Back

HIV-associated non-Hodgkin lymphoma tumor-microenvironment axes differ by EBV status across cellular origins

Chadburn, A.; Aguilar Hernandez, M. M.; Dai, J.; Harrison, M.; Reinoso-Vizcaino, N. M.; Barry, A. P.; Hocke, E.; Abramson, K.; Chan, C.; Cesarman, E.; Luftig, M. A.; SoRelle, E. D.

2025-10-16 cancer biology
10.1101/2025.10.15.682084 bioRxiv
Show abstract

Non-Hodgkin Lymphoma (NHL) is the main cancer-related mortality for people living with HIV (PLWH). NHL genetic and molecular classifications have been intensely studied and correlated with clinical outcomes, but critical unanswered questions relevant to malignancy persist. For example, tumors positive for Epstein-Barr virus (EBV) are aggressive and account for 30-50% of HIV-associated NHLs, yet insights on the nature of EBV in NHL pathogenesis or potential therapeutic vulnerabilities have been limited. Here, we examined HIV-associated NHLs stratified by histopathologic classification, cell of origin (COO), molecular subtype, and EBV status using genome-wide spatial transcriptomic analyses. Tumor tissues in EBV+ HIV-NHLs displayed enriched expression of mitochondrial respiration and purine metabolism signatures versus EBV- tumors. Immune infiltration of EBV+ tumors was limited, and immune and stromal regions of EBV+ HIV-NHLs displayed upregulation of the gene for the iron exporter ferroportin (SLC40A1), indicating immunosuppressive macrophage polarization. An immunosuppressive SPP1-CD44 axis and oncogenic GAS6-AXL interaction between tumor and stroma were specifically predicted for EBV+ HIV-NHLs with plasmablastic features. Tumor microenvironment (TME) similarities to secondary lymphoid tissues such as inverse CXCL12-CXCL13 gradients and evidence of matching between tumor B cell phenotypes and reticular cell signatures were also observed. Finally, we developed a simple yet flexible approach to quantify expression gradients and cell proximity to annotated regions. Thus, this study highlights potential avenues for NHL therapy tailored by EBV status and provides a unique resource to examine tumor-TME interactions in the HIV-immunocompromised context. Key PointsO_LISpatial transcriptomic landscapes resolve cell-of-origin- and EBV-associated HIV-NHL heterogeneity including immunosuppressive myeloid responses to EBV+ tumors C_LIO_LISpatial gradients including CXCL12 and CXCL13 indicate that tumor-stromal interactions mimic lymphoid tissue organization and depend on B cell development stage C_LIO_LIDevelopment of a direction-agnostic method to calculate spatial expression gradients relative to annotated tissue and cell types of interest C_LIO_LIPredicted pharmacologic vulnerabilities in clinical samples are replicated in vitro and confirmed by small molecule screening C_LI

Matching journals

The top 10 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.