Fast and reliable detection of patients with severe and heterogeneous illnesses is a major goal of precision medicine^1,2. 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 legislation^4,5. 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.

Invasive fungal infections are very severe infections associated with high mortality rates, despite the availability of new classes of antifungal agents. Based on pathophysiological mechanisms and limited pre-clinical and clinical data, adjunctive immune-stimulatory therapy with interferon-gamma (IFN-γ) may represent a promising candidate to improve outcome of invasive fungal infections by enhancing host defence mechanisms.

Significance: Hitherto, both the autonomic nervous system and innate immune system were regarded as systems that cannot be voluntarily influenced. The present study demonstrates that, through practicing techniques learned in a short-term training program, the sympathetic nervous system and immune system can indeed be voluntarily influenced. Healthy volunteers practicing the learned techniques exhibited profound increases in the release of epinephrine, which in turn led to increased production of anti-inflammatory mediators and subsequent dampening of the proinflammatory cytokine response elicited by intravenous administration of bacterial endotoxin. This study could have important implications for the treatment of a variety of conditions associated with excessive or persistent inflammation, especially autoimmune diseases in which therapies that antagonize proinflammatory cytokines have shown great benefit.

Rationale: Sepsis is characterized by a dysregulated immune response to infection. Norepinephrine, the cornerstone vasopressor used in septic shock, may contribute to immune dysregulation and impact host defense. Objectives: To investigate effects of norepinephrine and the alternative vasopressor vasopressin on the immune response and host defense. Methods: Leukocytes from six to nine donors were stimulated in the presence or absence of norepinephrine and vasopressin. A total of 190 C57BL/6J mice received a continuous infusion of norepinephrine or vasopressin via microosmotic pumps and were challenged with LPS or underwent cecal ligation and puncture. Thirty healthy volunteers were randomized to a 5-hour infusion of norepinephrine, vasopressin, or saline and intravenously challenged with LPS. The relationship between the norepinephrine infusion rate and the use of β-blockers and plasma cytokines was assessed in 195 patients with septic shock. Measurements and Main Results: Norepinephrine attenuated the production of proinflammatory mediators and reactive oxygen species and augmented antiinflammatory IL-10 production both in vitro and in LPS-challenged mice. Norepinephrine infusion during cecal ligation and puncture resulted in increased bacterial dissemination to the spleen, liver, and blood. In LPS-challenged volunteers, norepinephrine enhanced plasma IL-10 concentrations and attenuated the release of the proinflammatory cytokine IFN-γ–induced protein 10. Vasopressin exerted no immunomodulatory effects across these experimental setups. In patients, higher norepinephrine infusion rates were correlated with a more antiinflammatory cytokine balance, whereas β-blocker use was associated with a more proinflammatory cytokine balance. Conclusions: Norepinephrine dysregulates the immune response in mice and humans and compromises host defense. Therefore, it may significantly contribute to sepsis-induced immunoparalysis, whereas vasopressin does not have untoward immunologic effects.

Septic shock is a major cause of death worldwide and a considerable healthcare burden in the twenty-first century. Attention has shifted from damaging effects of the proinflammatory response to the detrimental role of antiinflammation, a phenomenon known as sepsis-induced immunoparalysis. Sepsis-induced immunoparalysis may render patients vulnerable to secondary infections and is associated with impaired outcome. The immunoparalysis hypothesis compels us to reevaluate the current management of septic shock and to assess whether we are inadvertently compromising or altering the host immune response. In this perspective, we discuss the potential detrimental role of norepinephrine, the cornerstone treatment for septic shock, in sepsis-induced immunoparalysis. We provide a short overview of the current understanding of the immunologic pathophysiology of sepsis, followed by a detailed description of the immunomodulatory effects of norepinephrine and alternative vasopressors. We conclude that although the development of novel therapies aimed at reversing immunoparalysis is underway, the use of norepinephrine may aggravate the development, extent, and duration of sepsis-induced immunoparalysis. Current in vitro and animal data indicate that norepinephrine treatment exerts immunosuppressive and bacterial growth–promoting effects and may increase susceptibility toward infections. However, evidence in humans is circumstantial, as immunologic effects of norepinephrine have not been investigated properly in experimental or clinical studies. Alternatives such as vasopressin/selepressin, angiotensin II, and phenylephrine could have a fundamental advantage over norepinephrine with respect to their immunologic properties. However, also for these agents, in vivo immunologic data in humans are largely lacking. As such, human studies on the immunomodulatory properties of norepinephrine and viable alternatives are highly warranted.