Coronavirus Research Tracking - 27 August - Vaccine edition

Vaccine effectiveness, waning immunity, vaccine safety

This week due to the number of interesting papers, vaccine and non-vaccine related research have been split into separate tracking reports, and distributed separately.

Vaccine-related papers cover effectiveness, immune responses, correlates of protection, and how reporting vaccine hesitancy influences vaccination decisions.

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

Vaccine effectiveness over time

In a large US study the Pfizer/BioNTech vaccine remained effective in reducing the risk of hospitalisation over six months, but waning effectiveness against infection was found. Overall effectiveness against infection was 73%, and 90% against hospitalisation.

Vaccination showed high effectiveness against Delta infection (93%) in the first month, but declined to 53% after four months. Decreasing effectiveness against other variants was also observed. The authors interpret this as waning immunity. The paper has not yet been peer reviewed.

A news item in Science discusses the possibilities of waning vaccine effectiveness against infections, and whether booster shots are warranted. It notes that factors other than waning vaccine effectiveness or immune escape could contribute to a rise in post-vaccination infections. This includes changes in peoples’ behaviour after vaccination that could increase exposure risk.

Evidence of the effect of third vaccine doses is lacking from non-immunocompromised people. A third dose is likely to be beneficial, but the optimal timing is also uncertain. Booster shots can also reduce access to vaccines by unvaccinated or partially vaccinated people, so raise questions of equity.

Effectiveness against Delta

A US study finds that vaccination provides population protection against the Delta variant. Between May and July one quarter of infections in Los Angeles County were in fully vaccinated people. In July the infection rate of unvaccinated people was five times higher than for those who were vaccinated, and hospitalisation rates were 29 times higher. 

For Delta variant infections viral loads (based on PCR cycles) were similar between unvaccinated, partially and fully vaccinated people. However, the paper notes that this method of estimating viral load is not standardised, so care is required in how Ct values are used. The paper was published in the Morbidity and Mortality Weekly Report.

A Bahraini study of 1 million people found the four vaccines deployed reduced infections, severe Covid-19 and deaths, including those involving the Delta variant. The four vaccines were Pfizer/BioNTech, Astra-Zeneca/Oxford, Sinopharm and Sputnik V. Over half the vaccinees received the Sinopharm vaccine.

Compared to Pfizer/BioNTech recipients, individuals vaccinated with Sinopharm had a higher risk of post-vaccination infections, hospitalisations, intensive care, and death. This effect was greater for those over 50. The four vaccines were not all administered in the same time period, and socioeconomic details of recipients was not collected, which may influence the results. The paper has not yet been peer reviewed.

An Israeli study found that natural infection appears to provide longer lasting and stronger immunity against the Delta variant than vaccination with the Pfizer/BioNTech vaccine. For those originally vaccinated in January or February, the risk of an infection later in the year with the Delta variant was 13-fold higher that for those who had been infected with another strain in those months.

Naive vaccinees also had a greater risk of hospitalisation if subsequently infected. Having an earlier infection and then one vaccine dose improved protection. The study probably underestimated asymptomatic infections, which could affect calculations of levels of protection. The paper has not yet been peer reviewed.

Vaccinated people may be less infectious

A Dutch study of healthcare workers found that those vaccinated but subsequently infected may have reduced shedding of the virus, compared with unvaccinated cases. Four different vaccines were covered by the study. Viral loads decreased quickly over the first three days after symptom onset in vaccinated cases.

In addition, successful recovery of infectious virus was lower from samples obtained from vaccinated people, compared to those from the unvaccinated, indicating vaccinations can reduce infection risks.

A limitation of the study is that it compared vaccinated and unvaccinated cases at different times. The unvaccinated data was collected earlier in the pandemic (because there were too few unvaccinated healthcare workers to include later on), while the vaccinated cases were studied more recently, when the Delta variant was predominant. 

Viral loads were also found to be higher in symptomatic cases than asymptomatic ones. The effectiveness of the different vaccines was not assessed. The paper has not yet been peer reviewed

A Chinese study also found that vaccinated people infected with the Delta variant had a lower probability of infecting others. However, this is based on a small number of cases.

The study also found that people infected with the Delta variant begin shedding viral particles about two days before symptoms develop. This helps explain the rapid spread of the variant. Shedding began on average 4 days after infection, while symptoms appeared after about 5.8 days.  

Nearly three quarters of secondary infections were estimated to have occured in the pre-symptomatic stage. Those infected with the Delta variant tended to have higher viral loads (median Ct values of 23, compared with 36.5 for the wild type virus). 

The paper has not yet been peer reviewed. A news article in Nature also discusses these results. 

