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

expert reaction to a study on the results of phase 3 human trials of the Oxford/AZ COVID-19 vaccine

The results of the phase 3 clinical trials of the Oxford AstraZeneca COVID-19 vaccine candidate have been published in The Lancet.

This Roundup accompanied an SMC Briefing.


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

“The Oxford/AstraZeneca research has been conducted at such remarkable, overlapping pace that earlier decisions have had to be revised (e.g. in respect of administering a second or booster dose; eligibility by adult age-group; choice of placebo).  In addition, the research programme has had to contend with quality-control issues which led to the earlier-randomized volunteers in the UK trial COV002 having received, as an inadvertent combination, low dose/standard dose (LD/SD) rather than SD/SD.

“Hence the UK COV002 efficacy trial featured as two sub-trials: first, 1367/2741 randomized volunteers received LD/SD (aged 18-55 years, enrolled between May 31 and June 10, 1357/1367 (99%) had greater than 8 weeks between Ox/AZ vaccine doses); thereafter 2377/4807 randomized volunteers received SD/SD (aged 18-55 years, enrolled between June 9 to July 20, 1407/1879 (75%) had greater than 8 weeks between Ox/AZ vaccine doses; aged 56+ years, enrolled from August 8).

“Based on 8895 volunteers randomized in Brazil and UK between SD/SD and control, vaccine efficacy for SD/SD was reported as 62% (95% CI: 41% to 76%) for primary symptomatic COVID-19 disease so that efficacy below 50% was not excluded.  The protocol variation to administer a second dose came into play rather late so that most volunteers in the UK trial received their booster dose more than 8 weeks after their initial vaccine.

“International regulators, with whom protocol variations were discussed, may – like the triallists themselves – react to the dossier of unblinded data now at hand, or may wish to defer decision-making until the research-team itself can clarify any intriguing science (immunology or vaccinology) underlying the empirical evidence that, in the UK trial, LD/SD afforded better protection than LD/LD did.

“At least two routes to resolution may be underway: via final rather than interim analysis of Lancet-published studies; and a new trial designed specifically to test the inadvertent but serendipitous combination of LD/SD.  A new trial may also seek to control (or test) the time-interval between vaccine-doses.

“All volunteers, irrespective of their received combination of active doses, contribute to appraisal of vaccine safety about which there is much reassurance.”


Dr Charlie Weller, Head of Vaccines at Wellcome, said:

“Today marks another key milestone in the Covid-19 vaccine journey, with the first peer-reviewed set of data from a Phase III Covid-19 vaccine trial shared openly with the world. Although we await the trial completion and full data, it is highly encouraging to see the data behind the interim results announced last month, including an analysis of the different dosing regimens. This suggests that this vaccine could prevent asymptomatic disease.

“It’s important to remind ourselves that there are still important questions to be answered on the efficacy of this vaccine across different age groups and in different settings, which we will only be able to fully understand once Phase III is completed.

“Safety is the single most important consideration when developing any vaccine, which is why regulatory bodies such as the MHRA, FDA and EMA will thoroughly assess any vaccines before they rolled out. Therefore, it remains critically important that this trial can be formally completed, and regulators can begin to independently and rigorously assess the full data set.

“We should applaud the University of Oxford /AstraZeneca team for transparently sharing their full data set and we hope to see the independent assessment and publication of data from the other successful vaccine candidates in due course.

“The reassurance this published data provides is particularly welcome as the University of Oxford/AstraZeneca vaccine can be easily administered in existing healthcare systems around the world. It can be stored at fridge temperature and can use existing delivery mechanisms, giving it a distinct advantage for swift and successful global rollout.

“It is remarkable to be at a stage where we have multiple vaccines on the horizon. But we must not be complacent. If we are to produce the billions of doses needed for the world, we will need a range of vaccines that work across different groups and settings. We must keep investing in the next generation of vaccines and make sure that they are fairly available globally, with all countries prioritising vaccination of those most at risk of serious illness from Covid-19 and healthcare workers.”


