Part 2

Part 1 of this article addressed the issue of partially inoculating a broader population due to limitations in the vaccine supply.

Part 2 considers the question: will partial immunization promote the emergence of variant SARS-CoV-2 strains?

The decision to partially vaccinate the Canadian population has raised concerns about the emergence of resistant variant strains of SARS-CoV-2. Dr. Caroline Quach, chair of the National Advisory Committee on Immunization (NACI), has stated that there is a risk of variant strains with partial vaccination, but NACI considers this to be unlikely since individuals will have sufficient immunity within a few weeks of their first vaccination. However, Dr. Quach added, “we could be wrong.” (Serebrin J. Canadian Press, 2 March 2021. www.thestar.com/news/canada/2021/02/03/fear-that-delayed-second-covid-19-vaccine-dose-could-lead-to-new-variants-realistic.html).

How these variants have emerged has not been determined. The pandemic strain (D614G) is a variant that supplanted the Wuhan strain and soon became dominant due to enhanced infectivity of the upper respiratory tract (Plante et al. Nature 2021;592:116-121). Countless variants have since arisen due to the global proliferation of the virus, but most do not confer an evolutionary advantage. As for the current variants of concern (B.1.1.7, B.1.351, P.1), the New and Emerging Respiratory Virus Threats Advisory Group (NERVTAG) in the UK has speculated that they may have emerged due to exposure to current therapies, or to viral persistence in immunocompromised individuals in whom partial immunity may allow for prolonged viral shedding (www.gov.uk/government/publications/sars-cov-2-immunity-escape-variants-7-january-2021).

Resistance to COVID-19 therapies is akin to antibiotic resistance, in that resistant strains exposed to an anti-infective environment emerge due to selective pressure. This has already been seen with some treatments for COVID-19. A single point mutation is sufficient to produce resistance (e.g. E484K) to monoclonal antibodies used to treat COVID-19 infection (Greaney et al. Cell Host Microbe 2020;S1931312820306247). Distribution of bamlanivimab, approved in the U.S. in November, was halted as monotherapy only four months later due to the increase in resistant variants. Several variants have demonstrated bamlanivimab resistance, including the South Africa (B.1.351), Brazil (P.1), California (B.1.427 and B.1.429) and New York (B.1.256) variants, which harbour E484K and/or L452R substitutions. Resistance following administration of convalescent serum has also been documented (Kemp et al. http://medrxiv.org/lookup/doi/10.1101/2020.12.05.20241927).

Vaccine resistance differs from treatment resistance in two respects. Prophylactic vaccines are administered to an uninfected host so selective pressure does not apply. Secondly, if a vaccinated individual subsequently becomes infected with a variant strain, the vaccination-induced immune response would likely be sufficiently broad to defeat the resistance mechanisms that the variant had evolved (Kennedy & Read. PNAS USA 2018; 115:12878-12886).

As a result, vaccine resistance is exceedingly rare. One of the few examples is an early formulation of the hepatitis B vaccine, which used a limited antigen (hepatitis B surface antigen, HBsAg). Cases of HBsAg-negative and escape variants have been reported but appeared to be due to vaccination in previously-infected individuals (Carman et al. Lancet 1990;336:325-329). However, this cautions against a narrow range of epitopes in vaccine development.

NERVTAG has stated that variants are unlikely to proliferate to the same extent in a fully vaccinated individual or be transmitted to the population at large. But the group acknowledges that delaying the second dose would be expected to increase somewhat the probability of vaccine resistance emerging but could not estimate what that risk might be.

Partial immunity from one vaccine dose may be sufficient to prevent severe COVID outcomes in most people (see NeuroSens, COVID vaccination: is one dose better than nothing? April 13, 2021). While the immune response to one dose is not robust, it is likely within the lower range of immunity seen with natural infection. NERVTAG adjudged that a single dose of vaccine does not generate a new risk with respect to variants.

It should be noted that NERVTAG was considering a scenario in which the AstraZeneca vaccine was delayed >12 weeks, slightly beyond the recommended 4-12 weeks between doses. The consequences of arbitrarily extending the mRNA vaccine dosing interval to 16 weeks are unknown. It may be that a prolonged delay will reduce the booster effect against the dominant strain or lower the already reduced effectiveness against variant strains. More worryingly, partial immunity may allow for more prolonged viral shedding and another wave of transmission. To this might be added the concern of what might be termed global partial immunity, since most of the world lacks access to the current vaccine supply.

NERVTAG concluded that immune escape variants will emerge, but this will most likely be driven by increasing population immunity. This is akin to the variant strains of influenza that regularly emerge to exploit the niche that is created when a population is vaccinated against the dominant strain. It remains to be determined if this process will be accelerated by partial immunization. The short-term gain of partial immunity may well increase the likelihood over the longer term of the pandemic becoming endemic.