Coronavirus Research Tracking - 30 July

This article was published on
July 30, 2021

This explainer is more than 90 days old. Some of the information might be out of date or no longer relevant. Browse our homepage for up to date content or request information about a specific topic from our team of scientists.

It is easy to get swamped by all the research on coronavirus. New Zealand’s Science Media Centre is keeping track of much of it so you don’t have to. The Research Tracker is prepared by Dr Robert Hickson for the Science Media Centre New Zealand. This week emerging but mixed results on waning effectiveness of the Pfizer vaccine, frequency of breakthrough infections, adverse reactions to vaccines, and vaccine safety for adolescents. Non-vaccine papers include Covid-19-associated neurological damage, recalculating deaths in India, and recent changes in selection pressure on the virus.

It is easy to get swamped by all the research on coronavirus. New Zealand’s Science Media Centre is keeping track of much of it so you don’t have to. The Research Tracker is prepared by Dr Robert Hickson for the Science Media Centre New Zealand. This week emerging but mixed results on waning effectiveness of the Pfizer vaccine, frequency of breakthrough infections, adverse reactions to vaccines, and vaccine safety for adolescents. Non-vaccine papers include Covid-19-associated neurological damage, recalculating deaths in India, and recent changes in selection pressure on the virus.


What our experts say

Context and background


Subscribe to SMC-NZ's Coronavirus Research Tracker.

Vaccine-related papers

Indications of declines in vaccine effectiveness over six months

Data from Israel’s Ministry of Health indicates that Pfizer/BioNTech vaccine effectiveness may decline substantially after several months. In June & July infection rates for adults fully vaccinated in January were higher than those vaccinated more recently.

Overall vaccine effectiveness against infection was calculated to be 39% in June/July. Effectiveness for those vaccinated in January was only 16%, compared with 75% for those vaccinated in April. Confidence intervals for some months are, however, large. In contrast, effectiveness against severe Covid-19 remained high according to that data release.

However, further analyses reported yesterday (but not published) indicates that effectiveness in preventing severe Covid-19 in those over 60 in Israel vaccinated in January has decreased from 97% to 81%. This was reported by the newspaper Haaretz. This has led to Israel’s government asking those over 60 to get a third Pfizer/BioNTech dose.

There is no scientific paper based on the data yet. The Delta variant was increasing in prevalence in Israel during June & July. These results are being contrasted in some media articles with UK results on very good effectiveness against the Delta published last week in the New England Journal of Medicine (reported in last week’s Tracker). However, the methodologies and levels of infection differ between the two studies and countries, making direct comparisons unreliable. The UK study also didn’t examine how time since vaccination may influence effectiveness.

Another UK study, published in The Lancet, found antibody levels following vaccination with the Pfizer/BioNTech vaccine decreased two-fold, or more, by 70 days after the second dose. For the AstraZeneca/Oxford vaccine the decline was five-fold. However, T cell trends and the effects on infection and disease severity were not determined.

A large multi-country placebo-controlled trial found that the efficacy of the Pfizer/BioNTech vaccine against symptomatic infection declined slightly between two and six months after the second dose. Efficacy against symptomatic infections decreased from 91.7% to 83.7%. The study involved over 40,000 participants, aged 16 and over. Inter-country variation in efficacy was observed.

Efficacy against severe Covid-19 was around 97%, and was not reported to decrease. In South Africa the vaccine was effective in protecting against symptomatic infections from the Beta variant. Prevention of asymptomatic infections was not assessed. Previous infection without vaccination was calculated to provide a level of protection of  about 72%. The paper has not yet been peer reviewed.

Waning vaccine effectiveness is a “watch this space” issue, with more analyses, potentially contradictory, likely to be published over the coming weeks and months.

“Breakthrough Infections” not common but require monitoring

An Israeli study of fully vaccinated healthcare workers found 2.6% subsequently became infected. Infected workers had lower neutralising antibody levels before infection than a comparable group. Neutralizing antibody titers within the week before infection correlated with the viral load. The infections were either asymptomatic (33%) or resulted in mild disease. About one fifth had symptoms lasting more than six weeks.

