Longitudinal analysis of PD-1 receptor occupancy in spleen and tumor-infiltrating lymphocytes in a murine model.

Abstract

e14580 Background: Anti-PD-1 antibodies restore T-cell-mediated anti-tumor immune responses by blocking the interaction between PD-1 and programmed cell death ligand 1 (PD-L1). Following anti-PD-1 administration, PD-1 receptor occupancy (RO) on circulating T cells has been shown to vary widely in a dose-independent manner. We have previously demonstrated that subset-specific PD-1 RO on circulating T cells is associated with clinical outcomes. Given that tumor-infiltrating lymphocytes (TILs) are more directly involved in anti-tumor immunity than peripheral T cells, RO on TILs is expected to be more precisely correlated with anti-tumor activity. However, research to date has focused almost exclusively on peripheral T cells, and an analysis of PD-1 RO on TILs has not yet been reported. This study aimed to clarify the distinct kinetics of PD-1 RO on TILs and systemic T cells. Methods: BALB/c mice bearing CT-26 colon tumors were treated with a single dose (150 μg) of rat anti-PD-1 antibody on day 10 post-tumor inoculation. Spleen and TILs were harvested at 24 hours, 1 week, and 2 weeks post-injection. PD-1 RO was assessed via flow cytometry using anti-rat IgG to detect in vivo-bound antibody and anti-mouse PD-1 to assess total receptor expression. Based on the antibody binding status, T-cell subsets (CD8+, CD4+, and Foxp3+ Tregs) were classified into three functional states: saturated (PD-1−IgG+), representing complete receptor blockade where in vivo antibody prevents subsequent PD-1 detection; unsaturated (PD-1+IgG+), representing partial binding; and unbound (PD-1+IgG−). Matched-pair analysis was performed between splenic and intratumoral T cells within each individual mouse. Results: At 24 hours post-administration, PD-1 receptor saturation was approximately 40% in both the spleen and tumor-infiltrating lymphocytes (TILs) across total CD3+ T cells. RO on splenic T cells remained relatively stable at 40% after 1 week, and then decreased to 10% by 2 weeks. In contrast, RO on TILs exhibited a much more rapid decline, with RO dropping sharply to approximately 4% at 1 week and then to 2% by 2 weeks. Matched-pair analysis of individual mice confirmed that RO was significantly lower in TILs than in splenic T cells as early as 1 week post-injection. Despite this overall trend, inter-individual and subset-specific variability was also noted. Conclusions: Our results demonstrate that the kinetics of PD-1 blockade within the tumor environment are distinct from those in systemic circulation. This discrepancy suggests that systemic monitoring may not fully reflect the actual state of PD-1 blockade within the tumor microenvironment. We are currently investigating the correlation between these intratumoral RO dynamics and anti-tumor responses to identify more precise parameters for therapeutic success.