This week we saw the first results from Moderna's mRNA based vaccine, learnt that past infection with other coronaviruses including the common cold may offer some immunity against SARS-CoV-2, and saw the benefits of face masks and universal testing to prevent transmission in hospitals and care homes.

Vaccines

A phase 1 trial of mRNA-1273, an mRNA based vaccine from Moderna encoding the spike protein of SARS-CoV-2 showed that the vaccine induced an immune response. More than half of the participants had mild adverse effects, such as fatigue, chills, and headache, but the vaccine was otherwise safe. The trial included just 45 healthy adults, and did not test whether vaccination protected from infection, so further research is needed.

Immunology

Past infection with SARS generates T cells that recognise parts of SARS-CoV-2, even after 17 years after the original infection. T cell reactivity against SARS-CoV-2 was also seen in some people who had no past history of SARS or SARS-CoV-2, suggesting that another betacoronavirus infection, such as from a common cold coronavirus, may offer some T cell based immunity. The T cell response targeted the viral nucleoprotein and non-structural proteins and at least one of the epitopes was not seen in known common cold coronaviruses, suggesting that past infection with an as-yet unknown coronavirus may offer some protection from SARS-CoV-2. Further research is needed.

Circulating follicular helper T cells (cTFH) that target the spike protein of SARS-CoV-2 were identified, and their activity shown to correlate with the potency of the neutralising antibody response. cTFH may therefore be a biomarker of potency for vaccines entering the clinic.

Three “immunotypes” were identified in recovered patients based on their immune response to SARS-CoV-2. Some patients had a robust T and B cell response and others did not, with implications for immunity and vaccine development. Another study found that severe COVID-19 patients had an impaired type I interferon response, and an active proinflammatory response.

A proteome microarray was used to show that IgM and IgG antibodies are generated against several SARS-CoV-2 proteins, including the spike, the N protein, and ORF9b and NSP5.

The structural motifs that determine binding of the coronavirus spike protein to antibodies were identified through co-crystallisation.

Social distancing

A model of the outbreak in France found that physical distancing and mask wearing will not prevent a subsequent outbreak when lockdown is lifted, but that these interventions could be combined with shielding of vulnerable people to reduce fatal cases and not overwhelm ICU capacity.

Earlier implementation of lockdown was associated with the greatest decrease in COVID-19 incidence, according to an analysis of various interventions in 149 countries or regions across the world.  Physical distancing reduced COVID-19 incidence by 13% and public transport closure didn’t have any additional effect once other measures were in place.

Transmission

A comprehensive analysis of the outbreak in Wuhan using a new model named SAPHIRE found that 87% of cases before 8th March were undiagnosed, suggesting an important role for asymptomatic and mild cases. They also estimated the reproductive number at the start of the Wuhan outbreak as 3.54, which reduced to 0.28 after interventions. The role of those with no or mild symptoms makes continued surveillance challenging.

Whole genome sequencing of SARS-CoV-2 was used to track the outbreak in near to real-time in the Netherlands, and informed the public health response. Most early cases were in the south of the country and were imported from Italy and Austria, especially from ski resorts.

Economics

Gasoline demand has fallen as lockdown has been implemented across the world. A new machine-learning based model predicts that gasoline demand will recover by October 2020.

Risk of severe disease

An online calculation tool for patient triage at hospital admission was developed using a collection of 10 markers of severe disease. Another prediction model was developed using 7 machine learning techniques.

Treatments

Single chain Llama antibodies, known as nanobodies, can be generated against the coronavirus spike protein and block its interaction with ACE-2, and so could be used as a potential therapeutic.

Remdesivir binds to the active site of most coronavirus polymerases and reduces viral loads in mice, providing more support for its efficacy as a potential anti-viral.

Genomics

A nomenclature for strains of SARS-CoV-2 was developed, to allow researchers to analyse and discuss mutations in the virus as they are discovered. Viral lineages were divided into active, unobserved, or inactive to determine which strains are still causing infections and into two main lineages: A and B, with lineage A more closely related to the presumed ancestor virus in bats. At the time of writing, there are 81 viral lineages divided into A, B, and B.1, with B.1 the predominant global lineage.

Polymorphisms in the ACE-2 and TMPRSS2 genes were identified, and may affect the risk of catching the virus.

Virology

An interactive online platform of SARS-CoV-2 host interactions and potential drug target interactions was published. Named CoVex, they used the platform to identify several potential drugs that were predicted to target these host-virus interactions.

Pregnancy

The incidence of stillbirth at St George’s Hospital in London increased during the pandemic, but none of the mothers affected were infected with SARS-CoV-2. This suggests that an unwillingness to attend hospital or some other indirect factor was responsible.

A single case of transplacental transmission of SARS-CoV-2 was observed, and was associated with neurological symptoms in the baby. Vertical transmission from mother to child seems to be uncommon, but possible.

Healthcare

21% of healthcare workers at Mass General Brigham, a US hospital, tested positive for SARS-CoV-2 before mask wearing was implemented, falling to 11% after universal masking of all healthcare workers and patients. This was despite cases continuing to rise in the community, suggesting that mask use was effective at reducing hospital transmission.

Closing schools risks increasing absenteeism amongst healthcare workers who have children. A model of this in the US predicts that it would be cheaper to provide child care to all healthcare workers than bear the costs of healthcare worker absenteeism during school closures.

Diagnostics

Universal testing in care homes, rather than testing based on symptoms, found that almost 40% of residents were positive for SARS-CoV-2, more than half of whom had no symptoms. This shows the importance of universal testing of those at risk, and not basing testing on the presence of symptoms.

Another study found that symptom based screening of patients for COVID-19 is not effective at discriminating between patients infected with SARS-CoV-2 or other respiratory viruses, according to a study from a US health system.

A rapid point of care nucleic acid amplification testing protocol was developed and used to screen hospital patients, reducing time-to-bed placement by 6 hours.

Fatality rate

The mortality rate in hospitals in Lombardy, Italy, was very high, at 53%, with 87% requiring invasive mechanical ventilation. Older age and male sex were both associated with mortality, as was a history of COPD, high cholesterol, and type II diabetes.

In the US the mortality rate of those admitted to hospital was 35%, but there was wide variation between hospitals. Older age, male sex, obesity, coronary artery disease, and cancer were all associated with mortality, as was admission to a hospital with fewer intensive care beds. Another US study found that 17% of patients died, and also looked at risk factors for severe disease.

Clinical findings

Two studies described the disease course in patients with COVID-19 and tuberculosis, one from Singapore and another from China, highlighting the challenges in managing co-infections of these respiratory diseases.