Increased oral Epstein-Barr virus shedding with HIV-1 co-infection is due to a combination of B cell activation and impaired cellular immune control

Epstein-Barr virus (EBV) infection is transmitted by saliva and is a major cause of cancer in people living with HIV/AIDS as well as in the general population. To better understand the determinants of oral EBV shedding we evaluated the frequency and quantity of detectable EBV in the saliva in a prospective cohort study of 85 adults in Uganda, half of whom were co-infected with HIV-1. Participants were not receiving antiviral medications, and those with HIV-1 co-infection had a CD4+ T cell count >200 cells/mm3. Daily, self-collected oral swabs were collected over a 4-week period. Compared with HIV-1 uninfected participants, co-infected participants had an increased frequency of oral EBV shedding (IRR=1.27, 95% CI=1.10-1.47). To explain why EBV oral shedding is greater in HIV-1 co-infected participants, we developed a stochastic, mechanistic mathematical model that describes the dynamics of EBV, infected cells, and antiviral cellular immune responses within the tonsillar epithelium, and examined parameter-specific differences between individuals of different HIV-1 infection statuses. We fit the model to our observational data using Approximate Bayesian Computation. After fitting, model simulations showed high fidelity to daily oral shedding time-courses and matched key summary statistics. Examination of the model revealed that higher EBV loads in saliva are driven by B cell activation causing EBV lytic replication in the tonsils, in combination with a less effective EBV-specific cellular immune response. Thus, both these factors contribute to higher and more frequent EBV shedding in HIV-1 co-infected individuals compared to HIV-1 uninfected individuals. These conclusions were further validated by modelling daily oral EBV shedding in a 26-participant North American cohort. Our results provide insights into the determinants of EBV shedding and implicate B cell activation to be a potential therapeutic target to reduce EBV replication in HIV-1 co-infected individuals at high risk for EBV-related malignancies.