AI Identifies Pre-existing Antimicrobial Antibody Profile That May Predict Immune Response to Vaccination

Researchers analyzing antibody profiles in thousands of individuals have discovered that pre-existing antibodies to common microbes can predict the strength of new vaccine responses. The Arizona State University (ASU) team measured antibodies against 185 antigens—including those from common viruses, bacteria, and targets associated with autoimmune diseases—in blood samples from 4,000 immunosuppressed and healthy individuals.

The researchers then used artificial intelligence to analyze antibody patterns in samples collected before and after COVID-19 vaccination, identifying antibody signatures that helped distinguish strong vaccine responders from weak ones. In particular, they found that pre-existing antibodies to common microbes consistently predicted post-vaccination antibody responses in both healthy and immunosuppressed individuals.

These “sentinel antibodies,” the researchers suggest, may represent biomarkers of immune responsiveness to vaccination. “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 lead Joshua LaBaer, PhD, executive director of the Biodesign Institute at ASU and director of the Virginia G. Piper Center for Personalized Diagnostics. “This suggests that some people may be more immune-ready than others.”

The team’s approach is one of the first to use a broad, pre-vaccine antibody “fingerprint” to assess immune readiness. Unlike some prediction methods that rely on genetic analyses, this strategy uses antibody patterns in blood, which may be easier to adapt for clinical use.

LaBaer and colleagues reported their findings in Cell Press Blue, in a paper titled “Pre-vaccine sentinel antibodies predict blunted vaccine responses,” stating that their results “… identify pre-existing antimicrobial antibody profiles as scalable biomarkers of humoral immune responsiveness and provide a framework for predicting vaccine responses before immunization.”

Vaccines protect most people from serious illness, but the strength of that protection can vary considerably from one person to another. Before a vaccine ever enters the body, the immune system may already hold clues to how strongly it will respond. Age, sex, genetics, prior illnesses, and underlying health conditions have all been linked to how strongly people respond to vaccines. People with immune-compromising conditions are often at higher risk of weaker responses. But even within these groups, outcomes can differ sharply.

Usually, scientists evaluate vaccine response after the shot has been given by measuring whether the immune system has produced antibodies against the target. For their newly reported study, LaBaer and team asked whether antibody patterns already present in the blood might predict an individual’s immune readiness and response to vaccination.

The team looked at antibody responses to 185 antigens, including SARS-CoV-2 antigens, other common viral and bacterial antigens, and targets associated with autoimmune diseases. To do this, the researchers analyzed 8,687 samples from 4,089 participants, including 2,445 healthy volunteers and 1,644 people with conditions or treatments linked to immune suppression, such as HIV, multiple myeloma, solid organ malignancy, autoimmune disease, inflammatory bowel disease, and solid organ transplantation.

They found that several immunosuppressed groups were more likely to have blunted responses to COVID-19 vaccination. But those categories were imperfect predictors. Some immunosuppressed participants mounted strong responses, while about 5% to 6% of healthy participants demonstrated weak responses. The results did find that higher levels of certain preexisting antibodies, including antibodies to common bacteria and viruses, such as Staphylococcus aureus, respiratory syncytial virus, and human respirovirus 3, were associated with stronger COVID-19 vaccine responses.

The researchers describe these as “sentinel” antibodies because they may indicate a person’s baseline immune readiness. They are not necessarily fighting the vaccine target directly. Instead, they may reflect how responsive the antibody-producing arm of the immune system is likely to be. “These broadly prevalent antimicrobial antibodies represent sentinel antibodies that may serve as biomarkers of system-level humoral immune competence,” the team stated.

The researchers then asked whether the full antibody fingerprint, not just a few individual markers, could help identify people likely to have weak vaccine responses. A deep-learning model analyzed patterns across the antibody panel, combining measurements into a broader immune profile. “Using global antimicrobial antibody profiles, we developed a deep-learning predictive model that stratified individuals according to their likelihood of mounting blunted vaccine responses,” they explained.

The study highlights a key strength of AI in health research: its ability to find subtle, predictive patterns in millions of biological data points that might otherwise remain hidden. The approach suggests that vaccine readiness may be better understood by looking at the immune system as a whole, rather than focusing only on a single disease or a single antibody.

The work also highlights the value of newer technologies that can measure large numbers of antibody responses at once. Instead of asking whether someone has antibodies to one pathogen, the method can scan a wider immune landscape, capturing patterns formed by many previous encounters with viruses, bacteria, and other immune targets.

The researchers say the findings could have implications beyond COVID-19 if they are validated in additional studies and with other vaccines. Sentinel antibody profiling could help guide vaccine testing, vaccine development, and clinical care for people at risk of weak immune responses. “Together, these findings identify pre-existing antimicrobial antibody profiles as scalable biomarkers of humoral immune responsiveness and provide a framework for predicting vaccine responses before immunization,” the authors wrote in summary.

The approach might eventually help doctors identify patients who need additional vaccine doses, closer follow-up, or alternative protective measures. It could also help researchers better understand why some people respond well to vaccination while others do not. The work points toward a future in which vaccine decisions could be guided by a person’s own immune readiness.