A bit of hit and miss – how well antibody responses in pre-clinical models resemble human responses

From the Overbaugh Lab, part of the Human Biology Division

Many of us now vaxxed, boosted and living relatively normal lives in the nearing post-pandemic world may find it hard to recall the early days of the vaccine roll out. Perhaps you’ve tried (understandably) to block those days from your memory, but maybe you can remember obsessively checking the effectiveness of Moderna and Pfizer vaccines in early-stage clinical trials, wondering who would win the race to be the first available COVID-19 vaccine. But what happened before these vaccines even made it into humans? Rest assured, the COVID-19 vaccine (and vaccines in general) have to go through a lot of checks ensuring both their safety and their efficacy before they can be given to people. “Non-human primates (especially macaques) are considered the gold standard animal model for studying infectious disease in humans,” explains graduate student Alex Willcox from the lab of Dr. Julie Overbaugh, part of the Human Biology Division. In the case of the COVID-19 vaccine, “researchers started using macaques as a pre-clinical model for SARS-CoV-2 infection and vaccination early in the pandemic. In order to use this data to predict vaccine outcomes in humans, it is important to know if the antibody responses in macaques and humans are similar,” Dr. Overbaugh adds. In a recent paper led by Willcox and published in Plos Pathogens, the Overbaugh team asked how similar the antibody responses of vaccinated human and macaques were to SARS-CoV-2 infection. Specifically, they used a method developed in their lab called Phage DMS, allowing them “to compare antibody binding responses across the Spike protein, which is the target of antibodies that block infection and kill infected cells,” notes Dr. Overbaugh.

Willcox explains that a few months into the pandemic they began to wonder “how well antibody responses in non-human primates resemble the human antibody response to SARS-CoV-2 vaccination and infection.” They leveraged human and macaque serum samples collected from collaborators early in the pandemic to perform this study. Using these banked samples from prior studies allowed them to maximize the information gained from these early studies to inform the value of the preclinical model. The Overbaugh team had already developed a tool, Phage DMS, that they knew could be applied to these samples to help answer their question. Phage-DMS, or phage-based deep mutational scanning, “involves making every possible amino acid mutation to each locus along a protein sequence and assaying how each mutation affects antibody binding,” states Willcox. Willcox et al. used this technique to compare which part of the Spike protein is being targeted by antibody responses in humans and macaques that received the COVID-19 vaccine as compared to convalescent humans and macaques, or those that were infected with SARS-CoV-2. This method allows researchers to generate escape profiles showing which mutations in the Spike protein reduced antibody binding and which mutations enhanced binding. Thus, the authors could identify two important aspects of the antibody response: they could define the place where the antibodies bind and they could further define the mutations that “escape” binding of each antibody.  The mutations identified by this method are “commonly used to predict the impact of a new mutation in the SARS-CoV-2 Spike protein,” states Willcox.

Enrichment of Spike protein regions identified by antibodies in humans and preclinical models
Enrichment of Spike protein regions identified by antibodies in humans and preclinical models Image taken from original article

In this study “the main goal was to explore what macaque antibody responses look like and compare them to human responses,” exclaims Willcox. The researchers began with two comparisons: vaccinated macaques vs. vaccinated humans and convalescent macaques vs. convalescent humans. The authors found “broad similarities between groups, but also important differences in the recognition of certain epitopes and the antibody escape profiles,” notes Willcox. For instance, the Overbaugh team found that both species recognize major epitopes in certain regions of the Spike protein, including the C-terminal domain (CTD), the fusion peptide, and the stem helix-heptad repeat 2 regions. Important differences between groups included a response to certain epitopes in the N- and C-terminal domains in vaccinated humans but not in vaccinated pre-clinical models, as well as recognition of a CTD epitope and epitopes flanking the fusion peptide in convalescent macaques, but not convalescent humans. Intriguingly, the most similarity was found between vaccinated humans and convalescent macaques. Willcox theorizes that this similarity may be due to both groups having “two strong exposures to Spike: the vaccinated human received two doses of the Moderna vaccine, and the convalescent macaques were infected twice with SARS-CoV-2 in an experimental setting, so the dose they received was likely much higher than in natural infection.” She continues, “My hypothesis is that repeated exposure to a strong antigenic stimulus caused the antibody responses to converge in these two groups.” The researchers also found more variability than expected between individuals in addition to “evidence of pre-existing antibody responses to SARS-CoV-2 in some of the macaque samples, suggesting prior exposure to other coronaviruses,” Willcox mentions. Something to consider is that like humans, these clinical models are pretty genetically diverse – a major difference from other model organisms such as flies or mice. Collectively, these factors are all important to take into consideration “when interpreting results from macaque studies,” Willcox notes. She goes on to say, “I also hope our study is just a starting point. Now that we are much farther along in the pandemic, we can leverage many more banked sample sets collected by collaborators that offer new opportunities including longitudinal analysis of antibody responses in response to different combinations of vaccination and infection.”

This work was supported by the National Institutes of Health, the National Institute of Allergy and Infectious Disease, the National Institutes of Health Office of Research Infrastructure Programs, HDT Bio Corp internal funds, the Howard Hughes Medical Institute, and the Simons Institute.

UW/Fred Hutch Cancer Consortium members Dr. Julie Overbaugh, Dr. Frederick Matsen, and Dr. Denise Galloway contributed to this work.

Willcox AC, Sung K, Garrett ME, Galloway JG, Erasmus JH, Logue JK, Hawman DW, Chu HY, Hasenkrug KJ, Fuller DH, Matsen Iv FA, Overbaugh J. Detailed analysis of antibody responses to SARS-CoV-2 vaccination and infection in macaques. PLoS Pathog. 2022 Apr 11;18(4):e1010155. doi: 10.1371/journal.ppat.1010155. PMID: 35404959; PMCID: PMC9022802.