Early treatment of COVID-19 with anakinra guided by soluble urokinase plasminogen receptor plasma levels: a double-blind, randomized controlled phase 3 trial.

Early increase of soluble urokinase plasminogen activator receptor (suPAR) serum levels is indicative of increased risk of progression of coronavirus disease 2019 (COVID-19) to respiratory failure. The SAVE-MORE double-blind, randomized controlled trial evaluated the efficacy and safety of anakinra, an IL-1α/ß inhibitor, in 594 patients with COVID-19 at risk of progressing to respiratory failure as identified by plasma suPAR ≥6 ng ml-1, 85.9% (n = 510) of whom were receiving dexamethasone. At day 28, the adjusted proportional odds of having a worse clinical status (assessed by the 11-point World Health Organization Clinical Progression Scale (WHO-CPS)) with anakinra, as compared to placebo, was 0.36 (95% confidence interval 0.26-0.50). The median WHO-CPS decrease on day 28 from baseline in the placebo and anakinra groups was 3 and 4 points, respectively (odds ratio (OR) = 0.40, P < 0.0001); the respective median decrease of Sequential Organ Failure Assessment (SOFA) score on day 7 from baseline was 0 and 1 points (OR = 0.63, P = 0.004). Twenty-eight-day mortality decreased (hazard ratio = 0.45, P = 0.045), and hospital stay was shorter.

Circulating Osteopontin Levels and Outcomes in Patients Hospitalized for COVID-19.

BACKGROUND: Severe coronavirus disease 2019 (COVID-19) is the result of a hyper-inflammatory reaction to the severe acute respiratory syndrome coronavirus 2. The biomarkers of inflammation have been used to risk-stratify patients with COVID-19. Osteopontin (OPN) is an integrin-binding glyco-phosphoprotein involved in the modulation of leukocyte activation; its levels are associated with worse outcomes in patients with sepsis. Whether OPN levels predict outcomes in COVID-19 is unknown. METHODS: We measured OPN levels in serum of 341 hospitalized COVID-19 patients collected within 48 h from admission. We characterized the determinants of OPN levels and examined their association with in-hospital outcomes; notably death, need for mechanical ventilation, and need for renal replacement therapy (RRT) and as a composite outcome. The risk discrimination ability of OPN was compared with other inflammatory biomarkers. RESULTS: Patients with COVID-19 (mean age 60, 61.9% male, 27.0% blacks) had significantly higher levels of serum OPN compared to healthy volunteers (96.63 vs. 16.56 ng/mL, p < 0.001). Overall, 104 patients required mechanical ventilation, 35 needed dialysis, and 53 died during their hospitalization. In multivariable analyses, OPN levels ≥140.66 ng/mL (third tertile) were associated with a 3.5 × (95%CI 1.44-8.27) increase in the odds of death, and 4.9 × (95%CI 2.48-9.80) increase in the odds of requiring mechanical ventilation. There was no association between OPN and need for RRT. Finally, OPN levels in the upper tertile turned out as an independent prognostic factor of event-free survival with respect to the composite endpoint. CONCLUSION: Higher OPN levels are associated with increased odds of death and mechanical ventilation in patients with COVID-19, however, their utility in triage is questionable.

Swarm Learning for decentralized and confidential clinical machine learning.

Fast and reliable detection of patients with severe and heterogeneous illnesses is a major goal of precision medicine1,2. Patients with leukaemia can be identified using machine learning on the basis of their blood transcriptomes3. However, there is an increasing divide between what is technically possible and what is allowed, because of privacy legislation4,5. Here, to facilitate the integration of any medical data from any data owner worldwide without violating privacy laws, we introduce Swarm Learning-a decentralized machine-learning approach that unites edge computing, blockchain-based peer-to-peer networking and coordination while maintaining confidentiality without the need for a central coordinator, thereby going beyond federated learning. To illustrate the feasibility of using Swarm Learning to develop disease classifiers using distributed data, we chose four use cases of heterogeneous diseases (COVID-19, tuberculosis, leukaemia and lung pathologies). With more than 16,400 blood transcriptomes derived from 127 clinical studies with non-uniform distributions of cases and controls and substantial study biases, as well as more than 95,000 chest X-ray images, we show that Swarm Learning classifiers outperform those developed at individual sites. In addition, Swarm Learning completely fulfils local confidentiality regulations by design. We believe that this approach will notably accelerate the introduction of precision medicine.