Mixing vaccines: taking two doses of different vaccines seriously!

This week has seen research supporting the idea that having the two doses from different vaccines would be more effective than two doses of the same vaccine.  This idea is being taken seriously and thus this week there were announcements of two trials:

  • In the UK with AstraZeneca and Pfizer vaccines as the different doses
  • In Azerbaijan & UAE using AstraZeneca and Sputnik vaccines as the different doses. 

In this post I discuss the background and what may be the potential benefits of such ‘vaccine mixing’.

The case for these 2 combinations are different

1. Pfizer/AstraZeneca

Rationale for a Pfizer/AstraZeneca 2-dose vaccine regime

  • Pfizer and AstraZeneca  vaccines are both designed to produce the spike protein that when injected leads to immunity to the real virus
  • The spike protein produced by these 2 vaccines should be identical as the genetic instructions were based on the same genetic code, they just work in different ways
  • Theoretically then the body’s immune response to either of these vaccines should be similar.
  • Readers may recall that we have two immune responses to vaccines:  
    • Antibodies 
    • T cells*
  • These 2 immunity weapons both need to work to get full protection
  • It was thus interesting that a (not peer reviewed) paper published  a week ago showed that there may be differences between these two types response  between the 2 vaccine:
    • Pfizer is better at antibodies
    • AstraZeneca is better at T cells

*Some readers may have seen the recent TV series “It’s a Sin” about HIV/AIDS.  The pneumonias that caused some of the deaths caused are a consequence of the failure of some T cells (CD8) that protect against infections in people infected by HIV

  • Thus having these 2 different vaccines can have a theoretical advantage
  • A second consideration relates to side effects:
    • Side effects from the second dose of the Pfizer vaccine were reported this week to be worse than after the first dose
    • Switching vaccine to the AstraZeneca could possibly protect against this 
  • There are also logistical  justifications for testing this mixing:
    • Given the complexity of how they are made, production of these vaccines is a precarious process and even with all the quality controls in place, something could go wrong, limiting the availability of the product
    • AstraZeneca vaccine is far cheaper to produce and distribute 
  • So knowing that one of these vaccines can substitute for the other can be very helpful

Rationale for  AstraZeneca/Sputnik two dose regime

First let me introduce you to the Sputnik vaccine!

  • When Russia announced the launch of their ‘Sputnik’ vaccine in August 2020, Western media were concerned that the necessary clinical trials had not been completed prior to its wider use.  
  • Thus it might seem strange that, only a few months later, formal clinical trials are going to start in collaboration with AstraZeneca  
  • Whatever the views about the political situation in Russia, this is an interesting vaccine!
  • Basically, it is the same type of vaccine as the AstraZeneca one: so-called ‘viral vector’ vaccines, and also planned to be given in 2 doses 
    • The  term ‘vector’ means that the active bit of the vaccine – the genetic instructions to make the spike protein – is carried on a harmless virus 
    • The harmless virus in Sputnik is one that affects humans whereas that in AstraZeneca is a monkey virus – though this difference probably doesn’t matter
    • However, whilst the AstraZeneca vaccine uses the same monkey virus in both doses, the two doses of the Sputnik vaccine uses two slightly different human viruses 

Is the Sputnik vaccine successful?

  • Results from the Sputnik vaccine were published in full in the Lancet this week
  • Note that the first reports of the success from the other vaccines have come from a press release or an early version of the results not checked by peer review 
  • The results of this clinical trial of Sputnik with placebo, comparing 2 doses 3 weeks apart with a placebo, were impressive
  • The number of cases in each age groups was too small for very accurate analysis, but there was no indication that Sputnik is less effective in people aged over 60 
  • They didn’t collect any data on asymptomatic cases so we can’t comment on whether Sputnik reduces transmission
  • Conclusion: The results from Sputnik, albeit in very different populations, are at least as impressive as the AstraZeneca vaccine 

Back to the rationale for mixing these vaccines 

  • There is a plausible biological pathway for understanding why a pairing of the two vaccines  may be better than just two doses of the AstraZeneca vaccine 

  • Why this is of interest is a little bit complicated, but is important to understand as it underpins why mixing these 2 vaccines may be useful!
  • Here is a picture of the AstraZeneca vaccine (or indeed the Sputnik vaccine)
  • It shows the two parts: the carrying virus and the bit to make the spike protein
  • When injected as a vaccine, the body produces antibodies against both parts of the vaccine as in the picture on the right
  • The antibodies against the spike protein (blue arrow) are what we need to fight the virus when it attacks
  • The antibodies against the carrying virus (red arrow) probably don’t matter
    • BUT these antibodies could theoretically attack the carrier virus in the second dose of the AstraZeneca vaccine
    • AND reduce its overall effectiveness
  • With the Sputnik vaccine, using a  different carrier at each dose reduces this risk
  • This is a theoretical advantage of the Sputnik over the AstraZeneca vaccine
  • And therefore this is also a theoretical advantage for combining the AstraZeneca and the Sputnik vaccines 

Challenges of doing ‘mixing’ studies

  • It is not that simple to know how best to study the benefits of mixing compared to sticking with one vaccine 
  • Given what we know already about these vaccines, it would be almost impossible to prove that a mixing regimes (mixed) was better than 2 doses of the same vaccine (same)
  • This is because all the individual vaccines:
    • Lead to a very large  reduction in the number of cases.  
    • The numbers that therefore in any trials that would need to be studied to show that ‘mixed’ was better than ‘same’ would be enormous!
  • My guess is that studies are more likely just to analyse the laboratory results of these different regimes, in the hope that if the lab outcomes were the same then the clinical benefits should follow – though this is not necessarily true

What about new variants of the virus?

  • All these vaccines were designed to work against the initial variants which may be an issue
  • However, the AstraZeneca but especially the Pfizer and Moderna vaccines can be very easily modified to work better against the new variants, so for some people their second dose (booster dose) could be in theory ‘a different vaccine’.
  • More importantly, if one vaccine is easier to produce in a new-variant-proof version, then this could well be the vaccine of choice for the second dose, independent of the vaccine in the first dose


  • This is a unique situation in public health; for most viral vaccines we are not faced with this choice or this dilemma
  • Thus far, the vaccines licensed for use all work extremely well in leading to enormous drops in the number of people becoming seriously ill, so there is not an immediate powerful case for mixing 
  • There is a scientific logic though for looking at mixing vaccines as being beneficial
  • It is indeed possible that new variants will mean that booster doses (second or even third!) might well come from a different vaccine
  • At the very least, studying  mixing of vaccines is sensible, if only to prove their success rates are just as good

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:

7 replies on “Mixing vaccines: taking two doses of different vaccines seriously!”

My question relates to mixing 2 vaccines when you have already had 2 doses of one. I have had 2 doses of Novavax, and I was offered the AZ this week. I turned it down bacouse by then I knew I had had the Novavax. Should I have thought about accepting it, and having one dose of the AZ?


Given there are now a number of licensed vaccines with good efficacy, is there not an ethical problem of running trials for any new vaccine (or combination of vaccines) with a placebo group? Without a placebo group, how do you test the efficacy of a new approach?


I agree Ken. I was asked not to unblind, but to take part in a double blind trial, starting at the end of March. I said a categoric no, partly because I was almost certain I had had the vaccine, but also because it might mean a 3 month wait if I hadn’t. Why would I do that?


Don’t think anyone is suggesting a placebo comparison trial as need to compare with an established regimen. One problem at the moment is we can only rank the different vaccines from their results from different trials with a comparison group. There is a statistical methodology which does allow to compare between (say) two active treatments from their separate placebo controlled trials, but it is not without complexity,


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s