None of the infected workers appeared to infect others. However, asymptomatic infections in vaccinated people may increase the risk of transmission because infection is not noticed. Just under fifteen hundred healthcare workers were included in the study, and most were young and healthy. The paper was published in the New England Journal of Medicine.

While not a scientific paper, an article in The Atlantic links to research describing the reasons why you can become infected after being vaccinated. It uses the metaphor of your body as a castle, with immunisation as reinforcement against assault.

Prior infection can have different vaccination outcomes depending on the vaccine

A Qatari study found that prior infection enhanced the protection from reinfection for the Pfizer/BioNTech vaccine, but not for the Moderna vaccine. The authors suggest that the difference may be due to the already high protection provided by the Moderna vaccine, with fewer vaccinated people becoming infected. The main variants circulating during the study were Alpha and Beta.

Overall, infection rates were low in vaccinated adults during the study so the results may not hold when infection rates are high, or in different populations. The paper has not yet been peer reviewed.

Adverse reactions to first vaccine dose may not lead to similar responses after the second

In a study of people who had an immediate allergic reaction to the first dose of either mRNA vaccine, reactions to the second dose were usually less. Out of 159 people who had adverse reactions to the first dose, only 20% reported reactions to the second, and these were mild and short lived. The trial was small and most of the participants were women, so the results may not be generally applicable. The paper was published in JAMA Internal Medicine.

Impurities may influence vaccine risks

A study found that the adenovirus-based AstraZeneca/Oxford vaccine contains more impurities than the Johnson & Johnson/Janssen adenovirus-based vaccine. For example, proteins from the cells used to produce the vaccine were found at higher levels in the AstraZeneca vaccine. The AstraZeneca vaccine was also found to increase blood vessel permeability. The impurities may contribute to the small but higher frequency of clotting-like adverse reactions seen for the AstraZeneca/Oxford vaccine. The paper has not yet been peer reviewed.

Third dose tested for Sinovac’s vaccine

A third dose of Sinovac Biotech’s CoronaVac vaccine given six to eight months after the second dose provided a strong boost in antibody levels in a small Phase 2 clinical trial. Coronavac is based on an inactivated SARS-CoV-2 virus. Neutralising antibody concentrations declined to near undetectable levels six months after the second dose, but the strong response after the third dose indicates good immune memory was established.

Giving a third dose one month after the second resulted in antibody levels only a third of those seen when it was given after six months. Further research is needed to determine optimal timings and conditions for a third dose. Reactions to the booster were relatively mild. Effectiveness of a third dose against different variants was not tested. The paper has not yet been peer reviewed.

Post-vaccination reactions can be similar to early Covid-19 symptoms

Care is required to distinguish post-vaccination symptoms from Covid-19 infection. Early stage Covid-19 symptoms can be similar to vaccination side-effects, so clinical diagnosis without PCR testing is unreliable. This research is part of a larger UK Covid Symptom Study, where participants note symptoms and health using an app.

Participants were not representative of the whole population, as they tend to be younger, better educated,healthcare workers, and female. The results may differ in countries or regions where Covid-19 is less prevalent, and fewer are vaccinated. The paper has not yet been peer reviewed.

Several vaccines appear safe for children over 12

A News item published in Nature discusses vaccination of children. It reports that the mRNA, and two chinese vaccines appear safe for adolescents (over 12 years), but further research on the safety for younger children is still underway. Vaccinating children can help further reduce transmission, and reduce the risk of new variants emerging. On the other hand, vaccinating children may not be fair if it is at the expense of vaccinating people at greater risk.

The need for more vaccines that directly target the respiratory and reproductive tracts

Drawing on the Covid-19 vaccine successes, a review article discusses the potential of injectable (or inhalable) mucosal vaccines for a range of other diseases. These are vaccines that trigger immune responses at mucosal sites, such as in the respiratory and reproductive tracts, which are common targets for a range of pathogens.

Currently, oral older style vaccines (attenuated and inactivated viruses or bacteria) are used and can be relatively ineffective, due to their method of delivery. mRNA and subunit vaccines that can be injected and more rapidly reach sites of infection have significant advantages. The paper was published in Nature Reviews Immunology.

