Early Treatment of COVID-19 Pneumonia with Anakinra Guided by Urokinase Plasminogen Receptor (preprint)

Early recognition of risk and start of treatment may improve unfavorable outcome of COVID-19. In the SAVE-MORE double-blind randomized trial, 594 patients with pneumonia without respiratory dysfunction at risk as defined by plasma suPAR (soluble urokinase plasminogen activator receptor) ≥ 6 ng/ml were 1:2 randomized to subcutaneous placebo or 100 mg anakinra once daily for 10 days; 85.9% were co-administered dexamethasone. After 28 days, anakinra-treated patients were distributed to lower strata of the 11-point World Health Organization ordinal Clinical Progression Scale (WHO-CPS) (adjusted odds ratio-OR 0.36; 95%CI 0.26–0.50; P < 0.001); anakinra protected from severe disease or death (≥ 6 points of WHO-CPS) (OR: 0.46; P: 0.010). The median WHO-CPS decrease in the placebo and anakinra groups was 3 and 4 points (OR 0.40; P < 0.0001); the median decrease of SOFA score was 0 and 1 points (OR 0.63; P: 0.004). 28-day mortality decreased (hazard ratio: 0.45; P: 0.045) and hospital stay was shorter. (Sponsored by the Hellenic Institute for the Study of Sepsis ClinicalTrials.gov identifier, NCT04680949)

Early Anakinra Treatment for COVID-19 Guided by Urokinase Plasminogen Receptor (preprint)

BackgroundIn a previous open-label trial, early anakinra treatment guided by elevated soluble urokinase plasminogen activator receptor (suPAR) prevented progression of COVID-19 pneumonia into respiratory failure. MethodsIn the SAVE-MORE multicenter trial, 594 hospitalized patients with moderate and severe COVID-19 pneumonia and plasma suPAR 6 ng/ml or more and receiving standard-of-care were 1:2 randomized to subcutaneous treatment with placebo or 100 mg anakinra once daily for 10 days. The primary endpoint was the overall clinical status of the 11-point World Health Organization ordinal Clinical Progression Scale (WHO-CPS) at day 28. The changes of the WHO-CPS and of the sequential organ failure assessment (SOFA) score were the main secondary endpoints. ResultsAnakinra-treated patients were distributed to lower strata of WHO-CPS by day 28 (adjusted odds ratio-OR 0.36; 95%CI 0.26-0.50; P<0.001); anakinra protected from severe disease or death (6 or more points of WHO-CPS) (OR: 0.46; P: 0.010). The median absolute decrease of WHO-CPS in the placebo and anakinra groups from baseline was 3 and 4 points respectively at day 28 (OR 0.40; P<0.0001); and 2 and 3 points at day 14 (OR 0.63; P: 0.003); the absolute decrease of SOFA score was 0 and 1 points (OR 0.63; P: 0.004). 28-day mortality decreased (hazard ratio: 0.45; P: 0.045). Hospital stay was shorter. ConclusionsEarly start of anakinra treatment guided by suPAR provides 2.78 times better improvement of overall clinical status in moderate and severe COVID-19 pneumonia. (Sponsored by the Hellenic Institute for the Study of Sepsis ClinicalTrials.gov identifier, NCT04680949)

Swarm Learning as a privacy-preserving machine learning approach for disease classification (preprint)

Identification of patients with life-threatening diseases including leukemias or infections such as tuberculosis and COVID-19 is an important goal of precision medicine. We recently illustrated that leukemia patients are identified by machine learning (ML) based on their blood transcriptomes. However, there is an increasing divide between what is technically possible and what is allowed because of privacy legislation. To facilitate integration of any omics data from any data owner world-wide without violating privacy laws, we here introduce Swarm Learning (SL), a decentralized machine learning approach uniting edge computing, blockchain-based peer-to-peer networking and coordination as well as privacy protection without the need for a central coordinator thereby going beyond federated learning. Using more than 14,000 blood transcriptomes derived from over 100 individual studies with non-uniform distribution of cases and controls and significant study biases, we illustrate the feasibility of SL to develop disease classifiers based on distributed data for COVID-19, tuberculosis or leukemias that outperform those developed at individual sites. Still, SL completely protects local privacy regulations by design. We propose this approach to noticeably accelerate the introduction of precision medicine.