EXPERT REACTION: AstraZeneca COVID-19 vaccine efficacy questioned

It is such a shame that a vaccine with so much evidence supporting its safety as well as efficacy continues to be shrouded in controversy. We heard very positive news almost a week ago that the EMA’s safety committee concluded that the vaccine was not associated with an increase in the overall risk of blood clots and therefore the benefits of receiving the vaccine continue to outweigh the risk of side effects.

This was followed just two days ago by the press release of what certainly appears to be promising results from the interim analysis of the US phase 3 trial that reported 79 per cent efficacy, comparable efficacy across ethnicity and age, and a favourable reactogenicity and overall safety profile.

It is important to note that this trial used a shorter dosing interval and it appears data support a higher efficacy with a longer interval as is the plan in Australia. Given these results have only been reported in the format of a press release we do need to interpret them in this context and await the formal reporting of these results in a peer-reviewed publication.

Unfortunately, after appearing to have all of the perceived issues addressed, the NIH statement yesterday has once again fuelled some scepticism with relation to this vaccine. This was based on some reported concern by the DSMB that outdated information may have been included thus providing an incomplete view of the efficacy. To me, this seems like a highly atypical approach to resolving an issue such as this and is very unfortunate in terms of the perception of this vaccine.

In response, AstraZeneca has stated that the numbers published were based on a pre-specified interim analysis with a data cut off of 17th February. They went on to say they have reviewed the preliminary assessment of the primary analysis and the results were consistent with the interim analysis. This to me does not signal cause for concern with the vaccine or the science supporting its use.

Perhaps, though, there is an opportunity to improve the communication of some of this information and resolve issues such as this directly rather than letting it play out in the press. What is clear is that regulatory approvals are not based on press releases and that the data available thus far has, and future data will continue to be, made available in full to regulators for review. Issues such as this therefore should not detract from our confidence with this vaccine nor its use in our or other countries.

As the only COVID-19 vaccine currently able to be manufactured in Australia, and the exciting news that the first four batches have been approved for release, we need to continue to support its use in an evidence-based manner with faith in our regulator the TGA, and not falsely sound the alarm over coincident events with no established link or based on perceived issues with how results are communicated in the press.

AstraZeneca has just released its latest COVID-19 vaccine trial results, which looked highly impressive: They supported their previous studies showing that their vaccine is highly protective against serious disease, and found better protection than their earlier trials against symptomatic infection.

However, there has clearly been a breakdown in process, with the US National Institutes of Health publicly criticising the company for not presenting the most complete data available, for reasons that remain unexplained. The company, in response, is in the process of redoing the statistics to include the most comprehensive data, and has suggested that the findings will not change in a meaningful way.

We can remain optimistic that this is the case, but the episode illustrates what can go wrong as these huge trials of new products are rushed out in time frames that were previously inconceivable in vaccine development. Public confidence is always fragile, and it is particularly important for accepted research processes to both be maintained and be seen to being maintained.

From an ethics viewpoint, there are at least three key takeaways from the current AZ-NIAID situation:

First, the National Institute of Allergy and Infectious Diseases (NIAID), who issued the statement, is distinct from the Food and Drug Administration (FDA) in the US, the latter being the group who regulate and permit drugs and vaccines to enter the market. On the other hand, the purpose of the NIAID is to support research, including clinical trials. The Data and Safety Monitoring Board (DSM) of the AZ COVID-19 vaccine trial went out of its way to contact NIAID, which as many have pointed out, is highly unusual, if not unprecedented. The purpose of a DSMB is to help uphold the soundness of a trial by being an independent means to evaluate safety-related data in order to protect the trial participants, first and foremost. That it felt the need to go directly to the NIAID, which could be argued is outside its remit, makes one wonder the extent to which the DSMB felt it had a duty of care beyond its specified goal of independently ensuring the safety of trial participants. Although the NIAID is separate from the FDA, in all likelihood, the FDA will pay special attention to the events that are currently unfolding. 

Second, it provides evidence of the perils that accompany 'science-by-press-release'. The purpose of peer-review, in part, is to allow other scientists to ask tough questions of the methods and results presented by the study sponsors, in this case Astra Zeneca. It provides a layer of scientific critique that, though different in scope and manner, can help supplement the analysis of the DSMB; this layer of critique has continuingly been bypassed in the name of quickly getting results in the hands of decision-makers. Of course, regulatory bodies such the FDA and the Therapeutic Goods Agency in Australia, often (always?) have access to data in an unencumbered manner that the general public does not. It's not clear that companies actually need to release data via press releases except to connect with members of the general public and politicians, thereby driving up demand (and stock shares) by raising hopes. The problem, then, is when the results released via press releases do not stand up to the scrutiny of peer-review or the review processes of national regulatory bodies; in such instances, one might argue a real harm has occurred. 

This leads to a third, and final, ethical challenge, namely, it may call into question the trust that the public has in the whole enterprise of vaccines. One thing is for there to be issues that are picked up in post-market surveillance (i.e., the process by which governments keep track of potential adverse outcomes and effectiveness of new technologies once it has been rolled out). This was the case a few months ago, for example, when Norway was investigating the deaths of elderly patients given the Pfizer/BioNTech vaccine. It is another matter altogether what (even the appearance of) impropriety does to public trust in the AZ vaccine, and vaccines in general, when the DSMB of the trial contacts the NIAID, which then  brings us back to the first point made above.

The US National Institutes of Health released a statement that AstraZeneca may have included outdated information from their initial studies, which may have provided an incomplete view of the efficacy data. This is really a governance and compliance issue regarding the ongoing reporting of data. It is doesn’t in any way suggest there was a problem with the data or indeed misrepresentation of data.

This statement should be seen as reassuring that governance compliance and documentation of results are in place and working. There are now in excess of  400 million vaccination doses given globally. We are continually seeing that those who have been vaccinated are not being admitted to hospital with an escalating pneumonia illness related to COVID-19, and no deaths. This is truly a remarkable result for medical science.

The Data Safety Monitoring Board (DSMB) in the United States has expressed concern that data provided by AstraZeneca from its initial COVID-19 clinical trial may have included outdated information that influenced the representation of the vaccine’s efficacy.
 
This appears to be either an oversight or delay in processing trial data rather than any perceived attempt at deception.

AstraZeneca have clarified which data were used for trial analysis and provided an indication that validation of interim data via statistical analysis will be made immediately to the DSMB once this is completed.
 
The concerns placed on this event are perhaps more of a reflection of specific regulatory guidelines placed on the use of the vaccine by the Food and Drug Administration (FDA) and Centres for Disease Control and Prevention (CDCP) in the United States.
 
Variations in the regulations that govern the use and supply of vaccines do exist between countries where the vaccine is being considered for use.

The concerns expressed by the FDA and CDCP are not ones which question the safety of the AstraZeneca vaccine, but rather how efficacious the product is in preventing COVID-19 transmission and mortality.