The association between SARS-CoV-2 infection and neuronal damage: A population-based nested case-control study (preprint)

ObjectiveTo assess whether severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection is associated with changes in plasma levels of neurofilament light chain (NfL), an extremely sensitive marker of neuroaxonal damage, in community-dwelling individuals. SettingThis study was embedded within the Rhineland Study, an ongoing community-based cohort study in Bonn, Germany DesignCross-sectional nested case-control study. ParticipantsParticipants were selected based on results from a previously conducted seroprevalence survey within the framework of the Rhineland Study. Cases were defined as those individuals who had had two positive confirmatory test results, including a recombinant spike-based immunofluorescence assay and a plaque reduction neutralization test (N=21). As controls, a random sample of individuals with a negative ELISA test result (Controls I, N=1117), and those with a borderline or positive ELISA test result who failed confirmatory testing (Controls II, N=68), were selected. Outcome measuresPlasma levels of NfL at the time of measurement, as well as change in plasma NfL levels compared to previously measured pre-pandemic levels ResultsAfter adjustment for age, sex and batch effects, serologically confirmed SARS-CoV-2 infection was neither associated with cross-sectional NfL levels, nor with the magnitude of change from pre-pandemic levels, compared to either of the two control groups. Similarly, after adjustment for age, sex and batch effects, self-reported neurological symptoms - including altered sense of smell or taste, headache, myalgia and fever - were not associated with changes in NfL levels in participants with a serologically confirmed SARS-CoV-2 infection (all p [≥] 0.56). ConclusionsOur findings indicate that mild-to-moderate coronavirus disease-19 is unlikely to be associated with a clinically relevant degree of neuroaxonal damage, even in those cases associated with neurological symptoms.

Seroprevalence and correlates of SARS-CoV-2 neutralizing antibodies: Results from a population-based study in Bonn, Germany (preprint)

Background Accurate estimates of SARS-CoV-2 seroprevalence are crucial for the implementation of effective public health measures, but are currently largely lacking in regions with low infection rates. This is further complicated by inadequate test performance of many widely used serological assays. We therefore aimed to assess SARS-CoV-2 seroprevalence in a region with low COVID-19 burden, especially focusing on neutralizing antibodies that presumably constitute a major component of acquired immunity. Methods We invited all individuals who were enrolled in the Rhineland Study, an ongoing community-based prospective cohort study in people aged 30 years and above in the city of Bonn, Germany (N=5427). Between April 24th and June 30th, 2020, 4771 (88%) of these individuals participated in the serosurvey. Anti-SARS-CoV-2 IgG levels were measured using an ELISA assay, and all positive or borderline results were subsequently examined through both a recombinant immunofluorescent assay and a plaque reduction neutralisation test (PRNT). Findings Seroprevalence was 0.97% (95% CI: 0.72-1.30) by ELISA and 0.36% (95% CI: 0.21-0.61) by PRNT, and did not vary with either age or sex. All PRNT+ individuals reported having experienced at least one symptom (odds ratio (OR) of PRNT+ for each additional symptom: 1.12 (95% CI: 1.04-1.21)). Apart from living in a household with a SARS-CoV-2 confirmed or suspected person, a recent history of reduced taste or smell, fever, chills/hot flashes, pain while breathing, pain in arms/legs, as well as muscle pain and weakness were significantly associated with the presence of neutralizing antibodies in those with mild to moderate infection (ORs 3.44 to 9.97, all p<0.018). Interpretation Our findings indicate a relatively low SARS-CoV-2 seroprevalence in Bonn, Germany (until June 30th, 2020), with neutralizing antibodies detectable in only one third of those with a positive immunoassay result, implying that almost the entire population in this region remains susceptible to SARS-CoV-2 infection.

Disease severity-specific neutrophil signatures in blood transcriptomes stratify COVID-19 patients (preprint)

The SARS-CoV-2 pandemic is currently leading to increasing numbers of COVID-19 patients all over the world. Clinical presentations range from asymptomatic, mild respiratory tract infection, to severe cases with acute respiratory distress syndrome, respiratory failure, and death. Reports on a dysregulated immune system in the severe cases calls for a better characterization and understanding of the changes in the immune system. Here, we profiled whole blood transcriptomes of 39 COVID-19 patients and 10 control donors enabling a data-driven stratification based on molecular phenotype. Neutrophil activation-associated signatures were prominently enriched in severe patient groups, which was corroborated in whole blood transcriptomes from an independent second cohort of 30 as well as in granulocyte samples from a third cohort of 11 COVID-19 patients. Comparison of COVID-19 blood transcriptomes with those of a collection of over 2,800 samples derived from 11 different viral infections, inflammatory diseases and independent control samples revealed highly specific transcriptome signatures for COVID-19. Further, stratified transcriptomes predicted patient subgroup-specific drug candidates targeting the dysregulated systemic immune response of the host.

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.