22 July 2021 Addendum to Episode 155 on Delta and vaccination

Thu, 07/22/2021 - 22:02

Dear colleagues,

I received some reactions and questions on the relation between the delta variant and vaccination.  Therefore, I feel it is appropriate to write an “addendum” with some additional info.  Some of it is straightforward and reassuring, but there is also contradiction on J&J as you will see.

Rather reassuring population studies in US and UK   

Ad Ep 155-1:  In the period between Dec 15 and April 30, hospital admission was provided to almost 12,000 COVID patients in the Michigan area.   The need for hospitalization or emergency care was 96% lower in fully vaccinated versus unvaccinated individuals. Nevertheless, in cases of breakthrough COVID-19 requiring hospital-based treatment, elderly patients with significant

comorbidities remain at high risk for severe outcomes regardless of vaccination status.

Clearly, this study precedes the delta wave.

Ad Ep 155-2: Follow-up REACT (REal-time Assessment of Community Transmision-1) in England: during the period from 20 May to 7 June, there was an exponential growth of infections, with a rapid replacement of the alpha variant by the delta variant. Nevertheless, the link between infection and hospitalization and death weakened. This remained the case for person above 65 (who were largely vaccinated), while in the younger (much less vaccinated) groups, hospitalizations were rising again. These findings are in keeping with the high efficacy of the vaccines (ChAdOx1-S and BNT162b2) in preventing hospitalisations and deaths among the fully vaccinated population.


Controversy about the J&J vaccine versus delta variant?


Two recent papers in the NYT provide seemingly contradictory messages about the activity of J&J against delta, based on in vitro data.  Therefore, I will analyze the publications referred to.


Ad Ep 155-3: is a preprint by scientists from J&J, showing at first view very decent neutralizing titers (usually > 1/100) 70 days after full vaccination against variants, including delta.  Nevertheless some vaccinated individuals had very weak titers, especially against beta and gamma (See Fig 3 p. 9).  The limitation is that only J&J vaccinated people are considered here.     


Ad Ep 155-4: is the preprint by Tada, I already discussed yesterday (as Ep 155-12).  It shows that the J&J vaccine induces much less neutralizing antibodies than the mRNA vaccines against several variants (especially beta, delta, delta plus and lambda). See Fig 1 p. 26


Both papers use a comparable setting: a similar neutralization test (based on pseudotyped lentiviruses) and the sera were taken between 2-3 months after fully vaccination.  Nevertheless the neutralization titers of the J&J vaccine are clearly, but not dramatically lower in the Tada paper and the difference with mRNA vaccinees is striking. 


So, the jury is out, since at present, I could not find strong data comparing breakthrough infections after mRNA versus J&J…..


What about the efficacy of other Adenovirus-based vaccine: Astra-Zeneca/Covishield?  


In Episode 155, I presented already data about breakthrough infections:

  • In UK, Astra-Zeneca has a protective efficacy of 60 % against delta infections, while Pfizer reaches 88 % (Ep 155-4).
  • In India, breakthrough with delta in fully Covishield vaccinated health care workers was 8.6 %, corresponding to 60 % efficacy, seen in UK  (Ep 155-6)
  • Another Indian study showed higher BTI (18.6 %) in fully Covishield vaccinated HCW, with 10/86 hospitalized, 2 in ICU and 1 death (Ep 155-7).  
  • According to an unpublished report of Public Health England (provided by a colleague): “the Pfizer/BioNTech jab was linked to a 94% vaccine effectiveness against hospital admission with the Delta variant after one dose and 96% after two doses, while the figures for the Oxford/AstraZeneca jab were71% and 92% respectively. “

So while Pfizer protects clearly better against delta infection, a full Astra-Zeneca (or Covishield?)  vaccination seems still very protective against hospitalization, according to UK data.  However, the second Indian study in HCW (Ep 155-7) is less reassuring…..


Ad Ep 155-5:   a very extensive description of the epidemiology, biology and immunology of delta in India, with a lot of supporting data:


  • Fig 1 f shows that convalescent sera from first UK wave have a strongly reduced neutralizing capacity against delta (5.7 X), but even more against beta (8.2 X);
  • Fig 2 shows a similar relative reduction of neut capacity  against delta vs wild type by  Covishield and Pfizer vaccinated sera, but neut titers from Pfizer vaccinees against delta are still 10 X  higher than those of Covishield vaccinees….  Moreover, neutralization capacity of several monoclonal antibodies directed against the receptor binding site and the N-terminal domain is decreased.  Importantly, the in vitro data suggest that the therapeutically used Regeneron antibody cocktail may be less efficacious against delta.   
  • Fig 3 and 4 show that the delta (and kappa) variant have increased infectivity, fusogenic capacity and syncytium- inducing activity in various relevant cell and organoid systems.
  • Fig 5 illustrates tens of breakthrough infections of fully Covishield vaccinated health care workers in several large Delhi hospitals at the time that delta was replacing other variants and sequencing showed that the odds of BTI by delta as compared to other variants was 4.5 times higher in vaccinated versus non-vaccinated subjects, confirming the higher fitness and escape from neut by the delta variant.


What about mRNA vaccines?


As already suggested in the previous paper, the delta variant also escapes partly from Pfizer vaccine, but clearly to a lesser extent than for Adstra-Zeneca/Covishield:


See Ad Ep 155-6 and Ad Ep 155-7: Remarkably:


  • The in vitro reduction of neutralization was more pronounced against viruses with E484K and N501Y/T (e.g., B.1.351 and P.1), followed by lineages with L452R (e.g., B.1.617.2) or with E484K (without N501Y/T).
  • Plasma from people who were infected and subsequently vaccinated retained better neutralizing capacity than those who were uninfected and vaccinated.



  • No doubt that the delta variant reduces vaccine efficacy in vitro, but this was equally (or even more) true for the beta and gamma variant.  The latter variants have probably a lower in vivo fitness (infectivity and replication capacity), explaining why delta seems more aggressive?
  • No doubt that mRNA vaccines remain superior against clinical delta disease, but Astra-Zeneca/Covishield is also protective, while the clinical data for J&J are not yet clear.