Pre-existing antibody patterns may reveal how strongly a person will respond to vaccination, helping identify individuals at risk of a blunted immune response. Credit: Graphic by Jason Drees for the Biodesign Institute at Arizona State University
A person's blood may already hold clues to how strongly they'll respond to a vaccine before they ever receive it, according to a new study that used artificial intelligence to identify patterns in pre-vaccination antibody levels.
For the study, published in Cell Press Blue, researchers measured antibodies against 185 antigens in 8,687 blood samples from 4,089 people. The samples, collected before and after COVID-19 vaccination, came from both healthy volunteers and people with conditions or treatments linked to immune suppression, such as HIV. A deep-learning model then analyzed patterns across the full antibody panel to distinguish strong vaccine responders from weak ones.
Age, sex, genetics, prior illness and underlying health conditions have all been linked to vaccine response, and people with immune-compromising conditions are generally considered more likely to respond weakly.
But the study found those categories were imperfect predictors on their own: some immunosuppressed participants mounted strong responses, while about 5 to 6% of healthy participants had weak ones.
Instead, the researchers found that higher levels of certain preexisting antibodies—including common microbes such as Staphylococcus aureus, RSV and human respirovirus 3—were associated with stronger COVID-19 vaccine responses. The researchers call these “sentinel” antibodies, since they aren't necessarily fighting the vaccine target directly but may instead reflect how responsive a person's antibody-producing immune system is likely to be overall.
“What our study found is that certain biomarkers, when analyzed with AI, can predict who is likely to respond well to a vaccine, even before they receive it,” said study author Joshua LaBaer, executive director of the Biodesign Institute at Arizona State University. “This suggests that some people may be more immune-ready than others.”
Because the method relies on antibody patterns in blood rather than genetic analysis, the researchers say it could be easier to adapt for clinical use than some existing prediction approaches.
If the findings hold up in future studies with other vaccines, sentinel antibody profiling could eventually help doctors identify patients who need additional vaccine doses, closer follow-up or alternative protective measures—pointing toward a future in which vaccination strategies are guided by a person's own immune readiness rather than applied uniformly.
Data from Arizona State University