ICOSL and miR-24 expression in DLBCL clinical tissues. (A) Three EBV-negative non-GC DLBCLs immunohistochemically stained for EBNA2 and ICOSL are shown. Cell membrane ICOSL positivity is in brown. These samples are negative for EBNA2. (B) Three cases of EBNA2-positive non-GC DLBCL and corresponding ICOSL expression are shown. Nuclear EBNA2 is shown in red. ICOSL is negative. (C) Scattered dot plot analysis representing ICOSL-positive cells in each case. The stained tissue sections were digitalized at a ×40 original magnification using Ventana DP200 Scan Scope. The number of positive cells was determined by manually counting 6 different areas and at least 100 cells in each area. For statistical significance, a 2-tailed unpaired t test was used; ∗P < .05, ∗∗∗∗P < .0001. (D) miR-24 expression and copy number in DLBCLs by RT-qPCR (left panel) and ddPCR (right panel). Statistical significance was calculated with two-way ANOVA and Tukey multiple comparisons test; ∗∗P < .01, ∗∗∗P < .001, ∗∗∗∗P < .0001. (E) miRNA-based immunotherapy of EBV-associated EBNA2-expressing B-cell lymphoma. A proposed model underscoring the use of PD-L1, targeting miR-34a mimics as suggested by our previous data³⁸ in combination with anti–miR-24 molecules to reconstitute ICOSL as an RNA-based immunotherapeutic approach (created with BioRender.com). Panel E is reproduced from Blandino et al,⁵⁶ licensed under a Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/).