The new pre-print published today presents data on how efficient the vaccine is against SARS-CoV-2 infection.  It suggests that the AstraZeneca/Oxford vaccine reduces viral load, and the length of time that infected people can carry the virus.  Moreover, it shows a similar level of effectiveness against symptomatic disease caused by the new UK variant of SARS-CoV-2 (called B.1.1.7) as it has shown against the previously circulating variant.  With cases from the B.1.1.7 virus variant occurring at a high rate in the UK, this initial report can provide reassurance that the AstraZeneca/Oxford vaccine can still provide similar levels of protection against the new circulating virus variant.

There are some differences between responses recorded to the two different virus variants in vaccinated individuals.  Notably that the overall levels of vaccine-induced antibody that recognise the two variations of virus are the same, but that neutralising antibody activity against the UK variant B.1.1.7 is reduced 9 fold.  This reminds us that antibody measurements may not always be the best measure of vaccine effectiveness.

We still need more data before we can conclude if this vaccine is able to reduce asymptomatic infection and transmission, so it is important to that vaccinated individuals still follow social distancing guidance.

Our knowledge of how the immune system works leads us to assume that vaccines will still be effective against minor variations of the SARS-CoV-2 virus.  But we cannot be complaisant.  As noted in the recent British Society for Immunology report, we must continue to implement a robust monitoring system of vaccine effectiveness against all new emerging virus variants globally to ensure we maintain high levels of protection and stay one step ahead of the virus.

‍

Genome sequence analysis of a subset of both symptomatic and asymptomatic Covid cases from the Oxford vaccine study, during which the ‘E Kent’ variant B.1.1.7 became more frequent (reaching 28% of the symptomatic cases), showed that clinical efficacy against the new variant was similar for both variants, though the point estimate was slightly lower against the new variant (74% vs 84%), with overlapping confidence intervals meaning that the results are indistinguishable statistically – though the numbers were lower for the variant samples.  In addition, they looked at the duration of PCR (Covid 19) positivity in the patients, who self administered a swab each week, and found that asymptomatic cases shed virus for a shorter time and at lower levels, implying that they may be less infectious.  They also compared the ability of the antibodies generated by the vaccine to neutralise the two strains of the virus, and showed that the neutralising antibody levels against the new variant were several fold lower than against the original variant.  This suggests that the neutralising antibody levels (so-called correlates of efficacy) do not necessarily predict clinical efficacy, although the trend was similar for both, i.e. lower efficacy and lower neutralising antibody levels against the new variant.  They suggest that other forms of immunity, in particular T cell immunity, may also be important.

The authors warn that new variants will certainly emerge, more so under the selection pressure of widespread vaccination, and briefly discuss the potential for anti vector antibodies (i.e. antibodies against the viral vector) reducing the effectiveness of the Oxford vaccine when given two or more times – their preliminary findings are that this does not appear to interfere with efficacy.  They mention that so-called heterologous prime-boost strategies (in which different vaccines are used for the first and second doses) are now being explored.

‍

Is this robust data; what are the strengths and limitations?

This information is in a non peer reviewed preprint of a submitted paper which describes work completed recently to investigate potential efficacy of the ChAdOx01-nCoV19 vaccine vs the emergent B.1.1.7 (UK) variant virus which has gradually replaced the initial strain in the UK.  As expected from this group, the data is discussed thoroughly.

What does the data suggest – is it convincing?

Although vaccine induced antibody was less effective against this strain than against a strain close to the parent Wuhan virus used to create the vaccine, the clinical effectiveness of the vaccine against symptomatic disease caused by the mutant strain is only slightly lower than against the prior strain (74 vs 84%) with the confidence limits overlapping.  Hence we can expect recipients of this vaccine to be protected against illness caused by the ‘UK’ variant.  The paper also contains information on estimated viral load and duration of virus shedding – based on cycle times and duration of PCR positivity – which suggests that duration of infectivity may also be reduced, suggesting a potential for reducing transmission.

Is this good news; are there caveats?

This is generally reassuring news, although keeping track of other strain types emerging during the ongoing outbreak will be important, particularly those containing the E484K mutant.

Any other comments about this preprint?

The discussion in the paper is a tad complicated by comments concerning possible false positive PCR test results.  This is speculative and does not detract from the otherwise clinically relevant finding of reduced symptomatic illness caused by the B.1.1.7 variant in the vaccinated population.

‍

So this is a welcome but not entirely surprising result.  We know that both the Pfizer and Moderna vaccines already retain their effectiveness against the original B.1.1.7 UK variant (VOC 202012/01) – which lacks the E484K mutation.

Next we need to see how it performs against the new B.1.1.7/E484K Kent variant.  It is likely to show a similar decreased efficacy against this variant – as was seen in the Pfizer and Moderna vaccines against other E484K mutants.

The other problem with the Oxford ChAd and other AdV-vectored vaccines – like Russian Sputnik V Ad26/Ad5 and Johnson&Johnson Ad26 – is how to deal with the inevitable host immune response to the AdV vector that may eventually neutralise the vaccine if repeated boosting is needed, e.g. to deliver the modified B.1.1.7/E484K-adapted S protein gene insert.

Heterologous AdV prime-boosting, as being explored now between the Oxford-AZ and Russian Sputnik V teams, may extend the use of these AdV vectors for a short while at least.

‍

This worthwhile study shows that the B.1.1.7 or ‘Kent’ variant of the Covid-19 coronavirus just as susceptible to the immunity generated by the Oxford/AstraZeneca vaccine as the versions of the virus that had been circulating previously.  This is unsurprising as the mutation does not make changes to the structure of the virus’ spike protein that would alter its ability to gain entry of the virus into human cells.

Most interestingly, the paper presents data showing that overall protection against disease does not reliably mirror the neutralising activity of antibodies produced by someone’s immune system.  It would appear that having the right type of antibodies is more important than their quantity.  Although there is no evidence that this vaccine does not work in over 65s, this finding calls into question this measure being used as a justification for administering the vaccine to people in that age group, where there isn’t much data demonstrating that they are actually protected.

‍

This is excellent news.  It reminds us that neutralising antibody titres are not all that matters.  Immunity to disease is what matters: antibody titres are part of this, but other factors including T-cell responses also matter.

In this study, participants were routinely swabbed to see if they had become infected.  They also provided information on any symptoms or illness they experienced.

Swabs from 256 participants were successfully sequenced – which means that the investigators knew exactly which viral variant they had been infected with.

Participants who had been infected with the virus’ B.1.1.7 variant were no more likely to become ill than participants who had the previous variants, against which the vaccine could be expected to work.  This is excellent news, as it indicates that the vaccine works effectively at preventing illness, even with the variant virus.

The study also found that the amount of virus found on swabs of vaccinated people was lower than you would have expected in non-vaccinated individuals, and the virus was present for less time – suggesting that the vaccine not only protects recipients against infection, but it also means they are less likely to become infected; and if they do have asymptomatic infection, they are less likely to transmit it to other people.  This is very important.  There are two things we need from vaccines: that they reduce the burden of disease and the pressure on health services by preventing serious illness – death, ICU admissions, and hospital admissions – and that they reduce transmission, so that we can get the effective R number below one, so the disease will die out, and so that there is less viral replication and thus less likelihood of variants of concern arising.

‍

The findings of this study are very reassuring.  The pre-print contains the first published clinical data showing that COVID-19 Vaccine AstraZeneca is effective at protecting against the B.1.1.7 variant, also known as the ‘Kent variant’.

It also provides data suggesting that this vaccine may reduce transmission of the disease.

‍