Coronavirus Research Tracking - 6 November
Epidemic risks, diagnostic tests, more infectious strains
In this week’s Research Tracker we highlight papers assessing epidemic and pandemic risks, novel diagnostic tests, and experimental evidence for greater infectiousness of the D614G variant.
The Research Tracker is prepared by Dr Robert Hickson for the Science Media Centre.
Understanding zoonotic risks
The pandemic is prompting further consideration of how to better assess risks of emergent diseases, and how to prevent them.
A pre-print paper discusses how data currently used to estimate zoonotic risks (animal to human infections) are highly biased, incomplete, and rapidly changing. Assumptions about what viruses are important to human and animal health also bias surveillance decisions.
The authors propose a more systematic and comprehensive sampling strategy. This should focus on the species and environments most likely involved in transferring viruses to humans and livestock.
Reducing pandemic risks
A report assessing the risks of future pandemics was recently published by the Intergovernmental Platform on Biodiversity and Ecosystem Services. It estimates that there are more than half a million animal viruses with the potential to infect humans. At least five new diseases emerge in people every year, any of which has the potential to spread and become pandemic. The report argues that there is an urgent need to change from responding to emerging diseases to preventing them.
Models for new epidemics
A paper published in Infection, Genetics & Evolution proposes an alternative to what the authors’ consider a common hypothesis for cross-species infection. Instead of a “spill over” model they propose a “circulation” model. The former involves microbial pathogens infecting a new host due to an epidemic in the original host increasing the chance of interaction (a spill over). A fortuitous “pre-adaptation” to the new host then leads to an epidemic in it too.
The authors note that there is no evidence linking SARS-CoV-2, SARS-CoV, or MERS to epidemics in possible original hosts. The “circulation model” hypothesises periodic infections of other species even when there is no epidemic in the original host. These initial cross infections don’t usually spread. Occasionally, however, post-infection selection may allow the pathogen to evade the new host’s immune response and so create an epidemic.
There is no evidence to support this for SARS-CoV-2 at the moment, and the direct ancestor of SARS-CoV-2 hasn’t been discovered. In my view, there are other models to consider as well - a circulation model with pre-adaptation, and a spill over model with post-infection selection.
Cough diagnostics
A paper published in IEEE Open Journal of Engineering in Medicine and Biology describes developing a cough audio dataset and using neural networks to identify people infected with Covid-19. It reports high sensitivity and specificity, even in asymptomatic people.
A caveat for this research is that it didn’t appear to test accuracy of discrimination between people with Covid-19 and those with coughs due to flu or colds.
Using wearable sensor data in diagnoses
Research published in Nature found that data from wearable devices (smartwatches and activity trackers) can help Covid-19 diagnoses. Distinguishing Covid-19 infections was more accurate when such data was combined with self-reported symptoms than just relying on symptoms alone. Larger studies are required to validate the findings.
Quick CRISPR-based diagnostic test
A prototype rapid method for detecting SARS-CoV-2 RNA in nasal swab samples is described in the Proceedings of the National Academies of Science. It uses CRISPR-Cas12 technology alongside electric fields and microfluidics to produce results in under 40 minutes. The authors suggest that their method is more amenable to automation than other CRISPR-based methods currently being developed. Further development is required to make it suitable for point-of-care use.
Another 30 minute test, not involving CRISPR, was described in an earlier issue of the Proceedings of the National Academy of Sciences. Developing rapid accurate SARS-CoV-2 tests is a hot research area, but few have been approved for widespread use so far.
Infection risk from contaminated surfaces is low
A study (not yet peer reviewed) tested several hundred frequently touched surfaces for SARS-CoV-2. These included door handles, rubbish bin lids, and pedestrian crossing buttons. Less than 10% gave positive results, and risk of infection was calculated to be less than 5 in 10,000.
The authors suggest monitoring such surfaces could provide an early warning of increasing numbers of Covid-19 cases.
Experimental support for a more infectious variant
In several previous issues of the tracker we have noted the debate over whether one SARS-CoV-2 variant called D614G is more infectious than the original virus. Two papers from different research groups provide experimental evidence that it can be.
A not yet peer-reviewed paper reports that in otherwise identical viruses the variant with glycine at position 614 (D614G) binds more strongly to the ACE2 receptor than one with aspartic acid (D614). D614G also infected some bronchial and epithelial cell cultures more effectively. When hamsters and ferrets were infected with both variants D614G reproduced much more quickly, accounting for more than 90% of the viral sequences subsequently recovered from the animals.
The second study, published in Nature, also reports superior replication by D614G in selected cell cultures. Hamsters infected with D614G had higher viral levels in nasal and throat fluid samples than hamsters infected with D614, but viral loads in their lungs were similar.
Both studies confirm earlier reports that there appears to be no difference in the severity of disease in animals infected by either variant.
Another variant spreads
Last week a paper (not yet peer reviewed) described a new viral variant that appears to have arisen in Spain. Since July it has spread to many European countries. It is unknown whether this strain has a selective advantage, or is just highlighting patterns of travel.
Sustained antibody levels and recovery time
Research published in Cell describes an association between antibody levels and recovery time in people with mostly mild Covid-19 symptoms. While most showed rapid decline in IgG antibodies over three months, about one quarter had stable or increasing antibody levels. This group, called “quick healers”, recovered about six days quicker than the others.
They also had more mutations in virus-specific memory B cell antibody genes, and had higher levels of CD4+ effector memory T cells. This indicates that there may be a distinct immunophenotype associated with rapid recovery from Covid-19. A limitation of the study was that it involved only 75 participants, and most were Caucasian women.
Long read: How the coronavirus hacks the immune system
While long The New Yorker magazine has a very readable overview of the immune system and how the virus disrupts it.