19 April 2021 Indian variants and protection against (re)infection

Mon, 04/19/2021 - 21:52

Dear colleagues,

I was asked to comment on the “Indian variant” and its clinical consequences. Very relevant question, but very little has appeared in the scientific literature until now, even not the “preprint” journals.  So, I mention some press papers and try to connect with some related preprints….

As usual, I found some other information, related to clinical presentation and protection by infection and vaccination.  

As always, this episode will not answer all your and my questions, but it provides some further provisional and growing insights, keeping us interested and busy…..   

New variants in India

Ep 131-1-to-4 are journalists impressions on the Indian variants

→ Is this “variant of concern” driving the (huge) second wave in India ?

→ Is it also responsible for different clinical presentation: more headache, conjunctivitis, diarrhea and discolored fingers and toes?  

→“The variant is more infectious and can infect many people together because it spreads faster in a group or within a family. However, no need for hospitalisation has been seen so far in most cases(???).”

Ep 131-5 is a more scientific consideration by William Haseltine

Lineage analysis shows that B.1.617 is really different from the other VOC: All of the 15 amino acid changes of B.1.617 differ from those of all other variants of concern when compared to the globally dominant D614G strain designated B.1 with two notable exceptions:

  • L452R = also found in “Californian variant B.1.427/429.  It increases affinity for the receptor and reduces affinity for antibodies in convalescent plasma and the therapeutic monoclonal antibody LY-CoV555 (from the Eli Lilly cocktail).
  • E484Q well known from the South-African and Brazilian variant (there rather E484K) can also contribute to both higher affinity for ACE2 and immune escape.
  • There are two mutations in the N-terminal domain (could also contribute to escape?) and one (P681R) near the S1/S2 cleavage site (could contribute to increased infectivity).

For more details and very nice illustrations see the paper.


Ep 131-6 and -7: more info on L452R in the context of the Californian variant: the higher affinity for ACE-2 and relative decrease of neutralization by convalescent and vaccinated plasma. In fact more comparable with British variant (2-6 X decrease), clearly not as dramatic decrease as with South-African variant ( in most reports > 10 X decrease).


Ep 131-8  Chakraborty et al announce in Lancet a rapid CRISPR-based assay to identify known variants.  Clearly, this might be a much-needed tool to track the spread of VOC rapidly on a large scale, even in resource limited settings.


Ep 131-9   TV Padma in Nature News explains that India is lagging in vaccination: 111 million first doses were given as of 12 April, on a total population of almost 1,400 million.  The main “workhorses” are COVASHIELD (in fact the Astra-Zeneca vaccine) and COVAXIN (an inactivated vaccine).   As discussed earlier, the phase 3 studies of Covaxin have not yet been published and it is not clear yet how effective both of these vaccines will be against for instance the B.1.617 “Indian variant”….


Clinical insights


Ep 131-10,11,12 Journalists reporting on a possible tendency that both in India and Brazil the patients in the present COVID wave, may be younger and that also children are disproportionally affected.  Moreover, the clinical presentation seems more severe with more serious pulmonary, cardiac, gastro-intestinal and renal symptoms. Obvious (as yet unanswered) question: are those phenomena related to the variants (Influence of vaccination in older people less likely in India and Brazil as compared to UK, US, EU, where most elderly is vaccinated).


Ep 131-13 a very nice analysis on two “phenotypes” in COVID ARDS: a major (< 75 %) “classical type” and a minor (> 25 %) “coagulopathy type”.  Both were indistinguishable for the respiratory parameters. The “coagulopathy type” was associated with increased end-organ failure (mainly cardiac and renal) and doubling of short term mortality. I am wondering whether it would also be associated with more damage to central nervous system and more “long-COVID” in survivors….  


Level of protection after infection or vaccination


Ep 131-14 studies a cohort of over 25,000 HCW in UK between June 2020 and January 2021.  The incidence of new SARS-CoV-2 infection was 7.6 and 57.3 per 100,000 person years in the originally seropositive versus seronegative subjects.  Hence a 84 % overall protection by infection-induced antibodies over a 7 months period.


Ep 131-15 is a more detailed study in over 3000 young US marine recruits with a very high incidence.  In fact over a 6 weeks period 48 % of originally seronegative and 10 % of originally seropositive recruits tested SARS-CoV-2 positive, hence also over 80 % protection. As could be expected, (1) infection was more likely with lower baseline full-length spike protein IgG titres than in those with higher baseline titres and (2)  infected seropositive participants had viral loads that were about 10-times lower than those of infected seronegative participants.


Although at first view reassuring, two remarks:

  • Reinfection is rather frequent upon high exposure.
  • These results are in young people.  What about aged subjects?


Ep 131-16 Khoury et al try to find out what level of neutralizing antibodies protects against infection.  This is a modeling exercise, based on large cohorts of convalescent and vaccinated subjects.  The obvious problem is that the neutralization assays and many other parameters (e.g. when to perform  PCR, definition of mild and severe disease etc).  So, the results have to be taken “cum grano salis”


  1. Consistent with Ep 131-14 and 15 they suggest that approximately 20% of the average convalescent antibody neutralization level provides 50% protection against infection and that 3% of mean convalescent neutralization levels suffices to protect 50 % against severe disease.  In other words to get reinfected resp. seriously ill after reinfection, your level of neutralizing antibodies should be low or very low….
  2. The decay of neutralizing antibodies after vaccination or infection is similar.  Implying that we will see increasing numbers of infections several months after vaccination.
  3. Fig 1 p. 15 provides an interesting comparison between (A) Neutralizing Ab, induced by different vaccines and “natural” (convalescent) Ab.  Not surprisingly, the mRNA as well as Novavax S protein score best.  The Sputnik would be just slightly better than convalescence.  The inactivated Covaxin and Coronavac as well as the “Western” Adeno’s (Astra-Zeneca and J§J) score less than convalescent plasma….. This order is also consistent for efficacy (panel C).     


That’s it for now….

Best wishes,