Coronavirus Research Tracking - 4 March
Three vaccine doses, long Covid & vaccination, vaccine safety & effectiveness in children, pandemic origins
This week, more research on the effectiveness of three doses, reduced risk of Long Covid after vaccination, vaccine effectiveness and safety in children and adolescents. Plus, papers on the lesser severity of Omicron compared with Delta, and factors that reduce the risk of household transmission.
In non-vaccine papers, results that support an animal market as the trigger point for the pandemic.
The tracker is shared with the COVID-19 Vaccine Media Hub.
The Research Tracker is prepared by Dr Robert Hickson for the Science Media Centre.
Vaccine-related papers
Three doses of Moderna show lower effectiveness against Omicron infections, compared with Delta
Three doses of the Moderna vaccine were found to be very effective at stopping Delta variant infections in adults, but less so against Omicron. Sixty days after the third dose effectiveness against Delta was 86%, and only 47.4% for Omicron. Effectiveness was slightly higher for those under 65, compared with those over that age. For immunocompromised people the three dose effectiveness against Omicron infection was even lower, 29.4%
Effectiveness against hospitalisation was high (nearly 100%) for both variants. Over 26,000 people with Covid-19 were included in the study. The paper was published in Nature Medicine.
Some decline in effectiveness of third doses against Omicron infection after one month
A third vaccine dose improved protection against symptomatic infection, but protection began to decline a little after four weeks, according to a UK study. Vaccine effectiveness against Omicron declined substantially after the second vaccine dose. A third dose improved effectiveness, although effectiveness was still substantially higher than at 10 weeks after the second dose.
The participants included those who had received two doses of the AstraZeneca/Oxford, Pfizer/BioNTech, or Moderna vaccines. They subsequently received the same or a different vaccine as the third dose 6 months or more after the second. Direct comparisons between primary doses were difficult because the groups receiving them were not always the same, and they were not all vaccinated at the same time. The paper was published in the New England Journal of Medicine.
Reduced risk of long Covid if vaccinated beforehand
A UK study found that the risk of long Covid is about 41% lower in people who had had two vaccine doses at least two weeks before their infection. For double-vaccinated (AstraZeneca/Oxford, or mRNA vaccines) participants 9.5% reported long Covid symptoms 12 weeks after infection, compared with 14.6% in the unvaccinated group.
Long Covid symptoms were self-reported, so further more controlled studies would be helpful. The study was undertaken before the Omicron variant spread. The paper has not yet been peer reviewed.
Pfizer/BioNTech vaccine may be less effective against infection in children under 12
A US study found a rapid decline in effectiveness of the Pfizer/BioNTech vaccine for younger children. Between mid-December and the end of January (when Omicron was dominant) effectiveness in preventing infection for children aged 5-to-11 declined from 68% to 12%, while for 12-to 17 year olds the effectiveness fell from 66% to 51%.
Effectiveness against hospitalisation did not decline as much. For the 5-to-11 year olds it declined from 100% to 48%, and from 85% to 73% for 12-to-17 year olds. However, hospitalised cases were small, and there was considerable variation by age group, so caution is required in estimating changes in effectiveness against hospitalisation.
Under 12’s received a lower vaccine dose than older age groups, which may at least partly explain the difference in effectiveness. But there may also be differences in exposure risks between the age groups. The study involved over one million fully vaccinated children with a three week interval between the two doses. In NZ the interval for children is 8 weeks. The paper has not yet been peer reviewed.
A paper published in JAMA Network Open reports the effectiveness of two doses of the Pfizer vaccine in 12-to-17 year olds against Delta infections for at least four months.
Three Pfizer doses appear generally safe for 12-to-17 year olds
A third dose of the Pfizer/BioNTech vaccine in 12-to-17 year olds does not elicit more severe adverse reactions than the first or second doses. This is based on US data from 2.8 million young people receiving a third dose, with adverse reactions assessed out to 11 weeks post-dose. Most adverse reactions were reported the day after the vaccination.
Of adverse reactions reported to the Vaccine Adverse Event Reporting System, 91.6% were non-serious. Confirmed cases of myocarditis were less frequent than after the second dose. Reports added to VAERS have not been clinically confirmed, and less serious events will be under reported. The paper was published in the Morbidity and Mortality Weekly Report.
Risk of more severe Covid-19 may increase five months after two vaccine doses
An Italian study found that the risk of admittance to ICU with Covid-19 increased after about five months after receiving the second dose. This appeared to be vaccine-independent (AstraZeneca/Oxford, Pfizer/BioNTech, Moderna, or one dose of the Johnson & Johnson.Janssen vaccine). This increased risk was associated with increasing numbers of infections in vaccinated people after this period.
However, overall very few of the Covid-19 cases in ICUs were vaccinated (2%). Males, and those over 60 were more likely to be hospitalised with Covid-19. The paper has not yet been peer reviewed.
Fatality risk significantly lower for Omicron than Delta
A UK study reports that the risk of death from an Omicron infection is 67% lower than for Delta. This is based on a model that accounted for various factors, such as age, sex, vaccination status, and previous infection. For those under 70 the risk was even lower (86% lower, versus 55% for over 70s).
