A study led by York University investigated how individuals with compromised immune systems respond to vaccines. The research focused on differences in vaccine-initiated immune responses between people living with HIV and HIV-negative individuals. The findings indicate the potential for personalized vaccination intervention strategies.
Methodology
Researchers utilized a dataset tracking individuals with and without HIV who received up to five doses of the COVID-19 vaccine over 100 weeks. The HIV-positive participants, located in the Greater Toronto Area, were undergoing antiretroviral therapy. A machine-learning method called random forest was applied to analyze 64 immune biomarkers. This analysis identified core differences between the groups, which were then used to generate 'virtual patients' to model immune responses.
Key Findings
- Saliva-based antibodies, particularly IgA, combined with white blood cell markers, were identified as key differentiators between the HIV-positive and HIV-negative groups.
- Machine-learning models accurately classified these differences with nearly 100% accuracy.
- Outliers were observed: A small subset of HIV-positive individuals displayed vaccine-induced immune responses indistinguishable from the HIV-negative group, suggesting effectively restored immune function in terms of vaccination response.
- Conversely, one individual in the healthy control group presented immune markers similar to those living with HIV, potentially indicating unidentified underlying immune issues.
Significance
Chapin Korosec, lead author, stated that the study represents a step toward personal vaccination intervention strategies by establishing a data-driven foundation for personalized vaccination and therapeutic design. Professor Jane Heffernan emphasized the complexity and individual variability of immune responses, highlighting the importance of personalized vaccination strategies.
"The study represents a step toward personal vaccination intervention strategies by establishing a data-driven foundation for personalized vaccination and therapeutic design."
Publication and Support
The study was published as a pre-print in the Journal Patterns and is slated to appear as a cover article. Support for the research was provided 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).