select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to preprint from Oxford looking at immunogenicity and efficacy of the Oxford/AstraZeneca vaccine with a 3-month gap between doses, and looking at whether data says anything about transmission

The University of Oxford have posted a preprint, an unpublished non peer reviewed paper, looking at the immunogenicity and efficacy of the Oxford/AstraZeneca vaccine with a 3-month gap between doses, and whether data says anything about transmission. 


Prof Sheila Bird, Formerly Programme Leader, MRC Biostatistics Unit, University of Cambridge, said:

“On reading the updated report on the efficacy of the Oxford/AstraZeneca virus-vectored vaccine, which now includes data from South Africa, my focus has been 4-fold: a) those who received dual vaccination with standard dose (SD/SD) as UK intends, b) those whose interval between 1st and 2nd dose was greater than 8 weeks, c) those aged 56+ years and d) the trial’s definition of overall vaccine efficacy.

“All cases with a positive nucleic acid amplification test (NAAT) were adjudicated for inclusion in the analysis by a blinded independent endpoint review committee.  There were 17,177 baseline seronegative trial participants in the UK (2 trials: 8 948), Brazil (1 trial: 6 753) and South Africa (1 trial: 1 476) who were eligible for inclusion in the efficacy analysis.  There were 619 documented NAAT + ve infections: 332 met the primary end-point of symptomatic infection > 14 days after 2nd dose [d) above].  Vaccine efficacy after a single standard dose (SD) from day 22 to day 90 post vaccination was also reported, see Table.

“Figure 3 in the updated report appears to suggest that no or few patients aged 56+ years had both an inter-dose interval of more than 8 weeks and had SARS-CoV-2 anti-spike IgG measured at 28 days after 2nd dose [b) and c) above].  Hence, any analyses from this of inter-dose intervals longer than 8 weeks (<6 weeks, 6-8 weeks, 9-11 weeks, >=12 weeks) in respect of SARS-CoV-2 anti-spike IgG should be limited to persons aged 18-55 years, as in the authors’ Table S4 (where 56+ years is limited to < 6 weeks versus 6-8 weeks): otherwise we risk confounding.

“For example, the first UK trial immunized 18-55 year olds only: 741/1077 were eligible for the primary analysis, all of whom seem to have had an inter-dose interval of at least 12 weeks (according to Figure 3).  In the Brazil trial, even of those aged 18-55 years, few had both a dose-interval longer than 11 weeks and measurement of SARS-CoV-2 anti-spike IgG; in the South Africa trial, no dose interval in Figure 3 exceeded 8 weeks.  Hence, within-trial information on extended dose-interval comes essentially from the second UK and Brazil trials and pertains to those aged 18-55 years, see Table S4.

“When this update appears in Lancet, I hope that a revised Figure 4 will strive to remove the confounding between age-group and longer inter-dose interval when plotting vaccine efficacy (95% CI) against geometric mean (95% CI) for A. anti-spike IgG at 28 days after 2nd dose when inter-dose interval is <6 weeks, 6-8 weeks, 9-11 weeks, >=12 weeks; and likewise for B. pseudovirus neutralization at 28 days after 2nd dose.  Outcome B were measured for even fewer participants than was A, see Table, and with apparently different sampling fractions per dose-interval.

“Finally, of course, what matters to patients during vaccine rollout is vaccine-effectiveness: the relative reduction in symptomatic SARS-CoV-2 cases that occur after the 1st dose was administered.  Tilted randomization, whereby one-third of 1st dose recipients are randomized to receive their 2nd dose according to the recommended schedule (1/22 days for the Pfizer/BioNTech messenger RNA vaccine) versus public-interest deferral to 12 weeks (maximum) is the only robust means of answering the public-interest question.”


Dr Penny Ward, Visiting Professor in Pharmaceutical Medicine at King’s College London and Chair of the Education and Standards Committee of the Faculty of Pharmaceutical Medicine, said:

“This preliminary non peer reviewed submission to the Lancet group provides an updated analysis from the participants in the ongoing trials of ChAdOx01-nCoV-19 in the UK, Brazil and South Africa including data from an additional 5,500 subjects and a further month of additional follow up (from 4 November to 7 December 2020).  The analysis looks at and compares the incidence of symptomatic COVID disease of any severity in the vaccinated and unvaccinated populations.  In the interim analysis previously reported the comparison depended on 131 cases reported 14 days or later following the second dose of the regimen, whereas in this assessment there are 332 cases reported 14 days or later following the second dose of which 84 were reported in vaccine recipients and 248 in the control group(s) giving an overall clinical effectiveness of 66.7%.

