved performance over its individual components. TRIAL REGISTRATION: EudraCT 2020-002594-10; ClinicalTrials.gov identifier, NCT04590586 ( https://clinicaltrials.gov/study/NCT04590586 ).