D91-08 T-cells And Antibodies Act Synergistically to Mediate Delayed Rejection of AAV6.2 Transduced Cells in the Mouse Lung

Abstract Rationale Lung-directed gene therapy offers a path to treat diseases such as cystic fibrosis, primary ciliary dyskinesia, alpha-1-antitrypsin deficiency, and lymphangioleiomyomatosis. A major barrier in systemic gene therapy has been pre-existing anti-vector immunity, which limits expression and prevents re-dosing. In contrast, the role of pre-existing immunity in mucosal AAV delivery is poorly defined. For two decades, animal models have failed to reproduce the T-cell-mediated rejection observed in humans. We sought to establish a model of anti-AAV immunity in the lung and test whether combined immune priming drives rejection of transduced cells. Methods BALB/c mice received intratracheal AAV6.2-Luciferase with or without systemic pre-existing antibodies, T-cells, or both. Antibodies were introduced by passive serum transfer from AAV6.2-GFP-immunized mice. Cross-reactive T-cells were generated by Ad-AAV8 immunization. Controls received PBS or Ad-GFP. Transgene expression was tracked by bioluminescent imaging. Flow cytometry assessed dendritic cell maturation and AAV- or luciferase-specific CD8+ T-cells. Antigen-specific T-cells were sorted for single-cell TCR + GEX RNA-seq to define repertoire and transcriptional states. Results Antibody transfer alone had little effect on transgene expression, while T-cell priming alone produced a modest early reduction that stabilized over time (figure 1). In contrast, the combination of antibodies and T-cells led to a distinct trajectory: expression initially paralleled T-cell primed controls but then diverged, culminating in delayed loss of signal. This loss correlated with the accumulation of luciferase- and AAV-specific CD8+CD69+CD103+ tissue-resident memory T cells (TRM) in the lung and bronchoalveolar lavage fluid. Dendritic cell profiling showed that both T-cell-primed groups had similar numbers of immature and mature DCs at early timepoints, but only mice with combined immunity exhibited persistence of mature CD11c+MHC-II+CD80+ DCs. Single-cell sequencing performed at the point of divergence revealed that luciferase-specific T cells were transcriptionally similar across groups, whereas AAV-specific T cells in the combined immunity condition displayed reduced expression of exhaustion markers and included a distinct subset enriched for dendritic cell activation signatures, consistent with active immunological synapse formation. Conclusions We demonstrate that both antibodies and T-cells are required to trigger CD8+ mediated rejection of AAV-transduced lung cells. Antibodies likely enhance cross-presentation of capsid antigen by dendritic cells, enabling robust T-cell activation. This model reproduces the delayed rejection seen in human trials and provides a platform to dissect cellular mediators of rejection. Defining these pathways will be critical for strategies to prolong transgene expression and to enable safe and effective lung-directed gene therapy. This abstract is funded by: NIH, Cystic Fibrosis Foundation