Risk of death was also lower for those receiving a third vaccine dose, compared to just two (AstraZeneca or either mRNA vaccine). Overall, only 0.03% of the infections resulted in deaths attributable to Covid-19. All subjects in the study had received at least one vaccine dose.
The study was based on the health records of more than 1 million adults who were infected in December 2021. Deaths from Omicron may be under-represented because some of those infected in late December may have subsequently died. The paper has not yet been peer reviewed.
Within households masks, isolation and vaccination can reduce transmission risk
A US study found that half of household contacts of a person with the Omicron variant became infected. Infection was more likely if the infected person did not self-isolate, did not wear a mask inside, and had not been fully vaccinated.
The study was based on 183 households. Two-thirds of the households had transmission events. Household transmission was lower (about 43%) when the index case had received three vaccine doses, or received two doses within the previous 5 months.
Participation in the study was voluntary, so there may be differences between those who chose or declined to participate. The study also did not confirm whether household transmission was from the index case. Asymptomatic cases may have been missed. The paper was published in the Morbidity and Mortality Weekly Report.
Risk of new variants may increase if physical distancing stopped too soon
A modelling study found that the risk of emergence of a vaccine-resistant viral variant can increase as vaccination rates increase. This occurs when social distancing requirements are removed before vaccination programmes end. However, the probability of a vaccine-resistant variant emerging is higher when vaccination rates are low.
The model makes a range of assumptions that are unlikely to hold, and the authors state that some of these assumptions will need to be revisited. The paper was published in Nature Human Behaviour.
Non-vaccine-related papers
Accumulating evidence for zoonotic origin
A paper by researchers from the Chinese Center for Disease Control and Prevention provides the results of 1380 environmental samples (surfaces, drains, and sewers) taken at the Huanan Market in early 2020.
Seventy three tested positive for SARS-CoV-2, mostly in the area of the market where wild animals were sold. The virus was not detected in 18 animal species, although as the paper below reports, several species being sold at the market were not tested. The paper has not yet been peer reviewed.
Two papers (sharing the same lead author) conclude that animal to human transmission at the Huanan market in Wuhan is the most probable source for the pandemic. Neither paper has been peer reviewed yet. These papers build on a perspective written by one of the lead authors and published in Science in November 2021.
The first paper describes the clustering of many of the December 2019 cases in Wuhan in the neighbourhoods around the Huanan Market. In January and February 2020 other clusters emerged further away from the market, suggesting the market was the epicentre.
Statistical analyses also showed that the two earliest SARS-CoV-2 lineages (A and B) were geographically associated with the market.
The paper also reports that seven species of mammals susceptible to SARS-CoV-2 were among live mammals being sold at the market in November and December 2019, although the Chinese CDC did not report test results for these species. The paper also emphasises that many of the SARS-CoV-2 positive samples in the Chinese CDC report were in the area of the market where live wild mammals were sold.
The paper does not provide definitive proof of a zoonotic origin, but the authors argue that the independent lines of evidence strongly support this hypothesis.
The second paper suggests that at least two separate zoonotic events occurred within weeks of each other at Huanan Market in November or December 2019. This is based on analyses of 787 genome sequences collected in Wuhan by mid February 2020, with the distinct A and B lineages being found to be associated with the market.
The authors suggest that there may have been other independent interspecies transmission events within the market, but these do not seem to have become established in people.
Using different mice strains helps identify genetic correlates for infection outcomes
Studies on laboratory mice often use only one or two inbred strains, but a recent study looked at Covid-19 infections in a bigger range of mouse strains. Eight different founding lines were bred with a strain containing the human ACE2 receptor. Covid symptoms in the progeny of these mice varied based on parental genotype, and better reflected symptom and outcome variation seen in people infected by SARS-CoV-2. This helps identify genetic factors associated with resistance or susceptibility to Covid-19.
Mice that survived had earlier interferon I responses, and more phased pro-inflammatory responses than mice strains that were less able to control disease progression. The authors note that some symptoms seen in the mice aren’t seen in people, so the mice aren’t a perfect model for studying factors affecting Covid-19 in humans. The paper has not yet been peer reviewed.
Global cases may be declining
The WHO Coronavirus Dashboard indicates declining global numbers of confirmed weekly cases, with late February cases down 16% from the previous week. Confirmed cases have been declining since late January.
The Dashboard does not indicate if there have been changes in reporting practices from countries since the Omicron variant emerged, which could lead to under reporting. However, weekly reported Covid-related deaths also decreased by 14% over the same period.
Low antibody response may facilitate emergence of new variants
A study of an avian coronavirus found that immune escape variants appeared in infected chickens with low antibody responses. This appeared to be due to polymorphisms in the spike protein. The results may be relevant to SARS-CoV-2.
Only small numbers of chickens were involved in the study, and they were injected with an inactivated virus rather than being naturally infected. The fitness of the mutants was not tested over the longer term. The paper was published in the Journal of Molecular Evolution.