Coronavirus Research Tracking - 23 April
post vaccination infections, pregnancy risks, comparing vaccines, delaying 2nd vaccine dose, immune responses, mutations, predicting the next pandemic
This Research Tracker includes reports on the risks of getting infected after vaccination, the safety of getting vaccinated when pregnant, early results from other vaccines, modelling the impacts of delayed vaccine shots, and communicating vaccination risks. There are also papers on the rarity of reinfections and catching Covid-19 when flying, different immune responses in people with mild and severe Covid-19, and in children vs adults, and how to improve the chances of spotting the next pandemic.
The Research Tracker is prepared by Dr Robert Hickson for the Science Media Centre.
Getting infected after being vaccinated
Reports of post-vaccination infections are emerging, though not common. These are being called “Breakthrough infections”, which can seem like an upbeat development.
A study of nearly 15,000 vaccinated residents and staff in Chicago nursing facilities found that infection after vaccination is not common. Only 22 fully vaccinated people became infected more than two weeks after mRNA vaccinations. Two thirds (14) of them were asymptomatic, six experienced mild to moderate symptoms, while two required hospitalisation, one of whom died. The study is published in this week’s CDC Morbidity and Mortality Weekly Report.
The case of two fully vaccinated women in the US are reported in the New England Journal of Medicine. Both subsequently developed mild Covid-19. One of the women had a virus with the E484K mutation, and very high viral loads, as well as strong antibody responses to the Moderna vaccine. The other patient didn’t have an infection that included the E484K mutation, and had a lower viral load. Her antibody response to the Pfizer vaccine is not reported.
Early results suggest low risks of mRNA vaccines for pregnant women
A preliminary study of women vaccinated in the later stages of pregnancy did not identify safety concerns. The women, largely health care workers, received the Moderna or Pfizer vaccines. Out of 221 reported adverse pregnancy-related events, 46 were miscarriages. However, this is similar to the miscarriage frequency seen in unvaccinated pregnant women. Further follow-up of vaccinated pregnant women is required, particularly those wo were vaccinated early in their pregnancy. The research was published in the New England Journal of Medicine.
Earlier research on vaccinating pregnant women was reported in the 26 March Tracker.
Vaccination boosts immune response generated by earlier infection
Another study confirms that vaccination of previously infected people boosts their immune responses. More vigorous antibody and B cell responses were seen after convalescent patients received either the Pfizer or Moderna vaccine.
Vaccination also reversed the creation of a dysfunctional set of B cells generated by the infection. This dysfunctional subset may be involved in Long Covid, so vaccinations may help alleviate their symptoms. The paper has not yet been peer reviewed.
B.1.1.7 variant well controlled by vaccination in Israel
An analysis of 300,000 PCR samples in Israel determined that the B.1.1.7 variant is 45% more transmissible than earlier variants. It became the dominant strain in under four weeks. However, the Pfizer/BioNTech vaccine, along with focused population testing, was able to prevent the further spread of the variant in elderly people. The paper was published in Cell.
CoronaVac reduces risk of symptomatic infection
A Brazilian study found that at least one dose of the Sinovac’s CoronaVac vaccine halved the risk of health care workers developing a symptomatic infection. The study used matched negative test cases as controls to estimate the vaccine’s effectiveness. The paper has not yet been peer reviewed.
DNA vaccine shows promise in controlling several variants of concern
The DNA vaccine INO-4800 (from Inovio Pharmaceuticals) produced neutralising antibodies and T cell responses against the main variants of concern (B.1.1.7, B.1.351, and P.1). Neutralising antibodies were lower against the B.1.351 variant but T cell responses appeared unaffected. The paper has not yet been peer reviewed.
Delaying second dose of mRNA vaccines can be beneficial
A model of Covid-19 transmission and vaccination indicates that delaying the second dose of the Moderna vaccine by 9 weeks (giving it 12 weeks rather than 4 weeks after the first shot) could improve vaccination effectiveness in reducing infections, hospitalisations, and deaths. This benefit remained even if the first dose’s efficacy declined. On the other hand more infections may occur if the second dose of the second Pfizer/BioNTech vaccine is delayed and efficacy of the first dose declines.
