Coronavirus Research Tracking - 29 April
Vaccine comparisons, three Pfizer doses, T-cell responses, pan-coronavirus vaccines, long Covid, viral loads, viral evolution
This week, comparing vaccines, the broader immune response after three Pfizer doses, different T-cell responses to infection and vaccination, and “universal” vaccines in development. In non-vaccine papers, the prevalence & persistence of long Covid, viral load isn’t a predictor of infectivity, coordinated antibody and T-cell responses over time, how Omicron’s structure influences transmissibility, and the potential effects of regulatory sequence mutations in the virus.
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.
Comparing immune responses between vaccines
A Nature news article discusses a paper (not yet peer reviewed) that directly compared immune responses to four vaccines. The two mRNA vaccines were compared over six months with the Johnson & Johnson/Janssen and Novavax vaccines using a range of criteria.
The purpose wasn’t to identify the best vaccine, but to understand immune responses in order to see how Covid vaccines could be improved. Only two doses of the mRNA vaccines were studied. Effectiveness against variants was not studied.
There were similarities and differences in the antibody, memory B cell, and T-cell responses between vaccines. The mRNA vaccines were the most immunogenic, although some responses declined over six months. In contrast, the Janssen vaccine tended to generate lower antibody and T-cell levels, but these remained stable over the 6 months.
The World Health Organisation’s Weekly epidemiological update - 27 April provides a table of the effectiveness of 5 vaccines in various populations against Omicron, with and without booster doses. These come from different studies, so direct comparisons are difficult.
Three Pfizer doses provide strong effectiveness for three months
Three doses of the Pfizer/BioNTech vaccine provide good effectiveness over three months against hospitalisation following an Omicron infection, but effectiveness then declines. The US study (published in The Lancet Respiratory Medicine) found that two doses of the Pfizer/BioNTech vaccine provided 41% effectiveness at least 9 months after the second dose. In the three months after a third dose effectiveness was 80-to-90%, then declined to around 55% by five months. Waning in effectiveness against less severe Covid was also observed.
More support for four Pfizer doses in older people
A fourth Pfizer/BioNTech dose significantly reduced the risk of death in those over 60. Compared to those of a similar age who had three doses, risk of hospitalisation was reduced by two-thirds, and risk of death was nearly 80% lower for those who had a fourth dose.
Assessments were made in the 40 days after receiving a fourth dose, which was given at least four months after the third. The authors note that further research on the safety of a fourth dose is required. The paper was published in Nature Medicine.
Strong and broad neutralisation after three Pfizer doses
A third mRNA dose neutralised the Omcron and its sub-variants well, in contrast to just two doses. Sera from patients who had an Omicron infection were also able to neutralise all Omicron sub-variants comparably well. Sera from these patients were also quite effective at neutralising the D614G and Alpha variants, but not Beta and Delta.
Serum samples were collected about five months after the third dose. There was often considerable variation in neutralisation ability in sera from different patients. Sample sizes for some comparisons were relatively small. Pseudotype viruses were used in the tests. The paper was published in Cell Host and Microbe.
A Japanese study found a significant decline in antibody neutralising activity within 8 months after the second Pfizer/BioNtech dose. Loss of neutralising ability against the Omicron variant was larger and quicker than against earlier variants.
A third dose significantly increased binding antibody and neutralising activity against Omicron, and other variants, at least within the first two weeks after a third dose. The third dose was also associated with greater cross-neutralisation of variants, indicating a broader immune response (as demonstrated in a paper included in the 15 April Tracker).
The study involved healthcare workers, and some comparisons of neutralising ability between different variants involved relatively small numbers of participants. The paper was published in Med.
T-cell responses differ between natural infections and vaccination
T-cell responses to a natural infection differ from that produced by mRNA vaccines. In the latter, T-cells that target the spike protein dominate, while natural infections lead to more T-cells that recognise other viral regions. However, a third vaccine dose leads to a broader CD8+ T-cell response directed against the spike protein, which may provide stronger protection against future variants.
