The stunning surge of SARS-CoV-2 infections after the emergence of the Omicron variant highlights a major challenge in fighting the virus: it’s a moving target. Keeping up requires that we rapidly recognize new variants and understand how they affect transmission and immunity. In one sense, our ability to watch the Omicron variant spread across the globe in real time represents the success of the scientific effort to track the virus’s evolution. But the struggle to contain that spread also highlights our limitations in responding to such changes. A challenge of particular concern is how to recognize and rapidly respond to variants that can evade pre-existing immunity to SARS-CoV-2. “A tremendous amount of experimental effort has been expended to characterize these SARS-CoV-2 variants in neutralization assays. Unfortunately, the rate at which new variants arise outstrips the speed at which these experiments can be performed,” explained Dr. Jesse Bloom, Professor in Fred Hutch’s Basic Sciences and Public Health Sciences Divisions, and colleagues in a new article on bioRxiv. In light of this limitation, Dr. Bloom’s group has previously aimed to get ahead of viral evolution by using high-throughput approaches to prospectively examine how a large number of individual mutations in the virus’s spike protein – the major antigenic target – affect antibody binding and neutralization. But this approach, too, has its limits. While it is valuable for studying individual mutations, when we start to consider combinations of mutations (Omicron has more than 30 separate mutations in its spike protein including 15 in the receptor-binding domain, the main target of neutralizing antibodies), the number of possible combinations is so mind-bogglingly large that “it is not feasible to experimentally characterize all combinations of mutations even via high-throughput approaches,” the authors wrote. In the current study, Dr. Bloom, graduate student Allie Greaney, and postdoctoral researcher Tyler Starr developed a tool to predict how any combination of mutations will impact SARS-CoV-2 antibody escape.
The goal of this work was to use existing experimental data to make predictions about how untested mutations would affect antibody binding and neutralization. The existing data was immense – comprehensive results of the impact of every possible single mutation in a key region of the spike protein on binding by 33 monoclonal antibodies – but was largely limited to single mutations. The authors first aggregated the data for the 33 monoclonal antibodies to simulate the polyclonal nature of human immunity. They then employed a mathematical “escape calculator” to predict how any arbitrary set of mutations would affect neutralization by combining the relative decrease in efficacy of each antibody as a result of each mutation. “Specifically, we reduce the contribution of each antibody [to escape prevention] by an amount that scales with how strongly that antibody targets each mutated site,” they explained.
The authors next tested the accuracy of their calculator by comparing their calculated binding score – an estimate of how much antibody binding to the spike protein remains after a given set of mutations in introduced – with experimentally validated impacts on neutralization by polyclonal serum against known SARS-CoV-2 variants. Incredibly, they found a strong correlation between their calculated binding score and the measured level of neutralization, indicating that their admittedly simple mathematical approach could accurately predict polyclonal antibody escape.
Finally, the group used their escape calculator to assess the Omicron variant. “The calculated binding score for the Omicron variant is much lower than any other SARS-CoV-2 variants of concern, indicating extensive antibody escape,” they wrote, a finding consistent with the now evident increased propensity of Omicron to infect vaccinated and previously infected people. Importantly, the calculator also identified several sites of concern within the spike protein that, if mutated, could drive additional escape by the Omicron variant. This simple but effective escape calculator, combined with continuing genomic surveillance of SARS-CoV-2, provides a powerful new tool in the attempts to keep up with the inevitable continued evolution of SARS-CoV-2.
This work was supported by the National Institutes of Health, the Gates Foundation, the Damon Runyon Cancer Research Foundation, and the Howard Hughes Medical Institute.
Greaney AJ, Starr TN, Bloom JD. An antibody-escape calculator for mutations to the SARS-CoV-2 receptor-binding domain. bioRxiv [Preprint]. 2021 Dec 7:2021.12.04.471236. doi: 10.1101/2021.12.04.471236. Update in: Cell Host Microbe. 2021 Jan 13;29(1):44-57.e9. PMID: 34909770; PMCID: PMC8669837.