A Front Row Seat on Evolution

In a recent perspective article in Science, researchers at the University of Pittsburgh School of Medicine discussed the recent evolution of SARS-CoV-2 (Science 2021;371:1306-8). The authors note in the first months of the pandemic the only major evolutionary adaption was the D614G mutation that enhanced affinity for the ACE2 site. This amino acid substation increased transmissibility and infectivity. Late last year new variants emerged with further enhancements in transmissibility: B.1.1.7 in the UK, B.1.351 in South Africa, and the P.1 and P.2 variants in Brazil.

Several of these variants of concern (B.1.1.7, B.1.351, and P.1) shared the same mutation in the receptor-binding domain of the S protein, N501Y. This mutation has been shown to increase virulence and transmissibility SARS-CoV-2 although recent studies suggest that it may not increase lethality in humans (Science 2020;369:1603-7; Lancet April 2021). The P.1 and B.1.351 lineages display substitution at a trio of sites in the receptor-binding domain (K417N, E484K, and N501Y). K417N and E484K are escape mutations that protect the virus from neutralizing antibodies (Cell Reports Medicine 2021;2:100255; Res Sq January 2021). In September, it appeared that Manaus, Brazil, had reached herd immunity with 75% of citizens infected (Science 2021;371:288-92). The P.1 variant with this trio of mutations subsequently emerged in Brazil last November, resulting in a surge of cases in Manaus. The genomic and mortality data suggest that the P.1 variant with this trio of mutations is 1.7 to 2.4 times more transmissible than wildtype, and that the prior infection with wildtype only provides 54%-79% protection (Science April 2021). Recently, the E484K mutation has also been observed in isolates of the B.1.1.7 lineage.

The evolutionary pressure for mutations that escape neutralizing antibodies is strongly opposed by the requirement to bind effectively to the human ACE2 receptor. One possible means to compensate for the loss of binding or infectivity that may occur in escape mutations is a second alteration in the S protein to retain binding affinity.

In a recent paper in Nature, Kemp and colleagues report on the in vivo evolution of SARS-CoV-2 in a patient unable to clear the infection due to lymphoma (Nature 2021;592:277-82). In this patient, SARS-CoV-2 acquired the D796H mutation, allowing escape from convalescent plasma that was given to the patient when it was clear that the patient could not clear the infection on his own. This mutation decreased affinity for the ACE2 receptor. The strains with the D796H mutations further evolved by deleting amino acids 69 and 70 (abbreviated Δ69-70) from the S protein peptide sequence. In vitro studies revealed that the virus bearing Δ69-70 and D796H displayed “modestly decreased sensitivity to convalescent plasma, while maintaining infectivity levels that were similar to the wild-type virus.” Individually, the D796H substitution was thought to have been the main source of reduced susceptibility to neutralizing antibodies, while the Δ69-70 deletion alone showed a two-fold greater infectivity relative to wild-type SARS-CoV-2. In summarizing their findings, the authors noted “These data reveal strong selection on SARS-CoV-2 during convalescent plasma therapy, which is associated with the emergence of viral variants that show evidence of reduced susceptibility to neutralizing antibodies in immunosuppressed individuals.”

Similarly, the N501Y mutation is one of several that arose in a U.S.-based immunocompromised patient who displayed viral replication for more than 20 weeks (N Engl J Med 2020;383:2291-3). In their NEJM correspondence, Choi et al. describe the case of a 45-year-old male with severe antiphospholipid syndrome who was on a complex regimen of cyclophosphamide, glucocorticoids, intermittent eculizumab and rituximab with anticoagulation therapy.

The phylogenetic analysis included in Choi et al. was “consistent with persistent infection and accelerated viral evolution.” One particularly salient observation was that amino acid alterations occurred primarily in the spike gene and the receptor-binding domain. Although the spike gene and receptor-binding domain only constitute 13% and 2% of the viral genome, respectively, they had 57% and 38% of the detected alterations. The overrepresentation of these mutations to these genes points to significant evolutionary pressure applied to them. Another important observation with implications for transmissibility was that infectious viral nasopharyngeal isolates were obtained from the patient on days 75 and 143.

These longitudinally obtained nasopharyngeal swabs provided insight into the evolution of the viral lineages within this patient. Samples taken on days 75 and 81 displayed lineages with the E484K mutation, while N501Y-bearing lineages were noted from day 128 until the last sampling (day 152).