Vaccination after an infection also led to a greater diversity of CD4+ and CD8+ T-cells that target conserved parts of the spike protein, so can enhance immunity. The paper was published in Nature Microbiology.
Unvaccinated people could pose a disproportionate infection risk to the vaccinated
A modelling study found that unvaccinated people can increase the risk of vaccinated people becoming infected when they mix. Even a small number of unvaccinated people can substantially increase infection risk in a largely vaccinated group.
The authors conclude that the assertion that vaccination decisions are best left to individuals is not in the general population’s best interest from a public health perspective. They advocate for strong public health actions to encourage vaccinations. The model is not a precise simulation of pandemic dynamics, nor does it take account of different behaviours between vaccinated and unvaccinated individuals. The paper was published in the Canadian Medical Association Journal.
Different types of “universal” coronavirus vaccines in development
A news article in Nature Reviews Drug Discovery discusses the range of “universal” coronavirus vaccines in development (at least 10). It points out that each of them uses a different definition for how universal it is. The article also notes that breadth of action could potentially result in weaker immune control if the proteins involved in cell infection are not effectively neutralised. Hopes are, however, reasonably high that a broad spectrum coronavirus vaccine is more achievable than one for flu.
Persistence of long Covid symptoms
One year after being discharged from hospital after Covid-19, just over 70% of 807 patients still reported Covid symptoms. Women, those who were obese, or had required mechanical ventilation were more likely to still have symptoms after one year. There was little improvement in symptoms between five and 12 months after leaving hospital.
The most common ongoing symptoms were fatigue, muscle pain, physically slowing down, poor sleep, and breathlessness. Those with severe or moderate cognitive impairment tended to have persistent systemic inflammation.
The results of this study may not be generalisable since a higher proportion of participants required mechanical ventilation when treated for Covid-19 than seen for hospitalised cases overall. The health of the participants pre-Covid could not be independently assessed, so the role of some pre-existing conditions in the results is less certain. The paper was published in The Lancet Respiratory Medicine.
Humoral and cellular immune responses coordinate to produce longer term immunity
High levels of specific T-cells soon after an infection correlate with low but long term persistence of neutralising antibodies. The study identified that some broadly neutralising antibodies can persist for over one year.
The authors conclude that the results indicate that post-infection the immune response continues to evolve, creating more diverse antibodies and more potent neutralising antibodies. The paper was published in Signal Transduction and Targeted Therapy.
Viral load isn’t a good predictor of the amount of infectious particles shed
A study of viral shedding found that there is a great deal of variability between individuals. Nasal and saliva samples from 60 people were collected daily and tested to see if they contained infectious viral particles. Shedding was not strongly correlated with the viral load in the individual, as measured by quantitative PCR. Older age was, however, associated with more infectious shedding.
Shedding can peak several days earlier in saliva than in nasal samples, so saliva testing may be a more effective testing strategy in the future.
The study was conducted during the Alpha variant wave. It found that levels of infectious viral shedding did not differ between the Alpha and earlier variants, so that the greater infectiousness of Alpha was not due to greater shedding.
The study participants were not representative of the general population, and none developed severe Covid-19. Sample collection may also not have been consistent between individuals. The paper was published in Nature Microbiology.
The structural mechanics of Omicron’s high transmissibility
An article in Scientific American describes how the structure of the Omicron variant makes it more contagious. Some of the mutations mean that vaccine-generated antibodies are less able to bind to it. While other mutations improved binding to the ACE2 receptor. In addition, the connection between the S1 and S2 subunit proteins has been strengthened, so the spike protein is less likely to break apart before cell attachment.
Regulatory mutations in SARS-CoV-2
The coronavirus may be able to evolve in ways like eukaryotic genes. Novel transcription regulatory sequences were detected when sequencing viral RNA from infections. Rather than changing proteins, they can lead to subgenomic mRNA with new functions, such as helping package the virus, inhibiting interferon production, or affect gene transcription (as in eukaryotes).
Convergent evolution was seen, where the same sequence was found in independent patient samples. The authors recommend genome sequencing pay attention to these types of mutations as well as ones in the structural genes. The paper has not yet been peer reviewed.