1 august 2021 Episode 158 CoronaVac heterologous prime boost and herd immunity

Sun, 08/01/2021 - 20:37

Episode 158 : Viral biology, vaccine efficacy, heterologous prime boost and herd immunity


Dear colleagues,

In this episode, we have several themes:

  • Starting  with some didactic papers;
  • Put the latest news on the forerunner inactivated vaccine CoronaVac in context;
  • Present data on the remarkable effectiveness of mRNA, also against delta;
  • First results on heterologous prime-boost studies, mainly meant to replace the second Astra-Zeneca by Pfizer;
  • Considerations on herd immunity and the need to vaccinate children.    

Papers see:

Ep 158-1: Nice short review in Nature on how to envisage the biology of SARS-CoV-2. It explains how SARS-CoV-2 enters cells, inhibits cellular RNA and interferon, transforms the endoplasmic reticulum, induces syncytia, is activated by furin.  All of these “details” hint for future treatment strategies. An explanation is given why chloroquine may inhibit the virus in some cells in vitro, but fails in vivo (minimal role of “endosomal pathway” in vivo).

Ep 158-2:  An opinion paper in Scientific American with as a central message:

Until now, based on the available data from the vaccine trials, much of the publichealth messaging around vaccination has focused on the individual benefit. And the efficacy of the COVID vaccines in protecting against illness is indeed remarkable.

But with mounting evidence that confirms effectiveness in reducing transmission, it is time to begin emphasizing the societal benefit—and the personal responsibility to avoid harming others.


Most of the references in this paper have been quoted in previous episodes, but it is good to have this “popularizing” paper at hand to convince your hesitant patients, especially those with frequent social contacts. 


Ep 158-3: In this short comment, evidence is provided for Pfizer (Ep 158-3 a) and Astra-Zeneca (Ep 158-3 b) that breakthrough infections are associated with relatively low S-specific IgG and neutralizing Ab levels.  While the difference between cases and (non-BTI) controls is evident, there is, unfortunately, no clear protective “threshold”.  


These studies do suggest that Ab levels, induced by vaccines, based on already validated platforms (e.g. next generation mRNA or Adeno against new variants) could be used as surrogate markers of protection, since it will be impossible to perform large scale placebo-controlled trials.  Adding a panel of convalescent plasma to compare vaccine-indiced Ab with infection-induced Ab is also good practice.

However, at this moment, trials and real world studies by different research groups often use different IgG ELISA’s and especially neutralization assays with different sensitivities.  Therefore, standardization is urgently needed, including panels of convalescent plasma after infection with different variants, panels of live (variant) viruses, SOPs for neutralization assays etc.  A task for WHO, CDC…?      


Inactivated CoronaVac


As you remember, this classical beta-propiolactone inactivated and AlOH formulated vaccine, produced by the Chinese company Sinovac and approved by WHO is the most widely distributed vaccine of this kind.   There is still little formally published clinical data however.  Let’s first reiterate what I mentioned in previous episodes (papers added again).


Ep 151-4: Preprint by Palacios in Brazil on Coronavac from the Chinese Sinovac Co.: it showed 50.7 % efficacy against symptomatic COVID.  All 6 severe cases were in the control arm.   Notes:

  • This study was on about 10,000 people, included mainly HCW,  also 5 % over 60 and 55 % subjects with underlying disease:
  • Vaccine Efficacy (VE) was the same (51 %) for individuals below and above 60 years.  It was lower (39 %) for those with cardiovascular disease, similar (49 %) for those with diabetes and higher (75 %) for obese subjects!  
  • Titers of neutralizing antibodies against the Brazilian variants (B.1.1.28, P1 and P2) were comparable, but fewer subjects in the older age group actually seroconverted!