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

“These findings from a first interim analysis of ChAdOx1 nCoV-19 reveal 5-10 years of vaccine development effort condensed into less than 8 months. Overlapping and pooling data from all studies, Phase  I, II and III, increased the speed of testing without compromising the scale of data required for evaluation of safety and efficacy.  We can now see that confounders including study, treatment group, age group and length of time on study were controlled for in the analysis.

“The authors acknowledge that the observed 90% efficacy in the low dose/standard dose group could be due to chance, but as this regimen also gave better protection against asymptomatic infection (58.9%) it’s more likely to be a real finding. This is exciting as a vaccine that prevents asymptomatic infection could reduce transmission of the virus.

“With follow-up time of only a few months there are still important questions to be answered on duration of efficacy and more data is needed to assess the low dose/standard dose regimen, efficacy in older adults and different ethnic groups but overall, these data are very positive and support the findings released a few weeks ago.”


Professor Fiona Watt, Executive Chair of UKRI’s Medical Research Council said:

“This publication is another landmark  for the vaccine. It gives us all reassurance that the vaccine is safe and effective. There were no cases of severe disease in those vaccinated. We’re another step closer to the vaccine becoming approved. One of the most exciting components of the Oxford vaccine is that it can be rolled out using existing infrastructure and doesn’t need to be stored at -70oC. The new vaccine builds on decades of MRC funded research and investment. We’re indebted to the people that have made this possible – not only the lead researchers but also the study volunteers, postdocs, technicians and administrators.”


Dr Sheuli Porkess, Chair of the Policy and Communications Group of the Faculty of Pharmaceutical Medicine, said:

“This is an interim analysis and collection of further data is ongoing. The review by the regulators is a critical step during which they will look at all the data relating to safety, efficacy and quality to determine whether the data meets their standards and if it can be granted an authorisation for use.  This would include defining which populations the vaccine would be suitable for and clarity on the dosing schedule (how often and at what dose). If granted, the conditions for any such approval would then need to be met by any subsequent roll-out for clinical use.”


Dr Julian Tang, Honorary Associate Professor/Clinical Virologist, Respiratory Sciences, University of Leicester, said:

“The Oxford-AstraZeneca vaccine has had its fair share of issues recently.

“We are now all familiar with the vaccine efficacy results from the low-dose (90%) vs. standard-dose (62%) regimens which are quite different. The combined efficacy of 70% doesn’t actually matter as when it comes down to it, you will get one or the other dosing regimen with that particular efficacy – though it may take a while before the low-dose regimen is potentially approved.

“A lower vaccine efficacy does run the risk of leaving a large proportion potentially unprotected despite being vaccinated. Those who have been vaccinated and think they are immune may behave more freely which may serve to spread the virus further if they do become infected.

“The other finding is that there is insufficient data as yet to confirm the vaccine efficacy against those over 55 years old. So if it is licensed by the MHRA, will it just be for the standard dose regimen but only for those aged 18-55 years? Or will they assume some level of efficacy (>50%) in those over 55 years also – given the current COVID-19 pandemic crisis?

“Also, this interim analysis did not contain sufficient data to demonstrate vaccine protection in BAME patients and those with comorbidities, both of which are known to be particularly vulnerable to severe COVID-19 disease.

“More data will come from the various arms of the trial to hopefully answer some of these outstanding questions to support its licensing.

“Its main advantage over the mRNA vaccines are its ease of storage and transport at normal fridge temperatures (2-8 C), as well as its much lower price.”


Dr Alexander Edwards, Associate Professor in Biomedical Technology, Reading School of Pharmacy, University of Reading, said:

“The vaccine is clearly safe, and has been into very large numbers of trial participants without any problems. It clearly provides protection that exceed expectations when the trials were planned.

“One question that arose from the initial press release surrounds a different dose used at an early stage- with a “half dose better than full dose” headline- this study now reports in more detail the numbers behind those headlines and terms these groups “SD/SD” and “LD/SD”. The dose and manufacturing batch information is explained fairly clearly in the supplementary information provided with the manuscript along with a great deal of trial details, and we must applaud and celebrate this transparency in sharing everything openly, including explaining that an initial batch had a different than expected dose (the LD/SD protocol).

