This past week we learnt a lot! There were several major studies covering everything from remdesivir in monkeys, a new inactivated vaccine, people sleeping more during lockdown, how interferon causes lung damage, and low levels of immunity against SARS-CoV-2 in Wuhan, Geneva, and Hong Kong.
14% of US adults reported serious psychological stress in April 2020, compared to just 4% in 2018, according to a national survey of almost 1500 people. Levels of stress were highest in young adults, those on low incomes, and Hispanics. 14% of the surveyed population also reported that they always or often felt lonely.
It was therefore a surprise that there was a reduction, rather than the predicted increase, in people visiting hospitals with psychiatric symptoms in Boston during the COVID-19 outbreak there. This is most likely due to an unwillingness to visit the hospital, and shows that telemedicine needs to be implemented.
Two papers looked at the effect of lockdown on sleep patterns. Sleep duration was increased during lockdown in Austria, Germany, and Switzerland, most likely because there was no longer a difference between work days and free days, a phenomenon known as social jetlag. Sleep quality, however, was reduced. Sleep duration also increased by 30 minutes in a sample of students in the US, and differences between the weekday and the weekend were reduced. Exposure to natural daylight and physical exercise should help increase the quality of sleep.
Macaques given remdesivir did not show any signs of respiratory distress when infected with SARS-CoV-2. They also had 100-fold less virus in their lungs than the untreated animals. Remdesivir was given early during infection, unlike in human trials where the drug is only given during hospitalisation, often a week or more after infection.
The hepatitis C antiviral, sofosbuvir, may be effective against SARS-CoV-2, based on similarities in their polymerases.
Yet another study found that hydroxychloroquine did not reduce the risk of death from COVID-19, either with or without azithromycin, but instead increased the length of hospital stay. 807 patients from a veterans’ hospital were assessed.
An inactivated SARS-CoV-2 vaccine, BBIBP-CorV, was developed from an isolate taken from a hospitalised COVID-19 patient. The virus was passaged in Vero cells and then inactivated with β-propionolactone. This inactivated vaccine induced antibodies in several animal models and protected macaque monkeys from disease; there was no evidence of antibody mediated enhancement of disease.
Conserved epitopes predicted to bind to HLA were identified, some of which were shared between coronaviruses, which should aid rational vaccine design.
Two studies in Nature looked at the impact of non-pharmaceutical interventions, including social distancing on the reproductive number and on the number of deaths likely averted. The first study, from Imperial College London, modelled the outbreak by tracking deaths in 11 European countries using data from the European Centre for Disease Prevention and Control. They found that the reproductive number was below 1 in all 11 countries, showing that the epidemic was slowing. The percentage of the population infected so far ranged from 0.46% in Norway to 8% in Belgium. They estimated that if the outbreak had not been controlled there would have been 3.1 million additional deaths in these 11 countries.
The second study from UC Berkeley used econometrics to estimate the impact of non-pharmaceutical interventions in six countries with large outbreaks and varying levels of restrictions: China, France, Italy, Iran, South Korea, and the US. They identified 1717 individual interventions across the countries and estimated that together they reduced the number of total cases by 46–77 million.
Even modest social distancing helps to flatten the curve and reduce cases, according to an analysis of the outbreak in Seattle.
Lockdown in Italy significantly reduced human-made ambient noise, allowing researchers for the first time to distinguish between artificial and natural seismic activity. Now that baseline seismic energy has been measured, this provides a tool to measure the impact of lockdown – by the reduction in noise.
The first serology studies, which look at how much of a given population has immunity to the virus, are now being published, and together show the continued susceptibility of much of the world to SARS-CoV-2. Less than 4% of people in Wuhan have antibodies against SARS-CoV-2. Antibodies against SARS-CoV-2 were detected in healthcare workers, their relatives, hotel staff, and hospital patients in Wuhan and several other cities. The prevalence of antibodies ranged from 3.8% in healthcare workers in Wuhan to 0.6% of residents of Chengdu, Sichuan. The researchers validated their antibody test and found that 95% of individuals that tested positive for the virus also had detectable IgG antibodies by day 16, with IgM responses peaking later, in an unusual pattern of immunity. Their testing for both IgG and IgM was wise, as some individuals only tested positive for one and not the other.
A second study used weekly serosurveys of a representative population of 2766 people in 1339 households in Geneva to estimate that between 5% and 11% of Geneva have been infected with the virus. Different members of the panel were tested each week, and so the results varied week by week. For every officially reported case, based on RT-PCT testing, there were more than 11 true infections, based on serology. Children were less likely to test positive, despite often living with a family member who had tested positive; those over 65 years were also less likely to test positive.
A third study failed to find a single positive sample in 1938 people from Hong Kong, when tested by neutralisation assay, although 3% tested positive for IgG antibodies by ELISA. The sensitivity of the anti-spike and anti-nucleoprotein ELISAs was 58% and 67% respectively, whereas the neutralisation assay was 91% sensitive. In contrast, 4% of asymptomatic returnees from Hubei province tested positive by ELISA or neutralisation test, showing the risk of imported cases.
Another study found that of 44 hospitalised COVID-19 cases tested, all developed antibodies 6 days after testing positive for the virus by RT-PCR. Isotype switching occurred rapidly, to IgG1 and IgG3. A study of a single patient found that most antibody responses are non-neutralising, away from the receptor binding domain, but at least one was neutralising, and so may be useful for further studies.