“This group are the only group to have systematically studied the incidence of asymptomatic infection by taking swabs for PCR testing weekly from UK based enrollees.  The incidence of asymptomatic infection was not substantively decreased but the incidence of PCR positive infection (symptomatic and asymptomatic) was reduced from 4% in the control population to 1.9% in the vaccine recipients – a 54% reduction in the burden of infection overall, which raises the potential for reduced transmission of infection and disease by vaccinated persons.

“The different doses and dosing interval investigated have been described previously and regrettably make the results complex to interpret, however the key outcomes are that protection against disease following a single dose is apparent starting from Day 21 (73-78% overall) following the first dose and lasting for at least 90 days (12.8 weeks).  This is associated with evidence of improved immune response following a longer between dose interval and overall vaccine efficacy following the 2 dose regimen is maximal (82.4%) in the 15% of subjects that had a 12 week interval between the first and second doses in the course.  These data confirm the analysis conducted during the MHRA and EMA reviews and justify the JCVI’s recommendations on dosing interval for this vaccine.

“The paper is not yet peer reviewed and some amendments to the publication are likely to result from review which will hopefully streamline and simplify the description of outcomes and remove some inconsistencies between tables text and figures in this preliminary report.  I look forward to the final published version.”


Dr Doug Brown, Chief Executive of the British Society for Immunology, said:

“This new pre-print, currently under review at The Lancet, provides reassuring data regarding the dosing schedule for the AstraZeneca/Oxford COVID-19 vaccine.  The team report that a longer gap of 3 months, compared with a gap of less than 6 weeks, between the first and second dose of the vaccine (such as that currently recommended in the UK) results in a vaccine efficacy of 82% after a second dose.  These findings were backed up by analysis of antibody responses in people aged 18-55.  After one dose of the vaccine, researchers found a vaccine efficacy of 76% when measured up to 90 days post-vaccination before the second booster dose had been given.

“This is positive news as it shows that just one dose of this vaccine generates good levels of immunity and that this protection does not seem to wane in the shorter term.  In immunology terms, this finding is not unexpected as we know that some other vaccines confer better immunity when the doses are more spread out.  Although further information is required to confirm these findings for older age groups, overall this new research should provide reassurance around the UK’s decision to offer the two doses of this vaccine 12 weeks apart.

“In addition to preventing people getting sick from COVID-19, the study also hints that the Oxford/AstraZeneca vaccine may be effective in stopping people being able to transmit the virus.  While this would be extremely welcome news, we do need more data before this can be confirmed and so it’s important that we all still continue to follow social distancing guidance after we have been vaccinated.

“While the results of this study show that a good level of immunity is still present at three months after one vaccine dose, it is important to remember that it is still the case that the highest and longest lasting protection from getting ill with COVID-19 will only be provided by getting two doses of the vaccine.  In order to get this protection, it is critical that all people eligible for COVID vaccination do return to get their second dose when asked to do so by their medical providers.”


Prof Jonathan Ball, Professor of Molecular Virology, University of Nottingham, said:

“These data confirm that the [chimp adenovirus] vaccine has an overall effectiveness – measured as a reduction in the number of cases of symptomatic SARS2 coronavirus infection – of around 67 percent.

“The data also shows that delaying the second dose potentially improves the protection seen after the receiving the second dose.

“The study also suggests a reasonable level of protection from symptomatic infection is offered by a single immunisation, and that this can last for several weeks, although the relatively low number of cases included in the analyses make the findings less robust.

“Whilst delaying the second dose of the AZ vaccine doesn’t adversely impact on overall effectiveness, such delayed dosing has not been trialled for the other approved vaccines; it might be reasonable to assume similar effects, but it can’t be guaranteed.

“Finally, the level of protection from asymptomatic infection following a single vaccine dose was very poor.  Although we don’t know the risk of onward transmission this must not be discounted, so it is essential that those receiving a single dose understand the risk of infection and subsequent spread and the need to complete the vaccine course; this is especially true for those who regularly encounter vulnerable people.”


Prof Helen Fletcher, Professor of Immunology London School of Hygiene & Tropical Medicine, said:

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

“This is a publication of the full analysis of the of the Oxford/AstraZeneca efficacy trial which follows on from the publication of the interim analysis on December 8th (Voysey et al).  This full analysis reports a vaccine efficacy of 66.7% which is consistent with the level of efficacy reported in the interim analysis.  The paper also includes additional data on vaccine efficacy following a single dose, variation in vaccine efficacy by interval between first and second dose and the potential of the vaccine to protect against viral transmission.  The data are robust although the authors acknowledge that analyses of vaccine efficacy by dosing interval is not pre-specified and the study was not specifically designed to address this question.

