Our understanding of the role of antibodies in infection was greatly increased this week. Researchers found that neutralising antibodies are made in most infected individuals, can be isolated from convalescent patients, and used to treat the infection in animal models. However, those with no symptoms had weaker antibody responses, with many losing their antibody response after two months.
Some new studies aim to better understand the antibody response to SARS-CoV-2 in people who have recovered from SARS-CoV-2. A study from Nature looked at the level of neutralising antibodies against the virus in 149 individuals who had been infected. They found that the majority of people had neutralising antibodies, but at a low level. Interestingly, different individuals had identical or similar neutralising antibodies that bound to the same region on the receptor binding domain of the coronavirus spike protein. These antibodies showed a very high level of neutralisation, despite the overall response being low, showing that a vaccine that targets these regions should be effective.
Another study isolated numerous neutralising antibodies from three convalescent COVID-19 patients. Some of these antibodies did not bind to the receptor binding domain, and so may not act via inhibition of the ACE-2 interaction. Many antibodies were potent neutralisers and some also neutralised SARS, as well as SARS-CoV-2.
A study in Nature Medicine looked at the level of antibodies in 37 asymptomatic individuals, and found that more than 80% of those with or without symptoms had IgG responses 3-4 weeks after exposure, but the titres were higher in those with symptoms. Titres declined in both groups by 8 weeks, so much so that 40% of the asymptomatic patients became seronegative, compared to 13% of the symptomatic group. Those without symptoms also had lower levels of circulating proinflammatory cytokines. The authors caution against the use of immunity passports, given the uncertain duration of this immunity, especially in those without symptoms.
A pair of papers in Science propose the use of an antibody cocktail to treat COVID-19 patients. The first paper described how these antibodies were generated, both from convalescent humans and humanised mice. The second paper showed that the spike protein of SARS-CoV-2 generates escape mutants when treated with antibodies, in a VSV pseudotype model in cell culture. Escape mutants, which prevent neutralisation, were not seen when a cocktail of four antibodies was used, as long as each bound to a different site in the protein.
A third paper in Science, from a different group, also isolated neutralising antibodies from convalescent patients. This study showed that neutralising antibodies protect Syrian hamsters from SARS-CoV-2 in an animal model of COVID-19.
Treatment of hospitalised COVID-19 patients with mavrilimumab, an antibody that blocks GM-CSF, was associated with improved clinical outcomes in a small prospective cohort study, showing that it may help reduce the hyperinflammatory symptoms seen in some severe patients.
The prone position (lying on the stomach) can improve oxygenation for patients hospitalised with COVID-19, according to a small study from a US hospital.
Another study found that hydroxychloroquine treated patients do no better than a control group.
6% of staff at a Belgian hospital tested positive for past infection with SARS-CoV-2, as measured by IgG antibodies (the IgM test had poor sensitivity). Caring for COVID-19 patients was not associated with a positive test result, but a household contact with the virus was, suggesting community rather than hospital transmission. 15% of those that tested positive had no symptoms.
Now that Wuhan is returning to work, the population is being screened for the virus by RT-PCR. The first such survey of almost 30,000 people found just 18 (0.061%) were positive. All of their contacts tested negative by either RT-PCR or serology.
Patient-collected lower nasal swabs were as effective as oropharangeal swabs taken by physicians, in a small study from the US. This could allow patients with suspected COVID-19 to test themselves, reducing costs and reducing the risk to healthcare workers.
A genome wide association study found two loci associated with COVID-19 respiratory failure in people from Spain and Italy. One locus contained six genes, including some known to be expressed in the lung. The second locus contained the ABO blood group, leading the researchers to find that blood group A was associated with a higher risk than the O group, although the mechanism for these effects, and whether they are anything more than an association, is unknown. There were no associations with HLA type, something that may have been expected for a viral infection.
At risk groups
In an extension of their earlier study for the UK, researchers estimated that 349 million people worldwide are at high risk of severe COVID-19 infection, 4% of the global population. 22% have at least one risk factor for severe infection. Less than 1% of those younger than 20 are at risk of severe disease, compared to 20% of those aged 70 years and above. Although the risks were lower for Africa, due to its younger population, the model does not take into account the availability of healthcare in each country.
The predicted severity of COVID-19 outbreaks in low and middle income countries was modelled. Low income countries tend to have a younger average age, which will reduce the impact of COVID-19, as well as less hospital beds, and fewer ventilators, which will both make the outbreak more severe.