“Within this, a lot of detail is provided around exactly how you measure and analyse the vaccine dose before putting it in a vial for use. This is a well-studied and well-known area in the field of vaccine manufacturing. Although at first sight you’d think you can just “count the virus particles”, what is produced when this type of vaccine is manufactured and purified is actually a mix of particles, and they can be measured in at least three ways: the amount of nucleic acid determined spectroscopically, the number of copies of viral RNA, and the number of infectious particles. There is nothing alarming or unusual about this but it is a real challenge; this is why vaccine manufacturing is tightly regulated and requires very specialised factories and expert bioprocess engineers. The world has never been better at manufacturing complex biological medicines- we have better than ever tools and equipment. This is also why innovation in vaccines remains vital- to develop vaccines that are easier to produce, ideally less complex and ideally chemically synthesised rather than ‘grown’. But often, the biologically complex vaccines just work better than simpler compositions- this is what our immune system evolved to recognise.

“What is surprising is that a small difference in composition would have any effect on the vaccine protection. The dose was chosen as having the optimal immune response in previous studies, but lower doses also produced similar immune responses in earlier studies. There isn’t a well-known immunological mechanism that would predict that “less is more”.

“However, this apparent difference in effectiveness between different batches will now certainly be a major focus, to make sure that we can make best use of this type of vaccine. We eagerly await further trial data from all ongoing vaccine trials including this one- over time the picture should become clearer. The “dose question” may be a distraction from the main findings of an effective vaccine that meets the target protection, and as a positive person my glass is half-full and the half-dose result will help us to better understand the best use of such vaccines. I remain reassured that vaccination programs have finally started us on a welcome path to recovery.


Prof Matt Sydes, Professor of Clinical Trials & Methodology, UCL, said:

“It is complicated to put 4 trials together, especially combining early phase (Phase I and II) studies with late phase (Phase III) trials, but the authors have reported clearly which trials contribute to which parts of the analysis. The authors recognise the issues in interpreting data which come from a mix of trials and phases, and with a mix of doses / treatment schedules, participant ages and follow-up lengths. The conclusions are modestly presented and seem reasonable from the data presented.

“In most instances, Phase I studies are small and non-randomised, but as these are vaccine trials, these are large and randomised. All 4 of the trials contribute to the safety assessments. Only the trials with a Phase III element (COV002 and COV003) contribute to efficacy analysis; the authors explain that this is because the two phase I trials have too few events (against a pre-defined number) to contribute yet.

“The error which led to the low-dose and standard-dose combination and the implications for analysis are well explained. The analysis plan was widely agreed, including with regulators. The authors are transparent about the error and how this was all discussed with regulators and say they will come back to this. The timing of the second dose of the vaccine compared to the first dose sounded like it was quite variable and was unclear to me on one reading.”

“From previous press releases, I had been genuinely uncertain whether we were only hearing about data up to 15 days after the second vaccine dose, but now it’s clear that all available follow-up is accounted for. We can see in the manuscript that median follow-up is 3.4 months for safety and 2.0 months for efficacy i.e. in the efficacy analysis half of the included patients have less than two months follow-up yet. The follow-up numbers also become clearer in the manuscript’s one figure: no participant had yet reached 100 days of follow-up for efficacy from the second dose. The longer-term elements of this pair of graphs (the participants still in the risk group towards the right of the graph) are driven by Low-dose/Standard-dose combo and younger patients so these patterns may change over time.

“The timings of events after the second dose is partly driven by potential for exposure: there were fewer cases around in the UK in the earlier part of this trial, so it is important to be driven more by the relative effect than the absolute effect. “Vaccine efficacy” is a summary of that relative effect.

“The safety signals seem okay with lower numbers of serious adverse events (SAE) and adverse events of special interest (AESI) on both arms and no obvious concerning increase with the vaccine. (Details in their supplement.)

“Unsurprisingly (and as planned), the patients in the study were younger than the general population and with fewer comorbidities.

“The press release mentions 23,745 participants which isn’t readily findable in the manuscript but can be calculated from Figure S1 in the supplement: 12021 vaccine plus 11724 control.

“These appear to be good findings. Like the authors, I look forward to learning about the duration of coverage. I would like to know what will trigger the main analyses; I suspect this is in the protocols but I couldn’t easily find this in the time available to me.”


