CD14- CD16+ monocyte PD-L1 prevents early tuberculosis progression and constrains reactivation under immune checkpoint therapy

Immune checkpoint blockade (ICB) has revolutionized cancer therapy, yet its unintended impact on chronic infections remains poorly understood. Here, we identify PD-L1 expression on CD14^- CD16⁺ monocytes as a critical determinant for preventing early tuberculosis (TB) progression. By integrating real-world adverse event data, Mendelian randomization (MR), and transcriptomic analysis, we establish a causal and cell-specific link between PD-L1 downregulation and early TB progression. Analysis of the FAERS database revealed that the PD-L1 inhibitor durvalumab is significantly associated with TB-related adverse events (reporting odds ratio = 7.81; 95% CI: 4.43-13.78; <i>P</i> = 1.10×10^-18). MR analysis confirmed that genetically elevated PD-L1 expression in CD14^- CD16⁺ monocytes confers protection against early TB progression (OR = 0.918, <i>P</i> = 0.042), independent of confounding or reverse causality. Transcriptomic profiling revealed that PD-L1^high CD14^- CD16⁺ monocytes exhibit enhanced antigen surveillance, whereas PD-L1^low cells display metabolic reprogramming associated with immune escape. Upstream regulatory analysis identified CDAN1, TCOF1, and LMO2 as transcriptional drivers of PD-L1, enriched in high-risk individuals. In <i>silico</i> drug prediction and molecular docking suggested several PD-L1-modulating compounds, including ruthenium, pomalidomide, zidovudine, and lycorine. Notably, several of these compounds (e.g., ruthenium complexes, pomalidomide, aziridine) have reported anti-mycobacterial effects, which supports the reliability of our screening results and their potential relevance to TB regulation. <i>In vitro</i> validation demonstrated that lycorine dose-dependently upregulates PD-L1 and inhibits <i>Mycobacterium tuberculosis</i> reactivation. Together, our findings define a mechanistic axis in CD14^- CD16⁺ monocytes that underpins early TB control and is vulnerable to PD-L1 blockade. Collectively, these findings align with the established notion that assessing latent tuberculosis infection before initiating immune-modulating therapies is essential for minimizing reactivation risk, and propose tractable molecular targets for preventing TB reactivation in immunocompromised hosts.