What does the data show and what are the implications?

“The data show that a single vaccine dose can provide protection of 76% for up to 90 days (Table 2) which is great news for those who have had their first dose of the Oxford/AstraZeneca vaccine and are waiting for the second dose.  There is not enough data to determine efficacy of single dose vaccination after 90 days.  When examining dosing intervals there is a trend for vaccine efficacy to increase as the interval between first and second dose increases.

“Vaccine efficacy after the second dose correlates with the levels of anti-spike IgG neutralising antibodies (Figure 4) and this is good news because we can monitor antibody levels as an immune correlate of vaccine efficacy.  The immune correlate after the first vaccine dose may be different and it would be important to explore this in future studies.

Is this good news for the 3-month dose gap, or is it not that simple?

“This data supports the approach taken by the UK which recommend a 3-month gap between first and second dose of the Oxford/AstraZeneca vaccine.

Is this good news for the vaccine and whether it prevents transmission, or is it not that simple?

“The data presented suggest a possibility that the vaccine could have an impact on transmission but further follow-up would be needed to confirm this.”

Question from journalist: The preprint says ‘Analyses presented here show that a single standard dose of the vaccine reduced PCR positivity by 67%, and that, after the second dose, the SD/SD schedule reduced PCR positivity by 49.5% overall.’  Does that mean the second dose makes it less effective re PCR positivity?

“It’s important to note the limitations of this study – there is limited length of follow-up after the second dose so we likely do not have robust data on the impact of the second dose on PCR positivity.”


Dr Gillies O’Bryan-Tear, Past Chair, Policy and Communications, Faculty of Pharmaceutical Medicine, said:

“These data from an additional analysis from the Oxford vaccine phase 3 study are encouraging in several ways: firstly they confirm data which had been referred to by JCVI when they announced the delayed dose policy, namely that efficacy increases with a longer dosing interval.  Efficacy after the first dose rose from 55% with an interval of less than 6 weeks, to 82% with an interval of 12 or more weeks.  The Oxford group noted that a similar effect has been seen with several other vaccines – Ebola, flu and malaria – and is a feature of the immune system which can take time to ‘spool up’ in response to a vaccine.  The data therefore support the JCVI recommendation to delay the second dose of the Oxford vaccine out to 12 weeks.  The data are not yet available for the Pfizer/BioNTech vaccine, but most commentators agree that it is likely to be the same with that vaccine, and indeed, other two dose vaccines.

“Secondly, the Oxford group released data on more than double the number of Covid cases from their study (332, up from 131 previously), and have shown efficacy after the first standard dose is 76% out to 90 days; and that efficacy of the two doses, given 12 weeks apart, is 82%.

“These figures are closer now to the efficacy figures seen with the mRNA and Sputnik vaccines, and underline what various researchers have been saying for a while: that the numbers, especially the point estimates of efficacy, may change as more data is looked at.

“As important as these new data are, the Oxford group has also reported the first definitive estimate of the effect of their vaccine on transmission: based on PCR tests on participants in the trial, whether or not they had symptoms: there was an estimated 67% reduction in viral expression after the first dose, in vaccine recipients, compared to unvaccinated participants.  Few data were provided on how this figure was calculated, for example, how many samples it represented; we await fuller data and the publication, which will appear in the Lancet shortly.

“The analyses announced today are probably post hoc analyses, i.e. analyses which were not planned in the initial protocol.  This can reduce their reliability, (because analyses may be ‘data-driven’ by what has been seen in the study, rather than a result of a priori hypothesis testing) but because of the large numbers enrolled in the study, most reviewers will have confidence in the findings.

“If the effect on transmission is confirmed for the Pfizer vaccine too, this would be a very positive, and could support information we heard from Matt Hancock yesterday suggesting there were signs of an effect of the vaccination campaign on case numbers in the over 80s, which he suggested indicate that a definite vaccine effect is being seen on the number of new cases in that age cohort (many of whom were vaccinated in December), over and above the effect of lockdown.

“If these vaccines reduce transmission to the extent reported, it will mean that the easing of social restrictions will be enabled sooner, than if we have to wait for herd immunity, (which may never in fact be achieved because of insufficient vaccine population coverage).  That would be the holy grail of the global vaccine rollout, and these data bring us one step closer.”