Prof Ian Jones, Professor of Virology, University of Reading, said:

“The data released today add a lot of flesh to the bones of the earlier press release. AZD1222 is undoubtedly safe and is also capable of preventing disease, so its general roll out would impact the epidemic. Its cost and availability are also powerful positives. But further trial data might be needed to explain why the lower dose group was significantly better protected than the standard dose group, not least as this is not what was reported for the phase 2/3 trial published in November. The number of viruses used in these doses is so high that a two-fold difference would not normally be expected to cause such a difference so it’s possible that something else, manufacturer or batch perhaps, could have played a role. Immunological measures, once completed, may shed some light on this. Subject to the dose or batch issue being resolved I would expect the vaccine to be approved and to play its part in bringing the pandemic to a close.”


Prof Deborah Dunn-Walters, Chair of the British Society for Immunology COVID-19 and Immunology Taskforce and Professor of Immunology at the University of Surrey said:

“Today’s publication in The Lancet of the pooled analysis of the interim phase 3 trial results of the University of Oxford/AstraZeneca COVID-19 candidate vaccine, ChAdOx1 nCoV-2019, is good news. The research team should be commended for publishing the interim results of this trial in a peer review journal so quickly to allow scrutiny from the scientific community.

“Phase 3 trials are designed to assess both the safety and effectiveness of vaccines in a large group of people who volunteer to take part. The finding that no individual who received the COVID-19 vaccine subsequently suffered from severe COVID-19 disease is particularly reassuring. Intriguingly, the team report a higher efficacy of up to 90% when using a halved first dose and standard second dose compared with giving two standard doses.  This finding warrants further investigation, particularly as there were differences in participants’ age profile between these parts of the trial – we know that for many other vaccines, age is often linked to effectiveness. The paper reports reassuring findings on the safety front with a good safety profile across the arms of the trial reported.

“While the findings reported today are positive, there are still more stages to go through before this vaccine can be approved for use.  In particular, we are reliant on the expert independent assessment by the UK’s Medicines and Healthcare products Regulatory Agency (MHRA).

“The researchers working on this vaccine, both in the UK and globally, should be applauded for their monumental efforts in developing this vaccine so quickly. The UK leads the world for the quality of our immunology research and this is another great example of how the community has come together to drive forward scientific discovery into this pandemic. Equally, we all owe a huge debt of gratitude to the many thousands of people who have volunteered to take part in these trials – they have played a critical part in the development of this vaccine.”


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

“These results reflect the headlines given in previous press releases, but with both a fuller explanation of what was done and more detail around the uncertainty in the estimates of efficacy.

“The unfortunate measurement technique problem that occurred with some doses produced in Italy is going to be seen by some as raising questions about this trial. It is clear however, that national and international regulators were fully apprised of what had happened and a plan for analysis was determined prior to knowledge of the results.

“The basic message that the overall efficacy across the trials that are reported here is about 70% but with a clear description of its uncertainty. The statistical uncertainty is that the efficacy could be as low as 55% or as high as 80%. The Pfizer/BioNTech and Moderna vaccines both have efficacies above 90% and are clearly more efficacious under trial conditions.

“The decision to report the subgroup that received a low dose as the prime [1st dose] followed by a standard dose as the boost [2nd dose] separately, was made prior to knowledge of the results and was expected to be about the same efficacy as giving two standard doses. This trial was originally intended to only have a single dose, but the early phase results made it clear that a worthwhile gain in efficacy would be obtained using two doses.

“The description of the reasons for having an unexpected low dose are set out clearly and there is some statistical evidence that the different dosing subgroups have different efficacy. However, to believe that the lower dose followed by a standard dose will actually have 90% efficacy must at best be regarded as provisional and further data will be needed to confirm this higher efficacy as the authors make clear. There is evidence from the asymptomatic infections that suggest the efficacy difference is real but considerable uncertainty remains and will only be fully clarified by extra data.

“What is clear is that whether the pre-planned pooling of all the results to obtain an efficacy of about 70% or the standard dose sub group of about 62% are taken as the principal results, the vaccine efficacy is well above the threshold set by regulators prior to knowing any results.