Ep 156-7b: Preprint by Ranzana and Hitchings on real world data 17 Jan-29 April 2021 Sao Paolo: > 43,000 adults, > 70 yrs

  • VE > 14 days after 2nd dose: 42 % symptomatic COVID; 49 % hospitalization and 71 % death
  • Clear decline in VE after 2nd dose with age:
    • Against COVID:                70-74 yrs: 62% VE; 75-79: 50 %; > 80: 28 %
    • Against hospitalization: 70-74 yrs: 80 % VE; 75-79: 69 %; > 80: 43 %
    • Against  death:                                 70-74 yrs: 86 % VE; 75-79: 87 %; > 80: 50 %


The new data


Ep 158-4: Alejandro Jara in NEJM large scale real world data on vaccine roll-out in over 1 million people in Chile between 2 Feb and 1 May 2021:

  • VE overall > 14 d after 2nd dose: 66 % symptom COVID; 87.5 % hospitalization; 86.3 % death.
  • VE in 60+:  66.6 % sump COVID; 85.3 % hospitalization; 86.5 % death.
  • Table 1 shows that the > 60 yrs old vaccinated Chileans consisted of 1.1 million of 60-69; 742,000 of 70-79 and 348,000 of 80 +.  There is no analysis as to whether the responses decline with advanced age.


Ep 158-5: A group from Thailand in medRxiv shows that CoronaVac induces similar S-specific Ab titers as are present in sera from convalescent COVID patients, while neutralizing titers against wild-type virus are clearly lower, but still clearly positive in CoronaVac vaccinees versus convalescent sera.   However, neut titers against alpha, beta and especially delta variants are strongly reduced.


Ep 158-6: A preprint by HongXing Pan on a phase 2 study to evaluate 3 doses of either 3 or 6 µg Coronavac either all three on an interval of a few weeks or with a delayed 3rd dose (after about 6 months).  It is evident that most vaccinated people lose their antibodies after a few months (even after 3 doses), whereas the delayed third dose acts as a strong booster. 



  • Real world data in Brazil and Chile indicate a good vaccine efficacy of CoronaVac.  In brazil, there is a clear decline with advanced age, which is not clear in Chile.
  • These studies were done between Jan and May, hence while the gamma variant was most dominant, but before the advent of Delta.
  • In vitro data suggest strongly reduced neutralizing Ab against alpha, beta and especially delta (gamma not testes)
  • Neut Ab induced by CoronaVac wane over a few months and a booster is needed.   


The remarkable effectiveness of mRNA vaccines


(For sake of clarity, I limit the discussion to full vaccination, unless indicated otherwise)


Ep 158-7: A very thorough case-control study in Ontario, nicely summarized on p. 35

  • According to vaccine type:
    • Against symptomatic: Pfizer 91%, Moderna 94 %
    • Against severe both 96 %
  • No clear difference according to age (> 70), if fully vaccinated, but lower efficacy of incomplete vaccination in older subjects.
  • Co-morbidities: sustained efficacy.
  • Alpha, beta, gamma VOC: sustained efficacy.


Ep 158-8: Refined retrospective study in Qatar, looking at effect of previous infection:

  • Without prior infection: Moderna slightly more effective: only 1.83 infect/10,000 versus 11.02/10,000 for Pfizer
  • With previous infection: similar efficacy: 1.66 Moderna vs 1.55 Pfizer

→  Will booster be more indicated in Pfizer?


Ep 158-9: Bernal paper in NEJM on effectiveness against VOC after full vaccination: “only modest differences between Astra-Zeneca ChADOx1 and Pfizer”.  Judge for yourself:

  • Against alpha: Pfizer 93.7% (95% CI, 91.6 to 95.3) vs A-Z: 74.5% (95% CI, 68.4 to 79.4);
  • Against delta:  Pfizer 88.0% (95% CI, 85.3 to 90.1) vs A-Z 67.0% (95% CI, 61.3 to 71.8)  


These data refer to “symptomatic infection”, while …. the numbers of cases and follow-up periods are currently insufficient for the estimation of vaccine effectiveness against severe disease, including hospitalization and death…. (p. 8).


Ep 158-10:  a recent comment in Science, quoting several studies that show a very high protection against any infection by both mRNA vaccines (in the 90-100 % range), including studies with alpha and beta VOC (but not yet delta) and with a hint in at least one study that Moderna potentially slightly outperforms Pfizer.


Conclusion: Data accumulate that mRNA vaccines are superior to Adenoviral and inactivated vaccines with regard to protection against any infection and non-severe COVID, also in case of VOC, including delta.  Nevertheless, protection against severe disease is also very strong by Adenoviral Astra-Zeneca, but clearly weaker by inactivated CoronaVac.  


