PD-1/PD-L1 expression and interaction by automated quantitative immunofluorescent analysis show adverse prognostic...

Programmed cell death protein 1/programmed cell death protein ligand1 (PD-1/PD-L1) interaction is an important immune checkpoint targeted by anti-PD-1/PD-L1 immunotherapies. However, the observed prognostic significance of PD-1/PD-L1 expression in diffuse large B-cell lymphoma treated with the standard of care has been inconsistent and even contradictory. To clarify the prognostic role of PD-1/PD-L1 expression and interaction in diffuse large B-cell lymphoma, in this study we used 3-marker fluorescent multiplex immunohistochemistry and Automated Quantitative Analysis Technology to assess the CD3⁺, PD-L1⁺, and PD-1⁺CD3⁺ expression in diagnostic samples and PD-1/PD-L1 interaction as indicated by presence of PD-1⁺CD3⁺ cells in the vicinity of PD-L1⁺ cells, analyzed their prognostic effects in 414 patients with de novo diffuse large B-cell lymphoma, and examined whether PD-1/PD-L1 interaction is required for the prognostic role of PD-1⁺/PD-L1⁺ expression. We found that low T-cell tissue cellularity, tissue PD-L1⁺ expression (irrespective of cell types), PD-1⁺CD3⁺ expression, and PD-1/PD-L1 interaction showed hierarchical adverse prognostic effects in the study cohort. PD-1/PD-L1 interaction showed higher sensitivity and specificity than PD-1⁺ and PD-L1⁺ expression in predicting inferior prognosis in patients with high CD3⁺ tissue cellularity ("hot"/inflammatory tumors). However, both PD-1⁺ and PD-L1⁺ expression showed adverse prognostic effects independent of PD-1/PD-L1 interaction, and PD-1/PD-L1 interaction showed favorable prognostic effect in PD-L1⁺ patients without high CD3⁺ tissue cellularity. Macrophage function and tumor-cell MYC expression may contribute to the PD-1-independent adverse prognostic effect of PD-L1⁺ expression. In summary, low T-cell tissue cellularity has unfavorable prognostic impact in diffuse large B-cell lymphoma, and tissue PD-L1⁺ expression and T-cell-derived PD-1⁺ expression have significant adverse impact only in patients with high T-cell infiltration. PD-1/PD-L1 interaction in tissue is essential but not always responsible for the inhibitory effect of PD-L1⁺/PD-1⁺ expression. These results suggest the benefit of PD-1/PD-L1 blockade therapies only in patients with sufficient T-cell infiltration, and the potential of immunofluorescent assays and Automated Quantitative Analysis in the clinical assessment of PD-1/PD-L1 expression and interaction.