“The total numbers who have been vaccinated with the ChAdOx vaccine determines the uncertainty in any detected harms. While the reasons for pausing this trial in relation to multiple sclerosis or transverse myelitis do not seem to have translated into evidence that the vaccine caused these problems, it is clear that very rare effects which occur less frequently than about one in 4000 people vaccinated would not necessarily be detected in a trial of this size. It is also obvious that for all the vaccines which are likely to be authorised for use in the near future that careful follow up in use in practice rather than in trial settings will need to be done. Whether the trial efficacy will translate into effectiveness in real world usage might depend on the success of keeping the cold chain as well as on other factors. It is also clear that adverse effects that only appear more than a few months after vaccination will not have been detected in any of the trials, however these are unusual with vaccines and generally so rare that the benefit of vaccination will outweigh the risks. This will need to be confirmed by careful follow up of all those vaccinated over a longer time.

“The UK is not in this pandemic on its own but must be aware of global health needs. This vaccine is going to be easier to administer in many situations, especially in low and middle income countries and this is of great importance.”


Dr Simon Clarke, Associate Professor in Cellular Microbiology, University of Reading, said:

“The report on the successful trial of the Oxford/AstraZeneca presents regulators with something of a dilemma.  Data are most compelling for the cohort who got half a dose of the vaccine in their first jab.  Not only does this seem to confer greater protection against disease, it is in this group that there is a reduction in asymptomatic transmission of the virus, something which is essential if herd immunity is to be obtained to get wider protection of the population.  Unfortunately, this cohort was relatively small, reducing the reliability of the findings – moreover it did not contain any older participants (age 55 or over) and it remains possible that if the regulators allowed the vaccine to be used in this manner, the most at risk group may not be protected.  The researchers have already indicated that further trials may be needed to determine whether this way of administering the vaccine is indeed effective.”


Dr Michael Head, Senior Research Fellow in Global Health, University of Southampton, said:

“This confirms the previously-released findings, namely that safety and effectiveness of this Oxford COVID-19 vaccine candidate are overall very good. It is possible that we will see this vaccine reviewed quite soon by the MHRA, and thus there may be the option of two available vaccines in the near future.

“However, the researchers were not yet able to fully assess how effective this vaccine is in elderly populations, so we are likely to see a continuation in the use of the Pfizer vaccine in older people. Due to the need for -70C storage, this will pose challenges particularly with taking the vaccine to care homes. However, for younger populations and for low- and middle-income settings, the Oxford candidate could be the vaccine of choice. It is stored at lower temperatures and is much cheaper than Pfizer or Moderna products.”

Possibly useful links –



‘Safety and efficacy of the ChAdOx1 nCoV-19 vaccine (AZD1222) against SARS-CoV-2: an interim analysis of four randomised controlled trials in Brazil, South Africa, and the UK’ by Voysey et al was published in The Lancet at 16:00 UK time on Tuesday 8th December.

DOI: 10.1016/S0140-6736(20)32661-1



Declared interests

Prof Sheila Bird: “SMB served on UK’s Medicines Commission in the 1990s and holds GSK shares.”

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

Dr Charlie Weller: “Wellcome co-founded and co-funds CEPI, which has provided support to the development and manufacture of the University of Oxford/AstraZeneca AZD1222 vaccine candidate.”

Dr Sheuli Porkess: “Director of Actaros Consultancy Ltd. and clients include the Association of the British Pharmaceutical Industry”

Dr Julian Tang: Dr Julian W Tang is a clinical virologist who supervises diagnostic virology testing and advises clinical teams and GPs on the management and infection control of viral infections – including advice on antiviral and vaccine safety and efficacy.

Dr Al Edwards: “I don’t work on current COVID vaccines nor have any commercial vaccine activity.”

Prof Matt Sydes: “No.”

Prof Ian Jones: “No conflicts.”

Prof Deborah Dunn-Walters: “No interests 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.”

Dr Simon Clarke: “No declarations.”

Dr Michael Head: “No declarations or conflicts of interest.”

None others received

in this section

filter RoundUps by year

search by tag