Heterologous prime-boost

Ep 158-11: The Oxford Com-COV phase ½ study: a systematic comparison of  homologous and heterologous combinations of AZ and Pfizer.


The most remarkable first result is a comparison of the geometric mean titer of Spike specific IgG ELISA at 1 month post boost:

Pfizer homologous: 14,080;  AZ homologous: 1,392

Pfizer + AZ hetero: 7,133; AZ+ Pfizer hetero: 12,906


T cell responses ELISPOT were also measured, with less clear differences:

Pfizer homol: 80; AZ homol: 50; Pfizer + AZ: 90; AZ + Pfizer: 185;


The authors conclude “non-inferiority”, but it is rather clear that AZ homologous is weak (both Ab and T cells), while either twice Pfizer or AZ + Pfizer induce high titers of Ab but the AZ + Pfizer induces the highest T cell responses. 


This was to be expected, based on many other experimental vaccine schemes with combinations of viral vector and nucleic acid vaccines.    


Ep 158-12: A similar phase 2 study in Germany concludes that: heterologous regimen:

  • Induced Spike-specific IgG, neutralizing antibodies, and spike-specific CD4 T-cells, which were significantly more pronounced than after homologous vector boost, and higher or comparable in magnitude to the homologous mRNA regimens.
  • Moreover also spike-specific CD8 T-cell levels after heterologous vaccination were significantly higher than after both homologous regimens.

See Fig 1 p. 19


Ep 158-13: A smaller phase 2 study, comparing AZ-Pfizer with homologous Pfizer reaches similar conclusions.  Moreover, they show also that the heterologous regimen induces slightly higher neutralizing antibodies against alpha, gamma and kappa VOC  (Fig 2F p. 8). (Unfortunately delta was not tested)


Ep 158-14: Real world data from Germany, showing that vaccinees who received the heterologous AZ-Pfizer had clearly higher neut Ab than either homol. Pfizer and much higher than homol AZ. (Measured with an ELISA format against WT virus).


Ep 158-15: Large population-based data from Denmark, showing that:

  • The heterologous AZ + Pfizer has clearly higher efficacy than AZ alone (not surprising)
  • The VE of AZ + Pfizer against any (symptomatic) infection = 88 %
  • No hospitalization or deaths in either the AZ only or the AZ + Pfizer group.



  • The phase 2 studies clearly show that AZ + Pfizer is much better in inducing immune responses than twice AZ and probably better than twice Pfizer.
  • Real world data on whether this regimen is really superior to the homologous regimens especially against delta variant are still awaited.


Herd immunity:

Ep 158-16: While only a part of the Brits is vaccinated, massive infections after lifting all restrictions is NOT a logical and ethical option reach “natural” herd immunity in UK, because:

  1. It will affect the younger (largely unvaccinated) population and expose them to complications and long-term effects of COVID
  2. High rates of transmission will disrupt school reopening.
  3. It will create fertile ground for vaccine-resistant variants to emerge.
  4. It will put additional stress on already exhausted health services.
  5. This strategy will disproportionally affect the most vulnerable minorities.


Ep 158-17: A complicated modeling exercise in France, showing that the delta variant really is a “ominous game changer” to reach herd immunity:


  • With the original less infectious strain, the present vaccination strategy would have sufficed to prevent a new wave.
  • With the delta variant, the current vaccination combined with unrestricted social interactions will lead to a fourth wave  


To reduce the hospitalization impact of the new wave there are several possible interventions:

1) Social restrictions;

2) Increase of the vaccination rate in age group 70-79, which is most affected by the new wave despite of the already large vaccination status in this group;

3) Increase of the vaccination rate in children and adolescent (1-19 yers), who are responsible for most of the spreading.  


Ep 158-18: Modeling in Shanghai: Herd immunity for new variants, such as Alpha or Delta, 12 can only be achieved with more efficacious vaccines and coverages above 80-90%, implying that the children and adolescents (3-17 years) should be included in vaccination campaigns.




All three papers clearly indicate that children should be vaccinated to reach herd immunity against the very infectious variants such as delta.