Prof Paul Hunter, Professor in Medicine, The Norwich School of Medicine, University of East Anglia, said:

“This paper includes important data and analyses on the impact of single doses of the Oxford/AZ vaccine and on the spacing between 1st and 2nd dose of vaccine.  The source data for these analyses are based on the phase 3 trials already reported but with a longer follow-up period and so more data to enable more precise estimates of effectiveness.  In my view the analyses presented here are sound and the conclusions drawn by the authors appropriate.

“Whilst there is a lot of information in the paper to me the most important findings are as follows:

“1. The data shows that the apparent increased efficacy of the half dose – standard dose regimen was in fact not due to the dose but to the increased spacing between the doses.  If the two doses were given less than 6 weeks apart the ultimate efficacy was only 55% but if the second dose was given at 12 or more weeks apart the efficacy was 82%.

“2. The analysis shows that the OxfordAZ vaccine is quite poor at preventing asymptomatic infection, though prevention of asymptomatic infection seems better with the longer gap but the confidence intervals are large.

“3. A single dose of the OxfordAZ provides good protection from 21 up to 90 days after the first shot 76% (59%, 86%) and importantly there is no evidence that is starting to decline in that time frame.  There were too few cases after 90 days to make any meaningful judgement about efficacy after that time.

“Taking all this evidence together the 12 week gap between first and second dose is clearly the better strategy as more people can be protected more quickly and the ultimate protective effect is greater.  Given the poor efficacy at preventing asymptomatic infections, the vaccine will not stop transmission of COVID but is will still go a long way to reduce the R value and transmission because there will be far fewer symptomatic infections and people who are symptomatic are rather more infectious than those who are asymptomatic.”

Question from journalist: The preprint says ‘Analyses presented here show that a single standard dose of the vaccine reduced PCR positivity by 67%, and that, after the second dose, the SD/SD schedule reduced PCR positivity by 49.5% overall.’  Does that mean the second dose makes it less effective re PCR positivity?

“I suspect this is because the confidence intervals are wide at this point and so these estimates are somewhat uncertain but could also be due to confounding because of different times and populations.”


Prof Azra Ghani, Chair in Infectious Disease Epidemiology, Imperial College London, said:

“This is a preprint that I assume is undergoing peer-review in a Lancet journal.  The authors have presented updated data from their trials in the UK, Brazil and South Africa – with most data coming from the UK and Brazil.  Additional data are now included from older age-groups although the numbers in these groups remain relatively small.

“The overall efficacy presented in the main results as pre-specified in the trial protocols and under the standard dosing schedule (SD/SD) is 63.1% (95% confidence interval 51.8%-71.7%), consistent with the interim efficacy reported late last year.

“The study authors have used the trial design to explore the impact that delaying the time between doses could have on the vaccine efficacy.  To do this they look at two groups: those that received two doses stratified by the gap between the doses, and those that only received a single dose.  They find that, in those that receive two doses, there is a trend towards higher efficacy with a  longer gap between doses.  They also report a higher overall efficacy with one dose (76%, 95% CI).

“Whilst it is tempting to interpret these estimates as indicative of a higher efficacy with a longer dosing gap, there are some important limitations that caution against this.

“Most importantly, the study was not designed to look at different dosing gaps or at one versus two doses.  This means that participants weren’t randomised and it is therefore quite possible that there are other things that are driving this apparent trend with dosing schedule.  This is evident from Table S2 which shows that participants who received a single dose were younger, more likely to be female, more likely to be a healthcare worker, more likely to be resident in Brazil and more likely to be white than those who received two doses.  In addition, those who received a single dose  were followed for a significantly longer period of time.

“This means that it is not sensible to compare the efficacy estimates from a single dose with those from a two dose.  Indeed, it is difficult to think of a biological explanation to support a higher efficacy from a single dose (76%) compared to across the trial as a whole (63%) – although if that comparison was made statistically it is unlikely to be important given the overlapping confidence intervals.

“The same data are not provided for the different dosing gaps as far as I can tell.  However, it is quite likely that there will be differences because the longer dosing gap was more likely to have occurred earlier in the trial (with the switch that was made from a one to a two dose schedule in the trial protocol) and the earlier trial participants were UK-based and younger.

“The only way in which this question can be answered robustly is in a prospective trial with different dosing schedules compared side-by-side.”


Prof Stephen Evans, Professor of Pharmacoepidemiology, London School of Hygiene & Tropical Medicine, said:

“These data are an extension of the previously published data on the Oxford/AZ vaccine.  They come from the same randomised trials as the original published paper and the data used to support authorisation of the vaccine in the UK and Europe.

