The way individuals respond to vaccines is far from uniform, a reality that has become increasingly clear during the COVID-19 pandemic. Now, a latest study led by York University is shedding light on the complex factors influencing vaccine efficacy, particularly in those with compromised immune systems. Researchers have harnessed the power of machine learning to identify key differences in immune responses between healthy individuals and people living with HIV, paving the way for more personalized vaccination strategies.
This research isn’t simply about identifying differences; it’s about understanding the intricate interplay of biological factors that dictate how our bodies react to immunization. The findings, published as a pre-print in the journal Patterns and slated to appear as a cover article on March 13th, 2026, suggest that saliva-based antibodies and white blood cell counts play a crucial role in distinguishing immune responses, offering a potential biomarker for predicting vaccine effectiveness. This operate represents a significant step toward a future where vaccination isn’t a one-size-fits-all approach, but rather a tailored intervention based on individual immune profiles.
Machine Learning Reveals Immune Signatures
The study, led by postdoctoral fellow Chapin Korosec, analyzed data from a cohort of individuals – both with and without HIV – who received up to five doses of a COVID-19 vaccine over a 100-week period. All participants living with HIV were from the Greater Toronto Area and were managing their condition with antiretroviral therapy. Researchers employed a machine-learning method called random forest to analyze 64 immune biomarkers triggered by the vaccine, creating “virtual patients” to model and predict immune responses. “By learning the structure of immune variability at scale, we move toward a data-driven foundation for personalized vaccination and therapeutic design,” explained Korosec.
The machine-learning model demonstrated remarkable accuracy – nearly 100 percent – in differentiating between the immune responses of individuals with HIV and those without. However, the analysis revealed intriguing outliers. Two individuals with HIV exhibited immune responses indistinguishable from the control group, suggesting their immune function was effectively restored. Conversely, one healthy individual displayed markers similar to those living with HIV, potentially indicating underlying, yet undetected, immune vulnerabilities.
The Role of Saliva Antibodies and White Blood Cells
The research team discovered that a specific combination of factors – saliva-based IgA antibodies coupled with white blood cell counts – created a distinct signature differentiating the two groups. This finding is significant, as prior research has indicated altered mucosal immunity in individuals living with HIV, and how We see impacted over both the short and long term. “We were able to reveal that saliva-based antibodies, particularly a type of antibody in the saliva called IgA, coupled with white blood cells, which have long been known to be associated with HIV status, create the signature difference between the two groups,” Korosec stated.
Individual Variability and Personalized Approaches
Professor Jane Heffernan, whose research at York University focuses on infectious disease modelling, emphasized the complexity of the immune response. “The immune response is very, very complicated,” she explained. “Sometimes something can act as an inhibitor of an arm of the immune response, but in other times it might be an activator. There is also a lot of individual variability among people with similar immune system status.” The study highlighted the importance of accounting for this variability when developing vaccination strategies.
The researchers identified subgroups within the HIV-positive cohort, further underscoring the require for personalized approaches. Using machine learning, mechanistic modelling, and “virtual patients,” they aim to uncover vital differences not only between groups but also among individuals, even considering immune components not directly measured in the data. This approach, Heffernan explained, is like “trying to discover the needle in a haystack, but with a clearer path to finding it.”
This research was supported by the National Research Council of Canada (NRC)-Fields Mathematical Sciences Collaboration Centre, the National Sciences and Engineering and Research Council of Canada, and Artificial Intelligence for Public Health (AI4PH). Korosec collaborated with researchers from the University of Toronto, St. Michael’s Hospital, Pennsylvania State University, and the NRC Digital Technologies Research Centre.
Looking ahead, this study offers a crucial foundation for refining vaccination strategies, particularly for individuals with compromised immune systems. Further research will focus on understanding the long-term maintenance of antibodies and the factors driving the observed variations in immune responses. As Korosec concluded, “This study moves us closer to understanding immune diversity in people living with HIV; how their responses compare to age-matched controls, how well antibodies are maintained over time, and why some individuals show strikingly different patterns.”
This research offers a promising step towards more effective and personalized vaccine interventions. Share your thoughts on the potential of AI in public health in the comments below.
Disclaimer: This article provides informational content and should not be considered medical advice. Consult with a qualified healthcare professional for any health concerns or before making any decisions related to your health or treatment.