The authors recommend that if there is a high rate of infections then the priority should be on rapid and widespread distribution of the first dose of the vaccine, and prioritising second doses to those most at risk (particularly in the case of the Pfizer vaccine). Prioritising administering the first dose would also be important when incidence is low.
The authors note that these conclusions may change as more information on vaccine effectiveness is gathered. The paper was published in PLOS Biology.
It’s hard to compare the effectiveness of different vaccines
The difficulty in comparing the effectiveness of different vaccines is discussed in a short article in The Lancet Microbe. There are several ways to assess efficacy and effectiveness, but the paper notes that “Relative Risk Reduction” is often used. This can be misleading because it doesn’t take account of variability in the background risks of becoming infected or ill.
Both clinical trial and field results of the effectiveness of different vaccines are hard to compare because of different methods and populations. The paper concludes that the suitability of a particular vaccine needs to consider the ability to prevent infection and reduce transmissibility, reduction in hospitalisations and death, along with considerations of safety, deployability, availability, and costs.
Communicating vaccination benefits and harms
The Winton Centre for Risk and Evidence Communication, based at the University of Cambridge, describes how to communicate the potential benefits and harms of the AstraZeneca vaccine. It developed charts for the potential benefits and harms for different age groups and different levels of exposure risk.
Don’t stop other protections after vaccination
A short communication in The Lancet Infectious Diseases uses a simple model to illustrate that an important measure to reduce the risk of SARS-CoV-2 escaping vaccine control is to keep the number of cases low. So, the article notes that non-pharmaceutical interventions shouldn’t be removed quickly after a vaccination campaign.
Reinfection is rare
Reinfection is a rare event, reports a study by the European Centre for Disease Prevention and Control. It reviewed studies of immunity following SARS-CoV-2 infection. Protection 14 days after an initial infection ranged from from 81% to 100%, and these effects can last seven months or more. These studies were undertaken before the emergence of the variants of concern.
Protection against reinfection was lower for people over 65 years. Studies also provided evidence that vaccination significantly reduces viral load and infections (both symptomatic and asymptomatic).
Immune responses of children with Covid-19 differ from adults
A Brazilian study of children with Covid-19 reported that the children had a different immune response than infected adults. The former generated higher levels of Dendritic Cells that did not target inflamed tissues. This may in part explain why children with Covid-19 tend to have more mild symptoms than adults. The children’s T cells also tended to target the virus’ membrane and nucleocapsid proteins rather than the spike protein, which may also result in stronger immune protection. The paper has not yet been peer reviewed.
White blood cell types differ between people with mild and severe Covid-19
Characterisation of more than three quarters of a million peripheral white blood cells from 130 with and without Covid-19 found certain types of cells and immune responses were common in cases of severe disease, and others were associated with milder infections. The study confirms some earlier research, but also contradicted other studies.
For example, unlike earlier research this study found that severe Covid-19 was associated with smaller populations of plasmablasts and plasma cells and lower levels of interferon alpha. Cellular differences were also found between men and women. The paper was published in Nature Medicine.
Californian variants may be more transmissible and better able to evade immune response
The variant B.1.526, which emerged in New York but is spreading in the US, is about 4 times less susceptible to two convalescent plasma and 3 to 4 times less susceptible to vaccinee sera. It has the E484K mutation seen in B.1.351 and the N501Y mutation which is thought to be responsible for the greater transmission rate seen in B.1.1.7. The transmission of B.1.526 and B.1.1.7 seem to be comparable in New York, which is a cause for concern. The study has not yet been peer reviewed.
Genomic surveillance essential to spot new mutations
Better genomic surveillance is required in the US (and some other countries) to detect new variants. Two variants that emerged in California last May- B.1.427 and B.1.429 – were not detected for several months and spread rapidly in parts of the US. These have three distinctive spike protein mutations - S13I, W152C, and L452R. Due to poor rates of genome sequencing in the US it is not clear if these mutations were acquired gradually or quickly.