“These extra data, together with analyses designed to look at particular aspects of the effect of the vaccine, are very helpful.  There is still some inevitable uncertainty around exact values of efficacy.  It must be remembered that the exact values vary with the time point when counting of events starts and when it ends.  The data definitely provide some evidence to suggest that the eventual protection from two doses of this vaccine are not worsened by having a longer than 28 or 42 day period between doses and tend to confirm what had been shown before, that if anything the eventual efficacy was better.  The data also do not provide any evidence that efficacy wanes after the first dose.

“There is evidence that the blood test results are reflected in the clinical outcomes.  As before there were no hospitalisations after 21 days in those vaccinated, while there were 15 in the unvaccinated.  There is still some uncertainty but the data are compatible with protection against severe disease being better than for mild disease.

“While these data should not be taken to say that all questions are answered, they certainly do not suggest that the JCVI advice on dose-spacing was in any way incorrect for this vaccine.  It does not directly address the same questions for other vaccines but there are no good reasons or data to suggest that they would have markedly dissimilar results.”


Dr David Matthews, Reader in Virology, University of Bristol, said:

“This is excellent news and very much what was expected from what we know about other vaccines.”



Preprint (not a paper): ‘Single dose administration, and the influence of the timing of the booster dose on immunogenicity and efficacy of ChAdOx1 nCoV-19 (AZD1222) vaccine’ by Merryn Voysey et al. was posted on the ‘Preprints with the Lancet’ preprint server.  This work is not peer-reviewed.


All our previous output on this subject can be seen at this weblink:


Declared interests

Prof Sheila Bird: “SMB wrote on 13 January to Matt Hancock about the need for tilted randomization to test the unprecedented policy-decision to defer to 12 weeks the 2nd dose of the Pfizer/BioNTech mRNA vaccine.  SMB hold GSK shares.  SMB is a member of the Royal Statistical Society’s COVID-19 Taskforce.  SMB served on UK’s Medicines Commission in the 1990s.”

Dr Penny Ward: “No COIs.  I am semi-retired, but I am owner/Director of PWG Consulting (Biopharma) Ltd a consulting firm advising companies on drug and device development.  Until July 2019 I was Chief Medical Officer of Virion Biotherapeutics, which was a company developing broad spectrum RNA therapy for the treatment/prevention of respiratory virus infections.  Between December 2016 and July 2019 I served as Chief Medical Officer of Virion Biotherapeutics Ltd, a company developing antiviral treatments for respiratory viral diseases.  Previous employee of Roche, makers of tocilizumab (anti IL6 antibody) and CMO of Novimmune, makers of empalumab (anti IFN gamma antibody).”

Dr Doug Brown: “Trustee of the Association of Medical Research Charities.”

Prof Jonathan Ball: “Receive funding to develop, and perform early-stage clinical trials of, a DNA-based COVID vaccine.”

Prof Helen Fletcher: “Director of International Development, UKRI.  Trustee of the Jenner Vaccine Foundation.”

Dr Gillies O’Bryan-Tear: “Former Head of Vaccine Clinical Development, GSK.  I am a pharmaceutical physician, semi-retired and have worked in a variety of fields, including oncology (cancer) drug development and vaccines development.  I have been appointed CMO of Scancell, a U.K. listed biotech Company which has a Covid-19 vaccine candidate in preclinical development.  Dr O’Bryan-Tear is Chief Medical Officer of Scancell plc, an early stage listed company developing vaccines for Covid-19.”

Prof Paul Hunter: “No conflicts of interest.”

Prof Azra Ghani: “Nothing to declare.”

Prof Stephen Evans: “No conflicts of interest.  I am funded (one day per week) by LSHTM.  They get funding from various companies, including Astra Zeneca and GSK but I am not funded by them, I have no involvement in obtaining funding from them and I am not an investigator on any grants obtained from them.  I am the statistician to the ‘meta-Data Safety and Monitoring Board’ for CEPI.  I am paid for my attendance at those meetings and will be paid expenses for travel if that occurs.  I am a participant in the Oxford/Astra Zeneca trial, and on 13th January 2021 learnt I had received the active vaccine.”

Dr David Matthews: “I have already published with the Oxford team on how their virus follows its pre-programmed genetic instructions using state of the art analysis of how the virus behaves in human cells ( However, I am independent of the Oxford team and I have no interests in any commercial aspects of this vaccine.”

None others received.

in this section

filter RoundUps by year

search by tag