Abstract 5310: Spatially resolved cell profiling of testicular diffuse large B-cell lymphoma unveils recurrent cellular neighborhoods associated with survival

Introduction: Testicular diffuse large B-cell lymphoma (T-DLCBL) is a rare and aggressive lymphoid cancer, accounting for 1% of all lymphomas. In recent years, the impact of the tumor microenvironment (TME) on DLBCL pathogenesis and survival has been recognized. We have previously shown that in T-DLBCL, the T cell-inflamed TME is associated with favorable survival. In this study, we aimed to dissect the T-DLBCL TME in more detail and characterize its spatial architecture at the single-cell resolution. Methods: We used a tissue microarray (TMA) containing formalin-fixed paraffin-embedded diagnostic T-DLBCL samples from 79 patients. TMA was stained with a 12-plex antibody panel to characterize B cells, T cells, macrophages, and immune checkpoint molecules. Image processing and quality control were performed by combining Ilastik and CellProfiler softwares, and nuclei segmented by a pretrained deep learning segmentation model nucleAIzer. Single cell features from the images were extracted using histoCAT, and cell phenotyping and cell neighborhood analyses were performed utilizing Python package SCIMAP. Results: We identified over 900,000 single cells, which were classified into 16 distinct phenotypes, including various subpopulations of B cells, T cells, and macrophages. In line with our previous findings, higher T cell proportions were associated with better outcomes. In addition, a subgroup of PD1+ B cells (13% of the whole cell population) was associated with worse outcomes independent of the treatment or international prognostic index (IPI). Neighborhood analysis revealed 10 recurrent cellular neighborhoods (RCNs) with differential cellular compositions, including B cell, macrophage, T cell, or immune checkpoint-rich RCNs. The most abundant was an RCN rich in B cells and macrophages (25% of all cells), followed by an immune-depleted RCN (17% of all cells) and an RCN rich in PD1+ B cells (16% of all cells). Interestingly, a subset of T-DLBCLs was characterized by a very high PD1+ B cell-rich RCN content (>50% of the total cellularity), which translated to poor survival independent of the treatment or IPI. In contrast, a T cell-rich RCN was associated with better outcomes. Finally, when we calculated the minimum average distance between the RCNs, we noticed that a shorter distance between the immune-depleted RCN and an RCN rich in T cells and non-immune cells was associated with better survival. Conclusions: In summary, the study provides detailed insights into the spatial architecture of the T-DLBCL TME, highlighting the significance of specific cell populations and their neighborhoods in predicting patient outcomes. Citation Format: Suvi-Katri Leivonen, Matias Autio, Marjukka Pollari, Marja-Liisa Karjalainen-Lindsberg, Teijo Pellinen, Sirpa Leppä. Spatially resolved cell profiling of testicular diffuse large B-cell lymphoma unveils recurrent cellular neighborhoods associated with survival [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5310.