Compared to other earlier variants the Californian variants have two-fold higher virus shedding. The L452R mutation made pseudoviruses more infective in lab tests, though not as much as the N501Y mutation present in B.1.1.7, B.1.351 and P.1. B.1.427 and B.1.429 also show some resistance to neutralisation by convalescent and vaccinee sera, and so they should be carefully monitored. The study was published in Cell.
Emergence of mutations in an immunocompromised patient
Immunocompromised patients receiving convalescent plasma may be breeding grounds for immune escape Covid-19 mutants. That’s the conclusion from a study (not yet peer reviewed) of one patient. It found that the infected patient, who was receiving various treatments including steroids and convalescent plasma, developed viruses with a variety of mutations in the spike protein’s receptor binding and N-terminal domains. One variant became dominant after the third weekly sample. This had mutations (such as E484K) that were more resistant to monoclonal antibodies and convalescent plasma.
Most people develop few viral mutants
A study of viral variants in over 1,300 patients found only one or two variants in most individuals, while a few had multiple variants. A more detailed investigation of mutations in 41 people showed that single point mutations were quickly generated and lost. The ratio of nonsynonymous (changing an amino acid) to synonymous mutations was consistent with strong purifying selection acting on the virus.
For transmissions within households between one and eight viral variants were passed on, but most failed to persist. The research was published in Science.
Structural analysis provides clues for the rapid spread of the 614G variation
A study of the structure of the spike protein suggests that the G-form of the D614G is more infectious due to the amino acid change increasing the protein’s binding encounters with the cell receptor. This change also exposes more neutralisation sites, making the G-form more susceptible to neutralizing antibodies. The paper was published in Science.
SARS-CoV-2 needs to mutate to infect ferrets
Domestic ferrets appear less susceptible to contracting Covid-19 from people than has been reported for laboratory and farmed ferrets (or mink). A study published in the Proceedings of the American Academy of Sciences proposes that at least two mutations in the virus spike protein are required for it to infect ferrets.
Study of NZ blood donors supports low prevalence of infection
A seroprevalence study of nearly 10,000 New Zealand blood donors indicated that few SARS-CoV-2 went undetected. Only 18 of the samples had antibodies against the virus, leading to an estimate of 0.1% seroprevalence, which is very low compared to many European countries and the US. Eight of the seropositive cases which were not associated with previous travel or infections, suggesting some undetected infections. The paper notes that blood donors aren’t ethnically or economically representative of New Zealanders as a whole, so extrapolating the results isn’t reliable. The paper has not yet been peer reviewed.
Getting infected while flying appears rare
A review of studies of transmission during air travel concludes that the risk of transmission is low. Flight crews were not included in the studies. The studies published between May and September 2020 covered about 1.4 billion passengers, of which there were just 2,866 index cases. The review estimates that the global risk of transmission during air travel may lie between 1 case for every 712,000 to 8 million travellers. The review has not yet been peer reviewed.
An experimental study of aerosol dispersion found that keeping a plane’s middle seats vacant reduced the relative exposure by 23% to 57%. The research was published in Morbidity and Mortality Weekly Report.
New strategy needed to assess future pandemic risks
Existing data for assessing the risks of zoonotic viruses are inadequate, reports a paper in PLOS Biology. This will lead to inaccurate predictions of the next pandemic. The researchers emphasise the need for more targeted focus on particular situations rather than wider scale sampling of viruses in the environment.
More effort should be directed to studying viruses in human communities exposed to wild animals (in work or living situations) and the animals these communities interact with, this includes at forest fringes and in disturbed environments. A better understanding of the frequency with which viruses are found in a range of hosts (even briefly) may also improve risk assessments.
Data Geeks - numbers of variants
A graphical representation of the number of sequenced genomes (from the GISAID database) organised by variant type is shown on Twitter. B.1.1.7 is the most common variant at this point in time, but this is due to the UK contributing a large proportion of the genome sequences to the database.