Johnson & Johnson/Janssen vaccine produces durable antibody responses

Neutralising and binding antibody levels remained stable in adults for at least 8 and 6 months respectively after vaccination. A booster shot given six months after the initial dose led to increases in the levels of binding antibodies, and did not generate safety concerns. Relatively small numbers of participants were included in the study. The paper has not yet been peer reviewed.

mRNA vaccines create durable immune memory

In a six month study mRNA vaccinations were found to induce durable immune memory. Binding and neutralising antibody levels declined but remained detectable over the time frame. Memory B cells increased in frequency following vaccination, and antibodies from these could bind the Alpha, Beta and Delta variants. CD4+ and CD8+ also tended to increase post-vaccination, with early CD4+ levels indicative of later antibody levels. 

The authors suggest that the declining antibody levels over time may result in increased risk of infection, but the strong Memory B cell and T cell responses may help explain reduced risk of severe Covid-19. The paper has not yet been peer reviewed.

A correlate of protection for the Moderna vaccine

Binding and neutralising antibody levels measured 57 days after receiving the first Moderna vaccine dose correlated with the degree of protection from infection for those who were fully vaccinated. Antibody levels measured at day 29 may also be a useful marker of the degree of protection. The paper has not yet been peer reviewed.

Different antibody dynamics between naturally infected and vaccinated people

An Israeli study found that the dynamics of antibody response differs between those who had been infected and those who had been vaccinated. Vaccinated adults (with Pfizer/BioNTech) initially developed higher antibody levels, but these declined faster than levels in unvaccinated infected adults. After six months 16% of the vaccinated group had undetectable antibody levels, compared with 11% in the convalescent group.

Neutralising antibodies and Memory B cells were not studied, and the level of protection against infection, or reinfection, was not examined. In addition, differences in behaviours between those who were or weren’t vaccinated, and decisions to get retested could have affected the results. The paper has not yet been peer reviewed

Vaccine safety in context

An Israeli study of 250,000 vaccinated adults put the risks of vaccination in context by comparing the risks of adverse effects from Covid-19. The risk of myocarditis increased by a factor of three following vaccinations with the Pfizer/BioNTech vaccine. The risk was greatest in young men. The study estimated that this could result in between 1 and 5 cases of myocarditis per 100,000 vaccinations.

However, SARS-CoV-2 infection increased the risk of myocarditis by a factor of 18, as well as increasing the risk of several other serious conditions.

An increase in the incidence of Bell’s palsy, resulting in 8 additional cases per 100,000 vaccinations, was also found. But the vaccine appears to reduce the risk of anaemia and intracranial hemorrhage, compared with unvaccinated people infected with SARS-CoV-2. The paper was published in The New England Journal of Medicine.

Immune responses can differ when the time between Pfizer doses increases

A Canadian study found that a longer interval between the first and second Pfizer/BioNTech doses increased some binding antibody concentrations. Neutralising antibody responses were similar in both groups. The first group had 3 to 6 weeks between first and second doses, while the other group had an interval of 8 to 12 weeks. CD4+ T cell responses were slightly reduced in the longer interval group, which could offset the stronger humoral immune response.

The study involved small numbers of participants, and persistence of the differences were not followed. They also tended to be younger and mostly female, characteristics which are known to influence immune responses. The paper has not yet been peer reviewed.

Some immunocompromised people can fail to produce antibodies after two vaccine doses

Forty percent of a group of immune-compromised patients either did not generate antibodies, or produced only low levels, in response to the Pfizer/BioNTech or AstraZeneca/Oxford vaccines. Eleven percent produced no detectable antibodies. In this UK OCTAVE trial patients with some conditions had much higher levels of non-response to the vaccines than others.

In contrast, T cell levels in the immune-compromised were similar to other vaccinated people. The impact of a third vaccine dose will be investigated. The paper has not yet been peer reviewed.

Pan-coronavirus antibodies

Previous SARS-CoV-1 infections appear to have generated neutralising antibodies that are restimulated by the Pfizer/BioNTech vaccine. They are able to neutralise variants of concern, as well as other sarbecoviruses. The findings indicate the feasibility of developing a pan-sarbecovirus vaccine. The paper was published in the New England Journal of Medicine.

The paper’s key results and implications are also covered in a news item in Science.

Stories of vaccine hesitancy reduce likelihood of getting vaccinated too

A New Zealand study found that exposing people to vaccine misinformation or to stories of vaccine hesitancy both reduced peoples likelihood to get vaccinated similarly. Sharing information about vaccine hesitancy was more common than sharing misinformation. The authors conclude that official messaging about vaccinations should be transparent and not avoid mentioning the limitations or adverse effects of vaccination. The paper was published in Frontiers in Communication.

Other SMC resources