Similarly, Avanzato et al. describe the history of a female patient with chronic lymphocytic leukemia and acquired hypogammaglobulinemia who had prolonged SARS-CoV-2 infection (Cell 2020;183:1901-12). In this case, infectious viral samples were obtained on days 49 and 70, while subgenomic RNA shedding was detected up to 105 days after initial diagnosis. No additional SARS-CoV-2 RNA was detected several weeks after a second course of convalescent plasma therapy. These authors also observed significant intra-host viral evolution, with a frequent turnover of dominant SARS-CoV-2 lineages. As a result, they concluded that in certain immunocompromised individuals, infectious viral shedding may occur for much longer than previously thought.

A recent South African study determined the different neutralization properties of plasma obtained from patients infected prior to (first wave) and after (second wave) the prevalence of B.1.351-lineage SARS-CoV-2 (Nature March 2021). The researchers noted that with the exception of one patient who had disease with an E484K alteration, all other first-wave patients had disease with no lineage-defining mutations for B.1.351.

The researchers found that the B.1.351 variant (i.e., second wave) virus was effectively neutralized by plasma from second-wave patients and from the lone first-wave patient with E484K-mutant disease. Conversely, other first-wave plasma typically showed a 15-fold decrease in neutralization of second-wave virus relative to second-wave plasma. Interestingly, the second-wave plasma showed only a 2.3-fold drop in neutralization of first wave virus relative to first-wave plasma. The lone plasma sample from E484K-only disease showed effective neutralization of both first- and second-wave viruses. From these data, the researchers postulate that future vaccines based on the sequences of variant-of-concern SARS-CoV-2 could retain neutralization potential against “other circulating SARS-CoV-2 lineages.”

In a recent report to the FDA, Janssen provided data from the Phase 3 ENSEMBLE (NCT04505722, COV30001) study evaluating efficacy, safety, and immunogenicity of a single-dose regimen of vaccine candidate Ad26.COV2.S (Table).

The vaccine efficacies for the prevention of moderate to severe COVID-19 at 28 days or more post-vaccination were as follows: USA – 72.0%, Brazil – 68.1%, South Africa – 64.0%. Additionally, the vaccine efficacies were even higher for the prevention of severe to critical cases of COVID-19 at 28 days or more post vaccination in the same geographies: USA – 85.9%, Brazil – 87.6%, South Africa – 81.7%. The vaccine efficacy values were somewhat lower in Brazil and South Africa, where disease bearing the E484K mutation were prevalent (the P.2 variant in Brazil has the E484K mutation but lacks the alterations at the K417 and N501 residues present in the P.1 variant). Even though vaccine efficacies were lower in E484K-prevalent areas, they were not significantly lower. However, more data need to be obtained before definitive conclusions can be drawn.

Israel has provided a real-life experiment of the role of vaccination in reducing SARS-CoV-2 evolution. The B.1.1.7 variant became the predominant strain in Israel in January, just the Pfizer/BioNTech BNT162b2 vaccine was aggressively rolled out to the entire population. As described by Munitz and colleagues, within two months, vaccination with BNT162b2 halted the spread of SARS-CoV-2, nearly ending SARS-CoV-2 transmission in Israel despite the increased infectiveness of the B.1.1.7 variant (Cell Reports Medicine April 2021). Stopping the spread of SARS-CoV-2 effectively prevents further evolution.

The University of Pittsburgh researchers noted in their conclusion that “the growing evidence for the emergence of immune escape mutations in protracted SARS-CoV-2 infection and for multiple, rapidly spreading variants should raise broad concern and action.” These authors expressed concern that half-measures such as the partial vaccine rollouts or incomplete immunizations could impede vaccination efforts via the development of escape variants in individuals with suboptimal levels of neutralizing antibodies. The most effective measure is to rapidly vaccinate as large a population as possible, while also increasing genotypic and phenotypic surveillance for emerging SARS-CoV-2 variants.

Meanwhile, both Pfizer and Moderna have developed and are now testing mRNA vaccines that encode the S protein with the known escape variants. If booster shots to address escape variants are required over the next year or so, it would be small price to pay to counter the evolution of SARS-CoV-2 and bring the COVID-19 pandemic to an end.

Do your part to advance the specialty by participating in a quality registry through the Anesthesia Quality Institute. Support patient safety, compliance and business objectives at asahq.org/registries.

Richard Simoneaux is a freelance writer with an MS in organic chemistry from Indiana University. He has more than 15 years of experience covering the pharmaceutical industry and an additional seven years as a laboratory-based medicinal chemist.