Do we know if vaccines can actually prevent transmission of Covid-19?

The full data from the Oxford/Astra Zeneca (AZ) vaccine studies published this week provided information on whether this vaccine, in addition to preventing people becoming ill, also prevented asymptomatic infections.  What does the AZ data show on this important question? 

Differences between AZ and Pfizer studies

• I have described in previous posts the differences between the two vaccines in terms of how they work
• They also had one important difference in their study design:
  • Participants in the  AZ vaccine study collected a nose and throat swab every week to test for any development of Covid-19 infection even if they did not then have symptoms 
  • By contrast, the Pfizer and Moderna studies only attempted to identify people who became ill following the vaccine

• I have also discussed their different ‘headline’ results, with the suggestion that the Pfizer vaccine may be more effective than the AZ vaccine at stopping serious infection
• What do the AZ data tell us about stopping asymptomatic infection?

These are the AZ data published this week

• There is a challenge in interpreting the AZ data, already widely discussed in the media.
• This is that the first recruits to the study had, inadvertently, only half the amount of the virus in their first shot.  All participants had the full amount of vaccine in their second shot
• Appropriately, in presenting their data, AZ divided the participants into two groups:
  • Those who received half/full dose
  • Those who received full/full dose
• The headline results showed a much greater efficacy (around 90%) in their half/full dose participants compared to around 60% in their full/full dose participants. 
• It is not clear if this was due to:
  • A statistical quirk and with larger numbers the difference would disappear
  • Differences in these two subgroups: the half/full dose participants were younger and were recruited at a different stage in the pandemic
  • Half/full dose might be a better strategy for some scientific reason 
• AZ are undertaking further studies to tease out which of these may be the most likely
• However, AZ have now published their data on asymptomatic infections and the results are shown below:

• Let me ‘talk you through’ this graph that I have drawn
  • The coloured bars show the % reduction in infection with the two dose regimes (compared to the infection risk in the dummy vaccine): but divided into the reduction for people with symptoms (blue bars) and those without symptoms (orange bars)
  • When combining the results from both dose regimes (‘All’), there was just over a 25% reduction in the numbers becoming infected but who were asymptomatic 
  • Importantly, there was also no evidence of reduction in risk in those who had the full/full regime (less than 5%)
  • Interestingly the reduction was much greater (60%) in those with the half/full dose regime.  However, this figure is misleading as the numbers were too small to give an accurate estimate of the true effect. Indeed, the statisticians  involved calculated that with larger numbers the benefit in terms of reduction could be as high as 83% or as low as 6%
• Conclusion: at the moment we have no evidence from any of the trials that, in comparison with no vaccine, the currently licensed vaccines can prevent asymptomatic infection

So, why does this matter?

Before discussing the vaccine results, I want to give some background about some of the issues surrounding asymptomatic viral infection

What does asymptomatic viral infection mean?

• Stating the obvious, it describes the state where an individual is capable of transmitting the virus to others but is not ill
• That implies that the virus has entered the human cells and started reproducing
• That is all a virus aims for in life, to make multiple copies of itself (limited ambition I know!)
• It does this by taking over the cell’s genetic machinery and starting to multiply
• We become sick because the cells have to respond to this attack and in this response the cells and tissues get damaged
• In some people though the virus seems to happily multiply but the response of the cells has no harmful effect on the body and there are no resulting symptoms

Is it easy to be sure that someone is asymptomatic?

• For diseases such as Covid-19 , it is not easy to define a border between having or not having symptoms.  This
  • might be easy if a fever is present
  • is much harder if someone reports a less specific problem such as feeling tired
• Further, to study this there needs to be a period of time that can be used as a cut-off to say symptoms are linked to the virus.  Typically,  this is often set at two weeks
• There is also the problem that  in some studies people are only asked about symptoms on the day of their swab test 

What about other viral infections?

• Because of concerns about transmission of virus from asymptomatic individuals, there has been much more attention paid in the Covid-19 pandemic to counting the number of asymptomatic cases
• Indeed, this is unique for major outbreaks of infection as normally numbers of cases refer just to those clinically diagnosed – mainly because:
  • Viral testing is expensive and difficult (as we know from the early stages in this pandemic)
• As a consequence, it hasn’t been done on this scale before 
• As an example, as far as seasonal flu is concerned there are  very few data
  • One estimate though  is that about 50% of cases are asymptomatic 

How frequently are people with Covid-19 asymptomatic?  

• Sadly, there is not a simple answer to this and the reported numbers vary considerably
• In the UK participants in the AZ study (and to be honest the paper is a bit confusing!) of the 354 people who tested positive during the trial, 69 (around 20%) were asymptomatic  
• We know that children and young adults are less likely to be ill than older people so  unless we know the age group tested, the ‘headline’ proportions are misleading.
• A recent report looked worldwide to find any research paper that could provide information on the frequency of asymptomatic people
• This found several studies, each of which gave a different estimate of the percent who were asymptomatic but these estimates varied widely
• This report also looked at results from studies not only from the general population but also from other groups with high rates of positive tests. The results are shown below

• Studies of contacts of infected people suggested between one in twelve up to half of these people who had themselves tested positive did not have symptoms
• Studies  of health workers who had tested positive, also produced very varying estimates – indeed as shown in this Table there was even one study when none of the workers had symptoms

Are people who are asymptomatic less likely to transmit infection to others?

OK vaccines may stop people becoming ill but might not stop them being infected with the virus. The key question is then: “Are such asymptomatic people less likely to pass the infection on?”

• A Swiss study in November compared the risk of contracting the infection from someone in your household, depending if the initial case had symptoms or not
  • Interestingly the rate of transmission from a household contact was 75% lower if the contact had no symptoms than if they had had symptoms
• Other studies have meticulously traced and tested all contacts of people who tested positive.  These studies then compared the rates of infection between the contacts of symptomatic with contacts of asymptomatic people.  
  • The results from the four studies I found are shown in the graph below

• To explain using the example from from Study A: around 4% of the contacts traced from symptomatic people (orange bar) tested positive compared with less than 1% of those from asymptomatic people (blue bar)
• The other 3 studies though produced different estimates (of course this happens all the time in epidemiology!) but overall the conclusion is that people who don’t have symptoms transmit the infection to fewer people
• Interestingly this may be because asymptomatic people
  • have less virus
  • shed less virus because they cough and sneeze less
  • may get rid of the virus more quickly 

What can I say in conclusion

• You are perhaps 4 times more likely to catch Covid-19 from a person who has symptoms 
• But you can catch Covid-19 from a person who is asymptomatic 
• Vaccines have yet to prove that they can reduce the chance of developing an asymptomatic infection: good for their risk but less useful for public health as such people could still transmit the infection
• It is thus going to be harder to achieve herd immunity with the vaccine alone
• This all points to the need for a very high vaccine take up to achieve the likely end of the risk of life threatening infection .

One final scientific thought!

• I am not an expert in the immunology of viral infections 
• The accepted wisdom is that antibodies work in part by stopping the virus entering cells and hence stop it multiplying. 
•  If vaccines succeed, as it seems likely that they do, and produce effective antibodies, why don’t these antibodies stop virus multiplication?
• I have done some reading and I have seen this question raised by others, but I have not seen a definitive answer apart from antibodies behaving differently in different people.

If you would like to receive email notifications of new posts on this blog just click this link and enter your email at the bottom of the home page: https://makingsenseofcovid19withs.com