Mice infected with an adenovirus expressing ACE-2 can be subsequently infected with SARS-CoV-2, providing a new animal model for the infection. The authors show that the interferon system, as well as T cells and neutralising antibodies, are necessary for virus clearance, as well as an upregulation of proinflammatory cytokines. Pooled plasma from human survivors led to rapid virus clearance, as did treatment with the anti-viral drug remdesivir.
A second paper in the same issue of Cell showed that the transduced mice accumulate virus in lungs and developed pneumonia and that passive transfer of monoclonal neutralising antibodies to the mice reduced disease.
Non-structural protein 9 (nsp9) of SARS-CoV-2 was crystallised and shown to be similar to that of SARS. An unexpected peptide-binding domain was discovered.
ORF8 of SARS-CoV-2 was predicted to be an immunoglobulin domain, and so may modulate host immunity.
A study of 420 healthcare workers who were deployed to Wuhan and who came into contact with COVID-19 patients found that none of them caught the virus during their deployment period. All were negative for symptoms, as well as negative by RT-PCR and serology, despite their close contact with patients, including aerosol generating procedures. All wore PPE.
Less than 1% of asymptomatic patients admitted to a Seattle hospital tested positive for SARS-CoV-2, after screening of all patients was introduced. For those with symptoms, 10% were positive for the virus.
38% of N95 respirators failed to fit properly, due to re-use by a sample of 68 healthcare staff, with duckbill masks more likely to fail than dome-shaped masks.
The COVID-19 outbreak in California was caused by at least 7 introductions into the state, including from Washington State within the US. There was limited transmission between communities, according to this genomic analysis from 36 patients.
Air travel volume from Wuhan correlated with the reported number of cases of COVID-19 in 49 locations across the world. Some locations with a high volume of travel from Wuhan have not yet reported large numbers of cases, suggesting the outbreak may be unrecognised there.
22 primary cases of COVID-19 caused more than 3000 cases of disease in Japan, grouped into 61 clusters. One third of the clusters were in healthcare facilities and most of the index cases were presymptomatic or asymptomatic at the time of transmission. As well as healthcare facilities, clusters were in care homes, restaurants or bars, and gyms.
Almost half the staff of a call centre in South Korea tested positive for SARS-CoV-2; none of their household contacts acquired the virus, due to effective contact tracing. Another outbreak occurred in sports facilities, also in South Korea.
Asymptomatic children with COVID-19 had high levels of viral RNA in saliva and faecal samples. Live virus was not tested for.
The incidence of SARS-CoV-2 in pregnant women in the UK was 4.9 per 1000 maternities. More than half of the pregnant women who tested positive were from a black, Asian or minority ethnic background and two thirds were overweight or obese. Five of the women (1%) died, and 5% of the infants tested positive by RT-PCR.
95% of pregnant patients with SARS-CoV-2 had no or mild COVID-19 disease, in a study from Spain. Caesarean delivery was associated with a worse outcome for the mother, and an increased risk of NICU admission for the infant.
Risk of severe disease
45% of adults in the US have at least one chronic condition that may put them at risk of severe COVID-19, much higher than seen in other countries so far. Age and D-dimer levels correlated with risk of severe disease in 1449 people hospitalised with COVID-19 in China.
The incidence of COVID-19 in children was less than 1%, according to a RT-PCR screen of paediatric surgical patients in the US. Half of the positive patients had no symptoms.
The immune response in 6 severely ill COVID-19 patients, including some with acute respiratory distress syndrome, was compared to 7 healthy controls. Several immune cell populations were depleted, and there was an increase in plasmablasts, part of the humoral immune response, as well as a novel immune cell population that the authors call developing neutrophils. Interestingly, they did not detect high levels of proinflammatory cytokines in peripheral monocytes, suggesting that these cells do not contribute to the putative cytokine storm observed in some patients.
Two studies in Science found that interferons produced during viral infection, especially IFN-λ, inhibited lung repair and increased the risk of a secondary bacterial infection, through the action of p53 and TLR-3. The first study looked at influenza infections in mice, and the second used synthetic viral RNA, and showed that IFN-λ is present in the lungs of COVID-19 patients. This cautions against the use of interferons as therapeutics against SARS-CoV-2.
The majority of global cases of SARS-CoV-2 have occurred along the 30° N to 50° N latitude corridor, where the temperature is between 5 and 11°C, and with low specific and absolute humidity. This may suggest seasonality of the virus in the future, although the study was written before the large increase in cases in Brazil, which lies outside this zone.
Hypokalaemia, or low potassium levels, were prevalent in COVID-19 patients in a study from China. The cause of this may be the loss of ACE-2 activity during infection, due to its interaction with the virus.
Diffuse alveolar damage and signs of thrombosis were the main pathology findings in lungs of fatal COVID-19 patients from Italy; diffuse alveolar damage was also the main finding in a study of fatal cases in the US.
More details of the clinical findings in paediatric inflammatory multisystem syndrome were described, with a study from New York and another from England. 58 children with paediatric inflammatory multisystem syndrome in England had persistent fever and inflammation, and many had a rash or diarrhoea, but only 22% met the diagnostic criteria for Kawasaki disease. The New York study looked at 17 children – all had prolonged fever, and most had elevated IL-6 and IL-10. Again the features overlapped with Kawasaki disease and toxic shock syndrome, but were distinct, suggesting that this is a new disorder.