Following the science

Please, follow the science (even if it has now become inconvenient and uncomfortable)

Today, I was made aware of the following publication:

https://www.medrxiv.org/content/10.1101/2021.07.01.21259833v1.full.pdf

This is one of the most abominable publications I’ve ever read. It’s simply unthinkable that this would pass peer review. I am truly sorry to say so, but this article is really an unprecedented violation and misrepresentation of the science. I have no choice but to react to this and copy the corresponding author (venky@nreference.net). Clearly, these guys have a mission that is all but inspired by science. I am truly wondering who is financing this type of fancy ‘machine-taught science’.

The corresponding author is the ‘rchief evangelist for nference’ and ‘has led multi-year R&D data-science alliances with global biopharma companies… ‘. This may already explain why pedantic messages are conveyed, which, unfortunately, are not supported by any data in this paper (‘This study demonstrates that mass vaccination may serve as an antigenic impediment to the evolution of fitter and more transmissive SARS-CoV-2 variants, emphasizing the urgent need to stem vaccine hesitancy as a key step to mitigate the global burden of COVID-19’) and why the title is misrepresenting the impact of mass vaccination.

Instead of ‘COVID-19 vaccines dampen genomic diversity of SARS-CoV-2: Unvaccinated patients exhibit more antigenic mutational variance’, the title should read: ‘COVID-19 vaccines dampen genomic diversity of SARS-CoV-2: Vaccinated patients exhibit diminished antigenic mutational variance’! This at least would reflect the increasing trend of convergence of immune escape mutations to a limited set of Sars-CoV-2 S protein-associated epitopes. This trend coincides, indeed, with mass vaccination and has extensively been documented by a number of molecular epidemiologists (https://pubmed.ncbi.nlm.nih.gov/33688681/) and is now considered a serious ‘Risk of rapid evolutionary escape from biomedical interventions targeting SARS-CoV-2 spike protein’ (https://pubmed.ncbi.nlm.nih.gov/33909660/).

In all of the section reporting on the results of this study, there is literally only one sentence that deals with the comparison of variants in vaccinates versus non-vaccinated patients (see under results: Whole-genome sequencing of SARS-CoV-2 genomes from unvaccinated and vaccinated individuals reveals different mutational profiles):

‘We find that the known B-cell epitopes exhibit more mutational diversity in the unvaccinated individuals than in the vaccinated individuals (Figure 4b,c).’

None of this is even further debated under Discussion! So what is justifying their title and gospel message?

My comments: OMG, really? Is it this what prompts them to preach the gospel of mass vaccination? Do they have any notable background in immunology or vaccinology? They seem to seek whatever data they can get their hands on to confirm the mantra of mass vaccination evangelized by the WHO: ‘The more viral replication, the more mutations and hence, the more mass vaccination will prevent new variants from arising’. Don’t they read what their peers are publishing (see reference above) and what I’ve been reiterating in my most recent contributions? (“Critical opinion article: Why is the ongoing mass vaccination experiment driving a rapid evolutionary response of SARS-CoV-2?” and “The chicken-and-egg problem”)? Don’t they understand that you cannot do mass vaccination with these C-19 vaccines without having the vaccinees exerting major S-directed immune pressure on the virus and without providing a steadily growing breeding ground for the selected immune escape variants? Don’t they understand that precisely this widespread immune selection pressure is now promoting convergent evolution of immune escape variants to the same antigenic sites within the S protein and thereby enabling the virus to adapt to its altered host environment? That this is eventually going to lead to dominance of variants of concern? That those are predominantly shed by vaccinees? (almost according to their own findings: ‘Furthermore, a larger fraction of the vaccinated (82.6%) cohort had alpha variant compared to the unvaccinated cohort (60%) (Table S1). Availability of a larger number of sequenced genomes in the future can help understand whether specific variants of concern are more likely to cause breakthrough infections’).

Of course, the spectrum of mutational diversity in non-vaccinated people will be higher as these individuals do not exert selective immune pressure on S protein, even if naturally infected. Moreover, non-vaccinated populations do not provide a suitable environment for immune escape variants to propagate as there is no competitive advantage for them to gain.

I am stunned at the immunological and epidemiologic illiteracy of the authors. If you really want to study the effect of mass vaccination on the evolution of mutations, you compare HEALTHY vaccinees with HEALTHY non-vaccinated people! Why? because the selective immune pressure (towards S protein) is highest in healthy vaccinees and (presumably) lowest in healthy non-vaccinated subjects (as they clear the infection rapidly). So, in terms of the degree of S-targeted immune selection pressure one expects the following gradation: unvaccinated <breakthrough vaccinated < healthy vaccinated or, in terms of (irrelevant!) mutational diversity: unvaccinated >breakthrough vaccinated > healthy vaccinated.

Mass vaccination at the height of a pandemic enables selection and adaptation of S-directed immune escape variants and leads to dominant circulation of variants of concern (VoCs), not the other way around as the authors try to make readers believe! Indeed, they erroneously pretend: ‘Dominance globally results in increased mutation of antigenic sites’! Conclusions as drawn by the authors reflect, indeed, the risk of bias and scientific misinterpretations when the analysis of complex phenomena such as the evolutionary dynamics of a pandemic is reduced to an exercise in computational stamp collection. Teaching machines how to read biological data can be extremely dangerous as lack of holistic insights lead to correlations that some tend to interpret as causation. For example, the current study results do not seem to take into account the impact of a number of tricky confounders, probably due to lack of knowledge in other biological fields (in this case: immunology, virology, evolutionary biology, vaccinology). For example: What about the impact of differences in the infectious pressure (number of active infections) in those different countries? What about the influence of the infectiousness of the circulating variants? Differences in the level of immune pressure? (e.g., which percentage of vaccinees only received a single injection of a 2-shot vaccine)? Differences in the total vaccine coverage rate? Differences in the speed of the mass vaccination campaigns? Differences in the type of vaccines used? Differences in flanking infection prevention measures?

Also their reasoning in regard of co-morbidities seems confusing. Although they did not find a statistically significant association between mutation count and rates of complications, they seem to suggest that co-morbidities resulted from Covid-19 rather than constituting a predisposing factor to Covid-19 disease. In this regard, the fact that vaccinated patients had lower rates of comorbidities would even suggest that vaccination was rendering them more susceptible to the disease! As underlying co-morbidities augment the risk for C19 disease, it is logical, indeed, that the majority of their unvaccinated subjects only contracted C19 disease provided they had an underlying health issue.

It’s amazing that in the key limitations of their study they didn’t even think about mentioning their full ignorance of the immunologic context that is deeply and massively altered by mass vaccination campaigns!

Their interpretation of how T cell responses may ‘buffer’ viral evolution is of course only hypothetical but nevertheless completely irrational. The large diversity of T cell peptides combined with the extensive heterogeneity of peptide-matching MHC class I alleles already explains why it is highly unlikely for a naturally infected population to place widespread selective immune pressure on a particular T cell epitope. I would not understand how degeneracy of T cell recognition would need to be involved in protecting against immune escape, especially not since degenerated T cell recognition has rather been associated with auto-immune disease (which is almost the opposite of immune escape!).

Last, their suggestion that T cell-based vaccines may offer a perspective to successful immune intervention is not based on sound scientific grounds either! Or do they think there is on or more promiscuous or universal MHC class I-dependent T cell epitope(s) that has the capacity to induce T cells capable of eliminating virus-infected cells? If this approach would work, we would already have several T cell-based vaccines on the market. It’s not the case as T cell antigens are only protective in individuals who are genetically predisposed to controlling the pathogen by virtue of their protective alleles.