Study design issues
Since vaccine-elicited anti-SARS-CoV-2 spike antibodies may themselves be harmful, current vaccine trial endpoints may be incorrectly chosen. The first phase of human testing of current COVID-19 vaccines are designed to be validated, at least partly, on immunogenicity, which is defined as elicitation of high titers of anti-pathogen antibodies by the vaccine. Immunogenicity measured in this way usually ignores the function and efficacy of vaccine-elicited antibodies, and is a low bar for vaccine candidates, which only remains in contention due to distant historical efforts, such as attempts to develop vaccines for encapsulated pathogens like Neisseria and Pneumococcus . Thus, in the case of COVID-19 disease, vaccine-elicited anti-SARS-CoV-2 may be detrimental or, at least, of uncertain benefit, making their appearance and titer a puzzling choice of milestone for vaccine advancement. The technological improvements that have been leveraged to produce the current SARS-CoV-2 vaccine candidates have nearly all been platform-based: e.g. lower cost and faster manufacturing of DNA or RNA vaccines, viral vectors, skin inoculation, etc. Accordingly, the question in the current vaccine trial of SARS-CoV-2 spike mRNA of whether the vaccine is immunogenic is a question entirely about the technology platform and not a predictor of efficacy of the vaccine. As for safety, public data on ongoing COVID-19 vaccine trials do not disclose whether antibody-mediated immunopathology is defined as an Adverse Event of Special Interest, and indeed several of the trials, and nearly all the trials outside the US, restrict Adverse Event monitoring to 28 days, which would be too soon to observe worsened COVID-19 disease due to exposure to circulating SARS-CoV-2 viruses (Source: clinicaltrials.gov). Accordingly, some of the current viral candidates will almost certainly pass a Phase I immunogenicity and safety test, even if they have significant delayed ADE or immunopathology liabilities. Similarly, Phase III trials may be incorrectly designed if their primary endpoint is reduction in infections rather than reduction in immunopathologic disease. If vaccine-worsened disease is indeed a relevant phenomenon for COVID-19 vaccines, one could easily imagine a scenario in which infections are reduced but immunopathologic disease incidence is not or is actually higher in vaccinees. Approval of such a vaccine would not be in the public’s best interest.
Current data on COVID-19 vaccines show no evidence of ADE of disease. Non-human primate studies of Moderna’s mRNA-1273 vaccine showed excellent protection with no detectable immunopathology11. Phase 1 trials of several vaccines have not reported any immunopathology in subjects administered the candidate vaccines, however these subjects were unlikely to have encountered circulating virus yet12. Nevertheless, all preclinical studies to date have been performed with the Wuhan or closely related strains of the virus, while a mutant D614G virus is now the most prevalent circulating form, and several observations suggest that this alternative form may be antigenically distinct from the Wuhan derived strain, not so much in composition, but in conformation of the viral spike and exposure of neutralization epitopes13-15. Similarly, Phase 1 and 2 clinical trials of vaccine candidates have not been designed to capture exposure of subjects to circulating virus after vaccination, which is when ADE/immunopathology is designed to occur. Thus, the absence of ADE evidence in COVID-19 vaccine data so far does not absolve investigators from disclosing the risk of enhanced disease to vaccine trial participants, and it remains a realistic, obvious potential risk to the subjects.