Introduction
The elicitation of antibodies, preferably neutralizing antibodies, is
the goal of nearly every current SARS-CoV-2 vaccine candidate, but
emerging data suggests that severe COVID-19 disease is associated with
the development of anti-SARS-CoV-2 serum antibodies1and that subjects who recover quickly have low or no anti-SARS-CoV-2
serum antibodies2. A recent study also revealed
IgG-mediated acute lung injury in vivo in macaques infected with
SARS that correlated with a vaccine-elicited, neutralizing antibody
response3. Inflammation and tissue damage in the lung
in this animal model recapitulated the inflammation and tissue damage in
the lungs of SARS-infected patients who succumbed to the disease. The
time course was also similar, with the worst damage occurring in delayed
fashion in synchrony with ramping up of the immune response. Remarkably,
neutralizing antibodies controlled the virus in the animal, but then
would precipitate a severe, tissue-damaging, inflammatory response in
the lung. This is a similar profile to immune-complex mediated disease
seen with respiratory syncytial virus (RSV) vaccines in the past,
wherein vaccinees succumbed to fatal enhanced RSV disease due to the
formation of antibody-virus immune complexes that precipitated harmful,
inflammatory immune responses. It is also similar to the clinical course
of COVID-19 patients, who experience their most severe morbidity and
mortality via inflammation in lung (including eosinophilic pulmonary
infiltrates), blood (coagulation), cardiac and other tissues at a
delayed phase after symptomatic infection, when the antibody response is
emerging but viral loads may be declining or absent. This picture, which
is vertically consistent from controlled SARS studies in primates to
clinical observations in SARS and COVID-19, suggests the risk that
vaccine candidates composed of the SARS-CoV-2 viral spike and eliciting
anti-SARS-CoV-2 antibodies place vaccinees at higher risk for more
severe COVID-19 disease when they encounter circulating viruses.
Indeed, studies in mice of prior SARS vaccines revealed this exact
phenotype, with four human vaccine candidates eliciting neutralizing
antibodies and protecting against SARS challenge, but viral re-challenge
of thus vaccinated animals resulting in immunopathologic lung
disease4. Independently, SARS/MERS vaccine candidates,
commonly exhibited antibody-dependent enhancement (ADE) associated with
high inflammatory morbidity in preclinical models, obstructing their
advancement to the clinic5,6. SARS ADE of both disease
in non-human primates and viral infection of cells in vitro was
clearly mapped to specific antibody-targeted SARS viral spike
epitopes7. This phenomenon was consistent across a
variety of vaccine platforms, including DNA, vector primes and
virus-like particles (VLP), irrespective of inoculation method (oral,
intramuscular, subcutaneous, etc). An unknown variable is how long this
tissue damage lasts, possibly resulting in permanent morbidity (e.g.
diabetes from pancreatic damage8).
Importantly, this ADE of disease , or immunopathology, that
distinguishes the morbidity, and likely the case fatality rate, of SARS,
MERS and COVID-19 disease from comparators, like influenza disease, may
be pathophysiologically independent from ADE of viral infection .
Both are anti-viral-antibody dependent, and the latter can be measuredin vitro , but the former may only be assessed in vivo .
Successful COVID-19 vaccines must incorporate rational strategies to
avoid both ADE of viral infection of cells in vitroand formation of immune complexes/immunopathology in vivo .
All current, reported COVID-19 vaccine candidates incorporate intact
viral spike domains, and none are designed to avoid both of types
of ADE. Although some vaccines, such as Pfizer’s BNT162b1, contain only
the receptor binding domain (RBD) of the viral spike, prior studies with
the closely related SARS virus have established that neutralizing
antibodies targeting the RBD exhibit ADE9.
The justifications for the whole spike vaccine immunogens in the current
COVID-19 vaccines currently in clinical trials insinuate an equivalence
of this COVID-19 pandemic to past influenza pandemics. This can be
reassuring, as effective vaccines for the influenza and H1N1 viruses
were rapidly achieved using the traditional recipe for vaccine
development. However, autologous immunity following influenza infection
is the rule rather than the exception, which is consistent with the
success of influenza vaccines in combatting influenza pandemics, like
H1N1, where the time frame is too short to allow significant viral
antigenic variation. However, immunity following SARS-CoV-2 infection is
less well established. Antigenic variation is indeed not the vaccine
challenge for the SARS-CoV-2 pandemic, contrary to frequent allusions
expressed in both scientific and lay public reports: the 108 sequences
of the SARS-CoV-2 spike protein in GenBank (accession date Mar 27, 2020)
are 100% identical, and only a single amino acid mutation has been
reported worldwide
(http://virological.org/t/first-report-of-covid-19-in-scotland/412),
and it is not in an antigen location. Instead, the better viral
comparator for SARS-CoV-2 is the non-pandemic RSV, with a similarly
conformationally metastable, but not particularly antigenically
variable, viral spike that has resisted an efficacious vaccine for more
than 50 years and precipitated a coronavirus-similar, vaccine-enhanced
disease syndrome10.