Data from six countries allowed researchers to build a model of how the virus affects different age groups. Those under 20 years were half as susceptible to infection as those over 20 years. The rate of asymptomatic infection was higher in the younger age groups, with only 21% of teenagers showing symptoms, compared to 69% of those aged 70 and over. The model was tested against the outbreak in four further locations and found to fit the data well. Similarly, cities with a higher median age tended to have worse outbreaks, and school closures were modelled as being less effective than for influenza.
A model of virus incidence in each of the 3124 US counties found that rural counties are likely to be worse affected than urban areas, based on the predicted number of cases with symptoms, and the number of severe cases. Rural counties were more likely to have an older population, as well as less hospital beds.
More than 40% of Latinos in the Baltimore-Washington area tested positive for SARS-CoV-2 during a screening programme in hospitals and health clinics. This compared to 9% of whites and 18% of black patients. The sample was not representative or randomised, so it is difficult to extrapolate the findings, but they may indicate that Latinos are at a greater risk of catching the virus, perhaps due to their occupations.
Age and smoking were the biggest predictors of severe COVID-19 disease in a study of 1859 hospitalised patients from Wuhan.
Contact tracing would need to identify a high proportion of contacts in order to keep the reproductive number below 1, in a model of the UK population. Contact tracing with social distancing measures and self-isolation was more effective than contact tracing alone.
A model of the outbreak in Guangzhou found that the secondary attack rate of COVID-19 in households was between 12-17%. Older household contacts were more likely to catch the virus than younger household contacts.
Of the 96 asymptomatic carriers on the Diamond Princess cruise ship, 11 subsequently developed symptoms, showing that they were in fact presymptomatic.
40% of the US population were estimated to be essential workers, with healthcare the largest industry. Black individuals were overrepresented in several essential industries, including transportation, public administration, and healthcare. 13% of households with essential workers were at a high risk of economic vulnerability, defined as low income, lack of health insurance, or an older household member.
The built environment influences the risk of catching SARS-CoV-2, according to a screening study of pregnant women in New York City. Women were most likely to test positive if they lived in buildings with high household membership, and least likely to test positive if they lived in buildings of high value. Neighbourhoods with more infections tended to have higher unemployment rates and greater overcrowding, although population density was not associated with infection.
Patients with COVID-19 who underwent surgery had a worse outcome than those without COVID-19, in a study from a hospital in Italy. Pulmonary and thrombotic complications were both seen in these patients. The authors recommend that surgery should be postponed in COVID-19 patients.
Four potential new anti-virals were identified in a FRET-based enzymatic assay in tissue culture.
The only randomised controlled trial of cloth masks versus surgical face masks was published in the BMJ in 2015 (it showed that cloth masks did not protect from influenza like illness, and indeed more people in the cloth masks group became infected than the control group, without masks). The authors have now written a brief commentary on their article, re-iterating their findings, and emphasising that their article does not condone wearing no PPE – rather healthcare workers should wear not only a surgical mask, but gloves and goggles. The authors also cautioned that all face masks are single use, and cloth masks must be disinfected between uses.
Sterilisation with hydrogen peroxide or chlorine dioxide can be used to treat N95, KN95 (the Chinese version of an N95) and surgical face masks, but they reduce the filtration efficiency. Each mask and treatment performed slightly differently.
The cryo-EM structure of the spike protein of a pig and a human coronavirus, SADS-CoV and HKU2, may be useful for those working on the SARS-CoV-2 spike protein.
A national survey of more than 5000 in the US found that knowledge of coronavirus symptoms and spread was lowest amongst African American participants, men, and people younger than 55 years. These groups were also less likely to wash their hands and more likely to leave the house often. More effort is needed to provide information to these groups.
Acute hypoxic injury was seen in the brains of 18 COVID-19 patients with neurological symptoms. There was no evidence of encephalitis or other brain injury related to the virus. In contrast, three cases of encephalopathy and encephalitis in Atlanta, Georgia, USA were described, together with post mortem findings.
The clinical characteristics of hospitalised patients in Detroit were described; more than 70% were African American and more than 90% had a comorbidity.
9 cases of co-infection with SARS-CoV-2 and Clostridioides difficile occurred in Detroit, both of which can cause digestive symptoms.