Non-vaccine-related papers

Delta variant risk

Public Health England’s 23 July updated risk assessment for the Delta variant notes the weight of evidence supporting greater transmissibility of the variant and reduced vaccine effectiveness. While there are concerns that the variant also results in greater disease severity and increases the risk of reinfection, there is currently insufficient data to confirm, or reject, these outcomes.

Cognitive impacts of Covid-19

People who have recovered from Covid-19 may have subsequent cognitive deficits. A large UK study, involving 80,000 people, compared intelligence test results with infection status. Those who had recovered from Covid-19 performed worse than expected in the cognitive tests, and those reporting more serious Covid-19 performed more badly in the tests. The results identify correlations, rather than causal links. The paper was published in EClinicalMedicine.

Covid-19 associated with nerve damage and loss

An indicator for the development of Long Covid may be nerve damage in the eye. A study of 40 Covid-19 patients found that nerve fibre loss in the cornea and an increase in Dendritic Cells is associated with Long Covid, particularly in patients showing neurological disorders. Covid-induced inflammation may be the cause of the damage. The damage can be detected by microscopy. The paper was published in the British Journal of Ophthalmology.

A UK study found that Covid-19 can lead to grey matter loss in the brain in regions associated with smell and taste. The study, which had brain scans from patients before and after the pandemic began, compared 394 recovered Covid-19 patients with 388 matched patients who had not been infected. Imaging revealed reduced thickness of tissue in several regions in the participants after infection. The participants were mostly of white ethnicity, so the study is not representative of the whole population. The paper has not yet been peer reviewed.

Absence of antibodies associated with reinfection

An Indian study found that the likelihood of reinfection was much lower in people who still had antibodies from an earlier infection. Only three out of 1170 people (0.3 %) who were still seropositive several months after an infection became reinfected. For those who had become seronegative (had no detectable antibodies), 127 out of 1068 (11.9%) became infected again. A limitation of this study is that some of the information was collected using phone interviews with participants rather than being derived from clinical data. The paper has not yet been peer reviewed.

Rapid Covid Tests may help infection control under certain conditions

An Indian modelling study found that low cost rapid Covid-19 antigen tests can be a cost effective means of helping control outbreaks under certain conditions, replacing the need for widespread use of PCR tests. The conditions include test sensitivity that is “not too low” (they don’t specify a value, though greater than 75% is likely), large numbers of people are tested daily, positive cases are isolated or quarantined, and non-pharmaceutical interventions are also used.

The study focussed on testing at least 0.5% of the population daily (that would mean about 25,000 New Zealanders per day), and seroprevalence of around 20% (which is high in a New Zealand context). The paper was published in PLOS Computational Biology.

Covid-19 mortality in India much higher than official estimates

An analysis of excess mortality in India concluded that the Covid-19 death rate is 7 to 8 times higher than the official figure. Three million, or more, may have died in India due to Covid-19. The paper has not yet been peer reviewed.

Increased selection pressure on the virus seen in genomes from late 2020 onwards

A significant change in selective forces acting on SARS-CoV-2 coincides with the spread of the Alpha, Beta and Gamma variants. While emerging independently at around the same time, and sharing some common mutations (notably the N501Y spike protein mutation), these variants have distinct sets of mutations and are regarded as separate lineages.

Genome analyses indicated that from about November 2020 there was a shift from largely neutral mutations acting on codons (which specify specific amino acids) to more frequent positive and negative selection. Convergent and divergent mutations are seen between the lineages, resulting in the same or divergent mutations at the same sites, respectively. Convergent mutations at 28 sites in these lineages suggest ongoing selection and adaptation. The patterns and nature of mutations seen in variants this year suggest that viral “fitness” may be rapidly increasing, which may lead to more infectious, resistant and/or serious variants. The paper has not yet been peer reviewed.

Adding mobility data to infection prediction models can improve results

Modelling indicates that combining social distancing requirements with mobility data from Google can provide a better predictor of infection trends. Use of mobility data helps to adjust for irregular patterns that are not covered by relying solely on public health measures. The paper was published in Scientific Reports.

Media briefing

Media Release

Expert Comments: 

No items found.


No items found.