Programmed Cell Death Ligand Expression Drives Immune Tolerogenesis across the Diverse Subtypes of Neuroendocrine Tumours

<b><i>Introduction:</i></b> A comprehensive characterization of the tumour microenvironment is lacking in neuroendocrine tumours (NETs), where programmed cell death-1 receptor-ligand (PD-1/PD-L1) inhibitors are undergoing efficacy testing. <b><i>Objective:</i></b> We investigated drivers of cancer-related immunosuppression across NETs of various sites and grades using multi-parameter immunohistochemistry and targeted transcriptomic profiling. <b><i>Methods:</i></b> Tissue microarrays (<i>n</i> = 102) were stained for PD-L1 and 2 and indoleamine deoxygenase-1 (IDO-1) and evaluated in relationship to functional characteristics of tumour-infiltrating T-lymphocytes (TILs) and biomarkers of hypoxia/angiogenesis. PD-L1 expression was tested in circulating tumour cells (CTCs, <i>n</i> = 12) to evaluate its relationship with metastatic dissemination. <b><i>Results:</i></b> PD-L1 expression was highest in lung NETs (<i>n</i> = 30, <i>p</i> = 0.007), whereas PD-L2 was highest in pancreatic NETs (<i>n</i> = 53, <i>p</i> &#x3c; 0.001) with no correlation with grade or hypoxia/angiogenesis. PD-L1⁺ NETs (<i>n</i> = 26, 25%) had greater CD4⁺/FOXP3⁺ and CD8⁺/PD1⁺ TILs (<i>p</i> &#x3c; 0.001) and necrosis (<i>p</i> = 0.02). CD4⁺/FOXP3⁺ infiltrate had the highest PD-L1/IDO-1 co-expressing tumours (<i>p</i> = 0.006). Grade 3 well-differentiated NETs had lower CD4⁺/FOXP3⁺ and CD8⁺/PD1⁺ TIL density (<i>p</i> &#x3c; 0.001), and NanoString immune profiling revealed enrichment of macrophage-related transcripts in cases with poorer prognosis. We identified PD-L1(+) CTC subpopulations in 75% of evaluated patients (<i>n</i> = 12). <b><i>Conclusions:</i></b> PD-L1 expression correlates with T-cell exhaustion independent of tumour hypoxia and is enhanced in a subpopulation of CTCs, suggesting its relevance to the progression of NETs. These findings support a potential therapeutic role for PD-L1 inhibitors in a subset of NETs.