Evusheld may be ineffective against the coming BQ.1 and BQ.1.1 wave

An awareness alert for those who are immunocompromised and provided Evusheld.

Correction: Initially I commented that the salt bridge for the K444T mutation was an interaction involving a cationic-anionic-cationic bridge in which the anion was a metal of some sort. However, after looking at the image further it’s clear the researchers are suggesting that the salt bridge is from an anionic carboxylate side chain (either from Glutamic Acid or Aspartic Acid), rather than an actual formation with a metal ion intermediary. An update has been made to the caption in italics to reflect this correction.

Evusheld is a combination monoclonal antibody injection and is the only combination monoclonal prophylactic/preventative available against SARS-COV2.

As a prophylactic agent its use is especially important for those who are immunocompromised and may not be able to mount a proper defense against SARS-COV2, making them more susceptible to worse outcomes if infected.

However, similar to each variant, subvariant, or whatever they’re called there are concerns that prior monoclonal antibody treatments may not be effective due to mutations in the virus’ spike protein—it’s the same argument with prior immunity and protection with each new variant (remember that most monoclonal antibodies have been sourced from people previously infected with COVID)

Right now it’s expected that the next wave of infections would come from the BQ.1 and BQ.1.1 strains of the virus.

From bostonglobe.com

The CDC’s COVID tracker shows the proportion of variants based on tests each week. Below the green bars are associated with BQ.1 and the weird, dark green color is BQ.1.1.

Over the past few weeks both appear to be increasing in prominence:

Among other things, I wanted to provide an update with these new variants and what the mutations may cause.

For now, I wanted to provide information on Evusheld as it relates to the possible BQ.1 and BQ.1.1 wave.

In my previous post on Evusheld I remarked that one of the monoclonals in this combination therapy, Tixagevimab, had reduced neutralizing capabilities against Ba.4/Ba.5 primarily due to the F486V mutation.

A Look at Evusheld

This mutation appears to have carried over from Ba.5 as it’s assumed that the BQ.’s branched from Ba.5 (not sure; not my main concern right now).

This would mean that Tixagevimab may not be effective against the BQ.’s.

However, new mutations in the BQ.’s also suggest that the other monoclonal, Cilgavimab, may lose effectiveness and therefore may render Evusheld ineffective.

The noteworthy mutations here are the R346S mutation and the K444T mutation. The former is a change from a positively charged Arginine to a neutral, alcohol residue Serine. The latter is a change from a positively charged Lysine residue to a neutral Tyrosine residue.

Modeling from a recent preprint¹ shows the interaction between Cilgavimab and the spike of premutation vs BQ mutations.

The modeling suggests that, in the case of the R346S mutation, the Arginine is involved in hydrogen bond formation with Cilgavimab. However, the mutation to a Serine removes that hydrogen bond formation and reduces the binding affinity of the antibody.

From Qu, et al. The guanidinium group (with the 3 blue dots noting the nitrogens) of Arginine is critical to forming multiple hydrogen bonds (dashed yellow lines) between residues of Cilgavimab. As can be noted, hydrogen bond formation appears to happen between Arginine of the spike and the Serine and Tyrosine residues of Cilgavimab. However, the Serine mutation loses the ability to form multiple hydrogen bonds. The model also suggests sterics may be at play as the Serine -OH sidechain is too short to interact with the light and heavy chains of the antibody.

For the K44T mutation the Lysine residue forms attachment with Cilgavimab through the formation of a salt bridge, which is broken with the Tyrosine residue due to loss of the positive charge:

The image on the left shows attachment of an antibody (heavy chain in cyan) with the spike protein (in green). The two red dots suggests oxygen atoms and indicates a possible carboxylic acid side chain while the blue dot suggests a nitrogen from the Lysine. The negative charge from the carboxylate can interact with the positive charge from the amine and form a salt bridge. However, the mutation to Tyrosine may break the salt bridge formation due to loss of the positive charge and preventing the antibody association.

Generally such models should be viewed with a bit of skepticism, but it at least provides some possible explanation to what’s happening.

