Heparin-binding protein levels predict unfavorable outcome in COVID-19 pneumonia: A post-hoc analysis of the SAVE trial.

We aimed to evaluate heparin-binding protein (HBP) as a marker of prognosis of unfavorable outcome in COVID-19 pneumonia. This was a post-hoc analysis of the SAVE clinical trial investigating anakinra treatment, guided by suPAR (soluble urokinase plasminogen activator receptor) levels ≥6 ng/ml, for the prevention of severe respiratory failure (SRF) in hospitalized patients with COVID-19 pneumonia. Baseline HBP plasma levels were measured in 534 patients by fluorescence dry quantitative immunoassay using the Jet-iStar 800 analyzer. Concentrations higher than 35 ng/ml predicted 30-day mortality with a moderate specificity of 53.3% and negative predictive value 78.1%; sensitivity was low (29.0%). After multivariate Cox analysis HBP higher than 35 ng/ml was an independent predictor of 30-day unfavorable outcome (HRadj, 1.77; 95% CI, 1.06-2.94; p: 0.028) and these patients were also at greater risk of death after 90 days (HR, 1.85; 95%CI, 1.25-2.74; p:0.002). The cut-off was not predictive of development of SRF, septic shock or AKI. Among patients with baseline HBP levels higher than 35 ng/ml, anakinra treatment was associated with decreased mortality (7.2%) versus comparators (18.1%; p < 0.001). Results confirm that HBP may be an early biomarker of poor outcome among pre-selected patients at risk from COVID-19 pneumonia.ClinicalTrials.gov registration NCT04357366.