All of this is in line with the FDA Fact Sheet for Evusheld, which indicates that serial passage studies of each individual monoclonal antibody led to the dominance of these mutations (emphasis mine):

Escape variants were identified following serial passage in cell culture of SARS-CoV-2 or replication competent recombinant vesicular stomatitis virus (VSV) expressing SARS-CoV-2 spike protein in the presence of tixagevimab or cilgavimab individually or in combination. Tixagevimab selected a variant expressing F486S in the spike protein with a >800-fold reduction in susceptibility to tixagevimab. Cilgavimab selected variants that expressed spike protein amino acid substitutions R346G, R346I, K444E, K444N, K444Q, K444R, K444T or N450D were each associated with a >200-fold reduction in susceptibility to cilgavimab. No escape variants to the tixagevimab and cilgavimab combination were selected.

In neutralization assays using recombinant VLPs pseudotyped with SARS-CoV-2 spike and harboring individual spike amino acid substitutions identified in circulating SARS-CoV-2, variants with reduced susceptibility to cilgavimab alone included those with R346I (>200-fold), K444E (>200-fold), K444Q (>200-fold), K444R (>200-fold), V445A (21- to 51-fold), G446V (4.2-fold), N450K (9.1-fold), or L452R (5.8-fold) substitutions. Variants with reduced susceptibility to tixagevimab alone included those with Q414R (4.6-fold), L455F (2.5- to 4.7-fold), G476S (3.3-fold), E484D (7.1-fold), E484K (6.2- to 12- fold), E484Q (3.0-fold), F486S (>600-fold), F486V (121- to 149-fold), Q493K (2.4- to 3.2-fold), Q493R (7.9-fold), E990A (6.1-fold), or T1009I (8.2-fold) substitutions. Variants harboring an E484K (2.4- to 5.4-fold), Q493R (3.4-fold), E990A (5.7-fold), or T1009I (4.5-fold) substitution exhibited low level reduced susceptibility to tixagevimab and cilgavimab in combination.

I specify the word individual here as Evusheld’s combination therapy is similar to Regeneron, in that one antibody targets the RBD of the spike (Tixagevimab) while the other binds peripherally (Cilgavimab) and likely provides steric hindrance. Both target nonoverlapping epitopes, and their combination use is likely intended to prevent both antibodies from losing effectiveness as mutations arise.

This is why Evusheld still has some protective benefits against BA.4/BA.5 even though Tixagevimab lost neutralization capabilities, as Cilgavimab was still able to bind.

However, similar to Regeneron and the coming of Omicron, the collection of mutations in the BQ.’s are enough to prevent both monoclonals from properly binding, meaning that Evusheld may not be effective against the coming BQ wave.

This is likely why the FDA provided an update to Evusheld’s Fact Sheet in early October, warning of the possible lack of effectiveness. This update likely included the serial passage studies from the excerpt above:

It’s important to remember that there are many people who are likely to be susceptible to SARS-COV2, including those who are immunocompromised.

Therefore, this post should serve almost as a PSA (“if you or your loved one has taken Evusheld”…) to alert people about the possible risk of increased susceptibility with the possible coming of the BQ.’s wave.

It’s important that medical professionals raise awareness, and that patients have discussions with their doctors in order to figure out how to possibly navigate the BQ.'s wave, if they are to arise.

---

If you enjoyed this post and other works please consider supporting me through a paid Substack subscription or through my Ko-fi. Any bit helps, and it encourages independent creators and journalists outside the mainstream.

Click on the image above or visit the website at https://ko-fi.com/moderndiscontent

1. Qu, P., Evans, J. P., Faraone, J., Zheng, Y. M., Carlin, C., Anghelina, M., Stevens, P., Fernandez, S., Jones, D., Lozanski, G., Panchal, A., Saif, L. J., Oltz, E. M., Xu, K., Gumina, R. J., & Liu, S. L. (2022). Distinct Neutralizing Antibody Escape of SARS-CoV-2 Omicron Subvariants BQ.1, BQ.1.1, BA.4.6, BF.7 and BA.2.75.2. bioRxiv : the preprint server for biology, 2022.10.19.512891. https://doi.org/10.1101/2022.10.19.512891